Abstract

It is a difficult technical challenge to design thermoelectric power generation systems that work optimally over a broad dynamic range of thermal input power. Conventional systems are designed to work optimally for a nominal operating condition, while maintaining the ability to operate at off nominal and extreme operating conditions without damage to the system. For systems that operate in a narrow range of thermal power conditions, thermoelectric waste heat recovery system design is simplified. However, for applications that do have a wide range of operating conditions, designs typically exhibit overall average efficiencies that are reduced by approximately 20% or more compared with that achievable for the thermoelectric material operating at peak efficiency. Both cars and trucks consume significant fuel at low mass flow rates. Since the ultimate goal of waste heat recovery systems is to minimize fuel consumption, it is critical that the recovery system be designed to operate near peak efficiency over the range of mass flow rates that make a significant contribution to overall power recovery. Such performance capability is especially important in city driving, and in hybrid vehicle applications. This paper describes a design concept that maximizes the performance for thermoelectric power generation systems in which the thermal power to be recovered is from a fluid stream (e.g., exhaust gas) subject to varying temperatures and a broad range of exhaust flow rates. The device is constructed in several parts, with each part optimized for a specific range of operating conditions. The thermoelectric system characteristics, inlet mass flow rates and fluid temperatures, and load and internal electrical resistances are monitored and generator operation is controlled to maximize performance. With this design, the system operates near optimal efficiency for a much wider range of operating conditions. Application of the design concept to an automobile is used to show the benefits to overall system performance.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call