CN105673909B - A kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal - Google Patents
A kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal Download PDFInfo
- Publication number
- CN105673909B CN105673909B CN201610224004.7A CN201610224004A CN105673909B CN 105673909 B CN105673909 B CN 105673909B CN 201610224004 A CN201610224004 A CN 201610224004A CN 105673909 B CN105673909 B CN 105673909B
- Authority
- CN
- China
- Prior art keywords
- control method
- data
- pressure signal
- frequency pressure
- air bleeding
- 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.)
- Expired - Fee Related
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16K—VALVES; TAPS; COCKS; ACTUATING-FLOATS; DEVICES FOR VENTING OR AERATING
- F16K31/00—Actuating devices; Operating means; Releasing devices
- F16K31/02—Actuating devices; Operating means; Releasing devices electric; magnetic
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Control Of Fluid Pressure (AREA)
Abstract
The present invention proposes a kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal, belongs to air bleeding valve control technology field.Information corresponding to the high-frequency pressure signal collected is carried out feature extraction by the control method;Then, the pressure signal for completing feature extraction is classified using the support vector cassification model trained, judging that water delivery declines in pipeline according to classification results whether there is air bag, and start degassing function.The active control system includes automatic inlet-outlet valve, high-frequency pressure sensor, wireless-transmission network and data processing centre;Automatic inlet-outlet valve 1 is fixedly installed on aqueduct, and high-frequency pressure sensor is fixedly installed in water delivery and declined on pipeline;Data processing centre is controlled automatic inlet-outlet valve in real time by wireless-transmission network;High-frequency pressure sensor carries out information transfer by wireless-transmission network and data processing centre.This method and system have the features such as monitoring precision is high, and exhaust efficiency is high.
Description
Technical field
The present invention relates to a kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal, belong to air bleeding valve control
Technical field processed.
Background technology
At present, operation principle of the automatic inlet-outlet valve commonly used on aqueduct in exhaust is, when in aqueduct
Air accumulation at the top of air bleeding valve, in valve body bubble accumulation make ball float (cylinder) with water level decreasing, so as to open exhaust pition,
After gas drains, water level rises, and ball float (cylinder) rises therewith, closes exhaust pition.This operation principle needs the sky in valve body
Gas is accumulated to a certain amount, and air bleeding valve could work, and now air bleeding valve nearby has a large amount of gases;Furthermore because air bleeding valve is pacified
The particularity of holding position, it has not been convenient to which regular maintenance, many air bleeding valves have lost due function in Practical Project.Therefore, need badly
Develop a kind of air bleeding valve active control technology for gas discharge in aqueduct.
The content of the invention
To solve technical problem present in above-mentioned prior art, the present invention proposes a kind of based on high-frequency pressure signal
Air bleeding valve Active Control Method and system, the technical scheme taken are as follows:
Information corresponding to the high-frequency pressure signal collected is carried out feature extraction by the control method;Then, instruction is utilized
The support vector cassification model perfected is classified to the pressure signal for completing feature extraction, judges water delivery according to classification results
Decline in pipeline and whether there is air bag, and start degassing function.
Preferably, the step of control method is:
Step 1: the principle trained according to SVMs, is divided into training number by the high-frequency pressure initial data collected
According to and test data;
Step 2: resampling and filtering process are carried out to training data described in step 1 and test data;
Step 3: the training data after resampling and filtering process and test data are marked;
Step 4: after to the training data and the short-time zero-crossing rate of test data, variance, Hilbert-Huang transform after mark
This four features of the comentropy of different frequency bands are extracted after the Energy distribution and WAVELET PACKET DECOMPOSITION of different frequency bands;
Step 5: the pressure signal for completing feature extraction is divided using the support vector cassification model trained
Class;
Step 6: judging that water delivery declines in pipeline according to support vector cassification result whether there is air bag, and start row
Airway dysfunction.
Preferably, training data described in step 1 occupies the 75% of initial data;The test data occupies original number
According to 25%.
Preferably, filtering process described in step 2 is using two methods of Wavelet Denoising Method and simple smooth filtering.
Preferably, the Wavelet noise-eliminating method selects the hard -threshold related to the data length of pressure signal to handle small echo
Coefficient, and training data and test data are handled by the wavelet coefficient after the reconstruct of Mallat algorithms.
Preferably, support vector cassification model described in step 5 is using the vector classification of C-SVC types, and uses radial direction base
Function is as kernel function.
