In practical engineering, equipment parameters are difficult to obtain and measurement data is limited. This paper proposes a PINN-Modelica data-driven modeling method that couples a neural network with a Modelica model encapsulated as an FMU. Instead of directly constructing governing-equation residuals in the proposed method, the FMU supplies a modular physicsbased constraint during training. The neural network predicts both terminal voltage and time-varying physical parameters, while the FMU computes a physics-based voltage from the predicted parameters. Data and physics losses jointly constrain the neural-network voltage output, and the sensitivity of the physics loss to the FMU input parameters is evaluated numerically by central finite differences and combined with neural-network backpropagation. A second-order Thevenin battery model is used as a proof-of-concept, and coupled simulation is implemented through the Functional Mockup Interface (FMI) standard. The framework provides a reusable approach for integrating Modelica physical models with data-driven parameter identification.

This session is chaired by: Prof. Liu Yanfang.
She is the professor at the school of transportation sciencen and technology of Beihang University. She received her Ph.D. in Aerospace Manufacturing Engineering from Beihang University. Her research focuses on intelligent connected vehicles and their applications. She has hosted multiple national level projects such as the National Natural Science Foundation of China, and published more than 109 academic papers. She received multiple scientific and technological awards, such as the First Prize for Technological Invention in the Mechanical Industry, the Special Prize for Technological Progress in the Chinese Mechanical Industry, and the First Prize for Science and Technology in the Chinese Automotive Industry.

With the development of electric drive systems for electric vehicles toward high power density and integration, electric drive assemblies integrating the drive motor, motor controller, and reducer have attracted lots of concerns. Under the combined effect of multiple heat sources within a compact space, the system heat dissipation of an electric drive system remains an open challenge, so it is necessary to analyze the temperature rise characteristics inside the motor. This paper examines in detail the distribution of internal losses in the motor and calculates the thermal resistances of the motor components. Then, a thermal–hydraulic coupled simulation platform on the Amesim platform for predicting the temperature rise of oil-cooled motors is developed based on the lumped-parameter thermal network method. A bench test is established to verify the accuracy of the simulation model. Experimental results show that the maximum error is 5.53%. This study provides a reference for the thermal analysis of motors in electric drive assemblies.