CN104597124B - Metal canned product quality on-line detection method based on sound frequency-spectrum fitting - Google Patents
Metal canned product quality on-line detection method based on sound frequency-spectrum fitting Download PDFInfo
- Publication number
- CN104597124B CN104597124B CN201410790471.7A CN201410790471A CN104597124B CN 104597124 B CN104597124 B CN 104597124B CN 201410790471 A CN201410790471 A CN 201410790471A CN 104597124 B CN104597124 B CN 104597124B
- Authority
- CN
- China
- Prior art keywords
- product
- acoustical signal
- spectrum
- qualified
- products
- 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
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a metal canned product quality on-line detection method based on sound frequency-spectrum fitting. The method comprises the following steps: applying excitation to the detected metal canned product to vibrate the product; acquiring the sound generated by the vibrations by using a microphone, sampling the acquired sound signal to obtain a digitized sound signal; processing the sound signal to obtain a sound signal frequency spectrum; carrying out curve fitting on the sound signal frequency spectrum to acquire a curve function and a coefficient set thereof {C}; normalizing {C} to obtain a new coefficient set {C'}; using {C'} as the input parameter of the product quality computation model to acquire the model operation result product quality R; computing the absolute value |D| of the difference between R and qualified product standard value delta; and only when |D| is less than the threshold T, determining that the detected product is qualified. The method can implement on-line high-speed detection on the metal canned product, effectively avoids the influence of the production technique difference of metal cans and the changes in the distance between the microphone probe and metal can on the detection result, and obviously enhances the detection accuracy.
Description
Technical field
The present invention relates to metal canned product quality detection field, more specifically, it is to be related to one kind to be based on sound spectrum
The metal canned product quality online test method of matching.
Background technology
Substandard product not only results in same batch products and is forced to reclaim, and brings massive losses to producer, affects producer
Reputation, and be likely to also can cause a series of dangerous situations when using substandard product.Therefore research product quality inspection
Survey significant, substandard products can be detected before dispatching from the factory, it is to avoid enterprises' loss, guarantee consumer safety.Canister is typically used as
Container is packed to product, can play sealing, the effect such as attractive in appearance, is widely used in food service industry, canister unqualified
Likely result in food quality to go wrong, and underproof food is easily led to consumer sick or is poisoned once eating by mistake.
Although nearly all metal canned products will carry out quality testing before dispatching from the factory, because some products are only taken out
Inspection, or the reason such as the missing inspection during full inspection, flase drop, minority substandard products are not detected and flow to market.Therefore study accurate
Really efficient metal canned product quality online test method is necessary.
Currently, with respect to metal canned product quality online test method, in addition to acoustic method, also machine vision method, ultrasonic
Ripple method, electric vortex method, optical method (or claiming reflection method) etc..Machine vision method is generally used for detecting product appearance such as date spray
Code, producer Logo etc., or judge product quality by detecting canister profile, and just cannot enter for on-deformable tank
Row is correct to be detected, and accuracy of detection is easily subject to the such environmental effects such as product vibration, light；Supercritical ultrasonics technology is by detecting liquid in tank
Whether body has bubble all can cause by mistake to judge that in tank, whether food goes bad, and the scope of application is little, rocks, to vibrate the bubble causing
Inspection, accuracy rate is low, and detection process needs medium, is not suitable for high speed on-line checking；Electric vortex method is also by detecting canister
Plane was seized whether deform to judge product quality, have identical shortcoming with machine vision method, accuracy of detection is easily vibrated by product
Impact, and for different metal materials, detect that performance is also variant；Collimated light beam is projeced into metal tin cover by optical method
On, the direction of propagation difference of reflected light is made due to the degree of deformation difference of lid, the reflection light quantity that photoelectric conversion device receives is just
Can be variant, thus producing different photoelectric currents, then the size by photocell detection photoelectric current, the size according to photoelectric current is sentencing
The canned product quality of disconnected metal, but material is uneven and measurand vibration all can bring measurement error.
In sum, all also there is certain weak point in the canned product quality online test method of existing metal；Right
In qualified products of the same race, the natural frequency of their own is relatively；And compare qualified products of the same race and substandard product, they
Natural frequency have notable difference.Therefore, each product is applied with a suitable pumping signal and cause its vibration, due to intrinsic
Frequency is different, vibrates the sound sending and also differs, so the quality of product can be judged with acoustic method.
