.106, 北京科技大学学报 第29卷 Simulation of Chinese characters learning with improved multi SOM net work DU Dapeng,MU Zhichun,CHEN Jing,FANG Xin Information Engineering School.University of Science and Technology Beijing.Beijing 100083.China ABSTRACI In order to simulate the Chinese character acquisition process,this paper set up a multilayer self- organizing maps(SOM)network model based on improved Kohonen network.The model's output maps, w hich adapt modified algorithm and expand neuron's neighborhood,were connected via associative links updated by Hebbian learning.After training the model could learn Chinese character architecture successfully and also do well in Chinese character component recognition.The simulation results demonstrated that the feasibility of fur- ther research in Chinese character acquisition and even Chinese language learning with this model was possible. KEY WORDS SOM network;multilayer;Chinese character learning;Chinese character architecture;Chinese character components (上接第89页) Torque ripple minimization in direct torque control of a permanent magnet syn chronous motor ZHA NG Chunmei,LIU Heping,MA Baozhu,CHEN Shujin,WANG Fangjun Information Engineering School.University of Science and Technology Beijing.Beijing 100083.China ABSTRACT The basic theory of direct torque control (DTC)for interior permanent magnet synchronous mo- tors (IPMSM)was studied.A new approach to direct torque control of an interior permanent magnet syn- chronous motor was presented to decrease torque ripple.In order to improve the static and dynamic performance of the system,a fuzzy logic idea was introduced into the system.In the fuzzy direct torque control system,the flux error and the flux error change were both properly divided into several fuzzy subsets.The acting time of the selected voltage vector and reverse voltage vector were modulated based on fuzzy inference.A speed estimation scheme was integrated with the proposed DTC scheme in order to achieve a fully sensorless high performance IPMSM drive.The scheme not only maintains constant switching frequency,but also reduces torque ripple, particularly at low speed.Simulation illustrates the operation and performance of the proposed fuzzy-logic-based controller. KEY WORDS permanent magnet synchronous motor:fuzzy logic;direct torque control;torque rippleSimulation of Chinese characters learning with improved mult-i SOM network DU Dapeng‚MU Zhichun‚CHEN Jing‚FA NG Xin Information Engineering School‚University of Science and Technology Beijing‚Beijing100083‚China ABSTRACT In order to simulate the Chinese character acquisition process‚this paper set up a multilayer self￾organizing maps （SOM） network model based on improved Kohonen network．The model’s output maps‚ which adapt modified algorithm and expand neuron’s neighborhood‚were connected via associative links updated by Hebbian learning．After training the model could learn Chinese character architecture successfully and also do well in Chinese character component recognition．The simulation results demonstrated that the feasibility of fur￾ther research in Chinese character acquisition and even Chinese language learning with this model was possible． KEY WORDS SOM network；multilayer；Chinese character learning；Chinese character architecture；Chinese character components （上接第89页） Torque ripple minimization in direct torque control of a permanent magnet syn￾chronous motor ZHA NG Chunmei‚LIU Heping‚MA Baoz hu‚CHEN Shujin‚WA NG Fangjun Information Engineering School‚University of Science and Technology Beijing‚Beijing100083‚China ABSTRACT The basic theory of direct torque control （DTC） for interior permanent magnet synchronous mo￾tors （IPMSM） was studied．A new approach to direct torque control of an interior permanent magnet syn￾chronous motor was presented to decrease torque ripple．In order to improve the static and dynamic performance of the system‚a fuzzy logic idea was introduced into the system．In the fuzzy direct torque control system‚the flux error and the flux error change were both properly divided into several fuzzy subsets．The acting time of the selected voltage vector and reverse voltage vector were modulated based on fuzzy inference．A speed estimation scheme was integrated with the proposed DTC scheme in order to achieve a fully sensorless high-performance IPMSM drive．The scheme not only maintains constant switching frequency‚but also reduces torque ripple‚ particularly at low speed．Simulation illustrates the operation and performance of the proposed fuzzy-logic-based controller． KEY WORDS permanent magnet synchronous motor；fuzzy logic；direct torque control；torque ripple ·106· 北 京 科 技 大 学 学 报 第29卷