CN202582898U

CN202582898U - Flush type bearing failure intelligent diagnosing device based on ARM and DSP

Info

Publication number: CN202582898U
Application number: CN 201120537269
Authority: CN
Inventors: 严如强; 赵锐
Original assignee: Southeast University
Current assignee: Southeast University
Priority date: 2011-12-20
Filing date: 2011-12-20
Publication date: 2012-12-05
Anticipated expiration: 2021-12-20

Abstract

The utility model discloses a flush type bearing failure intelligent diagnosing device based on an ARM and a DSP. The device comprises a signal input module connected with a vibration signal of a detected bearing, wherein the signal input module is connected with a core operation module, a serial communication module and a control module in sequence. A structure of DSP plus ARM is used, and an ARM sub-system and a DSP sub-system are communicated through an RS232 serial bus; and under the system structure, real-time sampling and calculation as well as the management and control of a system can be executed in parallel, and the two sub-systems are synchronized on task execution by virtue of communication between the two sub-systems. The dual-CPU structure-based system plays a critical role in protecting the real time property, and a multi-scale envelope spectrum analysis algorithm based on Morlet wavelet transformation is adopted in the core algorithm, and is realized in flush-type hardware for the first time, so that not only can the accuracy of bearing failure diagnosis be improved, but also the portability is realized.

Description

A kind of insert bearing intelligent fault diagnosing device based on ARM and DSP

Technical field

The present invention relates to a kind of bearing fault intelligent diagnostics device based on DSP and ARM double-core embedded type technology.

Background technology

The phenomenon of the rotating machinery cisco unity malfunction that causes by bearing fault as everyone knows, (such as finding that the problem that is caused by bearing has accounted for more than 40% of entire machine fault in the motor reliability study by general electronic corporation and the co-anchor of IEEE Industry Applications Society) of common occurrence.Thereby the exploitation of corresponding various bearing failure diagnosis systems obtains the common concern of industry member and academia with research.Bearing failure diagnosis system traditionally is based on classical Fourier analysis, comes fault is detected and diagnoses through the signal that records being carried out obtain after the Fourier transform characteristic frequency spectrums such as amplitude, phase place.The existence that has reflected bearing fault whether.Because the vibration signal that bearing fault produces in formation and evolution has transient state property and non-stationary; So Fourier transform is that the frequecy characteristic of a kind of equalization of Fourier pair signal in the whole time period representes (to be that it is to see that from overall angle the frequency spectrum of signal constitutes for the analysis of this type signal; And to need putative signal be stably on whole time shaft); It can not reflect the spectrum signature of signal in the local time zone, and this just defective has reflected the limitation of Fourier transform.And various non-stationary signal disposal routes such as time frequency analysis, wavelet transformation and Hilbert-Huang transform etc. can be used for the analysis of this type of signal, so research and development are significant for helping to improve accuracy rate of diagnosis based on the bearing failure diagnosis system of non-stationary signal processing.

Summary of the invention

In order to improve the accuracy rate of bearing failure diagnosis system, the invention provides a kind of insert bearing intelligent fault diagnosing device based on ARM and DSP.

For realizing above-mentioned purpose, the present invention adopts following technical scheme:

A kind of insert bearing intelligent fault diagnosing device based on ARM and DSP, it comprises the signal input module that links to each other with bearing vibration signal to be detected, signal input module links to each other with its main operational module, serial communication modular and control module successively.

Wherein, signal input module comprises sensor, charge amplifier, analog multichannel switch, low-pass filter and A/D modular converter; Bearing vibration signal to be detected is linked into the input end of sensor, and the output terminal of sensor links to each other with charge amplifier, analog multichannel switch, low-pass filter and A/D modular converter successively.

Wherein, described its main operational module comprises microprocessor, real-time clock module, the signal after the input termination signal input module of a microprocessor is handled, the output termination serial communication modular of a microprocessor; No. one microprocessor also links to each other with real-time clock module.

Wherein, control module comprises microprocessor, power module, storer and LCD touch-screen No. two; Described No. two microprocessors link to each other with power module, storer, LCD touch-screen and serial communication modular respectively.

