A temperature compensation control system for laser engraving machine based on multi-source temperature data fusion

Abstract

The invention discloses a temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion, comprising: a temperature acquisition module, comprising a plurality of temperature sensors arranged at key positions of the laser engraving machine, for acquiring overall temperature field distribution data; an algorithm module, comprising: an improved Kalman filter algorithm, for eliminating temperature data noise; a multi-point weighted fusion algorithm, for performing spatial distribution fusion calculation on multi-point temperature data; a thermal deformation prediction algorithm, integrating heat conduction, natural convection and radiation heat transfer models, for predicting overall system temperature changes; an adaptive compensation algorithm, for dynamically adjusting PID parameters according to real-time error feedback; and a compensation execution module, for performing real-time position compensation on the laser engraving machine according to the compensation amount output by the algorithm module.

Description

Technical Field

[0001] The invention relates to a temperature compensation system of a laser engraving machine, and in particular to a temperature compensation control system of a laser engraving machine based on multi-source temperature data fusion. Background Art

[0002] Laser engraving machine is a high-precision CNC processing equipment, widely used in industrial manufacturing, artistic creation and other fields. With the continuous improvement of processing accuracy requirements, thermal deformation compensation control has become a key technology to ensure processing quality. The temperature compensation control system of laser engraving machines on the market is usually composed of temperature sensors, data acquisition, compensation controllers and actuators, which are used to monitor and compensate for errors caused by thermal deformation in real time.

[0003] The existing temperature compensation system of laser engraving machines mainly adopts a single-point temperature measurement method, which obtains local temperature data by installing temperature sensors at key parts of the machine tool. Based on the collected temperature information, the system uses a pre-established linear compensation model to calculate the thermal deformation and perform position compensation. This compensation method can reduce the processing error caused by thermal deformation to a certain extent and improve the processing accuracy.

[0004] However, in the actual processing of laser engraving machines, due to the thermal effect of laser energy and changes in ambient temperature, the mechanical structure undergoes thermal deformation, affecting the processing accuracy. The laser beam generates a large amount of heat during operation, which is transmitted in the machine tool structure through heat conduction, heat convection, etc., resulting in a complex temperature field distribution. The existing single-point temperature measurement method cannot fully reflect the temperature distribution state of the entire system, the temperature information obtained is limited, and it is difficult to accurately describe the spatial distribution characteristics of thermal deformation.

[0005] In addition, the traditional linear compensation algorithm is too simplified and fails to consider the dynamic change characteristics of the temperature field and the nonlinear characteristics of thermal deformation. In the actual processing process, there is a complex thermal coupling relationship between the various components of the machine tool, and the thermal deformation has obvious hysteresis and nonlinear characteristics. Simple linear compensation models are difficult to accurately describe this complex thermal-mechanical coupling effect, resulting in unsatisfactory compensation effects.

[0006] At the same time, the existing compensation algorithm lacks adaptive capabilities and is difficult to cope with the thermal deformation compensation needs under complex working conditions. Under different processing conditions, such as changes in ambient temperature, the characteristics of thermal deformation will also change accordingly. Traditional fixed parameter compensation algorithms cannot automatically adjust the compensation strategy according to changes in working conditions, and it is difficult to ensure the stability of the compensation effect.

[0007] Therefore, it is of great practical significance to develop an intelligent temperature compensation control system based on multi-source temperature data fusion. This system should not only realize comprehensive monitoring of the temperature field and establish an accurate thermal deformation prediction model, but also have adaptive compensation capabilities to meet the strict requirements of modern precision machining on machining accuracy. Summary of the invention

[0008] The purpose of the present invention is to provide a temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion. The intelligent temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion has comprehensive temperature field detection and prediction functions, and has adaptive compensation capabilities.

