Electric Motor Nonlinear Black-Box Identification
Budget: £20 – £250 GBP
Project Title:
Nonlinear Black-Box System Identification for an Electric Motor (Based on Input-Output Data)
Project Overview:
We are seeking collaboration with someone experienced in nonlinear system identification, particularly using data-driven (black-box) techniques. The aim of this project is to develop a reliable nonlinear model of an electric motor using only input-output data, without access to internal system parameters.
Project Details:
• The system is a three-input, three-output electric motor.
• The data has been obtained from a Simulink simulation, but in future stages, real-world data may be used.
• The goal is to build an accurate and robust model that can predict motor output under varying conditions (including with added noise) based solely on input signals.
System Characteristics:
• Black-box approach (no knowledge of internal motor parameters)
• Nonlinear system, requiring advanced identification methods
• Important note: This project does not involve or allow the use of neural networks or AI models. Our focus is solely on classical nonlinear system identification techniques.
Preferred Methods:
• Nonlinear ARX
• NARMAX
• Hammerstein-Wiener
• Nonlinear subspace identification methods are particularly preferred, if you have experience in this area.
Requirements:
• Demonstrated knowledge and experience in nonlinear black-box system identification
• Ability to implement models in MATLAB or similar environments
• The developed model must be testable: we will provide new input signals, and the model’s output must closely match the outputs generated by the original Simulink model or a real motor
• The model must remain accurate, reliable, and robust under realistic and possibly noisy conditions
• All steps of the work must be fully documented, with clear explanations. Each stage should be transparent, traceable, and easy to follow for other collaborators or researchers.
Collaboration Terms:
• We are working within a limited time frame, but the project will proceed in clear, structured, and milestone-based steps.
• A competitive payment is available for the right contributor.
• Performance assurance of the model is a key condition for ongoing collaboration.
If you have relevant experience and are interested in working on this project, we would be very happy to hear from you. Please feel free to share any previous related work or experience.
Nonlinear Black-Box System Identification for an Electric Motor (Based on Input-Output Data)
Project Overview:
We are seeking collaboration with someone experienced in nonlinear system identification, particularly using data-driven (black-box) techniques. The aim of this project is to develop a reliable nonlinear model of an electric motor using only input-output data, without access to internal system parameters.
Project Details:
• The system is a three-input, three-output electric motor.
• The data has been obtained from a Simulink simulation, but in future stages, real-world data may be used.
• The goal is to build an accurate and robust model that can predict motor output under varying conditions (including with added noise) based solely on input signals.
System Characteristics:
• Black-box approach (no knowledge of internal motor parameters)
• Nonlinear system, requiring advanced identification methods
• Important note: This project does not involve or allow the use of neural networks or AI models. Our focus is solely on classical nonlinear system identification techniques.
Preferred Methods:
• Nonlinear ARX
• NARMAX
• Hammerstein-Wiener
• Nonlinear subspace identification methods are particularly preferred, if you have experience in this area.
Requirements:
• Demonstrated knowledge and experience in nonlinear black-box system identification
• Ability to implement models in MATLAB or similar environments
• The developed model must be testable: we will provide new input signals, and the model’s output must closely match the outputs generated by the original Simulink model or a real motor
• The model must remain accurate, reliable, and robust under realistic and possibly noisy conditions
• All steps of the work must be fully documented, with clear explanations. Each stage should be transparent, traceable, and easy to follow for other collaborators or researchers.
Collaboration Terms:
• We are working within a limited time frame, but the project will proceed in clear, structured, and milestone-based steps.
• A competitive payment is available for the right contributor.
• Performance assurance of the model is a key condition for ongoing collaboration.
If you have relevant experience and are interested in working on this project, we would be very happy to hear from you. Please feel free to share any previous related work or experience.