Publications

Automatic Identification of Teachers in Social Media using Positive Unlabeled Learning

With the emergence of online social media platforms, there has been a surge of teachers/educators turning to these platforms for professional purposes, e.g., supplementing their students' educational needs. Consequently, teachers in social media have been the subject of many educational studies. Despite the progress in this line of research, one of the major obstacles is the limited number of teachers being investigated. Current studies usually suffice to at most a few hundreds of surveyed teachers while there are thousands of other teachers online. To better understand teachers in online social media and enable modern machine learning approaches to process teacher-related data, we need to identify more teachers. Thus, this paper proposes a framework to automatically identify teachers on Pinterest– an image-based social media platform popular among teachers. We formulate the teacher identification problem as a positive unlabeled learning task where positive samples are a small set of surveyed teachers, and unlabeled samples are their connected users on Pinterest. We perform extensive experiments on a real dataset of teachers on Pinterest and show the effectiveness of our framework. We believe the proposed framework can potentially improve the quality of many research endeavors concerned with studying teachers in social media.

A novel energy management strategy for hybrid electric vehicles using deep reinforcement incentive learning

Finite-Time Control and Estimation for Complex and Practical Dynamical Systems

1Merchant Marine College, Shanghai Maritime University, Shanghai 201306, China 2Department of Mechanical and Aerospace Engineering, The Ohio State University, Columbus, OH 43202, USA 3Department of Engineering, Faculty of Engineering and Science, University of Agder, 4898 Grimstad, Norway 4School of Automation, Harbin Engineering University, Harbin 150001, China 5Deep Space Exploration Research Center, School of Astronautics, Harbin Institute of Technology, Harbin 150001, China

Mathematical Modeling and Numerical Optimization of Three Vehicle Crashes Using a Single-Mass Lumped Parameter Model

In this paper mathematical modelling of a vehicle crash based on a lumped parameter model is studied. The vehicle is modelled as a single mass connected to a non-linear spring and damper system. The characteristics of the non-linear behaviour of the model is identified with a hybrid Firefly/Harmony Search optimization algorithm that minimizes the deviation between experimental test data and a simulated response. The experimental data is taken from three crashes of an identical vehicle that crashes into a wall at different initial velocities. The aim of this paper is to find a piecewise-linear function for the spring and damper coefficients which is scaleable to reconstruct the three different experimental crashes at different impact velocities. Numerical results are provided to illustrate the applicability of the proposed algorithm. Three data sets will be used for parameter identification and a fourth data set will be used for verification

Improving Performance of Evolutionary Algorithms with Application to Fuzzy Control of Truck Backer-Upper System

We propose a method to improve the performance of evolutionary algorithms (EA). The proposed approach defines operators which can modify the performance of EA, including Levy distribution function as a strategy parameters adaptation, calculating mean point for finding proper region of breeding offspring, and shifting strategy parameters to change the sequence of these parameters. Thereafter, a set of benchmark cost functions is utilized to compare the results of the proposed method with some other well-known algorithms. It is shown that the speed and accuracy of EA are increased accordingly. Finally, this method is exploited to optimize fuzzy control of truck backer-upper system.

State and disturbance observers-based polynomial fuzzy controller

It is important to account for the gap between the dynamics of a plant to be controlled and its mathematical model to improve the control performance. Hence, a controller is designed based on this. In this study, the gap is termed as a lumped disturbance, and is considered with respect to the polynomial fuzzy model that is more effective to represent the plant dynamics. More specifically, a design of state observer is first proposed under the existence of the lumped disturbance, and this is then followed by the proposal of a disturbance observer to estimate the lumped disturbance for further use in the controller design. Finally, a controller is developed for the case in which the control state as well as the lumped disturbance are unavailable. Additionally, computer simulations are provided to illustrate the effectiveness of the proposed approach.

The Impact of the Subsidy Policy on Total Factor Productivity: An Empirical Analysis of China's Cotton Production

This paper develops one model to explore the relationship between the subsidy policy and the agricultural total factor productivity (TFP). It indicates that the agricultural TFP will be lower after the subsidy policy is implemented and there exists a negative relation between the subsidy and TFP, if subsidies are associated with the acreage. Using Malmquist index, this paper measures the changes of TFP in China's cotton production before and after the subsidy policy is implemented. The results verify that the subsidy policy could not increase but decrease the TFP of China's cotton production, not only in the whole country but also in major provinces of China. Based on the positive study, some policy implications are provided in the end of this paper.

