中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

11 records (took 0.018 seconds)

    SEGMENT-BASED CHANGE DETECTION METHOD IN MULTIVARIATE DATA STREAM
    2.
    发明申请
    SEGMENT-BASED CHANGE DETECTION METHOD IN MULTIVARIATE DATA STREAM 审中-公开
    多元数据流中基于分段的变化检测方法

    公开(公告)号：WO2009045312A1

    公开(公告)日：2009-04-09

    申请号：PCT/US2008/011104

    申请日：2008-09-25

    Applicant: SIEMENS CORPORATE RESEARCH, INC. ,  CHEN, Terrence ,  YUAN, Chao ,  SHEIKH, Abdul Saboor ,  NEUBAUER, Claus

    Inventor： CHEN, Terrence ,  YUAN, Chao ,  SHEIKH, Abdul Saboor ,  NEUBAUER, Claus

    IPC: G06K9/00 ,  G06K9/62 ,  G05B23/02

    CPC classification number: G06K9/00536 ,  G06K9/6284

    Abstract: A method and framework are described for detecting changes in a multivariate data stream. A training set is formed by sampling time windows in a data stream containing data reflecting normal conditions. A histogram is created to summarize each window of data, and data within the histograms are clustered to form test distribution representatives to minimize the bulk of training data. Test data is then summarized using histograms representing time windows of data and data within the test histograms are clustered. The test histograms are compared to the training histograms using nearest neighbor techniques on the clustered data. Distances from the test histograms to the test distribution representatives are compared to a threshold to identify anomalies.

    Abstract translation: 描述了用于检测多变量数据流中的变化的方法和框架。 通过在包含反映正常条件的数据的数据流中采样时间窗口来形成训练集。 创建直方图以总结每个数据窗口，并且将直方图中的数据进行聚类以形成测试分发代表以最小化训练数据的大部分。 然后使用表示数据的时间窗口的直方图来汇总测试数据，并且将测试直方图中的数据聚类。 将测试直方图与使用最近邻技术的聚类数据的训练直方图进行比较。 将测试直方图与测试分布代表的距离与阈值进行比较以识别异常。

    SYSTEM, DEVICE, AND METHODS FOR UPDATING SYSTEM-MONITORING MODELS
    3.
    发明申请
    SYSTEM, DEVICE, AND METHODS FOR UPDATING SYSTEM-MONITORING MODELS 审中-公开
    用于更新系统监测模型的系统，装置和方法

    公开(公告)号：WO2006026340A1

    公开(公告)日：2006-03-09

    申请号：PCT/US2005/030213

    申请日：2005-08-25

    Applicant: SIEMENS CORPORATE RESEARCH, INC. ,  YUAN, Chao ,  NEUBAUER, Claus ,  CATALTEPE, Zehra ,  MCCORKLE, Wesley ,  BRUMMEL, Hans-Gerd ,  FANG, Ming

    Inventor： YUAN, Chao ,  NEUBAUER, Claus ,  CATALTEPE, Zehra ,  MCCORKLE, Wesley ,  BRUMMEL, Hans-Gerd ,  FANG, Ming

    IPC: G05B23/02

    CPC classification number: G05B17/02 ,  G05B23/0243

    Abstract: A system (102) for updating a plurality of monitoring models is provided. The system (102) includes a model association module (202) that, for each of a plurality of monitored systems (104a, 104b, 104c) determines, an association between a particular monitored system and at least one of a plurality of estimation models. Each estimation model is based upon one of a plurality of distinct sets of estimation properties, and each set uniquely corresponds to a particular estimation model. The system also includes an updating module (204) that updates at least one of the estimation properties and propagates the updated estimation properties to each estimation model that corresponds to a distinct set containing at least one estimation property that is updated. The system further includes a model modification module (206) that modifies each estimation model that corresponds to a distinct set containing at least one estimation property that is updated.

    Abstract translation: 提供了一种用于更新多个监视模型的系统（102）。 系统（102）包括模型关联模块（202），对于多个被监视系统（104a，104b，104c）中的每一个确定特定的被监控系统与多个估计模型中的至少一个之间的关联。 每个估计模型基于多个不同的估计属性集合之一，并且每个估计模型唯一地对应于特定估计模型。 该系统还包括更新模块（204），其更新估计属性中的至少一个并且将更新的估计属性传播到对应于包含至少一个被更新的估计属性的不同集合的每个估计模型。 该系统还包括模型修改模块（206），其修改对应于包含至少一个被更新的估计属性的不同集合的每个估计模型。

    SUPERVISED FAULT LEARNING USING RULE-GENERATED SAMPLES FOR MACHINE CONDITION MONITORING
    4.
    发明申请
    SUPERVISED FAULT LEARNING USING RULE-GENERATED SAMPLES FOR MACHINE CONDITION MONITORING 审中-公开
    使用规则生成样本进行机器状况监控的监督故障学习

