CN113686922A - Online intelligent monitoring device and method for risks of reaction kettle - Google Patents
Online intelligent monitoring device and method for risks of reaction kettle Download PDFInfo
- Publication number
- CN113686922A CN113686922A CN202110716311.8A CN202110716311A CN113686922A CN 113686922 A CN113686922 A CN 113686922A CN 202110716311 A CN202110716311 A CN 202110716311A CN 113686922 A CN113686922 A CN 113686922A
- Authority
- CN
- China
- Prior art keywords
- reaction kettle
- machine learning
- learning model
- early warning
- risk
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012806 monitoring device Methods 0.000 title claims abstract description 15
- 238000006243 chemical reaction Methods 0.000 title claims description 98
- 239000004020 conductor Substances 0.000 claims abstract description 37
- 238000012544 monitoring process Methods 0.000 claims abstract description 36
- 238000001514 detection method Methods 0.000 claims abstract description 33
- 238000005260 corrosion Methods 0.000 claims abstract description 28
- 230000007797 corrosion Effects 0.000 claims abstract description 14
- 238000010801 machine learning Methods 0.000 claims description 51
- 230000006870 function Effects 0.000 claims description 19
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000005094 computer simulation Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 6
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 230000005611 electricity Effects 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract description 5
- 238000009413 insulation Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 230000029936 alkylation Effects 0.000 description 1
- 238000005804 alkylation reaction Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 239000000975 dye Substances 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000005984 hydrogenation reaction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 238000006396 nitration reaction Methods 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 239000000377 silicon dioxide Substances 0.000 description 1
- 238000005987 sulfurization reaction Methods 0.000 description 1
- 231100000331 toxic Toxicity 0.000 description 1
- 230000002588 toxic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
Abstract
The utility model relates to an online intelligent monitoring device and method of reation kettle risk, its electric conductor that is connected with reation kettle conductive housing electricity, with the detection module that the electric conductor electricity is connected, with the power module that the detection module electricity is connected, with the corrosion-resistant conductor that the power module electricity is connected, corrosion-resistant conductor is arranged in reation kettle and is not contacted with reation kettle conductive housing, detection module is arranged in real time monitoring wire loop's current value and output current signal, detection module is connected with the early warning module that is used for receiving current signal, the early warning module judges whether reation kettle is normal based on current signal, if no, early warning module output early warning signal. The method and the device have the effects of convenience in monitoring and high prediction accuracy.
Description
Technical Field
The application relates to the technical field of chemical industry, in particular to a device and a method for online intelligent monitoring of risks of a reaction kettle.
Background
Currently, the current practice is. In the industries of medicine, dye, medicine, food and the like, a reaction kettle is needed for completing the technological processes of sulfuration, nitration, hydrogenation, alkylation, polymerization, condensation and the like. As toxic and harmful dangerous chemicals are often filled in the reaction kettle, once the reaction kettle breaks down, solution in the reaction kettle leaks to destroy the production environment, and accidents such as fire, explosion and the like are caused to cause serious casualties and economic loss.
In order to reduce the occurrence of accidents, the factory needs to frequently and periodically perform comprehensive detection on the reaction kettle. During detection, the reaction kettle needs to stop working and be cleaned comprehensively, so that the working efficiency is greatly influenced; meanwhile, the service life of the reaction kettle can be shortened due to frequent change of the temperature; in addition, the detection process of the reaction kettle usually consumes time and labor, has high cost, and can cause some unnecessary damages to the reaction kettle. The regular high-frequency equipment detection can only reduce the occurrence of accidents to a certain extent, and can not well eliminate risks.
The existing online monitoring method, such as means for monitoring pressure, temperature and the like, can only monitor the condition of substance reaction, but cannot monitor the risk and damage condition of the reaction kettle. The damage and risk of the reaction kettle are the real hidden troubles of major accidents of the reaction kettle.
Disclosure of Invention
In order to conveniently monitor the self damage condition of the reaction kettle and improve the accuracy of prediction, the application provides an online intelligent monitoring device and method for the risk of the reaction kettle.
