CN116612836B - Tail gas amount prediction method and system for trifluoromethane production - Google Patents
Tail gas amount prediction method and system for trifluoromethane production Download PDFInfo
- Publication number
- CN116612836B CN116612836B CN202310882319.0A CN202310882319A CN116612836B CN 116612836 B CN116612836 B CN 116612836B CN 202310882319 A CN202310882319 A CN 202310882319A CN 116612836 B CN116612836 B CN 116612836B
- Authority
- CN
- China
- Prior art keywords
- amount
- trifluoromethane
- tail gas
- predicted
- chlorodifluoromethane
- 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.)
- Active
Links
- XPDWGBQVDMORPB-UHFFFAOYSA-N Fluoroform Chemical compound FC(F)F XPDWGBQVDMORPB-UHFFFAOYSA-N 0.000 title claims abstract description 257
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 101
- 238000000034 method Methods 0.000 title claims abstract description 39
- KRHYYFGTRYWZRS-UHFFFAOYSA-N Fluorane Chemical compound F KRHYYFGTRYWZRS-UHFFFAOYSA-N 0.000 claims abstract description 173
- VOPWNXZWBYDODV-UHFFFAOYSA-N Chlorodifluoromethane Chemical compound FC(F)Cl VOPWNXZWBYDODV-UHFFFAOYSA-N 0.000 claims abstract description 69
- 238000006243 chemical reaction Methods 0.000 claims abstract description 31
- 239000007789 gas Substances 0.000 claims description 160
- VEXZGXHMUGYJMC-UHFFFAOYSA-N Hydrochloric acid Chemical compound Cl VEXZGXHMUGYJMC-UHFFFAOYSA-N 0.000 claims description 99
- 229910000040 hydrogen fluoride Inorganic materials 0.000 claims description 41
- HEMHJVSKTPXQMS-UHFFFAOYSA-M Sodium hydroxide Chemical compound [OH-].[Na+] HEMHJVSKTPXQMS-UHFFFAOYSA-M 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 15
- 238000003062 neural network model Methods 0.000 claims description 13
- 230000009471 action Effects 0.000 claims description 8
- 239000003054 catalyst Substances 0.000 claims description 8
- 239000003344 environmental pollutant Substances 0.000 claims description 3
- 231100000719 pollutant Toxicity 0.000 claims description 3
- 238000005259 measurement Methods 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- 238000005303 weighing Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 239000012295 chemical reaction liquid Substances 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000002253 acid Substances 0.000 description 2
- 239000000356 contaminant Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 238000002329 infrared spectrum Methods 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 239000002912 waste gas Substances 0.000 description 2
- 238000000862 absorption spectrum Methods 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 239000003513 alkali Substances 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- BMXRGTLJFLAMCQ-UHFFFAOYSA-N chloro(difluoro)methane;fluoroform Chemical compound FC(F)F.FC(F)Cl BMXRGTLJFLAMCQ-UHFFFAOYSA-N 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000003317 industrial substance Substances 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003334 potential effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
Abstract
The invention provides a tail gas quantity prediction method and a tail gas quantity prediction system for producing trifluoromethane, and relates to the technical field of production of trifluoromethane. Acquiring data of a plurality of sensors in a continuous time period in a production process of the trifluoromethane; determining a first predicted tail gas amount using a first prediction model based on a video of a back-off of chlorodifluoromethane and hydrofluoric acid at the time of production of trifluoromethane and a plurality of sensor data for successive time periods during the production of trifluoromethane; obtaining a reaction video of a continuous time period in the production process of the trifluoromethane; determining a second predicted tail gas amount based on the reaction videos of successive time periods in the production process of the trifluoromethane using a second prediction model; the target predicted tail gas amount is determined based on the first predicted tail gas amount and the second predicted tail gas amount, and the method can accurately predict the tail gas amount of the trifluoromethane.
Description
Technical Field
The invention relates to the technical field of trifluoromethane production, in particular to a tail gas quantity prediction method and a tail gas quantity prediction system for trifluoromethane production.
Background
Trifluoromethane (also known as Freon-23 or R-23) is an important industrial chemical and is widely used in the fields of air conditioning, refrigeration, fire extinguishing agents and the like. However, in the production of trifluoromethane, a large amount of tail gas is produced. These tail gases are typically harmful substances such as hydrogen fluoride, hydrochloric acid, etc., which have potential effects on the environment and human health.
Therefore, the method for accurately predicting the tail gas amount in the production process of the trifluoromethane has important significance in the aspects of process optimization, environmental protection, capacity planning and the like. Traditionally engineers and researchers have relied primarily on experience and test data to estimate the amount of tail gas. However, this method has some limitations, such as time consuming, laborious, and sometimes inaccurate, and cannot fully take into account complex relationships of various factors.
How to accurately predict the tail gas amount of the trifluoromethane is a problem to be solved currently.
