Machine learning of materials design and state prediction for lithium ion batteries

With the widespread use of lithium ion batteries in portable electronics and electric vehicles,further improvements in the performance of lithium ion battery materials and accurate prediction of battery state are of increasing interest to battery researchers.Machine learning,one of the core technologies of artificial intelligence,is rapidly changing many fields with its ability to learn from historical data and solve complex tasks,and it has emerged as a new technique for solving current research problems in the field of lithium ion batteries.This review begins with the introduction of the conceptual framework of machine learning and the general process of its application,then reviews some of the progress made by machine learning in both improving battery materials design and accurate prediction of battery state,and finally points out the current application problems of machine learning and future research directions.It is believed that the use of machine learning will further promote the large-scale application and improvement of lithium-ion batteries.

Visual Collaborative Maintenance Method for Urban Rail Vehicle Bogie Based on Operation State Prediction

The bogie is a crucial component of urban rail vehicles,and its performance plays a decisive role in the safe operation of vehicles.Aiming at the intelligent operation and maintenance requirements of rail transit equipment,in this paper,it takes several key parts of the urban rail vehicle bogie system as research objects,such as motor bearings,frames,fasteners,etc.,and proposes a three-dimensional(3D)visual collaborative maintenance method.Firstly,a multi-sensor urban rail vehicle bogie running simulation experiment analysis platform was constructed,thereby establishing a database of running state and performance characteristics of the bogie in the whole life cycle.Then,the health status of key components of bogie was predicted by the state interval prediction model.Finally,the three-dimensional visual collaborative maintenance model proposed in this paper was integrated to realize the early warning of the bogie operation faults,3D precise guidance of automatic location and maintenance operation information,and collaborative sharing of visual information among multiple users.

Anti-Windup Control Strategy of Drive System for Humanoid Robot Arm Joint

To address the problems of torque limit and controller saturation in the control of robot arm joint,an anti-windup control strategy is proposed for a humanoid robot arm,which is based on the integral state prediction under the direct torque control system of brushless DC motor. First,the arm joint of the humanoid robot is modelled. Then the speed controller model and the influence of the initial value of the integral element on the system are analyzed. On the basis of the traditional antiwindup controller,an integral state estimator is set up. Under the condition of different load torques and the given speed,the integral steady-state value is estimated. Therefore the accumulation of the speed error terminates when the integrator reaches saturation. Then the predicted integral steady-state value is used as the initial value of the regulator to enter the linear region to make the system achieve the purpose of anti-windup. The simulation results demonstrate that the control strategy for the humanoid robot arm joint based on integral state prediction can play the role of anti-windup and suppress the overshoot of the system effectively. The system has a good dynamic performance.

Real-time K-TIG welding penetration prediction on embedded system using a segmentation-LSTM model

Keyhole tungsten inert gas(K-TIG)welding is capable of realizing single-sided welding and double-sided forming and has been widely used in medium and thick plate welding.In order to improve the accuracy of automatic weld identification and weld penetration prediction of robot in the process of large workpiece welding,a two-stage model is proposed in this paper,which can monitor the K-TIG welding penetration state in real time on the embedded system,called segmentation-LSTM model.The proposed system extracts 9 weld pool geometric features with segmentation network,and then extracts the weld gap using a traditional algorithm.Then these 10-dimensional features are input into the LSTM model to predict the penetration state,including under penetration,partial penetration,good penetration and over penetration.The recognition accuracy of the proposed system can reach 95.2%.In this system,to solve the difficulty of labeling data and lack of segmentation accuracy,an improved LabelMe capable of live-wire annotation tool and a novel loss function were proposed,respectively.The latter was also called focal dice loss,which enabled the network to achieve a performance of 0.933 mloU on the testing set.Finally,an improved slimming strategy compresses the network,making the segmentation network achieve real-time on the embedded system(RK3399pro).

Digital Twin-Driven Intelligent Construction:Features and Trends

Digital twin(DT)can achieve real-time information fusion and interactive feedback between virtual space and physical space.This technology involves a digital model,real-time information management,comprehensive intelligent perception networks,etc.,and it can drive the rapid conceptual development of intelligent construction(IC)such as smart factories,smart cities,and smart medical care.Nevertheless,the actual use of DT in IC is partially pending,with numerous scientific factors still not clarified.An overall survey on pending issues and unsolved scientific factors is needed for the development of DT-driven IC.To this end,this study aims to provide a comprehensive review of the state of the art and state of the use of DT-driven IC.The use of DT in planning,design,manufacturing,operation,and maintenance management of IC is demonstrated and analyzed,following which the driving functions of DT in IC are detailed from four aspects:information perception and analysis,data mining and modeling,state assessment and prediction,intelligent optimization and decision-making.Furthermore,the future direction of research,using DT in IC,is presented with some comments and suggestions.This work will help researchers gain in-depth and systematic understanding of the use of DT,and help practitioners to better promote its implementation in IC.

Further monopole effects in neutron-rich Sb isotopes

The level spectra of neutron-rich Sb isotopes have been investigated within a shell-model space containing cross-shell excitations and the intruder orbit i_(13/2).High-spin levels(27/2^(-))and(29/2^(-))in 135 Sb are taken over by the monopole effect induced by orbit i_(13/2).The ground state and excited levels in ^(136)Sb are well improved by considering the monopole correction between neutron orbits f_(7/2) and h_(9/2).The energy shrinking of the first excited state 5/2^(+)in ^(135,137)Sb isotopes is explained by theπd_(5/2) shift due to the attractiveπd5/2νf_(7/2) monopole interaction when increasingly more neutrons occupy orbit f_(7/2).The ground state of ^(139)Sb is predicted as 5/2^(+)owing to the shrinking of the 5/2^(+)states in Sb isotopes that causes ground state inversion when N=88.Further monopole effects extend the applicable range of the present Hamiltonian to nuclei with more neutrons above the N=82 shell.This Hamiltonian will be public,and researchers are encouraged to contact the authors.

Trajectory prediction of ballistic missiles using Gaussian process error model

Ballistic Missile Trajectory Prediction(BMTP)is critical to air defense systems.Most Trajectory Prediction(TP)methods focus on the coast and reentry phases,in which the Ballistic Missile(BM)trajectories are modeled as ellipses or the state components are propagated by the dynamic integral equations on time scales.In contrast,the boost-phase TP is more challenging because there are many unknown forces acting on the BM in this phase.To tackle this difficult problem,a novel BMTP method by using Gaussian Processes(GPs)is proposed in this paper.In particular,the GP is employed to train the prediction error model of the boost-phase trajectory database,in which the error refers to the difference between the true BM state at the prediction moment and the integral extrapolation of the BM state.And the final BMTP is a combination of the dynamic equation based numerical integration and the GP-based prediction error.Since the trained GP aims to capture the relationship between the numerical integration and the unknown error,the modified BM state prediction is closer to the true one compared with the original TP.Furthermore,the GP is able to output the uncertainty information of the TP,which is of great significance for determining the warning range centered on the predicted BM state.Simulation results show that the proposed method effectively improves the BMTP accuracy during the boost phase and provides reliable uncertainty estimation boundaries.