Support vector machines and a Kalman-like observer are used for fault detection and isolation in a variable speed horizontalaxis wind turbine composed of three blades and a full converter. The support vector approach ...Support vector machines and a Kalman-like observer are used for fault detection and isolation in a variable speed horizontalaxis wind turbine composed of three blades and a full converter. The support vector approach is data-based and is therefore robust to process knowledge. It is based on structural risk minimization which enhances generalization even with small training data set and it allows for process nonlinearity by using flexible kernels. In this work, a radial basis function is used as the kernel. Different parts of the process are investigated including actuators and sensors faults. With duplicated sensors, sensor faults in blade pitch positions,generator and rotor speeds can be detected. Faults of type stuck measurements can be detected in 2 sampling periods. The detection time of offset/scaled measurements depends on the severity of the fault and on the process dynamics when the fault occurs. The converter torque actuator fault can be detected within 2 sampling periods. Faults in the actuators of the pitch systems represents a higher difficulty for fault detection which is due to the fact that such faults only affect the transitory state(which is very fast) but not the final stationary state. Therefore, two methods are considered and compared for fault detection and isolation of this fault: support vector machines and a Kalman-like observer. Advantages and disadvantages of each method are discussed. On one hand, support vector machines training of transitory states would require a big amount of data in different situations, but the fault detection and isolation results are robust to variations in the input/operating point. On the other hand, the observer is model-based, and therefore does not require training, and it allows identification of the fault level, which is interesting for fault reconfiguration. But the observability of the system is ensured under specific conditions, related to the dynamics of the inputs and outputs. The whole fault detection and isolation scheme is evaluated using a wind turbine benchmark with a real sequence of wind speed.展开更多
Ⅰ. INTRODUCTION MULTILEVEL inverters are increasingly being used in high-power medium voltage applications due to their superior performance compared to two-level inverters, such as lower common-mode voltage, lower d...Ⅰ. INTRODUCTION MULTILEVEL inverters are increasingly being used in high-power medium voltage applications due to their superior performance compared to two-level inverters, such as lower common-mode voltage, lower dv/dt, lower harmonics in output voltage and current, and reduced voltage on the power switches.展开更多
This paper presents a new system identification approach using vector space base functions, and proposes two network structures based on Gamma sequence and Laguerre sequence. After analyzing and comparing these struct...This paper presents a new system identification approach using vector space base functions, and proposes two network structures based on Gamma sequence and Laguerre sequence. After analyzing and comparing these structures in detail, some simulation results to demonstrate the conclusions are given.展开更多
Complex industry processes often need multiple operation modes to meet the change of production conditions. In the same mode,there are discrete samples belonging to this mode. Therefore,it is important to consider the...Complex industry processes often need multiple operation modes to meet the change of production conditions. In the same mode,there are discrete samples belonging to this mode. Therefore,it is important to consider the samples which are sparse in the mode.To solve this issue,a new approach called density-based support vector data description( DBSVDD) is proposed. In this article,an algorithm using Gaussian mixture model( GMM) with the DBSVDD technique is proposed for process monitoring. The GMM method is used to obtain the center of each mode and determine the number of the modes. Considering the complexity of the data distribution and discrete samples in monitoring process,the DBSVDD is utilized for process monitoring. Finally,the validity and effectiveness of the DBSVDD method are illustrated through the Tennessee Eastman( TE) process.展开更多
In this article, we propose two control charts namely, the “Multivariate Group Runs’ (MV-GR-M)” and the “Multivariate Modified Group Runs’ (MV-MGR-M)” control charts, based on the multivariate normal processes, ...In this article, we propose two control charts namely, the “Multivariate Group Runs’ (MV-GR-M)” and the “Multivariate Modified Group Runs’ (MV-MGR-M)” control charts, based on the multivariate normal processes, for monitoring the process mean vector. Methods to obtain the design parameters and operations of these control charts are discussed. Performances of the proposed charts are compared with some existing control charts. It is verified that, the proposed charts give a significant reduction in the out-of-control “Average Time to Signal” (ATS) in the zero state, as well in the steady state compared to the Hotelling’s T2 and the synthetic T2 control charts.展开更多
