Publikationer

The growing population and demand for new public buildings contribute to increased energy consumption and greenhouse emienisdans. In Sweden, the largest amount of energy is consumed in school buildings, i.e., where schools form the highest number of public properties (30 million m ² ). In total, schools consumed 4 222 GWh of district heating and about 3 GWh of electricity for heating and other purposes in 2020. These figures lead to the realization of the need to apply effective measures to meet the European Green Deal target for 2030. Accurately forecasting energy usage is important for all stakeholders to conduct economic analysis and optimize decision-making. It is equally important in maintenance operations to allocate resources and enable the staff and students to adjust their behaviours and address the issues in buildings where peak forecasts occur. This paper develops and evaluates a power and district heating consumption for a single day and multiple days forecasting using Multivariate Recurrent Neural Network (RNN) -Long-Short term memory (LSTM) and convolutional neural networks (CNNs) and Autoencoders (AE), using daily real consumption data of six public schools provided by Skelefteå municipality in Sweden. The experimental results demonstrate that the hybrid model CNN-LSTM has achieved good accuracy compared to others, with RMSE and nRMSE error between 18%-25% and 5%-6% for electricity, respectively, and between 20%-30% RMSE and 5% nRMSE for district heating.

The development of accurate energy prediction models plays a significant role in achieving sustainability in smart cities. However, stakeholders such as municipalities face the problem of creating individual energy forecasting models for multiple building fleets which leads to an increased amount of computational resources and time spent to prepare each model. This research proposes a method using Hierarchical clustering with dynamic time warping to group similar buildings according to their consumption values and the integration of transfer Learning (TL) to share the model weights from a source building to other target buildings. Different TL models using only 20%, 40%, and 60% of the target data were tested against a standard workflow without TL for predicting electricity and district heating for several school buildings using a Multivariate LSTM model. The results show a small variation between the TL and the standard models; when trained on only 40% of the data, the models achieved an average of 0.24% RMSE improvement for district heating and a 1.23% for electricity, indicating a potential for reduced data requirements without sacrificing predictive accuracy and demonstrating TL’s efficiency to streamline the energy forecasting process for building fleets.

Technological developments, such as mobile augmented reality (MAR) and Internet of Things (IoT) devices, have expanded available data and interaction modalities for mobile applications. This development enables intuitive data presentation and provides real-time insights into the user’s context. Due to the proliferation of available IoT data sources, user interfaces (UIs) have become complex and diversified, while mobile devices have limited screen spaces. This state increases the necessity of design principles that help to secure sufficient user experience (UX). We found that studies of design principles for IoT-enabled MAR applications are limited. Therefore, we conducted a systematic literature review to identify existing design principles applicable to IoT-enabled MAR applications. From the state-of-the-art research, we compiled and categorized 26 existing design principles into seven categories. We analyzed the UIs of three IoT-enabled MAR applications with the identified design principles and user feedback gathered from each application’s evaluation to understand what design principles can be considered in designing these applications. Among the 26 principles, we find eight principles that are commonly identified as possible improvements for the applications based on their purposes. We demonstrate the practical use of the identified principles by redesigning the UIs, and we propose five new design principles derived from the application analysis. As a result, we summarized a total of 31 design principles, including the five new ones. We expect that our findings will give insight into the UX/UI design of IoT-enabled MAR applications for researchers, educators, and practitioners interested in UX/UI development.

Technological developments, such as mobile augmented reality (MAR) and Internet of Things (IoT) devices, have expanded available data and interaction modalities for mobile applications. This development enables intuitive data presentation and provides real-time insights into the user’s context. Due to the proliferation of available IoT data sources, user interfaces (UIs) have become complex and diversified, while mobile devices have limited screen spaces. This state increases the necessity of design principles that help to secure sufficient user experience (UX). We found that studies of design principles for IoT-enabled MAR applications are limited. Therefore, we conducted a systematic literature review to identify existing design principles applicable to IoT-enabled MAR applications. From the state-of-the-art research, we compiled and categorized 26 existing design principles into seven categories. We analyzed the UIs of three IoT-enabled MAR applications with the identified design principles and user feedback gathered from each application’s evaluation to understand what design principles can be considered in designing these applications. Among the 26 principles, we find eight principles that are commonly identified as possible improvements for the applications based on their purposes. We demonstrate the practical use of the identified principles by redesigning the UIs, and we propose five new design principles derived from the application analysis. As a result, we summarized a total of 31 design principles, including the five new ones. We expect that our findings will give insight into the UX/UI design of IoT-enabled MAR applications for researchers, educators, and practitioners interested in UX/UI development.

