Hair dryer surface treatment method based on intelligent multi-axis linkage technology

Abstract

The invention discloses a blower surface treatment method based on an intelligent multi-axis linkage technology, and relates to the technical field of intelligent manufacturing and automatic control. Comprising the following steps: S1, collecting a vibration signal in real time through a multi-source sensor network; s2, preprocessing the vibration signal to obtain a vibration component, and determining a dominant vibration source according to the frequency band energy of the vibration component; s3, generating a dynamic suppression strategy according to the dominant vibration source; and S4, nozzle track deviation is obtained in real time through a laser displacement sensor, and the proportional gain and the integral gain of a PID controller are adjusted. Frequency domain separation is carried out on multi-source coupling vibration signals, a leading vibration source is recognized from the signals, a suppression strategy is dynamically generated, meanwhile, the nozzle track deviation is corrected to be within 0.03 mm in real time through a laser displacement sensor, the problem of misjudgment caused by frequency spectrum aliasing in a traditional method is solved, and the method is suitable for large-scale popularization and application. And the track precision and the coating uniformity of special-shaped curved surface spraying are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent manufacturing and automatic control, and specifically to a surface treatment method for a hair dryer based on intelligent multi-axis linkage technology. Background Art

[0002] Traditional surface treatment methods for hair dryers often rely on single-axis operation, which has low efficiency and is difficult to achieve uniform treatment of complex surfaces. In addition, traditional manual or semi-automatic treatment methods are prone to inconsistent product quality and increase production costs.

[0003] With the progress of industrial automation and intelligent manufacturing technologies, intelligent multi-axis linkage technology has been widely applied and developed. This technology can achieve efficient and precise processing of complex curved surfaces by precisely controlling the synchronous operation of multiple motion axes.

[0004] Chinese Patent No. CN105425586A discloses an online active vibration suppression method for milling. First, an online active control architecture in the frequency domain is constructed based on an improved frequency domain LMS algorithm. Then, an iterative formula for the weight coefficient is derived. Again, a hybrid error convergence criterion is constructed by fusing the frequency point error and the global frequency domain error, which is used to control the iterative update of the weight coefficient of the algorithm and the determination of the control convergence point. Finally, taking the surface vibration of the milling workpiece as the control target and a large driving force piezoelectric excitation device as the secondary excitation source, the online active vibration suppression of milling is realized. This method is based on the frequency domain, and only one FFT operation is performed in one control cycle without IFFT operation, which can effectively ensure the control efficiency. The construction of the hybrid error criterion can effectively improve the self-adaptability and anti-interference ability of the algorithm to a certain extent.

[0005] In the multi-axis high-speed spraying operation of existing and similar surface treatment methods for hair dryer surface treatment equipment, since the vibration signal is caused by multi-source coupling interference, including motor electromagnetic noise (high frequency, 1000 - 5000 Hz), mechanical transmission gear meshing impact (medium frequency, 200 - 800 Hz), and high-speed air flow turbulence excitation (broadband, 50 - 2000 Hz), traditional passive shock absorption devices and single-frequency band suppression strategies will misjudge the dominant excitation source due to the inability to separate the mixed vibration components. For example, in the spraying scenario of an irregular curved surface, the broadband vibration caused by air flow turbulence will be superimposed on the medium-frequency vibration of mechanical transmission. Traditional FFT spectrum analysis is limited by insufficient frequency resolution and spectrum aliasing effects, and it is difficult to accurately identify the energy-concentrated frequency band. At the same time, after this error is transmitted to the active suppression system, it will cause the output of phase mismatch damping force, which not only cannot cancel the target vibration, but will instead amplify the energy of the local frequency band, resulting in fluctuations in the contact pressure between the nozzle and the workpiece, and ultimately causing spraying trajectory deviation and uneven coating thickness, seriously restricting the yield rate of high-precision surface treatment processes and the reliability of equipment. Summary of the Invention

[0006] The purpose of the present invention is to provide a surface treatment method for a hair dryer based on intelligent multi-axis linkage technology to solve the problems raised in the above background technology.

[0007] To achieve the above purpose, the present invention provides the following technical solutions: A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology, including:

[0008] S1: Through a multi-source sensor network, vibration signals are collected in real time, and the vibration signals include electromagnetic noise signals, mechanical impact signals, and airflow turbulence signals;

[0009] S2: Through a hybrid frequency-domain separation algorithm, the vibration signals are preprocessed to obtain corresponding vibration components, and the dominant vibration source is determined according to the frequency band energy of the vibration components, including:

[0010] S2.1: Adaptive wavelet packet decomposition: Determine the number of sub-bands and the theoretical frequency resolution through a preset decomposition layer, and obtain the frequency range and energy of each sub-band according to the number of sub-bands and the theoretical frequency resolution;

[0011] S2.2: Determine the dominant vibration source: Compare the energy ratios of the electromagnetic noise signal, mechanical impact signal, and airflow turbulence signal to determine the largest energy ratio, and the signal source corresponding to the largest energy ratio is the dominant vibration source;

[0012] S3: Generate a dynamic suppression strategy according to the dominant vibration source, and execute the dynamic suppression strategy through a multi-axis cooperation instruction;

[0013] S4: Through a laser displacement sensor, the nozzle trajectory deviation is obtained in real time, and the proportional gain and integral gain of the PID controller are adjusted according to the magnitude of the nozzle trajectory deviation. Specifically:

[0014] When the magnitude of the nozzle trajectory deviation is greater than 0.03 mm, adjust the proportional gain and integral gain of the PID controller, and repeat steps S1 - S4 until the nozzle trajectory deviation is not greater than 0.03 mm. Otherwise, the proportional gain and integral gain of the PID controller are not adjusted.

