Ensemble of self-organizing adaptive maps and dynamic multi-objective optimization for organic Rankine cycle (ORC) under transportation and driving environment

Efficient construction and optimization of mapping correlation of organic Rankine cycle (ORC) system under driving environment is the key to obtain the actual waste heat recovery limit. Under external and internal disturbances, the ORC operating characteristics have obvious uncertainty and nonlinearity. Based on driving conditions and ORC operation characteristics, this paper proposes an ensemble approach of self-organizing adaptive maps and dynamic multi-objective optimization for ORC under driving environment from the perspectives of coupling ORC integration system, variable data selection, parameter coupling correlation, adaptive structure design and multi-objective optimization. This approach can achieve efficient capture, construction and optimization of ORC dynamic characteristics in complex driving environment. Compared with direct modeling, input variables decreased by at least 69.23%. RMSE decreased by at least 66.06%. Approach can adjust operating parameters in real time according to the fluctuation of actual driving environment, break through the trade-off effect between thermal efficiency and emissions of CO2 equivalent (ECE), to keep the optimal state of thermal efficiency and ECE continuously. The approach proposed in this paper can provide a new idea for efficient construction and rapid optimization of ORC dynamic mapping association in driving environment. •Proposed self-organizing adaptive dynamic modeling for organic Rankine cycle (ORC).•An ensemble approach for multi-objective optimization of ORC in driving environment.•Approach proved to be effective and adaptive.•The driving environment data confirmed the robustness of our approach.