Sensor-based perfume bottle nozzle assembly line automatic control system and method

Abstract

The present application relates to the technical field of device control, in particular to a kind of perfume bottle nozzle assembly line automatic control system and method based on sensor.The method decomposes the original sequence of molten pool temperature into molten pool temperature oscillation sequence by filtering method, identifies oscillation event in each stable section and carries out smoothing, to extract the dynamic response mode that can represent the commonality and certainty of molten pool under certain thermal state, then compare the response characteristics between two stable sections, obtain the temperature rise response sensitivity combined with interval temperature rise amount, evaluate the qualification of welding result based on temperature rise response sensitivity.Furthermore, the degree of power attenuation is determined by analyzing the qualified welding result.The present application makes the control of laser power a self-adaptive process according to actual welding result, avoids the failure of strategy caused by unreasonable attenuation degree.

Description

Technical Field

[0001] This invention relates to the field of equipment control technology, specifically to an automatic control system and method for a perfume bottle nozzle assembly line based on sensors. Background Technology

[0002] In the automated production of customized pump heads for high-end perfume bottle nozzles, laser welding can be used to permanently seal the metal rings and metal pump bodies (aluminum, stainless steel, etc.) of the nozzle. The quality of this circumferential weld determines the product's airtightness and long-term reliability. Therefore, ensuring a high degree of consistency in weld quality for every product in mass production is a key technical challenge in this field.

[0003] In existing technologies, automated welding typically employs an open-loop control strategy, controlling the laser head to move along a preset path and executing fixed or programmed segmented laser parameters. However, when processing circumferential welds, an inherent process is unavoidable: as the welding path nears closure and the endpoint overlaps with the starting point, the heat generated throughout the welding process has already accumulated locally on the workpiece, causing the initial temperature of the overlapping area to be significantly higher than the temperature at the start of welding. If the initial welding parameters are continued, excessive energy input can easily lead to defects such as excessive penetration or weld collapse in the overlapping area.

[0004] To alleviate this problem, the industry commonly adopts a method of procedurally reducing laser power at the end of the welding process. The limitation of this method is that it cannot respond to changes in actual working conditions during production. For example, slight differences in thermal conductivity between different batches of raw materials, changes in contact thermal resistance between the workpiece and the fixture due to wear, and fluctuations in the ambient temperature of the production workshop can all cause the actual heat accumulation rate to deviate from the preset model, making this fixed parameter adjustment strategy frequently fail. Summary of the Invention

[0005] To address the technical problem that existing technologies fail to adapt to actual working conditions when reducing laser power at the end of welding, leading to the failure of parameter adjustment strategies, the present invention aims to provide a sensor-based automatic control system and method for perfume bottle nozzle assembly lines. The specific technical solution adopted is as follows:

[0006] This invention proposes a sensor-based automatic control method for a perfume bottle nozzle assembly line, the method comprising:

[0007] During a single welding cycle of the perfume bottle nozzle assembly, the angular position sequence of the laser head controller and the corresponding raw sequence of the molten pool temperature on the perfume bottle nozzle assembly are acquired in real time.

[0008] The original sequence of molten pool temperature is filtered to different degrees to obtain the molten pool temperature oscillation sequence and the weld reference temperature rise sequence;

[0009] The angular position sequence is divided into an initial stable segment and a final stable segment. In each stable segment, smoothing is performed based on the oscillation event to obtain the average oscillation response waveform of the initial stage and the average oscillation response waveform of the final stage corresponding to the molten pool temperature oscillation sequence.

[0010] On the weld reference temperature rise sequence, the interval temperature rise between two stable segments is obtained; the response characteristic difference between the average oscillation response waveform in the initial stage and the average oscillation response waveform in the final stage is obtained; the temperature rise response sensitivity is obtained based on the response characteristic difference and the interval temperature rise.

[0011] The welding of the perfume bottle nozzle is evaluated based on the temperature rise response sensitivity; the laser power is adjusted according to the change in temperature rise response sensitivity corresponding to qualified products.

[0012] Furthermore, the method for obtaining the original sequence of molten pool temperatures includes:

[0013] The original light intensity sequence of the molten pool region is collected by a dual-color spot detector, and the original temperature sequence of the molten pool is obtained based on the dual-color temperature measurement principle.

[0014] Furthermore, the method for obtaining the molten pool temperature oscillation sequence and the weld reference temperature rise sequence includes:

[0015] The original sequence of molten pool temperature is input into a preset high-pass filter, which is a fourth-order Butterworth high-pass filter. The filtering result is the oscillation sequence of molten pool temperature.