Gold-tinted week and Cai Dequan exist《Huanan's journal (natural science edition)》04 phase in 1987 delivers《Canned food
The Non-Destructive Testing of interior vacuum》In one literary composition, beam oscillating method is forced to apply excitation to can ends using the continuous scanning of multi-frequency, when strong
So that it may obtain the amplitude of maximum when the frequency of urgent vibration is close with the natural frequency of cover, the signal of sensor detection is maximum；
The signal that sensor detects delivers to detection means, and this signal is compared by detection means with a fixed level, according to comparing
Result is classified to qualified tank and unqualified tank and is rejected unqualified tank.
In the Chinese invention patent of Application No. CN85106707, proposition swept signal generator produces and output one accounts for
Empty ratio is 1：4 high power sweep square-wave signal deactivates solenoid, to produce the magnetic field of strong and weak change, and then makes to be detected
The cover of canned food produces amplitude, the forced resonance of frequency all changes, then sound cover being vibrated by the pick up on sensor is believed
Number it is transformed into the signal of telecommunication, is input to signal deteching circuit and is amplified, filters, then after useful signal detection, relatively more aobvious
Show in circuit and compare with fixed level, be substandard products higher than fixed level, be certified products less than fixed level.
Although above two scheme is all to make tested product vibration using excitation, thus sending sound, it is substantially
Whether the power of the acoustical signal according to time domain is qualified to judge tested product, and time domain voice is easily subject to noise jamming, and testing result
Sound signal collecting probe is compared with the distance of canister sensitive, there is weak point in actual applications.
Fu Jiangyun, Liu Yijun, Chen Jingyu, Tang Yongjin etc. exist《Computer measurement and control》09 phase in 2013 delivers《Base
Beer bottle interior pressure detecting system design online in Principles of Acoustics》In one literary composition, first use high-energy magnetic pulse excitation beer bottle cap, make
Bottle cap vibration sends sound, and sound is received and pass to signal processing system by microphone probe, and signal processing system is first to sound
Signal carries out FFT (Fast Fourier Transformation, fast fourier transform), obtains frequency-domain information, in frequency domain
Search out 15 amplitudes and the respective frequencies of maximum, by comparing whether these amplitudes and frequency meet frequency limit and amplitude circle
Limit and to learn pressure of the inside of a bottle situation, thus judging whether medicated beer is qualified.
The Chinese invention patent of Application No. CN201310106023 proposes to make container cover produce vibration using electromagnetic pulse
Thus sending sound, voice pickup module gathers this section of sound, and the voice data and removing then intercepting 400～3000Hz is less than
These voice datas are then converted to frequency domain data by 80 decibels of voice data, find two crests of highest in a frequency domain
Amplitude and this two crests frequency range, amplitude and frequency range to be judged vacuum degree of vessel as inspection data.
Above two scheme although having used the frequency domain information of sound, but simply simple select frequency domain peak value and its
Frequency is used as the basis for estimation of tested product, and the production technology difference of canister and microphone probe are become with the distance of canister
Change and testing result is all had an impact.
Content of the invention
It is an object of the invention to overcoming drawbacks described above of the prior art, provide a kind of gold based on sound spectrum matching
Belong to canned products quality online test method, improve the on-line checking speed of metal canned products, be prevented effectively from canister
The impact to testing result for the distance change of production technology difference and sound signal collecting probe and canister is hence it is evident that just improve
Really verification and measurement ratio.