[0009] The above technical objectives of the present invention are achieved through the following technical solutions:

[0010] A temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion comprises: a temperature acquisition module, comprising a plurality of temperature sensors arranged at key positions of the laser engraving machine, for acquiring overall temperature field distribution data; an algorithm module, comprising: an improved Kalman filter algorithm, for eliminating temperature data noise; a multi-point weighted fusion algorithm, for performing spatial distribution fusion calculation on multi-point temperature data; a thermal deformation prediction algorithm, integrating heat conduction, natural convection and radiation heat transfer models, for predicting overall system temperature changes; an adaptive compensation algorithm, for dynamically adjusting PID parameters according to real-time error feedback; and a compensation execution module, for performing real-time position compensation on the laser engraving machine according to the compensation amount output by the algorithm module.

[0011] The present invention is further configured as follows:

[0012] The filtering equation steps of the improved Kalman filtering algorithm include:

[0013] (1) Perform state prediction: X(k|k-1)=AX(k-1|k-1)+BU(k);

[0014] (2) Calculate the prediction error covariance: P(k|k-1)=AP(k-1|k-1)A T +Q;

[0015] (3) Calculate the Kalman gain: Kg(k)=P(k|k-1)H T [HP(k|k-1)H T +R] -1 ;

[0016] (4) Update state: X(k|k)=X(k|k-1)+Kg(k)[Z(k)-HX(k|k-1)];

[0017] (5) Update error covariance: P(k|k)=[I-Kg(k)H]P(k|k-1);

[0018] Where, X: system state vector, k: current moment, k-1: previous moment, X(k|k-1): prior estimate (predicted value) at moment k, X(k|k): posterior estimate (corrected value) at moment k, X(k-1|k-1): optimal estimate at moment k-1; A: state transfer matrix, describing how the system state is transferred from moment k-1 to moment k; B: control input matrix; H: observation matrix, mapping state space to observation space; I: identity matrix; U(k): control input vector; Z(k): actual measurement value; P(k|k-1): prior estimate error covariance; P(k|k): posterior estimate error covariance; P(k-1|k-1): estimate error covariance at previous moment; Q: process noise covariance matrix; R: measurement noise covariance matrix; Kg(k): Kalman gain; T : Matrix transpose; -1 : Matrix inversion.

[0019] The present invention is further configured as follows: the weighted fusion steps of the temperature data of the multi-point weighted fusion algorithm are: (1) obtaining the temperature value Ti of each measuring point; (2) determining the weight coefficient wi of each measuring point by the least squares method; (3) calculating the temperature value of any point in space: T(x, y, z) = Σ(wi·Ti) / Σwi-(q / k)Δx; wherein q is the heat flux density, k is the thermal conductivity, and Δx is the spatial distance.

[0020] The present invention is further configured as follows: the thermal deformation prediction algorithm step comprises:

[0021] (1) Solve the heat conduction equation:

[0022] ;

[0023] (2) Calculate the boundary heat exchange: q = h (Ts - T∞) + εσ (Ts 4 -T∞ 4 ), where h is the convective heat transfer coefficient, Ts is the surface temperature, T∞ is the ambient temperature, ε is the emissivity, and σ is the Stefan-Boltzmann constant.

[0024] The present invention is further configured as follows:

[0025] The steps of the adaptive compensation algorithm include:

[0026] (1) Calculate the compensation amount: C(t)=Kp·e(t)+Ki∫e(t)dt+Kd·de(t) / dt;

[0027] (2) Dynamically adjust PID parameters:

[0028] dKp / dt=γp·e(t)·|e(t)|

[0029] dKi / dt=γi·e(t)·∫e(t)dt

[0030] dKd / dt=γd·e(t)·de(t) / dt;

[0031] Among them, γp, γi, and γd are learning rate parameters.

[0032] The present invention is further configured as follows: the calculation formula for the thermal deformation is:

[0033]

[0034] Among them, ΔL is the length change, L 0 is the initial length of the component, α is the linear thermal expansion coefficient of the material, T is the current temperature, T 0 is the reference temperature, is the temperature gradient.