[0016] The original sequence of molten pool temperature is input into a preset low-pass filter, which is a second-order Butterworth low-pass filter. The filtering result is the reference temperature rise sequence of the weld bead.

[0017] Furthermore, the method for identifying the oscillation event includes:

[0018] A peak detection algorithm is used to identify the bands of the molten pool temperature oscillation sequence in each stable segment, and the bands are used as the oscillation events.

[0019] Furthermore, the smoothing based on oscillation events includes:

[0020] The time axis is aligned with the main peak of the band as a reference, and a point-by-point arithmetic average is performed to obtain a smooth result.

[0021] Furthermore, the method for obtaining the temperature rise over the specified interval includes:

[0022] The average value of the elements of the weld reference temperature rise sequence in the initial stable section is taken as the initial temperature reference value; the average value of the elements of the weld reference temperature rise sequence in the final stable section is taken as the final temperature reference value; and the difference between the final temperature reference value and the initial temperature reference value is taken as the interval temperature rise amount.

[0023] Furthermore, the method for obtaining the response feature differences includes:

[0024] The average oscillating response waveforms of the initial stage and the final stage are processed using the logarithmic decay method to obtain the molten pool response characteristics of the two waveforms. The molten pool response characteristics include the equivalent damping ratio and the natural angular frequency. The Euclidean distance between the molten pool response characteristics is taken as the difference between the response characteristics.

[0025] Furthermore, the method for obtaining the temperature rise response sensitivity includes:

[0026] The ratio of the difference in response characteristics to the temperature rise in the interval is used as the temperature rise response sensitivity.

[0027] Furthermore, adjusting the laser power based on changes in the temperature rise response sensitivity corresponding to qualified products includes:

[0028] The average temperature rise response sensitivity of a preset number of newly produced qualified perfume bottle nozzles is statistically analyzed, and the difference between the average temperature rise response sensitivity and a preset sensitivity threshold is used as a first risk amplification coefficient. The sum of the differences between the temperature rise response sensitivity of a preset number of newly produced qualified perfume bottle nozzles and the preset sensitivity threshold is used as a second risk amplification coefficient. The first risk amplification coefficient and the second risk amplification coefficient are weighted and fused to obtain the control parameters of the PI controller. The laser power is attenuated based on the control parameters.

[0029] This invention proposes a sensor-based automatic control system for a perfume bottle nozzle assembly line, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the steps of the sensor-based automatic control method for a perfume bottle nozzle assembly line.

[0030] The present invention has the following beneficial effects:

[0031] To adjust welding power according to actual welding conditions, this invention first considers the need to quantify the actual welding results of the product. Furthermore, considering that a perfume pump head is a small-volume product with a large curvature, the heat accumulation during welding does not merely raise the macroscopic temperature of the workpiece, but also alters the intrinsic properties of the molten metal, such as viscosity and surface tension, leading to an evolution in the dynamic response characteristics of the molten pool as a fluid system. Therefore, to effectively analyze these changes in intrinsic characteristics, this invention decomposes the original molten pool temperature sequence into a molten pool temperature oscillation sequence using a filtering method. This oscillation sequence contains numerous single oscillation events triggered by random factors. Therefore, this invention identifies and smooths these oscillation events in each stable segment to extract a dynamic response pattern that represents the common and deterministic characteristics of the molten pool under specific thermal conditions. The response characteristics between two stable segments are then compared, and the temperature rise interval is used to obtain the temperature rise response sensitivity. Based on this sensitivity, the passability of the welding result is evaluated. Furthermore, by analyzing the passable welding results, the degree of power attenuation is determined, making laser power control an adaptive process based on the actual welding results, avoiding strategy failure caused by unreasonable attenuation. Attached Figure Description

[0032] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a flowchart of an automatic control method for a perfume bottle nozzle assembly line based on sensors, provided as an embodiment of the present invention. Detailed Implementation

[0034] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a sensor-based automatic control system and method for perfume bottle nozzle assembly lines proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0036] The following description, in conjunction with the accompanying drawings, details the specific solution of a sensor-based automatic control system and method for perfume bottle nozzle assembly lines provided by the present invention.

[0037] Please see Figure 1 The diagram illustrates a flowchart of an automatic control method for a perfume bottle nozzle assembly line based on sensors, according to an embodiment of the present invention. The method includes:

[0038] Step S1: During a single welding cycle of the perfume bottle nozzle assembly, the angular position sequence of the laser head controller and the corresponding raw sequence of the molten pool temperature on the perfume bottle nozzle assembly are acquired in real time.