For achieving the above object, the technical scheme that the present invention provides is as follows：
A kind of online test method of the canned product quality of the metal based on sound spectrum matching, comprises the following steps：
S1, for metal canned products to be detected, therefrom choose multiple qualified products and be trained, set up qualified products
Calculation model of mass；
S2, using the qualified products calculation model of mass set up in step S1, quality is carried out to tested metal canned products
Detection；
The method setting up qualified products calculation model of mass in described S1 includes：
S101, choose N number of qualified products from metal canned products to be detected, to k-th qualified products (1≤k≤N)
Apply excitation, so that product is vibrated；
The acoustical signal that S102, collection described k-th qualified products vibration produce, adopts to the acoustical signal collected
Sample processes and obtains digitized acoustical signal；
S103, process acoustical signal, obtain acoustical signal spectrum；Acoustical signal spectrum is carried out curve fitting and obtains frequency spectrum after matching
Curve, processes to spectrum curve calculating after matching and obtains qualified curvilinear function FkAnd its acceptance coefficient collection { Ck}；
S104, to qualified coefficient set { CkBe normalized, obtain new acceptance coefficient collection { C 'k}；
S105, all qualified products to selection carry out above-mentioned steps S101～S104 operation, obtain N number of new acceptance coefficient
Collection { C 'i(1≤i≤N), with new acceptance coefficient collection { C 'i(1≤i≤N) respectively as |input paramete, qualified products standard value δ
As output parameter, product quality calculating network is trained, obtains qualified products calculation model of mass；
The method carrying out quality testing to tested metal canned products in described S2 includes：
S201, excitation is applied to detected metal canned products, make product produce vibration；
The acoustical signal that S202, collection product vibration produce, carries out sampling processing number to the acoustical signal collected
The acoustical signal of word；
S203, process acoustical signal, obtain acoustical signal spectrum；Acoustical signal spectrum is carried out curve fitting and obtains frequency spectrum after matching
Curve, processes to spectrum curve calculating after matching and obtains measured curve function F and its actual measurement coefficient set { C }；
S204, to actual measurement coefficient set { C } be normalized, newly surveyed coefficient set { C ' }；
S205, will newly survey coefficient set { C ' } as in step S1 set up qualified products calculation model of mass input ginseng
Number, computing obtains product quality parameters R of this product；
The absolute value D of the difference of S206, product quality parameters R of calculating detected product and qualified products standard value δ,
Relatively D and given threshold T；If D is less than threshold value T, detected product is qualified；Conversely, then detected product is unqualified.
Further, described qualified curvilinear function Fk, the computational methods of curvilinear function include in measured curve function F：?
Q frequency spectrum point is taken on one section of acoustical signal spectrum, line translation is clicked through to each frequency spectrum of this section of acoustical signal spectrum, by j-th frequency spectrum
Point transformation is (x for coordinate figurej, valj) point, wherein, j be frequency spectrum point sequence number, 1≤j≤Q, abscissa xj=(j-K)/TT,
K and TT is constant, depending on its value is according to concrete examined product, vertical coordinate valjSignal value for corresponding frequency spectrum point；To sit
It is designated as (xj, valj) series of points be sequentially connected constituent curve；Above-mentioned curve is fitted according to function f (x), is somebody's turn to do
The curvilinear function of section acoustical signal spectrum.
Further, described function f (x) is sample conditional function, rational function or exponential function.
Further, described product quality calculating network is neutral net or support vector machine.
Further, described acoustical signal adopts microphone probe to collect.
Further, described rational function is:F (x)=(a+bx)/(1+cx+dx2).
Compared with prior art, the beneficial effects of the present invention is：For detected metal canned products, by first counting
Calculate the acoustical signal spectrum that the vibration of grade metal canned products produces, carry out curve fitting the qualified curve letter of acquisition to acoustical signal spectrum
Number, the new acceptance coefficient collection being obtained using after the acceptance coefficient collection normalization of qualified curvilinear function is as |input paramete, qualified
Product standard value δ, as output parameter, is trained to product quality calculating network, establishes qualified products Mass Calculation mould
Type；Again to detected metal canned products, by the acoustical signal spectrum of tested metal canned products vibration generation, to acoustical signal frequency
Spectrum carries out curve fitting and obtains measured curve function, with the new actual measurement obtaining after the actual measurement coefficient set normalization of measured curve function
Respectively as the |input paramete of qualified products calculation model of mass, the operation result of model is product quality parameters R to coefficient set, will
Product quality parameters R compare size with absolute value | D | of the difference of qualified products standard value δ with given threshold T, by comparing
Result is judging product quality.Make the present invention can realize the online high speed detection of metal canned products, be prevented effectively from canister
Production technology difference and microphone probe and canister the impact to testing result for the distance change hence it is evident that improving correct
Verification and measurement ratio.
Brief description
Fig. 1 is a kind of metal based on sound spectrum matching of the present invention canned product quality online test method step
The flow chart of rapid S1；
Fig. 2 is a kind of metal based on sound spectrum matching of the present invention canned product quality online test method step
The flow chart of rapid S2；
Fig. 3 is acoustical signal spectrum and its matched curve of the bright described embodiment of this law；
Fig. 4 is the neural network structure figure of embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples to a kind of metal canned products based on sound spectrum matching of the present invention
Quality online test method is described further.
