CN117455451A - Recovery method of glyphosate hydrolysis solvent - Google Patents
Recovery method of glyphosate hydrolysis solvent Download PDFInfo
- Publication number
- CN117455451A CN117455451A CN202311415906.5A CN202311415906A CN117455451A CN 117455451 A CN117455451 A CN 117455451A CN 202311415906 A CN202311415906 A CN 202311415906A CN 117455451 A CN117455451 A CN 117455451A
- Authority
- CN
- China
- Prior art keywords
- tower
- tower top
- training
- parameter
- inflow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000005562 Glyphosate Substances 0.000 title claims abstract description 97
- XDDAORKBJWWYJS-UHFFFAOYSA-N glyphosate Chemical compound OC(=O)CNCP(O)(O)=O XDDAORKBJWWYJS-UHFFFAOYSA-N 0.000 title claims abstract description 97
- 229940097068 glyphosate Drugs 0.000 title claims abstract description 97
- 239000002904 solvent Substances 0.000 title claims abstract description 97
- 238000006460 hydrolysis reaction Methods 0.000 title claims abstract description 81
- 230000007062 hydrolysis Effects 0.000 title claims abstract description 79
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000011084 recovery Methods 0.000 title abstract description 46
- 238000012549 training Methods 0.000 claims description 69
- 230000004927 fusion Effects 0.000 claims description 68
- 239000013598 vector Substances 0.000 claims description 57
- 238000010586 diagram Methods 0.000 claims description 37
- 239000011159 matrix material Substances 0.000 claims description 37
- 238000013527 convolutional neural network Methods 0.000 claims description 26
- 230000003301 hydrolyzing effect Effects 0.000 claims description 14
- 238000010219 correlation analysis Methods 0.000 claims description 9
- 230000003247 decreasing effect Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 238000012546 transfer Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000004064 recycling Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 8
- 238000004422 calculation algorithm Methods 0.000 abstract description 8
- 238000004519 manufacturing process Methods 0.000 abstract description 8
- 238000012545 processing Methods 0.000 abstract description 8
- 239000002699 waste material Substances 0.000 abstract description 6
- 230000006978 adaptation Effects 0.000 abstract 1
- OKKJLVBELUTLKV-UHFFFAOYSA-N Methanol Chemical compound OC OKKJLVBELUTLKV-UHFFFAOYSA-N 0.000 description 63
- 230000006870 function Effects 0.000 description 23
- NEHMKBQYUWJMIP-UHFFFAOYSA-N chloromethane Chemical compound ClC NEHMKBQYUWJMIP-UHFFFAOYSA-N 0.000 description 22
- 238000009826 distribution Methods 0.000 description 20
- 230000008859 change Effects 0.000 description 15
- 239000000203 mixture Substances 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 239000000047 product Substances 0.000 description 9
- NKDDWNXOKDWJAK-UHFFFAOYSA-N dimethoxymethane Chemical compound COCOC NKDDWNXOKDWJAK-UHFFFAOYSA-N 0.000 description 8
- 229940050176 methyl chloride Drugs 0.000 description 8
- 238000011176 pooling Methods 0.000 description 8
- 230000004913 activation Effects 0.000 description 7
- 239000000306 component Substances 0.000 description 6
- 239000007789 gas Substances 0.000 description 6
- 239000012071 phase Substances 0.000 description 6
- VEXZGXHMUGYJMC-UHFFFAOYSA-N Hydrochloric acid Chemical compound Cl VEXZGXHMUGYJMC-UHFFFAOYSA-N 0.000 description 5
- 239000003463 adsorbent Substances 0.000 description 5
- 239000007788 liquid Substances 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 5
- 238000010992 reflux Methods 0.000 description 5
- 238000000926 separation method Methods 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000005406 washing Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 3
- 238000003795 desorption Methods 0.000 description 3
- 238000011049 filling Methods 0.000 description 3
- 238000010438 heat treatment Methods 0.000 description 3
- IXCSERBJSXMMFS-UHFFFAOYSA-N hydrogen chloride Substances Cl.Cl IXCSERBJSXMMFS-UHFFFAOYSA-N 0.000 description 3
- 229910000041 hydrogen chloride Inorganic materials 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000012731 temporal analysis Methods 0.000 description 3
- 238000000700 time series analysis Methods 0.000 description 3
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- NEHMKBQYUWJMIP-NJFSPNSNSA-N chloro(114C)methane Chemical compound [14CH3]Cl NEHMKBQYUWJMIP-NJFSPNSNSA-N 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 230000002363 herbicidal effect Effects 0.000 description 2
- 239000004009 herbicide Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000007670 refining Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000012896 Statistical algorithm Methods 0.000 description 1
- 239000003513 alkali Substances 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000004090 dissolution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000000746 purification Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 238000001179 sorption measurement Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000012808 vapor phase Substances 0.000 description 1
- 238000003809 water extraction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Feedback Control In General (AREA)
Abstract
The utility model discloses a recovery method of glyphosate hydrolysis solvent, it is through the overhead operating temperature value of sensor group real-time collection, overhead operating pressure value, tower cauldron operating pressure value and tower cauldron operating temperature value, and introduce data processing and analysis algorithm in the rear end and carry out the chronogenesis collaborative analysis of multiparameter data, thereby establish the association with the inflow velocity time sequence variation of glyphosate hydrolysis solvent inflow desolventizing tower, thereby come the inflow velocity value of flow of real desolventizing tower's multiparameter time sequence collaborative variation to control the glyphosate hydrolysis solvent inflow desolventizing tower in real time adaptation, recovery efficiency and quality are improved, and reduce the waste of resource and energy, in this way, can realize the high-efficient recovery and the recycle of glyphosate hydrolysis solvent, thereby improve production efficiency, reduce cost, and reduce the influence to the environment.
Description
Technical Field
The present application relates to the field of intelligent recovery, and more particularly, to a method for recovering glyphosate hydrolysis solvent.
Background
Glyphosate is a commonly used herbicide and is widely used in the agricultural and horticultural fields. In the production process of glyphosate, the glyphosate hydrolysis solvent plays an important role. The glyphosate hydrolysis solvent is a solvent for reaction and separation, which reacts with raw materials in the synthesis reaction of glyphosate and plays a role in dissolution and separation in the subsequent separation step.
