Search

Circular RNAs are closed endogenous RNAs that are involved in the progression of diverse tumors. Even with the most advanced combined treatments, patients with glioblastoma multiforme have a median survival time of15 months. This study aimed to investigate the roles of circular PLOD2 (circPLOD2) in glioma tumorigenesis and tumor development and to clarify its tumor-promoting effects by bioinformatics analysis and molecular experiments.To determine the characteristics of circPLOD2 expression, quantitative real-time polymerase chain reaction was conducted. Stable knockdown of circPLOD2 was implemented for functional assays. Cell Counting Kit-8 and colony formation assays were used to measure cell proliferation. Transwell assays and tube formation assays were used to evaluate cell invasion and angiogenesis abilities, respectively. An intracranial xenograft model was established to determine the function of circPLOD2 in vivo. Further biochemical and Western blot analyses were conducted to evaluate proteins associated with circPLOD2.circPLOD2 was upregulated in glioma tissues and cells. High expression of circPLOD2 was significantly associated with tumor size, World Health Organization grade, and molecular characteristics of glioma. circPLOD2 deregulation affected glioblastoma multiforme cell proliferation, invasion, and angiogenesis. Knockdown of circPLOD2 inhibited tumorigenesis in vivo. Further biochemical analysis showed that circPLOD2 was involved in oncogenic pathways and correlated with the expression of proteins related to proliferation, invasion, and angiogenesis.Our data indicate that circPLOD2 promotes glioma tumorigenesis and tumor development in vitro and in vivo and that suppressing circPLOD2 could be a novel therapeutic strategy for glioma.

The stochastic cloud cover on photovoltaic (PV) panels affects the solar power outputs, producing high instability in the integrated power systems. It is an effective approach to track cloud motion during short-term PV power forecasting based on data sources of satellite images. However, since temporal variations of these images are noisy and non-stationary, pixel-sensitive prediction methods are critically needed in order to seek a balance between forecast precision and huge computation burden due to a large image size. Hence, a graphical learning framework is proposed for intra-hour PV power prediction. By simulating the cloud motion using bi-directional extrapolation, a directed graph is generated representing the pixel values from multiple frames of historical images. The nodes and edges in the graph denote the shapes and motion directions of the regions of interest (ROIs) in satellite images. A spatial-temporal graph neural network (GNN) is then proposed to deal with the graph. Comparing with conventional deep-learning-based models, GNN is more flexible for varying sizes of input, in order to be able to handle dynamic ROIs. Referring to the comparative studies, the proposed method greatly reduces the redundancy of image inputs without sacrificing the visual scope, and slightly improves the prediction accuracy.

Due to the stochastic and non-stationary characteristics of wind speed, the wind power generation is highly uncertain and fluctuating, which significantly challenges the operation of the power system and the associated electricity market. In this article, a new spatial-temporal method is proposed for short-term wind power prediction based on image inputs and augmented convolutional network. First, the geographical locations of various wind farms and the relevant wind vectors are processed into a series of multiframe spatial-temporal wind images, which can be handled by the convolutional networks. Then, wind power conversion and prediction models are developed based on those networks, where recurrent paths and attention mechanism are introduced to enhance the model architecture. The testing results have validated the high performance of the proposed method within a forecast horizon of up to seven hours. In particular, even when the terrain information is not available, the implicit wind flow field within the original inputs can still be approximately learned by the proposed convolutional networks.

