[动力电池]融合车、路、人信息的电动汽车续驶里程估算

Aiming at driving cycle and driving style influenced significantly on remaining range estimation, which resulted in the remaining range estimation difficulty, a estimate method was adopted,which combined driving cycle with driver’s driving style,to estimation the driving range of electric vehicles. The typical driving condition fragments were selected based on large sample and traffic characteristics of Zhengzhou, which through principal component analysis and fuzzy C-means clustering technique.Under MATLAB/Simulink,a model to identify driving patterns was established based on principal component analysis and learning vector quantization by the typical driving condition fragments and driving style fuzzy recognition model, the remaining range estimation were carried out, which adopted joint simulation method,and the results of remaining range estimation were optimized by Kalman filtering. The simulation analysis and hardware in the loop test results show that driving range estimation for electric vehicles through vehicles,roads, and human information fusion method, reduces the remaining range estimation errors, and confirmes the feasibility of the method at the same time.     

方家山核电机组DCS时钟跳变的影响分析与研究

时钟系统的故障或异常极易触发机组运行事件,严重影响运行人员对机组的监盘和控制。以方家山核电厂DCS一层（I/A系统）时钟跳变引起二层（KIC系统）的历史功能异常事件为主线,介绍了方家山核电机组的DCS系统架构、授时方式、一二层数据传输及其时钟同步的规则。并在最小化模拟平台进行实验,详细阐述了DCS一层时钟突跳对二层的影响及原因。最后,对事件产生的原因进行深入的剖析和研究,找出设备系统设计的缺陷及不合理之处,并提出相应的优化和改进方案,以解决目前核电厂DCS时钟系统授时所存在的薄弱问题,提升DCS时钟系统授时的可靠性。     

融合车、路、人信息的电动汽车续驶里程估算

针对因行驶工况与驾驶风格对剩余续驶里程有显著影响而导致剩余里程估算困难的问题,采用工况识别与驾驶风格识别相结合的方式对电动汽车续驶里程进行估算。通过主成分分析和聚类分析,选出基于大数据、符合郑州市当地交通特征的典型行驶工况片段,并利用选取的典型行驶工况片段,在MATLAB/Simulink下建立主成分分析和学习矢量量化神经网络相结合的工况识别模型、驾驶风格模糊识别模型;通过联合仿真进行剩余续驶里程实时估算,并通过卡尔曼滤波对输出结果进一步优化。仿真分析及半实物测试结果表明,采用融合车、路、人信息的电动汽车续驶里程估算方法,不仅降低了剩余里程估算误差,同时也证实了方法的可行性。