[0035] In summary, the present invention has the following beneficial effects: This intelligent temperature compensation control system for laser engraving machines based on multi-source temperature data fusion overcomes the technical limitations of traditional laser engraving machines that only rely on single-point temperature measurement, and proposes a compensation control scheme based on multi-source temperature data fusion. Through the multi-point temperature sensor network arranged at the key parts of the machine tool, combined with the improved Kalman filter and multi-point weighted fusion algorithm, an accurate description of the temperature field distribution of the entire machine tool is achieved. In terms of thermal deformation prediction, the thermodynamic theory is combined with the boundary compensation algorithm, which not only considers the heat conduction equation, but also incorporates the natural convection and radiation heat transfer caused by the ambient temperature into the calculation model, thereby overcoming the problem that the traditional linear compensation model cannot accurately describe the complex thermal-mechanical coupling effect. This intelligent compensation method based on multi-source data fusion significantly improves the system's ability to predict and compensate for thermal deformation under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flowchart of an embodiment. DETAILED DESCRIPTION

[0037] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "page", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicating orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0038] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0039] In the description of the present invention, "multiple" means two or more than two, unless otherwise clearly defined. In the present invention, unless otherwise clearly defined and defined, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be internal communication between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0040] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0041] A temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion includes a temperature acquisition module for collecting temperature data at multiple points, an algorithm module for processing the temperature data at multiple points and calculating the compensation amount, and a compensation execution module for adjusting the compensation amount accordingly. Figure 1 A flowchart of the overall process.

[0042] The temperature acquisition module includes several high-precision PT100 temperature sensors installed on the laser engraving machine, with a sampling frequency of 10Hz and a measurement accuracy of ±0.1°C. According to the different weights of temperature field measurement, this application sets 4 measuring points on the spindle box, 4 measuring points on the workbench, 2 measuring points on the machine column, and 2 measuring points on the ambient temperature. The temperature of each measuring point is measured in real time by the temperature sensor, and the temperature data is collected into the algorithm module.

[0043] 2. Algorithm module

[0044] The algorithm module includes a filtering algorithm for filtering and eliminating noise from the collected multi-point temperature data, a multi-point fusion algorithm for weighting the multi-point temperatures, a thermal deformation prediction algorithm for predicting the temperature changes of the overall system, a boundary compensation algorithm for calculating the compensation amount for the ambient temperature boundary conditions, and an adaptive algorithm for calculating the compensation amount.

[0045] Among them, the filtering algorithm uses the improved Kalman filtering algorithm to eliminate temperature noise, and its filtering equation is:

[0046] ①X(k|k-1)=AX(k-1|k-1)+BU(k)

[0047] ②P(k|k-1)=AP(k-1|k-1)A T +Q

[0048] ③Kg(k)=P(k|k-1)H T [HP(k|k-1)H T +R] -1

[0049] ④X(k|k)=X(k|k-1)+Kg(k)[Z(k)-HX(k|k-1)]

[0050] ⑤P(k|k)=[I-Kg(k)H]P(k|k-1)

[0051] X: system state vector, k: current moment, k-1: previous moment, X(k|k-1): prior estimate (predicted value) at moment k, X(k|k): posterior estimate (corrected value) at moment k, X(k-1|k-1): optimal estimate at moment k-1; A: state transfer matrix, describing how the system state is transferred from moment k-1 to moment k; B: control input matrix; H: observation matrix, mapping state space to observation space; I: identity matrix; U(k): control input vector; Z(k): actual measurement value; P(k|k-1): prior estimate error covariance; P(k|k): posterior estimate error covariance; P(k-1|k-1): estimate error covariance at previous moment; Q: process noise covariance matrix; R: measurement noise covariance matrix; Kg(k): Kalman gain; T : Matrix transpose; -1 : Matrix inversion.

[0052] In the above, formula ① is the state prediction equation, which can predict the temperature change state in the next time state. Formula ② is the prediction error covariance equation, which is used to calculate the uncertainty generated in the prediction process. Formula ③ is the Kalman gain calculation, which determines the weight ratio of the predicted value and the measured value. Formula ④ is the state update equation, which is used to integrate the predicted value and the measured value; formula ⑤ is the update error covariance.

[0053] After the calculation of the above filtering algorithm, the measurement noise of the temperature sensor can be filtered out, providing smoother and more accurate temperature data, improving the measurement accuracy of the system, and improving the stability of subsequent compensation calculation.

