Batteries play a crucial role in electric vehicles since they are the only power sources of the vehicles. In order to guarantee that batteries can work efficiently and safely, battery management systems (BMS) are employed to measure, estimate and regulate battery states during the operation of electric vehicles. To do that, numerous voltage, current, and temperature sensors are required to be installed in the BMS. However, a large number of sensors may lead to some problems, e.g., high cost, reduced space, low efficiency, and high failure rates. To address these challenges, in this paper, a digital twin paradigm is proposed for the BMS to estimate and predict the battery states with only a voltage sensor. A multi-linear regression algorithm is utilized to build the regression model between battery voltage and the other variables. Experiment results show that the proposed digital twin model achieves over 90% prediction accuracy in practical applications.

Supercapacitors have been considered as a promising choice for public transportation electrification, e.g., catenary-free trams, as the only power source. Due to the frequent stops and drastic load changes of trams, the DC bus of supercapacitor storage systems exhibits significant voltage fluctuations, which degrades the reliable operation of trams. In this article, we propose a predictive set point modulation control method for supercapacitors storage systems, where the reference voltage is regulated in real time based on the prediction of DC bus voltage, which can effectively suppress the voltage fluctuations. We first model the supercapacitor storage systems based on the averaging method. Then, a cascade control framework is designed to regulate the DC bus voltage. A predictive set point modulation control method is designed to suppress the voltage fluctuations without changing the closed-loop system structure. We built a laboratory hardware platform to illustrate the performance of the proposed method. Both simulation and experiment results verify that the proposed method can effectively suppress the voltage fluctuations of the DC bus when compared with classical methods.

Optimal charging of supercapacitors with energy efficiency maximization is of significance for supercapacitor charging systems. However, most existing studies are focused on the optimal charging of a single superca-pacitor. In practical applications, supercapacitors are usually connected in series as packs. To address this problem, an optimal charging method is proposed for series-connected supercapacitor packs. By using it, the charging time can be specified by the user. Firstly, the existing charging methods for supercapacitor systems and their limitations are analyzed, especially when the charging time is limited. Then, a user-specified optimal charging method for supercapacitors with the user-specified charging time is designed, and the effectiveness of the proposed method in energy effectiveness maximization is rigorously proved. The performance of the proposed charging method is verified via extensive experimental results. The results show that the proposed charging method can effectively improve the energy efficiency under the limitation of charging time compared with the existing methods.

AbstractBraking control of urban rail vehicles with multiple carriages is critical to ensure the safe operation of urban rails. However, existing decentralized braking control methods lead to the inconsistent speed and excessive coupler force among carriages, which compromises the operation safety of urban rails. To address this issue, a cooperative braking strategy is proposed for urban rail vehicles based on Koopman model predictive control method. First, a cyber‐physical model is established, where the physical layer characterizes the dynamic model of multiple carriages using the Koopman operator. The cyber layer represents the communication topology of carriages with graph theory. Second, a cooperative braking controller is designed with the distributed model predictive control method. An optimisation problem is formulated to minimize the speed inconsistency and relative displacement difference among carriages. Third, extensive simulations under various scenarios, including normal and extreme operating conditions, are conducted to verify the effectiveness of the proposed method. Simulation results show that, compared with the classical decentralized method, the proposed braking method reduces the average relative displacement by 78% while reducing the speed error by 29%.

制动机均衡风缸控制系统具有部件种类多、退化过程复杂、工况多样等特点,给系统的健康状态评估带来挑战.提出一种结合长期和短期混合特征的健康状态评估方法,通过对均衡风缸系统关键部件进行寿命预测进行健康状态评估.结合系统工作模式切换对部件退化过程的影响提取累积物理动态指标,使用循环神经网络实现部件的长期累积寿命预测;在时间滑窗内提取局部时域和频域特征,表征系统短期波动对老化的影响,利用改进的轻量梯度提升机进行短期寿命预测;通过分段模型平均法进行长期和短期寿命预测结果的融合,得到健康状态评估结果.通过搭建制动机均衡风缸控制系统实验平台,进行加速老化实验,获取全寿命周期老化数据,对提出的混合健康状态评估方法进行仿真,验证所提方法的有效性.