Disclosure of Invention
The invention mainly solves the technical problem of accurately predicting the tail gas amount of the trifluoromethane.
According to a first aspect, the present invention provides a method for predicting the amount of tail gas from the production of trifluoromethane, comprising: obtaining a dumping video of chlorodifluoromethane and hydrofluoric acid, a dumping amount of chlorodifluoromethane and a dumping amount of hydrofluoric acid during production of the trifluoromethane; acquiring data of a plurality of sensors in a continuous time period in the production process of the trifluoromethane, wherein the production process of the trifluoromethane is to react chlorodifluoromethane with hydrofluoric acid under the action of a catalyst to generate the trifluoromethane and hydrochloric acid, and the plurality of sensors comprise a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph and an infrared spectrometer; determining a first predicted tail gas amount based on a back-off video of chlorodifluoromethane and hydrofluoric acid at the time of production of the trifluoromethane and a plurality of sensor data for successive time periods in the production process of the trifluoromethane using a first prediction model, the first predicted tail gas amount comprising a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid; obtaining a reaction video of a continuous time period in the production process of the trifluoromethane; determining a second predicted tail gas amount based on the reaction videos of successive time periods in the production process of the trifluoromethane using a second prediction model, the second predicted tail gas amount including a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid; a target predicted tail gas amount is determined based on the first predicted tail gas amount and the second predicted tail gas amount, the target predicted tail gas amount including a target predicted amount of hydrogen fluoride and a target predicted amount of hydrochloric acid.
Still further, the determining a predicted tail gas amount based on the first predicted tail gas amount and the second predicted tail gas amount includes: and giving different weights to the first predicted amount of the hydrogen fluoride and the second predicted amount of the hydrogen fluoride, carrying out weighted summation to obtain the target predicted amount of the hydrogen fluoride, giving different weights to the first predicted amount of the hydrochloric acid and the second predicted amount of the hydrochloric acid, and carrying out weighted summation to obtain the target predicted amount of the hydrochloric acid.
Still further, the first prediction model is a long-short-period neural network model, the input of the first prediction model is a feed-back video of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane and a plurality of sensor data of continuous time periods in the production process of the trifluoromethane, and the output of the first prediction model is a first prediction tail gas amount.
Further, the second prediction model is a long-short-period neural network model, the input of the second prediction model is a reaction video of a continuous time period in the production process of the trifluoromethane, and the output of the second prediction model is a second prediction tail gas amount.
Still further, the method further comprises: and determining the addition amount of sodium hydroxide based on the target predicted tail gas amount.
According to a second aspect, the present invention provides an exhaust gas amount prediction system for trifluoromethane production, comprising: the first acquisition module is used for acquiring the dumping video of chlorodifluoromethane and hydrofluoric acid, the dumping amount of chlorodifluoromethane and the dumping amount of hydrofluoric acid during the production of the trifluoromethane;
the second acquisition module is used for acquiring data of a plurality of sensors in a continuous time period in the production process of the trifluoromethane, wherein the production process of the trifluoromethane is that chlorodifluoromethane reacts with hydrofluoric acid under the action of a catalyst to generate the trifluoromethane and hydrochloric acid, and the plurality of sensors comprise a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph and an infrared spectrometer;
a first predicted tail gas amount determination module for determining a first predicted tail gas amount using a first prediction model based on a video of a dumping of chlorodifluoromethane and hydrofluoric acid at the time of production of the trifluoromethane and a plurality of sensor data for successive time periods in the production process of the trifluoromethane, the first predicted tail gas amount including a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid;
the third acquisition module is used for acquiring a reaction video of a continuous time period in the production process of the trifluoromethane;
a second predicted tail gas amount determination module for determining a second predicted tail gas amount using a second prediction model based on a reaction video of a continuous time period in the production process of the trifluoromethane, the second predicted tail gas amount including a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid;
and a target predicted amount determination module configured to determine a target predicted amount of tail gas based on the first predicted amount of tail gas and the second predicted amount of tail gas, the target predicted amount of tail gas including a target predicted amount of hydrogen fluoride and a target predicted amount of hydrochloric acid.
Still further, the target pre-measurement determination module is further configured to: and giving different weights to the first predicted amount of the hydrogen fluoride and the second predicted amount of the hydrogen fluoride, carrying out weighted summation to obtain the target predicted amount of the hydrogen fluoride, giving different weights to the first predicted amount of the hydrochloric acid and the second predicted amount of the hydrochloric acid, and carrying out weighted summation to obtain the target predicted amount of the hydrochloric acid.
Still further, the first prediction model is a long-short-period neural network model, the input of the first prediction model is a feed-back video of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane and a plurality of sensor data of continuous time periods in the production process of the trifluoromethane, and the output of the first prediction model is a first prediction tail gas amount.
