CN102183951A - Device for monitoring state of rotary bearing and diagnosing fault based on laboratory virtual instrument engineering workbench (Lab VIEW) - Google Patents

Abstract

The invention relates to a slewing bearing monitoring and fault diagnosis device based on a virtual instrument development platform (LabVIEW), including an industrial computer, a detection device, a voltage stabilized power supply, a data acquisition module, a data processing module and a fault diagnosis module. The industrial computer is equipped with A PCI data acquisition card including an analog input channel, an acceleration sensor and a signal conditioning circuit are arranged in the detection device. The signal conditioning circuit is used to amplify and filter the signal picked up by the acceleration sensor; the data acquisition module inputs the collected signal into the industrial computer through program control, and writes the DAQmx data acquisition driver program on the LabVIEW platform to control the sampling frequency of the signal. input and output etc. The data processing module includes signal denoising display, storage, and feature extraction. The data processing module eliminates noise interference, extracts the time domain and frequency domain feature quantities of the signal, and prepares for fault diagnosis; the fault diagnosis module is based on the extracted characteristics of the slewing bearing. The fault information is obtained from the data, and the diagnosis results and processing opinions are output. The device has the advantages of simple structure, high openness, good real-time performance, friendly human-machine interface, and can automatically identify the fault of the slewing bearing without the participation of diagnostic experts, and has high reliability.

Description

Slewing Bearing Condition Monitoring and Fault Diagnosis Device Based on Virtual Instrument Development Platform

technical field

The invention belongs to the field of mechanical equipment fault diagnosis, in particular to a slewing bearing monitoring and fault diagnosis device based on a virtual instrument development platform (LabVIEW).

Background technique

Large-scale slewing bearings are important components of various construction machinery and equipment, such as tower cranes, portal cranes, and excavators. Its failure or not directly affects the operation of the whole machine, and is also related to the safety of construction personnel and production equipment. However, due to its high price, difficult disassembly and assembly, and long maintenance cycle, it is necessary to carry out condition monitoring and early fault diagnosis of large slewing bearings, which can avoid sudden failures and unnecessary disassembly and inspection, thereby improving economic benefits.

During the operation of the slewing bearing, it is not uncommon for accidents to occur due to faults. Especially during the working process, the components of the slewing bearing are subjected to extrusion force or component wear, which causes the continuous change of its working state and makes it difficult to identify its faults. The traditional slewing bearing fault diagnosis is usually based on the engineer's experience judgment or offline signal analysis. The disadvantage of this approach is that it is inefficient in diagnosis and relies too much on the expertise of diagnostic experts.

In recent years, with the development of computer technology, microelectronics technology, sensor technology and artificial intelligence technology, intelligent fault diagnosis system has been applied to many large construction machinery equipment. This patent proposes a large-scale slewing bearing fault diagnosis system based on the NI virtual instrument platform LabVIEW, which can quickly and accurately locate the fault of the slewing bearing and automatically provide fault information.

Contents of the invention

The purpose of the present invention is to provide a slewing bearing state monitoring and fault diagnosis device based on the virtual instrument development platform LabVIEW. The device can diagnose common faults of slewing bearings online without the participation of fault diagnosis experts, and has a high accuracy rate.

The present invention utilizes the major achievements in the signal processing field, the artificial intelligence field and the computer field in recent years, and combines the virtual instrument technology represented by LabVIEW of NI Corporation of the United States to develop a slewing bearing state monitoring and fault diagnosis device based on LabVIEW.

The slewing bearing fault diagnosis device based on the virtual instrument development platform proposed by the present invention is composed of an industrial computer 1, a detection device 3, a voltage stabilized power supply 7, a data acquisition module 11, a data processing module 12 and a fault diagnosis module 13. In the industrial computer 1 A PCI data acquisition card 2 comprising an analog input channel is provided, an acceleration sensor 5 and a signal conditioning circuit 4 are arranged in the detection device 3, and the signal conditioning circuit 4 is used to amplify, filter, etc. the signal picked up by the acceleration sensor 5; The module 11 inputs the collected signal into the industrial computer 1 through program control, writes the DAQmx data acquisition driver program on the LabVIEW platform, and controls the sampling frequency, input and output of the signal. The data processing module 12 includes signal denoising display, storage, and feature extraction, all of which are developed based on the function library that comes with LabVIEW. The data processing module 12 eliminates the interference of noise, extracts the time domain and frequency domain feature quantities of the signal, and prepares for fault diagnosis; the fault diagnosis module 13 obtains fault information according to the extracted characteristic data of the slewing bearing, and outputs diagnosis results and treatment opinions.