The brake system has the characteristics of multi-component, multi working conditions and complex degradation process, which brings great challenges to its health condition assessment. As it is difficult for a single brake agent to make a comprehensive and accurate health condition assessment, a health condition assessment model based on multi-agent federated learning is proposed in this paper. The different agents train their brake health data under different working conditions and states, which ensures the accuracy of health condition assessment and the safety of data of each agent. Aiming at the problem that it is difficult to determine the credibility of agent data in the process of federated learning, a credibility of agent data scheme based on evidence theory is proposed, which not only reduces the cost of calculation and communication, but also further improves the accuracy of health condition assessment. Simulation results show that the scheme not only has higher prediction accuracy, but also can ensure the security of agent data compared with the conventional centralized training scheme.

The braking system is the key part of trains, and its full life-cycle of health status is essential to ensure the safety of trains. How to accurately assess real-time health status throughout the full life-cycle of the train braking system is a challenge. In this paper, a digital twin-driven hybrid estimate method for health status of the braking system is proposed. Firstly, an equivalent model of the braking system is built in the digital twin platform. Then, a hybrid method of fusing model and data is proposed to assess the health status. Finally, a cloud digital twin experimental platform for health status assessment of the braking system is built, and the health status is shown by visualization framework. The experiments verify the effectiveness and practicality of the proposed scheme.

Pneumatic valves are key components of the train electro-pneumatic braking system. In order to obtain health indicators of pneumatic valves and provide faults early-warning, this paper proposes a fault prognosis method using principal component analysis (PCA) and support vector regression (SVR). Two health indicators ( $$T^{2}$$ and SPE) of pneumatic valves are extracted through PCA method based on the full life cycle data set, which came from the joint simulation model. Second, a pneumatic valve fault prognosis model based on SVR is trained based on the health indicators. Combined with the working model of the train electro-pneumatic braking system, the proposed fault prognosis model can estimate the expected time of pneumatic valve fault time accurately. Results from a semi-physical simulation verification platform of DK-2 braking system indicate that the proposed method can effectively predict the occurrence of faults. This work can provide a scientific basis for the operation of braking system and maintenance strategy of pneumatic valves.

Recent years have seen traditional chemical factories are transforming into smart factories to improve the efficiency and reduce the cost. To do that, a large number of sensors are deployed on the production lines. However, in harsh chemical production environment, such as strong acid and high temperature, certain variables are very difficult to measure, since the corresponding sensors are costly and prone to failure. To address this challenge, we propose a cooperative sensorless perception approach to estimate hard-to-measure variables by exploring the correlation and regression among the estimated variables with the other easy-to-measure variables of the production line. Specifically, the production data is pre-processed through correlation analysis and data cleaning, which brings a valuable data set for modeling. Furthermore, a regression model based on random forest is built to forecast the targeted variables. We have verified our methodology in the practical $$AlF_3$$ production lines in a chemical company. Experiment results show that the accuracy of predicted results is higher than 95%, which verifies that the proposed method is feasible and reliable.

Long short-term memory network (LSTM) is a popular deep learning network method for estimating the state of health (SOH) of lithium-ion batteries. However, the hyperparameters in the network are usually difficult to predefine, which degrades the estimation accuracy in applications. To address this challenge, this paper proposes a data-driven estimation method based on the improved LSTM, where the network topology is estimated by the particle swarm optimization (PSO) algorithm. First, four health indicators in the charging and discharging process are selected. Grey relational analysis is further employed to quantify their correlations with the battery SOH. Then, an LSTM model is established to map the relationship between health factors and battery SOH. To tackle the parameter determination and slow convergence problems of the classical LSTM method, the particle swarm optimization algorithm is adopted to determine the key hyperparameters in the neural network, and the RMSProp training method and dropout technique are introduced to accelerate the convergence speed and avoid over-fitting problems. The experimental results show that the prediction accuracy is improved by at least 5 % when compared with the classical LSTM method.

