arXiv subjects

Jing Sun

Publications and source records attributed to Jing Sun.

At least 109 records · Page 6Linked to original sources

arXiv2021

A Unified Joint Maximum Mean Discrepancy for Domain Adaptation

Domain adaptation has received a lot of attention in recent years, and many algorithms have been proposed with impressive progress. However, it is still not fully explored concerning the joint probability distribution (P(X, Y)) distance for this problem, since its empirical estimation derived from the maximum mean discrepancy (joint maximum mean discrepancy, JMMD) will involve complex tensor-product operator that is hard to manipulate. To solve this issue, this paper theoretically derives a unified form of JMMD that is easy to optimize, and proves that the marginal, class conditional and weighted class conditional probability distribution distances are our special cases with different label kernels, among which the weighted class conditional one not only can realize feature alignment across domains in the category level, but also deal with imbalance dataset using the class prior probabilities. From the revealed unified JMMD, we illustrate that JMMD degrades the feature-label dependence (discriminability) that benefits to classification, and it is sensitive to the label distribution shift when the label kernel is the weighted class conditional one. Therefore, we leverage Hilbert Schmidt independence criterion and propose a novel MMD matrix to promote the dependence, and devise a novel label kernel that is robust to label distribution shift. Finally, we conduct extensive experiments on several cross-domain datasets to demonstrate the validity and effectiveness of the revealed theoretical results.

cs.LG↗

arXiv2020

Robust State of Health Estimation of Lithium-ion Batteries Using Convolutional Neural Network and Random Forest

The State of Health (SOH) of lithium-ion batteries is directly related to their safety and efficiency, yet effective assessment of SOH remains challenging for real-world applications (e.g., electric vehicle). In this paper, the estimation of SOH (i.e., capacity fading) under partial discharge with different starting and final State of Charge (SOC) levels is investigated. The challenge lies in the fact that partial discharge truncates the data available for SOH estimation, thereby leading to the loss or distortion of common SOH indicators. To address this challenge associated with partial discharge, we explore the convolutional neural network (CNN) to extract indicators for both SOH and changes in SOH ($Δ$SOH) between two successive charge/discharge cycles. The random forest algorithm is then adopted to produce the final SOH estimate by exploiting the indicators from the CNNs. Performance evaluation is conducted using the partial discharge data with different SOC ranges created from a fast-discharging dataset. The proposed approach is compared with i) a differential analysis-based approach and ii) two CNN-based approaches using only SOH and $Δ$SOH indicators, respectively. Through comparison, the proposed approach demonstrates improved estimation accuracy and robustness. Sensitivity analysis of the CNN and random forest models further validates that the proposed approach makes better use of the available partial discharge data for SOH estimation.

eess.SP↗

arXiv2020

Strong convergence order for the scheme of fractional diffusion equation driven by fractional Gaussion noise

Fractional Gaussian noise models the time series with long-range dependence; when the Hurst index $H>1/2$, it has positive correlation reflecting a persistent autocorrelation structure. This paper studies the numerical method for solving stochastic fractional diffusion equation driven by fractional Gaussian noise. Using the operator theoretical approach, we present the regularity estimate of the mild solution and the fully discrete scheme with finite element approximation in space and backward Euler convolution quadrature in time. The $\mathcal{O}(τ^{H-ρα})$ convergence rate in time and $\mathcal{O}(h^{\min(2,2-2ρ,\frac{H}α)})$ in space are obtained, showing the relationship between the regularity of noise and convergence rates, where $ρ$ is a parameter to measure the regularity of noise and $α\in(0,1)$. Finally, numerical experiments are performed to support the theoretical results.

math.NA↗

arXiv2020

High-order BDF fully discrete scheme for backward fractional Feynman-Kac equation with nonsmooth data