Further, the second prediction model is a long-short-period neural network model, the input of the second prediction model is a reaction video of a continuous time period in the production process of the trifluoromethane, and the output of the second prediction model is a second prediction tail gas amount.
Still further, the system is further configured to: and determining the addition amount of sodium hydroxide based on the target predicted tail gas amount.
The invention provides a tail gas amount prediction method and a tail gas amount prediction system for producing trifluoro methane, wherein the method comprises the steps of obtaining a dumping video of chlorodifluoro methane and hydrofluoric acid, a dumping amount of chlorodifluoro methane and a dumping amount of hydrofluoric acid during producing trifluoro methane; acquiring data of a plurality of sensors in a continuous time period in the production process of the trifluoromethane, wherein the production process of the trifluoromethane is to react chlorodifluoromethane with hydrofluoric acid under the action of a catalyst to generate the trifluoromethane and hydrochloric acid, and the plurality of sensors comprise a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph and an infrared spectrometer; determining a first predicted tail gas amount based on a back-off video of chlorodifluoromethane and hydrofluoric acid at the time of production of the trifluoromethane and a plurality of sensor data for successive time periods in the production process of the trifluoromethane using a first prediction model, the first predicted tail gas amount comprising a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid; obtaining a reaction video of a continuous time period in the production process of the trifluoromethane; determining a second predicted tail gas amount based on the reaction videos of successive time periods in the production process of the trifluoromethane using a second prediction model, the second predicted tail gas amount including a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid; the method can accurately predict the tail gas amount of the trifluoromethane by determining the target predicted tail gas amount based on the first predicted tail gas amount and the second predicted tail gas amount, wherein the target predicted tail gas amount comprises the target predicted amount of the hydrogen fluoride and the target predicted amount of the hydrochloric acid.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting the tail gas amount in the production of trifluoromethane according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an exhaust gas amount prediction system for trifluoromethane production according to an embodiment of the present invention.
Detailed Description
In an embodiment of the present invention, there is provided a method for predicting the amount of tail gas produced by trifluoromethane as shown in fig. 1, where the method for predicting the amount of tail gas produced by trifluoromethane includes steps S1 to S6:
and step S1, obtaining a dumping video of chlorodifluoromethane and hydrofluoric acid, a dumping amount of chlorodifluoromethane and a dumping amount of hydrofluoric acid during production of the trifluoromethane.
Chlorodifluoromethane has the chemical formula CHClF2 and is a colorless gas, also known as Freon 22. In the production of trifluoromethane chlorodifluoromethane is used as one of the raw materials. For example, chlorodifluoromethane has a purity of 99% or more, depending on the specifications provided by the supplier.
Hydrofluoric acid has the chemical formula of HF, is a strong acid and is also one of important raw materials in the production process of trifluoromethane. For example, hydrofluoric acid purchased from suppliers typically has a purity of 40% or more.
Pouring amount of chlorodifluoromethane: refers to the total amount of chlorodifluoromethane which is poured into the reactor over a period of time. Meter devices or weighing apparatus are commonly used for accurate measurement and registration. For example, the total mass of chlorodifluoromethane poured into the reactor in one hour is 1000 kg.
Pouring amount of hydrofluoric acid: refers to the total amount of hydrofluoric acid that is poured into the reactor over a period of time. Also, instrumentation or weighing devices are often used to accurately measure and record. For example, the total mass of hydrofluoric acid poured into the reactor in one hour is 500 kg.
The video of the pouring of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane is used for recording the video of the pouring of the chlorodifluoromethane and the hydrofluoric acid into a reactor during the production process. This can be achieved by an image capturing apparatus. For example, a camera is provided to capture the pouring process and record it in video format. The video of the dumping of chlorodifluoromethane and hydrofluoric acid during the production of trifluoromethane can display a variety of information that can be used to predict the production of trifluoromethane. For example, the pouring video of chlorodifluoromethane and hydrofluoric acid can show the pouring amount of chlorodifluoromethane and hydrofluoric acid, can show the pouring frequency of chlorodifluoromethane and hydrofluoric acid, can show the concrete time of pouring chlorodifluoromethane and hydrofluoric acid into the reactor, can also show the pouring operation flow of staff, the pouring operation flow includes opening and closing the valve, operating time, the pouring video provides visual information, can provide key data for analyzing the pouring behavior of chlorodifluoromethane and hydrofluoric acid in the production process of trifluoromethane, building the predictive model and monitoring the production process.
