CN102890750A - Data analysis method for transportation safety recorder - Google Patents

Abstract

The invention relates to a data analysis method for a transportation safety recorder. The method mainly comprises the following steps of: importing acquired data, performing time-domain processing on the acquired data, eliminating outliers, removing mean values, performing low-pass filter to filter out high-frequency signals, calculating acceleration root-mean-squares, calculating impact duration, obtaining power spectral density, obtaining an impact duration speed change amount-acceleration peak value statistical point scatter diagram, and applying data results. Acquired time-domain signals are subjected to time-domain preprocessing and low-pass filter processing first by the method provided by the invention for laying a foundation for the analysis of the acquired data; and by the method provided by the invention, dynamic conditions of goods in a transportation process are analyzed comprehensively, and damage conditions of the goods are judged in time, so that a foundation is laid for the improvement of logistics management.

Description

Transportation safety registering instrument data analysing method

Technical field

The invention belongs to the technical data analysis field of monitoring logistics transportation, especially vibration signal power spectral-density analysis, impact signal velocity variable statistics are specifically related to a kind of transportation safety registering instrument data analysing method.

Background technology

The transportation environment data of country are the basic datas of a country, have status of equal importance with geographical, hydrographic data, and each state all pays much attention to this data acquisition and arrangement.These data all will play important effect for improving conevying efficiency, the less wasting of resources, reduction damage rate of goods etc.According to the demand of transportation and packing process, carrying out the Product transport environmental test and launching the relevant parameters measurement is a very complicated and heavy job always.Along with the development of the new and high technologies such as electronic technology and computer technology, this working condition has greatly change, and it is much easier than past that the test of transportation environment and measurement become.Main cause is that light small-sized on-site parameters survey record device comes out, but this registering instrument can independently use and monitoring product logistics transportation medium power state.For realizing effective utilization of transportation safety registering instrument, it is very necessary and urgent that the image data of efficient statistic record instrument, the image data of Macro or mass analysis registering instrument just seem.

Summary of the invention

The object of the invention is to solve that the transportation safety monitor data gathers and the deficiency of analytical approach, a kind of transportation safety registering instrument data analysing method is provided.

The present invention solves its technical matters and takes following technical scheme to realize:

A kind of transportation safety registering instrument data analysing method, the hardware system of the method operation comprises safety recording instrument and analyzes main frame, safety recording instrument sends the acceleration information that collects to the analysis main frame by USB port, and it is as follows to the analytical approach step of image data to analyze main frame:

⑴ insert operating analysis main frame USB port with the safety recording instrument data line, and main frame is analyzed in the data importing that safety recording instrument gathers;

⑵ Outliers in Data Processing is rejected in the time domain pre-service of image data, and zero-mean is processed;

⑶ low-pass filtering: adopt the filtering of Butterworth filter device to the minimum high-frequency signal of goods infringement;

⑷ calculate acceleration-root-mean square;

⑸ draw power spectrum density: use period map method to draw the power spectrum density of vibration acceleration signal;

⑹ calculate shock duration: the starting point when first corresponding time of 5%-20% signal of getting acceleration signal shock response peak value is impact metre, terminal point when second the corresponding time of 5%-20% signal of getting acceleration signal shock response peak value is impact metre calculates shock duration;

⑺ draw the loose figure of shock duration velocity variable～acceleration peak value statistics point: to be that the maximal value of each acceleration signal is a little loose as the longitudinal axis counts as transverse axis, impact peak value scheme velocity variable in the shock duration;

⑻ data result is used: comprise the application of acceleration-root-mean square analysis, power spectrum density application and shock duration velocity variable～loose figure of acceleration peak value statistics point.

And Outliers in Data Processing is rejected in the time domain pre-service of described step ⑵ image data, and the concrete steps that zero-mean is processed are:

1. according to band limit relative method, the data of safety recording instrument acceleration collection are arranged sampled value time series { x according to time sequencing _i, i=1,2,3 ..., N, calculating sampling value seasonal effect in time series Fang Jun and mean square deviation: ${S_i}^2=\stackrel{\‾}{x_i^2}-{\stackrel{\‾}{x_i}}^2;$

2. Outliers in Data Processing is judged: if, x _i-kS _i＜x _I+1＜x _i+ kS _iX then _I+1Normal point, otherwise, x _I+1Be wild point, it rejected coefficient k value value between 3 to 5 for the data that are judged to be wild point from image data;

3. zero-mean is processed: average is the direct current distracter of collection signal, according to the Estimation of Mean of signal, before carrying out normalized set, average is removed.If { x _i, i=1,2,3 ..., N is the sampled value time series of acceleration, note sampled value average is Have:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{x}_i$

New signal sequence after zero-mean is processed is { u _i, i=1,2,3 ..., N has And the new signal serial mean is zero, namely

And described step (3) filtering is signal more than the 50HZ to the minimum high-frequency signal of goods infringement.

