Norwegian AI Directory
Description:
The focus artificial intelligence for energy related operations in technically advanced warehouses. The output of this PhD study will automate the energy related managements in a warehouse in order to reduce the operational cost and to improve the energy efficiency based on various data inputs and the-state-of-the-art deep neural network.
A successful solution will be able to automatically (1) clean data inputs, (2) propose accurate forecast of future events and (3) suggest corresponding actions based on the deep neural network. The suggested actions will be deployed through different actuators in the system and the behavior of the system will become a feedback (input) to the neural network again.
Traditionally, this type of management is played by human being on experience basis and now we expect to adopt AI for such management. In the beginning, the suggested actions from AI may go through a small degree of manual inspections and the final goal is to reduce and eliminate the involvement of human being in the management.
Deep neural networks play a pivotal role in the system because the deep neural network will forecast future events and propose corresponding actions based on the inputs from various data sources. Depending on the configuration of the warehouses and the application of the system, the data inputs can include, for example, weather forecasts, expected operating conditions, energy storage along with data from a sensor network inside the warehouse.
In this study, we consider primarily two areas: (1) improvement in energy efficiency and deduction in operating costs, and (2) optimization of thermal energy storage.
The algorithms will be practically applied in technically advanced warehouses in close collaboration with the warehouse owners and operators.
Project leader: Frank Skjærvik
Started: 2018
Ends: 2022
Category: Næringsliv
Sector: Næringsliv
Budget: 2021000
Institution: LOGIN EIENDOM AS
Address: Trondheim