Esperanza García Gonzalo

How to design a powerful family of particle swarm optimizers for inverse modelling

In this paper, we show how to design a powerful set of particle swarm optimizers to be applied in inverse modelling. The design is based on the interpretation of the swarm dynamics as a stochastic damped mass-spring system, the so-called particle swarm optimization (PSO) continuous model. Based on this idea we derived a family of PSO optimizers (GPSO, CC-PSO and CP-PSO) having different exploitation and exploration capabilities. Their convergence is related to the stability of their first (mean trajectories)- and second-order moments (variance and temporal covariance). Good parameter sets are located inside their first stability regions close to the upper border of their respective second stability regions where the attraction from the particles oscillation centre is very weak. In this region of weak attraction, both convergence to the global minimum and exploration of the search space are possible. Based on this idea, we have designed a particle–cloud algorithm where each particle in the swarm has different inertia (damping) and acceleration (rigidity) constants. We explored the performance of these algorithms for different PSO members using different benchmark functions, showing that the cloud algorithms have a very good balance between exploration and exploitation. Also, the cloud design helps to avoid two main drawbacks of the PSO algorithm: the tuning of the PSO parameters and the clamping of the particles velocities. We also present the lime and sand algorithm that changes the time step with iterations. This feature helps to avoid entrapment in local minima when the time step is increased, and enables exploration around the global best when the time step is decreased. All these designs are based on the theoretical analysis of the PSO dynamics. We explain how to use this knowledge to the solution and appraisal of inverse problems. Finally, we briefly introduce the combined use of PSO and model reduction techniques to allow posterior sampling in high dimensional spaces.

A New Predictive Model for the State-of-Charge of a High-Power Lithium-Ion Cell Based on a PSO-Optimized Multivariate Adaptive Regression Spline Approach

Batteries play a key role in achieving the target of universal access to reliable affordable energy. Despite their relevant importance, many challenges remain unsolved with regard to the characterization and management of batteries. One of the major issues in any battery application is the estimation of the state-of-charge (SoC). SoC, which is expressed as a percentage, indicates the amount of energy available in a battery. An accurate SoC estimation under realistic conditions improves battery performance, reliability, and lifetime. This paper proposes an SoC estimation method based on a new hybrid model that combines multivariate adaptive regression splines (MARS) and particle swarm optimization (PSO). The proposed hybrid PSO–MARS-based model uses data obtained from a high-power load profile (dynamic stress test) specified by the United States Advanced Battery Consortium (USABC). The results provide comparable accuracy to other more sophisticated techniques but at a lower computational cost.

Modeling the milling tool wear by using an evolutionary SVM–based model from milling runs experimental data

The main aim of this research work is to build a new practical hybrid regression model to predict the milling tool wear in a regular cut as well as entry cut and exit cut of a milling tool. The model was based on Particle Swarm Optimization (PSO) in combination with support vector machines (SVMs). This optimization mechanism involved kernel parameter setting in the SVM training procedure, which significantly influences the regression accuracy. Bearing this in mind, a PSO–SVM–based model, which is based on the statistical learning theory, was successfully used here to predict the milling tool flank wear (output variable) as a function of the following input variables: the time duration of experiment, depth of cut, feed, type of material, etc. To accomplish the objective of this study, the experimental dataset represents experiments from runs on a milling machine under various operating conditions. In this way, data sampled by three different types of sensors (acoustic emission sensor, vibration sensor and ...