A prior knowledge of residential load demand is critical for power system operations at the distribution level, such as economic dispatch, demand response and energy storage schedule. However, as residential customers perform more casual and active consumption behaviors, prediction of such highly volatile loads can be much harder. Owing to the development of sensor technology, micrometeorological data can be sampled with a high geographic resolution. Those data that represent the weather condition on the land surface show a strong relationship to the residential load evidently, whereas it remains unsolved on how to fully utilize those great number of datasets. This paper proposes a day-ahead probabilistic residential load forecasting method based on a novel deep learning model, named convolutional neural network with squeeze-and-excitation modules (CNN-SE), and micrometeorological data. The model can employ multi-channel input data with dissimilar weights, suitable for analyzing massive relevant input factors. A feature extraction method is adopted for customer consumption pattern based on sparse auto-encoder (SAE), which can help correct probabilistic forecasting results. A case study that covers 8 residential communities and 18 micrometeorological sites is conducted to validate the feasibility and accuracy of the proposed hybrid method.

Solar energy is a strongly intermittent renewable energy source, which is affected by varied meteorological conditions, and thus produces arbitrary power outputs in photovoltaic (PV) power generation. Complex weather variations make it challenging to develop an efficient PV power forecasting method. In this study, a graph modeling method is proposed for short-term PV power prediction. Unlike many conventional machine-learning models, the proposed model is capable of evaluating interconnections among various meteorological input factors. This study details the design and operation of graph modeling, including graph construction, node feature construction, message transfer, and readout. An entire model is established consisting of spectral graph convolution, multiple graphical edges and a hierarchical output manner. The testing results suggest that the proposed multi-graph model outperforms other benchmark models in terms of accuracy under day-ahead forecasting cases. Besides, the single-graph model achieves a reduced cost of training time comparing to deep-learning benchmark models. This work was supported by National Natural Science Foundation of China under Grant 52077062.

Lead free Sr3SmNa2Fe0.5Nb9.5O30 ceramic was synthesized and characterized, and high total energy storage density, recoverable energy storage density, energy storage efficiency and good thermal stability are synchronously achieved.

To explore and evaluate the performance of MRI-based brain tumor super-resolution generative adversarial network (MRBT-SR-GAN) for improving the MRI image resolution in brain tumors.A total of 237 patients from December 2018 and April 2020 with T2-fluid attenuated inversion recovery (FLAIR) MR images (one image per patient) were included in the present research to form the super-resolution MR dataset. The MRBT-SR-GAN was modified from the enhanced super-resolution generative adversarial networks (ESRGAN) architecture, which could effectively recover high-resolution MRI images while retaining the quality of the images. The T2-FLAIR images from the brain tumor segmentation (BRATS) dataset were used to evaluate the performance of MRBT-SR-GAN contributed to the BRATS task.The super-resolution T2-FLAIR images yielded a 0.062 dice ratio improvement from 0.724 to 0.786 compared with the original low-resolution T2-FLAIR images, indicating the robustness of MRBT-SR-GAN in providing more substantial supervision for intensity consistency and texture recovery of the MRI images. The MRBT-SR-GAN was also modified and generalized to perform slice interpolation and other tasks.MRBT-SR-GAN exhibited great potential in the early detection and accurate evaluation of the recurrence and prognosis of brain tumors, which could be employed in brain tumor surgery planning and navigation. In addition, this technique renders precise radiotherapy possible. The design paradigm of the MRBT-SR-GAN neural network may be applied for medical image super-resolution in other diseases with different modalities as well.

An increased interest has been observed in the application of soybean protein isolate (SPI) into O/W emulsion because of the amphipathic characteristics of SPI. However, at pH around 4.5, SPI was almost lost its hydrophilic characteristic, thus greatly limiting its application in emulsion under an acidic environment. Therefore, this drawback of SPI needs to be urgently solved. This study aims to investigate the effect of γ-polyglutamic acid (γ-PGA) on physicochemical properties of SPI-stabilized O/W emulsion. The results suggested that the interaction between γ-PGA and SPI improved the SPI solubility in solution, and increased emulsifying properties of SPI in the pH range of 4.0–5.0 via electrostatic interaction. Meanwhile, the charge neutralisation between SPI emulsions with γ-PGA was confirmed via ζ-potentiometry. With the presence of γ-PGA in emulsion at pH 4.0 and 5.0, the electrostatic complexation between SPI and anionic γ-PGA exhibited decreased the viscosity of SPI emulsion, which might be related to the phenomenon as indicated by the confocal laser scanning microscope measurements. Therefore, the electrostatic complexation between SPI and γ-PGA suggested that the promising potential of γ-PGA to be used in SPI-stabilized O/W emulsion under an acidic environment.