The integration of renewable energy sources (RES) into modern power grids has enabled decentralized energy generation at the community level, fostering peer-to-peer (P2P) energy trading among prosumers and microgrids. Accurate forecasting of household energy consumption and photovoltaic (PV) generation is critical for optimizing energy flows, enhancing grid reliability, and enabling cost-effective trading decisions. This paper presents an intelligent energy trading platform that integrates machine learning-based forecasting, battery-aware decision-making, and blockchain-enabled transactions to facilitate secure and efficient local energy exchange. Using historical smart meter and weather data from London households, multiple forecasting models including GRU, LSTM, Random Forest, and XGBoost were trained and evaluated. The GRU model achieved superior performance in predicting energy consumption, while Random Forest produced the most accurate PV generation forecasts. These predictions were combined with household battery levels to dynamically determine next-day operational roles: Buyer, Seller, Store, or Use Battery. Unlike conventional fixed-threshold approaches, the framework supports user-defined variable battery thresholds, allowing personalized energy management strategies. The proposed decision-making model achieved an accuracy of 90.72 % for one random block, and extended simulations across 29 different random household blocks confirmed its robustness with an average accuracy of 88.69 % (95 % CI: 87.9–89.6 %). In the trading phase, households participate in a decentralized energy trading platform powered by blockchain and smart contracts. Based on the next-day forecasts, a linear programming-based optimization algorithm matches buyer requests and seller offers to minimize the total system cost while ensuring fairness and efficient energy allocation. To assess its performance, the proposed optimization approach was compared against a greedy matching algorithm where sequential matching is done without a cost optimization and a grid baseline scenario where no storage/sharing of energy takes place. The optimized matching consistently achieved substantially lower trading costs across all households demonstrating superior efficiency, fairness, and scalability compared to the benchmark methods. All transactions are executed securely and transparently on the blockchain through Ethereum-based smart contracts, which automate energy trading, pricing, and settlement. A user-friendly web interface was developed to allow participants to monitor and interact seamlessly with the platform. Overall, this battery-aware, community-driven trading framework showcases how intelligent energy forecasting, cost-optimized decision-making, and blockchain-enabled trading can collectively enhance energy autonomy, cost savings, and renewable energy utilization at both the household and community levels.

Buildings account for a substantial share of global energy consumption. Enhancing energy efficiency in buildings is crucial for both environmental sustainability and economic efficiency. Accurate and interpretable forecasting of energy demand is particularly critical in cold-climate regions to ensure efficient energy management, reduce wastage, and provide stakeholders with transparent insights into decision-making. This thesis presents the application of interpretable deep learning technique to forecast daily heat consumption in a school located in Northern Sweden. It specifically focuses on enhancing prediction accuracy while maintaining model transparency. A Bidirectional Long Short-Term Memory (BiLSTM) neural network is designed to capture temporal dependencies in multivariate time series data, comprising weather variables, calendar features, and historical consumption. To enable model interpretability, SHAP(SHapley Additive exPlanations) is applied to analyze global feature contributions and provide local explanations for individual predictions.

Cybersecurity is becoming increasingly important as the number of IoT devices andthe adoption of Building Management Systems (BMS) in modern buildings continues to grow. Monitoring smart buildings of the future is becoming increasingly critical and BMS depends on making accurate decisions based on sensor data. This thesis investigates anomaly detection in a district heating system with a focus on set temperature that is strongly infuenced by the outside temperature. We evaluate two machine learning methods Long Short-Term Memory (LSTM) neural networks and Isolation Forest using real world BMS data from a school building in Piteå, Sweden. Performance is measured in terms of recall, precision, F1 score and computational requirements. The LSTM models employs a sliding-window approach with varied window lengths and both single-layer and stacked architectures, while the Isolation Forests model varies in tree count and contamination levels. In real-world systems such as district heating, anomalies are rare by nature. To support model evaluation synthetic anomalies are often introduced using diferent methods. In this work, anomalies were injected based on localized statistical deviations. The stacked LSTM achieved the best balance between precision, recall and system resources. LSTM detected correctly 95 of 211 anomalies with an average deviation of 1.31°C. Resource results reveal that sliding-window size and LSTM depth substantially impact memory and inference latency, highlighting trade-ofs for real-time deployment. The fndings of this study could support the implementation of real-time anomaly detection in district heating systems, improving operational efciency and security, thus enabling building managers and operators to take timely actions.

Building Management Systems (BMS) are essential for modern smart buildings, managing services such as HVAC, lighting, and access control. However, their reliance on open communication protocols like BACnet makes them vulnerable to faults and cyberattacks. Detecting anomalies in BMS traffic is therefore important for maintaining both security and reliability.

This thesis investigates the use of machine learning, specifically the Isolation Forest algorithm, for anomaly detection in BMS network traffic. Five weeks of real traffic data were collected from a live BMS using a Raspberry Pi configured as a passive sniffer. The captured pcap files were converted into CSV format, with features such as IP addresses, protocols, and packet lengths extracted for analysis. Isolation Forest was then trained and tested under four time-window configurations (1h, 6h, 24h, 7d) to evaluate how temporal granularity influences detection outcomes.

The results show that the algorithm successfully modeled the stable and repetitive baseline of BMS communication. Shorter windows identified more fluctuations, demonstrating higher sensitivity, while longer windows produced highly stable baselines but flagged few or no anomalies. No anomalies were strongly associated with specific protocols, and the overall traffic composition remained consistent across the dataset.

The findings highlight that even in the absence of labeled data, unsupervised learning can provide valuable insights into BMS traffic. While anomalies were limited in this case, the study demonstrates the potential of Isolation Forest as a lightweight baseline tool for anomaly detection, and it underlines the trade-off between sensitivity and stability in configuring such systems.