The following is a kind of canned product quality online test method of metal based on sound spectrum matching of the present invention
Preferred example, therefore limit protection scope of the present invention.
Refer to Fig. 1 and Fig. 2, in figure shows a kind of canned product of metal based on sound spectrum matching of the present invention
The flow chart of quality online test method, comprises the following steps：
S1, for metal canned products to be detected, therefrom choose multiple qualified products and be trained, set up qualified products
Calculation model of mass；
S2, using the qualified products calculation model of mass set up in step S1, quality is carried out to tested metal canned products
Detection；
As shown in figure 1, the method setting up qualified products calculation model of mass in described S1 specifically includes：
S101, from metal canned products to be detected, choose N number of qualified products, qualified products k (1≤k≤N) is applied
Excitation, makes product vibrate；
The acoustical signal that S102, collection described qualified products k vibration produce, is carried out at sampling to the acoustical signal collected
Reason obtains digitized acoustical signal；
S103, process acoustical signal, obtain acoustical signal spectrum, i.e. former section of frequency spectrum in Fig. 3；Curve is carried out to acoustical signal spectrum
Matching, obtains spectrum curve after the matching shown in Fig. 3, spectrum curve calculating after matching is processed and obtains qualified curvilinear function Fk
And its acceptance coefficient collection { Ck}；
S104, to qualified coefficient set { CkBe normalized, obtain new acceptance coefficient collection { C 'k}；
S105, all qualified products to selection carry out above-mentioned steps S101～S104 operation, obtain N number of new acceptance coefficient
Collection { C 'i(1≤i≤N), with new acceptance coefficient collection { C 'i(1≤i≤N) respectively as |input paramete, qualified products standard value δ
As output parameter, product quality calculating network is trained, obtains qualified products calculation model of mass；
Carry out quality testing to tested metal canned products in described S2 to specifically include：
S201, excitation is applied to detected metal canned products, make product produce vibration；
The acoustical signal that S202, collection product vibration produce, carries out sampling processing number to the acoustical signal collected
The acoustical signal of word；
S203, process acoustical signal, obtain acoustical signal spectrum；Acoustical signal spectrum is carried out curve fitting and obtains frequency spectrum after matching
Curve, processes to spectrum curve calculating after matching and obtains measured curve function F and its actual measurement coefficient set { C }；
S204, to actual measurement coefficient set { C } be normalized, newly surveyed coefficient set { C ' }；
S205, will newly survey coefficient set { C ' } as in step S1 set up qualified products calculation model of mass input ginseng
Number, computing obtains product quality parameters R of this product；
The absolute value D of the difference of S206, product quality parameters R of calculating detected product and qualified products standard value δ,
Relatively D and given threshold T；If D is less than threshold value T, detected product is qualified；Conversely, then detected product is unqualified.
As preferred, described qualified curvilinear function Fk, in measured curve function F curvilinear function circular tool
Body includes：Q frequency spectrum point is taken on one section of acoustical signal spectrum, line translation is clicked through to each frequency spectrum of this section of acoustical signal spectrum, will
J-th frequency spectrum point transformation is (x for coordinate figurej, valj) point, wherein, j be frequency spectrum point sequence number, 1≤j≤Q, abscissa xj=
(j-K)/TT, K and TT are constant, depending on its value is according to concrete examined product, vertical coordinate valjLetter for corresponding frequency spectrum point
Number value；Coordinate is (xj, valj) series of points be sequentially connected constituent curve；Above-mentioned curve is intended according to function f (x)
Close, obtain the curvilinear function of this section of acoustical signal spectrum.
In this embodiment, the coefficient set { C } of curvilinear function is { a, b, c, d }, right according to x=(x-min)/(max-min)
{ C } be normalized after new coefficient set { C ' } be a', b', c', d'}, wherein min and max be respectively { C } minima and
Maximum, value is a to x successively, b, c, d.
As preferred, described function f (x) selects sample conditional function or rational function or exponential function.
In this embodiment, rational function is：F (x)=(a+bx)/(1+cx+dx2).
Furthermore, described product quality calculating network selects neutral net or support vector machine.
Furthermore, described acoustical signal adopts microphone probe to collect.