The glyphosate hydrolysis solvent is typically a complex mixture comprising the hydrolysis product of glyphosate and other reaction byproducts. In order to improve the product quality and the resource utilization efficiency, the glyphosate hydrolysis solvent needs to be recovered and reused. The recovery of the glyphosate hydrolysis solvent not only can reduce the production cost, but also can reduce the influence on the environment. Recovery of glyphosate hydrolysis solvent typically involves a separation and purification process wherein a desolventizing column is a common apparatus used to separate the vapor phase mixture produced in the hydrolysis kettle. In the desolventizing tower, separation and recovery of different components can be achieved by adjusting the operating temperature and operating pressure. Common recovery components include methyl chloride, hydrogen chloride, methanol, methylal and water.
However, conventional solutions are usually based on manual experience in recovering the glyphosate hydrolyzing solvent, and are susceptible to the skill level and subjective factors of the operator during the operation, resulting in poor stability of the recovery process, and instability of the operation may lead to fluctuations and inconsistencies in recovery. In addition, the traditional scheme generally adopts a single-parameter control method, and cannot fully consider the comprehensive influence of a plurality of key factors, so that the recovery efficiency is low. That is, a single parameter control method may not sufficiently optimize the recovery process, resulting in an inefficient recovery of a portion of the solvent.
In addition, in the recovery process of the glyphosate hydrolysis solvent, the flow rate of the glyphosate hydrolysis solvent flowing into the desolventizing tower is generally controlled fixedly, so that the method cannot adapt to the reaction rate change under different conditions, and the solvent cannot be recovered to the maximum extent, thereby reducing the recovery efficiency. Meanwhile, the hydrolysis reaction of the glyphosate hydrolysis solvent is a complex process and is affected by a plurality of factors. The fixed inflow velocity control cannot adapt to component recovery under different conditions, so that instability of the recovery process is caused, quality of the solvent is inconsistent, and the product quality requirement is difficult to meet.
Thus, an optimized recovery scheme for glyphosate hydrolysis solvent is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a recovery method of a glyphosate hydrolysis solvent, which is characterized in that a tower top operation temperature value, a tower top operation pressure value, a tower bottom operation pressure value and a tower bottom operation temperature value are acquired in real time through a sensor group, and a data processing and analyzing algorithm is introduced at the rear end to perform time sequence collaborative analysis of multi-parameter data, so that an association relation with time sequence change of inflow flow velocity of the glyphosate hydrolysis solvent into a desolventizing tower is established, the inflow flow velocity value of the glyphosate hydrolysis solvent into the desolventizing tower is adaptively controlled in real time based on the multi-parameter time sequence collaborative change of the actual desolventizing tower, recovery efficiency and quality are improved, waste of resources and energy sources is reduced, and therefore, efficient recovery and reuse of the glyphosate hydrolysis solvent can be realized, production efficiency is improved, cost is reduced, and influence on environment is reduced.
According to one aspect of the present application, there is provided a method of recovering glyphosate hydrolysis solvent comprising:
obtaining inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points in a preset time period;
acquiring tower top operation temperature values, tower top operation pressure values, tower bottom operation pressure values and tower bottom operation temperature values at a plurality of preset time points in the preset time period;
performing multi-parameter time sequence correlation analysis on the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at a plurality of preset time points to obtain a tower bottom tower top parameter characteristic diagram;
carrying out data fusion on the inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points and the tower top parameter characteristic diagram of the tower to obtain tower top parameter characteristics of fused flow velocity characteristics; and
and determining that the inflow flow rate value at the current time point should be increased, decreased or kept unchanged based on the tower top parameter characteristics of the tower bottom fused with the flow rate characteristics.
Compared with the prior art, the recovery method of the glyphosate hydrolysis solvent has the advantages that the overhead operation temperature value, the overhead operation pressure value, the tower kettle operation pressure value and the tower kettle operation temperature value are collected in real time through the sensor group, the data processing and analysis algorithm is introduced at the rear end to carry out the time sequence collaborative analysis of multi-parameter data, so that the association relation with the time sequence change of the inflow flow rate of the glyphosate hydrolysis solvent into the desolventizing tower is established, the inflow flow rate value of the glyphosate hydrolysis solvent into the desolventizing tower is adaptively controlled in real time based on the multi-parameter time sequence collaborative change of the actual desolventizing tower, the recovery efficiency and quality are improved, the waste of resources and energy sources is reduced, and the efficient recovery and reutilization of the glyphosate hydrolysis solvent can be realized, so that the production efficiency is improved, the cost is reduced, and the influence on the environment is reduced.
Drawings
The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a flow chart of a method of recovering glyphosate hydrolysis solvent according to an embodiment of the present application;
FIG. 2 is a system architecture diagram of a method for recovering glyphosate hydrolysis solvent according to an embodiment of the present application;
FIG. 3 is a flow chart of a training phase of a method for recovery of glyphosate hydrolysis solvent according to an embodiment of the present application;
fig. 4 is a flow chart of substep S3 of a method for recovering glyphosate hydrolysis solvent according to an embodiment of the present application;
fig. 5 is a flow chart of substep S4 of a method for recovering glyphosate hydrolysis solvent according to an embodiment of the present application;
fig. 6 is a flow chart of substep S42 of the method of recovering glyphosate hydrolysis solvent according to an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.
Flowcharts are used in this application to describe the operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Conventional solutions are usually based on manual experience in the recovery of glyphosate hydrolyzing solvents, and are susceptible to the skill level and subjective factors of the operator during the operation, resulting in poor stability of the recovery process, while instability of the operation may lead to fluctuations and inconsistencies in recovery results. In addition, the traditional scheme generally adopts a single-parameter control method, and cannot fully consider the comprehensive influence of a plurality of key factors, so that the recovery efficiency is low. That is, a single parameter control method may not sufficiently optimize the recovery process, resulting in an inefficient recovery of a portion of the solvent. In addition, in the recovery process of the glyphosate hydrolysis solvent, the flow rate of the glyphosate hydrolysis solvent flowing into the desolventizing tower is generally controlled fixedly, so that the method cannot adapt to the reaction rate change under different conditions, and the solvent cannot be recovered to the maximum extent, thereby reducing the recovery efficiency. Meanwhile, the hydrolysis reaction of the glyphosate hydrolysis solvent is a complex process and is affected by a plurality of factors. The fixed inflow velocity control cannot adapt to component recovery under different conditions, so that instability of the recovery process is caused, quality of the solvent is inconsistent, and the product quality requirement is difficult to meet. Thus, an optimized recovery scheme for glyphosate hydrolysis solvent is desired.
