Skip main navigation

Cookies Notification

We use cookies on this site to enhance your user experience. By continuing to browse the site, you consent to the use of our cookies. Learn More
×

System Upgrade on Tue, May 28th, 2024 at 2am (EDT)

Existing users will be able to log into the site and access content. However, E-commerce and registration of new users may not be available for up to 12 hours.
For online purchase, please visit us again. Contact us at [email protected] for any enquiries.
SEARCH GUIDE  

You do not have any saved searches

  • articleNo Access

    Artificial neural network (ANN)-based optimization of a numerically analyzed m-shaped piezoelectric energy harvester

    In this research work, the M-shaped cantilever piezoelectric energy harvester is modeled and optimized using advanced artificial intelligence algorithms. The proposed harvester adopts a single structure geometrical configuration in which two secondary beams are being connected to the principal bimorph. Finite element analysis is carried out on COMSOL Multiphysics to analyze the efficiency of the proposed energy harvester. The influence of frequency, load resistance, and acceleration on the electrical performance of the harvester is numerically investigated to enhance the bandwidth of the piezoelectric vibrational energy harvester. Numerical analysis is also utilized to obtain the iterative dataset for the training of the artificial neural network. Furthermore, a genetic multi-objective optimization approach is implemented on the trained artificial neural network to obtain the optimal parameters for the proposed energy harvester. It is observed that optimization using modern artificial intelligence approaches implies nonlinearities of the system and therefore, machine learning-based optimization has shown more convincing results, as compared to the traditional statistical methods. Results revealed the maximum output values for the voltage and electrical power are 15.34 V and 4.77 mW at 51.19 Hz, 28.09 kΩ, and 3.49 g optimal design input parameters. Based on the outcomes, it is recommended to utilize this reliable harvester in low-power micro-devices, electromechanical systems, and smart wearable devices.