And described step ⑷ calculates acceleration-root-mean square for using time domain to ask the root mean square that collects acceleration signal:

$G_{\mathrm{RMS}}=\sqrt{\frac{1}{T}\underset{0}{\overset{T}{\∫}}{A_i}^2\mathrm{dt}}=\sqrt{\frac{1}{N}\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}{A_n}^2}$

Wherein, A _iBe acceleration sampling time sequential value, A _nBe sequential value in the acceleration sampling frequency domain, G _RMSBe the acceleration root-mean-square value in the time domain or in the frequency domain, according to the Parseval theorem, the acceleration root-mean-square value that time domain is interior or frequency domain is interior equates.

And the power spectrum density that described step ⑸ uses period map method to draw vibration acceleration signal is:

1. windowing process: for suppressing Power leakage, adopt Hanning window that the acceleration time domain signal that gathers is sampled and block;

2. Fourier transform: the acceleration signal after the windowing is carried out Fourier transform, obtain amplitude spectrum;

3. draw power spectrum density: with after each amplitude data in the amplitude spectrum behind the Fourier transform square divided by amount of frequency translation, both drawn final power spectral density results.

And, get among the described step ⑹ impulse response signal peak value first 5%, or 10%, or 15%, or the starting point of corresponding time of 20% signal when being impact metre, second 5% of getting the impulse response signal peak value, or 10%, or 15%, or the terminal point of corresponding time of 20% signal when being impact metre.

And velocity variable is in the shock duration among the described step ⑺, begins front 0.4D integration amount the 0.1D behind the shock pulse from shock pulse, and wherein, D is the shock duration that calculates among the step ⑹.

And the acceleration-root-mean square analysis of described step ⑻, power spectrum density application and shock duration velocity variable～acceleration peak value statistics loose figure of point are applied as:

1. acceleration-root-mean square analysis: the damaged condition of judging goods by the size of acceleration-root-mean square numerical value;

2. power spectrum density is used: power spectral density value is inputted the program-con-trolled computer of laboratory shaking table, realize the laboratory random vibration test of equivalent road examination, reach the purpose of accurate simulation test;

3. the application of shock duration velocity variable～loose figure of acceleration peak value statistics point: the damaged boundary curve map that contrasts shock duration velocity variable～loose figure of acceleration peak value statistics point and product, drop in a large number such as the statistics point among the loose figure of fruit dot in the damaged area of damaged boundary curve of product, product will suffer damage so, almost drop on such as the statistics point among the loose figure of fruit dot outside the damaged area of damaged boundary curve of product, then Product Status is good.

Advantage of the present invention and good effect are:

1, the inventive method is at first carried out the time domain pre-service to the time-domain signal that gathers, and low-pass filtering treatment is for the analysis of image data lays the foundation.

2, the inventive method has been carried out the loose diagram data calculating of acceleration-root-mean square calculating, power spectrum density calculating and shock duration velocity variable～acceleration peak value statistics point to image data, the dynamic situation of goods in the multianalysis transportation is in time judged the damage situations of goods.

3, the inventive method step is reasonable, and the operation result of image data is reliable, for the raising that improves the logistics management level lays the foundation.

4, the drawn power spectral density value of the inventive method provides valuable real data support for the contrast test in laboratory.

Description of drawings

Fig. 1 is the logic step block scheme of the inventive method.

Embodiment

Below in conjunction with accompanying drawing the embodiment of the invention is further described; it is emphasized that; embodiment of the present invention is illustrative; rather than determinate; therefore the present invention is not limited to the embodiment described in the embodiment; every other embodiments that drawn by those skilled in the art's technical scheme according to the present invention belong to the scope of protection of the invention equally.

A kind of transportation safety registering instrument data analysing method, the hardware system of the method operation comprise safety recording instrument and analysis main frame, and safety recording instrument sends the acceleration information that collects to the analysis main frame by USB port, and method step is as follows:

⑴ insert operating analysis main frame USB port with the safety recording instrument data line, and main frame is analyzed in the data importing that safety recording instrument gathers;

⑵ Outliers in Data Processing is rejected in the time domain pre-service of image data, removes average:

1. according to band limit relative method, the data of safety recording instrument collection are arranged sampled value time series { x according to time sequencing _i, i=1,2,3 ..., n calculating sampling value seasonal effect in time series Fang Jun and mean square deviation:

2. Outliers in Data Processing is judged: if, x _i-kS _i＜x _I+1＜x _i+ kS _iX then _I+1Normal point, otherwise, x _I+1Be wild point, for the data that are judged to be wild point it rejected from image data, Number k value value between 3 to 5;

