充电截止电压是大多数电动汽车用户充电都会经历的电压点。针对传统安时积分法忽略初始容量误差和电池老化等一系列待优化的问题,提出了双层集成极限学习机(extreme learning machine,ELM)算法,实现锂离子电池充电截止电压下的荷电状态(...充电截止电压是大多数电动汽车用户充电都会经历的电压点。针对传统安时积分法忽略初始容量误差和电池老化等一系列待优化的问题,提出了双层集成极限学习机(extreme learning machine,ELM)算法,实现锂离子电池充电截止电压下的荷电状态(state of charge,SOC)和健康状态(state of health,SOH)联合估计。首先,提取易测的电池健康特征(health indicator,HI),采用集成极限学习机映射HI及充电所需时间与SOH之间的关系。其次,用测得的HI估计难以在线测量的充电所需时间,对充电截止电压下安时积分法的SOC进行在线修正。该方法充分考虑了电动汽车用户初始充电状态的不确定性,指导电动汽车用户合理充电。此外,通过选择合适的集成ELM模型集成度,解决了单个ELM模型输出不稳定的问题。最后,选用NASA和CALCE数据集进行实验验证。验证结果表明,锂离子电池充电截止电压下SOC的估计均方根误差均小于1.5%,集成ELM相比于其他常见算法具有较高的训练、测试精度和较短的预测时间。展开更多
To obtain an accurate and robust soft sensor model in dealing with the increasingly complex industrial modeling data,an effective heterogeneous ensemble of extreme learning machines(HEELM)is proposed.Specifically,the ...To obtain an accurate and robust soft sensor model in dealing with the increasingly complex industrial modeling data,an effective heterogeneous ensemble of extreme learning machines(HEELM)is proposed.Specifically,the kernel extreme learning machine(KELM)and four common extreme learning machine(ELM)models that have different internal activations are contained in the HEELM for enriching the diversity of sub-models.The number of hidden layer nodes of the extreme learning machine is determined by the trial and error method,and the optimal parameters of the kernel extreme learning machine model are determined by cross validation.Moreover,to obtain the best output of the ensemble model,least squares regression is applied to aggregate the outputs of all individual models.Two complex data sets of practical industrial processes are used to test the HEELM performance.The simulation results show that the HEELM has a high prediction accuracy.Compared with the individual ELM models,bagging ELM ensemble model,BP and SVM models,the prediction accuracy of the HEELM model is improved by 4.5%to 8.7%,and the HEELM model can obtain better generalization capability.展开更多
Supposing that the overall situation is dug out from the distributed monitoring nodes, there should be two critical obstacles, heterogenous schema and instance, to integrating heterogeneous data from different monitor...Supposing that the overall situation is dug out from the distributed monitoring nodes, there should be two critical obstacles, heterogenous schema and instance, to integrating heterogeneous data from different monitoring sensors. To tackle the challenge of heterogenous schema, an instance-based approach for schema mapping, named instance-based machine-learning （IML） approach was described. And to solve the problem of heterogenous instance, a novel approach, called statistic-based clustering （SBC） approach, which utilized clustering and statistics technologies to match large scale sources holistically, was also proposed. These two algorithms utilized the machine-leaning and clustering technology to improve the accuracy. Experimental analysis shows that the IML approach is more precise than SBC approach, reaching at least precision of 81% and recall rate of 82%. Simulation studies further show that SBC can tackle large scale sources holisticalty with 85% recall rate when there are 38 data sources.展开更多
The purpose of this study was to establish factors that lead to poor integration of Information and communication technology （ICT） for teaching and learning in schools in Kenya, despite comprehensive policy, institu...The purpose of this study was to establish factors that lead to poor integration of Information and communication technology （ICT） for teaching and learning in schools in Kenya, despite comprehensive policy, institutional, infrastructural frameworks and capacity building by the Ministry of Education. The subject of this study was administered by use of questionnaires in three categories of public schools： national school, provincial schools and district schools. The respondents were students from each level that is from one, two, three and four and teachers based on the most offered subjects in the secondary schools. The computer assisted learning facilities were classified into computers, internet and content in optical media. In national school Internet based research, optical media content provided by Kenya Institute of Curriculum Development and Cyber School program for science subjects was used in learning. In provincial school, it lacks adequate computers, reliable Internet and content in optical media. In district school, it lacks adequate computer, no internet connection and content in optical media. A learner management system which can be accessed by all learners by use of any internet access devices like mobile phone access will be an ideal tool with over 4,000,000 mobile phone subscribers currently in Kenya.展开更多
In this paper, we study the problem of employ ensemble learning for computer forensic. We propose a Lazy Local Learning based bagging （L3B） approach, where base learners are trained from a small instance subset surr...In this paper, we study the problem of employ ensemble learning for computer forensic. We propose a Lazy Local Learning based bagging （L3B） approach, where base learners are trained from a small instance subset surrounding each test instance. More specifically, given a test instance x, L3B first discovers x＇s k nearest neighbours, and then applies progressive sampling to the selected neighbours to train a set of base classifiers, by using a given very weak （VW） learner. At the last stage, x is labeled as the most frequently voted class of all base classifiers. Finally, we apply the proposed L3B to computer forensic.展开更多
文摘充电截止电压是大多数电动汽车用户充电都会经历的电压点。针对传统安时积分法忽略初始容量误差和电池老化等一系列待优化的问题,提出了双层集成极限学习机(extreme learning machine,ELM)算法,实现锂离子电池充电截止电压下的荷电状态(state of charge,SOC)和健康状态(state of health,SOH)联合估计。首先,提取易测的电池健康特征(health indicator,HI),采用集成极限学习机映射HI及充电所需时间与SOH之间的关系。其次,用测得的HI估计难以在线测量的充电所需时间,对充电截止电压下安时积分法的SOC进行在线修正。该方法充分考虑了电动汽车用户初始充电状态的不确定性,指导电动汽车用户合理充电。此外,通过选择合适的集成ELM模型集成度,解决了单个ELM模型输出不稳定的问题。最后,选用NASA和CALCE数据集进行实验验证。验证结果表明,锂离子电池充电截止电压下SOC的估计均方根误差均小于1.5%,集成ELM相比于其他常见算法具有较高的训练、测试精度和较短的预测时间。
基金The National Natural Science Foundation of China(No.71471060)the Natural Science Foundation of Hebei Province(No.E2018502111)Fundamental Research Funds for the Central Universities(No.2019QN134).