Lithium-ion batteries have been employed extensively in many important applications in the electronics industry. For safety and reliability, it is extremely critical to get an accurate and early-stage remaining useful life prognostic of lithium-ion batteries. However, battery lifetime predictions are challenging due to the nonlinear battery degradation and the operational diversity among batteries. To increase the prediction accuracy, this paper proposes a hybrid framework combining the model-based method and data-driven method. In this framework, after estimating the battery capacity using online operating data, battery lifetime is predicted by the model-based empirical model as well as the data-driven support vector regression model. For the empirical model, its adaptability is improved by updating the parameters dynamically with particle filters. For the support vector regression model, its performance is optimized by an artificial bee colony algorithm. Finally, a fusion method with cascaded structure is proposed to integrate predictions from these two models, which boosts the prediction accuracy by iteratively exerting two concatenated Kalman filters. The generality and effectiveness of the proposed method are verified on battery data sets provided by NASA and our testing bench, respectively. The experimental results illustrate that the proposed method can improve the prediction accuracy of battery remaining lifetime, especially at the early stage. RMSE and MAE of the proposed hybrid framework are within 4 and 3.5. Compared with two existed hybrid methods, RMSE of prediction can be reduced by at least 7.6%. A reduction of not less than 5.9% in MAE of prediction is achieved.

Catenary-free trams have been considered a promising public transportation option for modern cities, where onboard supercapacitors are applied as the power source for the tram. Due to the frequent short-distance stops and drastic load changes, the DC bus exhibits significant voltage fluctuations, which affects the normal operation of loads. To address this challenge, in this paper, we propose a predictive set point modulation control for the onboard supercapacitors of catenary-free trams, where the DC bus voltage is predicted in real-time and the reference voltage of the closed-loop system is regulated accordingly to suppress the voltage overshooting. The system modeling of the onboard supercapacitors is presented first, and then the predictive set point modulation control is introduced in detail. A laboratory experimental platform is built to verify the effectiveness of the proposed method. Extensive experiment results show that the proposed method can effectively suppress the voltage fluctuations when compared with the classical control method.

The driving performance of electric vehicles seriously degrades due to the deterioration of lithium-ion batteries at low temperatures. Preheating lithium-ion batteries can effectively improve the driving range of electric vehicles at subzero temperatures. In this paper, an optimal pulse heating strategy is proposed for low-temperature heating of lithiumion battery. Firstly, this paper establishes a coupling model to describe the electro-thermal-aging behavior of battery. Secondly, the heating time and capacity loss jointly form a multi-objective optimization problem with the current constraint. The optimization problem is solved by using the particle swarm optimization(PSO) algorithm and the effect of weighting coefficient on heating performance is discussed to obtain the optimal pulse current. The results show that the proposed strategy can effectively reduce heating time without causing serious capacity reduction.

Normal operation of the pressure sensor is important for the safe operation of the locomotive electro-pneumatic brake system. Sensor fault diagnosis technology facilitates detection of sensor health. However, the strong nonlinearity and variable process noise of the brake system make the sensor fault diagnosis become challenging. In this paper, an adaptive unscented Kalman filter- (UKF-) based fault diagnosis strategy is proposed, aimed at detecting bias faults and drift faults of the equalizing reservoir pressure sensor in the brake system. Firstly, an adaptive UKF based on the Sage-Husa method is applied to accurately estimate the pressure transients in the equalizing reservoir of the brake system. Then, the residual is generated between the estimated pressure by the UKF and the measured pressure by the sensor. Afterwards, the Sequential Probability Ratio Test is used to evaluate the residual so that the incipient and gradual sensor faults can be diagnosed. An experimental prototype platform for diagnosis of the equalizing reservoir pressure control system is constructed to validate the proposed method.