The utility model provides a pair of online intelligent monitoring device of reation kettle risk adopts following technical scheme:
the utility model provides an online intelligent monitoring device of reation kettle risk, includes the electric conductor of being connected with the electrically conductive shell electricity of reation kettle, with the detection module that the electric conductor electricity is connected, with the power module that the detection module electricity is connected, with the corrosion-resistant conductor that the power module electricity is connected, corrosion-resistant conductor arranges in reation kettle and does not contact with the electrically conductive shell of reation kettle, detection module is arranged in real time monitoring wire loop's current value and output current signal, detection module is connected with the early warning module that is used for receiving current signal, the early warning module judges whether normal based on current signal reation kettle, if no, the early warning module outputs early warning signal.
By adopting the technical scheme, under the conventional condition, the inner wall of the reaction kettle is attached with the anti-corrosion insulating layer, and when the reaction kettle works, the conductive medium is in contact with the anti-corrosion insulating layer.
When the anticorrosion insulation layer is not damaged, a closed loop cannot be formed between the conductor and the anticorrosion conductor due to the anticorrosion insulation layer. When the insulating property of the insulating layer is excellent, the loop current detected by the detection module is close to zero.
When any one of the anticorrosive insulation layer is damaged, the conductive medium overflows, and with reation kettle conductive shell looks butt, at this moment, form the circular telegram return circuit because of conductive medium and detection module between electric conductor and the anticorrosive conductor, the current value that detection module detected takes place corresponding change, when the current signal who changes is acquireed by early warning module, early warning module can judge whether the inside damage takes place of reation kettle based on current signal, in case reation kettle takes place the damage, early warning module output early warning signal, when early warning signal notifies managers through the internet mode, just can be instant effectual take emergency measures, thereby effectively reduce the risk, and the safety factor is improved.
Optionally, the early warning module judges whether the reaction kettle is normal through the machine learning model, and outputs an early warning signal if the reaction kettle is abnormal.
By adopting the technical scheme, the process of machine learning is actually a process of determining unknown parameters through a certain method by using known samples. After the current information is acquired, the current information is compared with known sample information, whether the acquired current information is accurate or not can be indirectly known, and the anti-corrosion insulating layer in the reaction kettle is broken when the acquired current information is matched with the known sample information. Meanwhile, known sample data can be continuously optimized in an autonomous learning mode of the machine learning model, so that the monitoring accuracy can be improved.
Optionally, the detection module is an ammeter, and the ammeter is used for monitoring a current value in the wire loop in real time and outputting a current signal.
By adopting the technical scheme, the set ammeter can conveniently detect current information and send current signals conveniently.
Optionally, the power supply module is a direct current variable power supply, and the direct current variable power supply can adjust power of the power supply.
Through adopting above-mentioned technical scheme, the variable power of direct current that sets up can be according to the power of different reation kettle equipment regulation power to can effectively improve practicality and suitability.
The second purpose of the application provides an online intelligent monitoring method for the risk of the reaction kettle, which adopts the following technical scheme:
an online intelligent monitoring method for the risk of a reaction kettle comprises the following steps,
acquiring conductive medium sample information and conductive performance information;
monitoring the magnitude of a current value in a wire loop which can form a loop with a conductive medium sample in real time, and outputting a current signal to a machine learning model;
the machine learning model judges whether the reaction kettle is normal or not through the received current signal;
when the machine learning model judges that the reaction kettle is abnormal, outputting an early warning signal;
otherwise, it does not act.
By adopting the technical scheme, whether the reaction kettle is damaged or not can be indirectly judged through the machine learning model through the acquired current signal, and the judgment mode is simple and effective and has a prominent effect. In addition, under the actual condition, a large amount of simulation sample data can be directly obtained by acquiring sample data such as conductive medium sample information, conductive performance information, current information and the like in advance and adopting a computer simulation mode, and accordingly, a more accurate function model is convenient to establish, the authenticity and the accuracy of a simulation known sample can be effectively improved, and the monitoring accuracy of a machine learning model can be further improved.
Optionally, the machine learning model includes, but is not limited to, one or more of a support vector machine, a neural network.