This paper presents a corner-based image alignment algorithm based on the procedures of corner-based template matching and geometric parameter estimation. This algorithm consists of two stages: 1) training phase, and ...This paper presents a corner-based image alignment algorithm based on the procedures of corner-based template matching and geometric parameter estimation. This algorithm consists of two stages: 1) training phase, and 2) matching phase. In the training phase, a corner detection algorithm is used to extract the corners. These corners are then used to build the pyramid images. In the matching phase, the corners are obtained using the same corner detection algorithm. The similarity measure is then determined by the differences of gradient vector between the corners obtained in the template image and the inspection image, respectively. A parabolic function is further applied to evaluate the geometric relationship between the template and the inspection images. Results show that the corner-based template matching outperforms the original edge-based template matching in efficiency, and both of them are robust against non-liner light changes. The accuracy and precision of the corner-based image alignment are competitive to that of edge-based image alignment under the same environment. In practice, the proposed algorithm demonstrates its precision, efficiency and robustness in image alignment for real world applications.展开更多
The size and performance of a System LSI depend heavily on the architecture which is chosen. As a result, the architecture design phase is one of the most important steps in the System LSI development process and is c...The size and performance of a System LSI depend heavily on the architecture which is chosen. As a result, the architecture design phase is one of the most important steps in the System LSI development process and is critical to the commercial success of a device. In this paper, we propose a C-based variable length and vector pipeline (VVP) architecture design methodology and apply it to the design of the output probability computation circuit for a speech recognition system. VVP processing accelerated by loop optimization, memory access methods, and application-specific cir- cuit design was implemented to calculate the Hidden Markov Model (HMM) output probability at high speed and its performance is evaluated. It is shown that designers can explore a wide range of design choices and generate complex circuits in a short time by using a C-based pipeline architecture design method.展开更多
Use of multidisciplinary analysis in reliabilitybased design optimization(RBDO) results in the emergence of the important method of reliability-based multidisciplinary design optimization(RBMDO). To enhance the effici...Use of multidisciplinary analysis in reliabilitybased design optimization(RBDO) results in the emergence of the important method of reliability-based multidisciplinary design optimization(RBMDO). To enhance the efficiency and convergence of the overall solution process,a decoupling algorithm for RBMDO is proposed herein.Firstly, to decouple the multidisciplinary analysis using the individual disciplinary feasible(IDF) approach, the RBMDO is converted into a conventional form of RBDO. Secondly,the incremental shifting vector(ISV) strategy is adopted to decouple the nested optimization of RBDO into a sequential iteration process composed of design optimization and reliability analysis, thereby improving the efficiency significantly. Finally, the proposed RBMDO method is applied to the design of two actual electronic products: an aerial camera and a car pad. For these two applications, two RBMDO models are created, each containing several finite element models(FEMs) and relatively strong coupling between the involved disciplines. The computational results demonstrate the effectiveness of the proposed method.展开更多
聚类是大规模高维向量数据分析的关键技术之一.近年来,基于密度的聚类算法DBSCAN(density-based spatial clustering of applications with noise)因其无须预先指定聚类数量、能够发现复杂聚类结构并有效识别噪声点的特性,在数据分析领...聚类是大规模高维向量数据分析的关键技术之一.近年来,基于密度的聚类算法DBSCAN(density-based spatial clustering of applications with noise)因其无须预先指定聚类数量、能够发现复杂聚类结构并有效识别噪声点的特性,在数据分析领域得到了广泛应用.然而,现有的基于密度的聚类算法在处理高维向量数据时将产生极高的时间代价且面临维度灾难等问题,难以在实际场景中部署应用.此外,随着信息技术的发展,高维向量数据规模急剧增加,使用CPU进行高维向量聚类在时间代价和可扩展性等方面将面临更大的挑战.为此,提出一种GPU加速的高维向量聚类算法,通过引入K近邻(K-nearest neighbor,KNN)图索引加速DBSCAN的计算.首先,设计了GPU加速的并行K近邻图构建算法,显著降低了K近邻图索引的构建开销.其次,提出了基于层间并行的K-means树分区算法及基于广度优先搜索和核心近邻图的并行聚类算法,改进了DBSCAN算法的计算流程,实现了高并发向量聚类.最后,在真实向量数据集上进行了大量实验,并将所提出的方法与现有方法进行了性能对比.实验结果表明,所提方法在保证聚类精度的前提下,将大规模向量聚类的效率提高了5.7–2822.5倍.展开更多
航空标准数据具有结构复杂、语义严谨和跨文档引用频繁等特点,为实现高效、精准的知识获取与问答应用带来挑战,本文提出一种面向航空标准的大语言模型(Large language models,LLMs)迭代检索增强生成(Retrieval-augmented generation,RAG...航空标准数据具有结构复杂、语义严谨和跨文档引用频繁等特点,为实现高效、精准的知识获取与问答应用带来挑战,本文提出一种面向航空标准的大语言模型(Large language models,LLMs)迭代检索增强生成(Retrieval-augmented generation,RAG)方法,设计了基于结构路径感知的标准向量知识库构建与检索机制,结合标准文档的章节结构与标题链条构建支持语境追溯的知识库,并提出基于关键词与语义融合的知识检索机制。在此基础上,设计LLM驱动的自动迭代检索与生成机制,使模型能够自主判断是否需要发起子问题拆解与深层意图识别,并结合多轮检索与动态调度策略,实现问题拆解、信息获取、自主判断与生成控制的一体化闭环,提升对多知识点聚合型、语义递进型等复杂标准问答任务的生成质量与覆盖深度。实验基于7459份航空标准文档构建知识库,针对500条专家标注问答对,在4类涵盖不同参数规模、模型类型及中英文语言能力的主流开源大语言模型上开展对比实验。结果表明,对于中大型参数规模的大模型,此方法在回答准确性、覆盖度和表达质量等指标上均显著优于传统方法。在大模型DeepSeek-R1-70B上,双语评估替补(Bilingual evaluation understudy,BLEU)指标平均提升27.97%,模拟主观评分提升7.99%;在大模型Qwen-2.5-32B上,BLEU指标平均提升54.67%,模拟主观评分提升8.58%。本文所提方法不仅适用于航空标准场景,也可推广至适航规章、维修手册等其他航空结构化文档场景,以及法律、医疗等对回答效果、可信度与可溯源性要求极高的领域,为相关问答系统的构建提供通用的技术框架与实现路径。展开更多
