Improvement of wind power utilization through flexible operation of data center in wind parks

Wind energy's pivotal role in renewable transitions is hindered by grid integration challenges, notably limited distribution line capacity to handle wind-generated power efficiently. A solution to strengthen the effectiveness of wind power (WP) distribution systems involves integrating data centers (DCs) as flexible loads within WP plants. This approach introduces a novel dynamic load-shifting model aligning DC operations with wind energy availability, considering ambient temperature effects on cooling systems consumption. The model optimizes renewable energy (RE) usage and minimizes grid dependence by adjusting load schedules based on WP fluctuations, temperature changes, and electricity pricing. This model identifies specific controllable loads in DCs and employs adaptive and lexicographic optimization framework using Mixed-Integer Linear Programming to address system uncertainties. A case study highlights the model's effectiveness: WP share compared to not using DC integration surged to 91.3 % during peak wind periods, and WP utilization reached 88.9 % during low wind periods, significantly enhancing RE distribution efficiency compared to conditions without DC integration. Furthermore, the model optimizes cost efficiency by achieving a 52.6 % cost reduction during low-demand periods via strategic load shifting. Moreover, a stochastic solution analysis quantifies the benefit of incorporating uncertainty, revealing cost savings between 3.33 % and 16.63 % over deterministic models.