Fig. 4 shows the product quality calculating network that the present embodiment is selected, and the present embodiment selects three layers of BP neural network, its
Concrete structure includes：Input layer, hidden layer and output layer, wherein input layer include 4 node hidden layers and include 6 nodes, defeated
Go out layer and include 1 node；Each node of input layer is connected with each node of hidden layer respectively, each section of hidden layer
Point is all connected with the node of output layer respectively.New coefficient set { C ' } is inputted qualified products calculation model of mass by new coefficient set,
The operation result of qualified products calculation model of mass is product quality parameters R.
Embodiment described above only have expressed the several embodiments of the present invention, and its description is more concrete and detailed, but simultaneously
Therefore the restriction to the scope of the claims of the present invention can not be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, some deformation can also be made and improve, these broadly fall into the guarantor of the present invention
Shield scope.Therefore, the protection domain of patent of the present invention should be defined by claims.
Claims (6)
1. a kind of canned product quality of the metal based on sound spectrum matching online test method it is characterised in that include with
Lower step：
S1, for metal canned products to be detected, therefrom choose multiple qualified products and be trained, set up qualified products quality
Computation model；
S2, using the qualified products calculation model of mass set up in step S1, quality testing is carried out to tested metal canned products；
The method setting up qualified products calculation model of mass in described S1 includes：
S101, from metal canned products to be detected, choose N number of qualified products, excitation is applied to k-th qualified products, makes product
Product vibrate, wherein 1≤k≤N；
The acoustical signal that S102, collection described k-th qualified products vibration produce, is carried out at sampling to the acoustical signal collected
Reason obtains digitized acoustical signal；
S103, process acoustical signal, obtain acoustical signal spectrum；Acoustical signal spectrum is carried out curve fitting, and it is bent to obtain frequency spectrum after matching
Line, processes to spectrum curve calculating after matching and obtains qualified curvilinear function FkAnd its acceptance coefficient collection { Ck}；
S104, to qualified coefficient set { CkBe normalized, obtain new acceptance coefficient collection { Ck′}；
S105, all qualified products to selection carry out above-mentioned steps S101～S104 operation, obtain N number of new acceptance coefficient collection
{Ci', with new acceptance coefficient collection { Ci' respectively as |input paramete, qualified products standard value δ as output parameter, wherein 1≤i
≤ N, is trained to product quality calculating network, obtains qualified products calculation model of mass；
The method carrying out quality testing to tested metal canned products in described S2 includes：
S201, excitation is applied to detected metal canned products, make product produce vibration；
The acoustical signal that S202, collection product vibration produce, carries out sampling processing to the acoustical signal collected and obtains digitized
Acoustical signal；
S203, process acoustical signal, obtain acoustical signal spectrum；Acoustical signal spectrum is carried out curve fitting, and it is bent to obtain frequency spectrum after matching
Line, processes to spectrum curve calculating after matching and obtains measured curve function F and its actual measurement coefficient set { C }；
S204, to actual measurement coefficient set { C } be normalized, newly surveyed coefficient set { C ' }；
S205, will newly survey coefficient set { C ' } as in step S1 set up qualified products calculation model of mass |input paramete,
Computing obtains product quality parameters R of detected product；
The absolute value D of the difference of S206, product quality parameters R of calculating detected product and qualified products standard value δ, compares
D and given threshold T；If D is less than threshold value T, detected product is qualified；Conversely, then detected product is unqualified.
2. method according to claim 1 it is characterised in that：Described qualified curvilinear function Fk, bent in measured curve function F
The computational methods of line function include：Q frequency spectrum point, each frequency spectrum to this section of acoustical signal spectrum are taken on one section of acoustical signal spectrum
Click through line translation, j-th frequency spectrum point transformation is (x for coordinate figurej, valj) point, wherein, j be frequency spectrum point sequence number, 1≤j
≤ Q, abscissa xj=(j-K)/TT, K and TT are constant, depending on its value is according to concrete examined product, vertical coordinate valjFor correspondence
Frequency spectrum point signal value；Coordinate is (xj, valj) series of points be sequentially connected constituent curve；To above-mentioned curve according to letter
Number f (x) is fitted, and obtains the curvilinear function of this section of acoustical signal spectrum.
3. method according to claim 2 it is characterised in that：Described function f (x) be sample conditional function, rational function or
Exponential function.
4. method according to claim 1 it is characterised in that：Described product quality calculating network is neutral net or support
Vector machine.
5. method according to claim 1 it is characterised in that：Described acoustical signal adopts microphone probe to collect.
