The Probabilistic Deconvolution of the Distribution of Relaxation Times with Finite Gaussian Processes

Electrochemical impedance spectroscopy (EIS) is a tool widely used to study the properties of electrochemical systems. The distribution of relaxation times (DRT) has emerged as one of the main methods for the analysis of EIS spectra. Gaussian processes can be used to regress EIS data, quantify uncertainty, and deconvolve the DRT, but current implementations do not constrain the DRT to be positive and can only use the imaginary part of EIS spectra. Herein, we overcome both issues by using a finite Gaussian process approximation to develop a new framework called the finite Gaussian process distribution of relaxation times (fGP-DRT). The analysis on artificial EIS data shows that the fGP-DRT method consistently recovers exact DRT from noise-corrupted EIS spectra while accurately regressing experimental data. Furthermore, the fGP-DRT framework is used as a machine learning tool to provide probabilistic estimates of the impedance at unmeasured frequencies. The method is further validated against experimental data from fuel cells and batteries. In short, this work develops a novel probabilistic approach for the analysis of EIS data based on Gaussian process, opening a new stream of research for the deconvolution of DRT.

A Machine Learning Based Numerical Approach for Valve Seating Velocity Control in an Electromagnetic Camless System

Improving internal combustion engine performance is a significant concern over the past few decades for engine researchers and automobile manufacturers. One of the promising methods for improving the engine performance is variable valve actuation system with camless technology. In the camless system, the conventional spring-operated valve actuation mechanism is removed, and an actuator is used to independently control the valve events (lift, timing, and duration). Among different camless systems, electromagnetic variable valve actuation (EMVA) becomes more viable because of its faster valve operation. However, the major challenge is to control the valve seating velocity (velocity at which valve comes to rest during seating on the cylinder head) due to the absence of the cam mechanism. A sophisticated control system must be developed to achieve an acceptable valve seating velocity. In this study, a proportional-integral-derivative (PID) controller was used to control the EMVA system. A machine learning tool, i.e., genetic algorithm, and an iterative method, i.e., Ziegler-Nichols, were used to optimize the PID controller’s gain values. The valve lift profiles obtained using the Ziegler-Nichols method and the genetic algorithm were compared. It was found that the developed algorithm for the EMVA system can achieve faster rise time compared to the experimental results [25] utilized inverse square method. A parametric investigation was performed to verify the robustness of the PID controller with a change in temperature. It is concluded that the temperature rise may increase the resistance and inductance, but the controller with the updated gain values can control the EMVA system without affecting the performance parameter. The simulation was performed for both forward and backward strokes to investigate the valve seating velocity. It was found that the controller can achieve an acceptable valve seating velocity. Hence, the machine learning tool helps in optimizing the PID controller’s gain values to achieve faster valve operation with an acceptable valve seating velocity.