The Feynman-Kac equation governs the distribution of the statistical observable -- functional, having wide applications in almost all disciplines. After overcoming challenges from the time-space coupled nonlocal operator and the possible low regularity of functional, this paper develops the high-order fully discrete scheme for the backward fractional Feynman-Kac equation by using backward difference formulas (BDF) convolution quadrature in time, finite element method in space, and some correction terms. With a systematic correction, the high convergence order is achieved up to $6$ in time, without deteriorating the optimal convergence in space and without the regularity requirement on the solution. Finally, the extensive numerical experiments validate the effectiveness of the high-order schemes.

math.NA↗

arXiv2020

Importance Filtered Cross-Domain Adaptation

In Domain Adaptation (DA), the category-relevant losses usually occupy a dominant position, while they are usually built with hard or soft labels in existing models. We observed that hard labels are overconfident due to hard samples existed, and soft labels are ambiguous as too many small noisy probabilities involved, and both of them are easily to cause negative transfer. Besides, the category-irrelevant losses in Closed-Set DA (CSDA) paradigm fail to work in Open-Set DA (OSDA), and they also have to be in a category-relevant form, since target data samples are split into shared and private classes. To this end, we propose a newly-unified DA framework (i.e., Importance Filtered Cross-Domain Adaptation, IFCDA). Firstly, an importance filtered mechanism is devised to generate filtered soft labels to mitigate negative transfer desirably. Specifically, the soft labels are divided into confident and ambiguous ones. Then, only the maximum probability in each confident label is retained, and a threshold value is set to truncate each ambiguous label so that only prominent probabilities are reserved. Moreover, a general graph-based label propagation is contrived to attain soft labels in both CSDA and OSDA, where an extra component is embedded into label vector, so that it could detect target novel classes. Finally, the category-relevant losses in both scenarios are reformulated using filtered soft labels, while the category-irrelevant MMD loss in CSDA is reformulated as a form like class-wise MMD using newly-designed importance filtered soft labels. Notably, CSDA paradigm is a special case when all extra components are set to 0, thus the proposed approach is geared to both CSDA and OSDA. Comprehensive experiments on benchmark cross-domain object recognition datasets verify that the proposed approach outperforms several state-of-the-art methods in both scenarios.

cs.LG↗

arXiv2020

Sparsely-Labeled Source Assisted Domain Adaptation

Domain Adaptation (DA) aims to generalize the classifier learned from the source domain to the target domain. Existing DA methods usually assume that rich labels could be available in the source domain. However, there are usually a large number of unlabeled data but only a few labeled data in the source domain, and how to transfer knowledge from this sparsely-labeled source domain to the target domain is still a challenge, which greatly limits their application in the wild. This paper proposes a novel Sparsely-Labeled Source Assisted Domain Adaptation (SLSA-DA) algorithm to address the challenge with limited labeled source domain samples. Specifically, due to the label scarcity problem, the projected clustering is conducted on both the source and target domains, so that the discriminative structures of data could be leveraged elegantly. Then the label propagation is adopted to propagate the labels from those limited labeled source samples to the whole unlabeled data progressively, so that the cluster labels are revealed correctly. Finally, we jointly align the marginal and conditional distributions to mitigate the cross-domain mismatch problem, and optimize those three procedures iteratively. However, it is nontrivial to incorporate those three procedures into a unified optimization framework seamlessly since some variables to be optimized are implicitly involved in their formulas, thus they could not promote to each other. Remarkably, we prove that the projected clustering and conditional distribution alignment could be reformulated as different expressions, thus the implicit variables are revealed in different optimization steps. As such, the variables related to those three quantities could be optimized in a unified optimization framework and facilitate to each other, to improve the recognition performance obviously.

cs.CV↗

arXiv2020

Individual Cell Fault Detection for Parallel-Connected Battery Cells Based on the Statistical Model and Analysis