    公开(公告)号：WO2011034805A1

    公开(公告)日：2011-03-24

    申请号：PCT/US2010/048579

    申请日：2010-09-13

    Applicant: SIEMENS AKTIENGESELLSCHAFT ,  YUAN, Chao ,  NEUBAUER, Claus ,  HACKSTEIN, Holger

    Inventor： YUAN, Chao ,  NEUBAUER, Claus ,  HACKSTEIN, Holger

    IPC: G05B23/02

    CPC classification number: G05B23/0229 ,  G05B23/024 ,  G05B23/0283

    Abstract: A machine fault diagnosis system combines a rule-based predictive maintenance strategy with a machine learning system. A simple set of rules defined manually by human experts is used to generate artificial training feature vectors to portray machine fault conditions for which only a few real data points are available. Those artificial training feature vectors are combined with real training feature vectors and the combined set is used to train a supervised pattern recognition algorithm such as support vector machines (SVM). The resulting decision boundary closely approximates the underlying real separation boundary between the fault and normal conditions

    Abstract translation: 机器故障诊断系统将基于规则的预测维护策略与机器学习系统相结合。 人类专家手动定义的一套简单的规则用于生成人工训练特征向量，以描绘只有几个实际数据点可用的机器故障条件。 将这些人工训练特征向量与真实训练特征向量相结合，组合集合用于训练支持向量机（SVM）等监督模式识别算法。 所得到的决策边界与故障和正常条件之间的基本实际分离边界紧密相近

    BAYESIAN NETWORK FRAMEWORKS FOR BIOMEDICAL DATA MINING
    5.
    发明申请
    BAYESIAN NETWORK FRAMEWORKS FOR BIOMEDICAL DATA MINING 审中-公开
    贝耶斯网络生物医学数据挖掘框架

    公开(公告)号：WO2005119582A2

    公开(公告)日：2005-12-15

    申请号：PCT/US2005014718

    申请日：2005-05-02

    Applicant: SIEMENS CORP RES INC ,  CHENG JIE ,  NEUBAUER CLAUS

    Inventor： CHENG JIE ,  NEUBAUER CLAUS

    IPC: G06K9/62

    CPC classification number: G06K9/6296

    Abstract: A system (100) and method (200) for data classification are provided, the system including a processor (102), an adapter (112) in signal communication with the processor for receiving data, a filtering unit (170) in signal communication with the processor for pre-processing the data and filtering features of the data, a selection unit (180) in signal communication with the processor for learning a Bayesian network (BN) classifier and selecting features responsive to the BN classifier, and an evaluation unit (190) in signal communication with the processor for evaluating a model responsive to the BN classifier; and the method including receiving data (212), pre-processing the data (214), filtering features of the data (216), learning a BN classifier (218), selecting features responsive to the BN classifier (220), and evaluating a model responsive to the BN classifier (222).

    Abstract translation: 提供了一种用于数据分类的系统（100）和方法（200），该系统包括处理器（102），与处理器进行信号通信的适配器（112），用于接收数据;滤波单元（170），其与 所述处理器用于预处理所述数据和过滤所述数据的特征;与处理器进行信号通信的选择单元（180），用于学习贝叶斯网络（BN）分类器并响应于所述BN分类器选择特征;以及评估单元 190）与所述处理器进行信号通信，用于评估响应于所述BN分类器的模型; 并且所述方法包括接收数据（212），预处理数据（214），过滤数据的特征（216），学习BN分类器（218），响应于BN分类器（220）选择特征，以及评估 模型响应于BN分类器（222）。

    BAYESIAN NETWORK FRAMEWORKS FOR BIOMEDICAL DATA MINING
    6.
    发明申请
    BAYESIAN NETWORK FRAMEWORKS FOR BIOMEDICAL DATA MINING 审中-公开

    公开(公告)号：WO2005119582A3

    公开(公告)日：2005-12-15

    申请号：PCT/US2005/014718

    申请日：2005-05-02

    Applicant: SIEMENS CORPORATE RESEARCH, INC. ,  CHENG, Jie ,  NEUBAUER, Claus

    Inventor： CHENG, Jie ,  NEUBAUER, Claus

    IPC: G06K9/62

    Abstract: A system (100) and method (200) for data classification are provided, the system including a processor (102), an adapter (112) in signal communication with the processor for receiving data, a filtering unit (170) in signal communication with the processor for pre-processing the data and filtering features of the data, a selection unit (180) in signal communication with the processor for learning a Bayesian network (BN) classifier and selecting features responsive to the BN classifier, and an evaluation unit (190) in signal communication with the processor for evaluating a model responsive to the BN classifier; and the method including receiving data (212), pre-processing the data (214), filtering features of the data (216), learning a BN classifier (218), selecting features responsive to the BN classifier (220), and evaluating a model responsive to the BN classifier (222).

