CN102664630B - Synchronous acquisition method of multichannel signals inside mechanical systems - Google Patents

Synchronous acquisition method of multichannel signals inside mechanical systems Download PDF

Info

Publication number
CN102664630B
CN102664630B CN201210128636.5A CN201210128636A CN102664630B CN 102664630 B CN102664630 B CN 102664630B CN 201210128636 A CN201210128636 A CN 201210128636A CN 102664630 B CN102664630 B CN 102664630B
Authority
CN
China
Prior art keywords
data
interpolation
road
point
channel
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
Application number
CN201210128636.5A
Other languages
Chinese (zh)
Other versions
CN102664630A (en
Inventor
刘晓明
邵毅敏
韩术
徐驭航
王亮
胡磊
王姗姗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University
Original Assignee
Chongqing University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University filed Critical Chongqing University
Priority to CN201210128636.5A priority Critical patent/CN102664630B/en
Publication of CN102664630A publication Critical patent/CN102664630A/en
Application granted granted Critical
Publication of CN102664630B publication Critical patent/CN102664630B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Arrangements For Transmission Of Measured Signals (AREA)
  • Length Measuring Devices With Unspecified Measuring Means (AREA)

Abstract

A synchronous acquisition method of multichannel signals inside mechanical systems is characterized in that the method is performed in the following steps that each channel of sensors sends signals to each A/D channel of data acquisition and storages; each channel of sampling point data are stored in data acquisition and storages uniformly; the stored data are exported to a computer, sampling point data of each channel of A/D channel are separated to be preserved in a file respectively, and a corresponding channel of the sampling point data is stored in each profile; each channel of the sampling point data is subjected to 'piecewise interpolation', and discrete sequences are connected together to form piecewise data line of each channel of data; and all piecewise data lines after interpolation are connected together to form a data line of each channel of data, and then starting time of each channel of data lines are subjected to consistency processing. The synchronous acquisition method has the advantages of being small in size and capable of synchronously acquiring multi-channel A/D signals in closed and narrow mechanical transmission systems, and guaranteeing that the arrival time of actual measurement multi-channel signals is synchronous.