Preferably, the air bleeding valve Active Control Method comprises the following steps that:
Step 1: the principle trained according to SVMs, is divided into training number by the high-frequency pressure initial data collected
According to and test data, wherein, the training data occupies the 75% of initial data;The test data occupies initial data
25%;
Step 2: resampling and filtering process are carried out to training data described in step 1 and test data;Wherein, the filter
Ripple processing is using two methods of Wavelet Denoising Method and simple smooth filtering;Meanwhile the Wavelet noise-eliminating method selection and pressure signal
The related hard -threshold of data length handle wavelet coefficient, and by training data and test data by the reconstruct of Mallat algorithms after
Wavelet coefficient handled;
Step 3: the training data after resampling and filtering process and test data are marked;
Step 4: after to the training data and the short-time zero-crossing rate of test data, variance, Hilbert-Huang transform after mark
This four features of the comentropy of different frequency bands are extracted after the Energy distribution and WAVELET PACKET DECOMPOSITION of different frequency bands;
Step 5: the pressure signal for completing feature extraction is divided using the support vector cassification model trained
Class;The support vector cassification model uses RBF as kernel function using the vector classification of C-SVC types;
Step 6: judging that water delivery declines in pipeline according to support vector cassification result whether there is air bag, and start row
Airway dysfunction.
Preferably, the system includes automatic inlet-outlet valve 1, high-frequency pressure sensor 2, wireless-transmission network 3 and data
Processing center 4;The automatic inlet-outlet valve 1 is fixedly installed on aqueduct, and the high-frequency pressure sensor 2 is fixedly installed in
Water delivery declines on pipeline;The data processing centre 4 is controlled automatic inlet-outlet valve 1 in real time by wireless-transmission network 3;
The high-frequency pressure sensor 2 carries out information transfer by wireless-transmission network 3 and data processing centre 4.
Beneficial effect of the present invention:
When air is largely gathered near air bleeding valve, the degassing function with regard to air bleeding valve can be started, and can examine automatic
The degassing function of air inlet and exhaust valve whether there is;Verified by experimental data, in aqueduct down-comer, by above-described
The rate of accuracy reached of successfully identification to air bag in pipeline be present to 95% in method.For Practical Project, if increase data sample, can enter
One step improves the success rate of identification.
Brief description of the drawings
Fig. 1 is air bleeding valve Active Control Method flow chart of the present invention;
Fig. 2 is air bleeding valve active control system structural representation of the present invention;
(1, automatic inlet-outlet valve;2, high-frequency pressure sensor;3, wireless-transmission network;4, data processing centre).
Embodiment
With reference to specific embodiment, the present invention will be further described, but the present invention should not be limited by the examples.
The principle of air bleeding valve Active Control Method and system proposed by the present invention based on high-frequency pressure signal is:Water delivery
Air is detained accumulation due to buoyancy typically in pipeline is declined and forms air bag in pipeline, therefore is declining duct section air bleeding valve
One high-frequency pressure sensor of installation nearby, regularly passes through wireless network transmissions to Data processing by the pressure signal collected
The heart 4, data processing centre 4 are analyzed pressure signal to obtain the content situation of gas in pipelines under the pressure signal state,
If gas content has formed air bag in result display pipes, data processing centre 4 is signaled to air bleeding valve and indicates that it is opened
Open degassing function;During exhaust, continuous acquisition high-frequency pressure signal simultaneously still extracts tagsort, until point of pressure signal
Gas returns to acceptable tolerance state in class result display pipes.
Fig. 1 is the flow chart of air bleeding valve active of the present invention, the high-frequency pressure signal pair that the control method will collect
The information answered carries out feature extraction;Then, the pressure using the support vector cassification model trained to completion feature extraction
Signal is classified, and judging that water delivery declines in pipeline according to classification results whether there is air bag, and start degassing function.
Control method concretely comprises the following steps:
Step 1: the principle trained according to SVMs, is divided into training number by the high-frequency pressure initial data collected
According to and test data;
Step 2: resampling and filtering process are carried out to training data described in step 1 and test data;
Step 3: the training data after resampling and filtering process and test data are marked;
Step 4: after to the training data and the short-time zero-crossing rate of test data, variance, Hilbert-Huang transform after mark
This four features of the comentropy of different frequency bands are extracted after the Energy distribution and WAVELET PACKET DECOMPOSITION of different frequency bands;
Step 5: the pressure signal for completing feature extraction is divided using the support vector cassification model trained
Class;
Step 6: judging that water delivery declines in pipeline according to support vector cassification result whether there is air bag, and start row
Airway dysfunction, meanwhile, in the presence of having air bag, signal is sent by data processing centre 4 air bleeding valve is indicated, start exhaust work(
Can, the still continuous collecting high-frequency pressure signal, and the process for continuing to classify according to feature extraction, fluidised form is to pipeline during exhaust
In tolerance state be identified, untill there is no air bag in fluidised form classification results display pipes, close degassing function.