[0054] Among them, the multi-point fusion algorithm is mainly used for multi-point temperature field distribution calculation, mainly including the weighted fusion formula of temperature data:

[0055] T(x,y,z)=Σ(wi·Ti) / Σwi-(q / k)Δx

[0056] Ti is the temperature value of the i-th measuring point, and wi is the corresponding weight coefficient, which is determined by the least squares optimization method.

[0057] Through the above-mentioned weighted fusion calculation, the temperature data of multiple points are integrated according to the weights.

[0058] Among them, the thermal deformation prediction algorithm is based on thermodynamics theory, considering the heat conduction equation of the material as follows:

[0059]

[0060] T is the temperature field function, t is the time, D is the thermal diffusion coefficient, ρ is the material density, c is the specific heat capacity; q is the heat flux density.

[0061] x: lateral movement direction of the workbench;

[0062] y: longitudinal movement direction of the workbench;

[0063] z: the direction of the laser head rising and falling;

[0064] : represents the second-order partial derivative of temperature in the x direction, describing the heat conduction in the x direction;

[0065] : represents the second-order partial derivative of temperature in the y direction, describing the heat conduction in the y direction;

[0066] : represents the second-order partial derivative of temperature in the z direction, describing the heat conduction in the z direction;

[0067] By inputting multi-point data into the algorithm module and solving the above partial differential equation, the temperature field distribution of the overall system can be obtained.

[0068] Among them, the boundary conditions of the boundary compensation algorithm considering the ambient temperature take into account the natural convection and radiation heat transfer process, namely:

[0069] q = h (Ts - T∞) + εσ (Ts 4 -T∞ 4 )

[0070] The letters in the above formula have the following meanings:

[0071] q: total surface heat flux (heat flux), in W / m², represents the heat transfer rate per unit area

[0072] Convective heat transfer term h(Ts-T∞):

[0073] h: Convective heat transfer coefficient, in W / (m·K), describes the heat transfer capacity caused by air flow and is related to the air flow state, object shape, etc.

[0074] Ts: Surface temperature of an object, in K (Kelvin)

[0075] T∞: ambient temperature (far-field temperature), unit is K

[0076] Radiation heat transfer term εσ(Ts 4 -T∞ 4 ):

[0077] ε: radiance (emissivity), dimensionless, ranging from 0 to 1, describing the radiation ability of the surface of an object. For a perfect black body, ε=1, and for an actual object, ε<1. σ: Stefan-Boltzmann constant, with a value of 5.67×10 -8 W / (m²·K 4 )

[0078] Ts 4 : The fourth power of the surface temperature of an object

[0079] T∞ 4 : The fourth power of ambient temperature

[0080] The boundary compensation algorithm considering the ambient temperature can be used to calculate the heat loss of the overall structure of the computer tool, predict the distribution of the temperature field, and evaluate the heat dissipation effect. It can optimize the thermal deformation compensation calculation and improve the compensation accuracy.

[0081] After the temperature field distribution calculation is completed, the thermal deformation can be calculated to obtain the compensation required for the multi-point fusion temperature field. The specific formula is as follows:

[0082]

[0083] Where, ΔL: length change (thermal deformation), the unit is usually millimeter (mm)

[0084] L 0 : The initial length of the component, in millimeters (mm)

[0085] α: Linear thermal expansion coefficient of the material, in 1 / ℃ or K-1

[0086] T: current temperature in degrees Celsius (℃) or Kelvin (K)

[0087] T 0 : Reference temperature (initial temperature), in degrees Celsius (℃) or Kelvin (K)

[0088] K: Thermal conductivity

[0089] : Temperature gradient

[0090] The above formula can be used to predict the thermal deformation of each component of the laser engraving machine, and then calculate the compensation amount, and can be used to predict and evaluate the impact of thermal deformation.

[0091] After obtaining the thermal deformation amount of the multi-point fusion temperature field distribution, the compensation amount is calculated accordingly through an adaptive algorithm.