Fault diagnosis is extremely important to the safe operation of Lithium-ion batteries. To avoid severe safety issues (e.g., thermal runaway), initial faults should be timely detected and resolved. In this paper, we consider parallel-connected battery cells with only one voltage and one current sensor. The lack of independent current sensors makes it difficult to detect individual cell degradation. To this end, based on the high-frequency response of the battery, a simplified fault detection-oriented model is derived and validated by a physics-informed battery model. The resistance of the battery string, which is significantly influenced by the faulty cell, is estimated and used as the health indicator. The statistical resistance distribution of battery strings is first analyzed considering the distribution of fresh and aged cells. A fault diagnosis algorithm is proposed and the thresholds (i.e., 2 standard deviation interval) are obtained through statistical analysis. Monte Carlo simulation results show that the proposed fault diagnosis algorithm can balance false alarms and missed detections well. In addition, it is verified that the proposed algorithm is robust to the uniform parameter changes of individual battery cells.

eess.SY↗

arXiv2020

Engine and Aftertreatment Co-Optimization of Connected HEVs via Multi-Range Vehicle Speed Planning and Prediction

Connected vehicles (CVs) have situational awareness that can be exploited for control and optimization of the powertrain system. While extensive studies have been carried out for energy efficiency improvement of CVs via eco-driving and planning, the implication of such technologies on the thermal responses of CVs has not been fully investigated. One of the key challenges in leveraging connectivity for optimization-based thermal management of CVs is the relatively slow thermal dynamics, which necessitate the use of a long prediction horizon to achieve the best performance. Long-term prediction of the CV speed, unlike the V2V/V2I-based short-range prediction, is difficult and error-prone. The multiple timescales inherent to power and thermal systems call for a variable timescale optimization framework with access to short- and long-term vehicle speed preview. To this end, a model predictive controller (MPC) with a multi-range speed preview for integrated power and thermal management (iPTM) of connected hybrid electric vehicles (HEVs) is presented in this paper. The MPC is formulated to manage the power-split between the engine and the battery while enforcing the power and thermal (engine coolant and catalytic converter temperatures) constraints. The MPC exploits prediction and optimization over a shorter receding horizon and longer shrinking horizon. Over the longer shrinking horizon, the vehicle speed estimation is based on the data collected from the connected vehicles traveling on the same route as the ego-vehicle. Simulation results of applying the MPC over real-world urban driving cycles in Ann Arbor, MI are presented to demonstrate the effectiveness and fuel-saving potentials of the proposed iPTM strategy under the uncertainty associated with long-term predictions of the CV's speed.

eess.SY↗

arXiv2020

Integrated Power and Thermal Management of Connected HEVs via Multi-Horizon MPC

In this paper, a multi-horizon model predictive controller (MH-MPC) is developed for integrated power and thermal management (iPTM) of a power-split hybrid electric vehicle (HEV). The proposed MH-MPC leverages an accurate short-horizon vehicle speed preview and an approximate forecast over a longer shrinking horizon till the end of the driving cycle. This multiple-horizon scheme is developed to cope with fast and slow dynamics associated with power and thermal responses. The main objective of the proposed MH-MPC is to minimize fuel consumption and enforce the power and thermal constraints on the battery state-of-charge and engine coolant temperature, while meeting the driving (traction) and cabin air conditioning (heating) demands. The proposed MH-MPC allows for exploiting the engine coolant as thermal energy storage, providing more flexibility for the HEV energy flow optimization. The simulation results show that the proposed MH-MPC provides near-optimal results in reference to the Dynamic Programming (DP) solution with an affordable computational cost. Moreover, compared with a more conventional MPC strategy, the MH-MPC can leverage the speed previews with different resolutions effectively to achieve the desired performance with satisfactory robustness.

eess.SY↗

arXiv2020

Measuring the Quality of B Abstract Machines with ISO/IEC 25010

The B method has facilitated the development of software by specifying the design of software as abstract machines and formally verifying the correctness of the abstract machines. The quality of B abstract machines can significantly impact the quality of final software products. In this paper, we propose a set of criteria for measuring the quality of B abstract machines based on ISO/IEC 25010, which is one of the latest international standards for evaluating software quality in software engineering. These criteria evaluate abstract machines using a number of general-purpose and domain-independent equations and model checking techniques, so that the quality of abstract machines can be quantified as vectors. The proposed criteria are implemented as a B model quality evaluator, and they are explained and justified using a number of examples.