    MACHINE CONDITION MONITORING USING DISCONTINUITY DETECTION
    7.
    发明申请
    MACHINE CONDITION MONITORING USING DISCONTINUITY DETECTION 审中-公开
    使用不连续检测的机器状态监控

    公开(公告)号：WO2008127539A1

    公开(公告)日：2008-10-23

    申请号：PCT/US2008/003717

    申请日：2008-03-20

    Applicant: SIEMENS CORPORATE RESEARCH, INC. ,  NEUBAUER, Claus ,  YUAN, Chao

    Inventor： NEUBAUER, Claus ,  YUAN, Chao

    IPC: G05B23/02

    CPC classification number: G05B23/0232

    Abstract: Condition signals of machines are observed and one or more discontinuities are detected in the condition signals. The discontinuities in the condition signals are compensated for (e.g., by applying a shifting factor to models of the signals) and trends of the compensated condition signals are determined. The trends are used to predict future fault conditions in machines. Kalman filters comprising observation models and evolution models are used to determine the trends. Discontinuity in observed signals is detected using hypothesis testing.

    Abstract translation: 观察机器的状态信号，并在条件信号中检测到一个或多个不连续性。 条件信号中的不连续性被补偿（例如，通过将移位因子应用于信号的模型），并且确定补偿状态信号的趋势。 趋势用于预测机器中未来的故障状况。 包含观察模型和进化模型的卡尔曼滤波被用于确定趋势。 使用假设检验检测观察信号的不连续性。

    MACHINE CONDITION MONITORING USING PATTERN RULES
    8.
    发明申请
    MACHINE CONDITION MONITORING USING PATTERN RULES 审中-公开
    使用模式规则进行机器状况监测

    公开(公告)号：WO2008127535A1

    公开(公告)日：2008-10-23

    申请号：PCT/US2008/003642

    申请日：2008-03-20

    Applicant: SIEMENS CORPORATE RESEARCH, INC. ,  SIEMENS AKTIENGESELLSCHAFT ,  HACKSTEIN, Holger ,  NEUBAUER, Claus ,  YUAN, Chao

    Inventor： HACKSTEIN, Holger ,  NEUBAUER, Claus ,  YUAN, Chao

    IPC: G05B23/02

    CPC classification number: G05B23/0229

    Abstract: Pattern rules are created by comparing a condition signal pattern to a plurality of known signal patterns and determining a machine condition pattern rule based at least in part on the comparison of the condition signal pattern to one of the plurality of known signal patterns. A matching score based on the comparison of the condition signal pattern to one of the plurality of known signal patterns as well as a signal pattern duration is determined. The machine condition pattern rule is then defined for nonparametric condition signal patterns as a multipartite threshold rule with a first threshold based on the determined matching score and a second threshold based on the determined signal duration. For parametric signal patterns, one or more parameters of the signal pattern are determined and the machine condition pattern rule is further defined with a third threshold based on the determined one or more parameters.

    Abstract translation: 通过将条件信号模式与多个已知信号模式进行比较并至少部分地基于条件信号模式与多种已知信号模式之一的比较来确定机器状态模式规则来创建模式规则。 确定基于条件信号模式与多个已知信号模式中的一个的比较以及信号模式持续时间的匹配分数。 然后，根据所确定的匹配分数和基于确定的信号持续时间的第二阈值，将非参数条件信号模式定义为具有第一阈值的多部分阈值规则的机器状态模式规则。 对于参数信号模式，确定信号模式的一个或多个参数，并且基于所确定的一个或多个参数，用第三阈值进一步定义机器状态模式规则。

    USE OF SEQUENTIAL CLUSTERING FOR INSTANCE SELECTION IN MACHINE CONDITION MONITORING
    9.
    发明申请
    USE OF SEQUENTIAL CLUSTERING FOR INSTANCE SELECTION IN MACHINE CONDITION MONITORING 审中-公开
    在机器状态监测中使用连续选择的顺序聚类

    公开(公告)号：WO2007067521A1

    公开(公告)日：2007-06-14

    申请号：PCT/US2006/046361

    申请日：2006-12-05

    Applicant: SIEMENS CORPORATE RESEARCH, INC. ,  BALDERER, Christian ,  YUAN, Chao ,  NEUBAUER, Claus

    Inventor： BALDERER, Christian ,  YUAN, Chao ,  NEUBAUER, Claus

    IPC: G05B17/02 ,  G05B23/02

    CPC classification number: G05B17/02 ,  G05B23/0248 ,  G06K9/622 ,  G06K9/6256

    Abstract: A method is provided for selecting a representative set of training data for training a statistical model in a machine condition monitoring system. The method reduces the time required to choose representative samples from a large data set by using a nearest-neighbor sequential clustering technique in combination with a k d-tree. A distance threshold is used to limit the geometric size the clusters. Each node of the k d-tree is assigned a representative sample from the training data, and similar samples are subsequently discarded.

    Abstract translation: 提供了一种用于在机器状态监视系统中选择用于训练统计模型的代表性训练数据集合的方法。 该方法通过使用最近邻序列聚类技术与akd-tree组合，减少了从大型数据集中选择代表性样本所需的时间。 距离阈值用于限制集群的几何尺寸。 从训练数据中分配kd-tree的每个节点代表性样本，随后丢弃类似的样本。

Patent Agency Ranking