6. method according to claim 3 it is characterised in that：Described rational function is：F (x)=(a+bx)/(1+cx+
dx2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410790471.7A CN104597124B (en) | 2014-12-05 | 2014-12-17 | Metal canned product quality on-line detection method based on sound frequency-spectrum fitting |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2014107409781 | 2014-12-05 | ||
| CN201410740978 | 2014-12-05 | ||
| CN201410790471.7A CN104597124B (en) | 2014-12-05 | 2014-12-17 | Metal canned product quality on-line detection method based on sound frequency-spectrum fitting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104597124A CN104597124A (en) | 2015-05-06 |
| CN104597124B true CN104597124B (en) | 2017-02-22 |
Family
ID=53123044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410790471.7A Active CN104597124B (en) | 2014-12-05 | 2014-12-17 | Metal canned product quality on-line detection method based on sound frequency-spectrum fitting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104597124B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017114665A1 (en) * | 2015-12-30 | 2017-07-06 | Tetra Laval Holdings & Finance S.A. | Methods and apparatuses for packaging quality assurance |
| CN107748274B (en) * | 2017-09-30 | 2020-12-15 | 中国船舶重工集团公司第七0七研究所 | Method for detecting internal vacuum degree of accelerometer through servo circuit frequency sweep |
| SE541498C2 (en) | 2017-11-27 | 2019-10-22 | Acosense Ab | Method and system for determining process properties using active acoustic spectroscopy |
| CN108899048A (en) * | 2018-05-10 | 2018-11-27 | 广东省智能制造研究所 | A kind of voice data classification method based on signal Time-frequency Decomposition |
| CN109141919A (en) * | 2018-08-16 | 2019-01-04 | 杭州容大智造科技有限公司 | A kind of device and method using sound detection brake system |
| CN109060373A (en) * | 2018-08-16 | 2018-12-21 | 杭州容大智造科技有限公司 | A kind of equipment, method and automobile detecting rain brush |
| CN109141935A (en) * | 2018-08-16 | 2019-01-04 | 杭州容大智造科技有限公司 | A kind of equipment and automobile using sound detection tyre life |
| CN109115525A (en) * | 2018-08-16 | 2019-01-01 | 杭州容大智造科技有限公司 | A kind of equipment and automobile using current detecting tyre life |
| CN109141939A (en) * | 2018-08-16 | 2019-01-04 | 杭州容大智造科技有限公司 | A kind of device and method using sound detection suspension |
| CN109141882A (en) * | 2018-08-16 | 2019-01-04 | 杭州容大智造科技有限公司 | A kind of equipment, method and automobile using electricity consumption detection automobile bearing |
| CN109029994A (en) * | 2018-08-16 | 2018-12-18 | 杭州容大智造科技有限公司 | A kind of equipment, method and automobile using sound detection automobile bearing |
| CN109541031A (en) * | 2019-01-25 | 2019-03-29 | 山东农业大学 | Fruit hardness detection method based on acoustics and vibration characteristics |
| CN114850847A (en) * | 2021-02-04 | 2022-08-05 | 中国航发商用航空发动机有限责任公司 | Pipe joint assembling quality detection method |
| CN113159210B (en) * | 2021-04-30 | 2022-11-08 | 电子科技大学 | Object residual quantity detection model training method, object residual quantity detection system, object residual quantity detection device and storage medium |
| CN115616082B (en) * | 2022-12-14 | 2023-04-28 | 杭州兆华电子股份有限公司 | Keyboard defect analysis method based on noise detection |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN85106707A (en) * | 1985-09-02 | 1986-07-23 | 华南工学院 | A kind of method of vacuum in canned product non-destructive test and device |
| US6035718A (en) * | 1998-04-14 | 2000-03-14 | Coors Brewing Company | Acoustic bottle tester |
| US6182511B1 (en) * | 1998-04-14 | 2001-02-06 | Coors Brewing Company | Acoustic bottle tester and conveyor therefor |
| CN103230880A (en) * | 2013-03-28 | 2013-08-07 | 广州坚诺机械设备有限公司 | Rapid nondestructive detection apparatus of vacuum degrees of containers, and detection method thereof |
| CN103983325A (en) * | 2013-02-12 | 2014-08-13 | 荷兰普瑞玛燃气有限责任公司 | Method and program for determining the quantity of gas in a gas bottle using a smart phone, and a smart phone programmed for this purpose |
-
2014
- 2014-12-17 CN CN201410790471.7A patent/CN104597124B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN85106707A (en) * | 1985-09-02 | 1986-07-23 | 华南工学院 | A kind of method of vacuum in canned product non-destructive test and device |