Description

The inner multiple signals synchronous obtaining method of a kind of mechanical system
Technical field
The invention belongs to a kind of sensor signal to obtain and treatment technology, particularly the inner multiple signals synchronous obtaining method of a kind of mechanical system.
Technical background
The signal monitoring system of mechanical system key point for improving plant equipment reliability of operation and fail safe, can avoid various catastrophic failure, and extend electromechanical equipment useful life, reducing total life cycle maintenance cost etc. provides important leverage.Therefore, the civilian and military demand of the intelligent on-line monitoring system be suitable for grows with each passing day.Such as Aero-Space, nuclear power station etc. adhere to that dropping into a huge sum of money for a long time develops mechanical Information Development of Equipment Support system always.But traditional monitoring system be generally adopt the transducer for analyzing and monitoring is arranged on machine driven system casing bearing pedestal on carry out the method diagnosing Yu analyze, due to away from be concerned about signal source, so there is the problems such as very noisy interference, the faint transmission loss of damage signal energy, the complicated close coupling of signal.In order to solve above-mentioned crucial general character matter of science and technology and a difficult problem, best way is exactly transducer is directly installed on the required position detected and monitor, the change of sense mechanism internal system signal.The difficult point of the method is: bin enclosure, the installing space of particular surroundings machine driven system are narrow and small, rotation, high temperature, at a high speed, speed change, variable load, vibration, lubricating oil, lubrication oil gas, etc. factor, limit the application of routine monitoring technology.
Due to the These characteristics of particular surroundings machine driven system, the signals collecting that requirement is built in mechanical system and treatment facility have the feature of " small size ", because " small size " requires that circuitry is simplified as far as possible, the sampling adopting the multi-channel A/D passage in single CPU to complete multiple key point data in the mode of " poll " so, this just causes multi-channel A/D sampling difficulty more.Meanwhile, multiple signals need time synchronizing.
Summary of the invention
The present invention is just for an above-mentioned difficult problem, propose the inner multiple signals synchronous obtaining method of a kind of mechanical system, there is the advantage of " small size ", can realize synchronously obtaining multi-channel A/D signal in closed, narrow and small machine driven system, and can ensure that multiple signals reach time synchronized.
Sinc Interpolation Principle
Be provided with function g (x), sampling thheorem shows, when meeting following two conditions, just can from the discrete sample at equal intervals of g (x) undistorted reconstruction primary signal: 1) signal is with to limit, i.e. the highest frequency bounded of signal; 2) sample frequency meets nyquist sampling rate, namely must be greater than the twice of signal highest frequency to the sample frequency of signal.When meeting above condition, Reconstructed equation is as (1) formula:
g ( t ) = Σ n = - ∞ ∞ g ( n ) sin c ( t - n ) - - - ( 1 )
Above formula can regard the convolution of g (n) and sinc (n) as, and g (n) represents n-th sampled point of g (t), and sinc function definition formula is as shown in (2) formula:
sin c ( n ) = 1 , n = 0 sin ( πn ) πn , n ≠ 0 - - - ( 2 )
In theory, its value can be determined by formula (1) to the interpolation point of any one given t value, namely realize the Accurate Reconstruction of original signal g (x).But can find in practical application, because accumulated variables in formula is tending towards just infinite, formula (1) is infeasible in actual applications.Therefore, the maximum of accumulated variables n can be realized from just infinite to the transformation of finite number (L+1) by the method for windowing in sinc interpolation algorithm.
So, when the sampling period is T, convert to obtain formula (3) by interpolation formula (1) formula:
g ( t ) = Σ n = 0 L ω ( nT ) g ( nT ) sin [ π ( t - nT ) / T ] π ( t - nT ) / T ,
t = m T k , m = { 1,2,3 , . . . , L * k } , L , k ∈ N - - - ( 3 )
In formula, g (nT) is the discrete series be made up of sampled point, and wherein T is the sampling period; ω (nT) for the discrete sampling sequence of institute's added-time window, its length be (L+1); G (t) is the discrete series after sinc interpolation, t is the time index of this discrete series, k is the interpolation point number in a road A/D signal between every two sampled points, the value mode of k is as follows: k+1 is the integral multiple of port number CN, after interpolation, insert (k-1) individual point between every 2 sampled points, m is the natural number from 1 to L*k.
A kind of internal signal acquisition method for mechanical system provided by the invention, has the advantage of " small size ", can solve the multi-channel A/D signal acquisition difficult problem caused by the particular surroundings of machine driven system.
