公开(公告)号：US08498776B2

公开(公告)日：2013-07-30

申请号：US12620466

申请日：2009-11-17

Applicant: Satnam Singh ,  Vineet R. Khare ,  Rahul Chougule

Inventor： Satnam Singh ,  Vineet R. Khare ,  Rahul Chougule

IPC: G01M17/00

CPC classification number: G05B19/0428 ,  G05B2219/24076 ,  G05B2219/2637

Abstract: A system and method for fault diagnosis includes receiving information defining a relationship between failure modes and diagnostic trouble codes and extracting diagnostic trouble code data, including set times, frequency data and diagnostic trouble code sequence information for a plurality of diagnostic trouble codes relating to a plurality of failure modes. The system and method further include constructing a Markov chain using the diagnostic trouble code data for each of the plurality of failure modes, training the Markov chain to learn a set of state parameters using the diagnostic trouble code data, and computing a likelihood of a diagnostic trouble code sequence for each of the plurality of failure modes using the trained Markov chains.

Abstract translation: 用于故障诊断的系统和方法包括接收定义故障模式和诊断故障代码之间的关系的信息，并且提取诊断故障代码数据，包括用于多个诊断故障代码的多个诊断故障代码的设置时间，频率数据和诊断故障代码序列信息 的故障模式。 该系统和方法还包括使用针对多个故障模式中的每一个的诊断故障代码数据构建马尔可夫链，训练马尔可夫链以使用诊断故障代码数据学习一组状态参数，以及计算诊断的可能性 使用训练马尔科夫链的多个故障模式中的每一个的故障代码序列。

公开(公告)号：US20110118932A1

公开(公告)日：2011-05-19

申请号：US12620466

申请日：2009-11-17

Applicant: Satnam Singh ,  Vineet R. Khare ,  Rahul Chougule

Inventor： Satnam Singh ,  Vineet R. Khare ,  Rahul Chougule

IPC: G06F7/00 ,  G06F15/18 ,  G06N5/02 ,  G06F11/07

CPC classification number: G05B19/0428 ,  G05B2219/24076 ,  G05B2219/2637

Abstract: A system and method for fault diagnosis includes receiving information defining a relationship between failure modes and diagnostic trouble codes and extracting diagnostic trouble code data, including set times, frequency data and diagnostic trouble code sequence information for a plurality of diagnostic trouble codes relating to a plurality of failure modes. The system and method further include constructing a Markov chain using the diagnostic trouble code data for each of the plurality of failure modes, training the Markov chain to learn a set of state parameters using the diagnostic trouble code data, and computing a likelihood of a diagnostic trouble code sequence for each of the plurality of failure modes using the trained Markov chains.

Abstract translation: 用于故障诊断的系统和方法包括接收定义故障模式和诊断故障代码之间的关系的信息，并且提取诊断故障代码数据，包括用于多个诊断故障代码的多个诊断故障代码的设置时间，频率数据和诊断故障代码序列信息 的故障模式。 该系统和方法还包括使用针对多个故障模式中的每一个的诊断故障代码数据构建马尔可夫链，训练马尔可夫链以使用诊断故障代码数据学习一组状态参数，以及计算诊断的可能性 使用训练马尔科夫链的多个故障模式中的每一个的故障代码序列。

公开(公告)号：US20120303662A1

公开(公告)日：2012-11-29

申请号：US13114092

申请日：2011-05-24

Applicant: Asim Tewari ,  Vineet R. Khare ,  Sugato Chakrabarty

Inventor： Asim Tewari ,  Vineet R. Khare ,  Sugato Chakrabarty

IPC: G06F17/30

CPC classification number: G06Q10/10

Abstract: A method is provided for enhancing service diagnostics for root cause analysis of an identified problem in a vehicle. Service repair data of previously serviced vehicles is obtained from a memory storage device. The service data is compiled based on a service repair history for each vehicle. Each vehicle within the compiled service data having at least two service repairs performed within a predetermined period of time is identified. Combinations of parts serviced during each service repair are identified. A count is determined that indicates the number of times each combination appears in the compiled service data. The combinations having counts greater than a predetermined threshold are identified. A determination is made whether any of the combinations having counts greater than the predetermined threshold are present in the structural taxonomy database. A functional taxonomy database is updated by assigning the selected combinations to the function taxonomy database not present in the structural taxonomy database.

Abstract translation: 提供了一种用于增强对车辆中识别的问题的根本原因分析的服务诊断的方法。 从存储器存储装置获得先前服务车辆的维修数据。 服务数据是根据每个车辆的维修历史来编辑的。 识别在预定时间段内执行的具有至少两次服务维修的已编译服务数据中的每个车辆。 识别在每次维修期间服务的零件的组合。 确定指示每个组合在编译的服务数据中出现的次数的计数。 识别具有大于预定阈值的计数的组合。 确定在结构分类数据库中是否存在具有大于预定阈值的计数的组合中的任何组合。 通过将所选择的组合分配给结构分类数据库中不存在的功能分类数据库来更新功能分类数据库。