[0015] Furthermore, obtaining the frequency range and energy of each sub-band includes:

[0016] S2.1.1: Determine the decomposition layer and frequency resolution: Set the decomposition layer according to the preset decomposition layer and the number of signals actually collected per second, and determine the theoretical frequency resolution. Specifically:

[0017]

[0018] Where: N is the number of sub - bands, Δf the is the theoretical frequency resolution, f s is the sampling rate per second, and n is the preset decomposition level;

[0019] S2.1.2: Determine the frequency range and energy of the sub - bands: According to the decomposition level and the theoretical frequency resolution, determine the frequency range and energy of each sub - band, specifically:

[0020]

[0021] Where: f i is the frequency magnitude of the i - th sub - band, i is the index of the sub - band, f s is the sampling rate per second, n is the preset decomposition level, N is the number of sub - bands, E i is the energy of the i - th sub - band, W i，k is the k - th wavelet coefficient in the i - th sub - band, k is the index of the wavelet coefficient, N tot is the total number of sampling points of the signal;

[0022] S2.1.3: Perform energy screening: According to the energy of each sub - band, determine the energy proportion of each sub - band, and at the same time compare the energy proportion of the sub - band with the preset energy ratio, and perform energy screening on the sub - band according to the comparison result, specifically:

[0023] When the energy proportion of the sub - band is not less than the preset energy ratio, the energy of the sub - band is retained, otherwise, the energy of the sub - band is deleted.

[0024] Furthermore, according to the frequency range of each sub - band, determine the frequency difference of each sub - band, and compare the frequency difference of each sub - band with the theoretical frequency resolution, and at the same time, according to the comparison result, adjust the preset decomposition level to determine the final decomposition level, specifically:

[0025] When the frequency difference of the sub - band is greater than the theoretical frequency resolution, adjust the preset decomposition level and repeat step S2.1.2 until the frequency difference of the sub - band is not greater than the theoretical frequency resolution, otherwise, the preset decomposition level corresponding to the sub - band is the final decomposition level.

[0026] Furthermore, according to the resolution of the target frequency and the number of signal points actually collected per second, adjust the preset decomposition level, specifically:

[0027]

[0028] Where: n′ is the adjusted preset decomposition layer number, Δf tar is the resolution of the target frequency, f s is the sampling rate per second.

[0029] Furthermore, compare the theoretical frequency resolution with the preset frequency resolution, and determine the actual frequency resolution according to the comparison result, specifically:

[0030] When the theoretical frequency resolution is greater than the preset frequency resolution, the equivalent frequency resolution is obtained by the sliding window overlapping segmentation averaging method and the theoretical frequency resolution until the equivalent frequency resolution is not greater than the preset frequency resolution, then the equivalent frequency resolution is the actual frequency resolution; otherwise, the theoretical frequency resolution is the actual frequency resolution.

[0031] Furthermore, obtaining the equivalent frequency resolution includes:

[0032] M1: Determine the total number of sliding windows: Determine the number of sampling points in the window and the total number of sliding windows through the window time length, specifically:

[0033]

[0034] Where: N win is the number of sampling points in the window, T win is the time length of the window, M is the total number of windows, N tot is the total number of sampling points of the signal, ε is the overlapping ratio between windows, f s is the sampling rate per second;

[0035] M2: Obtain the equivalent frequency resolution: Determine the equivalent frequency resolution according to the theoretical frequency resolution and the total number of sliding windows, and at the same time compare the equivalent frequency resolution with the preset frequency resolution, and determine the final equivalent frequency resolution according to the comparison result, specifically:

[0036] When the equivalent frequency resolution is not greater than the preset frequency resolution, the equivalent frequency resolution is the final equivalent frequency resolution; otherwise, repeat steps M1 and M2, increase the window time length until the equivalent frequency resolution is not greater than the preset frequency resolution.

[0037] Furthermore, implementing the dynamic suppression strategy includes:

[0038] S3.1: Electromagnetic harmonic suppression: When the dominant vibration source is an electromagnetic noise signal, determine the real-time resonance frequency of the motor system through spectrum analysis, and adjust the driving frequency of the motor system according to the real-time resonance frequency. At the same time, determine the final driving frequency according to the comparison result between the sub-band energy ratio of the electromagnetic noise signal and the preset energy ratio threshold. Specifically:

[0039] When the sub-band energy ratio of the electromagnetic noise signal is greater than the preset energy ratio threshold, increase the driving frequency of the motor system until the sub-band energy ratio of the electromagnetic noise signal is not greater than the preset energy ratio threshold. The driving frequency corresponding to not greater than the preset energy ratio threshold is the final driving frequency. Otherwise, the driving frequency of the motor system is the final driving frequency;

[0040] S3.2: Acceleration smoothing: When the dominant vibration source is a mechanical shock signal, adjust the acceleration of the gearbox, and smooth the acceleration derivative of the gearbox by planning the motion trajectory with an S-curve;

[0041] S3.3: Turbulence attenuation: When the dominant vibration source is an air flow turbulence signal, obtain the output signal of the PID controller and adjust the flow rate and shaft speed of the air pump.