Content-based filtering E-commerce recommender system was discussed fully in this paper. Users＇ unique features can be explored by means of vector space model firstly. Then based on the qualitative value of products ...Content-based filtering E-commerce recommender system was discussed fully in this paper. Users＇ unique features can be explored by means of vector space model firstly. Then based on the qualitative value of products informa tion, the recommender lists were obtained. Since the system can adapt to the users＇ feedback automatically, its performance were enhanced comprehensively. Finally the evaluation of the system and the experimental results were presented.展开更多
文摘Support vector machines and a Kalman-like observer are used for fault detection and isolation in a variable speed horizontalaxis wind turbine composed of three blades and a full converter. The support vector approach is data-based and is therefore robust to process knowledge. It is based on structural risk minimization which enhances generalization even with small training data set and it allows for process nonlinearity by using flexible kernels. In this work, a radial basis function is used as the kernel. Different parts of the process are investigated including actuators and sensors faults. With duplicated sensors, sensor faults in blade pitch positions,generator and rotor speeds can be detected. Faults of type stuck measurements can be detected in 2 sampling periods. The detection time of offset/scaled measurements depends on the severity of the fault and on the process dynamics when the fault occurs. The converter torque actuator fault can be detected within 2 sampling periods. Faults in the actuators of the pitch systems represents a higher difficulty for fault detection which is due to the fact that such faults only affect the transitory state(which is very fast) but not the final stationary state. Therefore, two methods are considered and compared for fault detection and isolation of this fault: support vector machines and a Kalman-like observer. Advantages and disadvantages of each method are discussed. On one hand, support vector machines training of transitory states would require a big amount of data in different situations, but the fault detection and isolation results are robust to variations in the input/operating point. On the other hand, the observer is model-based, and therefore does not require training, and it allows identification of the fault level, which is interesting for fault reconfiguration. But the observability of the system is ensured under specific conditions, related to the dynamics of the inputs and outputs. The whole fault detection and isolation scheme is evaluated using a wind turbine benchmark with a real sequence of wind speed.
文摘Ⅰ. INTRODUCTION MULTILEVEL inverters are increasingly being used in high-power medium voltage applications due to their superior performance compared to two-level inverters, such as lower common-mode voltage, lower dv/dt, lower harmonics in output voltage and current, and reduced voltage on the power switches.
基金National Natural Science FundsNatural Science Funds of Jiangsu Province
文摘This paper presents a new system identification approach using vector space base functions, and proposes two network structures based on Gamma sequence and Laguerre sequence. After analyzing and comparing these structures in detail, some simulation results to demonstrate the conclusions are given.
基金National Natural Science Foundation of China(No.61374140)the Youth Foundation of National Natural Science Foundation of China(No.61403072)
文摘Complex industry processes often need multiple operation modes to meet the change of production conditions. In the same mode,there are discrete samples belonging to this mode. Therefore,it is important to consider the samples which are sparse in the mode.To solve this issue,a new approach called density-based support vector data description( DBSVDD) is proposed. In this article,an algorithm using Gaussian mixture model( GMM) with the DBSVDD technique is proposed for process monitoring. The GMM method is used to obtain the center of each mode and determine the number of the modes. Considering the complexity of the data distribution and discrete samples in monitoring process,the DBSVDD is utilized for process monitoring. Finally,the validity and effectiveness of the DBSVDD method are illustrated through the Tennessee Eastman( TE) process.
文摘In this article, we propose two control charts namely, the “Multivariate Group Runs’ (MV-GR-M)” and the “Multivariate Modified Group Runs’ (MV-MGR-M)” control charts, based on the multivariate normal processes, for monitoring the process mean vector. Methods to obtain the design parameters and operations of these control charts are discussed. Performances of the proposed charts are compared with some existing control charts. It is verified that, the proposed charts give a significant reduction in the out-of-control “Average Time to Signal” (ATS) in the zero state, as well in the steady state compared to the Hotelling’s T2 and the synthetic T2 control charts.
