Multi-data fusion meteorological prediction early warning method

Abstract

The invention discloses a multi-data fusion meteorological prediction early warning method, and the method comprises the steps: carrying out the assimilation of meteorological observation data througha multi-source 4D-WRF-EnSRF data assimilation system, enabling the assimilation data to serve as the input of a DeepConvLSTMs artificial intelligence prediction network, and achieving the training ofthe network and the real-time rainfall prediction of multiple data sources; finely dividing rainfall intervals, and setting early warning levels; seamlessly dividing the coverage area of each mobilemeteorological station; and planning a mobile meteorological station, a meteorological monitoring center, a multi-means publishing platform and a prediction and early warning information publishing system, and guaranteeing real-time linkage of the system architecture. Multi-data acquisition and fusion are realized, prediction and early warning results are uploaded and issued in real time, a to-be-rescued area is rapidly locked, a rescue scheme is issued, and rescue field communication is ensured; the weather forecasting and early warning timeliness and accuracy are improved, the early warninginformation publishing coverage rate is increased, the weather disaster prevention and reduction service capacity and the rescue capacity are enhanced, and life and property losses caused by rainstormdisasters and secondary derivative disasters of the rainstorm disasters are effectively reduced.

Description

technical field [0001] The invention belongs to the technical field of weather forecasting and artificial intelligence, and in particular relates to a multi-data fusion weather forecasting and early warning method. Background technique [0002] In recent years, the natural primitive ecology has been gradually destroyed, making heavy rain more likely to bring huge derivative disasters, including landslides, landslides, mudslides, and urban waterlogging, which in turn lead to secondary accidents in aviation, shipping, and transportation. Accurate real-time prediction of precipitation and accurate real-time communication of on-site rescue have always been an important task in the field of weather forecast and early warning. If effective forecasting and early warning, timely release of information, and good rescue communications can be achieved, losses can be greatly reduced. [0003] The existing traditional severe convective precipitation forecasting methods can be roughly di...

AI Technical Summary

Problems solved by technology

However, the optical flow estimation step and the radar echo extrapolation step are separated, so it is very difficult to determine the model parameters, and with the rapid reduction and disappearance of the spectral value, there are certain limitations in the prediction of real-time rainfall in the region

Images