Search

Wind speed and streamflow series always are nonlinear and unstable because the effects of chaotic weather systems. These inherent features make them difficult to forecast, especially in a changing environment. To improve forecasting accuracy, an innovation uncertainty forecasting architecture is developed by coupling data decomposition method, feature selection, multiple artificial intelligence (AI) techniques and composite strategy to do unstable time series forecasting. In the designed architecture, the AVMD (adaptive variational mode decomposition) is first applied to excavate implicit information from the original time series. Then, the random forest is utilized to select the suitable inputs for each mode. After that, the GPR (Gaussian Process Regression), a very famous probabilistic AI technique, is driven by various neural networks (ELM (Extreme Learning Machine), BP (Back Propagation Neural Networks), GRNN(Generalized Regression Neural Networks) and RBF (Radial Basis Function Neural Networks)) to produce both deterministic and probabilistic forecasting results in a nonlinear manner to play strengths of each other. The effectiveness and applicability of the proposed approach is verified by unstable wind speed data and streamflow data, and also compared with eleven related models. Results indicate that the proposed model not only improves the forecasting accuracy for deterministic predictions, but also provides more probabilistic information for decision making. The proposed method achieves significantly better performance than the traditional forecasting models both on wind speed forecasting and streamflow forecasting with at least 50% average performance promotion over all the eleven competitors. Comprehensive comparisons demonstrate the superior performance of the proposed method than the involved models as a powerful tool for unstable series forecasting.

High-ThroughPut Phenotyping (HTPP) is currently an active research topic in crop breeding. However, field-based HTPP still suffers from certain bottlenecks, such as small operating space and strong light variation of application scenarios. To address this problem, this paper proposes a solution for in-field HTPP. First, we develop an ultra-narrow phenotyping robot platform that can travel in-row and under canopy. Then, we deploy the Convolutional Neural Network (CNN) on our robot system to detect the maize stalks. Finally, we present an approach to calculate the stalk diameters based on RGB-D camera data. Here, CNN is used to detect maize stalks in RGB images, while the depth images are used to calculate the widths of the stalk bounding boxes, which are considered as the stalk diameters. The field experiment results show that the maximum deviation of stalk diameters measured by our approach is 0.007 m, RMSE is 0.003. To sum up, our robot system can be successfully applied in field HTPP.

The traditional image to enlarge algorithms include nearest neighbor interpolation, bilinear interpolation and high-order interpolation. In order to achieve super-resolution reconstruction of images, a new algorithm combining traditional algorithms and deep learning is proposed. The framework is divided into two parts. Firstly, the deep reconstruction of the low-resolution data is performed by the ability of deep learning to extract features automatically. Then, combining with the traditional interpolation reconstruction results, the deep learning algorithm is used again for training and learning, and finally the high-resolution reconstructed data is obtained. The algorithm is validated using an online public test data set. The results show that the algorithm has a significant effect on the MSE (mean squared error) and PSNR (Peak Signal to Noise Ratio). Compared with the traditional interpolation algorithm and the single deep learning algorithm, the proposed algorithm has higher performance. Moreover, the proposed algorithm is perfect for the reconstruction of the details, the outline is clear, and the high-quality image is obtained.

As an advanced ferroelectric material, barium titanate (BaTiO 3 ) crystal has an electro-optic effect higher than lithium niobate (LiNbO 3 ) crystal by 20-40 times, so it has been taken to be one of the best materials to realize the ultra-high bandwidth electro-optic modulators. In this paper, we model and simulate the transmission coefficient S 21 of the microwave signal to the frequency of drive microwave through an embedded waveguide-electrode structure, and further simulate the S 21 with a software tool - Sonnet. As a result, when the electrode length, electrode width, electrode thickness, electrode gap, BaTiO 3 thin film thickness, magnesium oxide (MgO) substrate thickness and embedded depth were 1mm, 80μm, 1μm, 8μm, 450nm, 500μm and 0.1μm, respectively, the electro-optic modulation bandwidths have reached 50-70GHz, and it can be optimized to 100GHz at 250μm MgO substrate thickness.

The encryption is independent from encoding, limiting the efficiency of information transmission and the miniaturization of baseband chips. It is necessary to find out the conjunction points between encryption and encoding. This paper designed a design a data transmission scheme-AESP algorithm by combining the AES(Advanced Encryption Standards) with polar codes, which has both encryption and error correction capabilities. This scheme integrated the encoding into the encryption process and proposed a new algorithm of key expansion. And the initial and final rounds are also changed to improve security. A 128-bit plaintext and an image are encrypted and decrypted, which achieves a good effect. AESP is 4% faster than the traditional scheme in serial state. Finally, AESP is performed in parallel on Hadoop platform. Results showed that AESP can reduce operation time by about 71% compared with serial computing.

Design and simulation of vertical pitch attitude and height control system of the ground effect vehicle based on active disturbance rejection control: As for the uncertainty of aerodynamic parameters such as fixed perturbation of the ground effect vehicle, atmospheric disturbance and wave noise, the second-order auto-disturbance rejection controller has been used to control the pitch attitude of the ground effect vehicle and the first-order active disturbance rejection controller has been used to control the altitude to achieve height stability control. The controller designed in this thesis has been verified by simulation in MATLAB, the simulation result showed that the proposed pitch angle ADRC and height ADRC are rational, anti-jamming and robust.

Objective To explore the effect of effective questioning on the teaching of “Toxicology” based on case-based teaching, and to accumulate experience for further promoting the reform of Toxicology. Methods We selected 317 undergraduates of preventive medicine in Grade 2012, 2013 and 2014, from Shaanxi University of Chinese Medicine as research objects. Two teaching cases were planned in the instructional design for Grade 2012, while six teaching cases were planned for Grade 2013. The instructional design for Grade 2014 implemented effective questioning on the basis of Grade 2013. The difference in test paper scores, grade point average (GPA), the difficulty level of the test paper and grade were analyzed before and after the increase of case-based teaching and effective questioning. Results The test paper scores and grade point averages increased gradually from Grades 2012, 2013 and 2014, and the difference was statistically significant (F=99.722, P

With the formation of the Four Vertical and Four Horizontal high speed railway (HSR) network in China, HSR has significantly improved Chinese regional accessibility. This paper evaluates the effect of HSRs on the economic growth in China using the data from 2007 to 2016 and uses the Gradient Boosting Regression Tree to carry out an empirical test. The paper argues that when analyzing the economic impact of new transportation means, both the construction of itself and improvement of transportation conditions should be taken into account. Therefore, the HSR was studied by operation mileages and its proportion to total railway mileage. The main conclusion from the paper is that Chinese HSR has a positive effect on economic growth. The difference of the purpose and background of HSR construction between China and other countries may be the reasons for the inconsistency of the research conclusions. Thus, Chinese government needs to optimize the HSR networks and improve transport services.

Collaborative learning technology has been widely applied in various disciplines in support of dialectical discourse and argumentation. Technologies not only allow multiple users to concurrently analyze and edit documents in the same virtual space but also present richer cues for collaboration awareness. However, researchers still know very little about how the synching mechanisms work together to support small group collaborative argument creation. This study investigates the experiences of 14 dyads that worked together in Critical Thinker, a web application that enables learners to collaborate on an argumentative task with a juxtaposed argumentation structure and multiple synchronous collaboration affordances. Our qualitative data showed that the learners are able to leverage different synching mechanisms in a complementary manner and appropriate the dialectic structure for collaborative argumentation.

We present an aptameric graphene field-effect transistor (GFET) nanosensor for sensitive and label-free detection of biomarkers in serum. The graphene nanosensor is serially functionalized with a polyethylene glycol (PEG) nanolayer and an aptamer for specific detection of a target analyte and effective rejection of background molecules in human serum. The analyte binds to the surface-immobilized aptamer, changing the carrier concentration in the graphene. The resulting changes in the conductance of the graphene is measured to determine the analyte concentration. Experimental results show that the device is capable of detecting immunoglobulin E (IgE) in serum in a clinically relevant range of 50 pM to 35 nM.

A soft sensor model based on generalized regression neural network (GRNN) is presented to be aimed at solving the problem that the key biological parameters in the microbial fermentation process are difficult to be real-time online measurement. GRNN contains a special linear output layer, which makes that its network structure possesses the adaptive certainty and output has noting to do with the initial weights, etc. The marine protease fermentation process is taken as an example, firstly through the analysis of the mechanism of marine protease fermentation process, the auxiliary and leading variables of soft sensor model are determined, then according to the GRNN algorithm steps, a GRNN soft sensor model is established for biological parameters of the marine protease fermentation, and is compared by simulation with the RBF neural network soft sensor model. The results show that, in comparison with RBF neural network, the GRNN soft sensor model has the faster convergence speed, higher precision and stronger generalization ability.

In order to solve the difficulties of online measurement in the marine protease fermentation process of crucial biological variables (such as biomass concentration, substrate concentration and enzyme activity, etc), a soft sensing method based on KPCA-RBF neural network is proposed by combining the kernel principal component analysis (KPCA ) with the radial basis function (RBF) neural network. Establishing the soft sensing model of KPCA-RBF neural network, KPCA is applied to compress data, and choose the nonlinear component as the input of RBF neural network and biomass concentration, substrate concentration, relative enzyme activity as the output. Simulation results indicate that this model has a higher accuracy, better tracking performance when compared with RBF and PCA-RBF neural network model. Therefore, the proposed method can satisfy the requirements of on-line measurement of biological parameters and is proved to be an efficient modeling method.

In this paper, the consistent fish tracking strategies have been present for the underwater surveillance system with multiple static cameras in the overlapping field of views (FOV). The discriminative appearance model has been first introduced to distinguish the swimming fish and the background by the superpixel. The centroid coordinate homographic mapping and the Speeded Up Robust Features (SURF) technique have been taken to capture and match the same fish from the perspectives among multiple cameras by a set of strategies. Simulation experiments have shown that the proposed method is less dependent on the exact motion detection of the fish and could perform robustly and stably at a high speed in case of the occlusion and appearance variation under the sea.

Based on the discussion of RFID technology and the Zigbee wireless sensor network technology, the author presents a method to attach electronic labels to the sensor, solving the integration problem of Zigbee wireless sensor network and RFID technology, implementing the application of RFID sensor network technology for monitoring the quality of agricultural products. Integrating RFID technology and the Zigbee wireless sensor network technology with low-cost electronic labels is worth to be applied to the development of the Internet of Things.

The purpose of this paper is to compare the biofidelity of LSTC and TRL legform impactor. A finite element model of simplified front vehicle structure was established. The simulation tests were conducted by using this simplified car model against two different legform impactors and a FE lower extremity, respectively at a speed of 40km/h. The simulation results from the two legform impactors were compare with that of the FE lower extremity. The compared outputs include tibia acceleration, shear displacement, knee bending angle. The test results show that the tibia acceleration of the impactors and the FE lower extremity are approximately consistent. As to the shear displacement and the knee bending angle, though the output of the impactors are not agreed so well with that of the FE lower extremity, the trend are comparable to some extent. The results of the FE lower extremity are higher than that of the legform impactors. The results of LSTC impactor has a better agreement with that from FE lower extremity.

This paper proposes a new geometric distortion correcting algorithm for wavelet-based watermarked image. The watermark is embedded into the low frequency sub-bands of the host image, which guarantees large embedding capacity and good invisibility. The modified Harris-Laplace detector is utilized to extract steady feature points used for image correcting. Simulation results show that the proposed algorithm is not only invisible but also robust against many image geometric attacks such as rotation, translation and scaling, etc.

Objective: Investigate the effects of TSA on MCF-7 cell growth in vitro and the expressions of fragile HiT (Fhit) and Survivin gene in human breast cancer cell strain MCF-7. Methods: Cultured cells were divided into control group and TSA groups, and treated with DMSO and different concentration of TSA (50, 100, 200, 500nmol/L) respectively. Then the cell activity was observed by try pan blue stain assay and the transcription level of Fhit and Survivin were analyzed by using real-time PCR. The change of Fhit protein was analyzed by western bolt. Results: The proliferation of breast cancer cell MCF-7 was inhibited significantly after culture with TSA. TSA inhibited the breast cancer cell growth in dose-dependent and time-dependent manners. The expression of Fhit mRNA and protein were all increased significantly (3.12 vs 1, P

Data mining for data stream becomes important in academic areas. Due to large-scale data, people utilize incremental learning approach to handle the data. In this paper, a modified Support Vector Machine (SVM) incremental learning model is proposed. Through experiments of selecting kernel function for the SVM method, we optimize several parameters. Real network dataset is used in our experiments to verify the model's feasibility and applicability. The experimental results show that the modified SVM incremental learning model can improve the accuracy of classification and increase performance.

This paper presents a method of using support vector machine (SVM) with Second Generation Wavelet kernel (2G Wavelet kernel) for classification of individual transmitters working in frequency-hopping spread spectrum (FH-SS) modulation. Simulation results show that this method can significantly improve the efficiency of the identification in comparison with those based on First Generation Wavelet kernel (1G Wavelet kernel). Meanwhile, it can also achieve a satisfactory classification rate. Furthermore, the relationship between the classification rate and SNR is also discussed.

tetherin cDNA was amplified from total RNA prepared from HeLa cells by RT-PCR, then cloned into the eukaryotic expression vector pEGFP-C3. The generated pEGFP-Tetherin construct was transfected into COS-1 to determine the expression and the subcellular localization of GFP-Tetherin fusion protein. Under CLSM the localization of GFP-Tetherin fusion protein distributed outside the nucleus and co-localized with TGN in COS-1 cells.

A low-voltage low-power bandgap voltage reference in 90nm CMOS is designed and simulated. The overall bandgap architecture is optimized to achieve high accuracy temperature and power supply independent voltage reference. It consists of the bandgap core circuit, op-amp, start-up circuit and output stage. The bandgap reference circuit provides reference voltage of 584.7mV ± 0.8mV with 1.2V ± 10% power supply and over −40°C to 125°C temperature ranges simultaneously. The total layout area including dummy structures is 100μm × 85μm.

This paper proposes a novel method of using wavelet kernel functions in Support Vector Machines (SVMs), and this method is applied to identification of individual communication transmitter which works in frequency-hopping spread spectrum modulation. The adoption of kernel function can improve the classification rate. The experimental results show how the recognition rates change with the parameters of wavelet kernel function. In a certain specific range, the classification rates maintain at a high level.

Accounting standards is a kind of incomplete contract and its residual control right belongs to enterprises' stakeholders. Stakeholders participating in accounting policy choice (SPAPC) is a method of distribution in residual control right of accounting standards. The strategy choice among stakeholders is a process of repeated game. Stakeholders choose strategy on the basis of the present value of their benefit in the game and they may adjust the strategy according to others' choice. The strategy change speed is related to rate of stakeholders choosing the strategy initially. The state of Nash equilibrium means that all stakeholders can choose the accounting policy which can reflect enterprises' income status factually. In the process of SPAPC, stakeholders constrain with each other, which can avoid inefficiency of accounting policy choice and earnings manipulation of management. It's helpful to solve the matching problem between residual control right and residual claim right and decrease institution change cost of accounting standards.

The inventory management of traditional Chinese medicines is an important part of GMP and GSP. For pharmaceutical trading enterprises (sales), after the drug delivery before selling, the drugs are mostly stored in the warehouse in order to ensure drug safety, effectiveness and quality control, which are the basic duties of pharmaceutical trading enterprises. How to ensure drug safety, effectiveness and quality control? How to make the drug management process successfully? How to reduce the cost? This paper attempts to explore these problems from the static and dynamic inventory strategy.

The processing of data from the database using data mining algorithms need more special methods. In fact, some redundancy and irrelevant attributes reduce the performance of data mining, so the problem of feature subset selection becomes important in data mining domain. This paper presentes a new algorithm which is called discrete binary differential evolution (BDE) algorithm to select the best feature subsets. The relativity of attributes is evaluated based on the idea of mutual information. Experiments using the new feature selection method as a preprocessing step for SVM, C&R Tree and RBF network are done. We find that the method is very effective to improve the correct classification rate on some datasets and the BDE algorithm is useful for feature subset selection.

Phytoremediation is a promising tool in removing pollutants from the environment or in rendering them harmless. Some previous reports showed that Fe deficiency in dicots induced the stimulated accumulation of a number of micronutrient metals other than Fe. The objective of this research was to determine the effect of Fe deficiency on the uptake of cadmium (Cd) by tomato (Lycopersicon. Esculentum Mill.) in hydroponic culture. Results showed that Fe deficiency induced cadmium uptake, biomass decrease and changes of pH and Eh in hydroponic culture. Under Fe deficiency status, the decrease in pH and the increase in Eh might result in higher cadmium availability. Values of bioconcentration factor (BCF) and translocation factor (TF) indicated that Fe-deficiency status affected cadmium accumulation and translocation in tomato. The results of this research would significantly improve the understanding of metal removal from the contaminated environment.

The work deal with the stabilization control of double inverted pendulum system. Inverted pendulum system is a complicated, nonlinear, unstable system of high order. Fuzzy control research for stabilization a double inverted pendulum at an upright position successfully is proposed based on weight variable fuzzy inputs. The weight variable fuzzy inputs is gained by combining the fuzzy control theory with the optimal control theory. The fuzzy control rules of a double inverted pendulum are given. In order to consider cart, the lower pendulum, the upper pendulum error information, three different fuzzy controller were designed in this paper. Simulation results show that the controller, which the upper pendulum is considered as main control variable, has higher accuracy and quicker convergence speed and higher precision, and simulation result is promising. The control result can be expanded the control of multilevel inverted pendulum, and have a guiding meaning in the control of other unstable system.

Particle swarm optimization (PSO) has been shown to be an efficient, robust and simple optimization algorithm. Aim at the shortcoming that the PSO algorithm falls into local optimization easily, in this paper fuzzy control theory is introduced into PSO (FPSO). Parameters may be dynamic adjusted themselves according to the optimization effect every time in this algorithm. Its ability of dynamic adjustment is strengthened, and the global optimization performance of the algorithm can be improved better. And in this paper, the improved algorithm is illustrated how could solve the problem, which exists in the raw material requirement model for production processing in the ore dressing plant. The experimental results are provided to support the conclusions drawn from the theoretical findings.