Uponor (Sweden)

In most typical situations, thermal energy storage (TES) systems, which incorporate sensible and latent storage capacities, are not effectively utilized within the overall functions of building energy management systems (BEMSs), which usually rely on classical rule-based control (RBC). This study addresses the challenge of overcoming this by featuring model predictive control (MPC). The proposed method is based on modeling a water tank-integrated phase change material (PCM) using data-driven linear approximation generated with sparse regression. Based on the control objective, the proposed MPC can address two control targets, either providing robust and fast-tracking to the TES charging/discharging setpoints or reducing the energy cost related to the building heating needs. The digital simulation of a two-day scenario, using real operation conditions, demonstrates the effectiveness of the proposed MPC framework, showing up to 57 % heating cost reduction compared to the RBC scenario. As the real-time control requirement is critical, the MPC computing time was evaluated to assess its potential for integration into real-world applications within BEMS. • A novel modeling method of hybrid TES using data-driven linear approximation. • One MPC framework for two targets: as a temperature regulator or energy manager. • Real-time control evaluation of linear MPC with nonlinear solver COBYQA. • A BMS integration solution using an embedded system leveraging GPU acceleration.

Current district heating networks are undergoing a sustainable transition towards the 4th and 5th generation of district heating networks, characterized by the integration of different types of renewable energy sources (RES) and low operational temperatures, i.e., 55 °C or lower. Due to the lower temperature difference between supply and return, it is necessary to develop novel methods to understand the loads accurately and provide operation scenarios to anticipate demand peaks and increase flexibility in the energy network, both for long- and short-term horizons. In this study, a hybrid machine-learning (ML) method is developed, combining a clustering pre-processing step with a multi-input artificial neural network (ANN) model to predict heat loads in buildings cluster-wise. Specifically, the impact of time-series data clustering, as a pre-processing step, on the performance of ML models was investigated. It was found that data clustering contributes effectively to the reduction of data training costs by limiting the training processes to representative clusters only instead of all datasets. Additionally, low-quality data, including outliers and large measurement gaps, are excluded from the training to enhance the overall prediction performance of the models.

This study has focused on evaluating the financial potential of wastewater and geothermal heat recovery systems in a multi-family building. The recovered heat was used to improve the performance of mechanical ventilation with heat recovery (MVHR) system during the coldest days in central Sweden. The main issue, which was targeted with these solutions, was to reduce frost formation in the system and hence increase its thermal efficiency. By looking at the life cycle cost over a lifespan of 20 years, the observed systems were being evaluated economically. Furthermore, statistical analyses were carried-out to counter the uncertainty that comes with the calculation. It was found that the studied wastewater systems have a high possibility of generating savings in this period, while the one fed by geothermal energy is less likely to compensate for its high initial cost. All designed systems however, managed to reduce operational cost by 35-45% due to lower energy usage.