Supercapacitors have recieved increasing attentions in emerging portable power applications. The charging process of supercapacitors significantly affects the performance of both supercapacitors and chargers. Considering the charging time of supercapacitors is typically limited in practical applications, in this paper, we propose an optimal charging method for supercapacits with the limited charging time. Firstly, we analyze existing cell balancing and charging circuits, and adopt the switched resistor circuit. Then, we design a user-interactive optimal charging method for supercapacitors where the charging time can be specified by the users. The energy efficiency maximization of the proposed charging method is proved rigorously. A simulation charging platform has been established to verify the effectiveness of the proposed charging method. The simulation results show that the proposed charging method can effectively improve the energy efficiency under charging time constraints when compared with existing methods.

Lithium-ion batteries have been widely used in many fields such as electric vehicles and smart grid. Accurately predicting its lifetime is crucial for ensuring safety and accelerating battery technological development. This study aims to develop an interpretable battery lifetime prediction method based on a machine learning model by explaining the features used in the model. Firstly, the battery charge-discharge cycle data is analyzed, and five key features related to lifetime are extracted from the discharge curve of the first 100 cycles. Then, extreme gradient boosting tree (XGBoost) is built to learn the relationship between the features and the lifetime, and its optimal parameters are obtained through grid search and five-fold cross-validation. TreeExplainer is used to calculate shapley value based on game theory to quantitatively interpret and reveal the important features contributing to lifetime. Experimental results on the latest battery dataset demonstrate that XGBoost can effectively predict battery lifetime. At the same time, quantitative analysis and interpretation provide the reasons for the model decision, which improves the credibility of the prediction method and even helps to have a deeper understanding of the battery degradation mechanism.

With the continuous promotion of the green transportation concept, supercapacitors have gained popularity for their excellent charging and discharging characteristics. However, the unreasonable management of supercapacitor will lead to poor safety and reliability of the supercapacitor system. Aiming at this problem, a supercapacitor cloud management system based on the digital twin is proposed in this paper, which improves the computing power and data storage capacity through cloud computing for the whole life cycle state monitoring of supercapacitors. Firstly, the mathematical modeling of the supercapacitor system is carried out, and the state parameters during the charging and discharging of the supercapacitor are obtained in real-time at the physical layer and uploaded to the data interaction network. Secondly, in the network layer, all data related to the supercapacitor are measured and transmitted to the cloud through IoT, and a digital twin cloud model of the supercapacitor management system is established to evaluate the usage status of the supercapacitor by using the real-time data in the data interaction network. Finally, the supercapacitor charging and discharging control strategy is adjusted in real-time based on the state evaluation results at the control layer and given a visualization window visible to users. The application of the supercapacitor system in the digital twin is explored by developing a parameter estimation algorithm suitable for cloud computing. The experimental results verify the feasibility and practicality of the proposed supercapacitor cloud management system.

Different from previous data-driven methods for lithium-ion battery State-of-Charge (SoC) estimation, this paper aims to develop a hierarchical SoC estimation method to address the data dependency issue and measurement noise interferences. In the off-line training layer, aging-aware features are extracted to improve SoC estimation accuracy throughout the entire battery life cycle. Extreme gradient boosting (XGBoost) is introduced to map the relationship between the extracted features and SoC for its strong nonlinear fitting ability. In the on-line estimation layer, Ampere-hour integral method is utilized to provide SoC reference to guarantee the stability of the proposed method. Meanwhile, to suppress the measurement noise, we adopt Kalman filter to correct the SoC value estimated by XGBoost. The superiority of the proposed method is proved under the random walk discharging experiment by comparing with the results of XGBoost, i.e., without Kalman filter. The proposed method improved the accuracy of lithium-ion battery SoC by 4% to 10%.