For achieving the above object, the inner multiple signals synchronous obtaining method of a kind of mechanical system of the present invention, so its key is following these steps to carry out:
Step one: the revolving part that data acquisition and memory and at least two transducers are fixed on mechanical system is inner, and each transducer is connected with an A/D data acquisition channel of described data acquisition and memory;
Each road transducer outputs signal each A/D passage of data acquisition and memory respectively;
Step 2: driving device system works, in revolving part rotary course, data acquisition and memory Zhong Ge road A/D data acquisition channel are to the sampling of each road sensor signal, and each road sampling number is according to being unifiedly stored to described data acquisition and memory;
Step 3: mechanical system quits work, by all storage statistical conversion of described data acquisition and memory in computer, each road A/D channel sample point data be separated, be stored in a file respectively, each file correspondence deposits sampling number certificate described in a road;
Step 4: to each road sampling number according to carrying out " piecewise interpolation " process, namely first by each circuit-switched data segmentation, every one piece of data comprises 32 sampled points, and adopt sinc interpolation method to carry out interpolation to every segment data, interpolation formula is as follows:
g ( t ) = Σ n = 0 L ω ( nT ) g ( nT ) sin [ π ( t - nT ) / T ] π ( t - nT ) / T ,
(3)
t = m T k , m = { 1,2,3 , . . . , L * k } , L , k ∈ N
In formula, g (nT) is the discrete series be made up of sampled point, and wherein T is the sampling period; ω (nT) for the discrete sampling sequence of institute's added-time window, its length be (L+1); G (t) is the discrete series after sinc interpolation, t is the time index of this discrete series, k is the interpolation point number in a road A/D signal between every two sampled points, the value mode of k is as follows: k+1 is the integral multiple of port number CN, after interpolation, insert (k-1) individual point between every 2 sampled points, m is the natural number from 1 to L*k;
Discrete series is coupled together, forms the segmented data line of each circuit-switched data;
Step 5: when taking off one section of sampled point, starting point is last point of the preceding paragraph, and carries out interpolation by step 4 process; Afterwards, the segmented data line after every section of interpolation coupled together, the starting point of next section overlaps with the end point of the preceding paragraph, forms the data wire of each circuit-switched data.In step 5, after interpolation, each road A/D sampled signal is reconstructed within the scope of certain error, and between each data point, the time difference comprised between sampling and interpolation point shortens to T/k, and namely temporal resolution is Δ i=T/k.
Step 6: after the description completing each circuit-switched data line and drafting, then carry out consistency treatment to the initial time of each circuit-switched data line, obtain various transducing signal corresponding waveform at one time, processing method is:
Because multi-channel A/D adopts " poll " mode equal interval sampling, need interpolation Hou Ge road A/D signal from T cplace is blocked, that is: cast out T cfront data, by T cas new time 0 point.T ccomputing formula is as shown in (4) formula, and in formula, T is the sampling period, and CN is the A/D port number participating in sampling.The calculating of Δ t is as shown in (5) formula, and wherein T is the sampling period:
T c=(CN-1)*Δt (4)
Δt = T CN - - - ( 5 )
Δ T ibe less than Δ t, Δ t is Δ T iintegral multiple, the data point of each road A/D signal, comprise sampling and interpolation point.
As long as Δ T in step 4 ibe less than Δ t, Δ t is Δ T iintegral multiple, the data point of each road A/D signal, comprise sampling and interpolation point, then can in time with Δ T ialignment, thus realize the synchronous of multi-channel A/D.
Remarkable result of the present invention is: provide the inner multiple signals synchronous obtaining method of a kind of mechanical system, have the advantage of " small size ", can realize synchronously obtaining multi-channel A/D signal in closed, narrow and small machine driven system.And can ensure that actual measurement multiple signals reach time synchronized, each road A/D signal averaging absolute error precision is 10 -4v; In test process mechanical signal dynamic range in, between the A/D signal of each road, average relative error precision is 0.01%.
Accompanying drawing explanation
Fig. 1 data acquisition and memory and transducer assembling schematic diagram
The structured flowchart of Fig. 2 data acquisition and memory
Fig. 3 derived data Store form schematic diagram
Fig. 4 the 0th road signal sampling point oscillogram
Oscillogram after Fig. 5 the 0th road signal interpolation
Fig. 6 tetra-road signal interpolation point choose figure
The synchronous schematic diagram of Fig. 7 multi-channel sampling signal
Figure 84 road A/D sampled signal data synchronous waveform design sketch
Fig. 9 A/D0 signal 2 hours test waveform figure