Accurate short-term solar irradiance forecasting is crucial for ensuring the optimum utilization of photovoltaic power generation sources. This study addresses this issue by proposing a spatiotemporal correlation model based on deep learning. The proposed model first applies a convolutional neural network (CNN) to extract spatial features from a two-dimensional matrix composed of meteorological parameters associated with a target site and its neighboring sites. Then, a long short-term memory (LSTM) network is applied to extract temporal features from historical solar irradiance time-series data associated with the target site. Finally, the spatiotemporal correlations are merged to predict global horizontal irradiance one hour in advance. The prediction performance and generalization ability of the proposed CNN-LSTM model are evaluated within a whole year, under diverse seasons and sky conditions. Three datasets are involved for case studies, which are collected from 34 locations spread across three different climate zones in Texas, USA. Moreover, the performance of the CNN-LSTM model is compared with those obtained using the CNN, LSTM, and other benchmark models based on five evaluation metrics. The results indicate that the proposed model has advantages over the other models considered and provides a good alternative for short-term solar radiation prediction.

Pseudogenes, which are long noncoding RNAs that originate from protein-coding genes, have been suggested to play important roles in disease. Although studies have revealed high expression of legumain (LGMN) in many types of tumors, the regulation of LGMN remains largely unknown. Here, we found that a novel LGMN pseudogene (LGMNP1) was upregulated in glioblastoma (GBM) tissues and high LGMNP1 expression in GBM cells enhanced proliferation and invasion. Biochemical analysis showed that cytoplasmic LGMNP1 functionally targeted miR-495-3p in a manner involving an RNA-induced silencing complex. Dual-luciferase reporter assays demonstrated that LGMN was a target of miR-495-3p, and LGMN was upregulated and positively correlated with LGMNP1 in GBM. Moreover, miR-495-3p was downregulated and negatively correlated with LGMNP1 in GBM tissues. Notably, the tumor-promoting effects of LGMNP1 upregulation could be alleviated by miR-495-3p mimics. Furthermore, GBM cells overexpressing LGMNP1 exhibited more aggressive tumor progression and elevated LGMN expression in vivo. Thus, our data illustrate that LGMNP1 exerts its oncogenic activity, at least in part, as a competitive endogenous RNA (ceRNA) that elevates LGMN expression by sponging miR-495-3p. CeRNA-mediated miRNA sequestration might be a novel therapeutic strategy in GBM.

In this study, the widely used varieties in China were selected to explore the low starch digestive properties of highland barley from three aspects of composition, starch structural orders, and kernel structure. Two digestive systems were constructed, with highland barley in flour and chyme forms, to investigate the contribution of these inherent properties to starch digestion under different physical forms. Compared with white and black highland barley, blue varieties showed poor chemical composition, lower starch crystallinity, and higher short-range structural order degree, but the starch digestibility as flour was only slightly higher. All varieties of chyme demonstrated lower starch digestibility compared to flour; however, blue highland barley chyme showed much lower starch digestibility than other chyme. Furthermore, kernel structural analysis showed that the blue highland barley kernel structure effectively inhibited digestion. These results indicated that the kernel structure had a critical influence on the starch digestibility of highland barley.