[0042] Furthermore, smoothing the acceleration derivative of the gearbox includes:

[0043] S3.2.1: Obtain the acceleration and deceleration times: Determine the acceleration time and deceleration time according to the preset jerk change rate and the maximum acceleration. Specifically:

[0044]

[0045] Where: t1 is the acceleration time, t3 is the deceleration time, a max is the maximum jerk, J max is the preset jerk change rate;

[0046] S3.2.2: Obtain the constant speed time: Determine the constant speed time according to the target operating speed and the acceleration time. Specifically:

[0047]

[0048] Where: t2 is the constant speed time, v tar is the target operating speed, t1 is the acceleration time, a max is the maximum jerk;

[0049] S3.2.3: Acceleration adjustment: Adjust the acceleration of the gearbox in real time according to the acceleration time, deceleration time, constant speed time and the preset jerk change rate to reduce the acceleration slope.

[0050] Furthermore, adjusting the flow rate and shaft speed of the air pump includes:

[0051] S3.3.1: Determine the adjustment coefficient: Compare the energy proportion of the air flow turbulence signal with the preset turbulence energy proportion threshold range, and determine the adjustment coefficient according to the comparison range. Specifically:

[0052] When the energy proportion of the air flow turbulence signal is greater than the preset maximum turbulence energy proportion threshold, increase the adjustment coefficient; when the energy proportion of the air flow turbulence signal is less than the preset minimum turbulence energy proportion threshold, decrease the adjustment coefficient; otherwise, the adjustment coefficient remains unchanged.

[0053] S3.3.2: Obtain the target flow rate: Determine the corresponding target flow rate according to the adjustment coefficient and shaft speed. Specifically:

[0054]

[0055] Where: Q tar is the target flow rate of the air pump, k is the adjustment coefficient, and v tua is the shaft speed of the air pump;

[0056] S3.3.3: Obtain the output signal of the PID controller: Obtain the output signal of the PID controller according to the target flow rate and actual flow rate of the air pump. Specifically:

[0057] u(t) = K p ·(Q tar - Q act ) + K i ·∫(Q tar - Q act )dt

[0058] Where: u(t) is the control output signal, K p is the proportional gain, Q act is the actual flow rate of the air pump, K i is the integral gain, t is the time, and Q tar is the target flow rate of the air pump.

[0059] Furthermore, adjusting the proportional gain and integral gain of the PID controller includes:

[0060] S4.1: Determine the nozzle trajectory deviation: Obtain the deviation between the actual position and the target position of the nozzle through the laser triangulation method. Specifically:

[0061]

[0062] Where: Δd is the deviation between the actual position and the target position of the nozzle, Δx is the displacement of the CCD pixel, L is the reference distance from the laser head to the nozzle, and f is the focal length of the lens;

[0063] S4.2: Adjust the proportional gain and integral gain: According to the deviation between the actual position and the target position of the nozzle, adjust the proportional gain and integral gain of the PID controller, specifically:

[0064]

[0065] Where: K′ p is the adjusted proportional gain, K′ i is the adjusted integral gain, K p0 is the initial proportional gain, α is the adjustment coefficient, Δd is the deviation between the actual position and the target position of the nozzle, and t is the time.

[0066] Compared with the prior art, the beneficial effects of the present invention are:

[0067] Firstly: The present invention performs frequency-domain separation on the obtained multi-source coupled vibration signals through the adaptive wavelet packet decomposition algorithm, identifies the dominant vibration source therefrom, and dynamically generates a suppression strategy. At the same time, through the laser displacement sensor, the nozzle trajectory deviation is corrected in real time to within 0.03 mm, solving the misjudgment problem caused by spectral aliasing in the traditional method and improving the trajectory accuracy and coating uniformity of the special-shaped surface spraying;

[0068] Secondly: The present invention adopts a resonant frequency avoidance strategy for electromagnetic noise, reduces the electromagnetic energy ratio from 42.5% to 3.2% by adjusting the motor drive frequency in real time, implements an S-curve acceleration planning for mechanical shocks, reduces the peak value of gear meshing impact by 67% by constraining the jerk change rate, establishes a flow-axis speed nonlinear control model for turbulent disturbances, and cooperates with the PID gain adaptive adjustment to attenuate the airflow pulsation energy by 81%. Description of the Drawings

[0069] Figure 1 is a schematic flow chart of the hair dryer surface treatment method of the present invention;

[0070] Figure 2 is a schematic flow chart of the adaptive wavelet packet decomposition of the present invention. Detailed Embodiments

[0071] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0072] In the multi-axis high-speed spraying operation of the hair dryer surface treatment equipment, since the vibration signal is caused by multi-source coupling interference, including motor electromagnetic noise (high frequency, 1000 - 5000 Hz), mechanical transmission gear meshing impact (medium frequency, 200 - 800 Hz), and high-speed air flow turbulence excitation (broadband, 50 - 2000 Hz), the traditional passive shock absorption device and single-band suppression strategy will misjudge the dominant excitation source due to the inability to separate the mixed vibration components. For example, in the case of spraying on an irregular curved surface, the broadband vibration caused by air flow turbulence will be superimposed on the medium-frequency vibration of mechanical transmission. The traditional FFT spectrum analysis is limited by insufficient frequency resolution and spectral aliasing effects, making it difficult to accurately identify the frequency band with concentrated energy. At the same time, after this error is transmitted to the active suppression system, it will cause the output of phase-mismatched damping force, which not only cannot cancel the target vibration, but will instead amplify the energy of the local frequency band, resulting in fluctuations in the contact pressure between the nozzle and the workpiece, and ultimately causing spraying trajectory deviation and uneven coating thickness, severely restricting the yield rate of high-precision surface treatment processes and the reliability of the equipment. And in this application, a multi-source sensor network is used to collect multi-source coupled vibration signals of electromagnetic noise, mechanical shock, and air flow turbulence in real time, and an adaptive wavelet packet decomposition algorithm is used to perform frequency-domain separation on the mixed signals. At the same time, the dominant vibration source (such as electromagnetic resonance, gear impact, or turbulence excitation) is accurately identified according to the proportion of sub-band energy, and a suppression strategy is dynamically generated (such as avoiding the resonance frequency band by adjusting the motor drive frequency, smoothing the mechanical acceleration using an S-curve, and adjusting the air pump flow rate and shaft speed in combination with a PID controller). At the same time, a laser displacement sensor is introduced for closed-loop feedback to correct the nozzle trajectory deviation to within 0.03 mm in real time, solving the misjudgment problem caused by spectral aliasing in the traditional method and improving the trajectory accuracy and coating uniformity of spraying on irregular curved surfaces.