| US6035718A (en) * | 1998-04-14 | 2000-03-14 | Coors Brewing Company | Acoustic bottle tester |
| US6182511B1 (en) * | 1998-04-14 | 2001-02-06 | Coors Brewing Company | Acoustic bottle tester and conveyor therefor |
| CN103983325A (en) * | 2013-02-12 | 2014-08-13 | 荷兰普瑞玛燃气有限责任公司 | Method and program for determining the quantity of gas in a gas bottle using a smart phone, and a smart phone programmed for this purpose |
| CN103230880A (en) * | 2013-03-28 | 2013-08-07 | 广州坚诺机械设备有限公司 | Rapid nondestructive detection apparatus of vacuum degrees of containers, and detection method thereof |
Non-Patent Citations (2)
| Title |
|---|
| Preliminary studies of a novel air-coupled ultrasonic inspection system for food containers;T.H. Gan et al.;《Journal of Food Engineering》;20021231;第53卷;第315-323页 * |
| 基于声学原理的啤酒瓶在线内压力检测***设计;付江云 等;《计算机测量与控制》;20131231;第21卷(第9期);第2362-2364,2368页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104597124A (en) | 2015-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104597124B (en) | Metal canned product quality on-line detection method based on sound frequency-spectrum fitting | |
| CN103230880B (en) | Rapid nondestructive detection apparatus of vacuum degrees of containers, and detection method thereof | |
| CN109387567A (en) | One kind being based on the modified increasing material manufacturing laser ultrasonic detection data processing method of velocity of wave | |
| CN108899048A (en) | A kind of voice data classification method based on signal Time-frequency Decomposition | |
| CN108008006B (en) | Welding seam defect detection method, device, equipment and system | |
| CN102625912A (en) | Fault detection method and system | |
| CN107367676A (en) | Partial discharge indicator based on audio intelligent identification | |
| CN117173461A (en) | Multi-visual task filling container defect detection method, system and medium | |
| CN102338682A (en) | Method for detecting pressures in sealed containers on line | |
| US8712704B2 (en) | Defect detection system and method | |
| CN113624834B (en) | Defect depth identification method and system based on marginal spectrum centroid detection | |
| CN114778690A (en) | Laser ultrasonic quantitative detection method for pore defects of additive part | |
| CN113607817A (en) | Pipeline girth weld detection method and system, electronic equipment and medium | |
| CN102346171B (en) | Flaw detecting system and method | |
| CN111272875A (en) | Apple brittleness nondestructive testing method based on vibration sound signals | |
| CN203163931U (en) | Whole-case container pressure rapid and nondestructive test equipment used in production lines | |
| CN203132762U (en) | Container-pressure rapid nondestructive detecting device used on production line | |
| CN203148623U (en) | Container vacuum degree rapid nondestructive test device used for production line | |
| CN110443301B (en) | Liquid dangerous goods identification method based on double-layer feature classification | |
| CN103196627B (en) | Device and method for fast and nondestructive testing of vacuum degree of full container load (FCL) container | |
| JP2001272385A (en) | Hammering test method using liner prediction coefficient method and hammering test device | |
| CN106442744A (en) | Ultrasonic signal processing method for removing response characteristics of ultrasonic probe | |
| CN103196615B (en) | Container pressure fast nondestructive test device and test method | |
| US11913911B2 (en) | System and method for dual pulse-echo sub-surface detection | |
| CN110749650B (en) | Anchor rod anchoring compactness grade evaluation method based on support vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 510000 building 15, 100 Hall of martyrs Road, Guangzhou, Guangdong Patentee after: GUANGDONG INSTITUTE OF INTELLIGENT MANUFACTURING Address before: 510070 building 13, 100 Hall of martyrs Road, Guangzhou, Guangdong Patentee before: Guangdong Institute of Automation |
|
| CP03 | Change of name, title or address | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 510000 building 15, 100 Hall of martyrs Road, Guangzhou, Guangdong Patentee after: Institute of intelligent manufacturing, Guangdong Academy of Sciences Address before: 510000 building 15, 100 Hall of martyrs Road, Guangzhou, Guangdong Patentee before: GUANGDONG INSTITUTE OF INTELLIGENT MANUFACTURING |
|
| CP01 | Change in the name or title of a patent holder |