CN204390253U - The site test system of Based PC CP pipeline acoustical signal - Google Patents
The site test system of Based PC CP pipeline acoustical signal Download PDFInfo
- Publication number
- CN204390253U CN204390253U CN201420865505.XU CN201420865505U CN204390253U CN 204390253 U CN204390253 U CN 204390253U CN 201420865505 U CN201420865505 U CN 201420865505U CN 204390253 U CN204390253 U CN 204390253U
- Authority
- CN
- China
- Prior art keywords
- pipeline
- signal
- acoustical signal
- receiving set
- nautical receiving
- 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
Abstract
A kind of site test system of Based PC CP pipeline acoustical signal.System comprises nautical receiving set, pretreater, data acquisition unit and primary processor; Nautical receiving set is connected with primary processor with data acquisition unit by pretreater successively.The site test system of the Based PC CP pipeline acoustical signal that the utility model provides achieves collection to acoustical signal in PCCP pipeline, process and feature identification, in the process that field experiment data are processed, mainly have employed wavelet transformation and carry out acoustical signal feature extraction, and the feature identification of acoustical signal is carried out by support vector machine, by field experiment and data analysis, the resolution accuracy of this system to local flaw signal can reach 98.33%.
Description
Technical field
The utility model belongs to Computer Control Technology field, particularly relates to a kind of site test system of Based PC CP pipeline acoustical signal.
Background technology
The production of Prestressed concrete cylinder pipe and PCCP pipeline and use have very long history.To design the earliest and what manufacture Prestressed concrete cylinder pipe is French Bang Na pipeline company.To the forties in 20th century, American-European countries also starts numerous and confused exploitation and manufactures Prestressed concrete cylinder pipe.The U.S. produces as Prestressed concrete cylinder pipe maximum in the world and uses country, and the length to the modern Prestressed concrete cylinder pipe pipeline used reaches 28000KM, and wherein maximum Prestressed concrete cylinder pipe road diameter reaches 7.6 meters.
The main method of traditional Prestressed concrete cylinder pipe protection is hand inspection, as listening methods and eye-observation method.Prestressed concrete cylinder pipe fault detect also have far-field eddy/transformer coupled (RFEC/TC), based on the method such as Acoustic detection harmony fiber laser arrays (AFO) of nautical receiving set.RFEC/TC method needs the water in emptying pipe before detection, therefore can consume a large amount of man power and materials.In addition, RFEC/TC method is a kind of off-line check method, and efficiency is lower.The fine detection method of acousto-optic is a kind of real-time detection method of pin-point accuracy, but the method cost is high, and is difficult in in-service Prestressed concrete cylinder pipe road install acousto-optic fibre again, can only be applied to newly-built pipeline.Because nautical receiving set is highly sensitive to detection acoustical signal, and only need in pipeline, lay a nautical receiving set at a certain distance, so not only accuracy of detection is high but also easy for installation based on the detection method of nautical receiving set.The method has obvious advantage relative to other several detection methods, and in Prestressed concrete cylinder pipe road, damaged context of detection has application prospect very widely.
For the situation of prior art, the utility model people is after having made a large amount of early-stage Study to PCCP, propose the PCCP yarn break inspect system based on nautical receiving set, its principle detects the acoustical signal real-time online of PCCP pipeline, identify local flaw signal wherein, utilize local flaw signal to realize the early warning of PCCP pipeline burst; Monitor and identify local flaw signal, inadequate when only leaning against the simulation test of carrying out in laboratory, site test on the spot must be carried out at the scene of PCCP pipeline, but also there is no the site test system of the Based PC CP pipeline acoustical signal of comparatively maturation at present.
Summary of the invention
In order to solve the problem, the purpose of this utility model is the site test system providing a kind of Based PC CP pipeline acoustical signal.
