• Energy Research
• engineering and technology close
• Chinese Academy of Sciences close

Abstract As one of the four major industrial raw materials in the world, natural rubber is closely related to the national economy and people’s livelihood. The analysis of natural rubber price and volatility can give hedging guidance to manufacturers and provide investors with uncertainty and risk information to reduce investment losses. To effectively analyses and forecast the natural rubber’s price and volatility, this paper constructed a hybrid model that integrated the bidirectional gated recurrent unit and variational mode decomposition for short-term prediction of the natural rubber futures on the Shanghai Futures Exchange. In data preprocessing period, time series is decomposed by variational mode decomposition to capture the tendency and mutability information. The bidirectional gated recurrent unit is introduced to return the one-day-ahead prediction of the closing price and 7-day volatility for the natural rubber futures. The experimental results demonstrated that: (a) variational mode decomposition is an effective method for time series analysis, which can capture the information closely related to the market fluctuations; (b) the bidirectional neural network structure can significantly improve the model performance both in terms of fitting performance and the trend prediction; (c) a correspondence was found between the predicted target, i.e., the price and volatility, and the intrinsic modes, which manifested as the impact of the long-term and short-term characteristics on the targets at different time-scales. With a change in the time scale of forecasting targets, it was found that there was some variation in matching degree between the forecasting target and the mode sub-sequences.

Abstract Although the “resource curse” hypothesis has been supported by many empirical studies focusing on the transnational and regional levels, there has always been a point of contention about the proper method of estimating and selecting natural resource indicators in literature. This paper proposes a mixed estimation method for panel data of 10 typical oil and gas resource-based cities in China from 1998 to 2015. The results show that resource industry agglomeration can mirror the distribution and dependence on the resource industry by location rather than by merely measuring the influence of the relative scale of resource extraction on economic growth. Using resource industry agglomeration as the main explanatory variable for regional economic growth verifies the resource curse hypothesis and shows the nonlinear characteristics of the negative correlation between resources and economic growth. Despite mostly similar indicator parameter estimates, marked differences exist in measured effects for material capital investments and technological innovation investments. The result of using resource industry agglomeration as the main explanatory variable is basically consistent with the economic theory and is more in line with observed reality than the alternative indicator. The research conclusions can provide evidence and data for index selection in the studies on the mediating effect of resource curse transmission and international comparison.

Abstract Biomass, as a renewable and sustainable energy resource, can be converted into environmentally friendly and practically valuable biofuels and chemical materials via pyrolysis. However, the process optimization and pyrolysis efficiency are restricted by the limited perception of the complicated mechanisms and kinetics for biomass pyrolysis. Here, to establish an in-depth mechanism model for biomass pyrolysis, we presented a novel investigation for the thermal evolutions and pyrolysis kinetics of the functional groups in peanut shell matrix by using in-situ Fourier transform infrared spectrometry (in-situ FTIR) and thermogravimetric analysis-Fourier transform infrared spectrometry-mass spectrometry (TG-FTIR-MS). The in-situ FTIR spectrum deconvolution for the solid matrix was innovatively introduced to identify and quantify the real-time evolution and thermal dynamics of the functional groups during peanut shell pyrolysis. The result for the first time proposed that the pyrolysis mechanisms of total OH at 20–380 °C, aliphatic C-Hn groups at 20–500 °C, C O groups at 260–500 °C, and C–O groups at 300–500 °C were dominant by diffusion and order-based chemical reactions. The TG-FTIR-MS analysis was conducted for the online monitoring of the released volatiles and gases, the amounts of which were in the sequence of C O > CO2 > aliphatic C–O-(H) > C–O-(C) in esters > aromatics > H2O > phenolic hydroxyl > aliphatic hydrocarbons > CO. The study established a novel methodology to evaluate the biomass pyrolysis mechanisms at the molecular level, which provided valuable information for developing advanced pyrolysis techniques on a large scale for sustainable ecosystem.

A simple low cost method is proposed for the quantitative analysis of residual alcohol in biodiesel through determination of the flash point, with which it is correlated. Methyl ester biodiesels from vegetable oils such as corn, soy and sunflower were prepared. The ethyl ester was obtained from soy oil and methyl biodiesel was also synthesized from bovine fat. In all cases it became very evident that there is a direct correlation between the flash point and the residual alcohol content in the prepared biodiesel. Therefore this parameter can be used to directly determine the residual alcohol content of the product.

Fashion image clustering is the key to fashion retrieval, forecasting, and recommendation applications. Manual labeling-based clustering is both time-consuming and less accurate. Currently, popular methods for extracting features from data use deep learning techniques, such as a Convolutional Neural Network (CNN). These methods can generate high-dimensional feature vectors, which are effective for image clustering. However, high dimensions can lead to the curse of dimensionality, which makes subsequent clustering difficult. The fashion images-oriented deep clustering method (FIDC) is proposed in this paper. This method uses CNN to generate a 4096-dimensional feature vector for each fashion image through migration learning, then performs dimensionality reduction through a deep-stacked auto-encoder model, and finally performs clustering on these low-dimensional vectors. High-dimensional vectors can represent images, and dimensionality reduction avoids the curse of dimensionality during clustering tasks. A particular point in the method is the joint learning and optimization of the dimensionality reduction process and the clustering task. The optimization process is performed using two algorithms: back-propagation and stochastic gradient descent. The experimental findings show that the proposed method, called FIDC, has achieved state-of-the-art performance.

Abstract Fabricating flexible vanadium dioxide (VO2) films is a serious challenge towards commercial applications. In this work, we proposed a new strategy to directly deposit VO2 thermochromic films on flexible substrates by using Cr2O3 structural template layer, which can serve as growth template for low temperature (~300 °C) deposition of VO2 films. The obtained crystalline VO2 films on flexible substrates show significant phase transition properties with narrow hysteresis loops. Optical and electrical characterizations have indicated phase transition features of the flexible VO2 films, with an excellent solar modulation ability (~60% at 2500 nm) and a resistivity change over 2 orders of magnitude. Meanwhile, the flexible VO2 film exhibits narrow hysteresis loops during the phase transition process, and the hysteresis loop widths are less than 1 °C. Flexibility and stability of VO2 film in this work has been demonstrated by a designed bending test, which can maintain stable thermochromic performance after over 5000 bending cycles. This work provided a facile strategy to fabricate VO2 films on flexible substrates as flexible electronic devices.

A multi-channel scan measurement instrument designed for characterizing the infrared obscuring performance of smoke screen at the wavelength of 10.6 µm is developed. It offers 1 to 5 sensors for multi points real time measurements of infrared transmissivity in the field. Data transmission is accomplished by wireless communication. The maximum detection range of the instrument is 500 meters with measurement accuracy better than ± 5%.

Abstract The equivalent electrical resistance of a cell with plane parallel electrodes is described by a simple mathematical model. The limits of this model were determined experimentally. The relation between the overpotential at the edge and at the middle of the electrode is established. The causes of dendrite occurrence at the edge of the electrode in galvanostatic cadmium deposition are discussed.