[0073] Example 1

[0074] Reference Figure 1 and Figure 2 According to this example, a hair dryer surface treatment method based on intelligent multi-axis linkage technology is provided. The hair dryer surface treatment method includes the following steps:

[0075] Step S1: Through the set multi-source sensor network, collect vibration signals in real time. The vibration signals include electromagnetic noise signals with a frequency range of 1000 - 5000 Hz, mechanical shock signals with a frequency range of 200 - 800 Hz, and air flow turbulence signals with a frequency range of 50 - 2000 Hz.

[0076] In this embodiment, high-frequency vibration sensors are installed on the non-driving side of the motor stator end. Specifically, the high-frequency vibration sensors are symmetrically installed at intervals of 120° centered on the motor axis to eliminate signal deviation caused by installation eccentricity. Further, intermediate-frequency vibration sensors are installed on the bearing seats of the input shaft and output shaft of the gearbox, that is, near the gear meshing point, to capture tooth surface impact signals. At the same time, three-axis acceleration sensors are used to obtain the radial, axial, and tangential vibrations of the gearbox. Further, broadband vibration sensors are installed on the inner wall of the nozzle air chamber flow channel, that is, at a position not more than 50 mm away from the air flow outlet and close to the turbulence source.

[0077] That is to say, high-frequency vibration signals at the motor end, namely electromagnetic noise signals, are obtained through high-frequency vibration sensors. Intermediate-frequency vibration signals at the gearbox, namely mechanical impact signals, are obtained through intermediate-frequency vibration sensors. Broadband vibration signals in the nozzle air chamber, namely air flow turbulence signals, are obtained through broadband vibration sensors.

[0078] Step S2: Through the hybrid frequency domain separation algorithm, preprocess the mixed signals of the electromagnetic noise signals, mechanical impact signals, and air flow turbulence signals obtained in step S1 to obtain the corresponding vibration components therefrom, decompose the obtained vibration components, and calculate and obtain their corresponding frequency band energies to determine the dominant vibration source. Specifically as follows:

[0079] Step S2.1: Adaptive wavelet packet decomposition. That is, through the preset decomposition level, determine the corresponding number of sub-bands and the theoretical frequency resolution, and thus obtain the frequency range and energy of each sub-band according to the determined number of sub-bands and the theoretical frequency resolution. Specifically as follows:

[0080] Step S2.1.1: Determine the decomposition level and frequency resolution. That is, according to the preset decomposition level, set the decomposition level, which is the number of sub-bands. At the same time, according to the number of signal points actually collected per second, determine its corresponding theoretical frequency resolution, specifically:

[0081]

[0082] Where: N is the number of sub-bands, Δf the is the theoretical frequency resolution, f s is the sampling rate per second, and n is the preset decomposition level.

[0083] In the process of specific implementation, the preset decomposition level is set to 5-layer decomposition. That is to say, the number of sub-bands is 32 sub-bands, and at the same time, the sampling rate per second is 10 kHz, so its theoretical frequency resolution is 156.25 Hz.

[0084] Step S2.1.2: Determine the frequency range and energy of the sub - bands. That is, according to the number of sub - bands and the theoretical frequency resolution obtained in Step S2.1.1, determine the frequency range and corresponding energy of each sub - band. Specifically:

[0085]

[0086] Where: f i is the frequency magnitude of the i - th sub - band, i is the index of the sub - band, f s is the sampling rate per second, n is the preset decomposition level, N is the number of sub - bands, E i is the energy of the i - th sub - band, W i，k is the k - th wavelet coefficient in the i - th sub - band, k is the index of the wavelet coefficient, N tot is the total number of signal sampling points.

[0087] It should be noted that according to the frequency range obtained for each sub - band, determine the corresponding frequency difference for each sub - band, and compare the frequency difference of each sub - band with the theoretical frequency resolution in Step S2.1.2. Then, according to the comparison result, adjust the corresponding preset decomposition level. Specifically as follows:

[0088] Step W1: According to the frequency range obtained for each sub - band, obtain the corresponding frequency difference for each sub - band. Specifically:

[0089]

[0090] Where: Δf i is the frequency difference of the i - th sub - band, i is the index of the sub - band, f s is the sampling rate per second, n is the preset decomposition level.