The implementation method of the step S1 comprises the following steps: a. configuring an image pickup apparatus: in the process of producing trifluoromethane, an appropriate image pickup apparatus such as a high-definition camera or a monitoring camera is configured. The proper position of the image pickup device is ensured, and the details of the pouring operation can be comprehensively recorded. b. And (3) pouring: according to the process requirements and operating rules, chlorodifluoromethane and hydrofluoric acid are slowly poured into the reactor. The operator needs to carefully operate to ensure that the material pouring process is performed stably. c. Recording a material pouring video: the configured camera equipment is set to record the dumping process of chlorodifluoromethane and hydrofluoric acid in real time. The video should clearly record the whole process of pouring, including the speed, mode and form of the pouring liquid. d. The amount of chlorodifluoromethane to be poured was recorded: in the process of pouring, meter equipment (such as a flowmeter) or a weighing device is used for accurately measuring and recording the pouring amount of chlorodifluoromethane. For example, a weighing device may be used to record the difference in mass of the container before and after pouring to determine the amount of chlorodifluoromethane to be poured. e. Recording the pouring amount of hydrofluoric acid: likewise, during the pouring process, the amount of the poured hydrofluoric acid is accurately measured and recorded using an instrument device or a weighing apparatus. For example, a flow meter may be used to record the flow of hydrofluoric acid and calculate the total throw-away amount. f. Sorting and archiving data: and (5) sorting and archiving the recorded material pouring video, and ensuring the integrity and accessibility of the data. Meanwhile, the pouring amount of chlorodifluoromethane and the pouring amount of hydrofluoric acid are associated with corresponding time stamps so as to facilitate subsequent data processing and analysis. Through the steps, the dumping video of chlorodifluoromethane and hydrofluoric acid and the dumping amount of chlorodifluoromethane and hydrofluoric acid in the production process of the trifluoromethane can be obtained. These data will provide the basis for subsequent data processing and model output.
And S2, acquiring data of a plurality of sensors in a continuous time period in the production process of the trifluoromethane, wherein the production process of the trifluoromethane is to react chlorodifluoromethane with hydrofluoric acid under the action of a catalyst to generate the trifluoromethane and hydrochloric acid, and the plurality of sensors comprise a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph and an infrared spectrometer.
Under the action of a catalyst, chlorodifluoromethane reacts with hydrofluoric acid to generate trifluoromethane and hydrochloric acid, and the chemical equation is CHClF2+HF- & chF3+HCl.
Temperature sensor: for measuring the temperature in the reactor. For example, by placing a temperature sensor inside the reactor, the temperature change inside the reactor can be monitored in real time. The unit of sensor output data is typically in degrees celsius (°c).
A pressure sensor: for measuring the pressure inside the reactor. For example, a pressure sensor is mounted on the reactor wall, and the pressure change inside the reactor can be monitored and recorded in real time. The unit of sensor output data is typically pascal (Pa) or other pressure unit.
PH sensor: for measuring the acidity and alkalinity of the reaction liquid. For example, a pH sensor is added to the reaction liquid, and the degree of acid or alkali of the reaction liquid can be monitored and recorded in real time. The sensor output data is typically PH.
Flow sensor: for measuring the flow of a fluid. For example, flow sensors are installed on chlorodifluoromethane and hydrofluoric acid feed lines to monitor feed fluid flow in real time. The unit of sensor output data is typically either volumetric flow (cubic meter/hour, mTime) or mass flow (kg/hour, kg/h).
Gas chromatograph: apparatus for analysing a gas component. For example, a gas chromatograph is provided at the outlet of the reactor, and the content of various components in the generated gas can be monitored and analyzed in real time. The sensor output data are peak areas or peak height values of the various components.
Infrared spectrometer: apparatus for analyzing infrared absorption spectra of a substance. For example, an infrared spectrometer is arranged at the outlet of the reactor, so that the infrared spectrum of various compounds in the generated gas can be monitored and analyzed in real time. The sensor output data are light intensity values at different wavelengths.
The implementation method of the step S2 comprises the following steps: a. and (3) installing a sensor: according to the technological requirements and monitoring requirements, temperature sensors, pressure sensors, PH sensors, flow sensors and the like are arranged at proper positions on the reactor and related pipelines. Ensuring that the sensor is able to accurately measure and record the variable of interest. b. And (3) connecting a sensor: the sensor is connected with the data acquisition system, so that the measurement signal of the sensor can be transmitted to the data acquisition system. c. And a data acquisition system: a data acquisition system is configured to receive and store the sensor measurement data. The data acquisition system may be a computer or a dedicated data acquisition device. d. Data recording and storage: the data acquisition system is arranged to record the measurement data of the sensor at regular time. For example, it may be arranged to record data once per second and store the data in a database. f. Configuring a gas chromatograph and an infrared spectrometer: a gas chromatograph and an infrared spectrometer are arranged at the outlet of the reactor as required. Ensuring that these devices are able to accurately analyze the composition and infrared spectrum of the generated gas. g. Connect and control gas chromatograph and infrared spectrometer: the gas chromatograph and the infrared spectrometer are connected with a data acquisition system, and the data output by the equipment can be ensured to be acquired in real time. h. Data recording and storage: the data acquisition system is arranged to record the output data of the gas chromatograph and the infrared spectrometer at regular time. For example, it may be arranged to record data once per second and store the data in a database. Through the steps, the continuous time period data of the sensors such as temperature, pressure, PH value, flow, gas chromatograph, infrared spectrometer and the like in the production process of the trifluoromethane can be obtained. The data can help to know the variation trend and the correlation of various variables in the production process, optimize the production parameters and the quality control, and provide a basis for subsequent data processing and model output.
