Hitachi, Ltd.

Ahmed Khairy Farahat

Chetan Kumar Gupta

Kosta Ristovski

Dipanjan Ghosh

Chi Zhang

Wei T Huang

Maria Teresa Gonzalez Diaz

Hamed Khorasgani

Shuai Zheng

Marcos Vieira

Wei Huang

Hsiu-Khuern Tang

Haiyan Wang

Mahbubul Alam

Qiyao Wang

Adriano Arantes

Tsubasa Watanabe

Lasitha Vidyaratne

Onur Atan

Aman Kumar

Xian Yeow Lee

Tomoaki Hiruta

Arman Hasanzadehmoghimi

Lijing Wang

Karan Aggarwal

Walid Shalaby

Hideaki Suzuki

Systems and methods for constructing an equipment troubleshooting tree involve receiving an unstructured text source and using a parsing tool to segment it into sections that are associated with equipment components, units, or processes. If the parsing tool fails, a pre-trained language model is used to identify the sections using the extracted listing as content. Metadata, including section-specific text and page ranges, are obtained for each section. A non-static process enhances similarity matching, resulting in segmented data. An indexing process embeds this data, along with the metadata, into index nodes to improve retrieval efficiency. The pre-trained language model generates a catalog of false, causes, or repair recommendations, which are structured as index notes in the troubleshooting tree. In response to a user query, the constructed troubleshooting tree is generated and output after a grounding process cross-references it with the unstructured text to ensure accuracy.

Systems and method for constructing troubleshooting trees

A method for detecting an anomaly in time series sensor data. The method may include identifying a noisiest cycle from the time series sensor data; for an evaluation of the noisiest cycle indicative of the anomaly being detected at a confidence level above a threshold, providing an output associated with the noisiest cycle as being the anomaly; and for the evaluation of the noisiest cycle indicative of the anomaly being detected at the confidence level not above the threshold: identifying a cycle from the time series sensor data having a most differing shape; and providing the output associated with the cycle having the most differing shape as being the anomaly.

Versatile anomaly detection system for industrial systems

In some examples, a computer system may receive sensor data and failure data for equipment. The system may determine, for the equipment, a plurality of time between failure (TBF) durations that are longer than other TBF durations for the equipment. The system may determine, from the sensor data corresponding to operation of the equipment during the plurality of TBF durations, a plurality of measured sensor values for the equipment. Additionally, the system may determine a subset of the measured sensor values corresponding to a largest number of the TBF durations of the plurality of TBF durations. The system may further determine at least one operating parameter value for the equipment based on the subset of the measured sensor values. The system may send a control signal for operating the equipment based on the operating parameter value and/or a communication based on the operating parameter value.

Determining an operating envelope for controlling equipment

Example implementations involve fault detection and isolation in industrial networks through defining a component as a combination of measurements and parameters and define an industrial network as a set of components connected with different degrees of connections (weights). Faults in industrial network are defined as unpermitted changes in component parameters. Further, the fault detection and isolation in industrial networks are formulated as a node classification problem in graph theory. Example implementations detect and isolate faults in industrial networks through 1) uploading/learning network structure, 2) detecting component communities in the network, 3) extracting features for each community, 4) using the extracted features for each community to detect and isolate faults, 5) at each time step, based on the faulty components provide maintenance recommendation for the network.

System and method for maintenance recommendation in industrial networks

Example implementations described herein involve systems and methods for generating an ensemble of deep learning or neural network models, which can involve, for a training set of data, generating a plurality of model samples for the training set of data, the plurality of model samples generated from deep learning or neural network methods; and aggregating output of the model samples to generate an output of the ensemble models.

Method for reproducibility of deep learning classifiers using ensembles

In some examples, a computer system may receive historical repair data and may extract features from the historical repair data for use as training data. The computer system may determine, from the historical repair data, a repair hierarchy including a plurality of repair levels which includes repair actions as one of the repair levels. Furthermore, the computer system may train the machine learning model, which performs multiple tasks for predicting values of individual levels of the repair hierarchy, by tuning parameters of the machine learning model using the training data.

