基于强化学习的JLQ模型的直接自适应最优控制

研究离散时间跳变线性二次(JLQ)模型的直接自适应最优控制问题.将强化学习的理论和方法应用于JLQ模型,设计基于Q函数的策略迭代算法,以优化系统性能.在系统参数以及模态跳变概率未知的情况下,Q函数对应的参数矩阵,可通过观察给定策略下系统行为,应用递归最小二乘算法在线估计.基于此参数矩阵,可构造出新的策略使得系统性能更优.该算法可收敛到最优策略.     

基于Q2406B的车载智能终端的研究与设计

该文介绍了在智能交通系统中起到重要作用的车载智能终端的研究与设计。该智能终端是以GPRS／GPS等技术为基础,以Q2406B无线CPU为核心的一种嵌入式系统。文中给出了该嵌入式移动智能终端的软硬件设计方案。     

Invariance properties of RAMS and linear time

An operation on integers isLTTC if it is computable in linear time on a Turing machine (using the dyadic or binary representation of integers). AnLTTC-RAM (respectivelyI-RAM) is a RAM which only uses LTTC operations (respectively operations in the setI).The address-free time complexity measure of a RAM evaluates execution times using the logarithmic cost criterion but assumes that addressing operations are performed for free.     

Space-efficient recognition of sparse self-reducible languages

Mahaney and others have shown that sparse self-reducible sets have time-efficient algorithms, and have concluded that it is unlikely that NP has sparse complete sets. Mahaney's work, intuition, and a 1978 conjecture of Hartmanis notwithstanding, nothing has been known about the density of complete sets for feasible classes until now. This paper shows that sparse self-reducible sets have space-efficient algorithms, and in many cases, even have time-space-efficient algorithms. We conclude that NL, NC ^k , AC ^k , LOG(DCFL), LOG(CFL), and P lack complete (or even Turing-hard) sets of low density unless implausible complexity class inclusions hold. In particular, if NL (respectively P, _k , or NP) has a polylog-sparse logspace-hard set, then NLSC (respectively PSC, _k , or PHSC), and if P has subpolynomially sparse logspace-hard sets, then PPSPACE.     

Circuits constructed with MOD q gates cannot compute “AND” in sublinear size

Algebraic techniques are used to prove that any circuit constructed with MOD _q gates that computes the AND function must use (n) gates at the first level. The best bound previously known to be valid for arbitraryq was (logn).     

An introduction to TMN

The Telecommunication Management Network standard (ITU-T M.3010), the Network Magagement Forum OMNIPoint program, and many related developments are summarized. TMN and OMNIPoint are presented as a means to achieve integrated network management in the telecommunications industry by calling for all management systems to include an interoperable interface which permits each system to be integrated into a larger management hierarchy. By means of such interfaces, hierarchical integration of management systems can be achieved within the administration of a single service provider. Interoperable interfaces also make possible the interconnection of multiple service providers and the connection of customer to service provider, resulting in the possibility of industry wide integration of management infrastructures at all levels.     

基于知识图谱的智能问答意图识别联合模型

针对现有意图识别联合模型在专业领域知识图谱问答中容易发生识别领域实体以及问句分类错误的情况,提出一个结合了领域知识图谱的意图识别联合模型。该模型有三步,将领域知识图谱中实体对应的本体标签以及本体间关系导入训练数据集,形成包含本体标签的知识文本以及额外包含本体关系的知识文本图;通过字符级嵌入和位置信息嵌入将包含了本体标签的知识文本转化成嵌入表示并依据知识文本图创建实体关系可视矩阵,明确知识文本各成分的相关程度;将嵌入表示和实体关系可视矩阵输入模型编码层进行模型的训练。以高速列车领域知识图谱为例,经过准确率和召回率的验证,以该方法训练出的模型在高速列车领域问答数据集的意图识别任务上取得了更好的表现。     

基于改进深度Q网络算法的多园区综合能源系统能量管理方法

多园区综合能源系统可通过多能互补互济显著提升运行经济性,然而园区之间的复杂互动、多能耦合决策会给多园区综合能源系统的能量管理带来决策空间庞大、算法难以收敛等挑战性问题。为解决上述问题,提出了一种基于改进深度Q网络(modified deep Q network,MDQN)算法的多园区综合能源系统能量管理方法。首先,采用独立于园区的外部气象数据、历史互动功率数据,构建了基于长短期记忆(long short-term memory,LSTM)深度网络的各园区综合能源系统外部互动环境等值模型,降低了强化学习奖励函数的计算复杂度;其次,提出一种基于k优先采样策略的MDQN算法,用k-优先采样策略来代替ε贪心策略,克服了大规模动作空间中探索效率低下的问题;最后,在含3个园区综合能源系统的算例中进行验证,结果表明MDQN算法相比原DQN算法具有更好的收敛性与稳定性,同时可以提升园区经济效益达29.16%。     

Q235焊接钢板受剪声发射时频特性分析

Q235钢板在工程中应用广泛,为了研究焊接对于Q235钢板性能的影响,开展了焊接试样和完整试样的受剪实验,利用数字声发射系统对实验过程的声发射信号进行实时采集。根据声发射计数率划分不同阶段,采用魏格纳-维尔分布,平滑伪魏格纳-维尔分布,对信号进行时频分析。实验结果表明:受剪实验中,焊接试样与完整试样的声发射特性明显不同,主要表现在受剪过程的各阶段中声发射信号能量集中的频带不同,弹性变形阶段信号时频带主要集中在80~100 k Hz附近;钢板塑性变形阶段时频带集中在100~140 k Hz范围内;焊缝塑性变形阶段,时频能量主要集中的频带范围在300~350 k Hz之间,部分能量较弱的信号分布在100 k Hz附近频段。