In many cultures, rice is let to cool for a while after cooking before being consumed, which results in some retrogradation of the starch. The in vitro digestibility of 16 cooked rice starches after extended cold (4 °C) storage, which leads to extensive retrogradation, was investigated from the perspective of their starch molecular fine structure. Size-exclusion chromatography (SEC) and fluorophore-assisted carbohydrate electrophoresis (FACE) were used to characterize the starch chain-length distributions (CLDs) and whole molecular size distributions. The retrograded starch gel network was studied by differential scanning calorimetry and scanning electron microscopy. Pearson correlation analysis with in vitro digestibility revealed that the starch digestion rate was positively correlated with cell size and cell wall thickness within the formed gel network, which was mainly derived from the interactions of amylose short-medium chains and not strongly related to amylopectin molecular structure. This suggests that rather than just altering amylose content, modification of amylose fine molecular structure (e.g. increasing the amount of short-medium amylose chains) can also slow rice starch digestibility.

Background Malignant glioma is one of the essentially incurable tumors with chemoresistance and tumor recurrence. As a histone methyltransferase, SUV39H2 can trimethylate H3K9. SUV39H2 is highly expressed in many types of human tumors, while the function of SUV39H2 in the development and progression of glioma has never been elucidated. Methods RT-qPCR and IHC were used to test SUV39H2 levels in glioma tissues and paired normal tissues. The clinical relevance of SUV39H2 in glioma was analyzed in a public database. Colony formation assays, CCK-8 assays, and flow cytometry were conducted to explore the role of SUV39H2 in the growth of glioma cells in vitro. A cell line-derived xenograft model was applied to explore SUV39H2’s role in U251 cell proliferation in vivo. Sphere formation assays, RT-qPCR, flow cytometry, and IF were conducted to illustrate the role of SUV39H2 in the stemness and chemosensitivity of glioma. Luciferase reporter assays and WB were applied to determine the function of SUV39H2 in Hh signaling. Results SUV39H2 was highly expressed in glioma tissues relative to normal tissues. SUV39H2 knockdown inhibited cell proliferation and stemness and promoted the chemosensitivity of glioma cells in vitro. In addition, SUV39H2 knockdown also significantly inhibited glioma cell growth in vivo. Moreover, we further uncovered that SUV39H2 regulated hedgehog signaling by repressing HHIP expression. Conclusions Our findings delineate the role of SUV39H2 in glioma cell growth and chemosensitivity as a pivotal regulator of the hedgehog signaling pathway and may support SUV39H2 as a potential target for diagnosis and therapy in glioma management.

Here, we reveal a new food gel formed on simple pH reduction of a water-soluble crosslinked corn bran arabinoxylan complex. This is different from low pH gelling high-methoxyl pectin that requires high sugar content, and it is similar in gelling property to low acyl gellan gum though is readily soluble in water. Alkaline-solubilized corn bran arabinoxylan (CAX) with two levels of residual bound ferulic acid was treated with laccase, a crosslinking enzyme, to produce two soluble, crosslinked CAX (SCCAX) complexes of different sizes (avg. 3.5 and 4.5-mer). Both of the SCCAXs formed gels at pH 2, with the larger, more heavily feruloylated SCCAX forming the stronger gel. Gels showed shear-thinning behavior and a thermal and pH reversible property. A gel forming mechanism was proposed to occur through noncovalent crosslinking including hydrogen bonds and hydrophobic interaction among the SCCAX complexes. This mechanism was supported by structural characterization of crosslinked CAX complexes using a Zeta-sizer and FT-IR spectroscopy. Applications of SCCAX gels might be where low pH low sugar gels are desired or a beverage containing SCCAX might be taken with gelling occurring in the low pH environment of the stomach, as well as in other food gels and as a drug delivery matrix.

Soaking is an important factor determining the final quality of cooked rice. Although the water content, distribution, and properties of cooked rice have been studied, the structuring role and migration path of water during soaking remain unresolved. In this study, the mobility and combinative state of molecular chains of starch and protein in soaked rice were investigated using proton nuclear magnetic resonance. The proton populations with different T2 reflected molecular mobility and interactions of water, starch, protein, and non-starch polysaccharides, which were affected by water content and hydration. The spaces between granules and cell walls and cracks were explored by scanning electron microscopy as these are the main channels to penetrate. However, the distribution and morphology of these channels showed significant differences at different temperatures, causing different penetrate paths. With the time and temperature increasing, water increasingly penetrated through inhomogeneous cell walls and then along different channels into the inner areas of grains, and the instrumental texture and rheological measurements showed significant structural changes and phase transition in the grains.

