| With the increasing development of automotive electronic technology, a growing number of electrical and electronic products are widely used in automotives to meet the demand for the modern automotives in comfort, intelligence, safety, energy-saving, environmental protection, etc, which makes the automotive electromagnetic compatibility(EMC) facing serious challenges. Therefore, automotive EMC technology has become the research focus for automotive manufacturers and research institutions at home and abroad. As the main body of automotive electronic network, automotive wiring harness ties different electrical and electronic equipments together, and the types and quantity of automotive wiring harness are growing rapidly with the wide application of electrical and electronics equipment. The EMC research for automotive wiring harness usually commences on crosstalk, electromagnetic radiation and electromagnetic radiation sensitivity. Thereinto, automotive wiring harness crosstalk can not only cause transfer error or data-packets dropout on the wire and the false trigger of function equipment, but also effect the electromagnetic radiation and electromagnetic radiation sensitivity of the electrical and electronic equipments and harness wire. Therefore, the research on automotive wiring harness crosstalk has important significance, which can provide strong theoretical basis for solving automotive EMC and avoiding EMC problems in the early stages of the vehicle design. And this will be helpful in shortening the development period and reducing the cost of automobile research. Founded by specialized research fund for the doctoral program of higher education The Modeling Research of Automotive EMC Based on Random Statistical Techniques(No. 20110061110059), this paper focuses on the related analysis and forecast method for automotive wiring harness crosstalk, including the accurate extraction of distribution parameters, the time domain transient analysis and frequency domain analysis, the statistical characteristics prediction of distributed parameters and automotive wiring harness crosstalk. The main contents are as follows.By treating the system consisted of automotive wiring harness and chassis as multiconductor transmission line system, and using the classical electromagnetic field theory(Maxwell’s equations) and circuit analysis theory(Kirchhoff’s law), the multi-conductor transmission line equations are derived from the view of "field" and "circuit", the physical meaning of the distribution parameters is also elaborated. The automotive wiring harness can be treated as ideal lossless transmission lines according to its characteristics. Then the analytical expressions are derived by the electromagnetic field theory and the mirror method, which can be used to fast calculate the per-unit-length capacitance matrix and inductance matrix for automotive wiring harness, closely followed by the discussion of its application range. In order to overcome the lack of considering the proximity effect in analytical method, the numerical method expressions fitting for calculating the per-unit-length capacitance matrix and inductance matrix for automotive wiring harness are derived with the combination of moment method and mirror method. At the same time, the related factors influencing the per-unit-length distribution parameters of automotive wiring harness are also discussed in this paper.The time domain transient analysis of automotive wiring harness crosstalk is established on multi- conductor transmission line equations in time domain. However, the traditional time domain finite difference(FDTD) method has many shortages in computational efficiency and computational accuracy, influencing its engineering application. Against the low computational efficiency of traditional Leapfrog FDTD, the discrete parameters of π-type and T-type multi-conductor transmission line are expressed in matrix. Furthermore, the loop operations are replaced by matrix operations in the process of computing algorithm, and the calculation efficiency has been greatly improved. On account of the spurious oscillations, wide transition area in discontinuous solution and other shortages for the restriction of traditional Leapfrog FDTD, The upwind FDTD format is applied to the time domain transient analysis of automotive wiring harness crosstalk, so as to solve the above problems by using the characteristics that the difference direction of upwind FDTD is selected according to the positive or negative eigenvalues of the coefficient matrix, and the computational accuracy of automotive wiring harness crosstalk is improved in time domain. At the same time, the related factors influencing time domain transient analysis of automotive wiring harness crosstalk are discussed.Frequency domain analysis of automotive wiring harness crosstalk is achieved by solving multi-conductor transmission line equations in frequency domain. Transfer function is very important to analyze the crosstalk in frequency domain. However, the derivation process and expression form of transfer function in frequency domain obtained by the exact solution of automotive wiring harness crosstalk is too cumbersome, and its simplified expression also has harsh application restriction. In this paper, the transfer function with simple expression form and less restrictive conditions for automotive wiring harness crosstalk in frequency domain is got by the combination of matrix operation and solving multi- conductor transmission line equations in frequency domain, and the time domain and the frequency domain of automotive wiring harness crosstalk is linked by the transfer function in frequency domain. Meanwhile, the related factors affecting the frequency domain analysis of automotive wiring harness crosstalk are discussed.Polynomial chaos expansion(PCE) method is a kind of rigorous and efficient statistical analysis method. It expands the stochastic process by using orthogonal polynomial base, and then transfers the stochastic characteristics of stochastic process to the orthogonal polynomial coefficients by the properties of orthogonal polynomial. Finally, the related statistical characteristics of stochastic processes are got by orthogonal polynomial items and its coefficients. For the uncertainty of distributed parameter and automotive wiring harness crosstalk caused by installation, vehicle movement state and road condition, the predictions of statistical characteristics for distributed parameters and crosstalk in the time/frequency domain, such as mean value, mean square deviation, reasonable range and probability density function, can be accomplished by PCE method. Taking the 14-core automotive wiring harness model as example, the prediction of statistical characteristics for distributed parameter and automotive wiring harness crosstalk are calculated by using polynomial chaos expansion method, when the distribution types of variables for interfering line and disturbed line, such as height above ground and relative distance, are gauss distribution or uniform Distribution. The prediction of statistical characteristics for crosstalk is proceed in time domain and frequency domain, and then compared with the Monte Carlo(MC) method. The results show that the PCE method is very accurate and effective on the prediction of the statistical characteristics for distributed parameters and automotive wiring harness crosstalk. |
PDF Full Text Request