cs.SE↗

arXiv2019

Error estimates for backward fractional Feynman-Kac equation with non-smooth initial data

In this paper, we are concerned with the numerical solution for the backward fractional Feynman-Kac equation with non-smooth initial data. Here we first provide the regularity estimate of the solution. And then we use the backward Euler and second-order backward difference convolution quadratures to approximate the Riemann-Liouville fractional substantial derivative and get the first- and second-order convergence in time. The finite element method is used to discretize the Laplace operator with the optimal convergence rates. Compared with the previous works for the backward fractional Feynman-Kac equation, the main advantage of the current discretization is that we don't need the assumption on the regularity of the solution in temporal and spatial directions. Moreover, the error estimates of the time semi-discrete schemes and the fully discrete schemes are also provided. Finally, we perform the numerical experiments to verify the effectiveness of the presented algorithms.

math.NA↗

arXiv2019

Question Answering based Clinical Text Structuring Using Pre-trained Language Model

Clinical text structuring is a critical and fundamental task for clinical research. Traditional methods such as taskspecific end-to-end models and pipeline models usually suffer from the lack of dataset and error propagation. In this paper, we present a question answering based clinical text structuring (QA-CTS) task to unify different specific tasks and make dataset shareable. A novel model that aims to introduce domain-specific features (e.g., clinical named entity information) into pre-trained language model is also proposed for QA-CTS task. Experimental results on Chinese pathology reports collected from Ruijing Hospital demonstrate our presented QA-CTS task is very effective to improve the performance on specific tasks. Our proposed model also competes favorably with strong baseline models in specific tasks.

cs.CL↗

arXiv2019

Numerical algorithm for the model describing anomalous diffusion in expanding media

We provide a numerical algorithm for the model characterizing anomalous diffusion in expanding media, which is derived in [F. Le Vot, E. Abad, and S. B. Yuste, Phys. Rev. E {\bf96} (2017) 032117]. The Sobolev regularity for the equation is first established. Then we use the finite element method to discretize the Laplace operator and present error estimate of the spatial semi-discrete scheme based on the regularity of the solution; the backward Euler convolution quadrature is developed to approximate Riemann-Liouville fractional derivative and error estimates for the fully discrete scheme are established by using the continuity of solution. Finally, the numerical experiments verify the effectiveness of the algorithm.

math.NA↗

arXiv2019

Combined Energy and Comfort Optimization of Air Conditioning System in Connected and Automated Vehicles

In this paper, we propose a combined energy and comfort optimization (CECO) strategy for the air conditioning (A/C) system of the connected and automated vehicles (CAVs). By leveraging the weather and traffic predictions enabled by the emerging CAV technologies, the proposed strategy is able to minimize the A/C system energy consumption while maintaining the occupant thermal comfort (OTC) within the comfort constraints, where the comfort is quantified by a modified predictive mean vote (PMV) model adapted for an automotive application. A general CECO problem is formulated and addressed using model predictive control (MPC) and weather/traffic previews. Depending on the ways of exploiting the preview information and enforcing the OTC constraint, different MPCs are developed based on solving different variations of the general CECO problem. The CECO-based MPCs are then tested in simulation using an automotive A/C system simulation model (CoolSim) as the virtual testbed. The simulation results show that, over SC03 driving cycle, the proposed CECO-based MPCs outperform the baseline cabin temperature tracking controller, reducing the A/C system energy consumption by up to 7.6%, while achieving better OTC according to the PMV-based metrics. This energy saving in A/C system translates to 3.1% vehicle fuel economy improvement. The trade-off between energy efficiency and OTC for different control scenarios is also highlighted.

eess.SY↗

arXiv2019

Simultaneous Identification and Control Using Active Signal Injection for Series Hybrid Electric Vehicles based on Dynamic Programming