In the technical scheme of the application, a method for recycling the glyphosate hydrolysis solvent is provided. Fig. 1 is a flow chart of a method of recovering glyphosate hydrolysis solvent according to an embodiment of the present application. Fig. 2 is a system architecture diagram of a method for recovering glyphosate hydrolysis solvent according to an embodiment of the present application. As shown in fig. 1 and 2, a method for recovering a glyphosate hydrolyzing solvent according to an embodiment of the present application includes the steps of: s1, obtaining inflow flow velocity values of a plurality of preset time points of glyphosate hydrolysis solvent flowing into a desolventizing tower in a preset time period; s2, acquiring tower top operation temperature values, tower top operation pressure values, tower bottom operation pressure values and tower bottom operation temperature values at a plurality of preset time points in the preset time period; s3, performing multi-parameter time sequence correlation analysis on the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at a plurality of preset time points to obtain a tower bottom tower top parameter characteristic diagram; s4, carrying out data fusion on the inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points and the tower top parameter characteristic diagram of the tower bottom so as to obtain tower top parameter characteristics of fused flow velocity characteristics; and S5, determining that the inflow flow velocity value at the current time point should be increased, decreased or kept unchanged based on the tower top parameter characteristics of the tower bottom fused with the flow velocity characteristics.
In particular, in step S1, inflow flow rate values of the glyphosate hydrolyzing solvent into the desolventizing tower at a plurality of predetermined time points within a predetermined period of time are obtained. Among them, glyphosate is a widely used herbicide, and is commonly used in the agricultural and horticultural fields. The glyphosate hydrolysis solvent refers to a solvent used to dissolve or dilute glyphosate during the preparation or use of glyphosate. A desolventizing tower is a device for desorption or desorption of solutes in a gas or liquid. It is typically used in a regeneration step of the adsorption process to desorb the solute adsorbed on the adsorbent from the adsorbent for reuse of the adsorbent. The operating principle of the desolventizing tower is based on a desorption process, wherein solutes on the adsorbent are desorbed from the adsorbent surface by heating, depressurizing or other suitable operating conditions.
According to the embodiment of the application, the inflow flow rate value of the glyphosate hydrolysis solvent into the desolventizing tower at a plurality of preset time points in a preset time period can be obtained through the flow rate sensor. It is worth mentioning that the flow rate sensor is a device for measuring the fluid flow rate. Flow rate sensors are widely used in the fields of industrial automation, environmental monitoring, fluid control, laboratory research, and the like.
Specifically, in step S2, a top operation temperature value, a top operation pressure value, a bottom operation pressure value, and a bottom operation temperature value at a plurality of predetermined time points within the predetermined period of time are acquired. According to the embodiment of the application, the tower top operation pressure value and the tower bottom operation pressure value at a plurality of preset time points in the preset time period can be obtained through the pressure sensor; and acquiring tower top operation temperature values and tower bottom operation temperature values at a plurality of preset time points in the preset time period through a temperature sensor.
Notably, a pressure sensor is a device for measuring pressure. It can convert pressure into electrical signals for monitoring, control or data recording. The principle of operation of pressure sensors is generally based on the principle of force sensors. They use a sensing element (such as a strain gauge, capacitor or piezoresistive device) that deforms or generates an electrical signal when subjected to an external pressure. These signals, after amplification and processing, can be used to represent the magnitude of the pressure. Pressure sensors play an important role in many applications, such as industrial process control, liquid and gas measurement, automotive engine monitoring, medical devices, meteorological observations, and the like. A temperature sensor is a device for measuring temperature. It may convert the temperature into an electrical signal or other form of output signal for temperature monitoring, control or data recording. The choice of temperature sensor depends on the requirements of the application, including measurement range, accuracy, response time, environmental suitability, etc.
Specifically, in step S3, multi-parameter time-series correlation analysis is performed on the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at the plurality of predetermined time points to obtain a tower bottom tower top parameter characteristic diagram. In particular, in one specific example of the present application, the S3 includes: s31, arranging the tower top operation temperature values, the tower top operation pressure values, the tower bottom operation pressure values and the tower bottom operation temperature values at a plurality of preset time points into a tower bottom tower top parameter matrix according to a time dimension and a sample dimension; and S32, enabling the tower top parameter matrix to pass through a parameter time sequence correlation feature extractor based on a convolutional neural network model to obtain a tower top parameter feature map.
Specifically, the step S31 is to arrange the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at the plurality of predetermined time points into a tower bottom tower top parameter matrix according to a time dimension and a sample dimension. The real-time control precision of the inflow flow velocity value of the glyphosate hydrolysis solvent into the desolventizing tower is determined together, and the recovery efficiency and stability of different components are influenced by the consideration that the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value have time sequence dynamic change rules in the time dimension and the time sequence cooperative association relationship is also provided between the parameter data of the desolventizing towers. Therefore, in the technical scheme of the application, in order to better capture the multi-parameter time sequence related characteristics of the desolventizing tower, the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at a plurality of preset time points need to be further arranged into a tower top parameter matrix according to the time dimension and the sample dimension, so as to integrate the distribution information of the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value in the sample dimension and the time dimension.
Specifically, in the step S32, the tower top parameter matrix is passed through a parameter time sequence correlation feature extractor based on a convolutional neural network model to obtain a tower top parameter feature map. The tower top parameter matrix is subjected to feature mining in a parameter time sequence correlation feature extractor based on a convolutional neural network model so as to extract time sequence correlation feature distribution information in a time dimension among the tower top operation temperature value, the tower top operation pressure value and the tower top operation temperature value, thereby obtaining a tower top parameter feature map. More specifically, each layer using the parameter timing correlation feature extractor based on the convolutional neural network model performs the following steps on input data in forward transfer of the layer: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on the local feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the parameter time sequence correlation characteristic extractor based on the convolutional neural network model is the tower top parameter characteristic diagram of the tower, and the input of the first layer of the parameter time sequence correlation characteristic extractor based on the convolutional neural network model is the tower top parameter matrix of the tower.