The need of daily solar radiation (DSR) has been well recognized over the years because many countries consider solar energy into power grids. The essence of this paper is to apply the deep belief network (DBN) to estimating daily solar radiation on horizontal surfaces for Lhasa in China. Measured weather data (wind speed, sunshine duration, air dry-bulb temperature, air relative humidity) which cover the period of 1994–2015 are obtained and devoted to modeling. The statistical indicators (mean absolute bias error (MABE), root mean square error (RMSE) and correlation coefficient (R)) show that, the DBN model, optimized by testing the effect of multiple parameters, has more promising result when compared to back propagation (BP) network.

Photovoltaic (PV) electric power has been widely employed to satisfy rising energy demands because inexhaustible renewable energy is environmentally friendly. In order to mitigate the impact caused by the uncertainty of solar radiation in grid-connected PV systems, a hybrid method based on a deep convolutional neural network (CNN) is introduced for short-term PV power forecasting. In the proposed method, different frequency components are first decomposed from the historical time series of PV power through variational mode decomposition (VMD). Then, they are constructed into a two-dimensional data form with correlations in both daily and hourly timescales that can be extracted by convolution kernels. Moreover, the time series of residue from VMD is refined into advanced features by a CNN, which could reduce the data size and be easier for further model training along with meteorological elements. The hybrid model has been verified by forecasting the output power of PV arrays with diverse capacities in various hourly timescales, which demonstrates its superiority over commonly used methods.

Amylose is a linear polymer formed by α-1,4 linkages and one of the major components of starch. It is often used to encapsulate hydrophobic molecules of health benefit, such as fatty acids (FA), flavonoids or polyphenols. Amylose and linoleic acid form a complex when the amylose molecule is folded in a V-type conformation. However, it is still not clear how amylose and linoleic acid interact with each other and what is the conformation of the complex adopts under different conditions. Molecular dynamics (MD) simulation is a useful tool used to describe the structure of molecules and complex systems. In this work, MD simulations were used to illustrate conformational changes of the amylose-FA complex and the single amylose molecule, which was used as a reference, in water. During simulations, the conformation of the complex remained stable and the linoleic acid and the amylose molecules formed a helical structure. Furthermore, fatty acid molecules were located in cavity of the amylose helix with the hydrophobic tail inside it. Results also showed that 4C1 was the predominant ring conformation of glucose unit in amylose during interaction with fatty acids. Once the fatty acid was able to form the complex with amylose, the inter turn hydrogen bonds between O6 and neighboring O2/O3 groups were preferentially retained and allowed the complex to keep its helix conformation. Our analysis also shows that the amylose-linoleic acid complex is slightly less stable than the one formed between stearic acid and amylose.

Line loss is inherent in transmission and distribution stages, which can cause certain impacts on the profits of power-supply corporations. Thus, it is an important indicator and a benchmark value of which is needed to evaluate daily line loss rates in low voltage transformer regions. However, the number of regions is usually very large, and the dataset of line loss rates contains massive outliers. It is critical to develop a regression model with both great robustness and efficiency when trained on big data samples. In this case, a novel method based on robust neural network (RNN) is proposed. It is a multi-path network model with denoising auto-encoder (DAE), which takes the advantages of dropout, L2 regularization and Huber loss function. It can achieve several different outputs, which are utilized to compute benchmark values and reasonable intervals. Based on the comparison results, the proposed RNN possesses both superb robustness and accuracy, which outperforms the testing conventional regression models. According to the benchmark analysis, there are about 13% outliers in the collected dataset and about 45% regions that hold outliers within a month. Hence, the quality of line loss rate data should still be further improved.

