Model Predictive Control for the Electrification of Industrial Processes using Renewable Energy Sources and Storage

The green transition requires that industrial carbon dioxide emissions are eliminated. Deep decarbonization is a set of technologies to achieve this goal. One way to achieve deep decarbonization is by using different storages such that electrical energy can be used when needed. The electrical energy may be stored in batteries or using electrolysis as hydrogen and oxygen that may be reconverted in fuel cells to electricity. Resistive or induction heating may be used for thermal storage of electricity at various temperatures. Storage at high temperatures (above 1100C) is considered for high temperature processes such as cement production, which alone accounts for about 8% of global CO2 emissions. Decarbonizing the cement production process is challenging, because a good part of CO2 emissions is intrinsic in the chemical process. Cement is produced from limestone-based clinker, which is derived by calcining limestone at high temperature. The chemical reaction produces CO2 as a product. Additional CO2 emissions are generated by burning fuel in the calciner, in order to activate the reaction.
This project is part of the EUDP funded ECoClay project. The ECoClay project is a collaboration between FLSmidth, DTU, Technological Institute (DTI), cement producers, and Rondo. Rondo is a company who is a specialist in thermal storage systems for high temperature processes. Such a storage can overcome the problem of renewable energy intermittency in an industrial setting. The idea is that significant CO2 reduction in the cement process can be achieved by considering 2 improvements. First, the fuel burned in the calciner can be entirely substituted with an electric heat generator. Secondly, a good part of limestone-based clinker can be substituted by calcined clay. We develop and apply forecasting and optimal control methods for calcination of clay using renewable energy sources (solar and wind), thermal storage, and digitalization. Using Model Predictive Control (MPC), the project links renewable energy sources and the electrical power system, thermal storage, and the clay calcination process. The claim is that by controlling tightly the calcination process in the calciner, highly reactive and consistent clay can be obtained. By doing so, the calcinated clay can replace up to 40% of the clinker in the final product, resulting in a huge CO2 reduction. Substituting fuel with renewable heat makes the calcinated clay CO2 free.
In the first phase of the project, we focus on stochastic modeling and simulation of wind power production, thermal storage and the electrified clay calciner, based on the information provided by FLSmidth. Subsequently, a Model Predictive Control (MPC) algorithm based on the model and experimental data is developed. In the final part of the project, the software will be tested and implemented in the pilot plant built by FLSmidth in their R&D Center in Mariager (Denmark). Eventually, our MPC technology will allow the plant to operate as close as possible to the desired conditions, while minimizing energy consumption and costs and ensuring high product quality.