文摘This paper presents a corner-based image alignment algorithm based on the procedures of corner-based template matching and geometric parameter estimation. This algorithm consists of two stages: 1) training phase, and 2) matching phase. In the training phase, a corner detection algorithm is used to extract the corners. These corners are then used to build the pyramid images. In the matching phase, the corners are obtained using the same corner detection algorithm. The similarity measure is then determined by the differences of gradient vector between the corners obtained in the template image and the inspection image, respectively. A parabolic function is further applied to evaluate the geometric relationship between the template and the inspection images. Results show that the corner-based template matching outperforms the original edge-based template matching in efficiency, and both of them are robust against non-liner light changes. The accuracy and precision of the corner-based image alignment are competitive to that of edge-based image alignment under the same environment. In practice, the proposed algorithm demonstrates its precision, efficiency and robustness in image alignment for real world applications.
文摘The size and performance of a System LSI depend heavily on the architecture which is chosen. As a result, the architecture design phase is one of the most important steps in the System LSI development process and is critical to the commercial success of a device. In this paper, we propose a C-based variable length and vector pipeline (VVP) architecture design methodology and apply it to the design of the output probability computation circuit for a speech recognition system. VVP processing accelerated by loop optimization, memory access methods, and application-specific cir- cuit design was implemented to calculate the Hidden Markov Model (HMM) output probability at high speed and its performance is evaluated. It is shown that designers can explore a wide range of design choices and generate complex circuits in a short time by using a C-based pipeline architecture design method.
基金supported by the Major Program of the National Natural Science Foundation of China (Grant 51490662)the Funds for Distinguished Young Scientists of Hunan Province (Grant 14JJ1016)+1 种基金the State Key Program of the National Science Foundation of China (11232004)the Heavy-duty Tractor Intelligent Manufacturing Technology Research and System Development (Grant 2016YFD0701105)
文摘Use of multidisciplinary analysis in reliabilitybased design optimization(RBDO) results in the emergence of the important method of reliability-based multidisciplinary design optimization(RBMDO). To enhance the efficiency and convergence of the overall solution process,a decoupling algorithm for RBMDO is proposed herein.Firstly, to decouple the multidisciplinary analysis using the individual disciplinary feasible(IDF) approach, the RBMDO is converted into a conventional form of RBDO. Secondly,the incremental shifting vector(ISV) strategy is adopted to decouple the nested optimization of RBDO into a sequential iteration process composed of design optimization and reliability analysis, thereby improving the efficiency significantly. Finally, the proposed RBMDO method is applied to the design of two actual electronic products: an aerial camera and a car pad. For these two applications, two RBMDO models are created, each containing several finite element models(FEMs) and relatively strong coupling between the involved disciplines. The computational results demonstrate the effectiveness of the proposed method.
文摘聚类是大规模高维向量数据分析的关键技术之一.近年来,基于密度的聚类算法DBSCAN(density-based spatial clustering of applications with noise)因其无须预先指定聚类数量、能够发现复杂聚类结构并有效识别噪声点的特性,在数据分析领域得到了广泛应用.然而,现有的基于密度的聚类算法在处理高维向量数据时将产生极高的时间代价且面临维度灾难等问题,难以在实际场景中部署应用.此外,随着信息技术的发展,高维向量数据规模急剧增加,使用CPU进行高维向量聚类在时间代价和可扩展性等方面将面临更大的挑战.为此,提出一种GPU加速的高维向量聚类算法,通过引入K近邻(K-nearest neighbor,KNN)图索引加速DBSCAN的计算.首先,设计了GPU加速的并行K近邻图构建算法,显著降低了K近邻图索引的构建开销.其次,提出了基于层间并行的K-means树分区算法及基于广度优先搜索和核心近邻图的并行聚类算法,改进了DBSCAN算法的计算流程,实现了高并发向量聚类.最后,在真实向量数据集上进行了大量实验,并将所提出的方法与现有方法进行了性能对比.实验结果表明,所提方法在保证聚类精度的前提下,将大规模向量聚类的效率提高了5.7–2822.5倍.
基金Supported bythe Hunan Teaching Reformand Re-search Project of Colleges and Universities (2003-B72) the HunanBoard of Review on Philosophic and Social Scientific Pay-off Project(0406035) the Hunan Soft Science Research Project(04ZH6005)
文摘Content-based filtering E-commerce recommender system was discussed fully in this paper. Users＇ unique features can be explored by means of vector space model firstly. Then based on the qualitative value of products informa tion, the recommender lists were obtained. Since the system can adapt to the users＇ feedback automatically, its performance were enhanced comprehensively. Finally the evaluation of the system and the experimental results were presented.
