Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643825
Liang Tang, Mingzhong Yang, Xiaomin Wang
The firing process of roller kiln consists of several sub-processes and there exists unknown complex nonlinear mapping between the sub-process set points and the final firing quality. To meet this demand, a training algorithm for the radial basis function (RBF) network using GAP method based on the “significance” of a specified neuron is proposed in the paper. The training algorithm which uses GAP method to train the network has a number of advantages such as could be constructed and updated based on the new data sequentially collected from the real process in order to optimize the set point of each sub-process dynamically. Simulation results shows that this training system can work accurately and reliably.
{"title":"Modelling and optimization of the firing process for roller kiln using GAP-RBF neutral network","authors":"Liang Tang, Mingzhong Yang, Xiaomin Wang","doi":"10.1109/CINC.2010.5643825","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643825","url":null,"abstract":"The firing process of roller kiln consists of several sub-processes and there exists unknown complex nonlinear mapping between the sub-process set points and the final firing quality. To meet this demand, a training algorithm for the radial basis function (RBF) network using GAP method based on the “significance” of a specified neuron is proposed in the paper. The training algorithm which uses GAP method to train the network has a number of advantages such as could be constructed and updated based on the new data sequentially collected from the real process in order to optimize the set point of each sub-process dynamically. Simulation results shows that this training system can work accurately and reliably.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"451 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133270043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643890
Yuzhen Liu, Shoufu Li
In order to improve the ability of neighborhood search of differential evolutionary (DE) algorithm, we propose a new variant of DE with linear neighborhood search, called LiNDE, for global optimization problems (GOPs). LiNDE employs a linear combination of triple vectors taken randomly from evolutionary population. The main characteristics of LiNDE are less parameters and powerful neighborhood search ability. Experimental studies are carried out on a benchmark set, and the results show that LiNDE significantly improved the performance of DE.
{"title":"Differential Evolution with Neighborhood Search","authors":"Yuzhen Liu, Shoufu Li","doi":"10.1109/CINC.2010.5643890","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643890","url":null,"abstract":"In order to improve the ability of neighborhood search of differential evolutionary (DE) algorithm, we propose a new variant of DE with linear neighborhood search, called LiNDE, for global optimization problems (GOPs). LiNDE employs a linear combination of triple vectors taken randomly from evolutionary population. The main characteristics of LiNDE are less parameters and powerful neighborhood search ability. Experimental studies are carried out on a benchmark set, and the results show that LiNDE significantly improved the performance of DE.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133893219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643717
Wu Yong-hai, Liu Xiao-mei, Wang Feng
In allusion to a certain kind of bracket of work cabin overhead support equipment components of the special vehicle, taking OptiStruct software as optimization design platform, the topology optimization of the bracket is carried out by using variable density method. The geometry model of the bracket is rebuilt and the finite element analysis of the bracket is carried out, then the stress and deformation of before and after optimization are compared. The results show that the topology optimization method is an effective optimization method to carry out topology optimization of mechanical loading structure with FEA method. The model and approach adopted in this paper provide an important design reference to the mechanical structure and components lightweighting design.
{"title":"Topology optimization design of the bracket of a special vehicle based on FEA method","authors":"Wu Yong-hai, Liu Xiao-mei, Wang Feng","doi":"10.1109/CINC.2010.5643717","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643717","url":null,"abstract":"In allusion to a certain kind of bracket of work cabin overhead support equipment components of the special vehicle, taking OptiStruct software as optimization design platform, the topology optimization of the bracket is carried out by using variable density method. The geometry model of the bracket is rebuilt and the finite element analysis of the bracket is carried out, then the stress and deformation of before and after optimization are compared. The results show that the topology optimization method is an effective optimization method to carry out topology optimization of mechanical loading structure with FEA method. The model and approach adopted in this paper provide an important design reference to the mechanical structure and components lightweighting design.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127899270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643817
Long-qin Xu, Shuang-yin Liu
In terms of some problems existing in the process of large case base retrieval, combining tabu search method and the advantages of artificial fishschool algorithm, this paper proposes multilevel search strategy based on tabu artificial fishswarm algorithm. Tabu artificial fishswarm algorithm applies tabu table with a memory function to artificial fishswarm algorithm and uses different computing model in the similarity calculation according to properties of different types, effectively to avoid premature and blind search and other issues. Simulation results show that the algorithm outperforms other algorithms, it not only improves the retrieval accuracy and retrieval efficiency of the casebased reasoning system, but also is characterized by requiring not much with the initial values and parameters, diversity search and overcoming the local maximum, better coordinate the overall and local search capabilities and provides an effective retrieval method to retrieve the case of large case base.