[0092] The adaptive algorithm includes the improved PID control equation as follows:

[0093] C(t)=Kp·e(t)+Ki·∫e(t)dt+Kd·de(t) / dt

[0094] Among them, C(t) is the compensation amount, e(t) is the thermal deformation error, Kp, Ki, Kd are adaptive PID parameters, which are adjusted through real-time error feedback:

[0095] dKp / dt=γ p ·e(t)·|e(t)|

[0096] dKi / dt=γ i ·e(t)·∫e(t)dt

[0097] dKd / dt=γ d ·e(t)·de(t) / dt

[0098] The meanings of the letters in the formula are as follows:

[0099] Kp: Proportional coefficient

[0100] Ki: integration coefficient

[0101] Kd: differential coefficient

[0102] dKp / dt, dKi / dt, dKd / dt: The rate of change of these parameters over time

[0103] e(t): current error, that is, the difference between the target value and the actual value

[0104] |e(t)|: absolute value of error

[0105] ∫e(t)dt: The integral term of the error, indicating the cumulative error

[0106] de(t) / dt: The derivative of the error, indicating the rate of change of the error

[0107] γ p : Learning rate of the proportional term

[0108] γ i : Learning rate of the integral term

[0109] γ d : Learning rate of the derivative term

[0110] dKp / dt=γ p •e(t)•|e(t)|: The adjustment rate of the proportional coefficient is proportional to the current error and the absolute value of the error

[0111] dKi / dt=γ i •e(t)•∫e(t)dt: The adjustment rate of the integral coefficient is proportional to the current error and the accumulated error

[0112] dKd / dt=γ d •e(t)•de(t) / dt: The adjustment rate of the differential coefficient is proportional to the current error and the rate of error change

[0113] Through the above compensation formula, the control parameters can be adaptively adjusted according to the thermal deformation, and the compensation amount can be calculated and adjusted in a targeted manner according to the time algorithm for self-learning in different situations.

[0114] The result of calculating the compensation amount by the above algorithm is as follows:

[0115] X'=X+Cx

[0116] Y'=Y+Cy

[0117] Z'=Z+Cz

[0118] Among them, Cx, Cy, Cz are the compensation values ​​calculated in the x, y, z axis directions, X, Y, Z are the initial position coordinates when the system is started, and X', Y', Z' are the position coordinates after adjustment by the compensation values.

[0119] The compensation execution module includes a high-precision displacement sensor, whose resolution can be set to 0.1 μm, and realizes compensation through the position instruction of the CNC control system of the laser engraving machine to change the corresponding coordinate position.

[0120] In order to better understand the present invention, the present invention provides corresponding experiments to prove the temperature compensation mechanism of the present application. The main points are as follows:

[0121] 1. Setting conditions:

[0122] 1. Set environmental parameters: temperature: 25±1℃; relative humidity: 45±5%; air pressure: 101.325±0.5kPa.

[0123] 2. Set equipment parameters: laser power: 2000W; worktable size: 1500mm×1000mm; spindle speed: 12000r / min; positioning accuracy: ±2μm; repeat positioning accuracy: ±1μm.

[0124] 3. Measuring equipment: 12 PT100 temperature sensors (accuracy: ±0.1℃); 3 high-precision displacement sensors (resolution: 0.1μm); roundness meter (accuracy: 0.1μm); flatness measuring instrument (accuracy: 0.1μm).

[0125] 4. Static thermal characteristics test: duration: 4 hours; sampling interval: 10 minutes; measurement items: temperature of each measuring point, displacement of key positions

[0126] 5. Dynamic processing test: duration: 8 hours; sampling interval: 30 minutes; measurement items: temperature field, thermal deformation, processing accuracy

[0127] 6. Standard test piece processing: Material: 45# steel; Size: Φ100mm×50mm; Processing features: outer circle, end face, step surface

[0128] 2. Data Results

[0129] 1. Complete data of temperature field distribution:

[0130] 2. Thermal deformation data (uncompensated):

[0131] 3. Compensation control parameters adaptively adjust data:

[0132] 4. Position error data after compensation:

[0133] 5. Standard test piece processing accuracy data:

[0134] 3. Results Analysis

[0135] 1. Temperature field characteristics: The temperature rise of the spindle box is the largest, reaching 13.5°C; the temperature rise of the workbench is second, reaching 8.0°C; the temperature rise of the column is the smallest, reaching 5.0°C; the temperature field tends to be stable after 6 hours