Equipment repair management and execution

Example implementations described herein involve a new data-driven analytical redundancy relationship (ARR) generation for fault detection and isolation. The proposed solution uses historical data during normal operation to extract the data-driven ARRs among sensor measurements, and then uses them for fault detection and isolation. The proposed solution thereby does not need to rely on the system model, can detect and isolate more faults than traditional data-driven methods, can work when the system is not fully observable, and does not rely on a vast amount of historical fault data, which can save on memory storage or database storage. The proposed solution can thereby be practical in many real cases where there are data limitations.

System and method for data-driven analytical redundancy relationships generation for early fault detection and isolation with limited data

Example implementations involve a system for Predictive Maintenance using Generative Adversarial Networks for Failure Prediction. Through utilizing three processes concurrently and training them iteratively with data-label pairs, example implementations described herein can thereby generate a more accurate predictive maintenance model than that of the related art. Example implementations further involve shared networks so that the three processes can be trained concurrently while sharing parameters with each other.

System for predictive maintenance using generative adversarial networks for failure prediction

Example implementations described herein are directed to systems and methods for estimating the remaining useful life of a component or equipment through the application of models for deriving functions that can express the remaining useful life over time. In an aspect, the failure acceleration time point is determined for a given type of component, and a function is derived based on the application of models on the failure acceleration time point.

Deep long short term memory network for estimation of remaining useful life of the components

Example implementations described herein involve, for data having incomplete labeling to generate a plurality of predictive maintenance models, processing the data through a multi-task learning (MTL) architecture including generic layers and task specific layers for the plurality of predictive maintenance models configured to conduct tasks to determine outcomes for one or more components associated with the data, each task specific layer corresponding to one of the plurality of predictive maintenance models; the generic layers configured to provide, to the task specific layers, associated data to construct each of the plurality of predictive maintenance models; and executing the predictive maintenance models on subsequently recorded data.

Multi task learning with incomplete labels for predictive maintenance

Example implementations described herein involve systems and methods for conducting feature extraction on a plurality of templates associated with vibration sensor data for a moving equipment configured to conduct a plurality of tasks, to generate a predictive maintenance model for the plurality of tasks, the predictive maintenance model configured to provide one or more of fault detection, failure prediction, and remaining useful life (RUL) estimation.

Predictive maintenance for robotic arms using vibration measurements

Example implementations described herein involve a system for maintenance predictions generated using a single deep learning architecture. The example implementations can involve managing a single deep learning architecture for three modes including a failure prediction mode, a remaining useful life (RUL) mode, and a unified mode. Each mode is associated with an objective function and a transformation function. The single deep learning architecture is applied to learn parameters for an objective function through execution of a transformation function associated with a selected mode using historical data. The learned parameters of the single deep learning architecture can be applied with streaming data from with the equipment to generate a maintenance prediction for the equipment.

Deep learning architecture for maintenance predictions with multiple modes

Example implementations described herein are directed to systems and methods for predictive maintenance with health indicators using reinforcement learning. An example implementation includes a method to receive sensor data, operational condition data, and failure event data and generate a model to determine health indicators that indicate equipment performance based on learned policies, state values, and rewards. The model is applied to external sensor readings and operating data for a piece of equipment to output a recommendation based on the model.

Automatic health indicator learning using reinforcement learning for predictive maintenance

In some examples, a system may receive from a device, speech sound patterns corresponding to a voice input related to equipment. Further, the system may determine an identity of a person associated with the device, and may identify the equipment related to the voice input. Using at least one of the received speech sound patterns or a text conversion of the speech sound patterns, along with an equipment history of the identified equipment, as input to one or more machine learning models, the system may determine, at least partially, an instruction related to the equipment. Additionally, the system may send, to the device, the instruction related to the equipment as an audio file for playback on the device.

Management and execution of equipment maintenance

Example implementations described herein involve a system for maintenance recommendation based on data-driven failure prediction. The example implementations can involve estimating the probability of having a failure event in the near future given sensor measurements and events from the equipment, and then alerts the system user or maintenance staff if the probability of failure exceeds a certain threshold. The example implementations utilize historical failure cases along with the associated sensor measurements and events to learn a group of classification models that differentiate between failure and non-failure cases. In example implementations, the system then chooses the optimal model for failure prediction such that the overall cost of the maintenance process is minimized.

Growing community of inventors

Santa Clara, CA, United States of America

Ahmed Khairy Farahat