Advanced Materials Research Vols. 121-122

Abstract: In this article, we proposed a negative selection algorithm which based on hierarchical level cluster of self dataset CB-RNSA. First the self data set is clustered by different cluster radius, and then the self data are substituted by cluster centers to compare with candidate detectors to reduce the number of distance counting. In the detector creating process, the value of each detector property was restricted to a given value range so as to decrease the redundancy of detectors. The stimulation result shows that CB-RNSA is an effective algorithm for the creation of artificial immune detectors.

Abstract: The paper introduces the current research status of image-matching question and proposes a new improved genetic algorithm. It adopts a meme-gene co-evolution technique for the improvement of engineering performance. The simulation results show the improved GA has the higher stability and the better graph than the SGA.

Abstract: This paper proposed a feasible system for language identification (LID) and designed four different GMM-training approaches to improve the system performance by accuracy recognition rates. In our experiment, we used these model-training approaches to evaluation on the system performance, which utilizes Linear Prediction Cepstrum Coefficients (LPCC) and Gaussian Mixture Model (GMM), rely on a 10-language task. From all the results, we found an optimal approach for training GMM in LID system, which achieves high accuracy of 85.25%, and indicated that different GMM-training approaches have different performances for LID system, but an advisable training method that proposed in our paper can greatly improve the system performance.

Abstract: Pycnoporus cinnabarinus is a plant pathogen. It is common in many areas and is widely distributed throughout the world. Laccases of are some of the few oxidoreductases commercialized as industrial catalysts. In the present study, some characters of the amino acid sequence of P.cinnabarinus laccase (Lac1) were predicted and analyzed with the tools of bioinformatics. These results showed that the protein was composed of 20 kinds of amino acid; the theoretical pI of manganese peroxidase was 4.81 and the theoretical molecular weight of manganese peroxidase was 56292.0 Da; total number of atoms was 7806; the extinction coefficient was 58120 (280 nm). The N-terminal of the sequence considered was M (Met) and the estimated half-life was 30 hours (mammalian reticulocytes, in vitro). The instability index (II) was computed to be 34.50; this classifies the protein as stable. Aliphatic index was 82.64. Grand average of hydropathicity (GRAVY) was -0.063. There were 8 glycosylation sites, a signal peptide and conserved domains.

Abstract: The GUESS (Greenhouse Use of Energy Seedling Simulator) model is a lumped-parameter coupled dynamic simulation combining a carbon-based process model of seedling growth with a heat/mass transfer model of the greenhouse envelope. GUESS was created to provide nursery managers and engineers a tool to assess the impact of climate control decisions upon energy use and seedling growth rate. The source code for GUESS was written using Simulink and MATLAB. A test case was set up to validate the model. And experimented with different lighting targets to see which one offered the most growth per unit energy cost. The model was parameterized for Douglas fir production in Corvallis, OR Seedlings were started at 0.57g d.w and were harvested at 1.7g d.w Temperature regulated to 68.5±6.3°F Growth rate could be highly sensitive to carbon content conversion as well.Models can be very useful as simulation tools, but their utility depends highly upon the data used to parameterize them.

Abstract: Along with the deepening reform in power system, Electric Power Enterprise Marketing Management becomes an important management tool to ensure the effective operation of power enterprises. In this paper, on the basis of relevant principles, we have developed a market economy to adapt to the current situation of the development, marketing, and service sectors of the Power Marketing Strategy. At the same time, we designed to power enterprise marketing performance evaluation index system. And we introduced the quantitative analysis of power marketing effectiveness: AHP, according to the power of quantitative analysis method to evaluate and analyze marketing results.

Abstract: Bioregenerative life support system (BLSS) is one of the key technologies that must be solved before founding permanent base in space. The configuration and closure of BLSS largely depends on the astronauts’ diet. Thus, it’s necessary to set up a dietary management system for astronauts. A dietary management software system for BLSS is designed in this work. The nutritional requirements are used as constraints of the optimization process. Constrained nonlinear programming algorithm is used to accomplish the optimization process. The optimization results achieved by this software are compared with the standard nutritional needs. It will provide assistant decision support for BLSS designers.

Abstract: With the rapid development of information technology, ontology is gradually used for knowledge engineering and information science sphere. Agricultural Ontology mainly is composed of the concept and relationship between concepts in agricultural knowledge, as well as the computer can identify the composition of the formal description language. The goal of build the agricultural ontology is to form a common understanding of the agricultural information organizational structure, to known and analysis of knowledge in the field of agriculture so as to further establish the foundation for the agriculture semantic web. In this paper, we present a design principles and implementation process in vegetable products domain ontology. Through the collection and analysis of the field of vegetable products information as well as the use of inference engine, used to build vegetable domain ontology with OWL. In the information extraction prototype system, we achieved a semi-automatic filled with examples of vegetable products domain ontology. Through compared to the results before and after the use of inference engine, it show that vegetables and semi-automatic domain ontology building methods is feasible.

Abstract: The research actuality of Intrusion Detection System(IDS) were analyzed, Due to the defects of IDS such as high positive rate of IDS and incapable of effective detection of dispersed coordinated attacks on the time and space, the ideas of the multi-source information fusion were introduced in the paper, a multi-level IDS reasoning framework and prototype system were presented. The prototype adds analysis engine to the existing IDS Sensor, We used Bayesian Network as a tool for multi-source information fusion, and we used goal-tree to analyze the attempts of coordinated attacks and quantify the security risk of system. Compared to the existing IDS, the prototype is more integrated and more capable in finding coordinated attacks with lower false positive rate.

Abstract: This paper presents a series of methods based on data mining theory for information fusion system in order to avoid the ubiquitous problems of comprehensive evaluation in the process of information fusion. The methodology is developed by analyzing the significance factor that has impact on fusion performance based on gradual regression method and by analyzing the performance testing results of fusion by group using hierarchy clustering method. Through the analysis of performance testing results for distributed information fusion system, the factors that have effects on fusion performance and the correlation between various influencing factors have been discovered. Moreover, the performance of information fusion system is enhanced by adjusting the disadvantageous factors and improving the design methods.