# US 2021/0173974 A1 · Method for determining parameters of high-frequency vibrating mill with three grinding drums

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0173974 A1
- Document ID: US20210173974A1
- Status: Published application
- Kind: A1
- Application number: 17/045,938
- Filed: 2019-09-12
- Published: 2021-06-10
- Earliest claimed: 2018-12-17
- Assignees: Northeastern University
- Applicants: Northeastern University
- Inventors: Xueliang ZHANG; Chao LI; Shiju CUI; Zhiguo GAO; Hongliang YUE; Zhenmin LI; Hui MA
- Web page: https://patentgrep.com/patent/US20210173974A1

## CPC

- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- B02C 17/24: Driving mechanisms
- G06F 30/10: Geometric CAD
- B02C 17/14: Mills in which the charge to be ground is turned over by movements of the container other than by rotating, e.g. by swinging, vibrating, tilting {(mills provided with vibrators in general B02C19/16)}
- B02C 25/00: Control arrangements specially adapted for crushing or disintegrating
- G06F 2111/10: Numerical modelling
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- B02C: CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN ({household tools and machines for pulverising foodstuffs, e.g. coffee and spice mills A47J42/00; pharmaceutical mortars A61J3/02; mechanical processing of refuse and garbage B03B9/06; dressing mould materials by grinding B22C5/04}; obtaining metallic powder by crushing, grinding or milling B22F9/04; {recovery of plastics by disintegrating B29B17/00; crushing raw materials in starch making C08B30/02; beaters for papermaking D21D1/02; crushing devices specially for transport in mines E21F13/002; slag crushing devices F23J1/00; fuel milling devices in combustion apparatus F23K1/00; household devices for crushing coal F24B15/02; ice disintegrating devices F25C5/02})

## Abstract

A method for determining parameters of a high-frequency vibrating mill with three grinding drums is disclosed. The mathematic modeling is established by applying the average parameter method and transfer function method; the synchronization-stability capability coefficient curve, and the dimensionless coupling torque maximum value diagram of the system are obtained by the characteristic analysis of synchronization and stability. Finally, the curves of rotational velocity of motors, displacements of mass bodies, and phase difference between exciters are obtained by the simulation, and the correctness of the method is verified by the comparison of characteristic analysis and simulation. The parameters of the high-frequency vibrating mill of the present invention can lower the technical requirements of exciters, reduce the loss of the exciters, increase the service life of the mill.

## Claims

### Claim 1

A method for determining parameters of a high-frequency vibrating mill with three grinding drums, wherein a dynamic model of the vibrating mill comprises three exciters, four mass bodies and springs; mass body 1, mass body 2 and mass body 3 are respectively three grinding drums and are all connected with mass body 4 through springs, where mass body 4 is connected with foundation through a spring; the three exciters are respectively located on mass body 1, mass body 2, and mass body 3; the three exciters rotate in the same direction, and each exciter rotates about its rotation axis; a method for determining parameters of the exciters, comprising the following steps:
step 1: establishing a dynamic model and motion differential equations of the system
establishing two rectangular coordinate systems, the three exciters rotate in the same direction, and the motion of each body is divided into vibration in x and y directions and oscillation around the mass center, represented by x, y and ψ respectively;
selecting x, y, ψ, φ1, φ2 and φ3 as generalized coordinates, solving the kinetic energy, potential energy, and energy dissipation functions of a system, and substituting the functions into Lagrange equations to obtain differential equations of motion of the system;
M
1

x
¨
1
+
f
1

(
x
.
1
-
x
.
4
)
+
k
1

(
x
1
-
x
4
)
=
m
01

r

(
ϕ
.
1
2

cos


ϕ
1
+
ϕ
¨
1

sin


ϕ
1
)


M
2

x
¨
2
+
f
2

(
x
.
2
-
x
.
4
)
+
k
2

(
x
2
-
x
4
)
=
m
02

r

(
ϕ
.
2
2

cos


ϕ
2
+
ϕ
¨
2

sin


ϕ
2
)


