AS_02_2019

CONTROLLO tecnica Automazione e Strumentazione Marzo 2019 97 conflicting objective function. Information among MPC agents is transmitted and received once in each time interval. In this frame- work, according to the method studied in [5] , the logic/algebraic and dynamical system model is described in the form of a mixed logic dynamical model by which the logic constraints of system, such as charging/discharging of storages and power buying/selling from/to utility grid, can also be included in optimization problem. Let’s consider the given control scheme in υ Figure 2 . The com- fort condition is satisfied based onMPC2 by minimizing the comfort cost function as shown in equation (1) for the prediction horizon P in presence of the disturbances, user’s request estimation and the phys- ical and technical plant limitations. On the contrary, MPC1 decides how to choose the charging/discharging rates for each storage in order to guarantee the minimum required thermal energy for room heating system by minimizing the energy resources cost function as shown in equation (2) for the prediction horizon P. In the previous formulas, T zone and T inlet are the zone air tempera- ture and inlet water temperature passing through building pipe- lines respectively and T zone is the respective zone air temperature setpoint. P HP refers to the heat pump power consumption and P cost (k) is the total cost including selling and buying electricity power from/to utility grid. ΔP Batt (k) is the step changing in battery power exchange, which is relaxed by the weight of W Batt in order to prevent any damages or practical problems. Numerical results In this section, simulation tests as well as experimental vali- dation are presented. The tests witness the effectiveness of the proposed control approach. Specifically, the benefits of using of two control variables in the TES model (i.e., the tank mass flow-rate and the water temperature) and also the advantages deriving from the usage of different type of energy storages (electrical and thermal) are discussed. The total energy distributed for one day simulation among all stor- ages, generators and utility grid together with load demand are shown in υ Figure 3 . It shows the unit commitment, energy stor- ages, economic dispatch, sale and purchase of energy to/from utility grid based on a unitary buying and selling energy price. Similar test- ings have been performed in a smart grid laboratory at the National Technical University of Athens , using the ERIGrid Research Infrastructure funded by the European Union’s Horizon 2020 Research and Innovation Programme . Apart from the quality of the numerical results, the practical tests show the feasibility of the real-time implementation of distributed predictive control algorithms. Conclusions In this paper, an applicable economic solution is pro- posed to tackle efficiently the energy management problem in the context of smart buildings. The over- all solution is thus formulated as a distributed MPC with a coordi- nator, solving a mixed-integer quadratic programing optimization problem. DMPC approach, in comparison to classic control tech- niques, saves up to 12.5% in load energy consumption and it has 22% improvement of total cost benefit. The potential of proposed approach is also witnessed by an experimental validation test in a well-equipped smart grid research laboratory in Athens. References [1] S. Rastegarpour, M. Ghaemi and L. Ferrarmi, “A Predictive Control Strategy for Energy Manage- ment in Buildings with Radiant Floors and Thermal Storage”, in 2018 SICE International Symposium on Control Systems , Tokyo, Japan, 2018. [2] L. Ferrarini and G. Mantovani, “Modeling and control of thermal energy of a large commercial building”, in IEEE International Workshop on Intelli- gent Energy Systems (IWIES) , Vienna, Austria, 2013. [3] A. L. Nash, A. Badithela and N. Jain, “Dynamic modeling of a sensible thermal energy storage tank with an immersed coil heat exchanger under three operation modes”, Applied Energy , vol. 195, pp. 877-889, 2017. [4] M. Farina and R. Scattolini, “Distributed predic- tive control: A non-cooperative algorithmwith neigh- bor-to-neighbor communication for linear systems”, Automatica , vol. 48, no. 6, pp. 1088-1096, 2012. [5] A. Bemporad and M. Morari, “Control of systems integrating logic, dynamics, and constraints”, Auto- matica , vol. 35, no. 3, pp. 407-427, 1999. ref Figure 2 - Schematic diagram of DMPC approach Figure 3 - Total power distribution among all storages, generators and utility grid