Siemens Aktiengesellschaft

Siemens Corporation

Siemens Corporate Research, Inc.

Siemens Energy, Inc.

Siemens Medical Solutions Usa, Inc.

Siemens Westinghouse Power Corporation

Siemens Aktiengesell Schaft

Siemens Power Generation, Inc.

Neubauer, Claus

Chao Yuan

Claus Neubauer

Zehra Cataltepe

Amit Chakraborty

Holger Hackstein

Ming Fang

Hans-Gerd Johann Brummel

Leif Wiebking

Jie Cheng

Wesley McCorkle

Christian Balderer

Terrence Chen

Chellury R Sastry

Eberhard Ritzhaupt-Kleissl

Bernd Wachmann

Hans Weber

Siong Thye Goh

Stefan Galler

Matthew Evans

Abdul Saboor Sheikh

Tzu-Kuo Huang

Xinmin Cai

A computer-implemented method for performing machine condition monitoring for fault diagnosis includes collecting multivariate time series data from a plurality of sensors in a machine and partitioning the multivariate time series data into a plurality of segment clusters. Each segment cluster corresponds to one of a plurality of class labels related to machine condition monitoring. Next, the segment clusters are clustered into segment cluster prototypes. The segment clusters and the segment cluster prototypes are used to learn a discriminative model that predicts a class label. Then, as new multivariate time series data is collected from the sensors in the machine, the discriminative model may be used to predict a new class label corresponding to segments included in the new multivariate time series data. If the new class label indicates a potential fault in operation of the machine, a notification may be provided to one or more users.

Unifying semi-supervised approach for machine condition monitoring and fault diagnosis

A method for monitoring a condition of a system or process includes acquiring sensor data from a plurality of sensors disposed within the system (Sand S). The acquired sensor data is streamed in real-time to a computer system (Sand S). A discriminative framework is applied to the streaming sensor data using the computer system (Sand S). The discriminative framework provides a probability value representing a probability that the sensor data is indicative of an anomaly within the system. The discriminative framework is an integration of a Kalman filter with a logistical function (S).

Discriminative hidden kalman filters for classification of streaming sensor data in condition monitoring

A method and system for recognizing (and/or predicting) failures of sensors used in monitoring gas turbines applies a sparse coding process to collected sensor readings and defines the L-1 norm residuals from the sparse coding process as indicative of a potential sensor problem. Further evaluation of the group of residual sensor readings is perform to categorize the group and determine if there are significant outliers ('abnormal data'), which would be considered as more likely associated with a faulty sensor than noisy data. A time component is introduced into the evaluation that compares a current abnormal result with a set of prior results and making the faulty sensor determination if a significant number of prior readings also have an abnormal value. By taking the time component into consideration, the number of false positives is reduced.

Gas turbine sensor failure detection utilizing a sparse coding methodology

A method of predicting an amount of power that will be generated by a solar power plant at a future time includes: forecasting a value of a data variable at the future time that is likely to affect the ability of the solar power plant to produce electricity (S); computing a plurality of features from prior observed amounts of power generated by the power plant during different previous durations (S); determining a trending model from the computed features and the forecasted value (S); and predicting the amount of power that will be generated by the power plant at the future time from the determined model (S).

Combining multiple trending models for photovoltaics plant output forecasting

A method for solar forecasting includes receiving a plurality of solar energy data as a function of time of day at a first time, forecasting from the solar energy data a mode, where the mode is a sunny day, a cloudy day, or an overcast day, and the forecast predicts the mode for a next solar energy datum, receiving the next solar energy datum, updating a probability distribution function (pdf) of the next solar energy datum given the mode, updating a pdf of the mode for the next solar energy datum from the updated pdf of the new solar energy datum given the mode, forecasting a plurality of future unobserved solar energy data from the updated pdf of the mode, where the plurality of future unobserved solar energy data and the plurality of solar energy data have a Gaussian distribution for a given mode determined from training data.

Solar power forecasting using mixture of probabilistic principal component analyzers

A method for predicting failure modes in a machine includes learning () a multivariate Gaussian distribution for each of a source machine and a target machine from data samples from one or more independent sensors of the source machine and the target machine, learning () a multivariate Gaussian conditional distribution for each of the source machine and the target machine from data samples from one or more dependent sensors of the source machine and the target machine using the multivariate Gaussian distribution for the independent sensors, transforming () data samples for the independent sensors from the source machine to the target machine using the multivariate Gaussian distributions for the source machine and the target machine, and transforming () data samples for the dependent sensors from the source machine to the target machine using the transformed independent sensor data samples and the conditional Gaussian distributions for the source machine and the target machine.

