【文章內(nèi)容簡介】 0208 [13] 羅軍 計(jì)算機(jī)開關(guān)電源技術(shù)研究[j] 電腦編程技巧與維護(hù) 20111118 [14] 楊玉芳 項(xiàng)安 開關(guān)電源中的濾波技術(shù)[J] 中國新技術(shù)新產(chǎn)品 20111125 [15] 秦逸平 袁惠娟 基于TL494PWM控制的電動(dòng)車開關(guān)電源設(shè)計(jì)[J] 大眾科技 20111120 外文文獻(xiàn)及其翻譯Power Electronics Electromagnetic Compatibility The electromagnetic patibility issues in power electronic systems are essentially the high levels of conducted electromagnetic interference (EMI) noise because of the fast switching actions of the power semiconductor devices. The advent of highfrequency, highpower switching devices resulted in the widespread application of power electronic converters for human productions and livings. The highpower rating and the highswitching frequency of the actions might result in severe conducted EMI. Particularly, with the international and national EMC regulations have bee more strictly, modeling and prediction of EMI issues has been an important research topic. By evaluating different methodologies of conducted EMI modeling and prediction for power converter systems includes the following two primary limitations: 1) Due to different applications, some of the existing EMI modeling methods are only valid for specific applications, which results in inadequate generality. 2) Since most EMI studies are based on the qualitative and simplified quantitative models, modeling accuracy of both magnitude and frequency cannot meet the requirement of the fullspan EMI quantification studies, which results in worse accuracy. Supported by National Natural Science Foundation of China under Grant 50421703, this dissertation aims to achieve an accurate prediction and a general methodology. Several works including the EMI mechanisms and the EMI quantification putations are developed for power electronic systems. The main contents and originalities in this research can be summarized as follows. I. Investigations on General Circuit Models and EMI Coupling Modes In order to efficiently analyze and design EMI filter, the conducted EMI noise is traditional decoupled to monmode (CM) and differentialmode (DM) ponents. This decoupling is based on the assumption that EMI propagation paths have perfectly balanced and timeinvariant circuit structures. In a practical case, power converters usually present inevitable unsymmetrical or timevariant characteristics due to the existence of semiconductor switches. So DM and CM ponents can not be totally decoupled and they can transform to each other. Therefore, the mode transformation led to another new mode of EMI: mixedmode EMI. In order to understand fundamental mechanisms by which the mixedmode EMI noise is excited and coupled, this dissertation proposes the general concept of lumped circuit model for representing the EMI noise mechanism for power electronic converters. The effects of unbalanced noise source impedances on EMI mode transformation are analyzed. The mode transformations between CM and DM ponents are modeled. The fundamental mechanism of the onintrinsic EMI is first investigated for a switched mode power supply converter. In discontinuous conduction mode, the DM noise is highly dependent on CM noise because of the unbalanced diodebridge conduction. It is shown that with the suitable and justified model, many practical filters pertinent to mixedmode EMI are investigated, and the noise attenuation can also be derived theoretically. These investigations can provide a guideline for full understanding of the EMI mechanism and accuracy modeling in power electronic converters. (Publications: A new technique for modeling and analysis of mixedmode conducted EMI noise, IEEE Transactions on Power Electronics, 2004。 Study of differentialmode EMI of switching power supplies with rectifier frontend, Transactions of China Electro technical Society, 2006) II. Identification of Essential Coupling Path Models for Conducted EMI Prediction Conducted EMI prediction problem is essentially the problem of EMI noise source modeling and EMI noise propagation path modeling. These modeling methods can be classified into two approaches, mathematicsbased method and measurementbased method. The mathematics method is very timeconsuming because the circuit models are very plicated. The measurement method is only valid for specific circuit that is conveniently to be measured, and is lack of generality and impracticability. This dissertation proposes a novel modeling concept, called essential coupling path models, derived from a circuit theoretical viewpoint, means that the simplest models contain the dominant noise sources and the dominant noise coupling paths, which can provide a full feature of the EMI generations. Applying the new idea, this work investigates the conducted EMI coupling in an AC/DC halfbridge converter. Three modes of conducted EMI noise are identified by time domain measurements. The lumped circuit models are derived to describe the essential coupling paths based on the identification of the EMI coupling modes. Meanwhile, this study illustrates the extraction of the parameters in the aforedescribed models by measurements, and demonstrates the significance of each coupling path in producing conducted EMI. It is shown that the proposed method is very effective and accurate in identifying and capturing EMI features. The equivalent models of EMI noise are sorted out by just a few simple measurements. Under these approaches, EMI performance can be predicted together with the filtering strategies. (Publications: Identification of essential coupling path models for conducted EMI prediction in switching power converters, IEEE Transactions on Power Electronics, 2006。 Noise source lumped circuit modeling and identification for power converters, IEEE Transactions on Industrial Electronics, 2006) III. High Frequency Conducted EMI Source Modeling The conventional method of EMI prediction is to model the current or voltage source as a peri