Day-ahead residential load forecasting is crucial for electricity dispatch and demand response in power systems. Electrical loads are characterized by volatility and uncertainty caused by external factors, especially for individual residential loads. With the deployment of advanced metering infrastructure, the acquisition of electricity consumptions of multiple residential customers is available. This paper proposes a novel day-ahead residential load forecasting method based on feature engineering, pooling, and a hybrid deep learning model. Feature engineering is performed using two-stage preprocessing on data from each user, i.e., decomposition and multi-source input dimension reconstruction. Pooling is then adopted to merge data from both the target user and its interconnected users, in a descending order based on mutual information. Finally, a hybrid model with two input channels is developed by combining long short-term memory (LSTM) with self-attention mechanism (SAM). The case studies are conducted on a practical dataset containing multiple residential users. Performance of the proposed load forecasting method using data pools of different groups of users as well as different input forms is compared. The effectiveness of input dimension reconstruction and hybrid model is also validated. The overall results demonstrate the superiority of the proposed load forecasting method through comparison with other benchmark methods.

The composition and structure of starch are important indicators to evaluate the quality of rice, but the effect of protein on structural properties of rice has always been controversial. In the present study, the rheological properties and thermal properties were combined to investigate the effects of protein on mechanical and inner structure changes of rice kernel during cooking. The morphologic changes in overall form and inner structure of the kernel showed the limited and uneven gelatinization of starch, and reflected the good heating stability of protein. The comparison of molecular mobility in H2O/D2O reflected the special gelatinization behavior of starch and the hydrophobicity of protein during cooking. The similar proton distributions between starch, starch-protein mixture and rice flour indicated the weak interactions between starch and protein. The results of Fourier transform infrared spectra, thermogravimetric analysis and transmission electron microscope confirmed the wrapping effect of starch and surface interactions with protein.

The uncertainties of solar energy due to weather fluctuations will cause impact on photovoltaic (PV) integrated power system, limiting its further development. Hence, a novel probabilistic PV power forecasting approach based on lower upper bound estimation (LUBE) is proposed in this study, which consists of adaptive neuro-fuzzy inference system (ANFIS), modified grey wolf optimization (MGWO) algorithm and kernel density estimation (KDE). In the proposed method, two ANFIS models are combined as a group and optimized using MGWO. The output prediction intervals (PIs) from various groups are also merged together through KDE, which enhance the robustness and reliability of a sole optimized model. Owing to different quality indicators that measure diverse aspects of PIs, the proposed method achieves superior prediction results when compared with conventional forecasting methods.

A 52-year-old male who had chronic hypertension for several years presented with abrupt epistaxis. The CT scan revealed a 40 mm × 40 mm mass in the nasal cavity intended to the maxillary sinus and the base of skull. Nasal endoscope biopsy and serum/urinary catecholamine detection conformed an ectopic noradrenaline-secreting pheochromocytoma. The present research was to discuss the clinical characteristics of the rare pheochromocytoma and the palliative interventional embolization for it.

Owing to its importance in food and pharmaceutical applications, complexation of amylose with amphiphilic molecules has been experimentally and theoretically investigated. Most theoretical studies have involved only a single amphiphilic molecule as the guest. In the present work, Molecular Dynamics (MD) is utilized to study this complexation process when more than one amphiphilic molecule (linoleic acid) is complexing with an amylose fragment. Results showed that the amylose fragment complexed with linoleic acid to form a V-type structure, and more glucose residues participated when the number of linoleic acid molecules increased. Intra-molecular hydrogen bonds are the main interactions that stabilize the formation of the amylose helical structure. The preferred location of the linoleic acids molecules was in the central region of the amylose fragment, which induced a larger fluctuation of the two terminal regions of the complex. Alike when a single linoleic acid molecule is considered, the aliphatic tails of the linoleic acid molecules complexed with the hydrophobic cavity of the amylose fragment and the carboxyl heads were exposed to the polar water molecules.

