基于动态规划算法的额勒赛下游水电站防洪优化调度研究

Research on Optimal Flood Control Dispatching of the Lower Stung Russei Chrum Hydropower Station Based on Dynamic Programming Algorithm

水库的防洪调度需考虑很多因素,通常的常规调度无法充分利用水库的调蓄库容,故水库的防洪效益不高。为此,基于动态规划算法(DP),以最大削峰准则及最大防洪安全保证准则对额勒赛下游水电站进行优化调度,并采用动态水位约束及变动离散机制对DP算法进行有效降维。其中以最大削峰准则进行调度时整个调度过程的下泄流量比常规调度均匀,最大下泄流量比常规调度结果小1396m^(3)/s(降低了69.2%),对下游防护对象的安全更有利;以最大防洪安全保证准则进行调度时调洪最高水位比常规调度结果小0.38m,有效地保证了水库的安全;同时降维处理后模型的计算时间降低了40%,有效解决了维数灾问题,有助于梯级水库群联合调度的应用研究。

Many factors need to be considered in the flood control operation of the reservoir.The usual conventional operation can not make full use of the storage capacity of the reservoir,so the flood control benefit of the reservoir is not high.Based on the dynamic programming algorithm(DP),this paper uses the maximum peak-shaving criterion and the maximum flood control safety guarantee criterion to optimize the dispatching of the lower Stung Russei Chrum hydropower station,and uses dynamic water level constraints and variable discrete mechanisms to effectively reduce the dimensionality of the DP algorithm.Among them,when the maximum peak-shaving criterion is used for scheduling,the discharge flow of the entire scheduling process is even than that of conventional scheduling,and the maximum discharge flow is 1396 m^(3)/s(a decrease of 69.2%)less than the conventional scheduling result,which is more beneficial to the safety of downstream protection objects and maximizes flood prevention.The maximum flood control water level is 0.38 msmaller than the conventional method when dispatching according to the maximum flood control safety guarantee criterion,which effectively guarantees the flood control safety of the reservoir.At the same time,the calculation time of the model is reduced by 40%after dimensionality reduction,which effectively solves the problem of dimensionality disaster,which will help the application research of joint dispatch of cascade reservoirs.