Neural network-based learning and estimation of battery state-of-charge: A comparison study between direct and indirect methodology

Faced with the ever-increasing urban environmental pollution, the electric vehicles (EVs) have received increasing attention in the automotive industry. Lithium-ion batteries, serving as electrochemical power storage, have been extensively used in EVs because of the lightweight, no local pollution and high power density. The increasing awareness on the safe operation and reliability of the battery requires an efficient battery management system (BMS), among the parameters monitored by which, state-of-charge (SOC) is critical in preventing overcharge, deep discharge, and irreversible damage. This article investigates the neural network (NN)-based modeling, learning, and estimation of SOC by comparing two different methodologies, that is, direct structure with SOC as network output and indirect structure with voltage as output. Firstly, the nonlinear autoregressive exogenous neural network (NARX-NN) is introduced, in which SOC is directly deemed as an NN output for learning and estimation. Secondly, a radial basis function (RBF)-based NN with unscented Kalman filter (RBFNN-UKF) is proposed, in which the terminal voltage is used as output. Instead, SOC is deemed as an internal state which would be estimated indirectly based on the feedback error of voltage. Experimental results demonstrate that both estimators can achieve accurate SOC estimation for regular cases, in spite of the inaccurate initial conditions. However, the direct NN structure is revealed as not capable of dealing with the cases with sensor bias, which, however, can be well accommodated in the indirect structure by extending the sensor bias as an augmented state. Benefiting from the uncertainty augmentation and feedback compensation, the indirect RBFNN-UKF shows superiority over the direct estimation in the practical experiments, depicting a promising prospect in the future onboard EV-BMS application.     

State of charge estimation method based on the extended Kalman filter algorithm with consideration of time-varying battery parameters

In developing battery management systems, estimating state-of-charge (SOC) is important yet challenging. Compared with traditional SOC estimation methods (eg, the ampere-hour integration method), extended Kalman filter (EKF) algorithm does not depend on the initial value of SOC and has no accumulated error, which is suitable for the actual working condition of electric vehicles. EKF is a model-based algorithm; the accuracy of SOC estimated by this algorithm was greatly influenced by the accuracy of battery model and model parameters. The parameters of battery change with many factors and exhibit strong nonlinearity and time variance. Typical EKF algorithm approximates battery as a linear, time-invariant system; however, this approach introduces estimation errors. To minimize such errors, previous studies have focused on improving the accuracy of identifying battery parameters. Although studies on battery model with time-varying parameters have been carried out, few have studied the combination of time-varying battery parameters and EKF algorithm. A SOC estimation method that combines time-varying battery parameters with EKF algorithm is proposed to improve the accuracy of SOC estimation. Battery parameter data were obtained experimentally under different temperatures, SOC levels, and discharge rates. The results of parameter identification are made into a data table, and the battery parameters in the EKF system matrix are updated by looking up the data in the table. Simulation and experimental results shown that, average error of SOC estimated by the proposed algorithm is 2.39% under 0.9 C constant current discharge and 2.4% under 1.3 C, which is 1.91% and 2.35% lower than that of EKF algorithm with fixed battery parameters. Under intermittent discharge with constant current (1.1 C) and capacity (10%), the average error of SOC estimated by the proposed algorithm is 1.4%, which is 0.3% lower than that of EKF algorithm with fixed battery parameters. The average error of SOC estimated by the proposed algorithm under the New European Driving Cycle (NEDC) is 1.6%, which is 0.2% lower than that of EKF algorithm with fixed battery parameters. Relative to the EKF algorithm with fixed battery parameters, the proposed EFK algorithm with time-varying battery parameters yields higher accuracy.     

State of charge estimation for lithium-ion battery based on an intelligent adaptive unscented Kalman filter

Adaptive unscented Kalman filter (AUKF) has been widely used for state of charge (SOC) estimation of lithium-ion battery. The noise covariance of the conventional AUKF method is updated based on the innovation covariance matrix (ICM), which is estimated using the error innovation sequence (EIS). However, the distribution of EIS changes due to the time-varying noise, load current dynamics and modelling error, which will lead to inaccurate ICM estimation. Therefore, an intelligent adaptive unscented Kalman filter (IAUKF) method is proposed to detect the distribution change of EIS. Then, the ICM is estimated based on the EIS after the distribution change. Results show that the IAUKF method can improve SOC estimation accuracy significantly. Compared with that of the AUKF method, the root mean squared error and the mean absolute error of SOC based on the IAUKF method decrease by 43.70% and 72.37% under random walk discharge condition, respectively. In addition, the computation time of the IAUKF method slightly increases by 6.27% compared with that of AUKF method. Finally, the effect of initial parameters on the SOC estimation accuracy was analysed. The results indicate that proper algorithm tuning, such as initial window length of EIS for ICM update and the threshold value, can further improve the SOC accuracy based on the proposed IAUKF method. The proposed IAUKF method also shows high robustness against initial measurement noise covariance.     