[0091] Furthermore, compare the obtained frequency difference of the sub - band with its corresponding theoretical frequency resolution, and adjust the preset decomposition level according to the comparison result. Specifically:

[0092] When the obtained frequency difference of the sub - band is greater than the theoretical frequency resolution, execute Step W2, adjust the preset decomposition level, and repeat Step S2.1.2 until the obtained frequency difference of the sub - band is not greater than the theoretical frequency resolution. Otherwise, the preset decomposition level corresponding to this sub - band is the final decomposition level.

[0093] Step W2: Adjust the preset decomposition level according to the resolution of the target frequency and the number of actually collected signal points per second. Specifically:

[0094]

[0095] Where: n′ is the adjusted preset decomposition level, Δf tar is the resolution of the target frequency, f s is the sampling rate per second.

[0096] Step S2.1.3: Perform energy screening. That is, according to the energy of each sub-band obtained in step S2.1.2, determine the energy proportion of each sub-band, specifically:

[0097]

[0098] Where: w i is the energy proportion of the i-th sub-band, E i is the energy of the i-th sub-band, i is the index of the sub-band, and N is the number of sub-bands.

[0099] Furthermore, compare the energy proportion of each sub-band with a preset energy ratio (which is specifically set according to actual needs, so it is not specifically set in this embodiment), and according to the comparison result, perform energy screening of the sub-bands, specifically:

[0100] When the energy proportion of the sub-band is not less than the preset energy ratio, the energy of the sub-band is retained; otherwise, the energy of the sub-band is deleted.

[0101] During the specific implementation process, the sampling rate per second is set to 10 kHz, and the preset decomposition level is set to 5. Specifically, the number of sub-bands is 32, that is, the total energy of 32 sub-bands is 2000 V 2 , where the energy of the second sub-band is 850 V 2 , and the energy of the 32nd sub-band is 5 V 2 . That is to say, the energy proportion of the second sub-band is 42.5%, and its energy proportion is greater than the preset energy ratio (5%), so the energy of the second sub-band is retained. The energy proportion of the 32nd sub-band is 0.25%, and its energy proportion is less than the preset energy ratio (5%), so the energy of the 32nd sub-band is deleted.

[0102] Step S2.2: Determine the dominant vibration source. According to the adaptive wavelet packet decomposition in step S2.1, respectively obtain the energy proportions of the electromagnetic noise signal, the mechanical shock signal, and the airflow turbulence signal, and determine the maximum energy proportion among the energy proportions of the electromagnetic noise signal, the mechanical shock signal, and the airflow turbulence signal. The signal source with the maximum energy proportion is the dominant vibration source.

[0103] Step S3: Generate a corresponding suppression strategy according to the dominant vibration source determined in step S2.2. Specifically as follows:

[0104] Step S3.1: Electromagnetic harmonic suppression. That is, when the dominant vibration source is an electromagnetic noise signal, the real-time resonance frequency of the motor system is determined through spectrum analysis, and the driving frequency of the motor system is adjusted according to the determined real-time resonance frequency. Specifically:

[0105]

[0106] Where: f d is the driving frequency of the motor system, f res is the resonance frequency of the motor system, and d is the adjustment threshold.

[0107] Furthermore, repeat steps S2.1.1 - S2.1.3 to obtain the sub-band energy ratio of the electromagnetic noise signal, and compare the sub-band energy ratio of the electromagnetic noise signal with a preset energy ratio threshold (which is set according to specific actual requirements, so it is not specifically set in this embodiment). At the same time, according to the comparison result, adjust the driving frequency of the motor system to determine the final driving frequency. Specifically:

[0108] When the sub-band energy ratio of the electromagnetic noise signal is greater than the preset energy ratio threshold, increase the size of the adjustment threshold and increase the driving frequency of the motor system until the sub-band energy ratio of the electromagnetic noise signal is not greater than the preset energy ratio threshold. Then, the driving frequency corresponding to the not greater than the preset energy ratio threshold is the final driving frequency. Otherwise, the driving frequency of the motor system is the final driving frequency.

[0109] During the specific implementation process, the resonance frequency of the motor system is set to 1200 Hz, and the initial adjustment threshold is 50 Hz. That is to say, the prohibited interval of the driving frequency of the motor system is [1150, 1250]. That is to say, when the motor allows low-speed operation, the setting interval of the driving frequency of the motor system is [0, 1150). When the motor needs to maintain a high speed, the setting interval of the driving frequency of the motor system is [1250, f max , and this f max is the maximum driving frequency of the motor system.

[0110] Step S3.2: Acceleration smoothing. That is, when the dominant vibration source is a mechanical shock signal, adjust the acceleration of the gearbox. Specifically, plan the motion trajectory through an S-curve to make the acceleration derivative of the gearbox smooth. Specifically as follows:

[0111] Step S3.2.1: Obtain the acceleration and deceleration time. That is, determine the acceleration time and deceleration time according to the preset jerk change rate and maximum acceleration. Specifically:

[0112]

[0113] Where: t1 is the acceleration time, t3 is the deceleration time, a max is the maximum jerk, J max is the preset jerk change rate.

[0114] Step S3.2.2: Obtain the constant-speed time. That is, determine the constant-speed time according to the target operating speed and the acceleration time, specifically:

[0115]

[0116] Where: t2 is the constant-speed time, v tar is the target operating speed, t1 is the acceleration time, a max is the maximum jerk.

[0117] Step S3.2.3: Acceleration adjustment. That is, according to the acceleration time and deceleration time obtained in step S3.2.1, the constant-speed time obtained in step S3.2.2, and the preset jerk change rate, adjust the acceleration of the gearbox in real time to reduce the acceleration slope.

