Model verification and debugging of EOO models aided by model reduction techniques (Work in Progress Paper)

Equation-based object-oriented modeling approach significantly reduced effort needed for model implementation by releasing modeler of performing many error-prone tasks. An increasingly more complex models can be built, preferably from components of different model libraries. However, complexity of the models complicate the process of verification ‐ assuring that the model was implemented correctly and behaves as expected ‐ and possible subsequent debugging. A cause of error in a model with over 1000 different equations can be often hard to find by the deskchecking method. This requires the development of new modeling environment tools for model understanding and automated discovery of the fault causes. The difficulty of designing such tools in EOO modeling environments is linked to the difficulty of mapping simulation form to the model sources. Furthermore, debugging of complex models consisting of over thousand equations by traversing each equation may be very ineffective, especially when the fault has multiple and not very evident causes. A model reduction methods is proposed and discussed as a method of verification. With model reduction methods it is possible to identify the most important parts of the model which have contributed to the specific model behavior. Because model reduction can be performed on original model representation, the difficulty of mapping simulation form back to model source is avoided.