Abstract: One embodiment can provide a system for detecting anomaly for high-dimensional sensor data associated with one or more machines. During operation, the system can obtain sensor data from a set of sensor associated with one or machines, generate a first set of outputs by using a set of clustering models learned in parallel from the unlabeled sensor data and user-provided partial label information, generate a second set of outputs by using a set of feed-forward neural network (FNN) models learned in parallel from the first set of outputs and the unlabeled sensor data, and determine whether an anomaly is present in the operation of the one or more machines based on the second set of outputs and a user-specified threshold.

Abstract: A system and method are provided for determining a causal inference in a manufacturing process. During operation, the system can receive data associated with a processing system which includes a set of interconnected machines and an associated set of processes. The system can generate, based on the data, a graph indicating flows of outputs between the machines as part of the processes. The system can determine, based on a set of variables, one or more candidate clusters in the graph. The system can perform, based on one or more variables of interest, root cause analysis on the one or more candidate clusters by: applying an additive noise model to prune the one or more candidate clusters from the graph; and determining, based on the pruned graph, a candidate pathway likely to cause an issue in at least one process, thereby facilitating improved efficiency in the processing system.

Abstract: One embodiment can provide a system for detecting anomaly for high-dimensional sensor data associated with one or more machines. During operation, the system can obtain sensor data from a set of sensor associated with one or machines, apply data exploration techniques on the sensor data to automatically process sensor data to identify a subset of feature sensors from the available set of feature sensors, apply an unsupervised machine-learning technique to the identified subset of feature sensors and the target sensor to learn a set of pair-wise univariate models, and determine whether and how an anomaly occurs in the operation of the one or more machines based on the set of pair-wise univariate models.

Abstract: One embodiment of the present invention can provide a system for identifying a root cause of an anomaly event in operation of one or more machines is provided. During operation, the system can obtain sensor data from a set of sensors associated with the one or more machines, convert the sensor data into a set of sensor states, build an optimal DAG based on the set of sensor states to model causal dependency; determining, by using the DAG, a probability of an anomaly state of a target sensor given a state of a direct neighbor sensor, and determining a root cause of the anomaly event associated with the target sensor by back-tracking the anomaly state in the DAG.

Abstract: A system and method are provided to facilitate automated data imputation. During operation, the system generates a cluster model based on raw data obtained from sensors with multiple states, wherein the raw data includes missing values. The system replaces the missing values with first imputed data based on the cluster model. The system iterates, until a predetermined threshold has been reached, through a series of operations which include: updating the cluster model based on most recently imputed data; predicting outliers based on the cluster model; marking the outliers as null values to obtain filtered data; updating the cluster model based on the filtered data; and replacing the null values with second imputed data based on the cluster model.

Abstract: A system is provided for determining causes of faults in a manufacturing system. The system stores data associated with a processing system which includes machines and associated processes, wherein the data includes timestamp information, machine status information, and product-batch information. The system determines, based on the data, a topology of the processing system, wherein the topology indicates flows of outputs between the machines as part of the processes. The system determines information of machine faults in association with the topology. The system generates, based on the machine-fault information, one or more fault parameters which indicates frequency and severity of a respective fault. The system constructs, based on the topology and the machine-fault information, a system model which includes the one or more fault parameters, thereby facilitating diagnosis of the processing system.

Abstract: One embodiment can provide a system for estimating useful life of a load-bearing structure. During operation, the system can perform a degradation measurement on the structure to obtain degradation data for a predetermined time interval, apply a constraint convex regression model to the degradation data, estimate a total useful life (TUL) of the structure based on outputs of the constraint convex regression model, and predicting a remaining useful life (RUL) based on the TUL and a current time.

Abstract: A system is provided for determining a manufacturing network topology and fault propagation information. During operation, the system stores data associated with a processing system which includes machines and associated processes, wherein the data includes timestamp information, machine status information, and product-batch information. The system determines, based on the data, a network topology which corresponds to the processing system, wherein the network topology indicates flows of outputs between the machines as part of the processes. The system determines utilization information of a plurality of the machines of the processing system. The system displays one or more of the flows of outputs based on the utilization information, thereby facilitating diagnosis of the processing system.

Abstract: A system is provided for determining causes of faults in a manufacturing system. The system stores data associated with a processing system which includes machines and associated processes, wherein the data includes timestamp information, machine status information, and product-batch information. The system determines, based on the data, a topology of the processing system, wherein the topology indicates flows of outputs between the machines as part of the processes. The system determines information of machine faults in association with the topology. The system generates, based on the machine-fault information, one or more fault parameters which indicates frequency and severity of a respective fault. The system constructs, based on the topology and the machine-fault information, a system model which includes the one or more fault parameters, thereby facilitating diagnosis of the processing system.

Abstract: A system and method are provided for determining a causal inference in a manufacturing process. During operation, the system can receive data associated with a processing system which includes a set of interconnected machines and an associated set of processes. The system can generate, based on the data, a graph indicating flows of outputs between the machines as part of the processes. The system can determine, based on a set of variables, one or more candidate clusters in the graph. The system can perform, based on one or more variables of interest, root cause analysis on the one or more candidate clusters by: applying an additive noise model to prune the one or more candidate clusters from the graph; and determining, based on the pruned graph, a candidate pathway likely to cause an issue in at least one process, thereby facilitating improved efficiency in the processing system.

