汽车故障知识图谱构建及应用研究

知识图谱技术对汽车高效的故障诊断具有重要的意义,现有汽车故障知识图谱构建存在着实体识别模型效果不佳、无法解决嵌套实体等问题。针对上述问题,通过采用全词掩码的预训练语义模型、加入对抗训练和改进嵌套实体识别模型的方式提高实体识别模型效果,提出了一种改进的嵌套实体识别模型。实验结果表明,所提模型在F1指标、P指标、R指标上相比基线模型分别提高了3.56%、4.08%、3.05%,相比其他模型也有不同程度的提高,验证了所提模型对汽车维修领域实体识别具有显著效果。同时,基于构建的汽车故障知识图谱,实现了汽车故障知识智能问答原型系统,展示了知识图谱技术在汽车故障诊断与维护领域的应用前景。     

电气信息类专业数学系列课程的教学改革探索

针对电气信息类专业数学系列课程教学中存在的一些问题,通过开展翻转课堂式教学强化课前预习与课堂讨论环节,在数学概念的讲解中灵活运用迁移教学方法,建设典型案例驱动课堂教学,将对抽象数学知识的理解转化为工程应用问题,从而激发学生对数学课程的学习兴趣,提高学生运用数学知识解决复杂工程问题的能力,在多门数学课程的教学实践中取得了良好的教学效果.     

Volterra series based predistortion for broadband RF power amplifiers with memory effects

RF power amplifiers(PAs)are usually considered as memoryless devices in most existing predistortion techniques.However,in broadband communication systems,such as WCDMA,the PA memory effects are significant,and memoryless predistortion cannot linearize the PAs effectively.After analyzing the PA memory effects,a novel predistortion method based on the simplified Volterra series is proposed to linearize broadband RF PAs with memory effects.The indirect learning architecture is adopted to design the predistortion scheme and the recursive least squares algorithm with forgetting factor is applied to identify the parameters of the predistorter.Simulation results show that the proposed predistortion method can compensate the nonlinear distortion and memory effects of broadband RF PAs effectively.     

Virtual reality mobility model for wireless ad hoc networks

For wireless ad hoc networks simulation, node＇s mobility pattern and traffic pattern are two key elements. A new simulation model is presented based on the virtual reality collision detection algorithm in obstacle environment, and the model uses the path planning method to avoid obstacles and to compute the node＇s moving path. Obstacles also affect node＇s signal propagation. Considering these factors, this study implements the mobility model for wireless ad hoc networks. Simulation results show that the model has a significant impact on the performance of protocols.     

Efficient combination rule of Dezert-Smarandache theory

The Dezert-Smarandache theory （DSmT） is a useful method for dealing with uncertainty problems. It is more efficient in combining conflicting evidence. Therefore, it has been successfully applied in data fusion and object recognition. However, there exist shortcomings in its combination rule. An efficient combination rule is presented, that is, the evidence＇s conflicting probability is distributed to every proposition based on remaining the focal elements of conflict. Experiments show that the new combination rule improves the reliability and rationality of the combination results. Although evidences conflict another one highly, good combination results are also obtained.     

A survey of stackelberg differential game models in supply and marketing channels

Erratum to: J Syst Sci Syst EngDOI: 10.1007/s11518-007-5058-2The presentation of Table 2 in the original version of this article contained a few typos. The corrected Table 2 is given below.     

Design of complex FIR filters with arbitrary magnitude and group delay responses

To design approximately linear-phase complex coefficient finite impulse response (FIR) digital filters with arbitrary magnitude and group delay responses, a novel neural network approach is studied. The approach is based on a batch back-propagation neural network algorithm by directly minimizing the real magnitude error and phase error from the linear-phase to obtain the filter’s coefficients. The approach can deal with both the real and complex coefficient FIR digital filters design problems. The main advantage of the proposed design method is the significant reduction in the group delay error. The effectiveness of the proposed method is illustrated with two optimal design examples.     

基于双向投影法的概率语言多属性群决策

针对概率语言信息下的多属性群决策问题,首先基于概率语言术语集(probabilistic linguistic term sets, PLTS)的调整规则和语言尺度函数,定义了PLTS的新运算,并证明其满足代数运算的基本性质。然后,为解决现有向量模公式在下标对称语言下计算结果与实际不符的问题,提出一种改进的概率语言信息下的向量模公式。在此基础上定义方案与正、负理想解形成的向量表示式,并建立针对概率语言信息的双向投影法,由此提出新的贴近度公式。最后,通过应用算例的对比分析验证了本文所提方法的有效性和可行性。该方法充分考虑了PLTS的数据特征,在简化计算的同时避免了信息损失,为解决概率语言信息中数值差异较小的问题提供了新的方法和思路。     

可重构视频阵列处理器测试平台设计与实现

针对可重构视频阵列处理器的设计要求及传统测试方法测试视频编解码系统时速度慢、精度低和可观测性不强的问题。开发了基于Qt的用户界面,设计实现了以现场可编程门阵列(Field programmable gate-array,FPGA)为核心的软硬件协同测试平台。在PC端实现以软件仿真为基础的数据传输与图像重现,在FPGA端实现以可重构视频阵列处理器为基础的视频编解码算法并行映射。实验结果表明,在工作频率为100 MHz时,FPGA与PC之间可正确传输数据并满足算法测试时不同测试用例的更换需求,具有较好的可观测性。     

基于模型的多目标优化问题方法研究

基于模型的多目标优化方法目的是创新一种通过黑箱评估的多目标函数优化算法,该算法从解空间上的混合分布中迭代生成候选解,并根据采样解的控制数来更新混合分布,求解过程的搜索偏向于Pareto最优解的集合。算法在解空间上寻找混合分布,使得混合分布的每个分量都是以帕累托最优解为中心的简并分布,并且每个预计的Pareto最优解都通过一个阈值距离均匀地分布在Pareto最优解集上,实验通过几个基准函数和方法证明了该算法的性能。