[0039] The main hidden danger in the welding process of perfume bottle nozzles lies in the temperature accumulation in the overlapping area. Therefore, in order to determine the welding quality under actual working conditions, it is necessary to continuously collect temperature information of the product surface throughout the welding cycle and then analyze its temperature changes. In this embodiment of the invention, because the perfume bottle nozzle is a small product with a large curvature, the angular position sequence of the laser controller and the original sequence of the molten pool temperature can be collected and correlated within a single welding cycle. In subsequent processes, the welding stages can be divided according to the angular position sequence for the analysis of the overlapping area.

[0040] Preferably, in this embodiment of the invention, considering that directly using a single-band light intensity signal is easily interfered with by working conditions such as workpiece surface emissivity and welding fumes, and that the ratio of multi-band radiation intensity has a more stable functional relationship with temperature, the original light intensity sequence of the molten pool area is collected by a dual-color spot detector, and the original molten pool temperature sequence is obtained based on the dual-color thermometry principle. In this embodiment, a dual-color photodetector is deployed near the laser head, with its field of view aligned with the molten pool area. The original light intensity sequences and angular position sequences of two different monitoring bands are simultaneously collected at a fixed sampling frequency. Then, the instantaneous ratio of the two original light intensity sequences is calculated to obtain the light intensity ratio sequence. The original molten pool temperature sequence can then be obtained using a functional relationship pre-calibrated by a blackbody furnace or thermocouple. It should be noted that the method of calculating optical temperature based on the dual-color thermometry principle is a well-known technique among those skilled in the art. Other multi-band thermometry methods can also be used in other implementations of this embodiment, which will not be elaborated here.

[0041] In this embodiment of the invention, the sampling frequency is set to 10 kHz, and the resulting angular position sequence and the original melt pool temperature sequence have the same length, with elements at the same position corresponding to each other.

[0042] Step S2: Filter the original sequence of molten pool temperature to different degrees to obtain the molten pool temperature oscillation sequence and the weld reference temperature rise sequence.

[0043] During the ring welding process of perfume bottle nozzles, the acquired raw sequence of molten pool temperature is essentially a superposition of two processes with different time scales. The first is a rapid temperature oscillation on the millisecond scale caused by the keyhole effect, metal vapor pressure pulsation, and fluid surface tension oscillation. The second is a slow reference temperature rise on the second scale caused by the continuous input and diffusion of welding heat on the workpiece. Therefore, to independently analyze these two processes, the raw sequence of molten pool temperature first needs to be filtered to different degrees. Since the molten pool temperature oscillation sequence is a high-frequency oscillation, filtering to retain the high-frequency information yields the molten pool temperature oscillation sequence. The weld reference temperature rise sequence, on the other hand, is a reference information characterizing the overall trend; therefore, filtering to retain the overall trend of the signal yields the weld reference temperature rise sequence.

[0044] Preferably, in this embodiment of the invention, the method for obtaining the molten pool temperature oscillation sequence and the weld reference temperature rise sequence includes:

[0045] The original sequence of molten pool temperature is input into a preset high-pass filter, which is a fourth-order Butterworth high-pass filter. The filtering result is the oscillation sequence of molten pool temperature.

[0046] The original sequence of molten pool temperature is input into a preset low-pass filter, which is a second-order Butterworth low-pass filter. The filtering result is the reference temperature rise sequence of the weld bead.

[0047] In other implementations of this invention, other types of digital filters, such as Chebyshev filters or moving average-based filtering algorithms, can also be used to achieve similar signal separation effects.

[0048] Step S3: Divide the angular position sequence into an initial stable segment and a final stable segment; in each stable segment, smooth the oscillation event as a reference to obtain the initial stage average oscillation response waveform and the final stage average oscillation response waveform corresponding to the molten pool temperature oscillation sequence.

[0049] Perfume pump heads are customized workpieces with small volume and large curvature. The heat accumulation during the welding process not only raises the macroscopic temperature of the workpiece, but also changes the intrinsic properties of the molten metal, such as viscosity and surface tension. This leads to an evolution in the dynamic response characteristics of the molten pool as a fluid system, and this drastic evolution of intrinsic characteristics is the root cause of welding defects. To effectively reflect this dynamic response, this invention employs an oscillation event smoothing method to improve signal quality and extract a common, deterministic dynamic response pattern that represents the molten pool under specific thermal conditions.

[0050] First, the entire welding cycle needs to be segmented. Since the angular position represents the positional information of the laser during ring welding, segmentation can be performed based on the angular position sequence, dividing it into an initial stable segment and a final stable segment. In this embodiment of the invention, the initial stable segment is defined as the interval between 30° and 90°, and the final stable segment is defined as the interval between 270° and 330°. The two stable segments each represent a specific thermal state. Since the angular position sequence corresponds to the original sequence of the molten pool temperature, the molten pool temperature oscillation sequence and the weld reference temperature rise sequence can also be divided into corresponding subsequences based on the angular position sequence segmentation method.