In order to achieve the above object, the site test system of Based PC CP pipeline acoustical signal that the utility model provides comprises: nautical receiving set, pretreater, data acquisition unit and primary processor; Wherein: nautical receiving set is connected with primary processor with data acquisition unit by pretreater successively; Nautical receiving set is be fixedly mounted in PCCP pipeline by the flange at PCCP pipe joint place, for gathering the device of acoustical signal in pipeline; Pretreater is analog signal conditioner circuit, turns voltage, signal amplifies and filter function for realizing electric charge; It is primarily of charge amplifier, signal amplification circuit and bandpass filter three functional module compositions; Data acquisition unit is analog-digital commutator, for gathering simulating signal and converting thereof into digital signal; Primary processor is the arithmetic unit with storer, for realizing the storage of image data and follow-up data process and computing.
The RHS series Sphere Nominal nautical receiving set that described nautical receiving set selects Hangzhou acoustics to produce, model is RHS20A; The USB4432 data collecting card that described data acquisition unit adopts NI company to produce; Described primary processor adopts a portable computer.
Described primary processor is connected with ppu or external system by data network.
Described data acquisition unit is multi-channel data acquisition unit, and it connects multiple nautical receiving set by hyperchannel pretreater, and multiple nautical receiving set lays with a determining deviation at PCCP pipe interior, forms hydrophone, group.
The site test system of the Based PC CP pipeline acoustical signal that the utility model provides achieves collection to acoustical signal in PCCP pipeline, process and feature identification, in the process that field experiment data are processed, mainly have employed wavelet transformation and carry out acoustical signal feature extraction, and the feature identification of acoustical signal is carried out by support vector machine, by field experiment and data analysis, the resolution accuracy of this system to local flaw signal can reach 98.33%.
Accompanying drawing explanation
The composition schematic diagram of the site test system of the Based PC CP pipeline acoustical signal that Fig. 1 provides for the utility model.
The process flow diagram of the data analysing method that the site test system of the Based PC CP pipeline acoustical signal that Fig. 2 provides for the utility model adopts.
Fig. 3 is the time domain beamformer of four kinds of acoustical signals main in water-filled pipe.
Fig. 4 is 120 sample characteristics scatter diagrams.
Embodiment
Be described in detail below in conjunction with the site test system of the drawings and specific embodiments to the Based PC CP pipeline acoustical signal that the utility model provides.
As shown in Figure 1, the site test system of Based PC CP pipeline acoustical signal that the utility model provides comprises:
Nautical receiving set 1, pretreater 2, data acquisition unit 3 and primary processor 4; Wherein: nautical receiving set 1 is connected with primary processor 4 with data acquisition unit 3 by pretreater 2 successively; Nautical receiving set 1 is be fixedly mounted in PCCP pipeline 5 by the flange of the interface of PCCP pipeline 5, for gathering ducted acoustical signal; Pretreater 2 is analog signal conditioner circuit, turns voltage, signal amplifies and filter function for realizing electric charge; It is primarily of charge amplifier, signal amplification circuit and bandpass filter three functional module compositions; Data acquisition unit 3 is analog-digital commutator, for gathering simulating signal and converting thereof into digital signal; Primary processor 4 for having the arithmetic unit of storer, for realizing the storage of image data and follow-up data process and computing.
Native system detects acoustical signal mainly through nautical receiving set 1, the frequency range of general acoustical signal is 0-20KHZ, therefore the frequency range considering acoustical signal is needed when selecting nautical receiving set 1, in native system, the RHS series Sphere Nominal nautical receiving set that described nautical receiving set 1 selects Hangzhou acoustics to produce, model is RHS20A;
The USB4432 data collecting card that described data acquisition unit 3 adopts NI company to produce, its maximum sampling rate is 102.4kS/s, in actual measurement process, the frequency range of acoustical signal is 0-20KHZ, because Least sampling rate should be 40K, USB4431 data collecting card can meet the requirement of this sampling rate, and in actual samples process, the sampling rate of use is 44K.
Described primary processor 4 adopts a portable computer.
Described primary processor 4 is connected with ppu or external system by data network.