[0136] 2. Thermal deformation compensation effect: Uncompensated maximum thermal deformation: 123.2±0.9μm; Maximum residual error after compensation: 6.9±0.2μm; Compensation efficiency: 94.4%; System response time: 85ms

[0137] 3. Processing accuracy analysis: roundness retention: deviation increment <0.9μm; flatness retention: deviation increment <0.8μm; dimensional accuracy: maintained within the range of ±5μm; surface roughness: Ra value increment <0.3μm

[0138] 4. System stability: temperature compensation parameters are adaptively adjusted smoothly; 8 hours of continuous operation without abnormalities; compensation effect fluctuates over time <±0.3μm

[0139] IV. Conclusion

[0140] 1. Compensation effect: The temperature compensation system can reduce the error caused by thermal deformation from 120μm to 7μm, and the compensation efficiency is over 94%.

[0141] 2. Dynamic characteristics: system response time <100ms, compensation parameters are adaptively adjusted stably, and there is no overshoot.

[0142] 3. Processing accuracy: The processing accuracy fluctuation after compensation is controlled within the range of ±5μm, meeting the requirements of high-precision processing.

[0143] 4. System reliability: The 8-hour continuous operation test shows that the system has good stability and reliability, and the compensation effect is long-lasting and stable.

[0144] In order to more conveniently understand the calculation process of the present invention, this application illustrates and provides a complete analysis of the algorithm calculation process:

[0145] 1. Temperature data collection

[0146] Spindle box: 34.8±0.1℃

[0147] Workbench: 29.9±0.2℃

[0148] Column: 27.5±0.15℃

[0149] Ambient temperature: 25.0±0.1℃

[0150] 2. Kalman filter processing

[0151] (1) System parameters: A = 1 (state transition); H = 1 (observation matrix); Q = 0.001 (process noise); R = 0.01 (measurement noise)

[0152] (2) Spindle box filtering:

[0153] X(k|k-1)=34.7℃

[0154] P(k|k-1)=0.041

[0155] Kg(k)=0.804

[0156] X(k|k)=34.78℃

[0157] P(k|k)=0.008

[0158] Filtering results:

[0159] Spindle box: 34.78℃

[0160] Workbench: 29.85℃

[0161] Column: 27.48℃

[0162] Environment: 24.98℃

[0163] 3. Temperature field calculation (considering boundary conditions)

[0164] (1) Boundary heat flux calculation: convection heat transfer coefficient h = 12 W / (m²·K); radiation coefficient ε = 0.92; Stefan-Boltzmann constant σ = 5.67×10 -8 W / (m²·K 4 );Surface temperature Ts=31.22℃=304.37K;Ambient temperature T∞=24.98℃=298.13K

[0165] (2) Heat flux calculation:

[0166] q = h (Ts - T∞) + εσ (Ts 4 -T∞ 4 )

[0167] =12(304.37-298.13)+0.92×5.67×10 -8 (304.37 4 -298.13 4 )

[0168] =74.88W / m²

[0169] (3) Temperature field calculation:

[0170] Material parameters: ρ = 2700 kg / m³ (density); c = 896 J / (kg·K) (specific heat); k = 237 W / (m·K) (thermal conductivity); α = k / (ρc) = 9.8×10 -5 m² / s (thermal diffusivity).

[0171] Temperature gradient:

[0172] (4) Multi-point temperature fusion (considering boundary effects):

[0173] Weights: w1=0.4, w2=0.35, w3=0.25

[0174] T(x,y,z)=(0.4×34.78+0.35×29.85+0.25×27.48) / 1-0.316×Δx=31.22-0.00316=31.217℃

[0175] (5) Solving the heat conduction equation:

[0176]

[0177] Among them, the set time period Δt=1s, Δx=Δy=Δz=0.01m

[0178] =31.217+9.8×10 -5 ×(-0.316)+3.1×10 -5 =31.217℃

[0179] 4. Thermal deformation calculation

[0180] Characteristic length L 0 =800mm; linear expansion coefficient α=23×10 -6 / ℃; temperature difference ΔT=31.217-24.98=6.237℃;

[0181] Deformation calculation:

[0182] ΔL=L 0 αΔT=800×23×10 -6 ×6.237=0.1149mm=114.9μm

[0183] 5.PID compensation calculation

[0184] Initial parameters:

[0185] Kp=0.85,Ki=0.06,Kd=0.03

[0186] Learning rate: γ p =0.001,γ i =0.0001,γ d =0.0002

[0187] Error e(t)=114.9μm

[0188] Parameter update:

[0189] dKp / dt=0.001×114.9×114.9=13.202

[0190] dKi / dt=0.0001×114.9×∫114.9dt=1.377

[0191] dKd / dt=0.0002×114.9×(de / dt)=0.459

[0192] Updated data:

[0193] Updated PID parameters (Δt=1s):

[0194] Kp_new=Kp+dKp / dt×Δt=0.85+13.202×1=14.052

[0195] Ki_new=Ki+dKi / dt×Δt=0.06+1.377×1=1.437

[0196] Kd_new=Kd+dKd / dt×Δt=0.03+0.459×1=0.489

[0197] Take the X direction as an example:

[0198] Cx(t)=Kp_new·ex(t)+Ki_new·∫ex(t)dt+Kd_new·dex(t) / dt

[0199] in:

[0200] ex(t)=114.9×cos(θ x )μm(θ x is the angle between thermal deformation and X axis)

[0201] ∫ex(t)dt=114.9×cos(θ x )×Δtμm·s

[0202] dex(t) / dt=0.2×cos(θ x )μm / s

[0203] Assume the angle with the coordinate axis:

[0204] θ x =12°(X axis)

[0205] θ y =8°(Y axis)

[0206] θ z =5°(Z axis)

[0207] Substitute into the X direction calculation:

[0208] ex(t)=114.9×cos(12°)=112.3μm

[0209] ∫ex(t)dt=112.3×1=112.3μm·s

[0210] dex(t) / dt=0.2×cos(12°)=0.196μm / s

[0211] Cx=14.052×112.3+1.437×112.3+0.489×0.196=112.1μm (take 1 / 15 of the final compensation amount to avoid over-compensation)

[0212] Similarly, calculate the Y direction: ey(t)=114.9×cos(8°)=113.8μm; Cy=113.6μm;

[0213] Z direction: ez(t)=114.9×cos(5°)=114.4μm; Cz=114.3μm

[0214] Compensation calculation:

[0215] Cx=112.1μm

[0216] Cy=113.6μm

[0217] Cz=114.3μm

[0218] 6. Coordinate compensation execution

[0219] Initial position (100.000, 200.000, 300.000) mm

[0220] Coordinates after compensation:

[0221] X'=100.000+0.1121=100.1121mm

[0222] Y'=200.000+0.1136=200.1136mm

[0223] Z'=300.000+0.1143=300.1143mm

[0224] In summary, the present invention has the following beneficial effects: This intelligent temperature compensation control system for laser engraving machines based on multi-source temperature data fusion overcomes the technical limitations of traditional laser engraving machines that only rely on single-point temperature measurement, and proposes a compensation control scheme based on multi-source temperature data fusion. Through the multi-point temperature sensor network arranged at the key parts of the machine tool, combined with the improved Kalman filter and multi-point weighted fusion algorithm, an accurate description of the temperature field distribution of the entire machine tool is achieved. In terms of thermal deformation prediction, the thermodynamic theory is combined with the boundary compensation algorithm, which not only considers the heat conduction equation, but also incorporates the natural convection and radiation heat transfer caused by the ambient temperature into the calculation model, thereby overcoming the problem that the traditional linear compensation model cannot accurately describe the complex thermal-mechanical coupling effect. This intelligent compensation method based on multi-source data fusion significantly improves the system's ability to predict and compensate for thermal deformation under complex working conditions.

