Alessandra Parisio

Energy efficient building climate control using Stochastic Model Predictive Control and weather predictions

One of the most critical challenges facing society today is climate change and thus the need to realize massive energy savings. Since buildings account for about 40% of global final energy use, energy efficient building climate control can have an important contribution. In this paper we develop and analyze a Stochastic Model Predictive Control (SMPC) strategy for building climate control that takes into account weather predictions to increase energy efficiency while respecting constraints resulting from desired occupant comfort. We investigate a bilinear model under stochastic uncertainty with probabilistic, time varying constraints. We report on the assessment of this control strategy in a large-scale simulation study where the control performance with different building variants and under different weather conditions is studied. For selected cases the SMPC approach is analyzed in detail and shown to significantly outperform current control practice.

A Model Predictive Control Approach to Microgrid Operation Optimization

Microgrids are subsystems of the distribution grid, which comprises generation capacities, storage devices, and controllable loads, operating as a single controllable system either connected or isolated from the utility grid. In this paper, we present a study on applying a model predictive control approach to the problem of efficiently optimizing microgrid operations while satisfying a time-varying request and operation constraints. The overall problem is formulated using mixed-integer linear programming (MILP), which can be solved in an efficient way by using commercial solvers without resorting to complex heuristics or decompositions techniques. Then, the MILP formulation leads to significant improvements in solution quality and computational burden. A case study of a microgrid is employed to assess the performance of the online optimization-based control strategy and the simulation results are discussed. The method is applied to an experimental microgrid located in Athens, Greece. The experimental results show the feasibility and the effectiveness of the proposed approach.

Reducing peak electricity demand in building climate control using real-time pricing and model predictive control

A method to reduce peak electricity demand in building climate control by using real-time electricity pricing and applying model predictive control (MPC) is investigated. We propose to use a newly developed time-varying, hourly-based electricity tariff for end-consumers, that has been designed to truly reflect marginal costs of electricity provision, based on spot market prices as well as on electricity grid load levels, which is directly incorporated into the MPC cost function. Since this electricity tariff is only available for a limited time window into the future we use least-squares support vector machines for electricity tariff price forecasting and thus provide the MPC controller with the necessary estimated time-varying costs for the whole prediction horizon. In the given context, the hourly pricing provides an economic incentive for a building controller to react sensitively with respect to high spot market electricity prices and high grid loading, respectively. Within the proposed tariff regime, grid-friendly behaviour is rewarded. It can be shown that peak electricity demand of buildings can be significantly reduced. The here presented study is an example for the successful implementation of demand response (DR) in the field of building climate control.

Energy efficient microgrid management using Model Predictive Control

Microgrids are subsystems of the distribution grid which comprises small generation capacities, storage devices and controllable loads, operating as a single controllable system that can operate either connected or isolated from the utility grid. In this paper we present a preliminary study on applying a Model Predictive Control (MPC) approach to the problem of efficiently optimizing microgrid operations while satisfying a time-varying request and operation constraints. The overall problem is formulated using Mixed-Integer Linear Programming (MILP), which can be solved in an efficient way by using commercial solvers without resorting to complex heuristics or decompositions techniques. Then the MILP formulation leads to significant improvements in solution quality and computational burden. A case study of a typical microgrid is employed to assess the performance of the on-line optimization-based control strategy: simulation results show the feasibility and the effectiveness of the proposed approach.

Stochastic Model Predictive Control for Building Climate Control

In this brief paper, a Stochastic Model Predictive Control formulation tractable for large-scale systems is developed. The proposed formulation combines the use of Affine Disturbance Feedback, a formulation successfully applied in robust control, with a deterministic reformulation of chance constraints. A novel approximation of the resulting stochastic finite horizon optimal control problem targeted at building climate control is introduced to ensure computational tractability. This work provides a systematic approach toward finding a control formulation which is shown to be useful for the application domain of building climate control. The analysis follows two steps: 1) a small-scale example reflecting the basic behavior of a building, but being simple enough for providing insight into the behavior of the considered approaches, is used to choose a suitable formulation; and 2) the chosen formulation is then further analyzed on a large-scale example from the project OptiControl, where people from industry and other research institutions worked together to create building models for realistic controller comparison. The proposed Stochastic Model Predictive Control formulation is compared with a theoretical benchmark and shown to outperform current control practice for buildings.