By adopting the technical scheme, the support vector machine and the neural network are very good modeling tools, the neural network has very strong nonlinear fitting capacity, can map any complex nonlinear relation, and is simple in learning rule and convenient for computer implementation. The method has strong robustness, memory capability, nonlinear mapping capability and strong self-learning capability. Under the condition of a certain sample, the calculation amount of the support vector machine is smaller than that of the neural network, and the modeling precision is higher.
Optionally, the abstract model of the machine learning model includes:
，
wherein the content of the first and second substances,
the danger degree of the reaction kettle at the moment t is shown, the value range of t is 0-1S,
and n is the number of samples at a time point for the current signal acquired at the time point t.
By adopting the technical scheme, accidents and risks can be effectively reduced by acquiring the danger degree information of the reaction kettle in real time under the actual condition. Meanwhile, when the sampling data is enough, the function model of the abstract model can be effectively optimized, and therefore the monitoring accuracy is improved.
Optionally, the method further includes:
acquiring conductive medium sample information and conductive performance information for multiple times;
monitoring the magnitude of a current value in a wire loop which can form a loop with the conductive medium sample in real time according to the acquired conductive medium sample information and conductive performance information each time, and outputting a current signal to a machine learning model;
and the machine learning model optimizes the abstract model of the machine learning model according to the current signal sample information collected for many times.
By adopting the technical scheme, the accuracy of the function model of the machine learning model can be improved by acquiring a large amount of sample data information, so that the monitoring accuracy can be improved. In addition, when the number of the acquired samples is enough, a better machine learning model can be trained, and the machine learning model has better normalization capability.
Optionally, the method further includes:
the conductive medium sample information, the conductive performance information and the abstract function model information are obtained in a mode of combining computer simulation and real data.
By adopting the technical scheme, the mode of acquiring the sample data by combining computer simulation and real data can further improve the safety degree. The contrast people is damaged for causing reation kettle in a large number, just can acquire the mode of sample data, and the mode of sample is acquireed to this application, and is comparatively safe, and the rate of accuracy also can obtain the guarantee simultaneously.
Optionally, the sample information obtained by the machine learning model further includes different types of reaction kettles, chemical reactions of conductive media in the reaction kettles, and operating environments of the reaction kettles.
By adopting the technical scheme, when the current signal, different reaction kettle types, chemical reaction of the conductive medium in the reaction kettle, the operating environment of the reaction kettle and other sample data are acquired, and in addition, as many tests as possible are carried out, a better machine learning model can be trained.
In summary, the present application includes at least one of the following beneficial technical effects:
1. when the anticorrosion insulation layer is not damaged, a closed loop cannot be formed between the conductor and the anticorrosion conductor due to the anticorrosion insulation layer. When the insulating property of the insulating layer is excellent, the loop current detected by the detection module is close to zero. When any one of the anti-corrosion insulating layers is damaged, the conductive medium overflows and abuts against the conductive shell of the reaction kettle, at the moment, a power-on loop is formed between the conductive body and the anti-corrosion conductor due to the conductive medium and the detection module, the current value detected by the detection module is correspondingly changed, when the changed current signal is acquired by the early warning module, the early warning module can judge whether the inside of the reaction kettle is damaged or not based on the current signal, once the reaction kettle is damaged, the early warning module outputs an early warning signal, and when the early warning signal is notified to a manager through an internet mode, emergency measures can be taken timely and effectively, so that the risk is effectively reduced, and the safety coefficient is improved;
2. known sample data can be continuously optimized in an autonomous learning mode of a machine learning model, so that the monitoring accuracy can be improved;
3. the method for acquiring the sample data by combining the computer simulation and the real data can directly acquire a large amount of simulation sample data, and accordingly, a more accurate function model is convenient to establish, so that the authenticity and the accuracy of the simulation known sample can be effectively improved, and the monitoring accuracy of the machine learning model can be further improved.
Drawings
FIG. 1 is a schematic overall view of an on-line monitoring device in one embodiment;
FIG. 2 is a schematic diagram of current signaling in one embodiment;
FIG. 3 is a flow diagram of an online monitoring method in one embodiment.
Reference numerals: 1. a reaction kettle; 2. an electrical conductor; 3. a detection module; 4. a power supply module; 5. a corrosion-resistant conductor; 6. an early warning module; 7. and an anti-corrosion insulating layer.
