Nan Pan

Combined failure acoustical diagnosis based on improved frequency domain blind deconvolution

According to gear box combined failure extraction in complex sound field, an acoustic fault detection method based on improved frequency domain blind deconvolution was proposed. Follow the frequency-domain blind deconvolution flow, the morphological filtering was firstly used to extract modulation features embedded in the observed signals, then the CFPA algorithm was employed to do complex-domain blind separation, finally the J-Divergence of spectrum was employed as distance measure to resolve the permutation. Experiments using real machine sound signals was carried out. The result demonstrate this algorithm can be efficiently applied to gear box combined failure detection in practice.

Machine fault diagnosis based on Frequency-Domain Blind Deconvolution

On the basis of introducing the model of Frequency-Domain Blind Deconvolution (FDBD), key techniques in mechanical signal extraction were comprehensively related and analyzed in this paper, which include the methods of suppressing the difference between circular and partial convolution by coordinating the relationship between FFT size and length of each frequency bin or Modified Discrete Fourier Transform, the methods of removing the permutation indeterminacy (methods based on consistency of filter coefficients, DOA methods, Split Spectrum methods, methods based on Nonlinear Function, etc.), the applications of Complex-Domain Blind Separation Algorithms which based on explicit tensor eigenvalue decomposition and nonlinear functions. Aimed at vibration and acoustic signal feature extraction in complex environment and equipment with complex mechanical structures, the application values of FDBD and its research status in machinery condition monitoring and fault diagnosis were reviewed and summarized. Finally, the main problems which need to be studied further in this area were pointed out.

Vibration Signal Analysis of Bearing Based on EMD and Resonance Demodulation

In order to monitor the operation state and implement fault diagnosis of rolling bearing in rotating machinery, this paper presents a method of fault diagnosis of rolling bearing, which is based on EMD and resonance demodulation. Using EMD to decompose the signal, which comes from QPZZ-II experimental station, the components of intrinsic mode function (IMF) will be obtained. Then, calculating the correlation coefficient of each IMF component, the highest correlation coefficient of IMF component will be analyzed by resonance demodulation. Finally, the experimental results show that the method can accurately identify and diagnose the running state and bearing fault type.

Application of Frequency-Domain Blind Deconvolution in Mechanical Fault Detection

On the basis of summing up the Frequency-Domain Blind Deconvolution (FDBD), a method combine Complex-Domain FastICA algorithm and amplitude correlation was proposed to extract the typical defect signals from mechanical equipment. The application in combined failure rolling bearing acceleration signals demonstrate that, comparing with the existing Time-Domain Blind Signal Processing methods, FDBD has more advantages and better prospects in mechanical fault detection.

Bearing fault vibration diagnosis using frequency domain semi-blind extraction method

It is usually not easy to extract fault features from the vibration signals directly, since the complexity of the mechanical structure and the serious background interference in industry testing site. In order to deal with these kinds of monitoring problems, a mechanical failure diagnosis method based on reference signal frequency domain semi-blind extraction is proposed. In this method, dynamic particle swarm algorithm is used to construct improved multi-scale morphological filters which applicable to mechanical failure in order to weaken the background noises; thus reference signal unit semi-blind extraction algorithm is applied to do complex components blind separation band by band, coupled improved KL-distance of complex independent components are employed as distance measure to resolve the permutation; finally the estimated signal could be extracted and analysed by envelope spectrum method. Comparing to the time-domain blind deconvolution algorithm based on fuzzy clustering, it has several advantages such as more effectively and more accurately. Results from rolling bearing fault diagnosis experiment validate the feasibility and effectiveness of proposed method.

The Field Experiments for Mixed Vehicle-Carrying Signals Acquisition Based on the Compact RIO

The usability and reliability of conventional vehicle carrying system is not always satisfy when customers were using it, bugs occur occasionally. To solve the problem, this paper presented a series of field experiment plan to find and modify bugs so that its stability and usability would be guarantee. This paper introduced from the sensor calibration, Static floor check, vibrate test, braking test, long trip test to signal analysis, the high reliability of vehicle carrying system based on Compactrio is proved.

Study of Cluster-Footprint Analysis Testing Algorithm and System

The cluster-footprint left at a crime scene is one of the most common trace evidence in the field of criminal investigation. Through analyzing and calculating the plane footprint, the investigators can get some important feature information such as step size, age and height, etc. However, this way is difficult to apply to reality without any errors in the process of manual calculations. Information management is more complicated in the later, the efficiency is relatively low too. To solve above problems, a testing system is designed and developed to analyze cluster-footprint. This system is based on the LabVIEW platform, which can realize image acquisition, stride characteristics extraction, information management and so on. Actual analysis results show that this system can effectively achieve these functions, working accuracy and efficiency can meet requirements of criminal investigation process.

Study of Intelligent Evidence Management System

Based on the LabVIEW platform and SQL server database, the intelligent management system of physical evidence is developed, which implements the intellectualization and networking and makes the staff in the public security organization manage the evidence more conveniently, safely and efficiently. The key techniques including the design of evidence room, the hardware design and form controls of browsing information were introduced in detail. Finally, reliability, availability and stability of the system were proved through testing by the public security department for a period of time.

The Simultaneous Acquisition of the Mixed Vehicle-Carrying Signals Based on the CompactRIO

In order to realize a simultaneous acquisition of the mixed vehicle-carrying signals, this paper put forwards a new acquisition system that synchronous controls the digital IO module, the vibration-and-noise module, the electrical pressure module, the temperature-measuring module, as well as the CAN module, under the mode of FPGA, which is based on the embedded platform CompacRIO, and it could also be controlled by both PC and PDA. The operating procedures of this system are highly reliable and compatible, specifically designed to be people-oriented. This paper gives a full introductions to how the multi-mixed signals are simultaneously collected, the deepening of the FIFO on FPGA, and how it could be controlled by PDA. Finally, the efficiency and stability of the system would be proved through the practical pressure tests of data-acquisition.