Adaptive reliable output tracking of networked control systems against actuator faults

This paper investigates the reliable adaptive observer-based output tracking control problem for a class of networked control systems subject to actuator faults and external disturbances via equivalent-input disturbance technique. Notably, the reliable control design based on adaptive mechanism is implemented to compensate the on-line actuator faults automatically and an observer-based controller is introduced through communication networks to drive the output of controlled plant to track the output of a reference model. Moreover, due to the effect of network-induced delays and packet dropouts in the controller-to-actuator channel, the inputs of controlled plant and observer-based tracking controller are updated in an asynchronous way. Then, based on the asynchronous characteristic, the resulting closed-loop networked control system is formulated with two interval time-varying delays for obtaining the required result. In particular, the equivalent-input disturbance approach improves the disturbance rejection performance and it does not require any prior knowledge of the disturbances. By constructing a suitable Lyapunov–Krasovskii functional and using free-weighting matrix approach, a new set of sufficient conditions for the solvability of the addressed problem is derived in terms of linear matrix inequalities. At last, the proposed result is validated through two numerical examples and also a comparison study is presented which shows the effectiveness of the developed control scheme over some existing conventional control schemes.

Application of Genetic Algorithm on Parameter Optimization of Three Vehicle Crash Scenarios

This paper focuses on the development of mathematical models for vehicle frontal crashes. The models under consideration are threefold: a vehicle into barrier, vehicle-occupant and vehicle to vehicle frontal crashes. The first model is represented as a simple spring-mass-damper and the second case consists of a double-spring-mass-damper system, whereby the front mass and the rear mass represent the vehicle chassis and the occupant, respectively. The third model consists of a collision of two vehicles represented by two masses moving in opposite directions. The springs and dampers in the models are nonlinear piecewise functions of displacements and velocities respectively. More specifically, a genetic algorithm (GA) approach is proposed for estimating the parameters of vehicles front structure and restraint system for vehicle-occupant model. Finally, using the existing test-data, it is shown that the obtained models can accurately reproduce the real crash test data.

H<inf>∞</inf> filter design for time-delay Markovian jump systems

This paper investigates the H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> filtering problem for discrete time-delay Markovian jump systems with application to networked control systems. To design a full-order filter which ensures the stochastic stability and a prescribed H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> performance level for the filtering error system, the Scaled Small Gain (SSG) Theorem is developed for stochastic systems. By employing a two-term approximation to delayed state variables, the original system is transformed into an input-output form consisting of two subsystems. Based on the developed SSG Theorem and the proposed Lyapunov-Krasovskii Functional (LKF), the scaled small gains of the subsystems are analyzed to establish a new condition for the existence of the H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> filters. In addition, the corresponding H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> filter design scheme is proposed. Finally, a numerical example is presented to demonstrate effectiveness of the designed H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> filters.

Calculation of no‐load induced electromotive force of pole changing line‐start permanent magnet synchronous motor

Pole changing line‐start permanent magnet synchronous motor (LSPMSM) can significantly improve the starting capability of LSPMSM by decreasing the braking torque and pulsating torque during starting process. The fundamental reason for the elimination of braking torque is that there is no generating effect during minor‐pole starting process. That is to say, induced electromotive force (EMF) does not exist when the magnetic field generated by multi‐pole rotor PMs cut the minor‐pole stator winding. Therefore, the induced EMF of pole changing LSPMSM is analysed theoretically and calculated analytically. Considering that the magnetic flux per pole of the coils under the adjacent pole are different, the induced EMFs of a conductor, a full‐pitch coil, a coil group in minor‐pole stator winding are, respectively, calculated, and then the general analytical expression of the induced EMF of a phase winding with different rotor/stator pole ratio is derived. Then, the motors with different rotor/stator pole ratios are simulated based on finite element method to obtain the corresponding EMFs. Meanwhile, a 6/8 pole changing LSPMSM is taken as prototype. The consistency of the simulation, experiment, and the analytical calculation results effectively verify the validity of the derived general analytical expression and theoretical analysis.

A Learning-Based Vehicle-Cloud Collaboration Approach for Joint Estimation of State-of-Energy and State-of-Health

The state-of-energy (SOE) and state-of-health (SOH) are two crucial quotas in the battery management systems, whose accurate estimation is facing challenges by electric vehicles' (EVs) complexity and changeable external environment. Although the machine learning algorithm can significantly improve the accuracy of battery estimation, it cannot be performed on the vehicle control unit as it requires a large amount of data and computing power. This paper proposes a joint SOE and SOH prediction algorithm, which combines long short-term memory (LSTM), Bi-directional LSTM (Bi-LSTM), and convolutional neural networks (CNNs) for EVs based on vehicle-cloud collaboration. Firstly, the indicator of battery performance degradation is extracted for SOH prediction according to the historical data; the Bayesian optimization approach is applied to the SOH prediction combined with Bi-LSTM. Then, the CNN-LSTM is implemented to provide direct and nonlinear mapping models for SOE. These direct mapping models avoid parameter identification and updating, which are applicable in cases with complex operating conditions. Finally, the SOH correction in SOE estimation achieves the joint estimation with different time scales. With the validation of the National Aeronautics and Space Administration battery data set, as well as the established battery platform, the error of the proposed method is kept within 3%. The proposed vehicle-cloud approach performs high-precision joint estimation of battery SOE and SOH. It can not only use the battery historical data of the cloud platform to predict the SOH but also correct the SOE according to the predicted value of the SOH. The feasibility of vehicle-cloud collaboration is promising in future battery management systems.