3. zero-mean is processed: average is the direct current distracter of collection signal, according to the Estimation of Mean of signal, before carrying out normalized set, average is removed.If { x _i, i=1,2,3 .., N is sampled value (accekeration) time series of acceleration, note sampled value average is

Have:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{x}_i$

New signal sequence after zero-mean is processed is { u _i, i=1,2,3 ..., N has

And the new signal serial mean is zero, namely

⑶ low-pass filtering: adopt the filtering of Butterworth filter device to the minimum high-frequency signal of goods infringement, high-frequency signal is the above signal of 50HZ;

⑷ calculate acceleration-root-mean square: use time domain to ask the root mean square that collects acceleration signal:

$G_{\mathrm{RMS}}=\sqrt{\frac{1}{T}\underset{0}{\overset{T}{\∫}}{A_i}^2\mathrm{dt}}=\sqrt{\frac{1}{N}\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}{A_n}^2}$

Wherein, A _iBe acceleration sampling time sequential value, A _nBe sequential value in the acceleration sampling frequency domain, G _RMSBe the acceleration root-mean-square value in the time domain or in the frequency domain, according to the Parseval theorem, the acceleration root-mean-square value that time domain is interior or frequency domain is interior equates;

⑸ draw power spectrum density: use period map method to draw the power spectrum density of vibration acceleration signal;

1. windowing process: for suppressing Power leakage, adopt Hanning window that the acceleration time domain signal that gathers is sampled and block;

2. Fourier transform: the acceleration signal after the windowing is carried out Fourier transform, obtain amplitude spectrum;

3. draw power spectrum density: with after each amplitude data in the amplitude spectrum behind the Fourier transform square divided by amount of frequency translation, both drawn final power spectral density results;

⑹ shock duration is calculated: get the impulse response signal peak value first 5%, or 10%, or 15%, or the starting point of corresponding time of 20% signal when being impact metre, get the impulse response signal peak value second 5%, or 10%, or 15%, or the terminal point of corresponding time of 20% signal when being impact metre, calculate shock duration;

⑺ the loose figure of shock duration velocity variable～acceleration peak value statistics point: to be that the maximal value of each acceleration signal is a little loose as the longitudinal axis counts as transverse axis, impact peak value scheme velocity variable in the shock duration, velocity variable is in the described shock duration, begin front 0.4D integration amount the 0.1D behind the shock pulse from shock pulse, wherein, D is the shock duration that calculates among the step ⑹;

⑻ data result is used:

1. acceleration-root-mean square analysis: the acceleration-root-mean square that obtains among the above-mentioned steps ⑷ is the physical quantity of analyzing the oscillation intensity that vehicle-mounted cargo bears in transit, judges the damaged condition of goods by the size of acceleration-root-mean square numerical value;

2. power spectrum density is used: with the program-con-trolled computer of the power spectrum density input laboratory shaking table that obtains among the above-mentioned steps ⑸, realize the laboratory random vibration test of accurate road examination equivalence, reach the purpose of simulation test;

3. the application of shock duration velocity variable～loose figure of acceleration peak value statistics point: the damaged boundary curve map that contrasts shock duration velocity variable～loose figure of acceleration peak value statistics point and product, drop in a large number such as the statistics point among the loose figure of fruit dot in the damaged area of damaged boundary curve of product, product will suffer damage so, almost drop on such as the statistics point among the loose figure of fruit dot outside the damaged area of damaged boundary curve of product, then Product Status is good.

Claims (8)

1. transportation safety registering instrument data analysing method, the hardware system of the method operation comprises safety recording instrument and analyzes main frame, safety recording instrument sends the acceleration information that collects to the analysis main frame by USB port, it is characterized in that: it is as follows to the analytical approach step of image data to analyze main frame:

⑴ insert operating analysis main frame USB port with the safety recording instrument data line, and main frame is analyzed in the data importing that safety recording instrument gathers;

⑵ Outliers in Data Processing is rejected in the time domain pre-service of image data, and zero-mean is processed;

⑶ low-pass filtering: adopt the filtering of Butterworth filter device to the minimum high-frequency signal of goods infringement;

⑷ calculate acceleration-root-mean square;

⑸ draw power spectrum density: use period map method to draw the power spectrum density of vibration acceleration signal;

⑹ calculate shock duration: the starting point when first corresponding time of 5%-20% signal of getting acceleration signal shock response peak value is impact metre, terminal point when second the corresponding time of 5%-20% signal of getting acceleration signal shock response peak value is impact metre calculates shock duration;

⑺ draw the loose figure of shock duration velocity variable～acceleration peak value statistics point: to be that the maximal value of each acceleration signal is a little loose as the longitudinal axis counts as transverse axis, impact peak value scheme velocity variable in the shock duration;

⑻ data result is used: comprise the application of acceleration-root-mean square analysis, power spectrum density application and shock duration velocity variable～loose figure of acceleration peak value statistics point.

