欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对多变量过程均值异常，提出了选择性有向无环图支持向量机（DAG-SVM）集成，以之为模式识别工具对过程状态进行识别，以探测异常和判别异常源。集成结合Bagging方法的重复采样技术和对DAG-SVM结构的调整，对数据和模型进行双重扰动，以获得差异的候选个体；再通过二进制粒子群优化（BPSO）算法得到最优集成方案。离线仿真测试证明所提选择性DAG-SVM集成具有分类正确率和效率的双重优势；在线仿真测试表明基于选择性DAG-SVM集成的模型探测过程均值阶跃异常优于X^2图和BPN（误差反传神经网络）模型，且判别异常源比BPN模型更准确。针对实际齿轮加工过程数据的应用验证进一步证实了模型的有效性和实用性。
英文摘要：
Aiming at mean shifts in multivariate processes, a selective DAG--SVM ensemble was proposed, which was used as pattern recognition tool to construct an on--line detecting and diagno- sing model for the shifts. To produce diverse candidate components of the ensemble, a double dis- turbance mechanism of data and model was implemented, i.e. , throughthe Bootstrap sampling ap- proach from Bagging method to create different training data sets, and to randomly adjust the struc- ture of DAG--SVM. Finally, the BPSO（binary particle swarm optimization） algorithm was adopted to obtain the optimal ensemble. The simulative off--line test verify that the selective DAG--SVM en- semble have better performance on both classification rate and efficiency. The simulative on--line test shows that the model based on the selective DAG--SVM ensemble outperform z chart and BPN（back propagation neural network） model on detecting mean shifts, and identify the sources of shifts more accurately than the BPN model. An application examination on actual gear machining process data further verified the validity and practicability of this model.
同期刊论文项目
同项目期刊论文
Carbon Emissions Reducing Mechanisms on Integrated Process Parameter Optimization and Scheduling Pro
Hybrid discrete particle swarm optimization for multi-objective flexible job-shop scheduling problem
Modeling and impact factors analyzing of energy consumption in CNC face milling using GRASP gene exp
Advanced scatter search approach and its application in a sequencing problem of mixed-model assembly
Multi-objective artificial bee colony algorithm for simultaneous sequencing and balancing of mixed m
Simulation-based optimisation of logistics distribution system for an assembly line with path constr
An assemble-to-order production planning with the integration of order scheduling and mixed-model se
Realistic wrinkle generation for 3D face modeling based on automatically extracted curves and improv
Optimal Scheduling of Multiple-Line Collaborative Manufacturing Based on Teaching-Learning-Based Opt
An effective hybrid teaching¨Clearning-based optimization algorithm for permutation flow shop schedu
Computer-aided 3D human modeling for portrait-based product development using point- and curve-based
Point Cloud Data Defect Supplement Methods of Products with Complicated Shapes Based on Multiple Two
Multi-objective teaching¨Clearning-based optimization algorithm for reducing carbon emissions and op
Pareto based artificial bee colony algorithm for multi objective single model assembly line balancin
Two-sided assembly line balancing with operator number and task constraints: a hybrid imperialist co
A Method of Interactive Surface Free-Form Deformation Based on Control Point and Neighborhood Constr
A late acceptance hill-climbing algorithm for balancing two-sided assembly lines with multiple const
An effective hybrid honey bee mating optimization algorithm for balancing mixed-model two-sided asse
Simultaneous order scheduling and mixed-model sequencing in assemble-to-order production environment
期刊信息
位置：