Compared with prior art, beneficial effect of the present invention is following:

1. the present invention has used the structure of DSP+ARM; DSP selects 32 fixed-point DSP chip TMS320F2812 of the up-to-date release of American TI Company for use, the characteristics of utilizing high-speed computation and multiple of DSP to go up peripheral hardware accomplish seizure to input signal, sampling, conversion and with the work such as communication of ARM; ARM part is hardware devices such as core, the LCD of additional peripheral, touch-screen with the ARM9 family chip S3C2440A of Samsung company, in order to accomplish statistics to data, storage, with functions such as DSP and upper machine communication, man-machine interaction.Carry out communication through the RS232 universal serial bus between ARM subsystem and the DSP subsystem.Under this system architecture, the management of real-time sampling and calculating and system and control can executed in parallel, and through both sides' communication two sub-systems realized synchronously on task is carried out.This system based on dual-cpu structure has played key effect for the assurance of real-time.

2. core algorithm of the present invention has adopted the multiple dimensioned envelope spectrum analytical algorithm method based on the Morlet wavelet transformation, and is to realize at embedded hardware for the first time, has improved the accuracy of bearing failure diagnosis, has realized portability simultaneously.

3. this user interface utilizes QT/Embedded 4.5 to design, and the friendly interface of system is directly perceived, and is succinct, and our system is based on the most general instantly touch screen technology simultaneously, has applicability.

Description of drawings

Fig. 1 is a theory diagram of the present invention.

Embodiment

Below in conjunction with accompanying drawing and instantiation the present invention is described further.

As shown in Figure 1, a kind of insert bearing intelligent fault diagnosing device based on ARM and DSP of the present invention comprises signal input module 1, its main operational module 2, serial communication modular 3 and and control module 4.The input end of signal input module 1 links to each other with bearing to be detected, and the output terminal of signal input module 1 links to each other with its main operational module 2, and its main operational module 2 is connected with control module 4 through serial communication modular 3.

Signal input module 1 is that suitable digital signal is input to a microprocessor 21 in its main operational module 2 with bearing vibration signal transformation.Signal input module 1 comprises that sensor 11, charge amplifier 12, simulation multichannel leave 13, low-pass filter 14 and A/D modular converter 15; The input end of sensor 11 is connected with bearing vibration signal to be detected, and the output terminal of sensor 11 links to each other with charge amplifier 12, analog multichannel switch 13, low-pass filter 14 and A/D modular converter 15 successively; The output terminal of A/D modular converter 15 links to each other with a microprocessor 21 of its main operational module 2.

Sensor 11 in the signal input module 1 becomes electric signal output with bearing vibration conversion of signals to be detected; The signal of output is a simulating signal; Through charge amplifier 12 this faint electrical signal conversion is become A/D modular converter 15 needed input voltages; And analog multichannel switch 13 has guaranteed that total system can realize single channel and two-way collection; Low-pass filter 14 can be with the radio-frequency component elimination of input signal, and A/D modular converter 15 is that digital signal is exported to microprocessor 21 No. one with these analog signal conversion.

Sensor 11 can be selected the acceleration transducer of Denmark B&K company 4370 types for use; Charge amplifier 12 adopts DFH-6A type charge amplifier; Analog multichannel switch 13 adopts the analog multichannel switch of 8 channel cmos; Low-pass filter 14 adopts the integrated Butterworth SCF of CD4051LC04/MF4A.50 type monolithic, and A/D modular converter 15 adopts the MAX125 chip.

Its main operational module 2 mainly is the whole signal acquiring system of scheduling, and carries out data processing and data analysis computing.Its main operational module 2 comprises an interconnected microprocessor 21, real-time clock module 22, and power module is above-mentioned module for power supply.No. one microprocessor 21 links to each other with A/D modular converter 15, real-time clock module 22, and No. one microprocessor 21 also links to each other with No. two microprocessors 41 through serial communication modular 3.A microprocessor 21 is responsible for the collection of analog quantity, the A/D conversion, and real time data processing communicates with No. two microprocessors 41 in addition.No. one microprocessor 21 carries out the multiple dimensioned envelope spectrum analytical algorithm of the algorithm use of real time data processing based on the Morlet wavelet transformation.No. one microprocessor 21 adopts 32 fixed DSP microprocessor chip TMS320F2812.Real-time clock module 22 is the built-in circuit that microprocessor 21 provides accurate timing, is mainly used in timing, communication clock generator, time interrupt source or the like.

Serial communication modular 3 adopts the interface of RS-232 level transferring chip MAX2332 as RS232, realizes the conversion and the full-duplex communication of signal.

Control module 4 mainly is to be in charge of total system, and display of user interfaces is responsible for the mutual of user and system simultaneously, shows operation result--spectrogram.Control module 4 comprises microprocessor 41, power module 42, storer 43 and LCD touch-screen 44 No. two.No. two microprocessor 41 links to each other with power module 42, storer 43 and LCD touch-screen 44 respectively.