In the present invention, the PCI data acquisition card 2 adopts the PIC-6023E data acquisition plug-in card released by NI Company, and collects the picked-up signals into the industrial computer 1 through the standard computer interface PCI slot.

In the present invention, the acceleration sensor 5 includes a radial acceleration sensor and an axial acceleration sensor, which are uniformly arranged along the fixed circle of the slewing support respectively, and pick up the radial and axial vibration signals of each measuring point of the slewing support.

In the present invention, the fault diagnosis module 11 adopts BP neural network for fault classification, and extracts the mean square value of the signal as the input of the neural network.

In the present invention, the fault diagnosis module 13 adopts the method of calling the designed MATLAB neural network program through the MatlabScript node in the function library in the LabVIEW development platform to construct an online fault diagnosis system for the slewing bearing.

The beneficial effects of the present invention are:

1. Modular structure is adopted, which has high reliability. Friendly man-machine interface, easy to operate.

2. Fault diagnosis is real-time, and fault information is displayed automatically.

3. The system adopts virtual instrument technology, which has the characteristics of openness and can integrate fault diagnosis of various components.

Description of drawings

Fig. 1 is the structural representation of the present invention.

Fig. 2 is the arrangement diagram of the vibration measuring points of the slewing bearing of the present invention.

Fig. 3 is a schematic diagram of the program structure of the present invention.

Numbers in the figure: 1 is industrial computer, 2 is PCI data acquisition card, 3 is detection device, 4 is signal conditioning circuit, 5 is acceleration sensor, 6 is slewing bearing test platform, 7 is regulated power supply, 8 is outer ring, 9 is a rolling body, 10 is an inner ring, 11 is a data acquisition module, 12 is a data processing module, and 13 is a fault diagnosis module.

Detailed ways

Below in conjunction with accompanying drawing, will further describe the specific embodiment of the present invention.

Embodiment 1: The LabVIEW-based slewing bearing state monitoring and fault diagnosis device proposed by the present invention uses a virtual instrument LabVIEW development platform which is a graphical programming system developed by NI Corporation. It is a powerful integrated development environment with a huge function library, which completely integrates communication with hardware such as GPIB, VXI, PCI and plug-in data acquisition cards.

Accompanying drawing 1 is the block diagram of hardware structure of the present invention, comprises industrial computer 1, detection device 3 and stabilized voltage supply 7, is provided with a PCI data acquisition card 2 that comprises analog input channel in the industrial computer 1, is provided with acceleration in the detection device 3 Sensor 5 and signal conditioning circuit 4. The present invention takes the vibration acceleration of the slewing bearing as the monitoring parameter, so the arrangement of the acceleration sensor 5 is based on the vibration characteristics of the slewing bearing. Since the slewing bearing bears a large axial force and overturning moment, and each rolling element bears a different load, it is necessary to measure the axial and radial acceleration vibration signals of the slewing bearing separately, and arrange them evenly along the fixed ring . The layout of the acceleration sensor in the present invention is shown in Figure 2, and 4 measuring points are evenly arranged along the fixed circle.

The amplitude of the current or voltage signal obtained from the acceleration sensor 5 is very low, generally in the milliamp or millivolt level, which is not suitable for collection and transmission, and because there are many factors causing the vibration of the slewing bearing, there are many interferences, so it must be passed The signal conditioning circuit 4 amplifies and filters the signal detected by the acceleration sensor 5, and outputs a suitable voltage or current after shaping.

The rotation frequency of the slewing bearing is low, and the fault frequency of the inner and outer rings is relatively close. Therefore, high signal acquisition accuracy and resolution are required in the monitoring and diagnosis process. The PCI data acquisition card 2 in the present invention adopts the data acquisition card PCI-6023E based on the PCI slot introduced by NI Company, including 16 analog input channels, 8 digital I/O ports, and 8 vibration signals of the slewing ring using differential The input method captures the signal into the LabVIEW interface.

