A Recommendation System for Prosumers Based on Large Language Models.

As modern technologies, particularly home assistant devices and sensors, become more integrated into our daily lives, they are also making their way into the domain of energy management within our homes. Homeowners, now acting as prosumers, have access to detailed information at 15-min or even 5-min intervals, including weather forecasts, outputs from renewable energy source (RES)-based systems, appliance schedules and the current energy balance, which details any deficits or surpluses along with their quantities and the predicted prices on the local energy market (LEM). The goal for these prosumers is to reduce costs while ensuring their home's comfort levels are maintained. However, given the complexity and the rapid decision-making required in managing this information, the need for a supportive system is evident. This is particularly true given the routine nature of these decisions, highlighting the potential for a system that provides personalized recommendations to optimize energy consumption, whether that involves adjusting the load or engaging in transactions with the LEM. In this context, we propose a recommendation system powered by large language models (LLMs), Scikit-llm and zero-shot classifiers, designed to evaluate specific scenarios and offer tailored advice for prosumers based on the available data at any given moment. Two scenarios for a prosumer of 5.9 kW are assessed using candidate labels, such as Decrease, Increase, Sell and Buy. A comparison with a content-based filtering system is provided considering the performance metrics that are relevant for prosumers.

Developing a Mixed Neural Network Approach to Forecast the Residential Electricity Consumption Based on Sensor Recorded Data.

In this paper, we report a study having as a main goal the obtaining of a method that can provide an accurate forecast of the residential electricity consumption, refining it up to the appliance level, using sensor recorded data, for residential smart homes complexes that use renewable energy sources as a part of their consumed electricity, overcoming the limitations of not having available historical meteorological data and the unwillingness of the contractor to acquire such data periodically in the future accurate short-term forecasts from a specialized institute due to the implied costs. In this purpose, we have developed a mixed artificial neural network (ANN) approach using both non-linear autoregressive with exogenous input (NARX) ANNs and function fitting neural networks (FITNETs). We have used a large dataset containing detailed electricity consumption data recorded by sensors, monitoring a series of individual appliances, while in the NARX case we have also used timestamps datasets as exogenous variables. After having developed and validated the forecasting method, we have compiled it in view of incorporating it into a cloud solution, being delivered to the contractor that can provide it as a service for a monthly fee to both the operators and residential consumers.

A value sharing method for heterogeneous energy communities archetypes.

A novel value sharing (VS) method is proposed that distributes the energy communities (ECs) value based on the individual contribution to the total surplus/deficit. It considers the load-generation profile of each EC member and allocates a higher share to members who contribute to the EC revenue. The lowest share is received by the members with the highest demand that has to be supplied from the shared generation or from the grid, contributing to the EC cost. Several allocation methods are compared using the fairness index (FI), and, for setting the strategy of the EC using a decision model, as the strategy may vary over time, an objective function is defined as a combination between FI and self-sufficiency index using weighting coefficients. The methodology is implemented as an algorithm that automatically calculates and distributes the gain. For the proposed VS method, the FI is between 0.81 and 1.

How Fast to Avoid Carbon Emissions: A Holistic View on the RES, Storage and Non-RES Replacement in Romania.

The European Union targets aim to replace the non-renewable energy sources (non-RES) of coal, oil and gas (COG) generation with RES and storage (RES-S). The replacement of COG-generating units will lead to a decrease in CO2 emissions and a better living environment. Starting from this desideratum, in this paper, we create several scenarios to replace COG in Romania with RES-S, reconsider future energy mixes and engage with a more creative planning in order to meet the clean energy transition path. The energy shortages, especially in European countries after the Russian invasion of Ukraine, led many governments (including the Romanian, Polish, etc.) to think more about short-term supply issues and less about medium- and long-term power system planning. However, the decision makers of the European power systems have to decide how fast to avoid firing coal, how fast to adopt RES and how fast to invest in flexibility sources, including storage stations to enable a higher integration of RES. Therefore, in this paper, a holistic view to envision the RES and non-RES contribution to the load coverage in Romania for a smooth transition to a low-carbon economy is provided. The results show that an initial mix of wind, photovoltaic (PV) and storage systems is preferable to substitute 600 MW of installed power in coal-based power plants. Furthermore, the case of Poland-the European country with over 70% coal in its generation portfolio-is also presented as it can serve as a good example.

Feature engineering solution with structured query language analytic functions in detecting electricity frauds using machine learning.

Detecting fraud related to electricity consumption is usually a difficult challenge as the input datasets are sometimes unreliable due to missing and inconsistent records, faults, misinterpretation of meter reading remarks, status, etc. In this paper, we obtain meaningful insights from fraud detection using real datasets of Tunisian electricity consumption metered by conventional meters. We propose an extensive feature engineering approach using the structured query language (SQL) analytic functions. Furthermore, double merging of datasets reveals more dimensions of the data allowing better detection of irregularities in consumption. We analyze the results of several machine learning (ML) algorithms that manage cases of weakly correlated features and highly unbalanced datasets. The skewness of the target is approached as a regular characteristic of the input data because most of consumers are fair and only a small portion attempt to mislead the utility companies by tampering with metering devices. Our fraud detection solutions consist of combining classifiers with an anomaly detection feature obtained with an unsupervised ML algorithm-Isolation Forest, and extensive feature engineering using SQL analytic functions on large datasets. Several techniques for feature processing enhanced the Area Under the Curve score for Decision Tree algorithm from 0.68 to 0.99.

Evaluation of Dispersion Methods and Mechanical Behaviour of Glass Fibre Composites with Embedded Self-Healing Systems.

The present paper is focused on evaluating the most suitable dispersion method in the epoxy matrix of two self-healing systems containing dicyclopentadiene (DCPD) and 5-ethylidene-2-norbornene (ENB) monomers encapsulated in a urea-formaldehyde (UF) shell, prior to integration, fabrication and impact testing of specimens. Both microstructural analysis and three-point bending tests were performed to evaluate and assess the optimum dispersion method. It was found that ultrasonication damages the microcapsules of both healing systems, thus magnetic stirring was used for the dispersion of both healing systems in the epoxy matrix. Using magnetic dispersion, 5%, 7%, 10%, 12% and 15% volumes of microcapsules were embedded in glass fibre composites. Some of the samples were subjected to thermal cycling between -20 °C and +100 °C for 8 h, to evaluate the behaviour of both healing systems after temperature variation. Impact test results showed that the mechanical behaviour decreases with increasing microcapsule volume, while for specimens subjected to thermal cycling, the impact strength increases with microcapsule volume up to 10%, after which a severe drop in impact strength follows. Retesting after 48 h shows a major drop in mechanical properties in specimens containing 15% MUF-ENB microcapsules, up to total penetration of the specimen.