A级数据中心综合能源系统多目标优化设计和调度

Multi-objective Optimization Design and Scheduling of Integrated Energy System in A-level Data Center

采用分布式综合能源系统取代数据中心的传统供能系统,是降低数据中心运行费用的有效措施.考虑A级数据中心的供能可用度,设计了一种分布式综合能源系统,并基于混合整数非线性规划方法,建立了综合经济和环境效益的多目标优化设计和调度模型,选取ε-约束法对相互制约的多目标问题进行处理,在通用代数建模系统(GAMS)中建模并调用LINDOGLOBAL求解器进行优化求解.采用多维偏好分析线性规划决策法从非劣解集中选出最终解,并通过逼近于理想解的排序方法进行验证.算例结果表明,优化后的分布式综合能源系统能够保证A级数据中心的供能可用度,同时在降低年总费用和减排二氧化碳方面均具优势.

Replacing conventional energy supply systems by distributed integrated energy systems is considered as an effective alternative to reduce the operation expense of data centers.Considering the requirements of A-level data center for energy availability,a distributed integrated energy system is proposed,and a model for multi-objective optimization design and dispatch of economic and environmental benefits is established based on the mixed integer nonlinear programming approach.Furthermore,theε-constraint method is applied to solve the competing multi-objective problem,and an optimization program in GAMS using LINDOGLOBAL solver is developed.The final solution is selected from non-inferior solutions by linear programming techniques for multidimensional analysis of preference approach and verified by the technique for order preference by similarity to ideal solution approach.The results demonstrate that the optimized distributed integrated energy system is able to guarantee the energy availability of the A-level data center,and effectively achieves annual total cost saving and carbon dioxide emission reduction.