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
The revolving part that data acquisition and memory and at least two transducers are fixed on mechanical system is inner, and each transducer is connected with an A/D data acquisition channel of described data acquisition and memory;
Each road transducer outputs signal each A/D passage of data acquisition and memory respectively;
The tested object of embodiment is certain machine driven system engine output shaft, and its layout as shown in Figure 1.Test key point is in output shaft inwall, collection signal is needed to be the strain signal produced by the distortion of output shaft inner wall surface, in figure, engine output shaft is revolving part, 4 transducers are pasted on the different key point position of output shaft inner wall surface 4 respectively, each sensor output signal is the voltage analog signal that amplitude is less, be connected to data acquisition with multi-channel A/D data acquisition channel and memory device by wire, A/D expanding channels No. one, each road transducer exports.Data acquisition and memory to be sampled to sensor electrical signal by 4 road A/D data acquisition channels and are stored data.But in the present embodiment, in order to prove that multi-channel A/D signal data reaching high-precise synchronization after " piecewise interpolation " process, changes the connection of the A/D data channel input in transducer and A/D equipment.In the present embodiment, especially 4 road A/D inputs are connected to same transducer, to get same signal source to observe and to illustrate the synchronous effect of 4 road signals.
Data acquisition and memory are primarily of multiple sensor signal input interface, power module, data memory module, A/D sampling module, CPU, USB interface composition, and wherein A/D sampling module is integrated in CPU, general structure schematic diagram as shown in Figure 2:
Multiple sensor signal input interface circuit mainly comprises the element circuits such as input signal amplification, Signal Regulation, filtering, for the A/D sampling module in CPU provides the signal met the demands.Power module is responsible for the power supply that whole equipment provides stable, and primarily of battery, power-switching circuit, electric source monitoring circuit three part forms.Data memory module comprises NANDFLASH storage chip and peripheral circuit thereof, the K9NBG08U5A chip that NANDFLASH storage chip adopts Samsung to produce, and its capacity is 2G Bytes.CPU is the core microprocessors of equipment, selects the LPC2146 microprocessor of PHILIPS company, primary responsibility: control inner A/D sampling module, monitoring power supply, the memory module that reads and writes data management data, control USB interface to complete usb data communication.
A/D sampling module in CPU can provide at most 14 road signal sampling passages, sampling precision is 12bit, can measurement signal voltages scope 0-3.3V, each road A/D adopts the working method of " poll ", in chronological sequence hocket sampling successively on Ji Ge road, so need to adopt proposed method in the present invention synchronous to realize each road A/D sampled signal.USB interface controls by CPU, realizes the usb data communication between A/D equipment and computer.
Driving device system works, in revolving part rotary course, data acquisition and memory Zhong Ge road A/D data acquisition channel are to the sampling of each road sensor signal, and each road sampling number is according to being unifiedly stored to described data acquisition and memory;
Mechanical system quits work, and by all storage statistical conversion of described data acquisition and memory in computer, each road A/D channel sample point data be separated, be stored in a file respectively, each file correspondence deposits sampling number certificate described in a road;
Sensor output signal line and A/D equipment connection good after, need data acquisition and memory are installed and is fixed on engine output shaft inside, and sealed.Installation data gathers and after memory, data acquisition and memory is worked on power, and can start test afterwards.Start drive system, engine output shaft starts to rotate, and in the process, data acquisition and memory gather the signal by each road A/D data acquisition channel institute or acquisition in real time, and data are stored in data memory module after being changed by A/D.After obtaining the critical data wishing to obtain, stop drive system, open seal cover.Be connected with host computer computer, the statistical conversion recorded in data acquisition and memory data memory module in computer by the USB interface of USB data line by data acquisition and memory.
Export to the deposit data form of computer at first as shown in Figure 3, in Fig. 3, every lattice represent a byte.The data that 4 road A/D passages gather are staggered successively to be deposited, and each sampling number adds channel number C according to front meeting nas mark, subscript n=0 in the present embodiment, 1,2,3.Because every road A/D channel sample precision is 12bit, so a sampled point result binary system is deposited need 16bit, namely need with two byte representations, high byte DH mbefore leaving in, low byte DL mafter leaving in, subscript m is sampled point successively sequence number.
