An apparatus and method for predicting distress on a physical component. The method can include obtaining distress data. The distress data can be used to determine a distress rank. The distress rank can be compared to a distress output provided by a kernel that use parameters related to the physical component. The comparison can result in a prediction model for the physical component.

An unsupervised machine learning model can make prediction on time series data. Variance of time-varying parameters for independent variables of the model may be restricted for continuous consecutive time intervals to minimize overfitting. The model may be used in a control system to control other devices or systems. If predictions for the control system are for a higher granularity time interval than the current mode, the time-varying parameters of the model are modified for the higher granularity time interval.

A hybrid machine-learning and simulation-based system provides forecasting for an energy system. The system predicts day-ahead and real-time supply and demand, and prices of energy, and generates inputs to an optimization algorithm performed by an Independent System Operator (ISO) that affects behavior of electricity generators and electricity consumers to improve the economic efficiency of electricity grids, and reduce harmful emissions.

The present disclosure describes system and methods for inferring energy usage at multiple levels of granularity. One embodiment describes an industrial automation system including a first industrial automation component, a first sensor coupled to the first industrial automation component, in which the first sensor measures a first amount of power supplied to the first industrial automation component, a second industrial automation component that couples to the first industrial automation component, and an industrial control system that infers energy usage by the first industrial automation component and the second industrial automation component based at least in part on the first amount of power supplied to the first industrial automation component.

The current subject matter relates to controlling insects and includes a trap that can vary operation based at least on a target insect. For example, different insects are attracted to and/or repelled by different things, such as gases, orders, lights, and sounds. In general, things that attract a given insect are referred to as attractants and things that repel a given insect are referred to as repellents. By varying the operation of the trap, different attractants and repellants can be used to attract and/or repel specific insects. In addition, these attractants and repellants and/or characteristics thereof can these be varied dynamically based on, for example, time of day, proximity of individuals to the trap, proximity of insects to the trap, geographic location, and the like. Related apparatus, system, articles, and techniques are also described.

A grid independent operation control unit includes a load current estimator to estimate a load current supplied to stand-alone power system in accordance with an output current of the inverter and an output voltage, and a feedback controller configured to PWM control the inverter at a duty ratio feedback calculated to cause the inverter to output an output voltage command value in accordance with the output voltage and the load current. The feedback controller is configured to PWM control the inverter at a duty ratio feedback calculated for output of a normalized output voltage command value obtained by normalizing the output voltage command value with the DC bus voltage in accordance with normalized output voltage obtained by normalizing the output voltage with the DC bus voltage and normalized load current obtained by normalizing the load current with the DC bus voltage.

Prediction of electrical power system behavior, and related systems, apparatuses, and methods are disclosed. A controller includes a data storage device configured to store model data for time points of a time period of operation. The controller also includes a processor configured to determine current data for time points of a current time period of operation. The current time period corresponds to an early portion of the time period of the model data. The controller is also configured to fit the model data to the current data to produce predicted data, a future portion of the predicted data corresponding to time points occurring after the early portion of the time period of the model data. The controller is further configured to determine values for a set of control variables to effectuate a change to operation of the electrical power system based on the future portion of the predicted data.

Systems, devices, media, and methods are presented for controlling remote equipment in a network. A switch heater control system includes a weather modeling function. The system periodically obtains weather data according to a predetermined time interval. Based on the closest weather data set, the weather modeling function generates a hyperlocal forecast associated with each switch heater location. The system includes an active snowfall mode and a maintenance mode that controls heating based on an estimate of local snow depth, adjusted for wind conditions and passing trains. When the hyperlocal forecast indicates heating is required, the system calculates a melt duration, starts a timer, and transmits a start signal to the switch heater.

A computer-implemented method for determining a control trajectory for a robotic device. The method includes: performing an information theoretic model predictive control applying a control trajectory sample prior in each time step to obtain a control trajectory for a given time horizon; determining the control trajectory sample prior depending on a data-driven trajectory prediction model which is trained to output a control trajectory sample as the control trajectory sample prior based on an actual state of the robotic device.

A computer-implemented method for determining an appropriate control strategy for a mobile agent for an environment with one or more dynamic objects. The method includes: providing a number of different scenarios wherein to each of the scenarios a number of dynamic objects is associated, wherein for each of the scenarios, each of the dynamic objects is associated with a start, a goal and a behavior specification; providing a number of control strategy candidates for the mobile agent; benchmarking each of the control strategy candidates in any of the scenarios; selecting the control strategy for the mobile agent depending on the result of the benchmarking of the control strategy candidates.