Académique Documents
Professionnel Documents
Culture Documents
Energy
journal homepage: www.elsevier.com/locate/energy
State Key Laboratory for Manufacturing System Engineering, Xi'an Jiaotong University, Xi'an, 710049, China
Materials and Structures Centre, Department of Mechanical Engineering, University of Bath, Bath, BA27AY, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 June 2016
Received in revised form
17 November 2016
Accepted 9 December 2016
The complex dynamic behavior of nonlinear harvesters make it difcult to identify the optimum mechanical and electrical parameters for maximum output power, when compared to linear energy harvesting devices. In addition, the chaotic and multi-frequencies characteristics of responses under realistic
human motion excitations provide additional challenges for enhancing the energy harvesting performance, such as the traditional frequency domain method being inappropriate for optimum resistance
selection. This paper provides detailed numerical and experimental investigations into the inuence of
resistance on the efciency of nonlinear energy harvesting from human motions. Numerical simulations
under human motions indicate that optimum resistance of a nonlinear harvester can be attained to
maximize the power output. Moreover, simulations of linear and nonlinear harvesters under harmonic
excitations verify the effectiveness of frequency dominant method to obtain optimum resistance in the
absence of a change in the dynamic behavior of the harvester. However, numerical simulations and
experiments are the effective methods when the harvester shows complex dynamic characteristics.
Experimental measurements of harvested power under different motion speeds and resistances are in
agreement to the numerical analysis for the nonlinear harvester. The results demonstrate the effectiveness of the proposed resistance optimization method for nonlinear energy harvesting from human
motions.
2016 Elsevier Ltd. All rights reserved.
Keywords:
Energy harvesting
Piezoelectric
Nonlinear
Human motion
Optimum load resistance
1. Introduction
Vibration energy harvesting techniques have received considerable attention in recent decades due to its promising ability to
convert ambient vibration energy to useful electrical energy for the
supply of electricity for low-power consumption devices such as
sensors and wireless transceivers [1e3]. In particular, the use of
energy harvesting for body-worn or body-attached applications
has been the subject of a signicant amount of research interest as
it has the potential to power modern low-power sensor systems
and increase the mobility and independence of users [4e6].
Currently, there are a variety of transduction mechanisms based on
piezoelectric [7e9], electromagnetic [10,11], or thermoelectric [12]
effects for converting human kinetic energy and motion to usable
electric energy. Among them, piezoelectric vibration energy harvesting has been considered to be a promising method to harness
* Corresponding author.
E-mail address: caojy@mail.xjtu.edu.cn (J. Cao).
http://dx.doi.org/10.1016/j.energy.2016.12.035
0360-5442/ 2016 Elsevier Ltd. All rights reserved.
222
Fig. 1. (a) Schematic of the nonlinear energy harvester; (b) system swing a certain
angle.
become a linear energy harvester. For the linear conguration, according to Hamilton principle, the variational indicator (VI) should
be zero all the time through Lagrange function. It is shown as
Z t2
I
dEk dEp Ea dt 0
(1)
t1
Z
Z
8 E 0:5 r u_ udV
_
_ p 0:5mtip u_ t xtip ux
_ tip
0:5
rp u_ t udV
t
s
k
s
>
>
>
>
V
V
>
>
Zs
Z p
Z
>
>
<
Ep 0:5 S t TdVs 0:5 S t TdVp 0:5 E t DdVp
>
>
Vs
Vp
Vp
>
>
>
Nf
Nq
>
>
X
X
>
:E
duxi $fi xi
dv$qj
a
i1
j1
(2)
where Vs, Vp are the volume of the substrate and piezoelectric
layers, rs , rp are the material density of substrate and piezoelectric
layers. u is the vector of deection along x direction, while x and
mtip represent the position along the cantilever and the mass of tip
magnets. f is the applied force, v is the applied voltage and q is the
charge. Nf and Nq are the numbers of forces and charges applied to
the cantilever. Also, the subscript t signies the transpose of
matrix, while the subscripts s and p respectively represent the
substrate and piezoelectric layers. Further, S, T, D, E respectively are
the vector of strain, stress, electric displacement and electric eld,
and they satisfy the constitutive equation expressed by
T cE S eT E
D eS s E
(3)
ux; t
N
X
Z
Cp
(4)
C lM bK
8
< v tp ;
E 4dz vt 0;
:
v tp ;
ts =2 < dz < ts =2 tp
ts =2 < dz < ts =2
ts =2 tp < dz < ts =2
(6)
8
< 1 tp ;
4dz 0;
:
1 tp ;
ts =2 < dz < ts =2 tp
ts =2 < dz < ts =2
ts =2 tp < dz < ts =2
(7)
rs FxFt xdVs
Vs
mtip F xtip Ft xtip
00
Vp
00
d2z Fx cs Ft x dVs
K
Vs
(8)
Z
00
00
d2z Fx Ft x dVp
(9)
Vp
8
dUz; bt
>
< M zt C zt
_
qvt Ft
dz
>
: C vt
_ 0
vt=R qzt
p_
q
00
_ Kzt qvt Ft
Mzt C zt
_ 0
_ vt=R qzt
Cp vt
(13)
Fh b0 b1 zt b2 zt2 / bn ztn
(14)
rp FxFt xdVp
(12)
(5)
(11)
4T dz s 4dz dVp
Vp
fi xri t Fxrt
i1
223
_ Fh qvt Ft
Mzt C zt
_ 0
_ vt=R qzt
Cp vt
(15)
When the nonlinear conguration is applied to harvesting energy from human lower-limb motion, the swing motion of lowerlimb will drive the cantilever to swing a certain angle (shown in
Fig. 1(b)), which will result in a time-varying potential energy
function due to the gravity of the cantilever [34]. Thus the fundamental model can be given by
(16)
(10)
Vp
where eT31 is the piezoelectric coupling coefcient, also the capacitance of the system Cp is dened by
224
Fr z; bt Fh mg sinbt
(17)
Ft M wt,
where wt is the base displacement. Applying a
Fourier transform to the equation results in
8
2
2
>
< M U ZU jUCZU KZU qVU M U WU;
>
: Cp jUVU VU qjUZU 0;
R
b
(18)
ZU
M U2 WU
2
K MU2 C C jUq1=R jU
(19)
V2
2R
2R
M 2 q U6 W 2 U
2
2
2
2
KM U CC U2
KCp MCp U2 q CR U2
p
R
(20)
Ropt
v
u
2
u
u
K M U2 U2 C 2
u
u
2
t 2
U KCp MU2 Cp q2 C 2 C 2p U2
(21)
Additionally, the piezoelectric layers can be viewed as a capacitance for the piezoelectric energy harvester, thus on that condition
the equivalent resistance can be calculated by
R0opt
UCp
(22)
Fig. 2. (a) Relationship between optimum resistance and vibration frequency for a
linear harvester; (b) relationship between output power and resistance for linear
harvester under experimental harmonic excitation.
Fig. 3. Potential energy functions of the linear, bistable and tristable harvesters.
electromechanical model (Eq. (16)) and the potential energy function used for a nonlinear tristable energy harvester is shown in
Fig. 3, which also shows the potential energy functions of linear and
bistable systems as they will be compared later. During the
experimental process, human motion signals are measured at
different speeds of 4e8 km/h on a treadmill, and in the next numerical simulation the acceleration and swing angle for speed of
4 km/h and 8 km/h are considered as an input excitation. Fig. 4
shows the recorded acceleration and swing angle history at a
speed of 4 km/h and 8 km/h. It can be seen that the measured acceleration enlarges with an increase of motion speed and there is a
degree of quasi-periodicity and asymmetry. For a speed of 4 km/h,
the swing angle ranges from 60 to 15 , while from 80 to 5 for
speed of 8 km/h. It is also found that the increasing speed of motion
enlarges the backward swing angle and reduces the forward swing
angle.
The acceleration and swing angle data under speed of 4 km/h
and 8 km/h are used to excite the electromechanical model of the
nonlinear tristable energy harvester. The voltage response in the
condition of open circuit connection and the corresponding
225
Fig. 4. Acceleration and swing angle data at different motion speeds: (a) 4 km/h; (b) 8 km/h.
226
Fig. 5. Voltage response and the corresponding frequency spectrum under excitation of different motion speeds: (a) 4 km/h; (b) 8 km/h.
Fig. 6. Relationship between output power and resistance under excitation of different
motion speeds: (a) 4 km/h; (b) 8 km/h.
227
Fig. 7. Voltage response and corresponding power spectrum of the tristable harvester under excitation of different motion speeds for each optimum load resistance: (a) 4 km/h; (b)
8 km/h.
Fig. 8. Relationship between average output power and the load resistance of three harvesters under the two harmonic excitations. (a) LEH, 6.055 Hz; (b) LEH, 8.4 Hz; (c) BEH,
6.055 Hz; (d) BEH, 8.4 Hz; (e) TEH, 6.055 Hz; (f) TEH, 8.4 Hz.
228
Fig. 9. (a) Experiment setup on a treadmill; (b) enlarged view of harvesting device on
human leg.
4. Experimental verication
The experiment setup for verifying the performance of
nonlinear energy harvesting from human motion is shown in
Fig. 9(a), and (b) is the detailed setup of the harvesting device on
human leg. In the experiments, a nonlinear tristable energy
harvester is designed to verify the performance of energy harvesting from human motion. An accelerometer CXL04GP3 and an
angle sensor BWD-VG100 were used to collect the acceleration and
swing angle data of human motion on a treadmill. An oscilloscope
MSOX3052A with a probe resistance (R0) of 10 MU was applied to
Fig. 10. Experimental voltage response and frequency spectrum under excitation of different motion speeds: (a) 4 km/h; (b) 8 km/h.
229
Novel Energy Materials, Engineering Science and Integrated Systems (NEMESIS) (grant numbers 320963); the Program for New
Century Excellent Talents in University (grant numbers NCET-120453).
References
Fig. 11. Relationship between output power and resistance at different motion speeds
of for the nonlinear harvester.
2.31 MU, 1.3 MU and 1.67 MU. In addition, the maximum output
power generated in the experiments is 30.55m W at a speed of
7 km/h.
5. Conclusion
This paper numerically and experimentally investigates the
optimum load resistance of nonlinear energy harvesting from real
human motion excitation. A theoretical model of nonlinear energy
harvester with time-varying potential energy function is established considering the characteristics of human lower-limb motion
and the corresponding experimental system has been built to acquire the data used for numerical investigation. Numerical investigations of a nonlinear tristable energy harvesting from human
motion indicate that there always exists an optimum load resistance to maximize the average output power. For comparison,
simulations of linear, bistable, and tristable energy harvesters under harmonic excitation are carried out and results show that the
frequency domain analysis method can only be used to calculate
the optimum resistance in the absence of a change in the dynamic
behavior of the harvester. When the behavior of the cantilever
shows complex dynamic characteristics, such as the random and
chaotic phenomena, the calculation method by the combination of
simulation results and dominant frequency shown in power spectrum is a more effective way to determine the optimum resistance.
Experimental results under a variety of motion speed treadmill
tests and load resistances are in agreement with simulation results
to identify the optimum load resistance for nonlinear energy harvesting from human motion and demonstrate the effectiveness of
the proposed numerical method. In addition, the obtained
maximum output power of nonlinear energy harvesting from human motion under optimum load resistance is 30.55m W at speed of
7 km/h for running, which can be collected to supply electricity for
low-power consumption devices such as leadless pacemaker and
blood pressure sensor.
Acknowledgment
This work was supported by the National Natural Science
Foundation of China (grant numbers 51575426, 51421004,
51611530547); the National Key Scientic Instrument and Equipment Development Project (grant number 2012YQ03026101); the
[1] Bowen CR, Kim HA, Weaver PM, Dunn S. Piezoelectric and ferroelectric materials and structures for energy harvesting applications. Energy Environ Sci
2014;7:25e44.
[2] Harne RL, Wang KW. A review of the recent research on vibration energy
harvesting via bistable systems. Smart Mater Struct 2013;22:023001.
[3] Pellegrini SP, Tolou N, Schenk M, Herder JL. Bistable vibration energy harvesters: a review. J Intell Mater Syst Struct 2012;24:1303e12.
[4] Green PL, Papatheou E, Sims ND. Energy harvesting from human motion and
bridge vibrations: an evaluation of current nonlinear energy harvesting solutions. J Intell Mater Syst Struct 2013;24:1494e505.
[5] Rome LC, Flynn L, Goldman EM, Yoo TD. Generating electricity while walking
with loads. Science 2005;309:1725e8.
[6] Wei S, Hu H, He SY. Modeling and experimental investigation of an impactdriven piezoelectric energy harvester from human motion. Smart Mater
Struct 2013;22:105020.
[7] Xie L, Cai M. Increased piezoelectric energy harvesting from human footstep
motion by using an amplication mechanism. Appl Phys Lett 2014;105:
143901.
[8] Mateu L, Moll F. Optimum piezoelectric bending beam structures for energy
harvesting using shoe inserts. J Intell Mater Syst Struct 2005;16:835e45.
[9] Kluger JM, Sapsis TP, Slocum AH. Robust energy harvesting from walking vibrations by means of nonlinear cantilever beams. J Sound Vib 2015;341:
174e94.
[10] Papatheou E, Sims ND. Developing a hardware in-the-loop simulator for a
backpack energy harvester. J Intell Mater Syst Struct 2012;23:827e35.
[11] Ylli K, Hoffmann D, Willmann A, Becker P, Folkmer B, Manoli Y. Energy harvesting from human motion: exploiting swing and shock excitations. Smart
Mater Struct 2015;24:025029.
[12] Kim MK, Kim MS, Lee S, Kim C, Kim YJ. Wearable thermoelectric generator for
harvesting human body heat energy. Smart Mater Struct 2014;23:105002.
B, Baglio S, Trigona C. Improved energy
[13] Ferrari M, Ferrari V, Guizzetti M, Ando
harvesting from wideband vibrations by nonlinear piezoelectric converters.
Sens Actuat A Phys 2010;162:425e31.
[14] Masana R, Daqaq MF. Energy harvesting in the super-harmonic frequency
region of a twin-well oscillator. J Appl Phys 2012;111:044501.
[15] Harris P, Skinner W, Bowen CR, Kim HA. Manufacture and characterisation of
piezoelectric broadband energy harvesters based on asymmetric bistable
cantilever laminates. Ferroelectrics 2015;480:67e76.
[16] Zhao S, Erturk A. On the stochastic excitation of monostable and bistable
electroelastic power generators: relative advantages and tradeoffs in a
physical system. Appl Phys Lett 2013;102:103902.
[17] Mann BP, Sims ND. Energy harvesting from the nonlinear oscillations of
magnetic levitation. J Sound Vib 2009;319:515e30.
[18] Barton DAW, Burrow SG, Clare LR. Energy harvesting from vibrations with a
nonlinear oscillator. J Vib Acoust 2010;132:021009.
[19] Stanton SC, McGehee CC, Mann BP. Reversible hysteresis for broadband
magnetopiezoelastic energy harvesting. Appl Phys Lett 2009;95:174103.
[20] Erturk A, Inman DJ. Broadband piezoelectric power generation on high-energy
orbits of the bistable Dufng oscillator with electromechanical coupling.
J Sound Vib 2011;330:2339e53.
[21] Zhou S, Cao J, Erturk A, Lin J. Enhanced broadband piezoelectric energy harvesting using rotatable magnets. Appl Phys Lett 2013;102:173901.
[22] Zhou S, Cao J, Wang W, Liu S, Lin J. Modeling and experimental verication of
doubly nonlinear magnet-coupled piezoelectric energy harvesting from
ambient vibration. Smart Mater Struct 2015;24:055008.
[23] Oumbe Tekam GT, Kwuimy CA, Woafo P. Analysis of tristable energy harvesting system having fractional order viscoelastic material. Chaos 2015;25:
013112.
[24] Zhou S, Cao J, Inman DJ, Lin J, Liu S, Wang Z. Broadband tristable energy
harvester: modeling and experiment verication. Appl Energy 2014;133:
33e9.
[25] Li H, Qin W, Lan C, Deng W, Zhou Z. Dynamics and coherence resonance of tristable energy harvesting system. Smart Mater Struct 2016;25:015001.
[26] Cao JY, Zhou SX, Wang W, Lin J. Inuence of potential well depth on nonlinear
tristable energy harvesting. Appl Phys Lett 2015;106:173903.
[27] Kim P, Seok J. A multi-stable energy harvester: dynamic modeling and
bifurcation analysis. J Sound Vib 2014;333:5525e47.
[28] Cammarano A, Neild S, Burrow S, Wagg D, Inman D. Optimum resistive loads
for vibration-based electromagnetic energy harvesters with a stiffening
nonlinearity. J Intell Mater Syst Struct 2014;25:1757e70.
[29] Liang JR, Liao WH. Improved design and analysis of self-powered synchronized switch interface circuit for piezoelectric energy harvesting systems.
IEEE Trans Ind Electron 2012;59:1950e60.
[30] Roundy S, Leland ES, Baker J, Carleton E, Reilly E, Lai E, et al. Improving power
output for vibration-based energy scavengers. IEEE Pervasive Comput 2005;4:
28e36.
230
[31] Roundy S, Wright PK. A piezoelectric vibration based generator for wireless
electronics. Smart Mater Struct 2004;13:1131e42.
[32] Wang H, Meng Q. Analytical modeling and experimental verication of
vibration-based piezoelectric bimorph beam with a tip-mass for power harvesting. Mech Syst Signal Process 2013;36:193e209.
[33] Zhao S, Erturk A. Electroelastic modeling and experimental validations of
piezoelectric energy harvesting from broadband random vibrations of cantilevered bimorphs. Smart Mater Struct 2013;22:015002.
[34] Cao J, Wang W, Zhou S, Inman DJ, Lin J. Nonlinear time-varying potential
bistable energy harvesting from human motion. Appl Phys Lett 2015;107:
143904.