Notably, convolutional neural networks (Convolutional Neural Network, CNN) are a deep learning model that is primarily used to process data with a grid structure, such as images and audio. CNNs are widely used in the field of computer vision, and have excellent performance for tasks such as image classification, object detection, image segmentation, and the like. The core idea of CNN is to extract and learn the features of an image through a hierarchy of convolutional, pooled, and fully connected layers. The following are the main components of CNN: convolution layer: the convolution layer is a core component of the CNN, and extracts local features of the image through convolution operation. The convolution operation slides over the image using a set of learnable filters (also called convolution kernels) and performs a weighted summation of the local pixels at each location to generate a feature map. The convolution layer can learn the features with different scales and abstraction levels; activation function: the activation function introduces nonlinear transformation, increasing the expressive power of the model. Common activation functions include ReLU, sigmoid, tanh, etc.; pooling layer: the pooling layer is used for reducing the space size of the feature map and reducing the calculated amount. Common pooling operations include maximum pooling and average pooling, which reduces the size of the feature map by taking the maximum or average value of the local area; full tie layer: the fully connected layer connects the outputs of the previous convolutional layer and pooling layer into a fully connected neural network. The full connection layer realizes the combination and classification of the features through learning weights and biasing; dropout: dropout is a regularization technique for reducing overfitting. During training, dropout randomly sets the output of a portion of neurons to zero, forcing the model to learn more robust and generalized features. The CNN gradually extracts and combines the features of the images through the stacking of a plurality of convolution layers, pooling layers and full connection layers, and finally, the classification or other tasks of the images are realized. In the training process, the CNN optimizes network parameters through a back propagation algorithm, so that the CNN can be better adapted to training data.
It should be noted that, in other specific examples of the present application, the multi-parameter time-series correlation analysis may be performed on the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at the plurality of predetermined time points in other manners to obtain a tower bottom tower top parameter feature map, for example: collecting historical data of a tower top operation temperature value, a tower top operation pressure value, a tower kettle operation pressure value and a tower kettle operation temperature value; preprocessing the collected data, including data cleaning, abnormal value removal, missing value filling and the like. Ensuring the quality and integrity of data; and selecting parameters which need to be subjected to association analysis according to specific requirements and analysis targets. In this case, the overhead operating temperature value, the overhead operating pressure value, the column bottom operating pressure value, and the column bottom operating temperature value are selected as parameters for the correlation analysis; using appropriate statistical methods and algorithms, timing correlation analysis is performed on the selected parameters. Common methods include correlation analysis, time series analysis, regression analysis, etc.; and drawing a tower top parameter characteristic diagram of the tower kettle according to the analysis result. The relationship and trend of change between parameters may be demonstrated using a line graph, a scatter graph, a thermodynamic diagram, or the like. The feature map can intuitively display the relevance and change rule among parameters; and (3) interpreting and analyzing the characteristic diagram to understand the relation and influence among the parameters. According to the result of the characteristic diagram, further optimization and control strategy formulation can be performed to improve the operation efficiency and the system performance.
Specifically, in step S4, data fusion is performed on the inflow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at the plurality of predetermined time points and the tower top parameter characteristic map to obtain tower top parameter characteristics of the fused velocity characteristics. In particular, in one specific example of the present application, as shown in fig. 5, the S4 includes: s41, arranging inflow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points into inflow velocity time sequence input vectors according to a time dimension; s42, the inflow speed time sequence input vector and the tower top parameter characteristic diagram are used for obtaining a tower top parameter characteristic diagram with the fusion flow velocity characteristic through a data fusion module, and the tower top parameter characteristic diagram is used as the tower top parameter characteristic with the fusion flow velocity characteristic.
Specifically, the step S41 is to arrange the inflow velocity values of the glyphosate hydrolyzing solvent flowing into the desolventizing tower at the predetermined time points into inflow velocity time series input vectors according to the time dimension. It is considered that the inflow flow rate value due to the inflow of the glyphosate hydrolyzing solvent into the desolventizing tower is constantly changing in the time dimension, that is, the inflow flow rate values at the plurality of predetermined time points have a time-series correlation therebetween. Therefore, it is necessary to integrate the time-series distribution information of the inflow flow rate value of the glyphosate hydrolysis solvent into the desolventizing tower in the time dimension by arranging the inflow flow rate values of the glyphosate hydrolysis solvent into the desolventizing tower in the time dimension as the inflow speed time-series input vector.
Specifically, the step S42 is to use the inflow speed time sequence input vector and the tower top parameter feature map through a data fusion module to obtain a tower top parameter feature map with a fused flow velocity feature as the tower top parameter feature of the fused flow velocity feature. After obtaining time sequence correlation characteristic distribution information of the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value in a time dimension and time sequence distribution information of inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower, in order to fuse the time sequence correlation characteristic among multiple parameters of the tower bottom and the inflow flow velocity time sequence characteristic to perform better classification processing, in the technical scheme of the application, the inflow velocity time sequence input vector and the tower bottom tower top parameter characteristic diagram are further subjected to a data fusion module (MetaNet) to obtain the tower bottom tower top parameter characteristic diagram fused with the flow velocity characteristic. It should be appreciated that the data fusion module can use one-dimensional convolution sequence extraction coefficients on the inflow velocity value data to assist in classification of multi-parameter time-series correlation features extracted from the tower top parameter matrix. That is, the data fusion module enables the inflow velocity value time sequence distribution metadata to directly interact with the multi-parameter time sequence related features, directly controls the related features of each feature channel, helps the network concentrate on a specific part of each feature channel, and improves the accuracy of real-time control of inflow velocity values. In particular, in one specific example of the present application, as shown in fig. 6, the S42 includes: s421, the inflow speed time sequence input vector is subjected to linear correction through a ReLU function after passing through a first convolution layer of the data fusion module, so as to obtain a linear corrected inflow speed time sequence input vector; s422, the linear corrected inflow speed time sequence input vector passes through a second convolution layer of the data fusion module and then is processed through a Sigmoid function to obtain an inflow speed time sequence vector; s423, carrying out position point multiplication weighting on the inflow speed time sequence vector and the tower top parameter characteristic diagram to obtain the tower top parameter characteristic diagram of the fusion flow velocity characteristic.
More specifically, in S421, the inflow speed time sequence input vector is linearly corrected by a ReLU function after passing through the first convolution layer of the data fusion module, so as to obtain a linearly corrected inflow speed time sequence input vector. It should be noted that the ReLU function is a commonly used activation function, and is characterized in that when the input value is greater than 0, the output value is equal to the input value; when the input value is 0 or less, the output value is 0. In short, the ReLU function truncates the negative part to 0, leaving the positive part.