Described data acquisition unit 3 is multi-channel data acquisition unit, it connects multiple nautical receiving set 1 by hyperchannel pretreater 2, multiple nautical receiving set 1 lays with a determining deviation in PCCP pipeline 5 inside, forms hydrophone, group, and it can realize reception to PCCP pipe fracture of wire acoustical signal and early warning better.
As shown in Figure 2, the data analysing method that the site test system of Based PC CP pipeline acoustical signal that the utility model provides adopts comprises the following step performed in order:
Step 1) receive S01 stage of demblee form acoustical signal: primary processor 4 gathers by data acquisition unit 3 acoustical signal that nautical receiving set 1 receives, and therefrom extracts demblee form acoustical signal as input data;
Step 2) S02 stage of data prediction: filtering and noise reduction process are carried out to above-mentioned input data;
Step 3) wavelet character S03 stage of extracting: by wavelet decomposition, extract signal at the energy of different frequency section as its proper vector, form sample;
Step 4) S04 stage of selecting of support vector machine (SVM) optimized parameter: select optimum supporting vector machine model parameter;
Step 5) S05 stage of svm classifier: utilize supporting vector machine model to classify to above-mentioned input sample of data;
Step 6) determine whether S06 stage of local flaw signal: the result according to classification judges whether input sample of data belongs to the type of local flaw signal, if judged result is "Yes", then next step enters the S07 stage, otherwise enters next step S08 stage;
Step 7) system sends S07 stage of early warning signal: system sends early warning signal instruction, terminates this test;
Step 8) S08 stage of system worked well: system cloud gray model is normal, terminates this test.
In step 2) to step 8) in, described computing and operation all complete having in calculation function primary processor 4, or complete on the ppu be connected with primary processor 4.
In step 2) in, described data prediction comprises: filtering and wavelet de-noising;
Described filtering can arrange low-frequency cut-off frequency and high-frequency cut-off frequency, and what wave filter was here selected is Butterworth filter, and the exponent number of its median filter also can Lookup protocol;
Described wavelet de-noising also can realize carrying out filtering to signal, and in processing procedure, the type of small echo and the number of plies of process can be arranged; Meanwhile, small echo can also be utilized to carry out feature extraction to signal; The cardinal principle that small echo carries out feature extraction to signal is the time frequency analysis function that make use of small echo, and the feature of signal at time domain and frequency domain place can be segmented by wavelet analysis; What commonly use in wavelet analysis is the layering carrying out signal with wavelet packet, the frequency of different frequency-domain segment is divided into by signal, the component of signal of each frequency band can a feature of characterization signal, the signal of different frequency section is carried out feature extraction as the proper vector of this signal, finally can obtain the proper vector of this signal.
Wavelet and wavelet packets decomposition and reconstruction, due to the time frequency analysis characteristic that it is excellent, has a wide range of applications in the feature extraction of signal; In order to fracture of wire acoustical signal and other undesired signals being distinguished, often kind of signal feature separately just must be extracted to carry out Signal analysis; What the utility model adopted is that db6 small echo carries out 8 layers of wavelet decomposition to signal, finally obtains the signal energy normalization result under 9 different frequency bands; Using the proper vector of the result of this energy normalized as each signal.
As shown in Figure 3, nautical receiving set 1 water-filling PCCP pipeline 5 can in the four class acoustical signals that receive, Fig. 3 shows the time domain waveform of these four kinds of acoustical signals, and wherein fracture of wire is the principal mode of PCCP tube failure, therefore can realize the early warning to PCCP pipe to the accurate judgement of PCCP pipe fracture of wire acoustical signal; The nautical receiving set (group) laid with a determining deviation in PCCP pipeline 5 inside can realize reception to PCCP pipe fracture of wire acoustical signal and early warning; But, in the real-time monitoring system of reality, nautical receiving set 1 is except receiving local flaw signal, also can receive other undesired signals: the sound of the inside swelling produced due to pressure instability as inner in PCCP pipe, the knocking of artificial construction or interference on the sound that tube circumference people walks and duct wall; These signals all may be aliasing in real fracture of wire acoustical signal and make local flaw signal be difficult to identify in the monitoring of actual PCCP pipeline 5; These signals are the same with local flaw signal belongs to demblee form signal, may be identified by system by mistake.