{"title":"Case retrieval strategies of tabu-based artificial fish swarm algorithm","authors":"Long-qin Xu, Shuang-yin Liu","doi":"10.1109/CINC.2010.5643817","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643817","url":null,"abstract":"In terms of some problems existing in the process of large case base retrieval, combining tabu search method and the advantages of artificial fishschool algorithm, this paper proposes multilevel search strategy based on tabu artificial fishswarm algorithm. Tabu artificial fishswarm algorithm applies tabu table with a memory function to artificial fishswarm algorithm and uses different computing model in the similarity calculation according to properties of different types, effectively to avoid premature and blind search and other issues. Simulation results show that the algorithm outperforms other algorithms, it not only improves the retrieval accuracy and retrieval efficiency of the casebased reasoning system, but also is characterized by requiring not much with the initial values and parameters, diversity search and overcoming the local maximum, better coordinate the overall and local search capabilities and provides an effective retrieval method to retrieve the case of large case base.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121401132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643886
Jingyong Tanga, Li Dong
In this paper, we propose a new super-memory gradient method for unconstrained optimization problems. The global convergence and linear convergence rate are proved under some mild conditions. The method uses the current and previous iterative information to generate a new search direction and uses Wolfe line search to define the step-size at each iteration. It has a possibly simple structure and avoids the computation and storage of some matrices, which is suitable to solve large scale optimization problems. Numerical experiments show that the new algorithm is effective in practical computation in many situations.
{"title":"A new super-memory gradient method for unconstrained optimization","authors":"Jingyong Tanga, Li Dong","doi":"10.1109/CINC.2010.5643886","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643886","url":null,"abstract":"In this paper, we propose a new super-memory gradient method for unconstrained optimization problems. The global convergence and linear convergence rate are proved under some mild conditions. The method uses the current and previous iterative information to generate a new search direction and uses Wolfe line search to define the step-size at each iteration. It has a possibly simple structure and avoids the computation and storage of some matrices, which is suitable to solve large scale optimization problems. Numerical experiments show that the new algorithm is effective in practical computation in many situations.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114652334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643814
Jinxian Lin, Yiqing Lv
Finding out the longest common subsequences (LCS) is one of the most important problems of Bioengineering. But there is still not excellent algorithm for computing all of the longest common subsequences (ALCS). An existing fast algorithm for LCS was studied and improved in this paper. We combined the algorithm with MPI cluster and computed all the longest common subsequences of two random generate gene sequences on MPI cluster. The experiment results show that computing ALCS on MPI cluster is an efficient and practicality method.
{"title":"Computing all longest common subsequences on MPI cluster","authors":"Jinxian Lin, Yiqing Lv","doi":"10.1109/CINC.2010.5643814","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643814","url":null,"abstract":"Finding out the longest common subsequences (LCS) is one of the most important problems of Bioengineering. But there is still not excellent algorithm for computing all of the longest common subsequences (ALCS). An existing fast algorithm for LCS was studied and improved in this paper. We combined the algorithm with MPI cluster and computed all the longest common subsequences of two random generate gene sequences on MPI cluster. The experiment results show that computing ALCS on MPI cluster is an efficient and practicality method.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127551087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643718
Liu XiFa
Low carbon economy has been becoming the mainstream in the development of world economy. The paper elaborates the close relationship between low carbon economy and advancing the innovative management, and sets forth about increasing efficiency and profit to achieve sustainable development through the innovation in thinking, technology, system and integration which would help reduce consumption, emission and pollution.
{"title":"Low carbon economy and innovative management","authors":"Liu XiFa","doi":"10.1109/CINC.2010.5643718","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643718","url":null,"abstract":"Low carbon economy has been becoming the mainstream in the development of world economy. The paper elaborates the close relationship between low carbon economy and advancing the innovative management, and sets forth about increasing efficiency and profit to achieve sustainable development through the innovation in thinking, technology, system and integration which would help reduce consumption, emission and pollution.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126714554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643857
Huang Dongyuan, Chen Xiaoyun
Choosing optimal hyperparameters for Support Vector Machines(SVMs) is quite difficult but extremely essential in SVM design. This is usually done by minimizing estimates of generalization error such as the k-fold cross-validation error or the upper bound of leave-one-out(LOO) error. However, most of the approaches concentrate on the dual optimization problem of SVM. In this paper, we would like to consider the task of tuning hyperparameters in the primal. We derive a smooth validation function from the k-fold cross-validation, then tune hyperparameters by minimizing the smooth validation function using Quasi- Newton optimization technique. Experimental results not only show that our approach is much faster and provides more precise results than grid search method, but also demonstrate that tuning hyperparameters in the primal would be more efficient than in the dual due to advantages provided by the primal.