And step S3, determining a first predicted tail gas amount based on the dumping video of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane and a plurality of sensor data of continuous time periods in the production process of the trifluoromethane by using a first prediction model, wherein the first predicted tail gas amount comprises a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid.
The first predicted amount of off-gas includes a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid.
The first prediction model is a long-short term neural network model. The Long-Short Term neural network model includes a Long-Short Term neural network (LSTM). The long-term and short-term neural network model can process sequence data with any length, capture sequence information and output results based on the association relationship of front data and rear data in the sequence. The first prediction model comprehensively considers the characteristics of the association relationship among the chlorodifluoromethane and the pouring video of hydrofluoric acid in the production process of the trifluoromethane at each time point and a plurality of sensor data of continuous time periods in the production process of the trifluoromethane, and finally determines the first prediction tail gas quantity. The first prediction model can be obtained by training a training sample through a gradient descent method.
The input of the first prediction model is the dumping video of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane and a plurality of sensor data of continuous time periods in the production process of the trifluoromethane, and the output of the first prediction model is the first prediction tail gas amount.
The first predictive model may be utilized to predict the content of hydrogen fluoride and hydrochloric acid in the tail gas during the production of trifluoromethane based on the feed-down video and the sensor data.
And S4, obtaining a reaction video of a continuous time period in the production process of the trifluoromethane.
In some embodiments, a camera may be mounted on a wall in the reactor of the trifluoromethane. In some embodiments, the camera may be an industrial-grade high-temperature-resistant camera. The camera can comprehensively capture the reaction process of chlorodifluoromethane and hydrofluoric acid in the reactor.
The reaction video for a continuous period of time during the production of trifluoromethane can display a variety of information such as, for example, 1, transparency and concentration of the gas inside the reactor: the video of the reaction over successive time periods during the production of trifluoromethane can show the transparency and concentration changes of the gases inside the reactor. The amount of exhaust gas from the trifluoromethane is generally related to the change in the gas inside the reactor, and thus this information can be used to determine the amount of exhaust gas. 2. Gas flow rate and discharge conditions of the exhaust gas conduit: the gas flow rate and the discharge condition of the exhaust gas pipeline can be used for judging the amount of the trifluoromethane exhaust gas. As an example, a high and steady gas flow may indicate a higher amount of exhaust gas. 3. Exhaust emissions seen in the reactor: the off-gas of the trifluoromethane may be discharged from the outlet of the reactor or other apparatus. By observing the exhaust emission in the reaction video, the relative magnitude of the amount of exhaust can be inferred. 4. Contaminants near the reactor: other pollutants may accompany the process of producing the exhaust gas of trifluoromethane. The video of the reaction over successive time periods in the process of producing trifluoromethane may show other contaminants.
The video of the reaction at successive time periods in the production process of trifluoromethane is used as important visual data for analyzing the dynamic changes of the reaction process to determine the amount of exhaust gas.
And step S5, determining a second predicted tail gas amount based on the reaction video of the continuous time period in the production process of the trifluoromethane by using a second prediction model, wherein the second predicted tail gas amount comprises a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid.
The second prediction model is a long-short-period neural network model, the input of the second prediction model is a reaction video of a continuous time period in the production process of the trifluoromethane, and the output of the second prediction model is a second prediction tail gas amount.
The second prediction model can be used for determining a second predicted tail gas amount by carrying out reaction videos in a continuous time period.
And S6, determining a target predicted tail gas amount based on the first predicted tail gas amount and the second predicted tail gas amount, wherein the target predicted tail gas amount comprises a target predicted amount of hydrogen fluoride and a target predicted amount of hydrochloric acid.
In some embodiments, the first predicted amount of hydrogen fluoride and the second predicted amount of hydrogen fluoride may be weighted differently, and then weighted and summed to obtain the target predicted amount of hydrogen fluoride, and the first predicted amount of hydrochloric acid and the second predicted amount of hydrochloric acid may be weighted differently, and then weighted and summed to obtain the target predicted amount of hydrochloric acid.
In some embodiments, the amount of sodium hydroxide added may also be determined by an exhaust treatment model based on the predicted amount of tail gas.