In step 3) in order to real local flaw signal and other undesired signals effectively can be distinguished, from frequency domain, power feature extraction can be carried out to these signals; Because a base attribute of frequency domain character originally acoustical signal; Its frequency energy feature of dissimilar acoustical signal is also different; Compared with traditional Fourier transform, wavelet transformation has a lot of advantage, and signal can be carried out refinement analysis in the local of time (space) frequency by it; This multiscale analysis to signal is realized by flexible shift operations, signal can be made after refinement to realize time subdivision at high frequency treatment, in the segmentation of low frequency place frequency, thus automatically can meet the demand of time frequency signal analysis, and any details of signal can be focused on; Wavelet transformation is particularly useful for astable type signal; For PCCP early warning system, local flaw signal and other undesired signals all belong to the astable signal of demblee form, by wavelet decomposition, can extract the energy of signal in different frequency section as its proper vector.
In step 4) in, the parameter of described supporting vector machine model is penalty parameter c and the kernel functional parameter g of support vector machine:
Support vector machine is exactly first by the nonlinear transformation defined by inner product kernel function, the input space is transformed to a higher dimensional space, and then in this space, ask the sorting technique of (broad sense) optimal classification surface; The optimum configurations of support vector machine have impact on the performance of support vector machine greatly, and its major parameter has: penalty parameter c and kernel functional parameter g; Optimum penalty parameter c and kernel functional parameter g is how selected to be the difficult points that support vector machine is optimized; In the utility model, selected K to divide the method for cross validation (K-CV), its main thought is that raw data is divided into k group, each subset is selected to make checking collection respectively, remaining does training set, the classification accuracy of k group checking collection can be obtained like this, the performance index as differentiating are averaging to it; Penalty parameter c and kernel functional parameter g are increased progressively value successively, calculates the result of each K-CV respectively, just think that when obtaining best K-CV value penalty parameter c now and kernel functional parameter g are best, supporting vector machine model is now optimum.
In step 5) in, before carrying out svm classifier, first to carry out feature extraction to signal; Get the test sample book of 120 samples as SVM, in these 120 samples, 1-30 is fracture of wire acoustical signal, and 31-60 is for gushing acoustical signal, and 61-90 is knock signal, and 91-120 to walk acoustical signal for people.
Fig. 4 is the feature extraction result of 120 sample datas, and the data difference of db3, db5, db6 and ca8 frequency range is comparatively obvious as we can see from the figure, and effective feature extraction is that Data classification provides good foundation.
Claims (4)
1. a site test system for Based PC CP pipeline acoustical signal, is characterized in that: the site test system of described Based PC CP pipeline acoustical signal comprises:
Nautical receiving set (1), pretreater (2), data acquisition unit (3) and primary processor (4); Wherein: nautical receiving set (1) is connected with primary processor (4) with data acquisition unit (3) by pretreater (2) successively; Nautical receiving set (1) for be fixedly mounted in PCCP pipeline (5) by the flange of PCCP pipeline (5) interface, for gathering the device of acoustical signal in pipeline; Pretreater (2) is analog signal conditioner circuit, turns voltage, signal amplifies and filter function for realizing electric charge; It is primarily of charge amplifier, signal amplification circuit and bandpass filter three functional module compositions; Data acquisition unit (3) is analog-digital commutator, for gathering simulating signal and converting thereof into digital signal; Primary processor (4) for having the arithmetic unit of storer, for realizing the storage of image data and follow-up data process and computing.
2. the site test system of Based PC CP pipeline acoustical signal according to claim 1, is characterized in that: the RHS series Sphere Nominal nautical receiving set that described nautical receiving set (1) selects Hangzhou acoustics to produce, and model is RHS20A; The USB4432 data collecting card that described data acquisition unit (3) adopts NI company to produce; Described primary processor (4) adopts a portable computer.
3. the site test system of Based PC CP pipeline acoustical signal according to claim 1, is characterized in that: described primary processor (4) is connected with ppu or external system by data network.
