An integral analytic process from quantification to propagation based on limited uncertain parameters is investigated to deal with practical engineering problems. A new method by use of the smallest interval-set/hyper...An integral analytic process from quantification to propagation based on limited uncertain parameters is investigated to deal with practical engineering problems. A new method by use of the smallest interval-set/hyper-rectangle containing all exper- imental data is proposed to quantify the parameter uncertainties. With the smallest parameter interval-set, the uncertainty sponse and the least favorable response propagation evaluation of the most favorable re- of the structures is studied based on the interval analysis. The relationship between the proposed interval analysis method （IAM） and the classical IAM is discussed. Two numerical examples are presented to demonstrate the feasibility and validity of the proposed method.展开更多
An artificial immunity based multimodal evolution algorithm is developed to generate detectors with variable coverage for multidimensional intrusion detection. In this algorithm, a proper fitness function is used to d...An artificial immunity based multimodal evolution algorithm is developed to generate detectors with variable coverage for multidimensional intrusion detection. In this algorithm, a proper fitness function is used to drive the detectors to fill in those detection holes close to self set or among self spheres, and genetic algorithm is adopted to reduce the negative effects that different distribution of self imposes on the detector generating process. The validity of the algorithm is tested with spherical and rectangular detectors, respectively, and experiments performed on two real data sets （machine learning database and DAPRA99） indicate that the proposed algorithm can obtain good results on spherical detectors, and that its performances in detection rate, false alarm rate, stabih＇ty, time cost, and adaptability to incomplete training set on spherical detectors are all better than on rectangular ones.展开更多
基金Project supported by the National Natural Science Foundation of China (No. 11002013)the 111 Project (No. B07009)the Defense Industrial Technology Development Program of China(Nos. A2120110001 and B2120110011)
文摘An integral analytic process from quantification to propagation based on limited uncertain parameters is investigated to deal with practical engineering problems. A new method by use of the smallest interval-set/hyper-rectangle containing all exper- imental data is proposed to quantify the parameter uncertainties. With the smallest parameter interval-set, the uncertainty sponse and the least favorable response propagation evaluation of the most favorable re- of the structures is studied based on the interval analysis. The relationship between the proposed interval analysis method （IAM） and the classical IAM is discussed. Two numerical examples are presented to demonstrate the feasibility and validity of the proposed method.
文摘An artificial immunity based multimodal evolution algorithm is developed to generate detectors with variable coverage for multidimensional intrusion detection. In this algorithm, a proper fitness function is used to drive the detectors to fill in those detection holes close to self set or among self spheres, and genetic algorithm is adopted to reduce the negative effects that different distribution of self imposes on the detector generating process. The validity of the algorithm is tested with spherical and rectangular detectors, respectively, and experiments performed on two real data sets （machine learning database and DAPRA99） indicate that the proposed algorithm can obtain good results on spherical detectors, and that its performances in detection rate, false alarm rate, stabih＇ty, time cost, and adaptability to incomplete training set on spherical detectors are all better than on rectangular ones.