文摘To obtain an accurate and robust soft sensor model in dealing with the increasingly complex industrial modeling data,an effective heterogeneous ensemble of extreme learning machines(HEELM)is proposed.Specifically,the kernel extreme learning machine(KELM)and four common extreme learning machine(ELM)models that have different internal activations are contained in the HEELM for enriching the diversity of sub-models.The number of hidden layer nodes of the extreme learning machine is determined by the trial and error method,and the optimal parameters of the kernel extreme learning machine model are determined by cross validation.Moreover,to obtain the best output of the ensemble model,least squares regression is applied to aggregate the outputs of all individual models.Two complex data sets of practical industrial processes are used to test the HEELM performance.The simulation results show that the HEELM has a high prediction accuracy.Compared with the individual ELM models,bagging ELM ensemble model,BP and SVM models,the prediction accuracy of the HEELM model is improved by 4.5%to 8.7%,and the HEELM model can obtain better generalization capability.
基金Projects(2007AA01Z126, 2007AA01Z474) supported by the National High-Tech Research and Development Program of ChinaProject(NCET-06-0928) supported by the Program for New Century Excellent Talents in University
文摘Supposing that the overall situation is dug out from the distributed monitoring nodes, there should be two critical obstacles, heterogenous schema and instance, to integrating heterogeneous data from different monitoring sensors. To tackle the challenge of heterogenous schema, an instance-based approach for schema mapping, named instance-based machine-learning （IML） approach was described. And to solve the problem of heterogenous instance, a novel approach, called statistic-based clustering （SBC） approach, which utilized clustering and statistics technologies to match large scale sources holistically, was also proposed. These two algorithms utilized the machine-leaning and clustering technology to improve the accuracy. Experimental analysis shows that the IML approach is more precise than SBC approach, reaching at least precision of 81% and recall rate of 82%. Simulation studies further show that SBC can tackle large scale sources holisticalty with 85% recall rate when there are 38 data sources.
文摘The purpose of this study was to establish factors that lead to poor integration of Information and communication technology （ICT） for teaching and learning in schools in Kenya, despite comprehensive policy, institutional, infrastructural frameworks and capacity building by the Ministry of Education. The subject of this study was administered by use of questionnaires in three categories of public schools： national school, provincial schools and district schools. The respondents were students from each level that is from one, two, three and four and teachers based on the most offered subjects in the secondary schools. The computer assisted learning facilities were classified into computers, internet and content in optical media. In national school Internet based research, optical media content provided by Kenya Institute of Curriculum Development and Cyber School program for science subjects was used in learning. In provincial school, it lacks adequate computers, reliable Internet and content in optical media. In district school, it lacks adequate computer, no internet connection and content in optical media. A learner management system which can be accessed by all learners by use of any internet access devices like mobile phone access will be an ideal tool with over 4,000,000 mobile phone subscribers currently in Kenya.
基金the National High Technology Research and Development Program(863) of China(No.2007AA01Z456)the National Natural Science Foundation of China(No.60703030)
文摘In this paper, we study the problem of employ ensemble learning for computer forensic. We propose a Lazy Local Learning based bagging （L3B） approach, where base learners are trained from a small instance subset surrounding each test instance. More specifically, given a test instance x, L3B first discovers x＇s k nearest neighbours, and then applies progressive sampling to the selected neighbours to train a set of base classifiers, by using a given very weak （VW） learner. At the last stage, x is labeled as the most frequently voted class of all base classifiers. Finally, we apply the proposed L3B to computer forensic.