The exhaust treatment model is a deep neural network model comprising a deep neural network (Deep Neural Networks, DNN). And the input of the waste gas treatment model is the predicted tail gas amount, and the output of the waste gas treatment model is the addition amount of sodium hydroxide. The sodium hydroxide can absorb hydrogen fluoride and hydrochloric acid in the tail gas. Sodium hydroxide may be added to the exhaust treatment solution to absorb hydrogen fluoride and hydrochloric acid in the exhaust gas.
Based on the same inventive concept, fig. 2 is a schematic diagram of an exhaust gas amount prediction system for producing trifluoromethane according to an embodiment of the present invention, where the exhaust gas amount prediction system for producing trifluoromethane includes:
a first obtaining module 21, configured to obtain a video of pouring chlorodifluoromethane and hydrofluoric acid, a pouring amount of chlorodifluoromethane, and a pouring amount of hydrofluoric acid during production of trifluoromethane;
a second obtaining module 22, configured to obtain data of a plurality of sensors in a continuous time period during a process of producing trifluoromethane, where the process of producing trifluoromethane is to react chlorodifluoromethane with hydrofluoric acid under the action of a catalyst to generate trifluoromethane and hydrochloric acid, and the plurality of sensors include a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph, and an infrared spectrometer;
a first predicted tail gas amount determination module 23 for determining a first predicted tail gas amount using a first prediction model based on a video of the dumping of chlorodifluoromethane and hydrofluoric acid at the time of production of the trifluoromethane and a plurality of sensor data for successive time periods in the production process of the trifluoromethane, the first predicted tail gas amount including a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid;
a third obtaining module 24, configured to obtain a reaction video of a continuous time period in the production process of the trifluoromethane;
a second predicted tail gas amount determination module 25 for determining a second predicted tail gas amount using a second prediction model based on the reaction videos of successive time periods in the production process of the trifluoromethane, the second predicted tail gas amount including a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid;
a target predicted amount determination module 26 for determining a target predicted amount of exhaust gas based on the first predicted amount of exhaust gas and the second predicted amount of exhaust gas, the target predicted amount of exhaust gas including a target predicted amount of hydrogen fluoride and a target predicted amount of hydrochloric acid.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. A method for predicting the amount of tail gas from the production of trifluoromethane, comprising:
the method for obtaining the pouring video of chlorodifluoromethane and hydrofluoric acid, the pouring amount of chlorodifluoromethane and the pouring amount of hydrofluoric acid during the production of the trifluoromethane comprises the following steps of: the method comprises the following steps of pouring amount of chlorodifluoromethane and hydrofluoric acid, frequency of pouring of chlorodifluoromethane and hydrofluoric acid, concrete time for pouring chlorodifluoromethane and hydrofluoric acid into a reactor and a pouring operation flow of staff;
acquiring data of a plurality of sensors in a continuous time period in the production process of the trifluoromethane, wherein the production process of the trifluoromethane is to react chlorodifluoromethane with hydrofluoric acid under the action of a catalyst to generate the trifluoromethane and hydrochloric acid, and the plurality of sensors comprise a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph and an infrared spectrometer;
determining a first predicted tail gas amount by using a first prediction model based on a dumping video of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane and a plurality of sensor data of continuous time periods in the production process of the trifluoromethane, wherein the first predicted tail gas amount comprises a first predicted amount of hydrogen fluoride and a first predicted amount of hydrochloric acid, the first prediction model is a long-short-period neural network model, the input of the first prediction model is the dumping video of chlorodifluoromethane and hydrofluoric acid during the production of the trifluoromethane and the plurality of sensor data of continuous time periods in the production process of the trifluoromethane, and the output of the first prediction model is the first predicted tail gas amount;
acquiring a reaction video of a continuous time period in a production process of the trifluoromethane, wherein various information displayed by the reaction video of the continuous time period in the production process of the trifluoromethane comprises: transparency and concentration of gases inside the reactor, gas flow rate and emission conditions of the exhaust gas pipeline, visible exhaust gas emission in the reactor, pollutants in the vicinity of the reactor;
determining a second predicted tail gas amount by using a second prediction model based on the reaction videos of continuous time periods in the production process of the trifluoromethane, wherein the second predicted tail gas amount comprises a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid, the second prediction model is a long-short-period neural network model, the input of the second prediction model is the reaction videos of continuous time periods in the production process of the trifluoromethane, and the output of the second prediction model is the second predicted tail gas amount;
a target predicted tail gas amount is determined based on the first predicted tail gas amount and the second predicted tail gas amount, the target predicted tail gas amount including a target predicted amount of hydrogen fluoride and a target predicted amount of hydrochloric acid.
2. The method of predicting the amount of tail gas from the production of trifluoromethane as recited in claim 1, wherein said determining a target predicted amount of tail gas based on said first predicted amount of tail gas and said second predicted amount of tail gas comprises: and giving different weights to the first predicted amount of the hydrogen fluoride and the second predicted amount of the hydrogen fluoride, carrying out weighted summation to obtain the target predicted amount of the hydrogen fluoride, giving different weights to the first predicted amount of the hydrochloric acid and the second predicted amount of the hydrochloric acid, and carrying out weighted summation to obtain the target predicted amount of the hydrochloric acid.
