Soft Measurement Techniques for Practical Solid Oxide Fuel Cell Stack Temperatures

In the face of “double carbon”, “green energy” and other calls of the times, Solid Oxide Fuel Cell (SOFC) as a kind of environmentally friendly and efficient hydrogen power generation device, the core device of the electric reactor is manufactured on the basis of high precision and high sealing process. The high temperature working condition of SOFC system brings a lot of thermal safety problems to the stack, and if the stack temperature of SOFC can be measured accurately, it can play an important role in monitoring and warning in the actual use of SOFC system. However, due to the completely sealed characteristics of the stack, we cannot embed thermocouples inside the physical SOFC stack to obtain the temperature, this paper proposes a technique to indirectly measure the temperature inside the SOFC stack using observable and measurable inputs and outputs, select the optimal operating variables using mutual information of data, screen out the optimal system type, and utilize multiple linear regression algorithms to complete this paper's temperature soft measurement technique.