Input-Output Finite-Time Stability of Discrete-Time Impulsive Switched Linear Systems with State Delays

No abstract is provided for this article.

Stochastic H∞ filtering for neural networks with leakage delay and mixed time-varying delays

This paper deals with the problem of H ∞ filtering for stochastic neural networks (SNNs) with a mixed of time-varying interval delays, time-varying distributed delays, and leakage delays. A novel quintuple integral Lyapunov–Krasovskii functional (LKF) is constructed to improve the performance of the SNN. Sufficient criteria can be obtained by applying the linear matrix inequality (LMI) approach and developing a new mathematical analysis, which ensures the filtering error system is asymptotically stable in the mean square. Finally, simulation results are provided to show the superiority and usefulness of the proposed method.

An LMI approach to H&lt;inf&gt;&amp;#x221E;&lt;/inf&gt; synchronization of second-order neutral master-slave systems

The H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">infin</sub> synchronization problem of the master and slave structure of a second-order neutral master-slave systems with time-varying delays is presented in this paper. Delay-dependent sufficient conditions for the design of a delayed output-feedback control are given by Lyapunov-Krasovskii method in terms of a linear matrix inequality (LMI). A controller, which guarantees H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">infin</sub> synchronization of the master and slave structure using some free weighting matrices, is then developed. A numerical example has been given to show the effectiveness of the method.

A resource-aware sliding mode control approach for Markov jump systems

The problem of sliding mode control (SMC) for a class of Markov jump systems (MJSs) is addressed in this paper based on a resource-aware triggering mechanism which realizes computational resources saving and disturbance attenuation simultaneously. By introducing the self-triggered policy, the next execution time is pre-computed for sampling, updating and executing by relying on the latest sampled information. Then, the switching surface and the related dynamics of the original MJSs are obtained by means of a self-triggered sampling scheme. To guarantee both the system stability and the desired disturbance attenuation performance, sufficient conditions are presented in terms of linear matrix inequalities. Moreover, to ensure the time finiteness of the predefined switching surface reachability and satisfy the desirable sliding motion performance, an SMC law is proposed. The validity and superiority of the developed scheme are demonstrated via a simulation example.

Stabilization of positive switched systems with time-varying delays under asynchronous switching

No abstract is provided for this article.

A fuzzy programming method for optimization of autonomous logistics objects

Recently several studies have explored the realization of autonomous control in production and logistic operations. In doing so, it has been tried to transmit the merit of decision-making from central controllers with offline decisions to decentralized controllers with local and real-time decision makings. However, this mission has still some drawbacks in practice. Lack of global optimization is one of them, i.e., the lost chain between the autonomous decentralized decisions at operational level and the centralized mathematical optimization with offline manner at tactical and strategic levels. This distinction can be reasonably solved by considering fuzzy parameters in mathematical programming to meet the required tolerances for autonomous objects at operational level. This claim is recommended and partially experimented in this paper. An assembly scenario is modeled by a discrete-event simulation, in which autonomous pallets carry products throughout the system. This scenario is optimized with regard to its objectives in a simulation, while fuzzy parameters in optimization programming can consider autonomous decisions done at operational level.

Adaptive Event-Triggered Consensus of Multiagent Systems on Directed Graphs

This article systematically studies consensus of linear multiagent systems (MASs) on directed graphs through adaptive event-triggered control. It presents innovative adaptive event-triggered state-feedback protocols with novel composite event-triggering conditions. Two specific designs in terms of different event-triggering conditions and laws of adaption are first discussed for linear MASs on strongly connected directed graphs, which are then extended to general directed graphs that contain a spanning tree. Moreover, another adaptive event-triggered protocol is proposed for solving leader-follower consensus that tracks a leader of a bounded control input. The protocols inherit the merits of both adaptive control and event-triggered control: the protocols can be implemented in a fully distributed way, since the Laplacian is avoided in design, and each agent only needs to know the relative information between neighbors at discrete instants determined by event-triggering conditions. Compared with the existing related results, the proposed protocols are applicable for linear MASs on general directed graphs, and moreover, the time-dependent term in the event-triggering conditions is allowed to be a class of positive L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> functions. Two numerical examples clearly verify the effectiveness of the proposed protocols.