基于粒子群人工蜂群算法的灌区渠-塘-田优化调配耦合模型

为了缓解农业水资源供需矛盾,减少南方季节性干旱造成的损失,该文结合南方灌区渠、塘、田地复杂的水量转化关系,建立了以灌溉区域效益最大为目标的渠-塘优化调控与田间多种作物优化配水相结合的耦合模型;根据该模型的特点,提出了模型求解的粒子群-人工蜂群混合算法。将模型应用于漳河灌区,并与常规的渠-塘调控和作物配水模型和只考虑作物优化配水模型相比较。结果表明,渠-塘-田地协同调配效果显著,耦合模型计算所得特别枯水年(降雨频率为95%)下灌溉用水效益比采用其他2种模型分别提高了20.7%和6.9%。粒子群-人工蜂群混合算法能快速求解该优化调配耦合模型,有利于解决多水源、长距离输配水、库塘共同调控等复杂情况下的高效用水模型的求解问题。

Coupled allocation model for optimizing water in canal-pond-field based on artificial bee colony and particle swarm algorithm

Abstract: With the development of Chinese society and economy, contradiction between water supply and demand has become increasingly prominent. Consequently, the problem of agricultural water shortage becomes more serious. What's more, seasonal droughts have frequently taken place in the south of China in recent years, resulting in a serious impact on agricultural production. In order to alleviate the contradiction between water supply and demand of agricultural, to reduce losses caused by seasonal droughts, we proposed a model to optimize the allocation of water resources. Based on the consideration of the complex relationship of water conversion between the canals, ponds and fields in the southern irrigation area and the regulatory role of ponds which is less taken into consideration in the models built before, canal-pond optimal regulation has been proposed. Water flowed into ponds from canals in lower-intensity periods, which were used to irrigate crops in peak periods in the mode of canal-pond optimal regulation. Then an optimal operation model for coupling canal-pond regulation and water allocation between crops has been set up, with the goal of maximizing the economic benefits in the whole irrigated region. Channel water diversion and crop irrigation water at each period were treated as decision variables in the model. The responses of different crops to water deficit during the same period, the responses of the same crop to water deficit in different periods and the regulatory role of ponds were all taken into consideration in the model. Problems solved by the model were high-dimensional, complex, non-linear optimization problems. And the PSO-ABC hybrid algorithm was used to solve the model according to the characteristics of the model. Artificial bee colony algorithm is one of the current best evolutionary algorithms with advantages of simple principle, easy implementation, less parameters and quick convergence speed. But it still suffered from the problems of premature convergence when it came to the high-dimensional complex optimization problem. However, the particle swarm optimization algorithm has a strong ability to jump out of the local extremum. Then the PSO-ABC hybrid algorithm was proposed and used to solve these problems. The coupled model was applied to Zhanghe irrigation region, and compared with the following two models. In model 1, ponds supplied water first and then canals and water distribution between crops accorded to the proportion of crop water demand, which meant that neither canal-pond optimal regulation nor water allocation optimization between crops was considered. In model 2, ponds also supplied water first and only water allocation optimization between crops was considered. The results showed that the proposed model was remarkable in optimizing aspect. In three typical years, the number of water shortage periods decreased significantly. And the total water shortage periods were 91.4% lower than that of the model 1. In special dry years (rainfall for rate 95%), agricultural benefits were 20.7% and 6.9% higher than that of the other two models, respectively. In dry years (rainfall for rate 75%), agricultural benefits were 10.3% and 3.2% higher than that of the other two models, respectively. PSO-ABC hybrid algorithm is less time consuming, and it has advantages to solve the optimal model in consideration of multiple water sources, long distance water transmission and reservoir -pond co-regulation.