Each circuit-switched data is separated by computer, is stored in each file respectively, and each file correspondence deposits the data of a road signal.In file, each sampling number represents with decimal number according to after conversion, and in file, often row deposits a sampling number certificate, and each data are successively deposited by acquisition time, to represent the size of file respective signal when different time points.High byte DH mwith low byte DL mundertaken to decimal numeral conversion by formula 6, V in formula reffor the reference voltage of outside input, be 3.3V, n in the present embodiment being A/D sampling precision figure place, is 12bit, V in the present embodiment decfor the decimal number after conversion.
V dec = DH m × 256 + DL m 2 n × V ref - - - ( 6 )
Primary signal highest frequency f m=1KHz, for meeting nyquist sampling theorem, selects sample frequency f s=4KHz, the sinc function time window length (L+1) that order uses is 32 points, namely adopts the sinc function of 32 to carry out " piecewise interpolation " initial data.
At f swhen=4KHz and sinc window get 32, add starting point, a road signal is 32 sampled points, and corresponding time span is 7.75ms, and between every 2 sampled points, the time interval is Δ T=250 μ s, is described below for the 0th road signal data.As shown in Figure 4, in figure, abscissa line is the time to the waveform drawn by 32 direct lines of sampled point of the 0th road signal, and unit is millisecond (ms), and the longitudinal axis is the voltage magnitude of signal, and unit is millivolt (mv).
To each road sampling number according to carrying out " piecewise interpolation " process, namely first by each circuit-switched data segmentation, every one piece of data comprises 32 sampled points, adopts sinc interpolation method to carry out interpolation to every segment data;
These 32 sampled point formation sequence g (nT) are existing according to interpolation formula sinc interpolation method recovery original signal.Interpolation 7 points between every 2 sampling points, after making interpolation, waveform temporal resolution when ensureing that actual signal frequency components is constant improves, i.e. every 2 time interval Δ T after interpolation iless, by above-mentioned situation Δ T i=31.25 μ s.
g ( t ) = Σ n = 0 L ω ( nT ) g ( nT ) sin [ π ( t - nT ) / T ] π ( t - nT ) / T ,
t = m T k , m = { 1,2,3 , . . . , L * k } , L , k ∈ N - - - ( 3 )
In formula, g (nT) is the discrete series be made up of sampled point, and wherein T is the sampling period; ω (nT) for the discrete sampling sequence of institute's added-time window, its length be (L+1); G (t) is the discrete series after sinc interpolation, and t is the time index of this discrete series, and k and interpolation are counted relevant, and after interpolation, insert (k-1) individual point between every 2 sampled points, m is the natural number from 1 to L*k.
Discrete series is coupled together, forms the segmented data line of each circuit-switched data;
Data wire waveform after interpolation as shown in Figure 5.Comparison diagram 4 and Fig. 5 visible, after sinc interpolation processing, after the 0th road signal interpolation, waveform is rounder and more smooth, and according to Sinc Interpolation Principle, after interpolation, waveform is attached most importance to structure waveform, close to true waveform.
After processing 4 road A/D sampled signal data successively, each road signal waveform than before interpolation close to true waveform, and temporal resolution is Δ T i=31.25 μ s.Multi-channel A/D the sampled signal that rises to of temporal resolution synchronously provides necessary condition.
When taking off one section of sampled point, starting point is last point of the preceding paragraph, and carries out interpolation by step 4 process; Afterwards, the segmented data line after every section of interpolation coupled together, the starting point of next section overlaps with the end point of the preceding paragraph, forms the data wire of each circuit-switched data.
Get one section of waveform that time span is 1ms, with the same time for transverse axis illustrates the position of 4 road A/D sampled points respectively by 4 reference axis, 4 road sampled points choose respectively as Fig. 6 (a), (b), c (), shown in (d).Tu6Zhong Ge road sampled point all represents with dot, can see, each road sampled point does not all align in time, and reason is: sample frequency is f s=4KHz, then the sampling interval is Δ T=250 μ s; Again because four road A/D adopt " poll " mode equal interval sampling, therefore the interval of every adjacent two-way A/D sampled point in same " poll " cycle Δ T should be
In Fig. 6, a, b, c, d tetra-points are respectively the starting point of four road sampled points, and due to second and third, the relative first via of first sampling point on four tunnels has the delay of Δ t, 2 Δ t, 3 Δ t respectively, cannot by the signal before demosaicing.Because Δ t is in us level in this embodiment, and need the generation time section of the inner key signal of the machine driven system engine output shaft of test to start latter 1 hour in machine driven system, therefore can block from d point completely, the forward part of corresponding for 4 circuit-switched data d points time is given up.The d point corresponding time is intercepting time T c, T ccomputing formula is as shown in (4) formula.F in formula ssample frequency is, CN is the A/D port number participating in sampling.The calculating of Δ t is as shown in (5) formula, and wherein T is the sampling period.
