Reservoir downstream water level prediction method based on deep learning model

Abstract

The invention discloses a reservoir downstream water level prediction method based on a deep learning model. The method comprises the following steps: screening downstream water level related factorsby adopting a maximum information coefficient; optimizing by adopting a genetic algorithm on the basis of correlation analysis to obtain an optimal feature combination among single correlation factors; constructing a deep learning model (CNNLSTM) based on a convolutional neural network and a long-term and short-term memory network by taking the optimal feature combination of the downstream water level related factors as input; and training a CNNLSTM model weight variable by adopting an Adam gradient optimization algorithm, and taking the trained CNNLSTM as a reservoir downstream water level prediction model. According to the prediction method, the correlation factors of the downstream water level are finely considered, the feature combination of the correlation factors is optimized, the deep learning prediction model is adopted, the prediction precision of the downstream water level is effectively improved, and the prediction method plays a crucial role in accurately calculating the generated output in reservoir dispatching.

Description

Technical field [0001] The present invention relates to the technical field prediction downstream of the reservoir water level, particularly to a downstream water reservoir depth prediction method based learning model. Background technique [0002] Hydroelectric energy is a clean, cheap and renewable green energy. Accurately calculate the power output of reservoir reservoir operation is a necessary part of great significance for the realization of flood control, power generation, water supply and comprehensive benefits such as shipping reservoir. Downstream water level is one of the factors related to the calculation of output power. Currently, reservoir operation, the discharged volume of investigation usually tail water level downstream water flow curve. However, the end of stage-discharge curves are generally built in the early libraries available, along with reservoir operation for many years, the relationship between tailwater level (ie, the downstream water level) and Disch...

AI Technical Summary

Problems solved by technology

[0006] (1) The downstream water level obtained traditionally through the tail water level and flow relationship curve has errors, which has an adverse effect on reservoir regulation

[0007] (2) Recalibrating the relationship curve of tailwater level and flow can reduce the error of downstream water level, but it is time-consuming, laborious and expensive, and it is not an efficient method

[0008] (3) Although the hydrodynamic simulation method can obtain the downstream water level with relatively high accuracy, it has the problems of complex modeling and time-consuming solution

[0009] (4) The accuracy of ordinary machine learning models to predict downstream water levels is limited

[0013] (3) The deep learning model is widely used in the fields of speech recognition and image processing, but it is rarely used in the field of water level prediction in the downstream of the reservoir, and the reference literature is limited

Images