Transferring failure samples using conditional models for machine condition monitoring

Systems and methods construct a 2 dimensional (2D) gas concentration map (image) from data from one or more laser based absorption projection paths. Embodiments use a Bayesian approach to construct a 2D gas concentration map of a combustion region cross section plane by modeling the 2D map as a Gaussian Process (GP). The GP models the global correlation among all pixels in the 2D map. Data from one or more projection paths is propagated to all pixels in the map instead of just local pixels. The correlation among map pixels is used to propagate projection path data from map pixels traversed by a projection path to other map pixels.

Bayesian approach for gas concentration reconstruction based on tunable diode laser absorption spectroscopy

A generalized pattern recognition is used to identify faults in machine condition monitoring. Pattern clusters are identified in operating data. A classifier is trained using the pattern clusters in addition to annotated training data. The operating data is also used to cluster the signals in the operating data into signal clusters. Monitored data samples are then classified by evaluating confidence vectors that include substitutions of signals contained in the training data by signals in the same signal clusters as the signals contained in the training data.

Generalized pattern recognition for fault diagnosis in machine condition monitoring

A machine fault diagnosis system is provided. The 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. The resulting decision boundary closely approximates the underlying real separation boundary between the fault and normal conditions.

Supervised fault learning using rule-generated samples for machine condition monitoring

A method for predicting sensor output values of a sensor monitoring system, includes providing a set test input values to a system of sensors, and one or more known sensor output values from the sensor system, where other sensor output values are unknown, calculating, for each unknown sensor output value, a predictive Gaussian distribution function from the test input values and the known output sensor values, and predicting each unknown output yby integrating over a product of the predictive Gaussian distribution function and a conditional Gaussian distribution of the unknown output sensor values with respect to the test input values and other unknown output sensor values. A mean and covariance of the predictive Gaussian distribution function are determined from a training phase, and a hyperparameter of the conditional Gaussian distribution are determined by another training phase.

System and method for conditional multi-output regression for machine condition monitoring

A method for predicting sensor output values of a machine sensor monitoring system includes providing a set of input sensor data X and a set of output sensor data Y for a plurality of sensors the monitor the performance of a machine, learning a functional relationship that maps the input sensor data to the output sensor data by maximizing a logarithm of a marginalized conditional probability function P(Y|X) where a dependence of the output sensor data Y with respect to unknown hidden machine inputs u has been marginalized, providing another set of input sensor data X', and calculating expected values of the output sensor data Y&#x2032; using the input sensor data X&#x2032; and the marginalized conditional probability function P(Y|X&#x2032;), where the calculated expectation values reflect the dependence of the output sensor data Y&#x2033; with respect to the unknown hidden machine inputs u.

System and method for modeling conditional dependence for anomaly detection in machine condition monitoring

In a machine condition monitoring technique, related sensors are grouped together in clusters to improve the performance of state estimation models. To form the clusters, the entire set of sensors is first analyzed using a Gaussian process regression (GPR) to make a prediction of each sensor from the others in the set. A dependency analysis of the GPR then uses thresholds to determine which sensors are related. Related sensors are then placed together in clusters. State estimation models utilizing the clusters of sensors may then be trained.

Method and apparatus for creating state estimation models in machine condition monitoring

In a machine condition monitoring technique, a sensor reading is filtered using a switching Kalman filter. Kalman filters are created to describe separate modes of the signal, including a steady mode and a non-steady mode. For each new observation of the signal, a new mode is estimated based on the previous mode and state, and a new state is then estimated based on the new mode and the previous mode and state. In the steady mode, evolution covariances of both the observed signal and the rate of change of that signal are low. In the non-steady mode, the evolution covariance of the observed signal is set to a higher value, permitting the observed signal to vary widely, while the evolution covariance of the rate of change of the signal is maintained at a low level.

Robust filtering and prediction using switching models for machine condition monitoring

Machine condition monitoring on a system utilizes a wireless sensor network to gather data from a large number of sensors. The data is processed using a multivariate statistical model to determine whether the system has deviated from a normal condition. The wireless sensor network permits the acquisition of a large number of distributed data points from plural system modalities, which, in turn, yields enhanced prediction accuracy and a reduction in false alarms.

Multivariate analysis of wireless sensor network data for machine condition monitoring

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.

Growing community of inventors

Plainsboro, NJ, United States of America

Chao Yuan