Wherein, the training data mentioned in step 1 occupies the 75% of initial data;The test data occupies original number
According to 25%;The filtering process mentioned in step 2 is using two methods of Wavelet Denoising Method and simple smooth filtering;Wavelet Denoising Method side
Method selects the hard -threshold related to the data length of pressure signal to handle wavelet coefficient, and training data and test data are pressed
Wavelet coefficient after the reconstruct of Mallat algorithms is handled.The support vector cassification model mentioned in step 5 uses C-SVC
Type vector classification, and using RBF as kernel function.
Fig. 2 is air bleeding valve active control system structural representation of the present invention, should the exhaust based on high-frequency pressure signal
Valve active control system includes automatic inlet-outlet valve 1, high-frequency pressure sensor 2, wireless-transmission network 3 and data processing centre 4;
Automatic inlet-outlet valve 1 is fixedly installed on aqueduct, and high-frequency pressure sensor 2 is fixedly installed in water delivery and declined on pipeline;Number
Automatic inlet-outlet valve 1 is controlled in real time by wireless-transmission network 3 according to processing center 4;High-frequency pressure sensor 2 passes through nothing
Transmission network network 3 carries out information transfer with data processing centre 4.
Although the present invention is disclosed as above with preferred embodiment, it is not limited to the present invention, any to be familiar with this
The people of technology, without departing from the spirit and scope of the present invention, various changes and modification, therefore the protection of the present invention can be done
What scope should be defined by claims is defined.
Claims (7)
1. a kind of air bleeding valve Active Control Method based on high-frequency pressure signal, it is characterised in that the control method is to adopt
Information corresponding to the high-frequency pressure signal collected carries out feature extraction;Then, the support vector cassification model trained is utilized
The pressure signal for completing feature extraction is classified, judging that water delivery declines in pipeline according to classification results whether there is air bag,
In the presence of water delivery declines and has air bag in pipeline, start degassing function.
2. air bleeding valve Active Control Method according to claim 1, it is characterised in that as follows the step of the control method:
Step 1: according to SVMs train principle, by the high-frequency pressure initial data collected be divided into training data and
Test data;
Step 2: resampling and filtering process are carried out to training data described in step 1 and test data;
Step 3: the training data after resampling and filtering process and test data are marked;
Step 4: to different after the training data and the short-time zero-crossing rate of test data, variance, Hilbert-Huang transform after mark
This four features of the comentropy of different frequency bands are extracted after the Energy distribution and WAVELET PACKET DECOMPOSITION of frequency band;
Step 5: the pressure signal for completing feature extraction is classified using the support vector cassification model trained;
Step 6: judging that water delivery declines in pipeline according to support vector cassification result whether there is air bag, when water delivery down-comer
In the presence of having air bag in road, start degassing function.
3. air bleeding valve Active Control Method according to claim 2, it is characterised in that training data occupies described in step 1
The 75% of initial data;The test data occupies the 25% of initial data.
4. air bleeding valve Active Control Method according to claim 2, it is characterised in that filtering process described in step 2 uses
Simple smooth filtering method.
5. air bleeding valve Active Control Method according to claim 2, it is characterised in that support vector cassification described in step 5
Model uses RBF as kernel function using the vector classification of C-SVC types.
6. air bleeding valve Active Control Method according to claim 1, it is characterised in that comprise the following steps that:
Step 1: according to SVMs train principle, by the high-frequency pressure initial data collected be divided into training data and
Test data, wherein, the training data occupies the 75% of initial data;The test data occupies the 25% of initial data;
Step 2: resampling and filtering process are carried out to training data described in step 1 and test data;Wherein, at the filtering
Reason is using two methods of Wavelet Denoising Method and simple smooth filtering;Meanwhile the Wavelet noise-eliminating method is using the number with pressure signal
Wavelet coefficient is handled according to the related hard -threshold of length, and by training data and test data by small after the reconstruct of Mallat algorithms
Wave system number is handled;
Step 3: the training data after resampling and filtering process and test data are marked;
Step 4: to different after the training data and the short-time zero-crossing rate of test data, variance, Hilbert-Huang transform after mark
This four features of the comentropy of different frequency bands are extracted after the Energy distribution and WAVELET PACKET DECOMPOSITION of frequency band;
Step 5: the pressure signal for completing feature extraction is classified using the support vector cassification model trained;Institute
Support vector cassification model is stated using the vector classification of C-SVC types, and using RBF as kernel function;
Step 6: judging that water delivery declines in pipeline according to support vector cassification result whether there is air bag, when water delivery down-comer
In the presence of having air bag in road, start degassing function.
