Search

As the scale of the power system continues to expand, the power equipment fault rate gradually increases, which puts forward higher requirements for transformer fault diagnosis. Through fault diagnosis technology, faults can be found in advance during transformer operation, measures can be taken in time to reduce the possibility of accidents. It is found that the composition and content of dissolved gas in oil are closely related to the types of defects, and the composition and content of dissolved gas in transformer oil can play an important role in the prediction of transformer operating state and fault diagnosis. In order to obtain higher prediction accuracy, not only the composition and content of gas in oil, but also the correlation between fault types and characteristic quantities should be considered. In this paper, BP neural network algorithm based on association rules is adopted. Apriori algorithm of association rules reveals the association rules by mining high frequency terms in the data set of characteristic quantities. Apriori algorithm can effectively explore the confidence degree of association rules between feature quantity and running state and apply it as a weight value to the prediction link. Through the training steps of repeated forward training, reverse transfer and weight adjustment, the transformer operation state is finally determined.

Support vector machine (SVM) based on classification is applied for fault diagnosis of the automotive engine. The basic idea is to identify the information by using the trained SVM model to classify new fault samples. The data from the engine simulation model by AMESim software are fault features extracted, and these fault characteristic parameters have statistical property and specific physical meaning. Principal component analysis (PCA) is used to reduce the dimensions and redundancy of the data, and then these data are normalized as the input of SVM. The proposed method achieves accurate fault classification because SVM has good performance of classification and generalization ability, which is verified by the result of combining MATLAB/SIMULIK and AMESim. And the simulation results indicates that the proposed SVM based fault diagnosis method has achieved the better performance than the Artificial Neural Network, meeting the requirements of real-time diagnosis of the automotive engine.

A controller is designed based on BP neural network in this paper, which is applied to compensate the nonlinearity of electromagnetic valve of ABS, such as time-delay, etc. Thus, an intelligent controller is used to improve the quality of pressure adjustment in ABS. Data for training BP neural network is obtained from actual ABS. The cases for pressure-increase and pressure-decrease are studied, respectively. And the parameters of BP neural network are optimized. Finally, experiments including hardware in the loop are executed to verify the effectiveness of the provided design with the low adhesion road condition. And the quality of ABS is improved.

The 16 kb/s LD-CELP speech coding algorithm is submitted by AT&T to CCITT for the 16 kb/s speech coding standard which can be used in many applications. After having investigated the algorithm carefully and simulated it in FORTRAN and DSP TMS320C30 assembly language respectively the authors find that the difficulty implementing the 16 kb/s LD-CELP algorithm on a TMS320C30 is the large computational load and rather uneven distribution among 8 vectors in the same adaptation cycle. Some techniques are used in the program to reduce the computational load and a recursive window is used to compute autocorrelation functions. At last, the authors employ about 90% processor time of a TMS320C30 to implement the 16 kb/s LD-CELP coding algorithm in real-time. >

A new scheme to handle the echo canceller as an on-side job of a LD-CELP codec is studied in the paper. The authors give a novel approach to integrate the echo canceller (CCITT G.165) into the LD-CELP codec, such as CCITT G.728 "Coding of speech at 16 kbps using low-delay code-excited linear prediction" codec. >

Because the impulse response of an acoustic echo path may be several hundred milliseconds long, acoustic echo cancellation is a long-standing problem and hasn't been solved effectively. In order to solve this problem, in this paper an acoustic echo canceller is presented and a new self-orthogonalizing alternative constrained partitioned frequency-domain block LMS (PBFLMS) algorithm is put forward. This algorithm has good convergence ability, small misadjustment, low computation complexity and small delay. In addition, according to the application of the acoustic echo canceller, this algorithm is modified. Finally the whole echo canceller is implemented by using a single TMS320C30 processor in the ATD-C30 simulator and satisfying results are acquired.

Summary form only given. The conjugated organic ultra-thin films of 2,2-bithiophene(PT), 3-methythio-phene (PMT) and 3-bromothiophene (PBrT) are prepared dy electrochemical polymerization. The electrochemical properties of the single layer films well studied. And the absorption spectra of the films in reduction and oxidation conditions were measured respectively. The data of thermo-emf confirmed that the film of PBrT is p-type behaviore, while the films of PT and PMT exhibited the feature of weak donor. The single-layer and bilayer film cells, with ITO/polymer/Au and ITO/polymer1/polymer2/Au sandwich structure, were composed of PT, PMT and PBrT. No distinct reatifying behaviore was observed all of the single-layer film cells. This fact suggests that there should be no any juction or Schottky berrier between ITO and polymer or Au and polymer interface. The bilayer film cells, e.g. ITO/PBrT/PMT/Au and ITO/PBrT/ PT/Au, show excellent rectifying ability similar to a p-n junction diode and some extent of photo-electric conversion. The rectifying ratio reach about 200. However, the cells of ITO/PT/PBrT/Au and ITO/PMT/PBrT did not show rectifying characteristic. The phenomenon should be attributed to the overoxidation during electropolymerization.