HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Theses

Prediction for land smart management of the renewable photovoltaic energy & integration based on a network of IoT LoRaWAN sensors

Abstract : The implementation of decentralized electrical micro-grids base on renewable energies is a challenge for the power supplies of telecommunication infrastructures. These power sources supply from off-grid photovoltaic and micro-wind systems are more suitable for remote or isolated sites like telecom antennas considering the local installation feasibilities and renewable resource availabilities. However, the intermittence and the variability of renewable sources require to ensure the balance between the production and the consumption of energies. In addition, the system sizing requires a trade-off between size, investment cost, safety and reliability of the power supply due to the intermittency. This thesis aims to develop a model characterizing the global solar irradiation variability to predict PV productions from data recorded on sites and over a long period. We validated the calibration methods of the Weibull function allowing a reliable prediction of the global irradiation in a semi-continental climate (mainland French sites). More precisely, we experimentally determine the Weibull parameters for obtaining a reliable global irradiation forecast prediction by considering the sun incident radiation during time periods. This forecast allows to estimate the yearly energy generation and its prediction during the time evolution from a PV plant in considered sites. The originality of the proposed approach is to obtain reliable predictive yearly PV energy distributions which also can be used to generate synthetic data times series of the global irradiation. The proposed model is based a parametrized mathematical formula providing reliable prediction results and which can be integrated to a real-time data acquisition system. To validate the proposed approach for the prediction of the PV power source supplying of telecommunication infrastructures, we developed an IoT sensors network for real-time acquisition to instrument a telecom antenna allowing a wide area network coverage. We demonstrate that deployed networked IoT sensors based on LoRaWAN protocol on a traditional telecommunication tower operating under real conditions communicate reliably without affecting the tower’s functions while keeping the data quality of the sensors. The proposed LoRaWAN network is used for the data acquisition of the weather parameters performing the proposed forecast of the real-time PV production. The software implementation of the proposed prediction model interfaced to weather sensors allows a real-time and intelligent management of deployed renewable energy systems. Future work is also discussed to develop renewable energies on a wide territory with a semi-continental climate.
Document type :
Theses
Complete list of metadata

https://hal.univ-lorraine.fr/tel-03632971
Contributor : Thèses Ul Connect in order to contact the contributor
Submitted on : Wednesday, April 6, 2022 - 4:21:01 PM
Last modification on : Thursday, April 7, 2022 - 3:15:30 AM

Identifiers

  • HAL Id : tel-03632971, version 1

Collections

Citation

Ollé Michel Kam. Prediction for land smart management of the renewable photovoltaic energy & integration based on a network of IoT LoRaWAN sensors. Engineering Sciences [physics]. Université de Lorraine, 2021. English. ⟨NNT : 2021LORR0197⟩. ⟨tel-03632971⟩

Share

Metrics

Record views

11