T c≥(CN-1)*Δt (4)
Δt = T CN - - - ( 5 )
As long as Δ T in step 4 ibe less than Δ t, Δ t is Δ T iintegral multiple, the data point of each road A/D signal, comprise sampling and interpolation point, then can in time with Δ T ialignment, thus realize the synchronous of multi-channel A/D.As mentioned before, the multi-channel A/D sampled signal that rises to of temporal resolution synchronously provides necessary condition, for the ease of describing in detail, getting 0 ~ 312.5us section of time shaft, each sampled point and interpolation point position being directly marked on time shaft, as shown in Figure 7.Tu7Zhong Ge road sampled point dot represents, because four road A/D adopt " poll " mode equal interval sampling, therefore A/D0 first sampled point is at 0us place, A/D1 first sampled point is at 62.5us place, A/D2 first sampled point is at 125us place, A/D3 first sampled point is at 187.5us place, and namely the interval of every adjacent two-way A/D sampled point in same " poll " cycle Δ T should be for A/D0, between its 2 sampled points being positioned at 0us and 250us, the time interval is Δ T=250 μ s, by inserting 7 points after sinc interpolation between two sampled points, interpolation point lays respectively at: 31.25us, 62.5us, 93.75us, 125us, 156.25us, 187.5us, 218.75us place, i.e. every 2 time interval Δ T after interpolation i=31.25 μ s.Visible, Δ T/ Δ T i=8, namely interpolation Hou Mei road Signal temporal resolution brings up to 8 times before interpolation.
In the present embodiment, due to A/D0 ~ A/D3 collection is same signal source, and 4 road signals have been accurate to Δ T by temporal resolution after sinc interpolation i=31.25 μ s, each interpolation point amplitude also close to actual value, so on the corresponding time point of the sampled point of each road A/D signal, corresponding interpolation point can be found to align with it at other 3 road A/D signals, thus finally achieve the synchronous of 4 road A/D signals.Synchronous effect can be gone out see by the red dotted line perpendicular to transverse axis in Fig. 7, and at time point 62.5us, 125us, 187.5us, 250us, 312.5us place, each road A/D signal sampling point and interpolation point all achieve alignment.
To sum up, through the process of step 4 and step 5,4 road A/D sampled signal data waveforms reach synchronous, as shown in Fig. 8 (a), (b), (c), (d).
A/D signal corresponding reference axis in each road in Fig. 8, transverse axis is time shaft, and initial time is the time to chopping in step 5, and time span is 7.75ms, and the longitudinal axis is waveforms amplitude.After recovery longer section of original signal waveform, four tunnel signal waveforms are almost completely overlapping, and synchronous effect is obvious.
In order to say the feasibility of multi-channel A in the present invention/D signal synchronizing method, simultaneously also for the ease of work such as the post analysis of signal and process, so the absolute error be necessary to analyze between the signal of each road after synchronization and relative error.Error analysis object is 4 road A/D signals, and time span is 77.5ms.
1. mean absolute error
With Δ T i=31.25 μ s get a little in the time interval, by the mean absolute error MAD (Mean Aosolute Difference) between the signal of formula 7, Qiu Ge road.
MAD ( I , J ) = 1 S Σ j = 1 S | I j - J j | - - - ( 7 )
In formula 7, I, J represent the A/D signal that two-way is different, and j is with Δ T i=31.25 μ s are the data point subscript obtained in the time interval, I jand J jthe amplitude that the different A/D signal of two-way is put at one time, S is always counting of obtaining.Because time span is 77.5ms in Fig. 9, so the S=249 that always counts.
After four road A/D signals are synchronous, MAD is as shown in table 1.
MAD after table 1 four road A/D signal is synchronous
2. relative error
Relative error RE refers to that between the signal of each road, mean absolute error is relative to the size of A/D signal dynamic range within the whole testing time, computational methods as shown in Equation 8:
RE ( I , J ) = MAD ( I , J ) ( V MAX - V MIN ) - - - ( 8 )
In the present embodiment, testing time length is 2 hours because multi-channel A/D signal source is identical, so can by the total dynamic range of A/D0 waveform determination signal, its whole waveform as shown in Figure 9,
As seen from Figure 9, V mAX=2140mv, V mIN=1525mv, then calculate relative error between the A/D signal of each road by formula X as shown in table 2.
RE between the A/D signal of table 2 four road
Conclusion: after multi-channel A/D signal is synchronous, between the A/D signal of each road, mean absolute error precision is 10 -4v, in test process mechanical signal dynamic range in, between the A/D signal of each road, average relative error precision is 0.01%.