Additional file 1: Table S1. Sequences of shRNA. Table S2. Sequences of RT-PCR primers. Table S3. Antibodies used in Immunoblotting, Immunohistochemistry and Immunofluorescence experiments.

Antifreeze proteins (AFPs) were extracted from cold-acclimated oat by vacuum infiltration–centrifugation. Ammonium precipitation, ion exchange, and size-exclusion chromatography were used successively to purify the oat AFPs (AsAFPs), and the proteins were identified using matrix-assisted laser desorption/ionization time of flight tandem mass spectrometry. The results showed that an ammonium-sulfate saturation of 50–100 % was optimal for supernatant precipitation. After purification, AsAFP was found to have achieved an electrophoretic purity and its thermal-hysteresis activity was measured at 1.24 °C (15 mg ml−1). The mass fingerprinting and sequencing results indicated the homology of AsAFP and endochitinase. Recrystallization and melting resistance of ice cream were improved by AsAFP (0.1 %). Moreover, the addition of AsAFP (0.1 %) in ice cream increased the glass transition temperature (Tg) from −29.14 to −27.74 °C.

Postharvest decay and senescence results in significant losses in harvested mushrooms. Integrated management, utilizing different eco-friendly, non-chemical methods, represents a strategy for controlling postharvest decay and reducing economic losses in horticultural crops that is being actively developed and commercialized. In the present study, the use of the biocontrol yeast Pichia kluyveri combined with UV-C treatment was evaluated for its ability to control postharvest decay of king oyster mushrooms (Plerrotus eryngii) caused by Lactococcus lactis subsp. lactis. As stand-alone treatments, the use of P. kluyveri or UV-C significantly reduced disease development, relative to untreated mushrooms. The control effects on L. lactis decay and natural infection were significantly improved by the combination of P. kluyveri and UV-C, compared to either treatment alone. Treatment of mushrooms with P. kluyveri or UV-C alone and in combination also induced the activity the antioxidant enzymes of catalase and superoxide dismutase in treated mushrooms, and also enhanced the accumulation of antimicrobial substances, such as total phenolics and flavonoids. These data indicate that the mode of action by which UV-C enhances the biocontrol efficacy of P. kluyveri may be partially attributed to the elicitation of host defense responses in mushrooms. Integrating the use of a biocontrol agent with UV-C treatment represents an effective strategy for controlling postharvest decay in mushrooms.

The outputs of photovoltaic (PV) power are random and uncertain due to the variations of meteorological elements, which may disturb the safety and stability of power system operation. Hence, precise day-ahead PV power forecasting is crucial in renewable energy utilization, as it is beneficial to power generation schedule and short-term dispatch of the PV integrated power grid. In this study, a novel day-ahead PV power forecasting approach based on deep learning is proposed and validated. Firstly, two novel deep convolutional neural networks (CNNs), i.e. residual network (ResNet) and dense convolutional network (DenseNet), are introduced as the core models of forecasting. Secondly, a new data preprocessing is proposed to construct input feature maps for the two novel CNNs, which involves historical PV power series, meteorological elements and numerical weather prediction. Thirdly, a meta learning strategy based on multi-loss-function network is proposed to train the two deep networks, which can ensure a high robustness of the extracted convolutional features. Owing to the learning strategy and unique architectures of the two novel CNNs, they are designed into relatively deep architectures with superb nonlinear representation abilities, which consist of more than ten layers. Both point and probabilistic forecasting results are provided in the case study, demonstrating the accuracy and reliability of the proposed forecasting approach.