A kind of 7. system for realizing air bleeding valve Active Control Method described in claim 1, it is characterised in that the system bag
Include automatic inlet-outlet valve (1), high-frequency pressure sensor (2), wireless-transmission network (3) and data processing centre (4);It is described automatic
Air inlet and exhaust valve (1) is fixedly installed on aqueduct, and the high-frequency pressure sensor (2) is fixedly installed in water delivery and declines pipeline
On;The data processing centre (4) is controlled automatic inlet-outlet valve (1) in real time by wireless-transmission network (3);The height
Frequency pressure sensor (2) carries out information transfer by wireless-transmission network (3) and data processing centre (4).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610224004.7A CN105673909B (en) | 2016-04-12 | 2016-04-12 | A kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610224004.7A CN105673909B (en) | 2016-04-12 | 2016-04-12 | A kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105673909A CN105673909A (en) | 2016-06-15 |
CN105673909B true CN105673909B (en) | 2018-02-06 |
Family
ID=56309732
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610224004.7A Expired - Fee Related CN105673909B (en) | 2016-04-12 | 2016-04-12 | A kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105673909B (en) |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201643793U (en) * | 2009-12-04 | 2010-11-24 | 北京卫星环境工程研究所 | Gas-liquid separating system for liquid nitrogen |
CN102135186B (en) * | 2010-01-21 | 2013-01-02 | 天津市国威给排水设备制造有限公司 | Anti-theft intelligent air release valve |
CN201891929U (en) * | 2010-12-17 | 2011-07-06 | 宁波华平金属制品有限公司 | Automatic air valve of pipeline |
CN202561203U (en) * | 2012-04-17 | 2012-11-28 | 陈文萍 | Fine bubble pipeline system |
JP6225817B2 (en) * | 2014-04-28 | 2017-11-08 | 横河電機株式会社 | Valve remote control device |
-
2016
- 2016-04-12 CN CN201610224004.7A patent/CN105673909B/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
CN105673909A (en) | 2016-06-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106653032B (en) | Based on the animal sounds detection method of multiband Energy distribution under low signal-to-noise ratio environment | |
CN106295588A (en) | The automatic identifying method of leukocyte in a kind of leucorrhea micro-image | |
CN104268883A (en) | Time-frequency spectrum curve extracting method based on edge detection | |
CN110008932A (en) | A kind of vehicle violation crimping detection method based on computer vision | |
CN105673909B (en) | A kind of air bleeding valve Active Control Method and system based on high-frequency pressure signal | |
CN113269720B (en) | Defect detection method, system and readable medium for straight welded pipe | |
CN206803263U (en) | A kind of oil smoke intelligence control system | |
CN105741855A (en) | Steward calling and responding detection method based on audio analysis | |
CN101656070B (en) | Voice detection method | |
CN104631611A (en) | Rainwater quality control regulation and storage system and control method | |
CN204227780U (en) | A kind of centrifugal oil separator | |
CN109614887A (en) | A kind of vehicle whistle classification method based on support vector machines | |
CN204125794U (en) | A kind of paper machine dryer ponding automatic discharge device | |
US20230184619A1 (en) | Underground pipe leak detection system and method | |
CN111652834A (en) | Steel coil strapping tape identification method based on support vector machine | |
CN216418758U (en) | Skid-mounted carbon dioxide trapping device suitable for offshore platform facility | |
CN202834714U (en) | Condensation water recovery device capable of avoiding water accumulation in exhaust pipe | |
CN211346401U (en) | Vacuum low-temperature dehydration equipment | |
CN201265495Y (en) | Combined pumping apparatus by floating water of water and pressure of running vehicle to road | |
CN204352739U (en) | A kind of gas analyzer cool-down dehumidification device | |
CN102886186A (en) | Automatic water and sewage draining device for air-water separation | |
CN109692788A (en) | Glue collection device | |
CN208152196U (en) | A kind of exhaust apparatus of bend pipe | |
CN206207016U (en) | A kind of gas safety drainage arrangement | |
CN104264525A (en) | Automatic early-warning and elimination system for water accumulation in drying cylinder of paper machine, and control method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180206 Termination date: 20210412 |
|
CF01 | Termination of patent right due to non-payment of annual fee |