{"title":"Tuning SVM hyperparameters in the primal","authors":"Huang Dongyuan, Chen Xiaoyun","doi":"10.1109/CINC.2010.5643857","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643857","url":null,"abstract":"Choosing optimal hyperparameters for Support Vector Machines(SVMs) is quite difficult but extremely essential in SVM design. This is usually done by minimizing estimates of generalization error such as the k-fold cross-validation error or the upper bound of leave-one-out(LOO) error. However, most of the approaches concentrate on the dual optimization problem of SVM. In this paper, we would like to consider the task of tuning hyperparameters in the primal. We derive a smooth validation function from the k-fold cross-validation, then tune hyperparameters by minimizing the smooth validation function using Quasi- Newton optimization technique. Experimental results not only show that our approach is much faster and provides more precise results than grid search method, but also demonstrate that tuning hyperparameters in the primal would be more efficient than in the dual due to advantages provided by the primal.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127505326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643781
Li Yan-qing, Huang Zhi-ying, Ge Jing-feng, Jia Qiang, Chen Jing, Fan Yu-zhong
Ecological environment security had aroused more and more attentions, and became the focus of human society. The thesis took western mountain areas in Shijiazhuang District for study area, established ecological security prediction indexes under the model of Pressure-State-Response(PSR), and predicted standardized value of each index by using artificial neural network model, then predicted the development trend of ecological security condition in the next 9 years through ecological security calculation and trend curve. Results were as follows: the overall ecological security degree would present rising trend, it would be “better” state in 2007∼2015, and the maximum value would be 0.6940 in 2015, increased by 74.96% than that of 1996 and 21.58% than that of 2006. However, the ecological security degrees would be less than 0.8 each year, showing that the regional ecological environment problems were not completely solved, should take further measures to protect the ecological environment.
{"title":"Research on ecological security prediction in mountain areas based on the artificial neural network —Taken western mountain areas in Shijiazhuang District for example","authors":"Li Yan-qing, Huang Zhi-ying, Ge Jing-feng, Jia Qiang, Chen Jing, Fan Yu-zhong","doi":"10.1109/CINC.2010.5643781","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643781","url":null,"abstract":"Ecological environment security had aroused more and more attentions, and became the focus of human society. The thesis took western mountain areas in Shijiazhuang District for study area, established ecological security prediction indexes under the model of Pressure-State-Response(PSR), and predicted standardized value of each index by using artificial neural network model, then predicted the development trend of ecological security condition in the next 9 years through ecological security calculation and trend curve. Results were as follows: the overall ecological security degree would present rising trend, it would be “better” state in 2007∼2015, and the maximum value would be 0.6940 in 2015, increased by 74.96% than that of 1996 and 21.58% than that of 2006. However, the ecological security degrees would be less than 0.8 each year, showing that the regional ecological environment problems were not completely solved, should take further measures to protect the ecological environment.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126431611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-11-22DOI: 10.1109/CINC.2010.5643889
Lieya Yan, L. Xu
This paper made a further research based on the extended unitary diagonalization of skew self-conjugate matrix. Based on the definition of skew selfconjugate matrix, we discussed some properties of skew self-conjugate matrix, and gave the necessary and sufficient condition for determining whether ; is a right eigenvalue of matrix in the quaternion field. By means of the Schur theorem in real quaternion field, we proved that skew self-conjugate can be extended unitary diagonalized. Furthermore, if A and B are invertible and commutative, then A + B and A−1 + B−1 can be extended unitary diagonalized at the same time.
本文在斜自共轭矩阵的扩展酉对角化的基础上作了进一步的研究。从斜自共轭矩阵的定义出发,讨论了斜自共轭矩阵的一些性质,给出了判定斜自共轭矩阵是否成立的充分必要条件;是四元数域中矩阵的右特征值。利用实四元数域上的Schur定理,证明了斜自共轭可以扩展酉对角化。更进一步,如果A和B可逆且可交换,则A + B和A - 1 + B - 1可以同时扩展酉对角化。
{"title":"Study of diagonalization on skew self-conjugate matrix in quaternion field","authors":"Lieya Yan, L. Xu","doi":"10.1109/CINC.2010.5643889","DOIUrl":"https://doi.org/10.1109/CINC.2010.5643889","url":null,"abstract":"This paper made a further research based on the extended unitary diagonalization of skew self-conjugate matrix. Based on the definition of skew selfconjugate matrix, we discussed some properties of skew self-conjugate matrix, and gave the necessary and sufficient condition for determining whether ; is a right eigenvalue of matrix in the quaternion field. By means of the Schur theorem in real quaternion field, we proved that skew self-conjugate can be extended unitary diagonalized. Furthermore, if A and B are invertible and commutative, then A + B and A−1 + B−1 can be extended unitary diagonalized at the same time.","PeriodicalId":227004,"journal":{"name":"2010 Second International Conference on Computational Intelligence and Natural Computing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123692517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}