Publicação
Smart system for monitoring and controlling energy consumption by residential loads
| Resumo: | This work proposes the development of a smart system for monitoring and controlling surplus energy consumption by residential loads connected to smart plugs. Data processing, storage and monitoring tools will be used, specifically Node-RED, InfluxDB, Grafana and Home Assistant. The latter allows remote control of devices through its interface. Through these tools, the user can visualize excess power data, which, in this work, are obtained through a hardware prototype that uses the ESP32 as a microcontroller and is responsible for measuring voltage and current of energy to analyze negative power and positive or surplus. Based on this measurement, load control tests were carried out based on the ratio of available energy to the power required by the load, using the Home Assistant interface. Then, this work focused on performing the excess power prediction by the linear regression method, a Machine Learning approach. A dataset with values of consumption and generation of photovoltaic energy in a residence was used and forecast analyzes were performed. The predicted results meet the expected objectives, although it is reasonable to conclude that further studies can still be done to ensure the robustness of the prediction model. |
|---|---|
| Autores principais: | Dias, Paloma Greiciana de Souza |
| Assunto: | Surplus energy Smart plug Machine learning |
| Ano: | 2023 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Instituto Politécnico de Bragança |
| Idioma: | inglês |
| Origem: | Biblioteca Digital do IPB |
| Resumo: | This work proposes the development of a smart system for monitoring and controlling surplus energy consumption by residential loads connected to smart plugs. Data processing, storage and monitoring tools will be used, specifically Node-RED, InfluxDB, Grafana and Home Assistant. The latter allows remote control of devices through its interface. Through these tools, the user can visualize excess power data, which, in this work, are obtained through a hardware prototype that uses the ESP32 as a microcontroller and is responsible for measuring voltage and current of energy to analyze negative power and positive or surplus. Based on this measurement, load control tests were carried out based on the ratio of available energy to the power required by the load, using the Home Assistant interface. Then, this work focused on performing the excess power prediction by the linear regression method, a Machine Learning approach. A dataset with values of consumption and generation of photovoltaic energy in a residence was used and forecast analyzes were performed. The predicted results meet the expected objectives, although it is reasonable to conclude that further studies can still be done to ensure the robustness of the prediction model. |
|---|