利用“列表法”分析金属与盐溶液反应后的滤液滤渣

云南省初中学业水平考试中,有关金属与盐溶液反应后的滤液滤渣问题一直是学生的难点。结合金属活动性顺序,利用"列表法"分析,可将复杂问题简单化,将反应后的滤液滤渣成分直观展示出来,有效提高学生做题的准确性,特别对理解能力较弱的同学帮助很大。     

基于EKF和UKF算法非均匀介质热物性参数重建

将无迹卡尔曼滤波技术(unscented Kalman filter, UKF)用于求解一维介质热物性参数反演问题；也对利用扩展卡尔曼滤波技术(extended Kalman filter, EKF)反演一维介质中热导率问题进行了研究。首先给出了正问题模型，然后详细介绍了EKF算法和UKF算法的基本原理。最后为了验证当前算法的可行性，采用UKF算法重建了介质内部随位置变化的热导率，并采用EKF算法重建了介质内部随时间变化的热导率。计算结果表明，UKF算法和EKF算法均能较为准确地反演介质的热导率。为了减小重建结果的时间滞后，建议使用较小的测量误差协方差R。     

隔膜滤板膜片用改性聚丙烯的制备

研究了聚丙烯品种、弹性体和生产工艺对改性聚丙烯微观结构、力学性能的影响,开发了3种具有不同性能特点的隔膜滤板膜片用改性聚丙烯。结果表明:产品具有优异的拉伸强度、断裂拉伸应变和耐温性能;成型的膜片经过高压充水多次开合不发生破裂,满足隔膜式压滤机的使用要求。     

并联有源滤波器的改进重复控制策略研究

针对三相整流负载产生的6k±1次谐波,提出静止坐标系下的改进型6k±1重复控制策略。同时将比例控制与改进型重复控制相结合,设计改进重复控制并联比例的复合控制结构。为减小改进重复控制中延时环节不为整数的影响,采用基于Lagrange插值近似的FIR滤波器逼近分数延时特性。最后对系统进行稳定性分析和详细设计方法进行推导。通过Matlab仿真验证改进重复控制策略能有效跟踪6k±1次谐波且具有良好的补偿效果,动态响应较快。     

一种射频抗电磁干扰的矿用红外温度传感器

设计了一种适用于复杂条件下实施井下温度探测的射频式红外温度传感器。借助于光学系统、光电探测器实时感应井下温度并提取温度信号,同时在DC-DC电源隔离模块中实施EMI滤波以抗电磁干扰,处理后的信号经过A/D转换、运算器放大后显示在LCD上,实施读取数据并记录在传感器的存储片上,实现了矿井复合灾害条件下的温度测试,测试精度控制在±2%。     

非织造过滤介质孔径及截留性能的研究

对应用于液固分离和气固分离的非织造过滤介质的孔径和截留性能测试方法做了研究,指出采用气体流量法进行孔径性能测试,可同时得到非织造过滤介质的最大孔径、平均孔径及孔径分布,全面表征了其孔径性能;并对用于液固和气固分离2个不同领域的非织造过滤介质采用不同的方法做了截留性能测试。同时,对比非织造过滤介质的孔径和截留性能,对二者之间的关联做了探索,为实际工业过程中非织造过滤介质的选用和评价提供了指导。     

Observer‐based fuzzy adaptive quantized control for uncertain nonlinear multiagent systems

This paper focuses on consensus quantized control design problem for uncertain nonlinear multiagent systems with unmeasured states. Every follower can be denoted through a system with unmeasurable states, hysteretic quantized input, and unknown nonlinearities. Fuzzy state observer and Fuzzy logic systems are employed to estimate unmeasured states and approximate unknown nonlinear functions, respectively. The hysteretic quantized input can be split into two bounded nonlinear functions to avoid chattering problem. By combining adaptive backstepping and first‐order filter signals, an observer‐based fuzzy adaptive quantized control scheme is designed for each follower. All signals exist in closed‐loop systems are semiglobally uniformly ultimately bounded, and all followers can accomplish a desired consensus results. Finally, a numerical example is employed to elaborate the effectiveness of proposed control strategy.