Hybrid electric vehicles (HEVs) have an over-actuated system by including two power sources, a battery pack and an internal combustion engine. This feature of HEV is exploited in this paper to simultaneously achieve accurate identification of battery parameters/states. By actively injecting current signals, state of charge, state of health, and other battery parameters can be estimated in a specific sequence to improve the identification performance when compared to the case where all parameters and states are estimated concurrently using the baseline current signals. A dynamic programming strategy is developed to provide the benchmark results about how to balance the conflicting objectives corresponding to identification and system efficiency. The tradeoff between different objectives is presented to optimize the current profile so that the richness of signal can be ensured and the fuel economy can be optimized. In addition, simulation results show that the Root-Mean-Square error of the estimation can be decreased by up to 100% at a cost of less than 2% increase in fuel consumption. With the proposed simultaneous identification and control algorithm, the parameters/states of the battery can be monitored to ensure safe and efficient application of the battery for HEVs.

eess.SY↗

arXiv2019

Robust Hierarchical MPC for Handling Long Horizon Demand Forecast Uncertainty with Application to Automotive Thermal Management

This paper presents a robust hierarchical MPC (H-MPC) for dynamic systems with slow states subject to demand forecast uncertainty. The H-MPC has two layers: (i) the scheduling MPC at the upper layer with a relatively long prediction/planning horizon and slow update rate, and (ii) the piloting MPC at the lower layer over a shorter prediction horizon with a faster update rate. The scheduling layer MPC calculates the optimal slow states, which will be tracked by the piloting MPC, while enforcing the system constraints according to a long-range and approximate prediction of the future demand/load, e.g., traction power demand for driving a vehicle. In this paper, to enhance the H-MPC robustness against the long-term demand forecast uncertainty, we propose to use the high-quality preview information enabled by the connectivity technology over the short horizon to modify the planned trajectories via a constraint tightening approach at the scheduling layer. Simulation results are presented for a simplified vehicle model to confirm the effectiveness of the proposed robust H-MPC framework in handling demand forecast uncertainty.

math.OC↗

arXiv2019

Effects of restrained degradation on gene expression and regulation

The effects of carrying capacity of environment $K$ for degradation (the $K$ effect for short) on the constitutive gene expression and a simple genetic regulation system, are investigated by employing a stochastic Langevin equation combined with the corresponding Fokker-Planck equation for the two stochastic systems subjected to internal and external noises. This $K$ effect characterizes the limited degradation ability of the environment for RNA or proteins, such as insufficient catabolic enzymes. The $K$ effect could significantly change the distribution of mRNA copy-number in constitutive gene expression, and interestingly, it leads to the Fano factor slightly larger than 1 if only the internal noise exists. Therefore, that the recent experimental measurements suggests the Fano factor deviates from 1 slightly (Science {\bf 346} (2014) 1533), probably originates from the $K$ effect. The $K$ effects on the steady and transient properties of genetic regulation system, have been investigated in detail. It could enhance the mean first passage time significantly especially when the noises are weak and reduce the signal-to-noise ratio in stochastic resonance substantially.

physics.bio-ph↗

arXiv2019

The Sequential Algorithm for Combined State of Charge and State of Health Estimation of Lithium Ion Battery based on Active Current Injection

When State of Charge, State of Health, and parameters of the Lithium-ion battery are estimated simultaneously, the estimation accuracy is hard to be ensured due to uncertainties in the estimation process. To improve the estimation performance a sequential algorithm, which uses frequency scale separation and estimates parameters/states sequentially by injecting currents with different frequencies, is proposed in this paper. Specifically, by incorporating a high-pass filter, the parameters can be independently characterized by injecting high-frequency and medium-frequency currents, respectively. Using the estimated parameters, battery capacity and State of Charge can then be estimated concurrently. Experimental results show that the estimation accuracy of the proposed sequential algorithm is much better than the concurrent algorithm where all parameters/states are estimated simultaneously, and the computational cost can also be reduced. Finally, experiments are conducted under different temperatures to verify the effectiveness of the proposed algorithm for various battery capacities.

eess.SY↗