M
3

x
¨
3
+
f
3

(
x
.
3
-
x
.
4
)
+
k
3

(
x
3
-
x
4
)
=
m
03

r

(
ϕ
.
3
2

cos


ϕ
3
+
ϕ
¨
3

sin


ϕ
3
)


M
4

x
¨
4
-
f
1

(
x
.
1
-
x
.
4
)
-
k
1

(
x
1
-
x
4
)
-
f
2

(
x
.
2
-
x
.
4
)
-
k
2

(
x
2
-
x
4
)
-
f
3

(
x
.
3
-
x
.
4
)
-
k
3

(
x
3
-
x
4
)
+
f
4

x
.
4
+
k
4

x
4
=
0


M
1

y
¨
1
+
f
1

(
y
.
1
-
y
.
4
)
+
k
1

(
y
1
-
y
4
)
=
m
01

r

(
ϕ
.
1
2

sin


ϕ
1
-
ϕ
¨
1

cos


ϕ
1
)


M
2

y
¨
2
+
f
2

(
y
.
2
-
y
.
4
)
+
k
2

(
y
2
-
y
4
)
=
m
02

r

(
ϕ
.
2
2

sin


ϕ
2
-
ϕ
¨
2

cos


ϕ
2
)


M
3

y
¨
3
+
f
3

(
y
.
3
-
y
.
4
)
+
k
3

(
y
3
-
y
4
)
=
m
03

r

(
ϕ
.
3
2

sin


ϕ
3
-
ϕ
¨
3

cos


ϕ
3
)


M
4

y
¨
4
-
f
1

(
y
.
1
-
y
.
4
)
-
k
1

(
y
1
-
y
4
)
-
f
2

(
y
.
2
-
y
.
4
)
-
k
2

(
y
2
-
y
4
)
-
f
3

(
y
.
3
-
y
.
4
)
-
k
3

(
y
3
-
y
4
)
+
f
4

y
.
4
+
k
4

y
4
=
0


J


ψ
¨
+
f
ψ

ψ
.
+
k
ψ

ψ
=
l
0

r

∑
i
=
1
3

m
i

[
ϕ
¨
i

cos

(
ϕ
i
-
β
i
)
-
ϕ
.
i
2

sin

(
ϕ
i
-
β
i
)
]


J
oi

ϕ
¨
i
+
f
di

ϕ
.
i
=
m
oi

r
[
y
¨
i

cos


ϕ
i
-
x
¨
i

sin


ϕ
i
-
l
0

ψ
¨


cos

(
ϕ
i
-
β
i
)
+
l
0

ψ
.
2

sin

(
ϕ
i
-
β
i
)
]
(
1
)
wherein
m0i—mass of each exciter (i=1˜3); mi—mass of each mass body (i=1˜4);
M1=m1+m01, M2=m2+m02, M3=m3+m03, and M4=m4+m01+m02+m03
J=Jm4+(m01+m02+m03)(r2+l2)—moment of inertia of a vibrating rigid body;
Joi=moiri2—moment of inertia of each exciter (i=1˜3);
k
ψ
=
1
2

k
4

(
l
x


3
2
+
l
y


3
2
)
;
ki(i=1˜4)—spring stiffness coefficient;
f
ψ
=
1
2

(
f
1
+
f
2
+
f
3
)

(
l
x


1
2
+
l
x


2
2
+
l
y


1
2
+
l
y


2
2
)
+
1
2

f
4

(
l
x


3
2
+
l
y


3
2
)
-
1
2

(
f
1
+
f
2
+
f
3
)

(
l
x


1

l
x


2
+
l
y


1

l
y


2
)
;
fi(i=1˜4)—damping coefficient;
φi—phase angle of exciter i (i=1˜3); {dot over (φ)}i—angular velocity of exciter i (i=1˜3); {circumflex over (φ)}i—angular acceleration of exciter i (i=1˜3);
step 2: synchronization analysis
letting the phase difference between exciter 1 and exciter 2 be 2α1, the phase difference between exciter 2 and exciter 3 be 2α2, and the average phase difference among the three exciters be φ;
ϕ
1
-
ϕ
2
=
2

α
1
,
ϕ
2
-
ϕ
3
=
2

α
2
,
and


ϕ
=
1
3

∑
i
=
1
3

ϕ
i
rearranging the above equations to obtain
ϕ
1
=
ϕ
+
4
3

α
1
+
2
3

α
2
=
ϕ
+
υ
1


ϕ
2
=
ϕ
-
2
3

α
1
+
2
3

α
2
=
ϕ
+
υ
2


ϕ
3
=
ϕ
-
2
3

α
1
-
4
3

α
2
=
ϕ
+
υ
3
(
2
)
using the transfer function method to obtain the response of the system:
x
1
=
F
1

r


cos

(
ϕ
1
-
γ
1
)
+
F
5

r


cos

(
ϕ
2
-
γ
5
)
+
F
9

r


cos

(
ϕ
3
-
γ
9
)


x
2
=
F
2

r


cos

(
ϕ
1
-
γ
2
)
+
F
6

r


cos

(
ϕ
2
-
γ
6
)
+
F
1

0

r


cos

(
ϕ
3
-
γ
1

0
)
x
3
=
F
3

r


cos

(
ϕ
1
-
γ
3
)
+
F
7

r


cos

(
ϕ
2
-
γ
7
)
+
F
1

1

r


cos

(
ϕ
3
-
γ
1

1
)


x
4
=
F
4

r


cos

(
ϕ
1
-
γ
4
)
+
F
8

r


cos

(
ϕ
2
-
γ
8
)
+
F
1

2

r


cos

(
ϕ
3
-
γ
1

2
)
y
1
=
F
1

r


cos

(
ϕ
1
-
γ
1
)
+
F
5

r


cos

(
ϕ
2
-
γ
5
)
+
F
9

r


cos

(
ϕ
3
-
γ
9
)
y
2
=
F
2

r


cos

(
ϕ
1
-
γ
2
)
+
F
6

r


cos

(
ϕ
2
-
γ
6
)
+
F
1

0

r


cos

(
ϕ
3
-
γ
1

0
)
y
3
=
F
3

r


cos

(
ϕ
1
-
γ
3
)
+
F
7

r


cos

(
ϕ
2
-
γ
7
)
+
F
1

1

r


cos

(
ϕ
3
-
γ
1

1
)
y
4
=
F
4

r


cos

(
ϕ
1
-
γ
4
)
+
F
8

r


cos

(
ϕ
2
-
γ
8
)
+
F
1

2

r


cos

(
ϕ
3
-
γ
1

2
)
ψ
=
r
m

r
l

r
μ
ψ

l
e

∑
i
=
1
3

sin

(
ϕ
i
-
β
i
+
γ
1

3
)
determining intermediate parameters:
a
=
M
4

M
0
3

ω
m


0
8
-
(
3

f
0
2

M
4

M
0
+
9

f
0
2

M
0
2
+
3

k
0

M
0
3
+
3

M
4

M
0
2

k
0
+
k
4

M
0
3
)

ω
m


0
6
+
(
3

f
0
2

k
4

M
0
+
6

k
0
2

M
0
2
+
15

f
0
2

k
0

M
0
+
3

M
4

M
0

k
0
2
+
f
0
4
+
3

k
4

M
0
2

k
0
+
3

f
0
2

M
4

k
0
)

ω
m


0
4
-
(
3

f
0
2

k
0
2
+
3

k
4

M
0

k
0
2
+
M
4

k
0
3
+
3

k
0
3

M
0
+
3

f
0
2

k
4

k
0
)

ω
m


0
2
+
k
4

k
0
3


b
=
-
(
3

M
4

M
0
2

f
0
+
4

f
0

M
0
3
)

ω
m


0
7
+
(
6

M
4

M
0

k
0

f
0
+
15

f
0

M
0
2

k
0
+
3

k
4

M
0
2

f
0
+
6

f
0
3

M
0
+
f
0
3

M
4
)

ω
m


0
5
-
(
6

k
4

M
0

k
0

f
0
+
12

f
0

k
0
2

M
0
+
3

f
0
3

k
0
+
3

M
4

k
0
2

f
0
+
f
0
3

k
4
)

ω
m


0
3
+
(
f
0

k
0
3
+
3

k
4

k
0
2

f
0
)

ω
m


0


c
1
=
-
M
4

M
0
2

ω
m


0
6
+
(
2

M
4

M
0

k
0
+
3

k
0

M
0
2
+
f
0
2

M
4
+
6

f
0
2

M
0
+
k
4

M
0
2
)

ω
m


0
4
-
(
M
4

k
0
2
+
4

k
0
2

M
0
+
2

k
4

M
0

k
0
+
f
0
2

k
4
+
5

f
0
2

k
0
)

ω
m


0
2
+
k
4

k
0
2
+
k
0
3


d
1
=
2

f
0

M
4

M
0
+
4

f
0

M
0
2
)

ω
m


0
2
-
(
2

f
0

k
4

M
0
+
2

f
0

M
4

k
0
+
2

f
0
3
+
10

f
0

M
0

k
0
)

ω
m


0
3
+
2

f
0

k
4

k
0
+
4

f
0

k
0
2
e
1
=
f
0
2

M
0

ω
m


0
4
-
(
k
0
2

M
0
+
3

f
0
2

k
0
)

ω
m


0
2
+
k
0
3
f
1
=
-
(
2

f
0

M
0

k
0
+
f
0
3
)

ω
m


0
3
+
3

f
0

k
0
2

ω
m


0
h
1
=
f
0
2

M
0

ω
m


0
4
-
(
k
0
2

M
0
+
3

f
0
2

k
0
)

ω
m


0
2
+
k
0
3
p
1
=
-
(
2

f
0

M
0

k
0
+
f
0
3
)

ω
m


0
3
+
3

f
0

k
0
2

ω
m


0
u
1
=
(
k
0

M
0
2
+
2

f
0
2

M
0
)

ω
m


0
4
-
(
3

f
0
2

k
0
+
2

k
0
2

M
0
)

ω
m


0
2
+
k
0
3
z
1
=
f
0

M
0
2

ω
m


0
5
-
(
f
0
3
+
4

f
0

M
0

k
0
)

ω
m


0
3
+
3

f
0

k
0
2

ω
m


0
c
2
=
e
1
,
d
2
=
f
1
,
e
2
=
c
1
,
f
2
=
d
1
,
h
2
=
e
1
,
p
2
=
f
1
,
u
2
=
u
1
,
z
2
=
z
1
,

c
3
=
e
1
,
d
3
=
f
1
,
e
3
=
e
1
,
f
2
=
f
1
,
h
3
=
c
1
,
p
3
=
d
1
,
u
3
=
u
1
,
z
3
=
z
1
γ
1
=
{
arctan

bc
1
-
ad
1
ac
1
+
bd
1
,
ac
1
+
bd
1
>
0
π
+
arctan

bc
1
-
ad
1
ac
1
+
bd
1
,
ac
1
+
bd
1
<
0
,

γ
2
=
{
arctan

be
1
-
af
1
ae
1
+
bf
1
,
ae
1
+
bf
1
>
0
π
+
arctan

be
1
-
af
1
ae
1
+
bf
1
,
ae
1
+
bf
1
<
0
γ5=γ2, γ6=γ1, γ7=γ2, γ8=γ4, γ9=γ2, γ10=γ2, γ11=γ1, γ12=γ4
wherein γi(i=1˜4)—lag angle;
M
=
(
M
0
M
0
M
0
M
4
)
,
K
=
(
k
0
-
k
0
k
0
-
k
0
k
0
-
k
0
-
k
0
-
k
0
-
k
0
3

k
0
+
k
4
)


Δ

(
ω
2
)
=
(
k
0
-
M
0

ω
2
-
k
0
k
0
-
M
0

ω
2
-
k
0
k
0
-
M
0

ω
2
-
k
0
-
k
0
-
k
0
-
k
0
3

k
0
+
k
4
-
M
0

ω
2
)
(
5
)
wherein, M—mass coupling matrix, K—stiffness coupling matrix, and Δ(ω2) is an eigenvalue equation,
letting the eigenvalue equation be equal to 0, i.e., Δ(ω2)=0:
k03k4−k03ωm02M4−3ωm02M0k03−3k02ωm02M0k4+3k02ωm04M0M4+6ωm04M02k02+3k0ωm04M02k4−3k0ωm06M02M4−3ωm06M03k0−ωm06M03k4+ωm08M03M4=0
obtaining inherent frequency:
ω
1
=
ω
2
=
ω
0
=
k
0
M
0


ω
3
=
3

M
0

k
0
+
k
0

M
4
+
k
0

M
4
+
(
3

M
0

k
0
+
k
0

M
4
)
2
+
M
0

k
4

(
6

M
0

k
0
+
M
0

k
4
-
2

k
0

M
4
)
M
0

M
4


ω
4
=
3

M
0

k
0
+
k
0

M
4
+
k
0

M
4
-
(
3

M
0

k
0
+
k
0

M
4
)
2
+
M
0

k
4

(
6

M
0

k
0
+
M
0

k
4
-
2

k
0

M
4
)
M
0

M
4
(
6
)
when the three exciters can operate synchronously, {dot over (φ)}=ωm0, taking the second derivative of x1, x2, x3, y1, y2, y3 and ψ to time t, which is substituted into the equations with respect to φi, then integrating the equations over φ=0˜2π, after taking the average value and arranging the equations, the equilibrium equations of the three exciters are obtained as follows:
Te01−f01ωm0=TL1
Te02−f02ωm0=TL2
Te03−f03ωm0=TL3 (7)
wherein
TL1=Tu{2F2[sin(2α1+2α2)+sin(2α1)]cos γ2+2F2[cos(2α1+2α2)+cos(2α1)]sin γ2−l0F13[sin(2α1+2α2−β1+β3)+sin(2α1−β1+β2)]cos γ13+2F1 sin γ1−l0F13[cos(2α1+2α2−β1+β3)+cos(2α1−β1+β2)]sin γ13−l0F13 sin γ13}
TL2=Tu{2F2[sin(2α2)−sin(2α1)]cos γ2+2F2[cos(2α2)+cos(2α1)]sin γ2+l0F13[sin(2α1−β1+β2)−sin(2α2−β2+β3)]cos γ13+2F1 sin γ1−l0F13[cos(2α1−β1+β3)+cos(2α2−β2+β3)]sin γ13−l0F13 sin γ13}
TL3=Tu{−2F2[sin(2α1+2α2)+sin(2α2)]cos γ2+2F2[cos(2α1+2α2)+cos(2α2)]sin γ2+l0F13[sin(2α1+2α2−β1+β3)+sin(2α2−β2+β3)]cos γ13+2F1 sin γ1−l0F13[cos(2α1+2α2−β1+β3)+cos(2α2−β2+β3)]sin γ13−l0F13 sin γ13} (8)
the differences of output torques among the exciters are:
ΔT12=TL1−TL2=Tu{4F2 sin(2α1)cos γ2−2l0F13 sin(2α1−β1+β2)cos γ13+l0F13[sin(2α2−β2+β3)−sin(2α1+2α2−β1+β3)]cos γ13+l0F13[cos(2α2−β2+β3)−cos(2α1+2α2−β1+β3)]sin γ13+2F2[sin(2α1+2α2)−sin(2α2)]cos γ2+2F2[cos(2α1+2α2)−cos(2α2)]sin γ2} (9)
ΔT23=TL2−TL3=Tu{4F2 sin(2α2)cos γ2−2l0F13 sin(2α2−β2+β3)cos γ13+l0F13[sin(2α1−β1+β2)−sin(2α1+2α2−β1+β3)]cos γ13−l0F13[cos(2α1−β1+β2)−cos(2α1+2α2−β1+β3)]sin γ13+2F2[sin(2α1+2α2)−sin(2α1)]cos γ2−2F2[cos(2α1+2α2)−cos(2α1)]sin γ2} (10)
arranging the above two equations to obtain:
Δ

