Abstract: A building management system includes connected equipment configured to measure a plurality of monitored variables and a predictive diagnostics system configured to receive the monitored variables from the connected equipment; generate a probability distribution of the plurality of monitored variables; determine a boundary for the probability distribution using a supervised machine learning technique to separate normal conditions from faulty conditions indicated by the plurality of monitored variables; separate the faulty conditions into sub-patterns using an unsupervised machine learning technique to generate a fault prediction model, each sub-pattern corresponding with a fault, and each fault associated with a fault diagnosis; receive a current set of the monitored variables from the connected equipment; determine whether the current set of monitored variables correspond with one of the sub-patterns of the fault prediction model to facilitate predicting whether a corresponding fault will occur; and determin

Abstract: In various embodiments, a relevance application quantifies how a trained neural network operates. In operation, the relevance application generates a set of input distributions based on a set of input points associated with the trained neural network. Each input distribution is characterized by a mean and a variance associated with a different neuron included in the trained neural network. The relevance application propagates the set of input distributions through a probabilistic neural network to generate at least a first output distribution. The probabilistic neural network is derived from at least a portion of the trained neural network. Based on the first output distribution, the relevance application computes a contribution of a first input point included in the set of input points to a difference between a first output point associated with a first output of the trained neural network and an estimated mean prediction associated with the first output.

Abstract: A method for quantizing a neural network includes modeling noise of parameters of the neural network. The method also includes assigning grid values to each realization of the parameters according to a concrete distribution that depends on a local fixed-point quantization grid and the modeled noise and. The method further includes computing a fixed-point value representing parameters of a hard fixed-point quantized neural network.

Abstract: The present invention discloses a node flow optimization distribution method for improving the accuracy of transient hydraulic simulation of a water supply in-series pipeline. The present invention optimizes the flow distribution coefficients of intermediate nodes to minimize the impact thereof on the calculation and analysis of transient flow. Further, the simplified error generated by the node flow distribution can be quantified and evaluated by the control threshold of the simplified errors to achieve effective control of the simplified process. In addition, the simplified operation of the method of the present invention is carried out sequentially from the intermediate node with the smallest simplified error, which effectively overcomes the potential defect of the conventional node flow distribution that leads to a significant reduction in the accuracy of the model, and can ensure the reliability and accuracy of the simplified operation of the same-diameter in-series pipeline.

Abstract: An early earthquake detection method may comprise acquiring a frame image from a camera; acquiring a vibration signal from the frame image; removing a noise signal due to vibration of the camera from the vibration signal; acquiring a motion signal obtained by magnifying subtle motions from the noise signal-removed vibration signal; extracting vibration characteristics from the motion signal; estimating an occurrence of an earthquake by extracting a peak signal from the vibration characteristics; and determining whether an earthquake occurs by receiving earthquake estimation information from at least one other camera located within a certain range.

Abstract: A system and method for analyzing structural heath data includes a structural body, structural health sensors, and first and second computer systems. The structural health sensors are configured to sense data regarding structures of the structural body. The first computer system is configured to collect the sensed data as the structural health data. The second computer system that includes a user interface and display, and is configured to receive the structural health data and provide interactive transformational analysis of the structural health data. The interactive transformational analysis provides, on the display of the second computer system, a visual representation of the structural health data over time.

Abstract: A method for modeling a subsurface volume includes receiving a plurality of ordered seismic images including representations of objects in the subsurface volume, generating flow fields based on a difference between individual images of the plurality of ordered seismic images, and identifying the objects in the seismic images based on the flow fields and the plurality of ordered seismic images.

Abstract: The disclosed embodiments relate to a system that compactly stores time-series sensor signals. During operation, the system receives original time-series signals comprising sequences of observations obtained from sensors in a monitored system. Next, the system formulizes the original time-series sensor signals to produce a set of equations, which can be used to generate synthetic time-series signals having the same correlation structure and the same stochastic properties as the original time-series signals. Finally, the system stores the formulized time-series sensor signals in place of the original time-series sensor signals.

Abstract: Described herein are embodiments for systems and methods to incorporate skip-gram convolution to extract non-consecutive local n-gram patterns for comprehensive information for varying text expressions. In one or more embodiments, one or more recurrent neural networks are employed to extract long-range features from localized level to sequential and global level via a chain-like architecture. Comprehensive experiments on large-scale datasets widely used for the text classification task were conducted to demonstrate the effectiveness of the presented deep skip-gram network embodiments. Performance evaluation on various datasets demonstrates that embodiments of the skip-gram network are powerful for general text classification task set. The skip-gram models are robust and may be generalized well on different datasets, even without tuning the hyper-parameters for specific dataset.

Abstract: Systems and methods include a method for generating a high-resolution advanced three-dimensional (3D) transient model that models multiple wells by integrating pressure transient data into a static geological model. A crude 3D model is generated from a full-field geological model that models production for multiple wells in an area. A functional 3D model is generated from the crude 3D model. An intermediate 3D model is generated by calibrating the functional 3D model with single-well data. An advanced 3D transient model is generated by calibrating multi-well data in the functional 3D model.