Detailed Description
The present application is described in further detail below with reference to figures 1-3.
The embodiment of the application discloses online intelligent monitoring device of reation kettle risk for whether detect reation kettle 1 takes place the damage.
Referring to fig. 1, the reactor comprises a conductor 2, a detection module 3, a power supply module 4 and a corrosion-resistant conductor 5 which are connected in series in sequence through wires, the conductor 2 is connected with a conductive shell of a reaction kettle 1, the conductor 2 is made of a conductive material, specifically, a copper block and the like, in this embodiment, the material and the shape of the conductor 2 and the placement position of the conductor 2 on the reaction kettle 1 are not limited, but the conductor 2 can be electrically connected with the conductive shell of the reaction kettle 1 and can achieve the effect of ensuring safety.
The corrosion-resistant conductor 5 contacts with the conductive medium in the reaction kettle 1 through the discharge port of the reaction kettle 1, meanwhile, the corrosion-resistant conductor 5 does not contact with the conductive shell of the reaction kettle 1, in an actual situation, the corrosion-resistant insulating layer 7 is nested in the reaction kettle 1, the corrosion-resistant insulating layer 7 is attached to the inner wall of the reaction kettle 1, and the conductive medium is in the accommodating channel formed by the corrosion-resistant insulating layer 7. The corrosion-resistant insulating layer 7 may be glass-lined, which is glass containing high silica and is firmly adhered to the inner surface of the reaction vessel 1 by high-temperature burning. Glass lining has strong stability and metal strength.
The detection module 3 is specifically an ammeter, the ammeter is used for monitoring the current value in a wire loop where the ammeter is located in real time and outputting a current signal, when the anti-corrosion insulating layer 7 is not damaged, the electric conductor 2 and the anti-corrosion conductor 5 are enabled to be not conducted due to the insulating effect of the anti-corrosion insulating layer 7, and the current value of the ammeter is close to zero at the moment. The power supply module 4 is a direct current variable power supply, and the power of the power supply can be adjusted by the direct current variable power supply. In this regard, the power supply module 4 is adjusted to a suitable power for the unused reactor 1 equipment and the conductive medium loaded in the reactor 1.
When any one part of the anti-corrosion insulating layer 7 is damaged, the conductive medium overflows from the damaged part and is abutted against the conductive shell of the reaction kettle 1, at the moment, a power-on loop is formed between the conductive body 2 and the anti-corrosion conductor 7 due to the conductive medium and the detection module 3, the current value detected by the detection module 3 is correspondingly changed, and after the detection module 3 detects the changed current value, the changed current signal is synchronously output.
The early warning module 6 comprises a machine learning model, the early warning module 6 judges whether the reaction kettle 1 is normal or not through the machine learning model, and if not, an early warning signal is output. And known sample data, namely a function model, is arranged in the machine learning model, and when the function model formed according to the acquired current information is matched with the function model of the known sample, the reaction kettle 1 is judged to be abnormal, namely the anti-corrosion insulating layer in the reaction kettle is damaged. When the two are not matched, the reaction kettle 1 is in a normal state.
When the machine learning model judges that the machine learning model is abnormal, the machine learning model outputs an early warning signal, and when the early warning signal is notified to a manager through an internet mode (wireless communication mode), the manager can conveniently take corresponding emergency measures in time, so that the risk is reduced, and the safety degree of monitoring is improved. In practice, the machine learning model includes, but is not limited to, one or more of a support vector machine, a neural network. To this end, the monitoring accuracy can be effectively improved.
The implementation principle of the online intelligent monitoring device for the risk of the reaction kettle in the embodiment of the application is as follows: when any one part of the anti-corrosion insulating layer 7 is damaged, the conductive medium overflows and abuts against the conductive shell of the reaction kettle 1, at the moment, a power-on loop is formed between the electric conductor 2 and the anti-corrosion conductor 7 due to the conductive medium and the detection module 3, the current value detected by the detection module 3 changes correspondingly, when a changed current signal is obtained by the machine learning model, a corresponding function model is generated, and when the function model is matched with a function model of known sample data, the machine learning model outputs a warning signal.