Abstract: One embodiment can provide a system for detecting faults in a machine. During operation, the system can obtain a dynamic signal associated with the machine, apply one or more signal-processing techniques to the dynamic signal to obtain frequency, amplitude, and/or time-frequency information associated with the dynamic signal, extract motion-insensitive features from the obtained frequency, amplitude, and/or time-frequency information associated with the dynamic signal, and determine whether a fault occurs in the machine based on the extracted features.

Abstract: One embodiment can provide a system for detecting anomaly for high-dimensional sensor data associated with one or more machines. During operation, the system can obtain sensor data from a set of sensor associated with one or machines, generate a first set of outputs by using a set of clustering models learned in parallel from the unlabeled sensor data and user-provided partial label information, generate a second set of outputs by using a set of feed-forward neural network (FNN) models learned in parallel from the first set of outputs and the unlabeled sensor data, and determine whether an anomaly is present in the operation of the one or more machines based on the second set of outputs and a user-specified threshold.

Abstract: One embodiment of the present invention can provide a system for identifying a root cause of an anomaly event in operation of one or more machines is provided. During operation, the system can obtain sensor data from a set of sensors associated with the one or more machines, convert the sensor data into a set of sensor states, build an optimal DAG based on the set of sensor states to model causal dependency; determining, by using the DAG, a probability of an anomaly state of a target sensor given a state of a direct neighbor sensor, and determining a root cause of the anomaly event associated with the target sensor by back-tracking the anomaly state in the DAG.

Abstract: One embodiment can provide a system for detecting anomaly for high-dimensional sensor data associated with one or more machines. During operation, the system can obtain sensor data from a set of sensor associated with one or machines, apply data exploration techniques on the sensor data to automatically process sensor data to identify a subset of feature sensors from the available set of feature sensors, apply an unsupervised machine-learning technique to the identified subset of feature sensors and the target sensor to learn a set of pair-wise univariate models, and determine whether and how an anomaly occurs in the operation of the one or more machines based on the set of pair-wise univariate models.

Abstract: One embodiment can provide a system for estimating useful life of a load-bearing structure. During operation, the system can perform a degradation measurement on the structure to obtain degradation data for a predetermined time interval, apply a constraint convex regression model to the degradation data, estimate a total useful life (TUL) of the structure based on outputs of the constraint convex regression model, and predicting a remaining useful life (RUL) based on the TUL and a current time.

Abstract: One embodiment can provide a system for detecting faults in a machine. During operation, the system can obtain a dynamic signal associated with the machine, apply one or more signal-processing techniques to the dynamic signal to obtain frequency, amplitude, and/or time-frequency information associated with the dynamic signal, extract motion-insensitive features from the obtained frequency, amplitude, and/or time-frequency information associated with the dynamic signal, and determine whether a fault occurs in the machine based on the extracted features.

Abstract: Systems and methods are provided for estimating power breakdowns for a set of one or more appliances inside a building by exploiting a small number of power meters and data indicative of binary power states of individual appliances of such set. In one aspect, a breakdown estimation problem is solved within a tree configuration, and utilizing a single power meter and data indicative of binary power states of a plurality of appliances. Based at least in part on such solution, an estimation quality metric is derived. In another aspect, such metric can be exploited in a methodology for optimally placing additional power meters to increase the estimation certainty for individual appliances to a desired or intended level. Estimated power breakdown and energy breakdown—individually or collectively referred to as consumption breakdown—rely on measurements and numerical simulations, and can be evaluated in exemplary electrical network utilizing binary sensors.

Abstract: Systems and methods are provided for estimating power breakdowns for a set of one or more appliances inside a building by exploiting a small number of power meters and data indicative of binary power states of individual appliances of such set. In one aspect, a breakdown estimation problem is solved within a tree configuration, and utilizing a single power meter and data indicative of binary power states of a plurality of appliances. Based at least in part on such solution, an estimation quality metric is derived. In another aspect, such metric can be exploited in a methodology for optimally placing additional power meters to increase the estimation certainty for individual appliances to a desired or intended level. Estimated power breakdown and energy breakdown—individually or collectively referred to as consumption breakdown—rely on measurements and numerical simulations, and can be evaluated in exemplary electrical network utilizing binary sensors.

Abstract: Systems and methods are provided for estimating power breakdowns for a set of one or more appliances inside a building by exploiting a small number of power meters and data indicative of binary power states of individual appliances of such set. In one aspect, a breakdown estimation problem is solved within a tree configuration, and utilizing a single power meter and data indicative of binary power states of a plurality of appliances. Based at least in part on such solution, an estimation quality metric is derived. In another aspect, such metric can be exploited in a methodology for optimally placing additional power meters to increase the estimation certainty for individual appliances to a desired or intended level. Estimated power breakdown and energy breakdown—individually or collectively referred to as consumption breakdown—rely on measurements and numerical simulations, and can be evaluated in exemplary electrical network utilizing binary sensors.

Abstract: Systems and methods are provided for estimating power breakdowns for a set of one or more appliances inside a building by exploiting a small number of power meters and data indicative of binary power states of individual appliances of such set. In one aspect, a breakdown estimation problem is solved within a tree configuration, and utilizing a single power meter and data indicative of binary power states of a plurality of appliances. Based at least in part on such solution, an estimation quality metric is derived. In another aspect, such metric can be exploited in a methodology for optimally placing additional power meters to increase the estimation certainty for individual appliances to a desired or intended level. Estimated power breakdown and energy breakdown—individually or collectively referred to as consumption breakdown—rely on measurements and numerical simulations, and can be evaluated in exemplary electrical network utilizing binary sensors.