[0118] Step S3.3: Turbulence attenuation. That is, when the dominant vibration source is the air flow turbulence signal, obtain the output signal of the PID controller and adjust the flow rate and shaft speed of the air pump. Specifically as follows:

[0119] Step S3.3.1: Determine the adjustment coefficient. That is, compare the energy ratio of the air flow turbulence signal with the preset turbulence energy ratio threshold range, and determine the adjustment coefficient according to the comparison range, specifically:

[0120] When the energy ratio of the air flow turbulence signal is greater than the preset maximum turbulence energy ratio threshold, increase the adjustment coefficient; when the energy ratio of the air flow turbulence signal is less than the preset minimum turbulence energy ratio threshold, decrease the adjustment coefficient; otherwise, the adjustment coefficient remains unchanged.

[0121] In this embodiment, the adjustment formula of the adjustment coefficient is specifically:

[0122]

[0123] Where: k new is the adjusted adjustment coefficient, k cur is the initial adjustment coefficient, W a is the energy ratio of the air flow turbulence signal, is the preset minimum large turbulence energy ratio threshold, is the preset minimum turbulence energy ratio threshold.

[0124] Step S3.3.2: Obtain the target flow rate. That is, determine the corresponding target flow rate according to the adjustment coefficient and shaft speed determined in step S3.3.1, specifically:

[0125]

[0126] Where: Q tar is the target flow rate of the air pump, k is the adjustment coefficient, and v tua is the shaft speed of the air pump.

[0127] Step S3.3.3: Obtain the output signal of the PID controller. That is, according to the target flow rate of the air pump and the actual flow rate of the air pump obtained in step S3.3.2, obtain the output signal of the PID controller, specifically:

[0128] u(t) = K p ·(Q tar - Q act ) + K i ·∫(Q tar - Q act )dt

[0129] Where: u(t) is the control output signal, K p is the proportional gain, Q act is the actual flow rate of the air pump, K i is the integral gain, t is the time, and Q tar is the target flow rate of the air pump.

[0130] Step S4: Obtain the nozzle trajectory deviation in real time through a laser displacement sensor, and adjust the proportional gain and integral gain of the PID controller according to the nozzle trajectory deviation. Specifically as follows:

[0131] Step S4.1: Determine the nozzle trajectory deviation. That is, through the laser triangulation method, obtain the deviation between the actual position and the target position of the nozzle, specifically:

[0132]

[0133] Where: Δd is the deviation between the actual position and the target position of the nozzle, Δx is the displacement of the CCD pixel, L is the reference distance from the laser head to the nozzle, and f is the focal length of the lens.

[0134] Specifically, when the deviation between the actual position and the target position of the nozzle is greater than 0.03 mm, then execute step S4.2 to adjust the proportional gain and integral gain of the PID controller, repeatedly obtain the control output signal of the PID controller, and at the same time repeat steps S1 - S41 until the deviation between the actual position and the target position of the nozzle is not greater than 0.03 mm.

[0135] Step S4.2: Adjust the proportional gain and integral gain. That is, according to the nozzle trajectory deviation determined in step S4.1, adjust the proportional gain and integral gain of the PID controller to re-obtain the output signal of the PID controller.

[0136] In this embodiment, the adjustment formulas for the proportional gain and integral gain of the PID controller are specifically as follows:

[0137]

[0138] Where: K' p is the adjusted proportional gain, K' i is the adjusted integral gain, K p0 is the initial proportional gain, α is the adjustment coefficient, Δd is the deviation between the actual position and the target position of the nozzle, and t is the time.

[0139] During the specific implementation process, if the deviation between the actual position and the target position of the nozzle is 0.05 mm, the corresponding deviation rate is 3 mm / s. At the same time, the initial proportional gain is set to 2 and the adjustment coefficient is set to 0.1. Then the adjusted proportional gain is 2.6 and the adjusted integral gain is 0.5.

[0140] Embodiment 2

[0141] This embodiment provides a surface treatment method for a hair dryer based on intelligent multi-axis linkage technology. The specific implementation method is the same as that of Embodiment 1. The difference is that during the determination of the theoretical frequency resolution, the theoretical frequency resolution is compared with the preset frequency resolution, and based on the comparison result, the actual frequency resolution is determined. The following is an example of the present invention in combination with the specific implementation manner of this embodiment.

[0142] In this embodiment, the theoretical frequency resolution is compared with the preset frequency resolution, and based on the comparison result, the actual frequency resolution is determined, specifically as follows:

[0143] When the theoretical frequency resolution is greater than the preset frequency resolution, the equivalent frequency resolution is obtained through the sliding window overlapping segmented averaging method and the theoretical frequency resolution until the equivalent frequency resolution is not greater than the preset frequency resolution. Then the obtained equivalent frequency resolution is the actual frequency resolution. Otherwise, the theoretical frequency resolution is the actual frequency resolution.

[0144] In this embodiment, the equivalent frequency resolution is obtained through the sliding window overlapping segmented averaging method and the theoretical frequency resolution. Specifically as follows:

[0145] Step M1: Determine the total number of sliding windows. That is, through the set window time length, the number of sampling points within the window is determined. At the same time, through the number of sampling points within the window, the total number of sliding windows is determined. Specifically as follows:

[0146]

[0147] Where: N winis the number of sampling points within the window, T win is the time length of the window, M is the total number of windows, N tot is the total number of sampling points of the signal, ε is the overlapping ratio between windows, f s is the sampling rate per second.