More specifically, in S422, the linear corrected inflow velocity time sequence input vector is processed through a Sigmoid function after passing through the second convolution layer of the data fusion module, so as to obtain an inflow velocity time sequence vector. It is worth mentioning that Sigmoid function, also called Logistic function, is a commonly used nonlinear activation function. It maps the input value to an output value between 0 and 1.
More specifically, in S423, the inflow velocity timing vector and the tower top parameter feature map are weighted by position point multiplication to obtain a tower top parameter feature map with the fusion flow velocity feature. It should be appreciated that the effect of weighting the inflow velocity timing vector and the tower top parameter profile by location point multiplication is to assign different importance or weights to different elements or features. By weighting the inputs, elements with higher weights may be more focused or emphasized during the fusion or processing, thereby affecting the final result.
It should be noted that, in other specific examples of the present application, the inflow speed time sequence input vector and the tower top parameter feature map may be used as the tower top parameter feature of the fusion flow velocity feature by other ways through a data fusion module, where the tower top parameter feature map of the fusion flow velocity feature is obtained, for example: collecting time sequence data of inflow speed of the glyphosate hydrolysis solvent into the desolventizing tower and historical data of a tower kettle top parameter characteristic diagram; preprocessing the collected data, including data cleaning, abnormal value removal, missing value filling and the like; the time series data of the inflow speed is converted into a time series input vector. The inflow velocity values within a time window may be formed into a vector as input; time alignment is carried out on the time sequence input vector and the data of the tower top parameter feature map of the tower kettle; and inputting the aligned time sequence input vector and the tower top parameter characteristic diagram of the tower kettle into a data fusion module. The data fusion module can adopt different methods, such as splicing, weighting average and the like, so as to fuse the information of the two methods together; and extracting tower top parameter characteristics of the tower kettle from the fused data. Features may be extracted using statistical methods, time series analysis, machine learning, and the like. Common features include mean, variance, maximum, minimum, trend, etc.; drawing a tower kettle top parameter characteristic diagram fused with the flow velocity characteristic according to the extracted characteristic; and reading and analyzing the characteristic diagram to understand the relation and influence among the characteristics. According to the result of the characteristic diagram, further optimization and control strategy formulation can be performed to improve the operation efficiency and the system performance.
It should be noted that, in other specific examples of the present application, the inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at the plurality of predetermined time points and the tower top parameter feature map may be further subjected to data fusion in other manners to obtain tower top parameter features with fused flow velocity features, for example: and collecting historical data of inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower and tower kettle top parameter characteristic diagrams. Ensuring the correspondence and consistency of the time stamps of the data; preprocessing the collected data, including data cleaning, abnormal value removal, missing value filling and the like. Ensuring the quality and integrity of data; the inflow flow velocity value of the glyphosate hydrolysis solvent flowing into the desolventizing tower is aligned with the data of the tower kettle top parameter characteristic diagram in time, so that the time stamps of the inflow flow velocity value and the data of the glyphosate hydrolysis solvent are consistent; and (3) fusing the inflow flow velocity value of the time-aligned glyphosate hydrolysis solvent flowing into the desolventizing tower with the tower top parameter characteristic diagram of the tower kettle. Simple data stitching or more complex fusion algorithms such as weighted averaging, feature extraction, etc. may be used; and extracting tower top parameter characteristics of the tower kettle from the fused data. Features may be extracted using statistical methods, time series analysis, machine learning, and the like. Common features include mean, variance, maximum, minimum, trend, etc.; and drawing a tower kettle top parameter characteristic diagram according to the extracted characteristics.
In particular, in step S5, it is determined that the inflow flow rate value at the current time point should be increased, decreased or kept unchanged based on the tower bottom top parameter characteristics of the fusion flow rate characteristics. That is, the tower top parameter feature map fused with the flow velocity features is passed through a classifier to obtain a classification result indicating that the inflow flow velocity value at the current time point should be increased, decreased or kept unchanged. Therefore, the inflow flow rate value of the glyphosate hydrolysis solvent flowing into the desolventizing tower can be adaptively controlled in real time based on the multi-parameter time sequence cooperative change of the actual desolventizing tower, so that the recovery efficiency and quality are improved, and the waste of resources and energy sources is reduced. Specifically, expanding the tower top parameter feature map fused with the flow velocity features into classification feature vectors based on row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
A Classifier (Classifier) refers to a machine learning model or algorithm that is used to classify input data into different categories or labels. The classifier is part of supervised learning, which performs classification tasks by learning mappings from input data to output categories.
The fully connected layer (Fully Connected Layer) is one type of layer commonly found in neural networks. In the fully connected layer, each neuron is connected to all neurons of the upper layer, and each connection has a weight. This means that each neuron in the fully connected layer receives inputs from all neurons in the upper layer, and weights these inputs together, and then passes the result to the next layer.
The Softmax classification function is a commonly used activation function for multi-classification problems. It converts each element of the input vector into a probability value between 0 and 1, and the sum of these probability values equals 1. The Softmax function is commonly used at the output layer of a neural network, and is particularly suited for multi-classification problems, because it can map the network output into probability distributions for individual classes. During the training process, the output of the Softmax function may be used to calculate the loss function and update the network parameters through a back propagation algorithm. Notably, the output of the Softmax function does not change the relative magnitude relationship between elements, but rather normalizes them. Thus, the Softmax function does not change the characteristics of the input vector, but simply converts it into a probability distribution form.
It should be appreciated that the convolutional neural network model-based parameter timing-related feature extractor, the data fusion module, and the classifier need to be trained prior to the inference using the neural network model described above. That is, the method for recovering the glyphosate hydrolysis solvent further comprises a training stage for training the parameter time sequence correlation feature extractor, the data fusion module and the classifier based on the convolutional neural network model.