Claims (1)

1. the inner multiple signals synchronous obtaining method of mechanical system, it is characterized in that following these steps to carry out:
Step one: the revolving part that data acquisition and memory and at least two transducers are fixed on mechanical system is inner, and each transducer is connected with an A/D data acquisition channel of described data acquisition and memory;
Each road transducer outputs signal each A/D passage of data acquisition and memory respectively;
Step 2: driving device system works, in revolving part rotary course, data acquisition and memory Zhong Ge road A/D data acquisition channel are to the sampling of each road sensor signal, and each road sampling number is according to being unifiedly stored to described data acquisition and memory;
Step 3: mechanical system quits work, by all storage statistical conversion of described data acquisition and memory in computer, each road A/D channel sample point data be separated, be stored in a file respectively, each file correspondence deposits sampling number certificate described in a road;
Step 4: to each road sampling number according to carrying out " piecewise interpolation " process, by each circuit-switched data segmentation, and carry out " piecewise interpolation " process, form the segmented data line of each circuit-switched data;
Described " piecewise interpolation " process is:
First by each circuit-switched data segmentation, every one piece of data comprises 32 sampled points, and adopt sinc interpolation method to carry out interpolation to every segment data, interpolation formula is as follows:
g ( t ) = Σ n = 0 L ω ( nT ) g ( nT ) sin [ π ( t - nT ) / T ] π ( t - nT ) / T ,
t = m T k , m = { 1,2,3 , K , L * k } , L , k ∈ N
In formula, g (nT) is the discrete series be made up of sampled point, and wherein T is the sampling period; ω (nT) for the discrete sampling sequence of institute's added-time window, its length be (L+1); G (t) is the discrete series after sinc interpolation, t is the time index of this discrete series, k is the interpolation point number in a road A/D signal between every two sampled points, the value mode of k is as follows: k+1 is the integral multiple of port number CN, after interpolation, insert (k-1) individual point between every 2 sampled points, m is the natural number from 1 to L*k; Discrete series is coupled together, forms each road, the segmented data line of data;
Step 5: when taking off one section of sampled point, starting point is last point of the preceding paragraph, and carries out interpolation by step 4 process; Afterwards, the segmented data line after every section of interpolation coupled together, the starting point of next section overlaps with the end point of the preceding paragraph, forms the data wire of each circuit-switched data;
Step 6: after the description completing each circuit-switched data line and drafting, then carry out consistency treatment to the initial time of each circuit-switched data line, obtain various transducing signal corresponding waveform at one time, its processing method is:
Because multi-channel A/D adopts " poll " mode equal interval sampling, need interpolation Hou Ge road A/D signal from T cplace is blocked, that is: cast out T cfront data, by T cas new time 0 point, T ccomputing formula is as shown in (4) formula, and in formula, T is the sampling period, and CN is the A/D port number participating in sampling;
The calculating of Δ t is as shown in (5) formula, and wherein T is the sampling period:
T c=(CN-1)*Δt (4)
Δt = T CN - - - ( 5 )
Δ T ibe less than Δ t, Δ t is Δ T iintegral multiple, the data point of each road A/D signal, comprise sampling and interpolation point;
In step 5, after interpolation, each road A/D sampled signal is reconstructed within the scope of certain error, and between each data point, the time difference comprised between sampling and interpolation point shortens to T/k, and namely temporal resolution is Δ T i=T/k.
CN201210128636.5A 2012-04-28 2012-04-28 Synchronous acquisition method of multichannel signals inside mechanical systems Expired - Fee Related CN102664630B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210128636.5A CN102664630B (en) 2012-04-28 2012-04-28 Synchronous acquisition method of multichannel signals inside mechanical systems

