描述逻辑SHIN的ABox一致性判定算法

为了判定描述逻辑SHIN的ABox一致性,提出了一种Tableau算法。给定TBox T、ABox A和角色层次H,该算法通过预处理将A转换成标准的ABox [A],按照特定的完整策略将一套Tableau规则应用于[A],直到将它扩展成完整的ABox [A]为止。A与T和H一致,当且仅当算法能产生一个完整且无冲突的ABox [A]。算法所采用的阻塞机制可以避免Tableau规则的无限次执行,该机制允许一个新个体被在其之前创建的任意新个体直接阻塞,而不仅仅局限于其祖先。通过对算法的可终止性、合理性和完备性进行证明,算法的正确性得以确认。     

一种基于语义Tableau的不相容知识处理方法

Tableau作为自动推理的有效方法之一,在许多人工智能领域中有重要的应用。在Tableau基础上,该文提出了Tableau开放和封闭的推理标准,应用于数据库实例不满足完整性约束的不相容关系数据库中,并对其进行修正。采用了逻辑程序的方法,对数据库进行修正,解决了传统修正方法丢失信息、出现不相容等问题。     

枢纽成本约束下的零担物流轴辐式网络设计

针对考虑枢纽建造成本和货物流的不确定的枢纽新建方案问题,引入全寿命周期理论,建立以轴辐式运营网络总成本最小化为目标的混合整数线性规划模型,并提出改进的最大最小后悔值的不确定性决策方法。通过算例来分析投资年限、枢纽干线折扣系数和不确定枢纽建造成本对零担物流（LTL）轴辐式网络的设计的影响。实验结果表明, 采用改进的不确定性决策方法得到的最优方案的运营成本比5个场景的运营成本平均降低了2.17%,表明基于改进的最大最小后悔值的不确定性决策方法,能够降低整个零担物流轴辐式运营网络总成本。     

模块化本体的Tableau算法及其性能优化

针对现有模块化本体推理方法通用性低、控制复杂等不足,提出一种基于服务的分布式Tableau算法。模块在进行一致性推理时,对关于外部概念的断言,将调用相应模块的服务进行推理,同一推理中的矛盾在定义相应概念的模块中得到捕获,采用优化技术改进算法的时间性能。实验结果表明,该算法使得模块在表述知识时能灵活引用外部概念,支持复杂的推理任务,具有较好的可伸缩性。     

基于GSPM的安全协议检验工具

介绍一个基于GSPM的安全协议验证的图形化工具。验证工具以GSPM模型为基础形式化地描述了安全协议,并引进线性时序逻辑刻画了安全协议的性质,用基于状态搜索的模型检测方法在安全协议的验证过程中找出漏洞。以简化的NSPK协议为例,描述了该工具如何验证安全协议,表明GSPM模型和验证算法的有效性和正确性。     

基于线性时序逻辑理论的仓储机器人路径规划

首先根据仓储物流环境的特点构建了可灵活扩展的仓储环境模型,并制定了适合仓储物流需求的机器人运动规则,使该模型能够适用于动态的仓储物流环境;其次采用线性时序逻辑任务公式描述具体的任务需求,使其可以适用于实际应用中更加复杂的任务;继而将任务需求与环境信息相融合,构建任务可行网络拓扑,避免分段任务搜索;然后采用Dijkstra算法在任务可行网络拓扑上搜索出最优路径,确保规划所得路径的最优性;最后将任务可行网络拓扑上的最优路径映射回加权切换系统,获得环境中满足任务需求的最优路径。与目前广泛使用的A~*算法相比,上述方法不仅能够满足复杂的任务需求,而且能够保证路径规划的最优性,而不是次优解。     

Redox properties of alkaline and acidic L zeolites as studied by EPR spectroscopy using TCNE and perylene as probe molecules

The oxidizing and reducing character of various exchanged L zeolites has been studied by EPR spectroscopy via the formation of perylene and tetracyanoethylene ion radicals, respectively, and correlated with the acido-basic properties of zeolites. In the series of alkaline L forms the reducing character increases from LiL to CsL in parallel with the calculated framework average negative charge on oxygen atoms (oxygen basicity). An opposite variation is obtained for the oxidizing properties. The exchange of K⁺ ions by H⁺, enhancing the acidity of zeolites, decreases the reducing and increases the oxidizing properties. The present work reveals the links which exist between the acidity of zeolites with oxidizing centers on one hand and the basicity with reducing character on the other hand. This observed parallelism between the redox properties and the acido-basicity of zeolites could be of great importance to explain reliably the catalytic behaviour of L zeolites as support. The first attempt has been made concerning the correlations between the changes in the redox properties and the selectivity for the aromatization of n-hexane on Pt alkaline L zeolite and the sulphur resistance of Pt/acidic or basic zeolites.     

Catalytic CO hydrogenation on potassic Fe/zeolite LTL

Zeolite LTL in the potassic form ( -LTL) was used as support of iron. The precursor was characterized by temperature programmed reduction (TPR) and Mössbauer spectroscopy (MS) at 298 and 15 K. The catalyst was obtained reducing the precursor with H₂ and its structural properties were studied by H₂ chemisorption, volumetric oxidation and MS under controlled conditions at 298 and 15 K. Measurements of activity and selectivity were carried out at 1 and 20 bar in the CO hydrogenation. Two different types of active sites were detected: Fe⁰ inside the zeolite channels and carburized iron on its external surface. The solid showed a high activity and selectivity towards light alkenes in comparison with other iron supported systems (Fe/SiO₂, Fe/Al₂O₃ and Fe/C). This behavior can be attributed to the high percentage of metallic iron promoted by potassium ions, located inside the zeolite channels. The external carbide crystals generate hydrocarbons in the diesel oil range.     

零担物流时序预测的SARIMA-GRU-BPNN组合模型及应用

针对零担物流的需求物流量显著季节性、非线性和随机性特征使其预测难度大的问题,提出一种零担物流时序预测的SARIMA-GRU-BPNN组合模型的预测方法。使用季节性分解模型将物流量分解为趋势、季节性及残差,对趋势分量采用季节性差分自回归移动平均模型（SARIMA）拟合线性变化,对季节性分量采用门控循环神经网络（GRU）拟合季节性变化,对残差分量采用反向传播神经网络（BPNN）拟合非线性及随机性变化,组合重构得到最终预测值。实验结果表明,与自身单一模型SARIMA、GRU及BPNN相比,均方根误差（RMSE）分别降低31.5%、34.5%及47.1%;与其他单一模型灰色模型、支持向量机、长短期记忆网络及多元线性回归相比,RMSE分别降低71.3%、68.9%、54.4%及70.7%;与组合模型ARIMA-GRU、ARIMA-BPNN及ARIMA-SVM相比,RMSE分别降低31.0%、43.0%及56.1%,且趋势和季节性分量预测模型拟合优度达到92%和99%,有效降低整体预测误差,提升了预测精度和模型稳健性。