With a capacity-based network tariff structure, consumers are encouraged to reduce their connection capacity to avoid higher costs. However, overloading beyond the administrative grid connection capacity limit would result in an increased connection capacity, thus prosumers have to pay the increased electricity bill for the rest of the year. Therefore, it is important to optimize the energy generation and consumption profiles of local energy communities (LECs) considering the comfort level of occupants. This work aims to reduce the overloading of the grid connection and increase the utilization of local renewable energy resources (RES) thus avoids being penalized throughout the year due to casual intermittent overloading in peak hours, even once in a year. The present work proposes a novel data-driven flexibility optimizer model for day-ahead scheduling of energy profiles for LECs, considering photovoltaic (PV) generation, heat pump (HPs), and cooling loads. The proposed methodology has been developed to explore the flexibility potentials from a university campus network which includes both electrical and heating/cooling systems in an integrated way. A two-layer optimization strategy is developed, to guard the occupant’s comfort level. Simulation has been performed for complete two months, considering winter and summer scenarios. A peak demand reduction of 16% has been observed with negligible energy usage differences between the proposed and the baseline case. Two types of flexibility indicators are estimated to give a deeper insight into the performance. Economical gains of 9 % and 16 % are estimated depending on the type and voltage level of the connection.

通过基于容量的网络资费结构，鼓励消费者减少其连接容量以避免更高的成本。然而，超过行政电网连接容量限制的超载会导致连接容量增加，因此产消者必须支付一年剩余时间增加的电费。因此，考虑到居住者的舒适度，优化当地能源社区 (LEC) 的能源生产和消费状况非常重要。这项工作旨在减少电网连接的过载并增加当地可再生能源（RES）的利用率，从而避免因高峰时段偶然间歇性过载而全年受到惩罚，甚至一年一次。目前的工作提出了一种新的数据驱动的灵活性优化器模型，用于 LEC 能源配置的日前调度，考虑到光伏 (PV) 发电、热泵 (HP) 和冷却负载。所提出的方法是为了探索大学校园网络的灵活性潜力，该网络以集成方式包括电气和加热/冷却系统。开发了两层优化策略，以保护乘员的舒适度。考虑到冬季和夏季情景，仿真已经进行了整整两个月。已观察到峰值需求减少了 16%，建议和基准案例之间的能源使用差异可以忽略不计。估计有两种类型的灵活性指标可以更深入地了解性能。