Time filter

Source Type

Zhang Y.K.,Chongqing Creation Vocational College
Applied Mechanics and Materials | Year: 2013

Lattice Discrete Particle Model (LDPM) provides accurate representation of concrete behavior under cycling loading both in tension and unconfined, confined, and hydrostatic compression. In this paper, LDPM is used to simulate tensile fracturing as demonstrated by the successful simulation of three points bending tests on notched specimens, the result show that it is able to reproduce pre-peak nonlinearity as well as softening post-peak behavior. It reproduces realistically the development of crack pattern due to tensile stresses and also could predict correctly size effect on specimen load carrying capacity. © (2013) Trans Tech Publications, Switzerland.

Chen W.,Chongqing Creation Vocational College | Wang G.,Ankang University
ICSESS 2012 - Proceedings of 2012 IEEE 3rd International Conference on Software Engineering and Service Science | Year: 2012

Citation semantic link network is combination of semantic link network and citation network, which make citation network has the abilities of semantic representation and semantic reasoning. According to semantic reasoning rules, we can get the implicit citation semantic link in network and the new semantic relations between citations. It provides a more efficient method for the citation network retrieval and the complex network theory. Based on citation semantic link network scheme, this paper proposes a method to calculate the similarity of citation semantic link network. This method is the key to discover community in citation semantic link network. Using this similarity calculation method, we can discover semantic community of various research fields in citation network, and foresee developing directions and hot spots of various research fields. © 2012 IEEE.

Yin S.,Southwest University | Chen W.,Chongqing Creation Vocational College | Wang W.,Southwest University
ICIC Express Letters | Year: 2011

In this paper, a typical network teaching system is taken as run-time environment. Data mining techniques and fuzzy theory are introduced into evaluation of network teaching process. Mining the records of web log of students' learning journey, combined it with data warehouse, fuzzy theory analysis, level analysis, and through reasoning of the expert system map daily performance to evaluation indices, and then through fuzzy comprehensive judgment to obtain evaluation results. It uses expert system to interpret evaluation results and generate reviews. Network teaching evaluation depends on its implementation. Evaluation factors are qualitative and quantitative indicators. Evaluation methods include evaluation of process and assessment of learning. Formative evaluation is paid more attention for its teaching and learning promoting function. Quality assurance is not just a result, it should be a dynamic process of intelligence assessment. It works like the human cognitive process. Though the experiment and comparison with traditional max-min synthesis evaluation method, we verify the feasibility of the method. ICIC International © 2011 ISSN 1881-803X.

Qiao X.A.,Chongqing Creation Vocational College | Liu J.,Chongqing Creation Vocational College
Advanced Materials Research | Year: 2014

The pattern recognition process control diagram, this paper puts forward a new method of training neural network. It only needs a small training data set can complete this work. This method is also compatible with the training algorithm, and get a better network performance. Pattern recognition success rate is very high in the larger parameter range, but also has some comparability. © (2014) Trans Tech Publications, Switzerland.

Zheng X.-J.,Chongqing Creation Vocational College | Jiao J.,Chongqing Creation Vocational College
Applied Mechanics and Materials | Year: 2014

The channel whose impulse response contains a large number of zero taps is sparse. Exploiting the sparsity of the channel, the estimation problem can be converted into an on-off keying (OOK) detection problem. Therefore, a sparse channel estimation algorithm was proposed. First, with an initial estimate, posed a detection problem by casting the determination of non-zero taps into an OOK communication problem; then took precise evaluation against the established non-zero taps position. In the simulation, the proposed algorithm was compared to the standard LSE via simulations in terms of MSE and BER. The results show that, the proposed algorithm can provide the best performance while maintaining moderate complexity. © (2014) Trans Tech Publications, Switzerland.

Discover hidden collaborations