A vehicle body side system has a second data communication unit that receives road surface data transmitted from a first data communication unit, a storage unit that stores teacher data, a road surface determination unit that determines a road surface state on a traveling road surface of a vehicle based on the road surface data and the teacher data, and a tire identification unit that identifies a predetermined type as an identification target and identifies which content the tire corresponds to in the type. The tire identification unit identifies the tire based on the identification data transmitted from the tire side device, and the road surface determination unit determines the road surface state using a teacher data corresponding to the tire identification result, when the road surface determination unit determines the road surface state.

A movable system for automatically monitoring the correlated wind and temperature filed of a bridge, including a bridge monitoring subsystem, a cloud server, and a client. The system monitors the meteorological parameters of a bridge surface and a temperature of a bridge structure, performs data analysis and processing on a cloud server, and performs visual data interaction by using a client. A bridge surface-specific meteorological parameter monitoring module is movable, such that the location of the sensor for meteorological data monitoring can be adjusted at any time to monitor an entire bridge deck in a time-sharing manner. A lower cantilever structure has an adjustable height, such that the sensor for meteorological data monitoring can track a height of a boundary layer of the bridge surface. A bridge structure-specific temperature monitoring module adopts distributed patch-type temperature sensors, which can detect the temperature of the bridge structure in all directions.

A method for employing a unified semi-supervised deep learning (DL) framework for turbulence forecasting is presented. The method includes extracting historical and forecasted weather features of a spatial region, calculating turbulence indexes to fill feature cubes, each feature cube representing a grid-based 3D region, and building an encoder-decoder framework based on convolutional long short-term memory (ConvLSTM) to model spatio-temporal correlations or patterns causing turbulence. The method further includes employing a dual label guessing component to dynamically integrate complementary signals from a turbulence forecasting network and a turbulence detection network to generate pseudo-labels, reweighing the generated pseudo-labels by a heuristic label quality detector based on KL-Divergence, applying a hybrid loss function to predict turbulence conditions, and generating a turbulence dataset including the predicted turbulence conditions.

A method for employing a unified semi-supervised deep learning (DL) framework for turbulence forecasting is presented. The method includes extracting historical and forecasted weather features of a spatial region, calculating turbulence indexes to fill feature cubes, each feature cube representing a grid-based 3D region, and building an encoder-decoder framework based on convolutional long short-term memory (ConvLSTM) to model spatio-temporal correlations or patterns causing turbulence. The method further includes employing a dual label guessing component to dynamically integrate complementary signals from a turbulence forecasting network and a turbulence detection network to generate pseudo-labels, reweighing the generated pseudo-labels by a heuristic label quality detector based on KL-Divergence, applying a hybrid loss function to predict turbulence conditions, and generating a turbulence dataset including the predicted turbulence conditions.

Data integration and distribution systems. A system includes a graphical user interface (GUI). Weather and market data are collected. A weather symbology including symbol elements linked to segments of the collected weather data and rules for generating weather symbology instructions are stored. The GUI is generated for display on a user device. A weather symbology instruction is determined based on at least one requested symbol element indicated in a weather data request and the rules. A weather forecast dataset is created from among the collected weather data based on the weather symbology instruction. A presentation package comprising the weather forecast dataset and the collected market data is generated such that the weather forecast dataset is integrated with the collected market data. The presentation package is presented on the GUI and updated concurrent with changes at least one of the weather data, the market data and user input.

Data integration and distribution systems. A system includes a graphical user interface (GUI). Weather and market data are collected. A weather symbology including symbol elements linked to segments of the collected weather data and rules for generating weather symbology instructions are stored. The GUI is generated for display on a user device. A weather symbology instruction is determined based on at least one requested symbol element indicated in a weather data request and the rules. A weather forecast dataset is created from among the collected weather data based on the weather symbology instruction. A presentation package comprising the weather forecast dataset and the collected market data is generated such that the weather forecast dataset is integrated with the collected market data. The presentation package is presented on the GUI and updated concurrent with changes at least one of the weather data, the market data and user input.

An object of the present invention is to improve the accuracy of prediction in a technique for predicting natural energy power generation amount, solar radiation amount or wind speed by using a statistical method based on machine learning. In order to achieve this object, provided is a prediction apparatus (10) including a feature value extraction unit (13) that extracts a feature value being a variation in time series from meteorological data from m (m is 2 or more) hours before a target time to the target time, and an estimation unit (first estimation unit (14)) that estimates a natural energy power generation amount, a solar radiation amount, or a wind speed at the target time based on the feature values over plural days.

An object of the present invention is to improve the accuracy of prediction in a technique for predicting natural energy power generation amount, solar radiation amount or wind speed by using a statistical method based on machine learning. In order to achieve this object, provided is a prediction apparatus (10) including a feature value extraction unit (13) that extracts a feature value being a variation in time series from meteorological data from m (m is 2 or more) hours before a target time to the target time, and an estimation unit (first estimation unit (14)) that estimates a natural energy power generation amount, a solar radiation amount, or a wind speed at the target time based on the feature values over plural days.

A single beam frequency modulated continuous wave radar for clear air scatter (CAS) detection and method of monitoring clear air scatterers are provided. CAS monitoring capabilities, including the ability to estimate wind velocity and direction, are obtained using data from a single defined width beam of energy that instead of being averaged is sampled at discrete time steps over a range of altitudes.

A single beam frequency modulated continuous wave radar for clear air scatter (CAS) detection and method of monitoring clear air scatterers are provided. CAS monitoring capabilities, including the ability to estimate wind velocity and direction, are obtained using data from a single defined width beam of energy that instead of being averaged is sampled at discrete time steps over a range of altitudes.