In this paper, a deep residual network based on convolutional block attention module (CBAM) is proposed, which is utilized for feature extraction of partially occluded face expression data. The proposed method overcom...
In this paper, a deep residual network based on convolutional block attention module (CBAM) is proposed, which is utilized for feature extraction of partially occluded face expression data. The proposed method overcomes the problem of localized occlusion face feature extraction by focusing on the regions and channels containing important information in the occluded face data through CBAM. Multi-task cascaded convolutional networks (MTCNN) are firstly utilized to localize the key regions of face emotion, and then deep emotion features are extracted by CBAM-ResNet network. The final emotion labels are generated. The effectiveness of this paper's method is verified on the RAF-DB dataset and the occluded CK+ dataset. The experimental accuracy in the RAF-DB dataset is 76.3%, which is 3.74% and 1.64% higher than the accuracy produced by the method of RGBT, and the WLS-RF, respectively. Application experiments are carried out in the real teaching scenario, which verifies the applicability of the algorithm in the real teaching scene.
This paper considers distributed online nonconvex optimization with time-varying inequality constraints over a network of agents. For a time-varying graph, we propose a distributed online primal–dual algorithm with c...
详细信息
control of systems with operating condition-dependent dynamics, including control moment gyroscopes, often requires operating condition-dependent controllers to achieve high control performance. The aim of this paper ...
详细信息
With the increasing penetration of renewable energy resources(RESs), the uncertainties of volatile renewable generations significantly affect the power system operation. Such uncertainties are usually modeled as stoch...
详细信息
With the increasing penetration of renewable energy resources(RESs), the uncertainties of volatile renewable generations significantly affect the power system operation. Such uncertainties are usually modeled as stochastic variables obeying specific distributions by neglecting the temporal correlations. Conventional approaches to hedge the negative effects caused by such uncertainties are thus hard to pursue a trade-off between computation efficiency and optimality. As an alternative, the theory of stochastic process can naturally model temporal correlation in closed forms. Attracted by this feature, our research group has been conducting thorough researches in the past decade to introduce stochastic processes within renewable power systems. This paper summarizes our works from the perspective of both the frequency domain and the time domain, provides the tools for the analysis and control of power systems under a unified framework of stochastic processes, and discusses the underlying reasons that stochastic process-based approaches can perform better than conventional approaches on both computational efficiency and optimality. These work may shed a new light on the research of analysis, control and operation of renewable power ***, this paper outlooks the theoretic developments of stochastic processes in future’s renewable power systems.
If a concatenated Friedkin-Johnsen model is used to describe the evolution of the opinions of stubborn agents in a sequence of discussion events, then the social power achieved by the agents at the end of the discussi...
详细信息
ISBN:
(数字)9781665467612
ISBN:
(纸本)9781665467629
If a concatenated Friedkin-Johnsen model is used to describe the evolution of the opinions of stubborn agents in a sequence of discussion events, then the social power achieved by the agents at the end of the discussions depends from the stubbornness coefficients adopted by the agents through the sequence of events. In this paper we assume that the agents are free to choose their stubbornness profiles, and ask ourselves what strategy should an agent follow in order to maximize its social power. Formulating the problem as a strategic game, we show that choosing the highest possible values of stubbornness in the early discussions leads to the highest possible social power.
Data-driven modeling of nonlinear dynamical systems often requires an expert user to take critical decisions a priori to the identification procedure. Recently, an automated strategy for data driven modeling of single...
详细信息
Data-driven modeling of nonlinear dynamical systems often requires an expert user to take critical decisions a priori to the identification procedure. Recently, an automated strategy for data driven modeling of single-input single-output (SISO) nonlinear dynamical systems based on genetic programming (GP) and tree adjoining grammars (TAG) was introduced. The current paper extends these latest findings by proposing a multi-input multi-output (MIMO) TAG modeling framework for polynomial NARMAX models. Moreover, we introduce a TAG identification toolbox in Matlab that provides implementation of the proposed methodology to solve multi-input multi-output identification problems under NARMAX noise assumption. The capabilities of the toolbox and the modeling methodology are demonstrated in the identification of two SISO and one MIMO nonlinear dynamical benchmark models.
Coke pushing current is an indicator to evaluate the difficulty of coke pushing operation. The higher coke pushing current is, the greater coke pushing resistance is, and the more difficult coke is to push out. Accura...
详细信息
Coke pushing current is an indicator to evaluate the difficulty of coke pushing operation. The higher coke pushing current is, the greater coke pushing resistance is, and the more difficult coke is to push out. Accurate prediction of current peak during future coke pushing operation can provide more time for production personnel to adjust production status and avoid difficult coke pushing in carbonization chamber. In this paper, a combination prediction model based on VMD (Variational Mode Decomposition) and improved ARIMA (Autoregressive Integrated Moving Average) models is proposed. Firstly, VMD algorithm is used to decompose time series of coke pushing current peak, de-noising data, and extracting main information of time series. Then, ARIMA model is used to predict mean change of linear elasticity and GARCH (Generalized Autore-gressive Conditional Heteroskedasticity) model is introduced to predict ARIMA model residual and improve heteroscedasticity of nonlinear part of time series, and then ARIMA-GARCH model is established. Finally, predicted value is obtained by the sum of each component prediction. The experimental results show that the proposed prediction model has a high prediction accuracy in the short-term prediction of coke pushing current peak. The scheme is applied to actual coking production to guide production of coke.
This paper considers distributed online convex optimization with adversarial constraints. In this setting, a network of agents makes decisions at each round, and then only a portion of the loss function and a coordina...
详细信息
The novel active simultaneously transmitting and reflecting surface (ASTARS) has recently received a lot of attention due to its capability to conquer the multiplicative fading loss and achieve full-space smart radio ...
详细信息
A general upper bound for topological entropy of switched nonlinear systems is constructed, using an asymptotic average of upper limits of the matrix measures of Jacobian matrices of strongly persistent individual mod...
详细信息
暂无评论