3. The method for predicting the amount of tail gas from the production of trifluoromethane as defined in claim 1, further comprising: and determining the addition amount of sodium hydroxide based on the target predicted tail gas amount.
4. An exhaust gas amount prediction system based on the exhaust gas amount prediction method of trifluoromethane production according to any one of claims 1 to 3, characterized by comprising:
the first acquisition module is used for acquiring a pouring video of chlorodifluoromethane and hydrofluoric acid, a pouring amount of chlorodifluoromethane and a pouring amount of hydrofluoric acid during production of the trifluoromethane, and various information displayed by the pouring video of the chlorodifluoromethane and the hydrofluoric acid comprises the following components: the method comprises the following steps of pouring amount of chlorodifluoromethane and hydrofluoric acid, frequency of pouring of chlorodifluoromethane and hydrofluoric acid, concrete time for pouring chlorodifluoromethane and hydrofluoric acid into a reactor and a pouring operation flow of staff;
the second acquisition module is used for acquiring data of a plurality of sensors in a continuous time period in the production process of the trifluoromethane, wherein the production process of the trifluoromethane is that chlorodifluoromethane reacts with hydrofluoric acid under the action of a catalyst to generate the trifluoromethane and hydrochloric acid, and the plurality of sensors comprise a temperature sensor, a pressure sensor, a PH sensor, a flow sensor, a gas chromatograph and an infrared spectrometer;
a first prediction tail gas amount determining module, configured to determine a first prediction tail gas amount by using a first prediction model based on a back-off video of chlorodifluoromethane and hydrofluoric acid during production of the trifluoromethane and a plurality of sensor data of a continuous time period in a production process of the trifluoromethane, wherein the first prediction tail gas amount includes a first prediction amount of hydrogen fluoride and a first prediction amount of hydrochloric acid, the first prediction model is a long-short-period neural network model, an input of the first prediction model is the back-off video of chlorodifluoromethane and hydrofluoric acid during production of the trifluoromethane and a plurality of sensor data of a continuous time period in a production process of the trifluoromethane, and an output of the first prediction model is the first prediction tail gas amount;
a third obtaining module, configured to obtain a reaction video of a continuous time period in a process of producing trifluoromethane, where the plurality of information displayed by the reaction video of the continuous time period in the process of producing trifluoromethane includes: transparency and concentration of gases inside the reactor, gas flow rate and emission conditions of the exhaust gas pipeline, visible exhaust gas emission in the reactor, pollutants in the vicinity of the reactor;
a second predicted tail gas amount determining module, configured to determine a second predicted tail gas amount by using a second prediction model based on a reaction video of a continuous time period in the production process of the trifluoromethane, where the second predicted tail gas amount includes a second predicted amount of hydrogen fluoride and a second predicted amount of hydrochloric acid, the second prediction model is a long-short-period neural network model, an input of the second prediction model is a reaction video of a continuous time period in the production process of the trifluoromethane, and an output of the second prediction model is the second predicted tail gas amount;
and a target predicted amount determination module configured to determine a target predicted amount of tail gas based on the first predicted amount of tail gas and the second predicted amount of tail gas, the target predicted amount of tail gas including a target predicted amount of hydrogen fluoride and a target predicted amount of hydrochloric acid.
5. The tail gas amount prediction system for trifluoromethane production of claim 4, wherein said target prediction determination module is further configured to: and giving different weights to the first predicted amount of the hydrogen fluoride and the second predicted amount of the hydrogen fluoride, carrying out weighted summation to obtain the target predicted amount of the hydrogen fluoride, giving different weights to the first predicted amount of the hydrochloric acid and the second predicted amount of the hydrochloric acid, and carrying out weighted summation to obtain the target predicted amount of the hydrochloric acid.
