Some Key Techniques Of Cable Length Measurement Based On Time Domain Reflectometry

All kinds of wires and cables have been widely used with the rapid development of the national economy. The wire and cable industry have developed rapidly, and gradually expanded the production scale and market share. Cable is an important commodity, and its length measurement precision is strictly formulated in the national standard. However, the issue of measuring precision in the wire and cable market becomes more and more prominent. It is significant to measure the cable length precisely, rapidly and economically. Compared with the traditional measurement methods, time domain reflectometry (TDR) technology has an advantage of non-destruction, portability and high-precision, which is an ideal cable length measurement method.Based on much and detailed analysis of the status of domestic and external research, the theory, methods and key technologies involved in the TDR cable length measurement are investigated in this dissertation. The main contents and innovative achievements of this dissertation are as follows:The theory of the TDR cable length measurement is investigated. The equivalent circuit model of cable is established according to the transmission line theory. The electromagnetic wave propagation theory and the regular pattern of the TDR signal propagating in cables are studied. To further illustrate the TDR cable length measurement technology, this dissertation develops the transmission line transient analysis method based on finite-difference time domain (FDTD).The cable length measurement experimental systems are designed based on the TDR principle. The principle error of the TDR cable length measurement has also been analyzed. The influence of measuring numbers, counting pulse frequency and traveling wave velocity on measuring accuracy is analyzed. Aiming at the key segment which affected the measuring precision, two kinds of high precision time interval measurement methods based on direct pulse-counting method and FPGA are proposed.The nonlinear error is analyzed and compensated. In order to reduce the nonlinearity caused by attenuation and dispersion effecting on the measuring accuracy of the TDR cable length measurement system, this dissertation establishes the transmission characteristic mathematical model of the TDR cable length measurement, and derives the regular pattern of the pulse signal propagating in cable. The nonlinear characteristic caused by attenuation and dispersion is analyzed, thus the corresponding compensation method of the nonlinear error is proposed by studying the law of nonlinear error. Finally, the measuring accuracy of the TDR cable length measurement system is improved.The velocity characteristics of the TDR cable length measurement are studied. The frequency characteristics of the traveling wave propagation velocity and dielectric constant are analyzed respectively. The effect of cable structure parameters, insulation materials, conductivity materials, and the cable crimp levels on traveling wave propagation velocity is observed and analyzed. The law of the traveling wave propagation velocity is studied at different temperature. The velocity temperature error compensation coefficients of some cables are obtained according to the experimental data by measuring the traveling wave velocity in these cables at different temperature. Finally, the TDR cable length measurement model based on genetic algorithm optimizing neural network is established in order to reduce the velocity effecting on the measuring accuracy of the TDR cable length measurement system,The TDR signal detecting and processing technology is studied. Aiming at the singular points detected by wavelet analysis is susceptible to be affected by noise and the measurement error is comparatively large, the reflected wave de-noising technology based on wavelet analysis has been investigated. The level discriminant function is built to indicate the optimal decomposition level. A new wavelet decomposition level adaptive selection algorithm is proposed. This algorithm is capable to adaptively choose the optimal decomposition level to achieve the optimal noise elimination effect. The results of simulation and engineering applications indicate that the wavelet de-noising processing with the decomposition level determined by the adaptive selection algorithm can effectively filter out the noise from the TDR signal and then improve the measuring accuracy of the TDR cable length measurement system. The measuring accuracy of the TDR cable length measurement system relies on the recognition of the reflected wave. In order to correctly identifying and detecting the reflected wave, this dissertation studies four reflected wave recognition algorithms, which are established according to the threshold method, centroid method, polynomial fitting method and the wavelet modulus maxima method respectively. The recognition accuracies of the reflected wave arrival time for these algorithms are compared through practical applications.