Abstract: In an approach for forecasting environmental occlusion events, a processor receives a spatio-temporal zone of interest. A processor collects data associated with the spatio-temporal zone of interest. A processor builds a machine-learning model using the data. A processor generates an occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and enriched mathematical operators. A processor outputs the occlusion probability map.

Abstract: A method for diagnosing analog circuit fault based on vector-valued regularized kernel function approximation, includes steps of: step (1) acquiring a fault response voltage signal of an analog circuit; step (2) carrying out wavelet packet transform on the collected signal, and calculating a wavelet packet coefficient energy value as a characteristic parameter; step (3) utilizing a quantum particle swarm optimization algorithm to optimize a regularization parameter and kernel parameter of vector-valued regularized kernel function approximation, and training a fault diagnosis model; and step (4) utilizing the trained diagnosis model to recognize circuit faults.

Abstract: A GNSS data collection system includes a pole mounted GNSS receiver and inclination sensors. A data collection module provides a data collection graphical user interface (GUI) visible on a hand-held data collector computer. The data collector computer is communicably coupled to the GNSS receiver and receives three-dimensional location data and inclination data for the range pole in real-time. A virtual level component uses the inclination data to display on the GUI real-time tilt information in the form of a virtual bubble level indicator. The inclination data and height of the range pole are used to calculate and display horizontal distance and direction to level the GNSS receiver.

Abstract: A system for a region including a processor to receive data including weather and component data. Generate a dataset for the region to identify weather events and a plurality of model forcings converted into tabular form. A machine learning (ML) model for each model forcing for each component uses dataset. Iteratively, for each weather event: identify for each component a corresponding model forcing with weather variables. Generate an output value for ML model corresponding to the identified components to the set of weather variables and update ML model. Receive observed data over time periods of an impending weather (IW) event. Iteratively, for each time period: identify, for each component a corresponding model forcing with weather variables and update ML model. Generate for the updated ML model, an output value predicting a component status as a failed or not failed for the time period.

Abstract: A computing system includes a first hardware element having a first accelerometer and a first gyroscope, and a second hardware element having a second accelerometer and a second gyroscope. The first and second hardware elements are moveable with respect to each other. The computing system recursively generates a result signal indicative of a relative orientation of the first and second hardware elements with respect to each other. The result signal may be generated by generating a first intermediate signal indicative of a angle between the first and second hardware elements based on signals generated by the first and second accelerometers and generating a second intermediate signal indicative of the angle based on signals generated by the first and second gyroscopes. The result signal indicative of the angle may be generated as a weighted sum of the first intermediate signal and the second intermediate signal. At least one of the first and second hardware elements is controlled by on the result signal.

Abstract: A battery degradation evaluation system includes a memory and a processor. The processor is configured to acquire state quantities of a battery mounted at a vehicle, derive probabilities, and evaluate degradation of the battery based on the derivation results. The probabilities are a short-term degradation probability of the battery degrading in a pre-specified short period, a medium-term degradation probability of the battery degrading in a period that is longer than the short period, and a long-term degradation probability of the battery degrading in a period that is longer than the medium period. When a number of the state quantities is smaller, the processor sets a higher weighting for a combined degradation probability for the short period and medium period, which is calculated from the long-term degradation probability. When the number of state quantities is larger, the processor sets a higher weighting for the short-term degradation probability.

Abstract: An apparatus for processing a signal for input to a neural network, the apparatus configured to: receive a signal comprising a plurality of samples of an analog signal over time; determine at least one frame comprising a group of consecutive samples of the signal, wherein the or each frame includes a first number of samples; for each frame, determine a set of correlation values comprising a second number of correlation values, the second number less than the first number, each correlation value of the set of correlation values based on an autocorrelation of the frame at a plurality of different time lags; provide an output based on the set of correlation values corresponding to the or each of the frames for a neural network for one or more of classification of the analog signal by the neural network and training the neural network based on a predetermined classification.

Abstract: Disclosed are an apparatus and method for calibrating analyte data. In an embodiment, a method of calibrating analyte data may include receiving first analyte data measured by a reference device, storing the received first analyte, calculating a calibration value by using an artificial intelligence (AI) calibration model having second analyte data measured by an analyte sensor and the stored first analyte data as inputs, and calculating the final analyte data by incorporating the calculated calibration value into the second analyte data.

Abstract: A method for identifying a fault event in an electric power distribution grid sector including one or more electric loads and having a coupling node with a main grid, at which a grid current adsorbed by said electric loads is detectable. The method allows determining whether a detected anomalous variation of the grid current, adsorbed at the electric coupling node, is due to the start of a characteristic transitional operating period of an electric load or is due to an electric fault.

Abstract: An automated system and method to investigate degradation of cathode materials in batteries via atomistic simulations, and in particular by simulating the creation of atomistic defects in the cathode material, which occurs during charge cycling. A systematic procedure relates the degradation of battery performance metrics to underlying structural changes due to atomic rearrangements within the material, for example through density functional theory simulations. The performance metrics modeled with this approach include the Open Cell Voltage (OCV) as well as the discharge capacity curve.