6. The tail gas quantity prediction system for the production of trifluoromethane according to claim 4, said system further operable to: and determining the addition amount of sodium hydroxide based on the target predicted tail gas amount.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310882319.0A CN116612836B (en) | 2023-07-19 | 2023-07-19 | Tail gas amount prediction method and system for trifluoromethane production |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310882319.0A CN116612836B (en) | 2023-07-19 | 2023-07-19 | Tail gas amount prediction method and system for trifluoromethane production |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116612836A CN116612836A (en) | 2023-08-18 |
| CN116612836B true CN116612836B (en) | 2023-10-03 |
Family
ID=87675096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310882319.0A Active CN116612836B (en) | 2023-07-19 | 2023-07-19 | Tail gas amount prediction method and system for trifluoromethane production |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116612836B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112001527A (en) * | 2020-07-29 | 2020-11-27 | 中国计量大学 | Industrial production process target data prediction method of multi-feature fusion deep neural network |
| JP2021091617A (en) * | 2019-12-06 | 2021-06-17 | 株式会社豊田中央研究所 | Gas supply apparatus, methane producing apparatus, and control method of gas supply apparatus |
| CN113720968A (en) * | 2021-07-29 | 2021-11-30 | 深圳市天得一环境科技有限公司 | Gas concentration detection method, device, system and storage medium |
| CN115186900A (en) * | 2022-05-31 | 2022-10-14 | 重庆大学 | Dynamic blast furnace gas production prediction method and system suitable for multiple working condition types |
| CN115204533A (en) * | 2022-09-16 | 2022-10-18 | 中国地质大学（北京） | Oil-gas yield prediction method and system based on multivariable weighted combination model |
| CN115356998A (en) * | 2022-08-31 | 2022-11-18 | 深圳能源环保股份有限公司 | Nitrogen oxide prediction method based on artificial intelligence and related device |
-
2023
- 2023-07-19 CN CN202310882319.0A patent/CN116612836B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2021091617A (en) * | 2019-12-06 | 2021-06-17 | 株式会社豊田中央研究所 | Gas supply apparatus, methane producing apparatus, and control method of gas supply apparatus |
| CN112001527A (en) * | 2020-07-29 | 2020-11-27 | 中国计量大学 | Industrial production process target data prediction method of multi-feature fusion deep neural network |
| CN113720968A (en) * | 2021-07-29 | 2021-11-30 | 深圳市天得一环境科技有限公司 | Gas concentration detection method, device, system and storage medium |
| CN115186900A (en) * | 2022-05-31 | 2022-10-14 | 重庆大学 | Dynamic blast furnace gas production prediction method and system suitable for multiple working condition types |
| CN115356998A (en) * | 2022-08-31 | 2022-11-18 | 深圳能源环保股份有限公司 | Nitrogen oxide prediction method based on artificial intelligence and related device |
| CN115204533A (en) * | 2022-09-16 | 2022-10-18 | 中国地质大学（北京） | Oil-gas yield prediction method and system based on multivariable weighted combination model |
Non-Patent Citations (1)
| Title |
|---|
| 基于时序数据的尾气监测分析平台研究;刘伟锋 等;现代制造工程(6);90-94 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116612836A (en) | 2023-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10670570B2 (en) | Method of analysing gas chromatography data | |
| RU2010109422A (en) | SYSTEM AND METHODS OF CONTINUOUS OPERATIONAL MONITORING OF A CHEMICAL OR OIL REFINING PLANT | |
| CA2923624C (en) | Non-intrusive sensor system | |
| Vieira et al. | Experimental investigation of horizontal gas–liquid stratified and annular flow using wire-mesh sensor | |
| JP2008076267A (en) | Accuracy control system, analyzer, and accuracy control method | |
| KR100867938B1 (en) | Prediction method for watching performance of power plant measuring instrument by dependent variable similarity and kernel feedback | |
| EP3913449A1 (en) | Analysis system and analysis method | |
| CN104849360A (en) | System for monitoring operating status of chromatograph | |
| CN109375016A (en) | Sulfur hexafluoride resolution characteristic test method at a temperature of different Superheated steam driers | |
| CN116612836B (en) | Tail gas amount prediction method and system for trifluoromethane production | |
| JP5271776B2 (en) | Automatic analyzer | |
| US20150168199A1 (en) | Computer based fluid flow velocity estimation from concentrations of a reacting constituent for products and services | |
| CN114730184A (en) | Computer system and method for estimating changes in fugitive emissions | |
| CN116985183A (en) | Quality monitoring and management method and system for near infrared spectrum analyzer | |
| US20160209310A1 (en) | Gas measurement apparatus | |
| CN205103155U (en) | Gaseous on -line measuring appearance suitable for organic looks | |
| US20150168197A1 (en) | Computer based fluid flow velocity estimation from concentrations of a reacting constituent for products and services | |
| CN101995390B (en) | Method for analyzing compositions of materials in reactor | |
| CN105259135B (en) | Suitable for real-time online without measuring point temperature-compensating near-infrared measuring method | |
| US10472640B2 (en) | Method of real-time prognosis of flooding phenomenon in packed column | |
| Conway et al. | How to get Lab Equivalent Oil Analysis 24/7. | |
| RU2530447C1 (en) | Method of analysing multicomponent gaseous medium of sealed containers with electronic devices and apparatus therefor | |
| CN106909773A (en) | It is a kind of that the method that elemental analyser models robustness is improved based on PLS | |
| CN115015037A (en) | Coal dressing suspension density detection method and device, storage medium and computer equipment | |
| Demo et al. | Corrosion management and asset tracking through improved airframe contaminant monitoring |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |