Editorial Type:
Article Category: Other
 | 
Online Publication Date: 01 Oct 2013

Piezoelectric Vibration-Based Energy Harvesters for Next-Generation Intelligent Tires

and
Page Range: 262 – 293
DOI: 10.2346/tire.13.410404
Save
Download PDF

ABSTRACT

Because of access limitations inside the tire, the use of batteries for sensor nodes embedded inside an intelligent tire is not practical. The high vibration levels inside a tire have the potential to generate electrical power using vibration-based energy-harvesting techniques. In this article, the feasibility of using an inertial vibrating energy harvester unit to power a sensor module for tire use is assessed. To predict the electrical power output of the generator, a generic analytical model based on the transfer of energy within the system has been derived. The vibration measurements taken from the test conducted using accelerometers embedded in the tire have been applied as an excitation to the model to predict the power output for a device of suitable dimensions and to study the feasibility of this concept. For the tire applications, a special compact harvester design has been proposed that is able to withstand large shocks and vibrations. Suitable mathematical models for different harvester configurations have been developed to identify the best configuration suited for use inside a tire. The harvester unit demonstrates power generation over a wide speed range and provides a distinct advantage in cost and flexibility of installation while extending the lifetime of the power supply for sensor data acquisition and communication. Results indicate the viability of the procedure outlined in the article.

Keywords: energy harvesting; intelligent tire; frequency tuning; artificial neural network (ANN)
FIG. 1
FIG. 1

Intelligent tire system for online tire monitoring.

FIG. 2
FIG. 2

Tire vibration waveform arranged in time series.

FIG. 3
FIG. 3

(a) Radial acceleration signal in time domain. (b) Radial acceleration signal in frequency domain.

FIG. 4
FIG. 4

Typical wireless data transmission system for intelligent tires.

FIG. 5
FIG. 5

(a) Velocity profile for the city-driving schedule. (b) Energy demand analysis for the city.

FIG. 6
FIG. 6

(a) Velocity profile for the highway-driving schedule. (b) Energy demand analysis for the highway cycle.

FIG. 7
FIG. 7

(a) Power limits as a function of the state of charge (SOC) of the cell. (b) Charge-discharge current limits as a function of the state of charge (SOC) of the cell.

FIG. 8
FIG. 8

(a) Instrumented tire assembly. (b) Accelerometer glued to the tire inner liner.

FIG. 9
FIG. 9

Data for the dynamic tests collected using the in-house mobile tire test rig at the Intelligent Transportation Laboratory (ITL), Virginia Tech.

FIG. 10
FIG. 10

(a) Cascade diagram showing the time series data for the tire radial acceleration at different translational speeds. (b) Cascade diagram showing the dependency of the tire vibration spectra on the translational speed.

FIG. 11
FIG. 11

(a) Load dependence study. (b) Pressure dependence study. (c) Speed dependence study. (d) Roughness dependence study.

FIG. 12
FIG. 12

(a) Identifying ideal operating frequency band for the harvester. (b) Comparison between radial acceleration PSD plots for low and high mu surface conditions.

FIG. 13
FIG. 13

Planned mounting location of the harvester.

FIG. 14
FIG. 14

(a) CAD drawing of the harvester casing. (b) Drawing of the harvester placed inside the casing.

FIG. 15
FIG. 15

31-direction: Charge collection in 3 direction and stress in 1 direction.

FIG. 16
FIG. 16

Schematic of the beam showing the parameters used in the simulations.

FIG. 17
FIG. 17

Connection method between the energy-harvesting device and the rectification circuitry.

FIG. 18
FIG. 18

(a) Output power versus electrical resistance. (b) Output current versus electrical resistance. (c) Output power versus frequency.

FIG. 19
FIG. 19

Harvester power dissipated in the optimal resistive load versus operating frequency and squared coupling factor.

FIG. 20
FIG. 20

Basic buck-boost AC-DC switch-mode power converter.

FIG. 21
FIG. 21

Load impedance estimation via neural networks.

FIG. 22
FIG. 22

Radial acceleration signal power on a per-revolution basis: (a) at 30 mph and (b) at 65 mph.

FIG. 23
FIG. 23

Domain extracted signal power: (a) at 30 mph, (b) at 45 mph, and (c) at 65 mph.

FIG. 24
FIG. 24

(a) Sample experimental data set. (b) System architecture for the ANN system.

FIG. 25
FIG. 25

Artificial training data set creation.

FIG. 26
FIG. 26

(a) Neuron model with R inputs. (b) Tan-sigmoid activation function.

FIG. 27
FIG. 27

Multilayer perceptron.

FIG. 28
FIG. 28

Comparison of experimental and ANN-predicted results.

FIG. 29
FIG. 29

Neural network–based impedance matching system for tire energy harvesting.

Contributor Notes

2 Corresponding author. Email: kbsingh@vt.edu
  • Download PDF