[0225] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion, characterized in that: include: The temperature acquisition module includes multiple temperature sensors arranged at key parts of the laser engraving machine to obtain the overall temperature field distribution data; Algorithm modules, including: Improved Kalman filter algorithm to eliminate temperature data noise; Multi-point weighted fusion algorithm is used to perform spatial distribution fusion calculation on multi-point temperature data; Thermal deformation prediction algorithm, integrating heat conduction, natural convection and radiation heat transfer models to predict the temperature change of the entire system; Adaptive compensation algorithm dynamically adjusts PID parameters based on real-time error feedback; A compensation execution module is used to perform real-time position compensation on the laser engraving machine according to the compensation amount output by the algorithm module; The steps of weighted fusion of temperature data of the multi-point weighted fusion algorithm are as follows: (1) obtaining the temperature value Ti of each measuring point; (2) determining the weight coefficient wi of each measuring point by the least squares method; (3) calculating the temperature value of any point in space: T(x, y, z) = Σ(wi·Ti) / Σwi-(q / k)Δx; wherein q is the heat flux density, k is the thermal conductivity, and Δx is the spatial distance.

2. The temperature compensation control system of a laser engraving machine based on multi-source temperature data fusion according to claim 1 is characterized in that: The filtering equation steps of the improved Kalman filtering algorithm include: (1) Perform state prediction: X(k|k-1)=AX(k-1|k-1)+BU(k); (2) Calculate the prediction error covariance: P(k|k-1)=AP(k-1|k-1)A T +Q; (3) Calculate the Kalman gain: Kg(k)=P(k|k-1)H T [HP(k|k-1)H T +R] -1 ; (4) Update state: X(k|k)=X(k|k-1)+Kg(k)[Z(k)-HX(k|k-1)]; (5) Update error covariance: P(k|k)=[I-Kg(k)H]P(k|k-1); Where, X: system state vector, k: current moment, k-1: previous moment, X(k|k-1): prior estimate (predicted value) at moment k, X(k|k): posterior estimate (corrected value) at moment k, X(k-1|k-1): optimal estimate at moment k-1; A: state transfer matrix, describing how the system state is transferred from moment k-1 to moment k; B: control input matrix; H: observation matrix, mapping state space to observation space; I: identity matrix; U(k): control input vector; Z(k): actual measurement value; P(k|k-1): prior estimate error covariance; P(k|k): posterior estimate error covariance; P(k-1|k-1): estimate error covariance at previous moment; Q: process noise covariance matrix; R: measurement noise covariance matrix; Kg(k): Kalman gain; T : Matrix transpose; -1 : Matrix inversion.

3. The temperature compensation control system of a laser engraving machine based on multi-source temperature data fusion according to claim 1 is characterized in that: The thermal deformation prediction algorithm steps include: (1) Solve the heat conduction equation: ; Where T is the temperature field function, t is the time, D is the thermal diffusion coefficient, ρ is the material density, c is the specific heat capacity; q is the heat flux density; x: the lateral movement direction of the workbench; y: the longitudinal movement direction of the workbench; z: the lifting direction of the laser head; : represents the second-order partial derivative of temperature in the x direction, describing the heat conduction in the x direction; : represents the second-order partial derivative of temperature in the y direction, describing the heat conduction in the y direction; : represents the second-order partial derivative of temperature in the z direction, describing the heat conduction in the z direction; (2) Calculate the boundary heat exchange: q = h (Ts - T∞) + εσ (Ts 4 -T∞ 4 ), where h is the convective heat transfer coefficient, Ts is the surface temperature, T∞ is the ambient temperature, ε is the emissivity, and σ is the Stefan-Boltzmann constant.

4. The temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion according to claim 1 is characterized in that: The steps of the adaptive compensation algorithm include: (1) Calculate the compensation amount: C(t)=Kp·e(t)+Ki∫e(t)dt+Kd·de(t) / dt; (2) Dynamically adjust PID parameters: dKp / dt=γp·e(t)·|e(t)| dKi / dt=γi·e(t)·∫e(t)dt dKd / dt=γd·e(t)·de(t) / dt; Among them, γp, γi, and γd are learning rate parameters.

5. The temperature compensation control system for a laser engraving machine based on multi-source temperature data fusion according to claim 3 is characterized in that: The thermal deformation prediction algorithm step also includes thermal deformation calculation, and the specific calculation formula is: ; Where ΔL is the length change, L0 is the initial length of the component, α is the linear thermal expansion coefficient of the material, T is the current temperature, T0 is the reference temperature, is the temperature gradient.

Images