The embodiment of the application also discloses a method applied to the online intelligent monitoring device for the risk of the reaction kettle, which is used for improving the monitoring accuracy.
An online intelligent monitoring method for the risk of a reaction kettle comprises the following steps,
s1, acquiring conductive medium sample information and conductive performance information;
the conductive medium sample information is the material added in the reaction kettle, and the conductive property information is the conductive property of the conductive medium sample. The conductive medium sample is selected according to actual conditions, but after each conductive medium is selected, the conductive performance information of the conductive medium sample needs to be acquired.
S2: monitoring the magnitude of a current value in a wire loop which can form a loop with a conductive medium sample in real time, and outputting a current signal to a machine learning model;
the machine learning model may be such a modeling tool that supports a vector machine or a neural network.
S3: the machine learning model judges whether the reaction kettle is normal or not through the received current signal;
the method and the device for modeling the data of the mobile terminal acquire the known sample source of the modeling data by adopting a computer simulation test mode. When the acquired current signal is taken as an input, the abstract model of the machine learning model is as follows:
，
wherein the content of the first and second substances,
the danger degree of the reaction kettle at the moment t is shown, the value range of t is 0-1S,
and n is the number of samples at a time point for the current signal acquired at the time point t.
To train the machine learning model, a large amount of sample data needs to be collected. The specific method for acquiring a large amount of sample data comprises the following steps:
acquiring conductive medium sample information and conductive performance information for multiple times;
monitoring the magnitude of a current value in a wire loop which can form a loop with the conductive medium sample in real time according to the acquired conductive medium sample information and conductive performance information each time, and outputting a current signal to a machine learning model;
and the machine learning model optimizes the abstract model of the machine learning model according to the current signal sample information collected for many times.
Through changing the electric conductivity information (actually simulating the damage degree of the reaction kettle) of the specific conductive medium sample for many times, the change of the current signal can be correspondingly driven, and the current signal every time can be sent to the machine learning model. When each damage degree is tested for hundreds, thousands or even thousands of times, a more accurate function model can be established. The abstract model function for acquiring a large amount of sample data is as follows:
wherein k represents the test times, and the value range of k is [1, k ]]K is a natural number, and when the kth test is performed,
for the current signal acquired at time t, n is the number of samples taken at a time point, per sample
The state of the reaction vessel at time t (degree of breakage) was shown.
Through a large amount of sample data acquisition, the machine learning model with higher accuracy can be trained, so that the machine learning model has better normalization capability. Meanwhile, the sample data can also comprise different reaction kettle types, chemical reactions of the conductive media in the reaction kettles and the operating environments of the reaction kettles, and when the obtained sample data also comprises different reaction kettle types, chemical reactions of the conductive media in the reaction kettles and the operating environments of the reaction kettles, the function model can be further optimized. The monitoring accuracy and the authenticity are improved.
S4: when the machine learning model judges that the reaction kettle is abnormal, an early warning signal is output, otherwise, the reaction kettle does not act.
When a function model formed based on the acquired current signal is matched with a known function model, the reaction kettle is proved to be damaged to a certain degree. In response to this, the condition of reation kettle can be judged indirectly. Meanwhile, managers can be warned in time through early warning signals.
The above embodiments are preferred embodiments of the present application, and the protection scope of the present application is not limited by the above embodiments, so: all equivalent changes made according to the structure, shape and principle of the present application shall be covered by the protection scope of the present application.
Claims (10)
1. The utility model provides an online intelligent monitoring device of reation kettle risk which characterized in that: the electric conductor comprises an electric conductor (2) electrically connected with a conductive shell of a reaction kettle (1), a detection module (3) electrically connected with the electric conductor (2), a power supply module (4) electrically connected with the detection module (3), and a corrosion-resistant conductor (5) electrically connected with the power supply module (4), wherein the corrosion-resistant conductor (5) is arranged in the reaction kettle (1) and is not contacted with the conductive shell of the reaction kettle (1), the detection module (3) is used for monitoring the current value in a wire loop in real time and outputting a current signal, the detection module (3) is connected with an early warning module (6) used for receiving the current signal, the early warning module (6) judges whether the reaction kettle (1) is normal or not based on the current signal, and if not, the early warning module (6) outputs an early warning signal.
