Application of robust optimized spatiotemporal load management of data centers for renewable curtailment mitigation

Renewable energy sources play a crucial role in the provision of power supply in active distribution networks. In networks that rely on renewable energy sources, it may become imperative to curtail the power generated by such units due to a variety of factors. The curtailment of renewable energy generation can be attributed to various factors such as overgeneration, voltage limitations at the network bus, and power limitations at the network line. As per the specified parameters, it is possible for the power generation of renewable units to exceed the load consumption at certain intervals throughout the operational duration. Hence, it is essential to devise a solution that mitigates the frequency and duration of power generation curtailments of renewable units. It is feasible to redistribute overgeneration of renewable energy sources from periods of excess to periods of insufficient capacity to meet demand. This study demonstrates that the provision of power to loads from the sub-distribution substation experiences a reduction in power supply and cost. It is feasible to transmit surplus power from a single bus to multiple other buses. The term used to refer to the ability to transfer power across different locations and time periods is known as spatiotemporal flexibility. The presence of data centers in distribution networks enables a reduction in costs and the frequency of power interruptions associated with renewable energy sources, through the utilization of spatiotemporal flexibility. The objective of this paper is to present a methodology for achieving temporal and spatial flexibility in data centers, with the aim of minimizing costs and curtailing the output generation of renewable energy units. Undoubtedly, this approach can also be executed by incorporating mobile batteries; however, it is noteworthy that mobile batteries entail additional costs on the network. Consequently, the implementation of the spatiotemporal flexibility approach in data centers has been adopted as a substitute for the utilization of mobile batteries. This approach, also referred to as virtual battery, effectively eliminates the costs associated with the use of mobile batteries. This study examines overgeneration of renewable units, bus voltage limitations, and line power restrictions on the standard IEEE 33-bus network. Three scenarios are considered, each involving changes in data center power in response to varying load profiles. The implementation of equitable power distribution across data centers, both in terms of temporal and spatial dimensions, as well as the adoption of flexible spatiotemporal flexibility scenario for data centers, have been observed. The findings of this study indicate that the approach employed in this paper leads to a decrease in costs and the output power of renewable units in the examined scenarios. The implementation of spatiotemporal flexibility mode has resulted in a reduction of power supply expenses ranging from 12 to 37 percent when compared to the basic case. Ultimately, the sensitivity analysis pertaining to the placement of data centers and the associated uncertainty index is presented.