4. the site test system of Based PC CP pipeline acoustical signal according to claim 1, it is characterized in that: (3 is multi-channel data acquisition unit to described data acquisition unit, it connects multiple nautical receiving set (1) by hyperchannel pretreater (2), multiple nautical receiving set (1) lays with a determining deviation in PCCP pipeline (5) inside, forms hydrophone, group.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201420865505.XU CN204390253U (en) | 2014-12-31 | 2014-12-31 | The site test system of Based PC CP pipeline acoustical signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201420865505.XU CN204390253U (en) | 2014-12-31 | 2014-12-31 | The site test system of Based PC CP pipeline acoustical signal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN204390253U true CN204390253U (en) | 2015-06-10 |
Family
ID=53362955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201420865505.XU Expired - Fee Related CN204390253U (en) | 2014-12-31 | 2014-12-31 | The site test system of Based PC CP pipeline acoustical signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN204390253U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107355687A (en) * | 2017-06-09 | 2017-11-17 | 昆明理工大学 | A kind of sewer pipe fault detection method |
-
2014
- 2014-12-31 CN CN201420865505.XU patent/CN204390253U/en not_active Expired - Fee Related
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107355687A (en) * | 2017-06-09 | 2017-11-17 | 昆明理工大学 | A kind of sewer pipe fault detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109827081B (en) | Buried drainage pipeline blocking fault and pipeline tee part diagnosis method based on acoustic active detection | |
| CN108648764B (en) | Rainfall measurement system based on rainwater knocking sound identification and measurement method thereof | |
| CN104819766A (en) | Envelope demodulation frequency band determination method based on harmonic-to-noise ratio | |
| CN103776903B (en) | A kind of wind electricity blade delamination detection method and detection system | |
| CN107314251B (en) | A kind of detection device and detection method of sewer pipe leakage failure | |
| CN102879714A (en) | Detection and positioning method for partial discharge of transformer | |
| CN109063762A (en) | A kind of line clogging fault recognition method based on DT-CWT and S4VM | |
| CN104089699A (en) | Substation equipment sound reconstruction algorithm | |
| CN103637821A (en) | Hepatic fibrosis degree Fisher identification method based on ultrasonic radio frequency (RF) time sequence | |
| CN110568073B (en) | Method for picking up impact signal in noise environment | |
| CN104568132B (en) | Reference signal constraint-based mechanical characteristic acoustic signal frequency-domain semi-blind extraction method | |
| CN112856248A (en) | Underground pressure pipeline leakage detection method, device and system and storage medium | |
| CN110057918B (en) | Method and system for quantitatively identifying damage of composite material under strong noise background | |
| CN108181059A (en) | Multiphase flow pipeline leakage acoustic signals recognition methods based on small echo signal | |
| CN105909979B (en) | Leakage acoustic characteristic extracting method based on Wavelet Transform Fusion blind source separation algorithm | |
| CN104504380A (en) | Onsite testing system based on PCCP pipeline acoustic signal and data analyzing method thereof | |
| CN105067048A (en) | Vibration interference resisting measuring method of digitalized vortex shedding flowmeter | |
| CN114721037A (en) | Method for detecting, positioning and identifying characteristics in pipeline by intelligent acoustic technology | |
| CN204390253U (en) | The site test system of Based PC CP pipeline acoustical signal | |
| CN102879715A (en) | Partial discharge detecting and positioning method for transformer | |
| CN111735593B (en) | Method for tracking, testing and analyzing vibration frequency of water turbine model machine | |
| CN105139852A (en) | Engineering machinery recognition method and recognition device based on improved MFCC (Mel Frequency Cepstrum Coefficient) sound features | |
| CN104633461A (en) | PCCP pipe failure forewarning and monitoring signal processing method based on sound signals | |
| CN105890738A (en) | Conflux vortex impact vibration identification method | |
| CN113484417B (en) | Pipeline corrosion detection method based on wavelet transformation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150610 Termination date: 20191231 |