Search

Accurate and reliable prediction of PM2.5 concentrations is the basis for appropriate warning measures, and a single prediction model is often ineffective. In this paper, we propose a novel decomposition-and-ensemble model to predict the concentration of PM2.5. The model utilizes Ensemble Empirical Mode Decomposition (EEMD) to decompose PM2.5 series, Support Vector Regression (SVR) to predict each Intrinsic Mode Function (IMF), and a hybrid algorithm based on Differential Evolution (DE) and Grey Wolf Optimizer (GWO) to optimize SVR parameters. The proposed prediction model EEMD-SVR-DEGWO is employed to forecast the concentration of PM2.5 in Guangzhou, Wuhan, and Chongqing of China. Compared with six prediction models, the proposed EEMD-SVR-DEGWO is a reliable predictor and has achieved competitive results. [ABSTRACT FROM AUTHOR]

In the past, the extreme value standardization of indicators, the traditional weighting method, and the multifactor comprehensive model of land intensive use inevitably linearly correlate the evaluation indicators with the evaluation objects, ignoring the direction differences of different indicators in different intervals. At the same time, these methods are also difficult to meet the change of evaluation index weight value with land use type, and cannot adapt to the actual situation of land use environment level and dynamic change. Considering the objectivity of nonlinear correlation moderate index and weight assignment, based on the standardization of quadratic function index and entropy assignment, this paper studies the intensive and dynamic use of land in development zones by different regions to improve the realistic fit of the evaluation model. The results show that the overall level of land intensive use in Chongqing center district and western Chongqing is better than that in northeast Chongqing and southeast Chongqing, roughly showing the state of "high in west and low in east." [ABSTRACT FROM AUTHOR]

As an effective reinforcement learning (RL) algorithm, Q-learning has been applied to energy management strategy of hybrid electric vehicle (HEV) in recent years. In the existing literatures, the values of three hyperparameters based on Q-learning are all given in advance, which are respectively exploratory rate ε, discount factor γ and learning rate α. However, different values of hyperparameters will influence on fuel economy of the vehicle and offline computation speed. In this paper, it is proposed that the method of optimization on hyperparameters adapting to driving cycle. Firstly, the mathematical model between three hyperparameters and iteration times is established based on inherent regularity of hyperparameters influencing on vehicle performance respectively. Secondly, it is determined that the optimal changing index k_index of iteration number based on Q-learning corresponding to typical driving cycles. Finally, the simulation model of Yubei District in Chongqing is constructed based on the method of Kullback-Leibler (KL) driving cycle identification. The simulation results indicate that equivalent fuel consumption of the proposed strategy is reduced by 0.4 % and the offline operation time is reduced by 6 s. It can be concluded that the proposed strategy can not only improve fuel economy of the vehicle, but also accelerate the computation speed. [ABSTRACT FROM AUTHOR]