基于Q-learning的不确定环境BDI Agent最优策略规划研究

BDI模型能够很好地解决在特定环境下的Agent的推理和决策问题,但在动态和不确定环境下缺少决策和学习的能力。强化学习解决了Agent在未知环境下的决策问题,却缺少BDI模型中的规则描述和逻辑推理。针对BDI在未知和动态环境下的策略规划问题,提出基于强化学习Q-learning算法来实现BDI Agent学习和规划的方法,并针对BDI的实现模型ASL的决策机制做出了改进,最后在ASL的仿真平台Jason上建立了迷宫的仿真,仿真实验表明,在加入Q-learning学习机制后的新的ASL系统中,Agent在不确定环境下依然可以完成任务。     

On the performance of band-limited asynchronous DS-CDMA over nakagami-m channels

In this paper we investigated the BER performance of DS-CDMA using various chip-waveforms, which include three time-limited chip-waveforms and two band-limited chip-waveforms. Closed-form formulae were derived for evaluating the achievable bit-error rate performance with the aid of the standard Gaussian approximation, when communicating over a Nakagami-m channel. The time-limited waveforms impose a low implementational complexity, since they maybe over sampled and read from a look-up table. However, they are outperformed by the frequency-domain raised-cosine waveform as well as the optimum waveform specifically designed by Cho and Lehnert for achieving the lowest possible bit error rate     

动态评价技术在塔河碳酸盐岩缝洞型油气藏中的应用

塔河油田奥陶系油气藏是大型碳酸盐岩溶洞型油气藏，其储渗空间主要为大小不同的溶洞、裂缝带、溶蚀孔隙等组成，该油气藏具有极强的非均质性，单纯用静态资料来认识这类油气藏是非常困难。文章提出利用生产动态资料和信息进行该类型油气藏研究的新思路，利用人工神经网络技术在处理非线性相关参数预测方面的优势，并以渗流理论为基础，结合试井成果，选用已知油井的产量、油嘴、油压、含水率、气油比、原油密度等6个开发动态参数作为样品输入数据，推导出影响油气藏开发的重要参数（地层系数）与生产信息的关系，建立了人工神经网络预测储层参数的结构模型。通过塔里木盆地塔河油气田实例研究，说明了利用动态信息评价油气藏技术在碳酸盐岩缝洞型油气藏储层预测和非均质性分析等方面具有较高的实用价值。     

DVB-T调制器外码编码器的FPGA设计

设计并实现了一种DVB—T调制器的外码编码器，着重介绍了码流自适应接口、能量扩散、RS编码器及卷积交织器的设计方法，该设计方案最终在Altera的FPGA上进行验证。     

Tracing of the 1st IEC Secretariat Assumed by China

Introduction The IEC central office informed in 7/543/AC that the secretariat of TC 7 would be taken over by the Chinese National Committee on January 10, 2003 and affirmed subsequently in 7/544/AC that the secretariat of TC 7 has been taken over by the Chinese National Committee which appointing secretary in Shanghai Electric Cable Research Institute as no objection has been raised by the Standardization Management Board members. It's the first IEC secretariat that assumed by China, with great significance, just as commented by the media that the commitment indicate undoubtedly China is to play a much more active and important role in the world especially after its entry into world trade organization as well as the trend of global economic integration.     

纳米AlN粉末的制备与烧结

利用低温燃烧合成前驱物制备出平均粒度为100 nm的AlN陶瓷粉末,比较了该粉末的常压烧结和放电等离子烧结的特性.实验表明:以合成的AlN粉末为原料,添加5%(质量比)Y2O3作为烧结助剂,在常压、流动N2气氛下1600℃保温3 h,制备出平均晶粒尺寸为4～8 μm、密度为3.28 g·cm-3的AlN陶瓷;将同样的粉末不加任何烧结助剂,采用SPS技术在1600℃保温4 min,得到密度为3.26 g·cm-3的AlN陶瓷,晶粒度约为1～2μm.     

川合137井大修井工艺技术浅析

在川合137井的修井作业中，运用了多种修井工具和工艺技术。通过内捞、外捞、强提、倒扣、磨铣、套铣、开钻新井眼和铣锥通井的方法，充分解放了油气层，达到了修井目的。其独特的想法、施工和自制工具的成功应用对今后大修井作业具有很高的参考价值。     

相关条件下基于N-P准则的分布式检测理论研究

本文针对并行网络、二元局部判决的情况,研究了在相关条件下基于N-P准则的分布式检测融合算法,给出了在联合概率密度已知和未知两种情况下的最优融合规则的理论推导及相应的解决方案,并在实验仿真的基础上,得到了部分有益的结论。     

多带抗噪声语音识别算法研究

根据Flether等人的研究,基于感知独立性假设的子带识别方法被用于抗噪声鲁棒语音识别。本文拓展子带方法,采用基于噪声污染假定的多带框架来减少噪声影响。论文不仅从理论上分析了噪声污染假定多带框架在识别性能上的潜在优势,而且提出了多带环境下的鲁棒语音识别算法。研究表明：多带框架不仅回避了独立感知假设要求,而且与子带方法相比,多带方法能更好的减少噪声影响,提高系统识别性能。