2. The on-line intelligent monitoring device for the risk of the reaction kettle according to claim 1, characterized in that: and the early warning module (6) judges whether the reaction kettle is normal or not through the machine learning model, and outputs an early warning signal if the reaction kettle is abnormal.
3. The on-line intelligent monitoring device for the risk of the reaction kettle as recited in claim 2, wherein: the detection module (3) is an ammeter used for monitoring the current value in the wire loop in real time and outputting a current signal.
4. The on-line intelligent monitoring device for the risk of the reaction kettle as claimed in claim 1, 2 or 3, wherein: the power supply module (4) is a direct-current variable power supply, and the power of the power supply can be adjusted by the direct-current variable power supply.
5. An online intelligent monitoring method for reactor risk, which is applied to the online intelligent monitoring device for reactor risk in any one of claims 1 to 4, and is characterized in that:
acquiring conductive medium sample information and conductive performance information;
monitoring the magnitude of a current value in a wire loop which can form a loop with a conductive medium sample in real time, and outputting a current signal to a machine learning model;
the machine learning model judges whether the reaction kettle is normal or not through the received current signal;
when the machine learning model judges that the reaction kettle is abnormal, outputting an early warning signal;
otherwise, it does not act.
6. The on-line intelligent monitoring method for the risk of the reaction kettle as recited in claim 5, wherein: the machine learning model includes, but is not limited to, one or more of a support vector machine, a neural network.
7. The online intelligent monitoring method for the risk of the reaction kettle as recited in claim 5 or 6, characterized in that: the abstract model of the machine learning model comprises:
wherein the content of the first and second substances,
the danger degree of the reaction kettle at the moment t is shown, the value range of t is 0-1S,
and n is the number of samples at a time point for the current signal acquired at the time point t.
8. The on-line intelligent monitoring method for the risk of the reaction kettle as recited in claim 5, wherein: further comprising:
acquiring conductive medium sample information and conductive performance information for multiple times;
monitoring the magnitude of a current value in a wire loop which can form a loop with the conductive medium sample in real time according to the acquired conductive medium sample information and conductive performance information each time, and outputting a current signal to a machine learning model;
and the machine learning model optimizes the abstract model of the machine learning model according to the current signal sample information collected for many times.
9. The online intelligent monitoring method for the risk of the reaction kettle as recited in claim 6 or 8, characterized in that: further comprising:
the conductive medium sample information, the conductive performance information and the abstract function model information are obtained in a mode of combining computer simulation and real data.
10. The on-line intelligent monitoring method for the risk of the reaction kettle as recited in claim 8, wherein: the sample information obtained by the machine learning model further comprises different reaction kettle types, chemical reactions of conductive media in the reaction kettles and the operating environment of the reaction kettles.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110716311.8A CN113686922A (en) | 2021-06-28 | 2021-06-28 | Online intelligent monitoring device and method for risks of reaction kettle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110716311.8A CN113686922A (en) | 2021-06-28 | 2021-06-28 | Online intelligent monitoring device and method for risks of reaction kettle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113686922A true CN113686922A (en) | 2021-11-23 |
Family
ID=78576801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110716311.8A Pending CN113686922A (en) | 2021-06-28 | 2021-06-28 | Online intelligent monitoring device and method for risks of reaction kettle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113686922A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1184931A (en) * | 1996-12-11 | 1998-06-17 | 唐秀家 | Method and apparatus for detecting and positioning leakage of fluid transferring pipeline |
