Carry-over storage year-end fluctuating level prediction method and system

Abstract

The invention discloses a carry-over storage year-end fluctuating level prediction method and system. The method comprises the following steps of: 1, establishing a carry-over storage year-end fluctuating level prediction model, and storing the carry-over storage year-end fluctuating level prediction model in a model library; 2, optimizing a scheduling target through a prediction server according to the maximum total cascade energy, and calculating and selecting a corresponding carry-over storage year-end fluctuating level prediction model; and 3, calling a model solution algorithm in an algorithm library through the prediction server, solving the carry-over storage year-end fluctuating level prediction model, and obtaining the year-end fluctuating level rule. Compared with the prior art, the method has the advantages that the solution process is simple, the benefits in the later scheduling period after the carry-over storage year-end are considered, and the accuracy of the prediction method can be improved; and compared with a BP neural network and other methods, the method is suitable for a reservoir with short runoff data series.

Description

technical field [0001] The invention relates to a method and system for predicting year-end drawdown water levels of multi-year regulating reservoirs, and belongs to the field of cascaded reservoir group optimization dispatching. Background technique [0002] According to the existing data, there are few studies on the year-end drawdown level of multi-year regulating reservoirs at home and abroad. On the one hand, multi-year regulating reservoirs are rare; on the other hand, the current dispatching system usually evaluates the quality of dispatching in units of years, and does not fully consider the influence of multi-year regulating reservoir year-end water levels on future dispatching. Therefore, this study Not enough attention has been paid. In recent years, with the large-scale development of cascade hydropower stations in the basin, the number of multi-year regulating reservoirs has gradually increased, and hydropower has gradually participated in the competition in th...

AI Technical Summary

Problems solved by technology

In view of the above problems, Zhang Wenyi (2006) improved the method. On the basis of using stepwise regression to find out the influencing factors of the year-end drawdown level, the genetic algorithm and BP neural network algorithm were used to carry out the multi-year regulation of the year-end drawdown level of the reservoir. In order to predict, since the training of artificial intelligence algorithms requires a large amount of sample data, it is difficult to meet the training requirements for reservoirs with short runoff data series

Zou Jin (2006) established a fuzzy multi-objective decision-making model to determine the year-end drawdown level of multi-year regulating reservoirs in cascade reservoirs based on the analysis of the relationship between the power generation of hydropower stations for multi-year regulating reservoirs and the storage capacity of the reservoirs. Yes, and the solution process is cumbersome

Guo Xiaming (2003) proposed a long-term optimal dispatching model of reservoirs considering the time-of-use electricity price in both peak and dry seasons. Under the set year-end drawdown level, the electricity price factor was added to determine the optimal power generation strategy with the largest power generation. However, this method Failure to take into account the post-regulation benefits of the multi-year regulation reservoir after the end of the year makes the model lack of basis for formulating the year-end drawdown level

Images