公开(公告)号：US20170212012A1

公开(公告)日：2017-07-27

申请号：US15328676

申请日：2014-07-30

Applicant: HITACHI, LTD.

Inventor： Munetoshi UNUMA ,  Takashi SAEKI ,  Shinya YUDA

IPC: G01M17/007

CPC classification number: G01M17/007 ,  B60W50/00 ,  G05B23/0227 ,  G05B23/0275 ,  G07C5/08

Abstract: A degradation cause estimation device is provided with a degradation detector for detecting the amount of degradation undergone by a device, a state observation device for detecting observation values for internal portions, values observed from outside, or device control and operation information, a degradation section detector for detecting a section undergoing degradation using the output values of the degradation detector, and a cause estimator for estimating the cause of the degradation using the device state observations for the degradation section, and outputs the cause for the degradation section.

公开(公告)号：US20160320262A1

公开(公告)日：2016-11-03

申请号：US15108078

申请日：2014-03-03

Applicant: HITACHI, LTD.

Inventor： Munetoshi UNUMA ,  Shinya YUDA ,  Takashi SAEKI ,  Hideaki SUZUKI

IPC: G01M5/00 ,  G01M13/00 ,  G01M17/007

CPC classification number: G01M5/0033 ,  G01M13/00 ,  G01M17/007

Abstract: An objective of the present invention is to associate condition information, such as (a) a location in a movement section at which fatigue occurred, (b) a type of work or operation during which fatigue occurred, and (c) a type of natural phenomenon experienced when fatigue occurred, with machine fatigue occurring when a machine is moved or placed in a work environment, natural environment, or other condition and to visualize and display the machine fatigue. To achieve this objective, this device for displaying the material fatigue of a machine is characterized by being provided with a means for observing stress variation in a time series using a sensor correlated with the stress variation or stress of a machine, a means for detecting stress variation in a single cycle, and a means for detecting the time when the stress variation occurred, and by outputting or displaying the stress variation value and time of occurrence.

Abstract translation: 本发明的目的是关联条件信息，例如（a）发生疲劳的运动部分中的位置，（b）发生疲劳的工作类型或操作类型，以及（c）一种自然的 疲劳发生时出现的现象，当机器移动或放置在工作环境，自然环境或其他状况下发生机器疲劳并且可视化和显示机器疲劳时。 为了达到这个目的，用于显示机器的材料疲劳的装置的特征在于具有用于使用与机器的应力变化或应力相关的传感器来观察时间序列中的应力变化的装置，用于检测应力的装置 单个周期的变化，以及用于检测应力变化发生时间的装置，以及通过输出或显示应力变化值和发生时间的装置。

公开(公告)号：US20200380447A1

公开(公告)日：2020-12-03

申请号：US16428141

申请日：2019-05-31

Applicant: Hitachi, Ltd.

Inventor： Qiyao WANG ,  Haiyan WANG ,  Susumu SERITA ,  Takashi SAEKI ,  Chetan GUPTA

IPC: G06Q10/06 ,  H04L29/08 ,  G06N20/00

Abstract: Example implementations described herein involve systems and methods involving a plurality of sensors monitoring one or more processes, the sensors providing sensor data, which can include determining a probability map of the sensor data from a database and a functional relationship between key performance indicators (KPIs) of the one or more processes and the sensor data; executing a search on the probability map to determine constrained and continuous ranges for the sensor data that optimize KPIs for the one or more processes based on the functional relationship; and generating a recommendation for the one or more processes that fit within the constrained and continuous range of the sensor data.

公开(公告)号：US20200380388A1

公开(公告)日：2020-12-03

申请号：US16428016

申请日：2019-05-31

Applicant: Hitachi, Ltd.

Inventor： Qiyao WANG ,  Shuai ZHENG ,  Ahmed FARAHAT ,  Susumu SERITA ,  Takashi SAEKI ,  Chetan GUPTA

IPC: G06N5/04 ,  G06N20/00

Abstract: Example implementations described herein are directed to constructing prediction models and conducting predictive maintenance for systems that provide sparse sensor data. Even if only sparse measurements of sensor data are available, example implementations utilize the inference of statistics with functional deep networks to model prediction for the systems, which provides better accuracy and failure prediction even if only sparse measurements are available.

公开(公告)号：US20180024544A1

公开(公告)日：2018-01-25

申请号：US15540275

申请日：2015-01-21

Applicant: Hitachi, Ltd.

Inventor： Munetoshi UNUMA ,  Yasuki KITA ,  Takashi SAEKI

IPC: G05B23/02 ,  G07C3/00 ,  H04L29/08 ,  G07C5/00 ,  B60R16/023

CPC classification number: G05B23/0283 ,  B60R16/0234 ,  G06Q10/20 ,  G07C3/00 ,  G07C5/006 ,  H04L67/12

Abstract: According to the present invention, the work status at an actual site is represented using a combination of component statuses. A degree of damage that takes into account the manner of use at the actual site is predicted by associating test data collected in a component status test. The objective of the present invention is to estimate, with high accuracy, the degree of damage to a device operating in an actual environment. In order to achieve this objective, this damage estimation device is provided with: a component status classification processing unit that classifies the operating status of a device; a component status characteristic amount storage unit that stores a characteristic amount for classification processing by the component status classification processing unit; and a damage degree storage unit for each component status, said damage degree storage unit storing the degree of damage to various regions of a device for each component status, wherein the degree of damage is measured in advance for each component status or is calculated using a simulation. The damage estimation device is characterized by collecting sensor/control data required for device status classification, classifying component status using the component status classification processing unit, calculating the classified result and the degree of damage corresponding to said result using the damage degree storage unit for each component status, and outputting the degree of damage.