T
0

1

2
T
u
=
τ
c

1

2

(
α
¯
1
,
α
¯
2
)
,
Δ

T
0

2

3
T
u
=
τ
c

2

3

(
α
_
1
,
α
_
2
)
(
11
)
wherein,
τc12=4F2 sin(2α1)cos γ2−2l0F13 sin(2α1−β1+β2)cos γ13+l0F13[sin(2α2−β2+β3)−sin(2α1+2α2−⊕1+β3)]cos γ13+l0F13[cos(2α2−β2+β3)−cos(2α1+2α2−β1+β3)]sin γ13+2F2[sin(2α1+2α2)−sin(2α2)]cos γ2+2F2[cos(2α1+2α2)−cos(2α2)]sin γ2 (12)
τc23=4F2 sin(2α2)cos γ2−2l0F13 sin(2α2−β2+β3)cos γ13+l0F13[sin(2α1−β1+β2)−sin(2α1+2α2−β1+β3)]cos γ13−l0F13[cos(2α1−β1+β2)−cos(2α1+2α2−β1+β3)]sin γ13+2F2[sin(2α1+2α2)−sin(2α1)]cos γ2−2F2[cos(2α1+2α2)−cos(2α1)]sin γ2 (13)
in the above derivation, τc12 (α1, α2) and rc23 (α1, α2) are the dimensionless coupling torques between motor 1 and motor 2 and between motor 2 and motor 3 respectively, and the constraint functions thereof are as follows:
|τc12(α1,α2)|≤τc12 max
|τc23(α1,α2)|≤τc23 max (14)
to sum up, and combined with the above formula, the synchronization criterion of the three exciters can be obtained:

Δ


T
012
T
u

≤
τ
c


12


max



Δ


T
023
T
u

≤
τ
c


23


max
(
15
)
the above formula shows that the absolute value of the difference between the dimensionless residual torques of any two exciters should be less than or equal to the maximum value of the dimensionless coupling torque;
step 3: stability criterion of synchronization status
obtaining the kinetic energy T and potential energy V of the whole system, which are respectively:
T
=
1
2

[
M
1

(
x
.
1
2
+
y
.
1
2
)
+
M
2

(
x
.
2
2
+
y
.
2
2
)
+
M
3

(
x
.
3
2
+
y
.
3
2
)
+
M
4

(
x
.
4
2
+
y
.
4
2
)
+
3

m
0

r
2

ω
0
2
+
J

ψ
.
2
]
(
19
)
V
=
1
2

{
k
1

[
(
x
1
-
x
4
)
2
+
(
y
1
-
y
4
)
2
]
+
k
2

[
(
x
2
-
x
4
)
2
+
(
y
2
-
y
4
)
2
]
+
k
3

[
(
x
3
-
x
4
)
2
+
(
y
3
-
y
4
)
2
]
+
k
4

[
(
x
4
)
2
+
(
y
4
)
2
]
}
(
20
)
average kinetic energy ET and average potential energy EV can be obtained in a single cycle, and average Hamilton action I of the system in one cycle is:
I
=
1
2

π

∫
0
2

π

(
T
-
V
)