2. transportation safety registering instrument data analysing method according to claim 1 is characterized in that: the time domain pre-service of described step ⑵ image data, reject Outliers in Data Processing, and the concrete steps that zero-mean is processed are:

1. according to band limit relative method, the data of safety recording instrument acceleration collection are arranged sampled value time series { x according to time sequencing _i, i=1,2,3 ..., N, calculating sampling value seasonal effect in time series Fang Jun and mean square deviation: ${S_i}^2=\stackrel{\‾}{x_i^2}-{\stackrel{\‾}{x_i}}^2;$

2. Outliers in Data Processing is judged: if, x _i-kS _i＜x _I+1＜x _i+ kS _iX then _I+1Normal point, otherwise, x _I+1Be wild point, it rejected coefficient k value value between 3 to 5 for the data that are judged to be wild point from image data;

3. zero-mean is processed: average is the direct current distracter of collection signal, according to the Estimation of Mean of signal, before carrying out normalized set, average is removed.If { x _i, i=1,2,3 ..., N is the sampled value time series of acceleration, note sampled value average is

Have:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{x}_i$

New signal sequence after zero-mean is processed is { u _i, i=1,2,3 ..., N has

And the new signal serial mean is zero, namely

3. transportation safety registering instrument data analysing method according to claim 1 is characterized in that: described step (3) filtering is signal more than the 50HZ to the minimum high-frequency signal of goods infringement.

4. transportation safety registering instrument data analysing method according to claim 1 is characterized in that: described step ⑷ calculates acceleration-root-mean square for using time domain to ask the root mean square that collects acceleration signal:

$G_{\mathrm{RMS}}=\sqrt{\frac{1}{T}\underset{0}{\overset{T}{\∫}}{A_i}^2\mathrm{dt}}=\sqrt{\frac{1}{N}\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}{A_n}^2}$

Wherein, A _iBe acceleration sampling time sequential value, A _nBe sequential value in the acceleration sampling frequency domain, G _RMSBe the acceleration root-mean-square value in the time domain or in the frequency domain, according to the Parseval theorem, the acceleration root-mean-square value that time domain is interior or frequency domain is interior equates.

5. transportation safety registering instrument data analysing method according to claim 1 is characterized in that: the power spectrum density that described step ⑸ uses period map method to draw vibration acceleration signal is:

1. windowing process: for suppressing Power leakage, adopt Hanning window that the acceleration time domain signal that gathers is sampled and block;

2. Fourier transform: the acceleration signal after the windowing is carried out Fourier transform, obtain amplitude spectrum;

3. draw power spectrum density: with after each amplitude data in the amplitude spectrum behind the Fourier transform square divided by amount of frequency translation, both drawn final power spectral density results.

6. transportation safety registering instrument data analysing method according to claim 1, it is characterized in that: get among the described step ⑹ impulse response signal peak value first 5%, or 10%, or 15%, or the starting point of corresponding time of 20% signal when being impact metre, second 5% of getting the impulse response signal peak value, or 10%, or 15%, or the terminal point of corresponding time of 20% signal when being impact metre.

7. transportation safety registering instrument data analysing method according to claim 1, it is characterized in that: velocity variable is in the shock duration among the described step ⑺, begin front 0.4D integration amount the 0.1D behind the shock pulse from shock pulse, wherein, D is the shock duration that calculates among the step ⑹.

8. transportation safety registering instrument data analysing method according to claim 1 is characterized in that: the acceleration-root-mean square analysis of described step ⑻, power spectrum density uses and being applied as of shock duration velocity variable～acceleration peak value statistics loose figure of point:

1. acceleration-root-mean square analysis: the damaged condition of judging goods by the size of acceleration-root-mean square numerical value;

2. power spectrum density is used: power spectral density value is inputted the program-con-trolled computer of laboratory shaking table, realize the laboratory random vibration test of equivalent road examination, reach the purpose of accurate simulation test;

3. the application of shock duration velocity variable～loose figure of acceleration peak value statistics point: the damaged boundary curve map that contrasts shock duration velocity variable～loose figure of acceleration peak value statistics point and product, drop in a large number such as the statistics point among the loose figure of fruit dot in the damaged area of damaged boundary curve of product, product will suffer damage so, almost drop on such as the statistics point among the loose figure of fruit dot outside the damaged area of damaged boundary curve of product, then Product Status is good.

Images