North China University of Technology

About: North China University of Technology is a education organization based out in Beijing, China. It is known for research contribution in the topics: Model predictive control & Control theory. The organization has 5707 authors who have published 4748 publications receiving 29334 citations.

Traffic Flow Prediction With Big Data: A Deep Learning Approach

Abstract: Accurate and timely traffic flow information is important for the successful deployment of intelligent transportation systems. Over the last few years, traffic data have been exploding, and we have truly entered the era of big data for transportation. Existing traffic flow prediction methods mainly use shallow traffic prediction models and are still unsatisfying for many real-world applications. This situation inspires us to rethink the traffic flow prediction problem based on deep architecture models with big traffic data. In this paper, a novel deep-learning-based traffic flow prediction method is proposed, which considers the spatial and temporal correlations inherently. A stacked autoencoder model is used to learn generic traffic flow features, and it is trained in a greedy layerwise fashion. To the best of our knowledge, this is the first time that a deep architecture model is applied using autoencoders as building blocks to represent traffic flow features for prediction. Moreover, experiments demonstrate that the proposed method for traffic flow prediction has superior performance.

The sixth visual object tracking VOT2018 challenge results

Abstract: The Visual Object Tracking challenge VOT2018 is the sixth annual tracker benchmarking activity organized by the VOT initiative. Results of over eighty trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years. The evaluation included the standard VOT and other popular methodologies for short-term tracking analysis and a “real-time” experiment simulating a situation where a tracker processes images as if provided by a continuously running sensor. A long-term tracking subchallenge has been introduced to the set of standard VOT sub-challenges. The new subchallenge focuses on long-term tracking properties, namely coping with target disappearance and reappearance. A new dataset has been compiled and a performance evaluation methodology that focuses on long-term tracking capabilities has been adopted. The VOT toolkit has been updated to support both standard short-term and the new long-term tracking subchallenges. Performance of the tested trackers typically by far exceeds standard baselines. The source code for most of the trackers is publicly available from the VOT page. The dataset, the evaluation kit and the results are publicly available at the challenge website (http://votchallenge.net).

Deadbeat Predictive Current Control of Permanent-Magnet Synchronous Motors with Stator Current and Disturbance Observer

Abstract: In order to optimize the current-control performance of the permanent-magnet synchronous motor (PMSM) system with model parameter mismatch and one-step control delay, an improved deadbeat predictive current control (DPCC) algorithm for the PMSM drive systems is proposed in this paper. First, the performance of the conventional predictive current control, when parameter mismatch exist, is analyzed, and then a stator current and disturbance observer (SCDO) based on sliding-mode exponential reaching law, which is able to simultaneously predict future value of stator current and track system disturbance caused by parameter mismatch in real time, is proposed. Based on this SCDO, prediction currents are used for replacing the sampled current in DPCC to compensate one-step delay, and estimated parameter disturbances are considered as the feedforward value to compensate the voltage reference calculated by deadbeat predictive current controller. Thus, a composite control method combining the DPCC part and current prediction and feedforward compensation part based on SCDO, called DPCC + SCDO method, is developed. Moreover, based on conventional exponential reaching law, a novel sliding-mode exponential reaching law is proposed to further improve the performance of the DPCC + SCDO method. Simulation and experimental results both show the validity of the proposed current control approach.

Model Predictive Current Control for PMSM Drives With Parameter Robustness Improvement

Abstract: In order to solve the parameter dependence problem in model predictive control, an improved model predictive current control (MPCC) method based on the incremental model for surface-mounted permanent-magnet synchronous motor drives is proposed in this paper. First, the parameter sensitivity of a conventional MPCC method is analyzed, which indicates that the parameter mismatches would cause prediction current error and inaccurate delay compensation. Therefore, an incremental prediction model is introduced in this paper to eliminate the use of permanent magnetic flux linkage in a prediction model. Among the parameter of the incremental prediction model, only inductance mismatch contributes to the prediction error, since the influence of resistance mismatch on the control performance is very small. Therefore, in order to improve the antiparameter-disturbance capability of the MPCC method, an inductance disturbance controller, which includes the inductance disturbance observer and inductance extraction algorithm, is presented to update accurate inductance information for the whole control system in real time. Finally, simulation and experimental results both show that the proposed method can effectively eliminate the influence of the parameter mismatches on the control performance and reduce the parameter sensitivity of the MPCC method.

Two-Vector-Based Model Predictive Torque Control Without Weighting Factors for Induction Motor Drives

Abstract: Conventional model predictive torque control (MPTC) suffers from weighing factor tuning work and relatively high torque ripple, due to the different units of torque and stator flux and the limited number of voltage vectors. This paper proposes an improved MPTC without the use of weighting factor. The torque and flux magnitude references are converted into an equivalent reference vector of stator flux, hence eliminating the weighting factor of stator flux in conventional MPTC. Furthermore, two voltage vectors are applied during one control period to achieve better steady-state performance. Different from prior method using an active vector and a zero vector, the selected voltage vectors may be two nonzero vectors in the proposed method, which provides more opportunities to reduce both torque and flux ripples. The durations of the selected voltage vectors are determined based on the principle of stator flux error minimization. Both simulation and experimental results are presented to validate the effectiveness of the proposed method.