d


φ
=
E
T
-
E
v
(
21
)
expressing the Hesse matrix of I as H to obtain:
H
=
[
d
1

1
d
1

2
d
2

1
d
2

2
]
where
d
11
=
∂
2

I
∂
α
_
1
2
=
-
2

r
2

{
[
4

k
0

F
1

F
4

cos

(
γ
1
-
γ
4
)
+
8

k
0

F
2

F
4

cos

(
γ
2
-
γ
4
)
+
4

ω
m


0
2

M
0

F
1

F
2

cos

(
γ
1
-
γ
2
)
-
4

k
0

F
1

F
2

cos

(
γ
1
-
γ
2
)
+
2

ω
m


0
2

M
0

F
2
2
+
2

ω
m


0
2

M
4

F
4
2
-
2

k
0

F
2
2
-
6

k
0

F
4
2
]
×
[
cos

(
2

α
1
)
+
cos

(
2

α
1
+
2

α
2
)
]
+
ω
m


0
2

JF
13
2

[
cos

(
2

α
1
-
β
1
+
β
2
)
+
cos

(
2

α
1
+
2

α
2
-
β
1
+
β
3
)
]
}
d
12
=
d
21
=
∂
2

I
∂
α
1

∂
α
2
=
-
2

r
2

{
[
4

ω
m


0
2

M
0

F
1

F
2

cos

(
γ
1
-
γ
2
)
+
4

k
0

F
1

F
4

cos

(
γ
1
-
γ
4
)
+
8

k
0

F
2

F
4

cos

(
γ
2
-
γ
4
)
-
4

k
0

F
1

F
2

cos

(
γ
1
-
γ
2
)
+
2

ω
m


0
2

M
4

F
4
2
+
2

ω
m


0
2

M
0

F
2
2
-
2

k
0

F
2
2
-
6

k
0

F
4
2
]

cos

(
2

α
1
+
2

α
2
)
+
ω
m


0
2

JF
13
2

cos

(
2

α
1
+
2

α
2
-
β
1
+
β
3
)
}
d
22
=
∂
2

I
∂
α
_
2
2
=
-
2

r
2

{
[
8

k
0

F
2

F
4

cos

(
γ
2
-
γ
4
)
+
4

ω
m


0
2

M
0

F
1

F
2

cos

(
γ
1
-
γ
2
)
-
4

k
0

F
1

F
4

cos

(
γ
1
-
γ
4
)
+
-
4

k
0

F
1

F
2

cos

(
γ
1
-
γ
2
)
+
2

ω
m


0
2

M
0

F
2
2
+
2

ω
m


0
2

M
4

F
4
2
-
2

k
0

F
2
2
-
6

k
0

F
4
2
]
×
[
cos

(
2

α
2
)
+
cos

(
2

α
1
+
2

α
2
)
]
+
ω
m


0
2

JF
13
2

[
cos

(
2

α
2
-
β
2
+
β
3
)
+
cos

(
2

α
1
+
2

α
2
-
β
1
+
β
3
)
]
}
letting
H1=d11
H2=d11d22−d12d21
the H matrix is positive definite and satisfies:
H1>0, H2>0 (23)
defining H1 and H2 as the stability capability coefficients of the system in synchronization condition, the above formulas are the expressions of the stability capability of the system, and when the above formulas (23) are satisfied, the system is stable.

### Claim 2 (depends on 1)

The method for determining parameters of a high-frequency vibrating mill with three grinding drums according to claim 1, wherein TL1, TL2 and TL3 are summed and divided by 3Tu to obtain the average dimensionless load torque of the three exciters, which is expressed as τa (α1, α2);
τ
a
=
1
3

T
u

∑
i
=
1
3


T
_
Li
=
1
3

{
4

F
2

sin


γ
2
[
cos

(
2

α
1
+
2

α
2
)
+
cos

(
2

α
1
)
+
cos


2

α
2
)
]
-
2

l
0

F
13

sin


γ
13

[
cos

(
2


α
1
+
2

α
2
-
β
1
+
β
3
)
+
cos

(
2

α
1
-
β
1
+
β
2
)
+
cos

(
2

α
2
-
β
2
+
β
3
)
]
+
6

F
1

sin


γ
1
-
3

l
0

F
13

sin


γ
13
}
(
16
)
the constraint function of the average dimensionless load torque of the three exciters is as follows:
|τa(α1,α2)|≤τa max (17)
the synchronization capability coefficient is defined as ζij (i, j=1, 2, 3, 4), denoted by:
ζ
12
=
τ
c


12


max
τ
a


max
,
ζ
23
=
τ
c


23


max
τ
a


max
(
18
)
the greater the synchronization capability coefficient is, the stronger the synchronization capability of the system will be, and the easier it is to achieve synchronization.
