Within the national funded project EnVisaGe (www.envisage-wuestenrot.de) a small Plus-Energy district mainly with single-family homes is built in Wüstenrot, Germany. A novel district heating system based on a cold-water heating grid supplies low temperature heat from a near surface geothermal system to the heat pumps with hot water storages in the connected buildings of the district. To reach the Plus-Energy level, highly energy efficient buildings close to passive house standard are built and equipped with large PV systems for onsite solar electricity generation. Some buildings are also equipped with electricity storages providing a capacity between 3 and 5 kWh each. Heat pumps with hot water storage offer the opportunity to shift electricity loads e.g. to periods with high solar radiation. Combined with the small electricity storages, this can significantly increase the PV self-consumption / self-sufficiency rate and help to reduce peak feed-in power to improve the grid compatibility [1] [2]. This complex energy management task requires innovative MPC (model based predictive control) schemes implemented in an appropriate monitoring and control system, which is developed and tested within the project. Apart from this local optimization on a single building level, decentralised heat pumps and batteries in a larger cluster can also deliver controllable electricity sinks as smart grid applications for electricity providers. To test these additional opportunities within the project, the heat pumps and batteries are connected to the virtual power plant of Vattenfall. Within the paper, the developed cloud-based data collection and management system is presented together with the MPC based system. Additionally, monitoring and simulation results are shown.