Harvester Optimizer

HarvesterOpt: Harvester Optimizer is Matlab code designed to solve optimization problems aimed at maximizing the energy recovery of bistable piezo-magneto-elastic energy harvesting systems. This tool leverages the Cross-Entropy method to enhance the performance and efficiency of energy harvesters, making it a valuable asset for researchers and engineers in the field of energy harvesting.

HarvesterOpt is the optimization module of the package STONEHENGE - Suite for Nonlinear Analysis of Energy Harvesting Systems, which can be downloaded here.

Table of Contents

• Overview
• Features
• Usage
• Documentation
• Videos
• Author
• Citing HarvesterOpt
• License
• Institutional support
• Funding

Overview

HarvesterOpt was developed to optimize the amount of energy recovered by a nonlinear bistable piezoelectric energy harvester. The underlying results are reported in the following publication:

• A. Cunha Jr, Enhancing the performance of a bistable energy harvesting device via the cross-entropy method, Nonlinear Dynamics, 103:137-155, 2021 DOI

Preprint available here.

Features

• Optimizes energy recovery in bistable piezo-magneto-elastic energy harvesting systems
• Utilizes advanced optimization algorithms for maximum efficiency
• Supports customization of optimization parameters and constraints
• Includes visualization tools for simulation results
• Offers well-documented, easy-to-understand code
• Designed with an educational approach for intuitive use

Usage

To get started with HarvesterOpt, follow these steps:

1. Clone the repository:

   git clone https://github.com/americocunhajr/HarvesterOpt.git
   

2. Navigate to the code directory:

   cd HarvesterOpt/HarvesterOpt-1.0
   

3. For direct search optimization, execute:

   HarvesterOpt_DirectSearch
   

4. For cross-entropy optimization, execute:

   HarvesterOpt_CrossEntropy
   

This package includes the following files:

• CrossEntropyOpt_Example.m. – script with an one dimensional optimization with the CE method
• HarvesterOpt_DirectSearch.m – script with a program to optimize the harveters via direct serach
• HarvesterOpt_CrossEntropy.m – script with a program to optimize the harveters via CE method
• PiezoMagBeam_ObjFunc.m. – function that defines the objetive function
• PiezoMagBeam_ObjFuncNoise.m – function that defines the objetive function with noise
• PiezoMagBeam_RHS.m – function that defines the harvester equations
• test01chaos.m – function to compute the 0-1 test for chaos
• trandn.m – function to draw samples from truncated distribution
• plot_ce_animation.m – function to plot an animation with CE search for the optimum
• graph_contour_pnt.m – function to plot a contourmap and a given point
• graph_binarymap.m – function to plot a binary contourmap

Documentation

The routines in HarvesterOpt are well-commented to explain their functionality. Each routine includes a description of its purpose, as well as inputs and outputs.

Videos

Animations illustrating the operation of the CE method can be seen on YouTube.

Author

• Americo Cunha Jr

Citing HarvesterOpt

If you use HarvesterOpt in your research, please cite the following publication:

• A. Cunha Jr, Enhancing the performance of a bistable energy harvesting device via the cross-entropy method, Nonlinear Dynamics, 103:137-155, 2021 https://doi.org/10.1007/s11071-020-06109-0

@article{CunhaJr2021p137,
   author  = {A. {Cunha~Jr}},
   title   = {Enhancing the performance of a bistable energy harvesting device via the cross-entropy method},
   journal = {Nonlinear Dynamics},
   year    = {2021},
   volume  = {103},
   pages   = {137-155},
   note    = {https://doi.org/10.1007/s11071-020-06109-0},
}

License

HarvesterOpt is released under the MIT license. See the LICENSE file for details. All new contributions must be made under the MIT license.

Institutional support

Funding