Shuhai Quan

A Novel Optimization Method for the Electric Topology of Thermoelectric Modules Used in an Automobile Exhaust Thermoelectric Generator

Based on Bi2Te3 thermoelectric modules, a kind of automobile exhaust thermoelectric generator (AETEG) with a single-column cold-source structure was designed. To enhance its net power and efficiency, the output performance of all the thermoelectric modules was tested with a temperature monitoring unit and voltage monitoring unit, and modeled using a back-propagation (BP) neural network based on various hot-source temperatures, cold-source temperatures, load currents, and contact pressures according to the temperature distribution of the designed heat exchanger and cooling system. Then, their electric topology (series or parallel hybrid) was optimized using a genetic algorithm to achieve the maximum peak power of the AETEG. From the experimental results, compared with when all the thermoelectric modules were connected only in series or parallel at random, it is concluded that the AETEG performance is evidently affected by the electric topology of all the single thermoelectric modules. The optimized AETEG output power is greatly superior to the other two investigated designs, validating the proposed optimized electric topology as both feasible and practical.

Study on online monitoring system of PEMFC resistance based on electrochemical impedance spectroscopy

The ohmic resistance of Proten Exchange Membrance Fuel Cell(PEMFC)is mostly decided by PEMs water content, which can directly affect the efficiency of electricity generation, so it is very important for the measurement and control of PEMFCs humidity by means of monitoring its internal resistance. However, the internal resistance is hard to measure exactly because of its non-linear and time-varying charateristic. To solve this problem, In this paper we propose a methodology based on Electrochemical Impedence Spectroscopy(EIS) to guide the hardware and software design of online montoring system of PEMFC internal resistance, and the experiment results prove that good effect is achieved by adopting this method.

Air compressor flow forecast in fuel cell based on elman NN

The air flow is an important parameter for the proton exchange membrane fuel cells (PEMFC) engine. Based on a modified Elman neural network, a model, which is used to forecast the air flow for fuel cell compressor, will be introduced. And the Elman neural network is used to learn and train simulation experiments of the network, and compared with simulation result of the BP neural network. Simulation result shows that the Elman neural network is precise and effective in air flow prediction.