Fig. 3 is a flow chart of a training phase of a method for recovery of glyphosate hydrolysis solvent according to an embodiment of the present application. As shown in fig. 3, a method for recovering a glyphosate hydrolyzing solvent according to an embodiment of the present application includes: a training phase comprising: s110, training data are acquired, wherein the training data comprise training inflow flow velocity values of a plurality of preset time points of glyphosate hydrolysis solvent inflow desolventizing towers in a preset time period, and training tower top operation temperature values, training tower bottom operation pressure values and training tower bottom operation temperature values of a plurality of preset time points in the preset time period; s120, arranging the training tower top operation temperature value, the training tower top operation pressure value, the training tower bottom operation pressure value and the training tower bottom operation temperature value of the plurality of preset time points into a training tower bottom tower top parameter matrix according to a time dimension and a sample dimension; s130, passing the training tower kettle tower top parameter matrix through the parameter time sequence correlation feature extractor based on the convolutional neural network model to obtain a training tower kettle tower top parameter feature map; s140, arranging the training inflow velocity values of the glyphosate hydrolysis solvent inflow desolventizing tower at a plurality of preset time points into training inflow velocity time sequence input vectors according to a time dimension; s150, the training inflow speed time sequence input vector and the training tower kettle tower top parameter feature map pass through the data fusion module to obtain a tower kettle tower top parameter feature map with training fusion flow speed features; s160, the tower top parameter feature map of the tower kettle with the flow velocity features integrated is trained to pass through a classifier so as to obtain a classification loss function value; s170, training the parameter time sequence associated feature extractor, the data fusion module and the classifier based on the convolutional neural network model based on the classification loss function value and through gradient descent direction propagation, wherein the training fusion flow velocity feature tower top parameter feature vector obtained after the training fusion flow velocity feature tower top parameter feature map is unfolded is subjected to class matrix regularized weight space exploration constraint optimization when the training weight matrix iterates each time.
Particularly, in the technical scheme of the application, when the tower top parameter matrix is obtained through the parameter time sequence correlation feature extractor based on the convolutional neural network model, the tower top parameter feature graph may include a correlation feature representation of a local correlation scale based on the convolutional neural network model of a tower top operation temperature value, a tower top operation pressure value and a tower top operation temperature value in a time-sample cross dimension, so that after the tower top parameter matrix is further passed through the data fusion module together with the inflow velocity time sequence input vector, the feature distribution representation of the tower top parameter feature graph in a probability distribution domain of a classifier can be further strengthened based on the time sequence distribution of an inflow velocity value, but the feature distribution representation of the local correlation of the tower top parameter feature graph of the fusion velocity feature in a parameter time sequence-sample cross local correlation context is further increased, and the tag of the tower top parameter feature graph with the fusion velocity feature in the diversified feature distribution representation in the probability distribution domain of the classification result is further increased, so that the tag in the probability distribution of the classification result is enriched in the probability distribution domain of the classification result, and the effect of the classifier in the classifier is further increased. Based on this, when classifying the tower top parameter feature map fused with the flow velocity feature by the classifier, the applicant of the present application performs a weight space exploration constraint based on regularization of a class matrix on the tower top parameter feature vector fused with the flow velocity feature obtained after expanding the tower top parameter feature map fused with the flow velocity feature during each iteration of the weight matrix, specifically expressed as:
Wherein V is a tower bottom tower top parameter characteristic vector of the training fusion flow velocity characteristic obtained after the tower bottom tower top parameter characteristic map of the training fusion flow velocity characteristic is unfolded, and V ′ Is a tower kettle top parameter feature vector of the optimized training fusion flow velocity feature obtained after the tower kettle top parameter feature map of the optimized training fusion flow velocity feature is unfolded, and V is a column vector, V ′ Is the vector of the row and,as a learnable domain transfer matrix, M represents the weight matrix of the last iteration, M ′ Representing the weight matrix after iteration, +.>Representing a matrix multiplication. 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 tower top parameter feature vector V of the fusion flow velocity feature, the regularized representation of the class matrix of the weight matrix M relative to the tower top parameter feature vector V of the fusion flow velocity feature 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 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 tower top parameter feature map of the fusion flow velocity feature in classification by a classifier is also improved. Therefore, the inflow flow velocity value of the glyphosate hydrolysis solvent flowing into the desolventizing tower can be adaptively controlled in real time based on the multi-parameter time sequence cooperative change condition of the actual desolventizing tower, so that the recovery efficiency and quality are improved, the waste of resources and energy sources is reduced, the efficient recovery and reuse of the glyphosate hydrolysis solvent are realized, the production efficiency is improved, the cost is reduced, and the influence on the environment is reduced.
In summary, the recovery method of the glyphosate hydrolysis solvent according to the embodiment of the application is clarified, the top operation temperature value, the top operation pressure value, the bottom operation pressure value and the bottom operation temperature value are collected in real time through the sensor group, and the time sequence collaborative analysis of multi-parameter data is carried out by introducing a data processing and analyzing algorithm at the rear end, so that the association relation with the time sequence change of the inflow flow rate of the glyphosate hydrolysis solvent into the desolventizing tower is established, the inflow flow rate value of the glyphosate hydrolysis solvent into the desolventizing tower is adaptively controlled in real time based on the time sequence collaborative change of the multi-parameter of the actual desolventizing tower, the recovery efficiency and quality are improved, the waste of resources and energy sources is reduced, and therefore, the efficient recovery and reutilization of the glyphosate hydrolysis solvent can be realized, the production efficiency is improved, the cost is reduced, and the influence on the environment is reduced.
In one example, the hydrolysis solution from the hydrolysis kettle is continuously fed into a desolventizing tower T1 at a temperature of 30-40 ℃, the desolventizing tower is a stripping tower, the tower top operation pressure is 125KPa (A), the tower top operation temperature is 87.3 ℃, the tower bottom operation pressure is 127KPa (A), the tower top operation temperature is 108 ℃, the tower top is provided with a gas phase mixture of methyl chloride, hydrogen chloride, methanol, methylal and water, the gas phase mixture is not condensed and directly fed into the lower part of an alkaline washing tower T2, and the tower bottom liquid of the desolventizing tower T1 is fed into a glyphosate refining section.
The alkaline washing tower T2 is operated at normal pressure, the operating pressure at the top of the tower is 115KPa (A), the operating temperature at the top of the tower is 78 ℃, the operating pressure at the bottom of the tower is 117KPa (A), the operating temperature at the bottom of the tower is 80 ℃, fresh alkali liquor enters from the top of the tower, hydrogen chloride in the mixed gas rising from the bottom of the tower reacts to obtain sodium chloride, a part of methanol also flows to the bottom of the tower in a condensing way,
part of tower bottom liquid of the alkaline washing tower T2 is circulated to the top of the tower of the T2, and the other part of the tower bottom liquid enters the middle part of the tower of the methyl chloride heavy-duty removing tower T3, and gas phase at the top of the alkaline washing tower T2 also enters the middle part of the tower of the methyl chloride heavy-duty removing tower T3.