The module of LabVIEW among the present invention is as shown in accompanying drawing 3, comprises: data acquisition module 11, data processing module 12, fault diagnosis module 13. The data acquisition module 11 is written with the function in DAQmx and is located in the measurement I/O function module of LabVIEW. Since the present invention is a real-time fault diagnosis, a continuous sampling method must be adopted. The functions of the data processing module 12 include signal denoising, display, storage, and feature extraction. During the operation of the slewing bearing, each component sends out its own specific feature signals. These signals vary with the damage or wear of a single component and the pressure. And change. Slewing bearings usually work in low-speed and heavy-load environments, and their low-frequency fault characteristic signals are often interfered by complex environmental noise, which affects the final diagnosis effect. The invention uses wavelet threshold noise removal method to eliminate noise interference in the vibration signal of the slewing bearing, selects db4 wavelet, performs seven-level decomposition, and removes noise through threshold function and wavelet reconstruction. This program is written in Matlab, and this program is called through the MatlabScript node in LabVIEW. The wavelet transform effectively filters out the noise, and the feature vector of the fault is extracted by calculating the root mean square value of the reconstructed wavelet coefficients.

The fault diagnosis module 13 of the present invention adopts a fault diagnosis method based on BP neural network, and utilizes its powerful nonlinear mapping function to realize fault classification. Studies have shown that a neural network with a 3-layer network topology (input layer, intermediate layer, output layer) can arbitrarily approximate any complex continuous function, so complex nonlinear mapping from feature space to fault space can be achieved. Since the present invention measures 8 vibration signals in total, the BP neural network has 8 input interfaces to input the root mean square of the 8 vibration signals respectively, and normalizes them as feature vectors. The operating states of the slewing bearing mainly include normal, pitting, loose bolts, structural deformation or cracking, and high local resistance, which are divided into five modes of encoding. Therefore, a total of 5 output ports are set in the BP neural network as the output vector of the fault. The design of BP neural network in the present invention is realized by programming in Matlab, and this program is called by MatlabScript node in LabVIEW.

The present invention needs to train different neural networks for different types of slewing bearings, so before the BP neural network is used for fault diagnosis, it is necessary to obtain the vibration acceleration signal of the slewing bearing fault, extract the feature vector, and design the network training program in Matlab, correct Network weights, and then perform fault diagnosis.

Claims (5)

1. A slewing bearing fault diagnosis device based on a virtual instrument development platform, consisting of an industrial computer (1), a detection device (3), a stabilized power supply (7), a data acquisition module (11), a data processing module (12) and Composed of a fault diagnosis module (13), it is characterized in that a PCI data acquisition card (2) including an analog input channel is provided in the industrial computer (1), and an acceleration sensor (5) and a signal conditioning circuit are provided in the detection device (3) (5) and the signal conditioning circuit (4), the signal conditioning circuit (4) is used to amplify and filter the signal picked up by the acceleration sensor (5); the data acquisition module (11) inputs the signal collected into the industrial computer through program control In (1), write the DAQmx data acquisition driver program in the LabVIEW platform to control the sampling frequency, input and output of the signal; the data processing module (12) includes denoising display, storage, and feature extraction of the signal, all based on the built-in The function library is developed; the data processing module (12) eliminates the interference of noise, extracts the time domain and frequency domain characteristic quantities of the signal, and prepares for the fault diagnosis; the fault diagnosis module (13) obtains according to the characteristic data of the extracted slewing bearing Fault information, output diagnosis results and processing advice. 2. the slewing bearing fault diagnosis device based on virtual instrument development platform according to claim 1, is characterized in that described PCI data acquisition card (2) adopts the PIC-6023E data acquisition plug-in card of NI company, through computer standard interface PCI The slot collects the picked-up signal into the industrial computer (1). 3. The slewing bearing fault diagnosis device based on a virtual instrument development platform according to claim 1, characterized in that the acceleration sensor (5) includes a radial acceleration sensor and an axial acceleration sensor, respectively along the fixed circle of the slewing bearing Evenly arranged to pick up the radial and axial vibration signals of each measuring point of the slewing bearing. 4. the slewing bearing fault diagnosis device based on virtual instrument development platform according to claim 1, is characterized in that described fault diagnosis module (11) adopts BP neural network to carry out fault classification, by extracting the mean square value of signal as neural network input of. 5. the slewing bearing fault diagnosis device based on virtual instrument development platform according to claim 1, is characterized in that described fault diagnosis module (13) adopts the MATLAB designed by the MatlabScript node in the function library in the LabVIEW development platform The method of neural network program is used to build an online fault diagnosis system for slewing bearings.

Images