[0148] In the process of specific implementation, the sampling rate per second is 10 kHz, the time length of the window is 0.2 s, so the number of sampling points within the window is 2000 points. At the same time, the total number of sampling points of the signal is 100000 points, and the overlapping ratio between windows is 50%. Therefore, the total number of sliding windows is 98. That is to say, the signal is divided into 98 windows, the number of sampling points set within each window is 2000 points, and the number of overlapping sampling points between adjacent windows is 1000 points.

[0149] Step M2: Obtain the equivalent frequency resolution. That is, according to the theoretical frequency resolution and the total number of windows of the sliding window, determine the equivalent frequency resolution, specifically:

[0150]

[0151] Where: Δf equ is the equivalent frequency resolution, Δf the is the theoretical frequency resolution, M is the total number of windows.

[0152] Furthermore, compare the obtained equivalent frequency resolution with the preset frequency resolution. When the equivalent frequency resolution is not greater than the preset frequency resolution, the obtained equivalent frequency resolution is the actual frequency resolution. Otherwise, repeat Step M1 and Step M2, and reset the window time length, that is, increase the window time length until the equivalent frequency resolution is not greater than the preset frequency resolution.

[0153] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended embodiments and their equivalents.

Claims

1. A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology, characterized in that, It includes: S1: Real-time collect vibration signals through a multi-source sensor network, where the vibration signals include electromagnetic noise signals, mechanical shock signals, and airflow turbulence signals; S2: Preprocess the vibration signals through a hybrid frequency-domain separation algorithm to obtain corresponding vibration components, and determine the dominant vibration source according to the frequency band energy of the vibration components, including: S2.1: Adaptive wavelet packet decomposition: Determine the number of sub-bands and the theoretical frequency resolution through a preset decomposition level, and obtain the frequency range and energy of each sub-band according to the number of sub-bands and the theoretical frequency resolution; S2.2: Determine the dominant vibration source: Compare the energy ratios of the electromagnetic noise signal, mechanical shock signal, and airflow turbulence signal to determine the largest energy ratio, and the signal source corresponding to the largest energy ratio is the dominant vibration source; S3: Generate a dynamic suppression strategy according to the dominant vibration source, and execute the dynamic suppression strategy through a multi-axis collaborative command; S4: Real-time obtain the nozzle trajectory deviation through a laser displacement sensor, and adjust the proportional gain and integral gain of the PID controller according to the magnitude of the nozzle trajectory deviation, specifically: When the magnitude of the nozzle trajectory deviation is greater than 0.03 mm, adjust the proportional gain and integral gain of the PID controller, and repeat steps S1 - S4 until the nozzle trajectory deviation is not greater than 0.03 mm. Otherwise, the proportional gain and integral gain of the PID controller are not adjusted.

2. The surface treatment method of a hair dryer based on intelligent multi-axis linkage technology according to claim 1, characterized in that, Obtain the frequency range and energy of each sub-band, including: S2.1.1: Determine the decomposition level and frequency resolution: Set the decomposition level according to the preset decomposition level and the number of signals actually collected per second to determine the theoretical frequency resolution, specifically: Where: N is the number of sub - bands, Δf the is the theoretical frequency resolution, f s is the sampling rate per second, and n is the preset decomposition level; S2.1.2: Determine the frequency range and energy of the sub-band: Determine the frequency range and energy of each sub-band according to the decomposition level and the theoretical frequency resolution, specifically: where: f i is the frequency magnitude of the i-th sub-band, i is the index of the sub-band, f s is the sampling rate per second, n is the preset decomposition level, N is the number of sub-bands, E i is the energy of the i-th sub-band, W i，k is the k-th wavelet coefficient in the i-th sub-band, k is the index of the wavelet coefficient, N tot is the total number of sampling points of the signal; S2.1.3: Perform energy screening: Determine the energy ratio of each sub-band according to the energy of each sub-band, and at the same time compare the energy ratio of the sub-band with a preset energy ratio, and perform energy screening of the sub-band according to the comparison result, specifically: When the energy ratio of the sub-band is not less than the preset energy ratio, the energy of the sub-band is retained. Otherwise, the energy of the sub-band is deleted.

3. The surface treatment method of a hair dryer based on intelligent multi-axis linkage technology according to claim 2, wherein, Determine the frequency difference of each sub-band according to the frequency range of each sub-band, compare the frequency difference of each sub-band with the theoretical frequency resolution, and at the same time adjust the preset decomposition level according to the comparison result to determine the final decomposition level, specifically: When the frequency difference of the sub-band is greater than the theoretical frequency resolution, adjust the preset decomposition level and repeat step S2.1.2 until the frequency difference of the sub-band is not greater than the theoretical frequency resolution. Otherwise, the preset decomposition level corresponding to the sub-band is the final decomposition level.

4. A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology according to claim 3, characterized in that, Adjust the preset decomposition level according to the resolution of the target frequency and the number of signals actually collected per second, specifically: Where: n′ is the adjusted preset decomposition level, Δf tar is the resolution of the target frequency, f s is the sampling rate per second.

5. The surface treatment method of a hair dryer based on intelligent multi-axis linkage technology according to claim 2, wherein Compare the theoretical frequency resolution with the preset frequency resolution, and determine the actual frequency resolution according to the comparison result, specifically: When the theoretical frequency resolution is greater than the preset frequency resolution, obtain the equivalent frequency resolution through the sliding window overlapping segmentation averaging method and the theoretical frequency resolution until the equivalent frequency resolution is not greater than the preset frequency resolution, then the equivalent frequency resolution is the actual frequency resolution; otherwise, the theoretical frequency resolution is the actual frequency resolution.

