Research on Performance of Adaptive Solver Based on Joint Simulation of Aviation Systems

In order to accelerate the development cycle of aviation systems and improve the development efficiency of serialized products, a method of integrating adaptive solvers is proposed for joint simulation systems based on the FMI standard. The method is divided into two types of solvers: fixed step solvers and variable step solvers. The fixed step solvers mainly include Euler and Runge Kutta methods, while the variable step solvers include CVODE and LSODAR methods. Taking the fuel system as the main research object, the efficiency of traditional solver integration and adaptive solver integration is compared in terms of solution time, convergence tolerance, and error. The results show that as the convergence tolerance decreases, the proposed adaptive solver method is more stable, with relatively smooth errors, and better accuracy than the traditional integrated solver method; From the perspective of solution time, under the same solution time, the adaptive solver can reach the set calculation end time quickly, at least 500 times as fast as the traditional combined solver method.