A mixed integer linear formulation for microgrid economic scheduling

Microgrids are subsystems of the distribution grid which comprises small generation capacities, storage devices and controllable loads, which can operate either connected or isolated from the utility grid. This paper studies the microgrid economic scheduling, i.e. the problem of optimize microgrid operations to fulfil a time-varying energy demand and operational constraints while minimizing the costs of internal production and imported energy from the utility grid. The problem is posed as a mixed-integer linear programming model. The key difference in the proposed modeling approach is that no complex heuristics or decompositions are used; the full model is formulated and solved in an efficient way by using commercial solvers. This leads to significant improvements in schedule quality and in computational burden. A case study of a typical microgrid is investigated: simulation results show the feasibility and the effectiveness of the proposed approach.

A scenario-based predictive control approach to building HVAC management systems

We present a Stochastic Model Predictive Control (SMPC) algorithm that maintains predefined comfort levels in building Heating, Ventilation and Air Conditioning (HVAC) systems while minimizing the overall energy use. The strategy uses the knowledge of the statistics of the building occupancy and ambient conditions forecasts errors and determines the optimal control inputs by solving a scenario-based stochastic optimization problem. Peculiarities of this strategy are that it does not make assumptions on the distribution of the uncertain variables, and that it allows dynamical learning of these statistics from true data through the use of copulas, i.e., opportune probabilistic description of random vectors. The scheme, investigated on a prototypical student laboratory, shows good performance and computational tractability.

Randomized Model Predictive Control for HVAC Systems

Heating, Ventilation and Air Conditioning (HVAC) systems play a fundamental role in maintaining acceptable thermal comfort and Indoor Air Quality (IAQ) levels, essentials for occupants well-being. Since performing this task implies high energy requirements, there is a need for improving the energetic efficiency of existing buildings. A possible solution is to develop effective control strategies for HVAC systems, but this is complicated by the inherent uncertainty of the to-be-controlled system. To cope with this problem, we design a stochastic Model Predictive Control (MPC) strategy that dynamically learns the statistics of the building occupancy and weather conditions and uses them to build probabilistic constraints on the indoor temperature and CO2 concentration levels. More specifically, we propose a randomization technique that finds suboptimal solutions to the generally non-convex stochastic MPC problem. The main advantage of this method is the absence of apriori assumptions on the distributions of the uncertain variables, and that it can be applied to any type of building. We investigate the proposed approach by means of numerical simulations and real tests on a student laboratory, and show its practical effectiveness and computational tractability.

Implementation of a Scenario-based MPC for HVAC Systems: an Experimental Case Study

Heating, Ventilation and Air Conditioning (HVAC) systems play a fundamental role in maintaining acceptable thermal comfort and air quality levels. Model Predictive Control (MPC) techniques are known to bring significant energy savings potential. Developing effective MPC-based control strategies for HVAC systems is nontrivial since buildings dynamics are nonlinear and influenced by various uncertainties. This complicates the use of MPC techniques in practice. We propose to address this issue by designing a stochastic MPC strategy that dynamically learns the statistics of the building occupancy patterns and weather conditions. The main advantage of this method is the absence of a-priori assumptions on the distributions of the uncertain variables, and that it can be applied to any type of building. We investigate the practical implementation of the proposed MPC controller on a student laboratory, showing its effectiveness and computational tractability.

Demand response for aggregated residential consumers with energy storage sharing

A novel distributed algorithm is proposed in this paper for a network of consumers coupled by energy resource sharing constraints, which aims at minimizing the aggregated electricity costs. Each consumers is equipped with an energy management system that schedules the shiftable loads accounting for user preferences, while an aggregator entity coordinates the consumers demand and manages the interaction with the grid and the shared energy storage system (ESS) via a distributed strategy. The proposed distributed coordination algorithm requires the computation of Mixed Integer Linear Programs (MILPs) at each iteration. The proposed approach guarantees constraints satisfaction, cooperation among consumers, and fairness in the use of the shared resources among consumers. The strategy requires limited message exchange between each consumer and the aggregator, and no messaging among the consumers, which protects consumers privacy. Performance of the proposed distributed algorithm in comparison with a centralized one is illustrated using numerical experiments.