The chloromethane heavy-removal tower T3 is operated at normal pressure, the tower top operation pressure is 155KPa (A), the tower top operation temperature is 62 ℃, the tower bottom operation pressure is 106KPa (A), and the tower bottom operation temperature is 72 ℃. Methyl chloride and methylal are extracted from the tower top, part of methanol enters a methyl chloride product tower T4, the reflux ratio R-2-8 at the tower top, and the mixture of the methanol and the water is extracted from the tower bottom and enters a low-pressure methanol tower T6.
The chloromethane product tower T4 is operated at normal pressure, the tower top operation pressure is 150KPa (A), the tower top operation temperature is-15 ℃, the tower bottom operation pressure is 151KPa (A), and the tower bottom operation temperature is 54 ℃. Methyl chloride extracted from the tower top enters a methyl chloride refining system, the reflux ratio R-1-5 at the tower top, methylal extracted from the tower bottom and the mixture of methanol and water enter a methylal tower T5.
The methylal product column T5 is operated at normal pressure, the column top operation pressure is 110KPa (A), the column top operation temperature is 45 degrees, the column bottom operation pressure is 151KPa (A), and the column bottom operation temperature is 67 degrees. And (3) extracting methylal azeotrope from the tower top, extracting qualified methanol from the tower bottom, wherein the reflux ratio of the tower top is R-2-8.
Methanol from the tower kettle of the chloromethane heavy removal tower T3 enters the middle part of the low-pressure methanol tower T6, the tower top operation pressure of the low-pressure methanol tower T6 is 250KPa (A), the tower top operation temperature is 90 degrees, the tower kettle operation pressure is 251KPa (A), and the tower kettle operation temperature is 100 degrees. Qualified methanol is extracted from the tower top, the reflux ratio R-2-8 of the tower top, the methanol steam of the low-pressure methanol tower is used as heating steam of a tower kettle reboiler of the methyl chloride heavy removal tower T3, the methyl chloride product tower T4 and the methylal product tower T5, and the methanol and water mixture extracted from the tower kettle enters the middle part of the high-pressure methanol tower T7.
The high-pressure methanol tower T7 has a tower top operation pressure of 550KPa (A), a tower top operation temperature of 115 degrees, a tower bottom operation pressure of 551KPa (A) and a tower bottom operation temperature of 151 degrees. Qualified methanol is extracted from the tower top, the reflux ratio R-2-8 of the tower top, and the methanol steam of the high-pressure methanol tower is used as heating steam of a tower kettle reboiler of the low-pressure methanol tower T6, and hydrochloric acid is prepared in a water extraction area of the tower kettle.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. A method for recovering a glyphosate hydrolysis solvent, comprising:
obtaining inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points in a preset time period;
acquiring tower top operation temperature values, tower top operation pressure values, tower bottom operation pressure values and tower bottom operation temperature values at a plurality of preset time points in the preset time period;
performing multi-parameter time sequence correlation analysis on the tower top operation temperature value, the tower top operation pressure value, the tower bottom operation pressure value and the tower bottom operation temperature value at a plurality of preset time points to obtain a tower bottom tower top parameter characteristic diagram;
carrying out data fusion on the inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points and the tower top parameter characteristic diagram of the tower to obtain tower top parameter characteristics of fused flow velocity characteristics; and
and determining that the inflow flow rate value at the current time point should be increased, decreased or kept unchanged based on the tower top parameter characteristics of the tower bottom fused with the flow rate characteristics.
2. The method for recovering a glyphosate hydrolyzing solvent according to claim 1, wherein the multi-parameter time-series correlation analysis is performed on the top operation temperature value, the top operation pressure value, the bottom operation pressure value and the bottom operation temperature value at the plurality of predetermined time points to obtain a bottom top parameter profile, comprising:
Arranging the tower top operation temperature values, the tower top operation pressure values, the tower bottom operation pressure values and the tower bottom operation temperature values at a plurality of preset time points into a tower bottom tower top parameter matrix according to the time dimension and the sample dimension; and
and (3) enabling the tower top parameter matrix of the tower kettle to pass through a parameter time sequence correlation feature extractor based on a convolutional neural network model so as to obtain a tower top parameter feature map.
3. The method for recovering a glyphosate hydrolyzing solvent according to claim 2, wherein the data fusion of the inflow flow rate value of the glyphosate hydrolyzing solvent into the desolventizing tower at the plurality of predetermined time points and the tower top parameter characteristic map is performed to obtain a tower top parameter characteristic of the fused flow rate characteristic, comprising:
arranging inflow flow velocity values of the glyphosate hydrolysis solvent flowing into the desolventizing tower at a plurality of preset time points into inflow velocity time sequence input vectors according to a time dimension;
and the inflow speed time sequence input vector and the tower top parameter characteristic diagram are used for obtaining a tower top parameter characteristic diagram of the fusion flow velocity characteristic through a data fusion module, and the tower top parameter characteristic diagram is used as the tower top parameter characteristic of the fusion flow velocity characteristic.
4. A method of recovering a glyphosate hydrolyzing solvent according to claim 3, wherein passing the inflow velocity time-series input vector and the tower top parameter profile through a data fusion module to obtain a tower top parameter profile of a fusion flow rate characteristic as a tower top parameter characteristic of the fusion flow rate characteristic comprises:
The inflow speed time sequence input vector is subjected to linear correction through a ReLU function after passing through a first convolution layer of the data fusion module, so that the inflow speed time sequence input vector after linear correction is obtained;
the linear corrected inflow speed time sequence input vector passes through a second convolution layer of the data fusion module and then is processed through a Sigmoid function so as to obtain an inflow speed time sequence vector;
and weighting the inflow speed time sequence vector and the tower top parameter characteristic diagram according to position point multiplication to obtain the tower top parameter characteristic diagram of the fusion flow velocity characteristic.
5. The method for recovering a glyphosate hydrolyzing solvent according to claim 4, wherein determining that the inflow flow rate value at the present time point should be increased, decreased or maintained based on the tower top parameter characteristic of the fusion flow rate characteristic comprises: and the tower top parameter feature map fused with the flow velocity features is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating that the inflow flow velocity value at the current time point should be increased, decreased or kept unchanged.