Solar energy plays an essential role in environment governance and resource protection, as it is totally pollution-free and extensively accessed. An accurate knowledge of solar radiation is beneficial to the deployments of solar energy constructions, photovoltaic and thermal solar systems. In this study, a deep learning method is proposed for estimating daily global solar radiation, which is constituted by embedding clustering (EC) and functional deep belief network (DBN). Based on the curve shapes of daily solar radiation, EC divides the overall dataset into different subsets, which can be modeled separately. Knowledge from empirical radiation models is also merged as the input of functional DBN. The model can be directly applied to solar estimation in various stations due to its strong nonlinear representation. The case study in China is adopted that involves radiation data from a total of 30 stations to validate the practicability and accuracy of the proposed method. From the results, the method obtains better estimation precision with empirical knowledge, achieving 1.706 MJ/m2 of mean absolute error (MAE), 2.352 MJ/m2 of root mean square error (RMSE) and 13.71% of mean absolute percentage error (MAPE) according to the average values at the 30 stations.

The quality of cooked rice and its influence factors have always been the focus of researches. However, the formative mechanisms of its eating quality and structural changes of rice during cooking have seldom been evaluated. In this study, sectional real-time cooking was performed by differential scanning calorimetry (DSC) and dynamic viscoelasticity analysis to monitor the phase transitions and mechanical changes of kernels, which exhibited different characteristics in different stages. Both glass transition and pasting behavior were captured, and showed more viscoelasticity of cooked rice at higher soaking temperatures. Meanwhile, the enzyme activity of rice during soaking was successfully measured by a rapid viscosity analyzer (RVA). Along with the differences of morphologies and crystalline structure at different soaking conditions, the findings of DSC and rheometer were further verified. This study provides effective methods to evaluate changes in rice during cooking and explains the mechanism of differences formed by the soaking temperature.

Wind energy is a commonly utilized renewable energy source, due to its merits of extensive distribution and rich reserves. However, as wind speed fluctuates violently and uncertainly at all times, wind power integration may affect the security and stability of power system. In this study, we propose an ensemble model for probabilistic wind speed forecasting. It consists of wavelet threshold denoising (WTD), recurrent neural network (RNN) and adaptive neuro fuzzy inference system (ANFIS). Firstly, WTD smooths the wind speed series in order to better capture its variation trend. Secondly, RNNs with different architectures are trained on the denoising datasets, operating as submodels for point wind speed forecasting. Thirdly, ANFIS is innovatively established as the top layer of the entire ensemble model to compute the final point prediction result, in order to take full advantages of a limited number of deeplearningbased submodels. Lastly, variances are obtained from submodels and then prediction intervals of probabilistic forecasting can be calculated, where the variances inventively consist of modeling and forecasting uncertainties. The proposed ensemble model is established and verified on less than one-hour-ahead ultra-short-term wind speed forecasting. We compare it with other soft computing models. The results indicate the feasibility and superiority of the proposed model in both point and probabilistic wind speed forecasting.

The effects of different processing methods were investigated, including on the fluidity, solubility, nutrient composition, digestibility, color and aroma constituent of instant brown rice powder. It showed that, except for flavonoids, different processing methods significantly influenced the indexes of the instant brown rice flour (p

Cereal grains are generally subjected to thermal treatment before consumption. However, few data are available on total phenolic content (TPC), degree of starch gelatinization, phenolic acids, and antioxidant capacity of extruded whole grains based on trolox equivalent antioxidant capacity (TEAC), DPPH radical scavenging activity, and ferric reducing antioxidant potential (FRAP). Four whole grains, brown rice, wheat, maize, and barley were examined in this study. The free and bound TPCchanged significantly (p < 0.05) after extrusion. Morever, heat and degree of starch gelatinization could affect the bound TPC. In bound fraction, the antioxidant capacity was directly influenced by TPC. Extrusion (110-140°C) can retain or improve the antioxidant capacity in its free and bound fractions. Phenolic acids are stable in their bound form. Results showed that the bound TPC and total phenolic acids have a strong relationship against antioxidant capacity, indicating that TPC is the major antioxidant in the bound fraction.