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210128636.5A CN102664630B (en) 2012-04-28 2012-04-28 Synchronous acquisition method of multichannel signals inside mechanical systems

Publications (2)

Publication Number Publication Date
CN102664630A CN102664630A (en) 2012-09-12
CN102664630B true CN102664630B (en) 2015-06-24

Family

ID=46774060

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210128636.5A Expired - Fee Related CN102664630B (en) 2012-04-28 2012-04-28 Synchronous acquisition method of multichannel signals inside mechanical systems

Country Status (1)

Country Link
CN (1) CN102664630B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103064671A (en) * 2012-12-19 2013-04-24 宁波三星电气股份有限公司 Downlink transaction processing method for electrical power terminal
CN110361608B (en) * 2019-05-31 2022-02-15 南京国电南自电网自动化有限公司 Multi-path asynchronous sampling data processing method based on sampling point geometric scaling
CN110739969A (en) * 2019-10-18 2020-01-31 唐智科技湖南发展有限公司 signal synchronous acquisition system
CN112129928B (en) * 2020-11-18 2021-02-26 震坤行网络技术(南京)有限公司 Method, computing device and computer storage medium for detecting state of lubricating oil

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1388942A2 (en) * 2002-08-07 2004-02-11 Zarlink Semiconductor Limited Conversion circuit, tuner and demodulator
CN101267207A (en) * 1999-03-24 2008-09-17 株式会社爱德万测试 A/D converter and calibration unit

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101267207A (en) * 1999-03-24 2008-09-17 株式会社爱德万测试 A/D converter and calibration unit
EP1388942A2 (en) * 2002-08-07 2004-02-11 Zarlink Semiconductor Limited Conversion circuit, tuner and demodulator

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于ARM的汽车传动轴安全监测***研发;韩术;《中国优秀硕士学位论文全文数据库》;20110315(第3期);第I140-403页 *
高艳等.阶比跟踪中的惠特克插值.《昆明理工大学学报(理工版)》.2007,第32卷(第3期),第30-34页. *

Also Published As

Publication number Publication date
CN102664630A (en) 2012-09-12

Similar Documents

Publication Publication Date Title
Wang et al. Efficient data reduction at the edge of industrial Internet of Things for PMSM bearing fault diagnosis
CN201742158U (en) Online monitoring device for power transformer
CN102664630B (en) Synchronous acquisition method of multichannel signals inside mechanical systems
CN201277910Y (en) Test parameter measurement system for aero engine
JP3537452B2 (en) Apparatus and method for compressing measurement data correlated with machine state
CN201107393Y (en) Synchronization / coordinating controller performance test platform
CN111382789B (en) Power load identification method and system based on machine learning
CN104819841B (en) Built-in-coding-information-based single sensing flexible angle-domain averaging method
CN104483011A (en) Rotary machinery multichannel vibration signal on-line detection and analysis system and method
CN102565542A (en) Capacitive equipment medium loss online monitoring method based on IEC61850-9-2 standard
CN103471703B (en) Non-aliasing reconstruction method of high-speed leaf undersampled leaf apex vibration signals
CN103472331A (en) Photovoltaic power generation fault diagnosis system based on photovoltaic physical model
CN113834563A (en) Mechanical vibration state signal acquisition and analysis system
CN103149046A (en) Multi-dimensional fault diagnosis method based on expert thinking
CN104122086A (en) Kriging model based wind turbine gearbox fault diagnosis method
CN104895747A (en) Multi-channel motor set vibration data collecting system based on STM
EP3819607A1 (en) Monitoring of a revolving component employing time-synchronized multiple detectors
CN103827683A (en) A method of determining stationary signals for the diagnostics of an electromechanical system
Peng et al. Speed estimation in planetary gearboxes: A method for reducing impulsive noise
CN201331430Y (en) Intelligent checking meter of monitoring and protecting instrument of steam turbine
CN102638265A (en) Internal signal acquisition method for mechanical system
CN103155476A (en) Quantizing sampled inputs using fixed frequency analog to digital conversions through interpolation
Li et al. A spectral self-focusing fault diagnosis method for automotive transmissions under gear-shifting conditions
Yi et al. High-order synchrosqueezing superlets transform and its application to mechanical fault diagnosis
CN104714161B (en) A kind of cable insulation data processing 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
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: 20150624

Termination date: 20180428