Artificial Electric Field Optimization: Charged Particles, Search, and Limitations
Summary
The article presents the Artificial Electric Field Algorithm (AEFA), a population-based optimization method inspired by electrostatic attraction. Candidate solutions are particles in a search space; their fitness determines normalized charges, and forces based on charge and distance guide changes in particle velocity and position. A Coulomb-like constant decays over iterations to shift the search from broader exploration toward local refinement. The article gives the main equations and a procedural outline, then discusses an MQL5 implementation and benchmark comparisons.
Reported strengths include convergence on smooth, low-dimensional functions when enough objective evaluations are available, with relatively few external parameters. The author also reports wide variation across benchmark results, weak performance on discrete problems, poor scalability, and implementation complexity. These are general optimization experiments, not evidence of profitability or suitability for a trading strategy. Results depend on the benchmark functions and implementation, and the article notes that its algorithm descriptions may include modifications from canonical versions.
Key ideas
- AEFA represents candidate solutions as charged particles whose positions encode points in the search space.
- Fitness relative to the current population’s best and worst values determines normalized particle charges.
- Attractive forces use particle charges and inter-particle distances to update acceleration, velocity, and position.
- A decreasing Coulomb constant is intended to alter the balance between broad search and local refinement.
- The reported experiments favor smooth, low-dimensional problems and identify weak discrete performance and low scalability.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.