6. The method for recovering glyphosate hydrolyzing solvent as recited in claim 5, further comprising a training step of: and the parameter time sequence correlation feature extractor, the data fusion module and the classifier are used for training the parameter time sequence correlation feature extractor, the data fusion module and the classifier based on the convolutional neural network model.
7. The method for recovering glyphosate hydrolyzing solvent as recited in claim 6, wherein said training step comprises:
obtaining training data, wherein the training data comprises training inflow flow velocity values of a plurality of preset time points of glyphosate hydrolysis solvent inflow desolventizing towers in a preset time period, and training tower top operation temperature values, training tower bottom operation pressure values and training tower bottom operation temperature values of a plurality of preset time points in the preset time period;
arranging the training tower top operation temperature values, the training tower top operation pressure values, the training tower bottom operation pressure values and the training tower bottom operation temperature values of the plurality of preset time points into a training tower bottom tower top parameter matrix according to the time dimension and the sample dimension;
the tower top parameter matrix of the training tower kettle passes through the parameter time sequence correlation characteristic extractor based on the convolutional neural network model to obtain a tower top parameter characteristic diagram of the training tower kettle;
arranging training inflow velocity values of the glyphosate hydrolysis solvent inflow desolventizing tower at a plurality of preset time points into training inflow velocity time sequence input vectors according to a time dimension;
the training inflow speed time sequence input vector and the training tower kettle tower top parameter feature map pass through the data fusion module to obtain a tower kettle tower top parameter feature map with training fusion flow speed features; and
The tower top parameter feature map of the tower kettle with the flow velocity features integrated is trained through a classifier to obtain a classification loss function value;
training the parameter time sequence associated feature extractor, the data fusion module and the classifier based on the convolutional neural network model based on the classification loss function value and through gradient descent direction propagation, wherein the training fusion flow velocity feature tower top parameter feature vector obtained after training fusion flow velocity feature tower top parameter feature map expansion is subjected to class matrix regularization-based weight space exploration constraint optimization when the training weight matrix iterates each time.
8. The method for recycling glyphosate hydrolysis solvent according to claim 7, wherein, when the weight matrix of the training is iterated each time, the tower top parameter feature vector of the training fusion flow rate feature obtained after the tower top parameter feature map of the training fusion flow rate feature is expanded is subjected to weight space exploration constraint optimization based on class matrix regularization by using the following optimization formula to obtain a tower top parameter feature map of the optimized training fusion flow rate feature;
wherein, the optimization formula is:
Wherein V is a tower bottom tower top parameter characteristic vector of the training fusion flow velocity characteristic obtained after the tower bottom tower top parameter characteristic map of the training fusion flow velocity characteristic is unfolded, and V ′ Is a tower kettle top parameter feature vector of the optimized training fusion flow velocity feature obtained after the tower kettle top parameter feature map of the optimized training fusion flow velocity feature is unfolded, and V is a column vector, V ′ Is the vector of the row and,as a learnable domain transfer matrix, M represents the weight matrix of the last iteration, M ′ Representing the weight matrix after iteration, +.>Representing a matrix multiplication.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311415906.5A CN117455451A (en) | 2023-10-30 | 2023-10-30 | Recovery method of glyphosate hydrolysis solvent |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311415906.5A CN117455451A (en) | 2023-10-30 | 2023-10-30 | Recovery method of glyphosate hydrolysis solvent |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117455451A true CN117455451A (en) | 2024-01-26 |
Family
ID=89594377
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311415906.5A Pending CN117455451A (en) | 2023-10-30 | 2023-10-30 | Recovery method of glyphosate hydrolysis solvent |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117455451A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117964029A (en) * | 2024-03-28 | 2024-05-03 | 内蒙古莱科作物保护有限公司 | Method for preparing p-chlorophenylglycine based on waste liquid generated in production of chlorfenapyr |
-
2023
- 2023-10-30 CN CN202311415906.5A patent/CN117455451A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117964029A (en) * | 2024-03-28 | 2024-05-03 | 内蒙古莱科作物保护有限公司 | Method for preparing p-chlorophenylglycine based on waste liquid generated in production of chlorfenapyr |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110231156B (en) | Service robot motion system fault diagnosis method and device based on time sequence characteristics | |
| Chen et al. | One-dimensional convolutional neural network-based active feature extraction for fault detection and diagnosis of industrial processes and its understanding via visualization | |
| CN117455451A (en) | Recovery method of glyphosate hydrolysis solvent | |
| CN109218134B (en) | Test case generation system based on neural style migration | |
| CN111292195A (en) | Risk account identification method and device | |
| CN108509833A (en) | A kind of face identification method, device and equipment based on structured analysis dictionary | |
| CN110298248A (en) | A kind of multi-object tracking method and system based on semantic segmentation | |
| CN112529638B (en) | Service demand dynamic prediction method and system based on user classification and deep learning | |
| CN115983126A (en) | Intelligent preparation method of iron isomaltose anhydride 1000 | |
| WO2024021254A1 (en) | Intelligent separation and purification system for electronic-grade chlorine trifluoride | |
| CN113887675A (en) | Gesture recognition method based on feature fusion of heterogeneous sensors | |
| CN117308077A (en) | Water supply control system for evaporator of nuclear power unit of reactor | |
| CN116740384A (en) | Intelligent control method and system of floor washing machine | |
| CN111898565A (en) | Forest smoke and fire real-time monitoring system and method based on robust multi-view | |
| CN116258990A (en) | Cross-modal affinity-based small sample reference video target segmentation method | |
| Zhu et al. | Multi-scale Transformer-CNN domain adaptation network for complex processes fault diagnosis | |
| Zhang et al. | Energy-based generative cooperative saliency prediction | |
| CN117530684B (en) | Blood glucose abnormality detection and early warning system and method based on health big data | |
| CN114882529A (en) | Double-branch cross-domain pedestrian re-identification method based on attention calibration | |
| CN117225091A (en) | System and method for monitoring running state of bag-type dust collector | |
| CN116739219A (en) | Melt blown cloth production management system and method thereof | |
| Hu et al. | Siamese network object tracking algorithm combining attention mechanism and correlation filter theory | |
| CN116205927A (en) | Image segmentation method based on boundary enhancement | |
| Yao et al. | Hierarchical structure-feature aware graph neural network for node classification | |
| CN116835540B (en) | Preparation method of phosphorus pentafluoride |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |