Search results

The model is based on a healthy balance equations purely theoretical and empirical data.
Simplified equations are derived through a synthesis of the measured data obtained from meteorological stations, as well as the literature.
The estimation method was successfully validated by comparison with real data from natural gas consumption for heating in winter in Turkey.
The weather data used for the analysis of energy determine the accuracy and the characteristics of the results.
Data were obtained from daily average temperatures measured during 9 years.

Using the core data, flakes, imaging logging, geochemicals analysis and lithologies-lithofacies characteristics of Yaoyingtai area in Yingcheng Formation, Changling fault depression, we research on the volcanic rock diagenesis, reservoir space types, characteristics and relationships.
Due to the matrix volume reduction form some intergranular small pore in the process of devitrification(Fig.1_G).

These experimental data are compared with results given by a finite elements simulation.
T., 1993 "Cure Cycle Optimisation for the Reduction of Processing-Induced Residual Stresses in Composite Materials" Journal of Composite Materials, 27(14), 1352-1378

The annual energy consumption data were compared to the heating and cooling load result of EnergyPlus, to verify simulation accuracy.
Therefore, thicker insulation material could decrease annual heating load, and gap of the reduction decreased at thicker insulation material.

The FER2013 data set was used to train the Convolutional Neural Network (CNN) to identify emotions.
The paper concluded that CNN outperforms SVM when large data was involved. 3.Principal Component Analysis (PCA) An unsupervised statistical method is used for dimensionality reduction.
This makes the process of data exploration and visualisation easier by simplifying the complexity of high-dimensional data without interrupting its trends and patterns.
It is an effective method for modeling sequential data since it does not rely on pre clustered data per frame.
The amount of storage required to handle data increases with its size.

The numerical results were compared with the experimental data, and then the results were discussed.
Comparing the angle curves, the numerical results of the swaying angle are close to the experimental data when the wave height is 0.1m.
While in the crest values of the swaying angle, the distinction exists between the numerical results and the experimental data when the wave height is 0.07m.
So it’s expressed in the result as the reduction of the crest value, especially to the crest value of angle when the pendulum flap swings back with the acting of reflected waves.

Application of Elman Neural Network with HP Filter in the Trend Supply of Self-provided Power Plant Forecasting in the Iron and Steel Industry LI Hong-juana, WANG Jian-junb, WANG Huac, MENG Huad Engineering Research Center of Metallurgical Energy Conservation & Emission Reduction, Ministry of Education, Kunming University of Science and Technology, Kunming 650093,China aemail: fxzwlihongjuan@163.com, bemail: jennings_email@yahoo.com.cn, cemail: isentc@163.com demail: 284051088@qq.com Keywords: HP filter, Elman neural network (ENN), Trend sequence Abstract.
The prediction results using practical production data show that using the proposed HP-Elman method that sample A 48, 60 points trend forecast average relative error are 0.37%, 0.47% and sample B 48, 60 points trend forecast average relative error are 0.82%, 1.03%，which can effectively for the trend forecast of self-provided power plant gas supply with a reliable prediction capacity.
Therefore, we build up the model according to self-provided power plant boiler gas supply actual data on some iron and steel enterprise, the forecast results show that the accuracy and time all can meet the actual requirements which combination the HP filter and Elman neural network, and have a good application prospect. 1 Model construct Due to the byproduct gas frequent change which lead to the self-provided power plant supply predicting is not accurate, prediction results implement is not strong, making gas plan feasibility is not high, so most of the iron and steel enterprise are based on experience estimate and made plans.
Based on this situation, through a large number of production data, model creating as follows, first of all, used the HP filter to get the trend of self-provided power plant gas supply, and then to reconstruct the long-term trend sequence.

Signal model Assuming the steering vector of desired signal is known, the STAP method, which is based on minimum mean square error, can be obtained by solving the optimization problem as (1) where is the weight vector of STAP，is the steering vector of desired signal, can be written as (2) wheredenotes Kronecker product，the time steering vector，，(is the wavelength, is Doppler frequency, and is speed), spatial steering vector ,(is the spacing interval between array elements, is the azimuth of target), means transpose operation, is covariance matrix of received data, which can be obtained from the received data (3) where , N and K are the number of array elements and coherent processing pulses respectively.
Generally speaking, is usually replaced by its sample variance matrix: (4) The weight vectors of optimal STAP can be obtained : (5) When the received data are processed by the proceeding weights, the ground clutter can be suppressed and the desired signal can pass without distortion.
If the data covariance matrix can be acquired accurately, the weight vector in (5) is not sensitive to the errors.
However, in practical applications， is estimated by the data, the weights are sensitive to the vector error.
Dimension reduction of space-time adaptive processing algorithms and its applications of airborne radar and mobile communication [D].

This is explained by the reduction of tensions’ intensity on the length unit and by the increase of the probability of various material flaws, in case of big dimensions, which both favour the deformation process
Active elements with different shapes of the stamped profile The experimental stand (Fig. 3) includes: a deformation machine (mechanical press PAI 16); a punch with a set of changeable elements; a force transducer U9B (complete bridge resistive transducer); a displacement transducer TIC 35.100 (semi-bridge inductive transducer); an electric tonometer N2321 associated with a multimeter M890C; a data acquisition board NI – USB 6008; a Sony Vaio laptop.
The data acquisition board has been configured for two analogical channels (to acquire and read data Fig. 6 and Fig. 7).
The data is acquired at a frequency of 1000Hz and 4000 recordings, thus resulting in a 4-seconds acquisition time.
Chart of the data acquisition program Shearing tests were carried out with pairs of different active elements, but whose profile lengths were the same (60 mm).

In this research, the fatigue strength of butt weld joints was investigated and evaluated based on the data obtained by use of the ultrasonic fatigue testing technique. 1.
The ultrasonic fatigue technique provides one practical means of generating ultra-high cycle fatigue data[21-22].
Method of Processing Data 2.1 Finite Element Analysis Used for Hot Spot Stress Finite element program (ANSYS 13.0) was used to analyze welded joints.
Results and Discussion The obtained S-N data of welded joint are shown in Fig.4.
The S-N data of base material are also presented in this figure for comparison.