No. two microprocessor 41 adopts ARM9 series flush bonding processor S3C2440A; No. two microprocessors 41 are as control module; Be in charge of total system, and the control touch-screen, operation result--spectrogram on touch-screen, shown; Be responsible for the mutual of user and system simultaneously, control stopping and starting of total system.

Power module 42 adopts integrated voltage stabilizing chip LM2596, and power module 42 is No. two microprocessor 41 power supplies.

Storer 43 is made up of flash memory NOR FLASH chip SST39VF160 and static memory chip SDRAM K4S5616332D, is that No. two little processing 41 provide storage space.

And LCD touch-screen 44 adopts four-wire ohm touch-screen and ADS7843 as electric resistance touch screen translation interface chip; The signal of four-wire ohm touch-screen is received the X+ of translation interface chip ADS7843, Y+, X-; Y-terminal, and the DCLK terminal of ADS7843, the CS terminal; The DIN terminal, DOUT, terminals P ENIRQ terminal, BUSY terminal link to each other with the PG mouth of ARM9 series flush bonding processor S3C2440A.No. two little processing 41 utilize four-wire ohm touch-screen translation interface chip ADS7843 to be connected with LCD touch-screen 44.

The course of work of the present invention: No. two microprocessor 41 brings into operation, display graphical user interfaces on LCD touch-screen 44, and the user comprises the sampling time according to some parameters of situation input of scene and bearing; The initial center frequency, by centre frequency, the change of frequency step-length; Import these parameters into microprocessor 21 No. one through serial communication modular 3 then, and notify a microprocessor 21 to begin sampling, signal input module 1 begins to gather vibration signal like this; Obtain electric signal through sensor 11; Electric signal through charge amplifier 12 obtains amplifying through low-pass filter 14 elimination radio-frequency components, obtains digital signal through A/D modular converter 15 and imports microprocessor 21 into No. one again; A microprocessor 21 combines the parameter of input; Employing obtains tested frequency spectrum data based on the multiple dimensioned envelope spectrum analytical algorithm of Morlet wavelet transformation, then data is passed to microprocessor 41 No. two through serial communication modular 3; No. two microprocessors 41 call display routine then, and spectrogram is presented at LCD touch-screen 44.Operating personnel can check failure-frequency according to spectrogram, have so just accomplished the process of a bearing failure diagnosis.

Wherein the course of work of a microprocessor 21 is following: earlier input signal is carried out Fast Fourier Transform (FFT) FFT; And then carry out inner product with the expression formula of Morlet small echo on frequency domain and calculate; Obtain the wavelet coefficient of this signal at frequency domain; Obtain the wavelet coefficient of this signal on time domain through inversefouriertransform IFFT again; Continuation takes absolute value to wavelet coefficient and obtains the small echo envelope, carries out the envelope spectrum that Fast Fourier Transform (FFT) FFT obtains small echo at last, thereby can obtain the failure-frequency composition of bearing according to envelope spectrum.

Claims

1. insert bearing intelligent fault diagnosing device based on ARM and DSP; It is characterized by: it comprises the signal input module (1) that links to each other with bearing vibration signal to be detected, and signal input module (1) links to each other with its main operational module (2), serial communication modular (3) and control module (4) successively.

2. a kind of insert bearing intelligent fault diagnosing device based on ARM and DSP according to claim 1 is characterized by: signal input module (1) comprises sensor (11), charge amplifier (12), analog multichannel switch (13), low-pass filter (14) and A/D modular converter (15); Bearing vibration signal to be detected is linked into the input end of sensor (11), and the output terminal of sensor (11) links to each other with charge amplifier (12), analog multichannel switch (13), low-pass filter (14) and A/D modular converter (15) successively.

3. a kind of insert bearing intelligent fault diagnosing device according to claim 1 and 2 based on ARM and DSP; It is characterized by: described its main operational module (2) comprises a microprocessor (21), real-time clock module (22); Signal after the input termination signal input module (1) of a microprocessor (21) is handled, the output termination serial communication modular (3) of a microprocessor (21); A microprocessor (21) also links to each other with real-time clock module (22).

4. a kind of insert bearing intelligent fault diagnosing device based on ARM and DSP according to claim 1 is characterized by: control module (4) comprises No. two microprocessors (41), power module (42), storer (43) and LCD touch-screen (44); Described No. two microprocessors (41) link to each other with, LCD touch-screen (44) and serial communication modular (3) with power module (42), storer (43) respectively.