Transient behavior investigation of a regenerative dual-evaporator organic Rankine cycle with different forms of disturbances: Towards coordinated feedback control realization

Yang Du, Tingting Liu, Yaxiong Wang, Kang Chen, Pan Zhao, Jiangfeng Wang, Yiping DaiEnergy 2021 · DOI

TL;DR

This paper tackles the challenge of controlling regenerative dual-evaporator organic Rankine cycles (RDORCs) under dynamic conditions. By developing a comprehensive transient model and a novel coordinated feedback control system, the authors demonstrate improved stability and efficiency compared to traditional simple ORC systems when faced with heat source disturbances.

What Problem Does This Paper Address?

Organic Rankine Cycles (ORCs) are a promising technology for recovering waste heat, particularly from large-scale industrial sources. However, real-world heat sources are rarely stable, often experiencing fluctuations. Traditional ORC systems can struggle to maintain optimal performance and stability under these dynamic conditions, potentially leading to safety issues. Dual-evaporator ORCs (DORCs) offer enhanced efficiency by recovering more heat, but their increased complexity makes transient behavior analysis and control even more critical. This paper specifically addresses the need for robust transient modeling and effective control strategies for a regenerative dual-evaporator ORC (RDORC) to ensure stable and efficient operation under various disturbances.

Methods

The authors developed a comprehensive transient model of the RDORC using the finite volume method. This approach is chosen for its accuracy in modeling heat exchangers, which are crucial components. The model incorporates various sub-models for heat exchangers, compressible vapor volumes, a turbine, pumps, and a working fluid storage tank. A key aspect is the development of a novel coordinated feedback control system employing two PI controllers. This system aims to maintain constant superheat degrees of the turbine inlet main vapor and resupplied vapor by adjusting the rotational speeds of the high-pressure and low-pressure pumps. The control system parameters were tuned using the Ziegler-Nichols method. The model was validated against experimental data from a simple ORC (SORC) system.

Datasets

Not discussed in the paper. The authors mention using the NIST Refprop 10.0 database for thermal physical properties of R1234ze, air, and water. They also refer to experimental data from a previous study [36] for model validation.

Key Insights and Ideas

  • RDORC's Resilience: Despite its increased complexity, the RDORC, when equipped with the proposed control system, demonstrates comparable or even superior transient response to disturbances compared to a simpler SORC.
  • Control System Effectiveness: The coordinated two-PI controller system effectively manages the superheat degrees of both evaporators simultaneously, ensuring stable operation even under significant heat source fluctuations.
  • Disturbance Impact Varies: Different types of disturbances (step, sinusoidal, random) have distinct impacts on system parameters. For instance, sinusoidal disturbances lead to oscillations in net power with a longer period than the input disturbance.
  • Thermal Inertia Matters: The thermal inertia of components, particularly heat exchangers, plays a significant role in dampening disturbances and influencing the response times of different parts of the system.
  • RDORC's Efficiency Edge: The RDORC consistently shows higher exergy efficiency than the SORC, even during transient periods, highlighting its advantage in waste heat recovery.

Key Results

  • Under a 10°C step decrease in heat source inlet temperature, the RDORC exhibited a 0.5°C larger undershoot and a 164s longer setting time for superheat degree compared to the SORC. However, its exergy efficiency enhancement increased from 4.13% to 4.51% after the disturbance.
  • During sinusoidal disturbances, the peak amplitude of the turbine main vapor inlet pressure in the RDORC was significantly higher (3.6%) than that of the resupplied vapor inlet pressure (0.82%). The system net power showed sinusoidal fluctuations with a peak amplitude of 5.59%.
  • Under random oscillations, the RDORC maintained stable operation with superheat degrees fluctuating around the design value of 5°C with a maximum deviation of 0.02°C. System exergy and thermal efficiencies fluctuated around their design values with maximum deviations of 2.2% and 0.37%, respectively.
  • The proposed control system successfully realized steady and efficient sliding-pressure operation of the RDORC.

Limitations and Future Work

  • Model Simplifications: The turbine and pump were modeled using steady-state conditions, neglecting their dynamic response due to their fast characteristics. This simplification might limit the model's accuracy for very rapid transient events.
  • Experimental Validation: While the model was validated against SORC experimental data, direct experimental validation of the RDORC system itself under transient conditions is not presented.
  • Specific Disturbances: The study focused on specific types and magnitudes of heat source disturbances. Further investigation into a wider range of disturbance scenarios and their impact on system performance would be beneficial.
  • Control System Optimization: While the Ziegler-Nichols method was used for tuning, further optimization of the PI controller parameters could potentially lead to even better transient performance.

The authors suggest that their work provides a foundation for further research into the transient behavior and control of RDORC systems, paving the way for their more widespread adoption in industrial waste heat recovery.

Summary updated

BibTeX
@inproceedings{du2021transient,
  author = {Yang Du and Tingting Liu and Yaxiong Wang and Kang Chen and Pan Zhao and Jiangfeng Wang and Yiping Dai},
  title = {{Transient behavior investigation of a regenerative dual-evaporator organic Rankine cycle with different forms of disturbances: Towards coordinated feedback control realization}},
  booktitle = {Energy},
  year = {2021},
  doi = {10.1016/j.energy.2021.121437},
}

Loading related content...

Figure