6. The surface treatment method of a hair dryer based on intelligent multi-axis linkage technology according to claim 5, wherein, Obtaining the equivalent frequency resolution includes: M1: Determine the total number of sliding windows: Determine the number of sampling points within the window and the total number of sliding windows through the window time length, specifically: Where: N win is the number of sampling points within the window, T win is the time length of the window, M is the total number of windows, N tot is the total number of sampling points of the signal, ε is the overlapping ratio between windows, f s is the sampling rate per second; M2: Obtain the equivalent frequency resolution: Determine the equivalent frequency resolution according to the theoretical frequency resolution and the total number of sliding windows, and at the same time compare the equivalent frequency resolution with the preset frequency resolution, and determine the final equivalent frequency resolution according to the comparison result, specifically: When the equivalent frequency resolution is not greater than the preset frequency resolution, the equivalent frequency resolution is the final equivalent frequency resolution; otherwise, repeat steps M1 and M2, increase the window time length until the equivalent frequency resolution is not greater than the preset frequency resolution.

7. A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology according to claim 1, characterized in that, Executing the dynamic suppression strategy includes: S3.1: Electromagnetic harmonic suppression: When the dominant vibration source is an electromagnetic noise signal, determine the real-time resonance frequency of the motor system through spectrum analysis, and adjust the driving frequency of the motor system according to the real-time resonance frequency. At the same time, determine the final driving frequency according to the comparison result between the sub-band energy ratio of the electromagnetic noise signal and the preset energy ratio threshold, specifically: When the sub-band energy ratio of the electromagnetic noise signal is greater than the preset energy ratio threshold, increase the driving frequency of the motor system until the sub-band energy ratio of the electromagnetic noise signal is not greater than the preset energy ratio threshold, then the driving frequency corresponding to not greater than the preset energy ratio threshold is the final driving frequency; otherwise, the driving frequency of the motor system is the final driving frequency; S3.2: Acceleration smoothing: When the dominant vibration source is a mechanical shock signal, adjust the acceleration of the gearbox and smooth the acceleration derivative of the gearbox by planning the motion trajectory with an S-curve. S3.3: Turbulence attenuation: When the dominant vibration source is an air flow turbulence signal, obtain the output signal of the PID controller and adjust the flow rate and shaft speed of the air pump.

8. A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology according to claim 7, characterized in that, Smoothing the acceleration derivative of the gearbox includes: S3.2.1: Obtain the acceleration and deceleration time: Determine the acceleration time and deceleration time according to the preset jerk change rate and the maximum acceleration, specifically: Where: t1 is the acceleration time, t3 is the deceleration time, a max is the maximum jerk, J max is the preset jerk change rate; S3.2.2: Obtain the constant speed time: Determine the constant speed time according to the target running speed and the acceleration time, specifically: Where: t2 is the constant-speed time, v tar is the target operating speed, t1 is the acceleration time, a max is the maximum jerk; S3.2.3: Acceleration adjustment: According to the acceleration time, deceleration time, constant-speed time, and preset jerk change rate, adjust the acceleration of the gearbox in real time to reduce the acceleration slope.

9. A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology according to claim 7, characterized in that, Adjust the flow rate and shaft speed of the air pump, including: S3.3.1: Determine the adjustment coefficient: Compare the energy proportion of the air flow turbulence signal with the preset turbulence energy proportion threshold range, and determine the adjustment coefficient according to the comparison range. Specifically: When the energy proportion of the air flow turbulence signal is greater than the preset maximum turbulence energy proportion threshold, increase the adjustment coefficient; when the energy proportion of the air flow turbulence signal is less than the preset minimum turbulence energy proportion threshold, decrease the adjustment coefficient; otherwise, the adjustment coefficient remains unchanged. S3.3.2: Obtain the target flow rate: Determine the corresponding target flow rate according to the adjustment coefficient and shaft speed. Specifically: Where: Q tar is the target flow rate of the air pump, k is the adjustment coefficient, and v tua is the shaft speed of the air pump; S3.3.3: Obtain the output signal of the PID controller: According to the target flow rate and actual flow rate of the air pump, obtain the output signal of the PID controller. Specifically: u(t) = K p ·(Q tar - Q act ) + K i ·∫(Q tar - Q act )dt Where: u(t) is the control output signal, K p is the proportional gain, Q act is the actual flow rate of the air pump, K i is the integral gain, t is time, Q tar is the target flow rate of the air pump.

10. A surface treatment method for a hair dryer based on intelligent multi-axis linkage technology according to claim 1, characterized in that, Adjust the proportional gain and integral gain of the PID controller, including: S4.1: Determine the nozzle trajectory deviation: Obtain the deviation between the actual position and the target position of the nozzle through the laser triangulation method. Specifically: Where: Δd is the deviation between the actual position and the target position of the nozzle, Δx is the displacement of the CCD pixel, L is the reference distance from the laser head to the nozzle, and f is the focal length of the lens. S4.2: Adjust the proportional gain and integral gain: According to the deviation between the actual position and the target position of the nozzle, adjust the proportional gain and integral gain of the PID controller. Specifically: Where: K' p is the adjusted proportional gain, K' i is the adjusted integral gain, K p0 is the initial proportional gain, α is the adjustment coefficient, Δd is the deviation between the actual position and the target position of the nozzle, and t is the time.

Images