US 12,117,796 B2 · Methods for quickly optimizing geometric error compensation data of translational axes of five-axis numerically controlled machine tools

Abstract

The embodiment of the present disclosure provides a method for quickly optimizing geometric error compensation data of a translational axis of a five-axis numerically controlled machine tool. The method includes: 1) establishing a volumetric positioning error model; 2) establishing an error database; 3) constructing a volumetric error compensation table; 4) establishing a compensation data optimization model to form an optimization model of three face diagonals and one body diagonal in a translational axis linkage mode; 5) completing iterative optimization and selection of the correction coefficients; 6) completing compensation of the geometric errors of the five-axis numerically controlled machine tool; and 7) iterating error correction data to the error database, performing linkage trajectory detection, presetting a positioning error threshold, and cycling the operations 2) to 6) to realize a machine tool precision guarantee system for periodic detection, optimization, and compensation.

A method for optimizing geometric error compensation data of a translational axis of a five-axis numerically controlled machine tool, wherein execution of the method is processor-based, comprising:

(1) establishing a volumetric positioning error model of a tool center point caused by a geometric error of the translational axis of the five-axis numerically controlled machine tool, wherein 18 geometric error vectors including three coordinate axes are:
an X-direction positioning error δx(i), a Y-direction straightness error δy(i), a Z-direction straightness error δz(i), a roll error θα(i), a pitch error θβ(i), and a deflection error θγ(i), i=X, Y, Z, and X, Y, and Z denote operating positions of the coordinate axes within a travel range;
(2) establishing an error database of the 18 geometric error vectors including the three coordinate axes and the establishing an error database of the 18 geometric error vectors including the three coordinate axes includes:
(2.1) identifying the geometric error of the translational axis through a thirteen trajectory line detection manner using a high-precision laser interferometer based on the volumetric positioning error model of the five-axis numerically controlled machine tool established in the operation (1);
(2.2) combining six geometric errors identified for each axis into a form of an error vector:
E
1
×
18
=
[
δ
x
⁢
(
X
)
,
…
,
θ
γ
⁢
(
X
)
︸
six
⁢
items
,
θ
x
⁢
(
Y
)
,
…
,
θ
γ
⁢
(
Y
)
︸
six
⁢
items
,
δ
x
⁢
(
Z
)
,
…
,
θ
γ
⁢
(
Z
)
︸
six
⁢
items
]
;
and
(2.3) extending each error into a position-dependent vector throughout a whole motion travel since the 18 geometric errors of the translational axis are interrelated with a motion position and dividing the motion position into N equal parts, wherein same geometric errors at different positions are all different, and the error database of the translational axis of the numerically controlled machine tool is expressed as follows:
Ebase
N
×
18
=
[
δ
x
(
X
)
1
,
…
,
θ
γ
(
X
)
1
,
δ
x
(
Y
)
1
,
…
,
θ
γ
(
Y
)
1
,
δ
x
(
Z
)
1
,
…
,
θ
γ
(
Z
)
1
…
δ
x
(
X
)
N
,
…
,
θ
γ
(
X
)
N
,
δ
x
(
Y
)
N
,
…
,
θ
γ
(
Y
)
N
,
δ
x
(
Z
)
N
,
…
,
θ
γ
(
Z
)
N
]
;
(3) constructing a volumetric error compensation table of a numerically controlled system;
(4) establishing a compensation data optimization model of a volumetric positioning error of the translational axis, adding correction coefficients k and d to the geometric error vectors obtained in the operation (2), and combining with a translational axis linkage trajectory positioning error model to form an optimization model of three face diagonals and one body diagonal in a translational axis linkage mode;
(4.1) adding the correction coefficients k and d to each error of the geometric error vectors obtained in the operation (2), wherein a basic form is as follows:
e_Adjust=ke+d
where e denotes any one of the geometric error vectors, eAdjust denotes a corrected error, and an expression for correcting an entire geometric error vector is as follows:
Ebase_Adjust
N
×
18
=
[
δ
x
⁢
(
X
)
1
︷
k
1
,
d
1
,
…
,
θ
γ
(
X
)
1
,
δ
x
(
Y
)
1
,
…
,
θ
γ
⁢
(
Y
)
1
︷
k
i
,
d
i
,
δ
x
(
Z
)
1
,
…
,
θ
γ
(
Z
)
1
…
δ
x
⁢
(
X
)
N
︸
k
1
,
d
1
,
…
,
θ
γ
(
X
)
N
,
δ
x
(
Y
)
N
,
…
,
θ
γ
⁢
(
Y
)
N
︸
k
i
,
d
i
,
δ
x
(
Z
)
N
,
…
,
θ
γ
(
Z
)
N
]
where EbaseAdjust is a corrected error database of the translational axis, ki and di denote the correction coefficients required for an ith error, a purpose of adding the correction coefficients is to realize a dynamic change of each error vector within a certain range, which refers to searching for upper and lower bounds by relying on the correction coefficients of ki and di, wherein ki completes scaling adjustment of data, and di completes overall offset adjustment of the data to complete an approximation of a real error, and the adjustment is performed simultaneously on all the geometric errors to achieve accurate error compensation;
(4.2) calculating positioning errors of four diagonals of an XY plane, an XZ plane, a YZ plane, and an XYZ body according to an expression of the volumetric positioning error of the tool center point obtained in the operation (1);
(4.3) obtaining corrected positioning errors of the four diagonals by substituting the geometric error EbaseAdjust after the correction coefficients are added into the positioning errors of the four diagonals calculated in the operation (4.2);
(4.4) obtaining machine tool linkage trajectory detection positioning error data by directly detecting the positioning errors of three face diagonals and one body diagonal during linkage of the translational axis using the laser interferometer;
and
(4.5) calculating differences between the operation (4.3) and the operation (4.4), and obtaining four optimization objectives of the diagonals by taking a smallest sum of squared differences as an optimization objective;
(5) numerically solving vectors K and D composed of the correction coefficients k and d based on a particle swarm optimization algorithm to complete iterative optimization and selection of the correction coefficients;
(6) generating the volumetric error compensation table of the numerically controlled system by combining the correction coefficient vector obtained in the operation (5) and the geometric error vector with the operation (3) and completing compensation of the geometric errors of the five-axis numerically controlled machine tool;
and
(7) iterating error correction data to the error database in the operation (2), performing linkage trajectory detection on the volumetric positioning error of the translational axis by setting a fixed period, presetting a positioning error threshold, and cycling the operations (2) to (6) to realize a machine tool precision guarantee system for periodic detection, optimization, and compensation.

319 more elements

Methods for quickly optimizing geometric error compensation data of translational axes of five-axis numerically controlled machine tools

Life of this patent

Methods for quickly optimizing geometric error compensation data of translational axes of five-axis numerically controlled machine tools

Life of this patent

Abstract

Claims8