| CN101096006A (en) * | 2007-05-14 | 2008-01-02 | 美晨集团股份有限公司 | Material-feeding automatic detecting device and the chemical producing system |
| CN105424293A (en) * | 2015-12-02 | 2016-03-23 | 深圳凌水环保科技股份有限公司 | Water supply-drainage pipe leak detection system and detection method |
| CN205656230U (en) * | 2016-04-21 | 2016-10-19 | 山东大地盐化集团有限公司 | Damaged online alarm device of enamel reactor |
| CN108443716A (en) * | 2018-02-06 | 2018-08-24 | 江阴市长龄机械制造有限公司 | A kind of fluid hose convenient for detection leakage point |
| CN111898669A (en) * | 2020-07-24 | 2020-11-06 | 大连重工机电设备成套有限公司 | Machine learning-based direct-current submerged arc furnace abnormal event early warning system |
| WO2021097604A1 (en) * | 2019-11-18 | 2021-05-27 | 株洲中车时代电气股份有限公司 | Multi-information fusion-based fault early warning method and device for converter |
-
2021
- 2021-06-28 CN CN202110716311.8A patent/CN113686922A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1184931A (en) * | 1996-12-11 | 1998-06-17 | 唐秀家 | Method and apparatus for detecting and positioning leakage of fluid transferring pipeline |
| CN101096006A (en) * | 2007-05-14 | 2008-01-02 | 美晨集团股份有限公司 | Material-feeding automatic detecting device and the chemical producing system |
| CN105424293A (en) * | 2015-12-02 | 2016-03-23 | 深圳凌水环保科技股份有限公司 | Water supply-drainage pipe leak detection system and detection method |
| CN205656230U (en) * | 2016-04-21 | 2016-10-19 | 山东大地盐化集团有限公司 | Damaged online alarm device of enamel reactor |
| CN108443716A (en) * | 2018-02-06 | 2018-08-24 | 江阴市长龄机械制造有限公司 | A kind of fluid hose convenient for detection leakage point |
| WO2021097604A1 (en) * | 2019-11-18 | 2021-05-27 | 株洲中车时代电气股份有限公司 | Multi-information fusion-based fault early warning method and device for converter |
| CN111898669A (en) * | 2020-07-24 | 2020-11-06 | 大连重工机电设备成套有限公司 | Machine learning-based direct-current submerged arc furnace abnormal event early warning system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2433493C (en) | Low cost, on-line corrosion monitor and smart corrosion probe | |
| CN107024629B (en) | State detection and evaluation system and state evaluation method for power oil-less equipment | |
| CN102494990B (en) | Sensor for measuring corrosion of inner wall of metal tank body and monitoring method | |
| CN210487949U (en) | Oil-immersed high-voltage equipment oil-light-sound-spectrum on-line monitoring device | |
| CN101261213B (en) | Electrochemical test probe for erosion critical characteristic on-line monitoring | |
| CN1289038A (en) | Intelligent analysts system and method for electric equipment filled with fluid | |
| CN108051032A (en) | The on-line monitoring system of the few oily equipment of substation | |
| CN109490811A (en) | A kind of oil pressure monitoring method of inverted type current transformer | |
| WO2018142102A1 (en) | System and method for monitoring a flange joint assembly | |
| CN104914144A (en) | Hydrogen concentration on-line monitoring device used for nuclear power plants | |
| CN110579587A (en) | On-line monitoring system for lubricating oil quality in steam turbine | |
| CN113686922A (en) | Online intelligent monitoring device and method for risks of reaction kettle | |
| KR20110122238A (en) | Fault diagnosis method and apparatus using headspace gases for transformer | |
| CN110068670A (en) | A kind of micro- water of transformer insulation oil and hydrogen gas system detection device and its detection method | |
| CN208847287U (en) | Condenser type adaptive universal liquid level emasuring device | |
| CN110057489A (en) | A kind of power transformer on-Line Monitor Device and method based on transient state oil pressure feature | |
| CN109870610A (en) | A kind of USB long-term safety detection method of warm and humid biasing salt pollution environment | |
| CN112904130B (en) | Online monitoring method, device, equipment, medium and system for line aging | |
| CN109269975A (en) | Underwater equipment corrosion potential data detection system | |
| CN201199222Y (en) | Electrochemistry test probe for on-line monitoring erosion critical characteristics | |
| CN113932949A (en) | Function detection probe and device for electric energy meter terminal temperature sensor | |
| CN115047037A (en) | Portable detection device and detection method based on microfluidic gas sensor | |
| CN202661588U (en) | On-line monitoring system of zinc oxide arrester | |
| CN210166340U (en) | Experimental device for realize ion concentration and real-time normal position monitoring of electrochemistry parameter | |
| CN112255367A (en) | Chemical instrument detection system and method |
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 |