[0051] Then, in each stable segment, the oscillation events of the molten pool temperature oscillation sequence are used as a reference for smoothing to obtain the average oscillation response waveform of the initial stage and the average oscillation response waveform of the final stage corresponding to the molten pool temperature oscillation sequence.

[0052] Preferably, in this embodiment of the invention, the method for identifying oscillation events includes:

[0053] A peak detection algorithm is used to identify the bands of the molten pool temperature oscillation sequence in each stable segment, and the bands are used as the oscillation events.

[0054] In one specific implementation of this invention, the peak detection algorithm may incorporate a filtering mechanism that compares the identified initial bands with preset peak thresholds and band length thresholds, removing initial bands that do not meet the thresholds to obtain the final bands. This filtering mechanism can filter out low-energy electrical noise or irrelevant disturbances.

[0055] Furthermore, for any stable segment within the molten pool temperature oscillation sequence, after identifying the oscillation events, the main peak of the band can be used as a reference for time axis alignment. After aligning all oscillation events, a point-by-point arithmetic average is performed on the time axis using an averaging method to obtain a smooth result. Ultimately, each stable segment will produce a single, smooth, typical response waveform, where the initial stable segment corresponds to the initial stage average oscillation response waveform, and the final stable segment corresponds to the final stage average oscillation response waveform.

[0056] Step S4: On the weld reference temperature rise sequence, obtain the interval temperature rise between two stable segments; obtain the response characteristic difference between the average oscillation response waveform of the initial stage and the average oscillation response waveform of the final stage; obtain the temperature rise response sensitivity based on the response characteristic difference and the interval temperature rise.

[0057] By extracting the sensor signals through the above steps, further feature analysis can be performed to compare temperature information changes between different stable segments. Since an unstable process prone to defects is not fundamentally characterized by changes in the molten pool response characteristics, but rather by its overly sensitive response to heat input, a larger temperature change indicates a more sensitive heat response, resulting in poorer welding quality. Therefore, this embodiment of the invention establishes a relationship between the change in response characteristics and the change in macroscopic temperature rise to quantify the temperature rise response sensitivity.

[0058] Firstly, this embodiment of the invention obtains the interval temperature rise between two stable segments on the weld reference temperature rise sequence. Since the weld reference temperature rise sequence represents the overall temperature trend, the obtained interval temperature rise can represent the macroscopic temperature rise change. Furthermore, this embodiment of the invention obtains the response characteristic differences between the two response waveforms by comparing the average oscillation response waveform of the initial stage and the average oscillation response waveform of the final stage. Combined with the interval temperature rise, the temperature rise response sensitivity can be obtained.

[0059] Preferably, in this embodiment of the invention, the method for obtaining the interval temperature rise includes:

[0060] The average value of the elements of the weld reference temperature rise sequence in the initial stable section is taken as the initial temperature reference value; the average value of the elements of the weld reference temperature rise sequence in the final stable section is taken as the final temperature reference value; and the difference between the final temperature reference value and the initial temperature reference value is taken as the interval temperature rise amount.

[0061] Preferably, in this embodiment of the invention, the method for obtaining response feature differences includes:

[0062] The average oscillating response waveforms of the initial stage and the final stage are processed using the logarithmic decay method to obtain the molten pool response characteristics of the two waveforms. The molten pool response characteristics include the equivalent damping ratio and the natural angular frequency. The Euclidean distance between the molten pool response characteristics is taken as the difference between the response characteristics.

[0063] The logarithmic decay method is a standard technique in the field of vibration engineering. Its specific principle is well known to those skilled in the art, and the specific formulas and meanings will not be elaborated here. The process is briefly described in this embodiment of the invention:

[0064] (1) For each average oscillatory response waveform, measure the amplitude of the first positive peak. And the amplitude of the second positive peak that follows. And record the time interval between these two peaks, i.e., the damped oscillation period. .

[0065] (2) Calculate the logarithmic decay rate : ; where ln is a logarithmic function with the natural constant as the base.

[0066] (3) Calculate the equivalent damping ratio based on the logarithmic decay rate. : .

[0067] (4) Calculate the equivalent natural angular frequency based on the damped oscillation period and the equivalent damping ratio. : .

[0068] It should be noted that, in this embodiment of the invention, to avoid excessive deviation in the Euclidean distance value, the Euclidean distance calculation formula in this embodiment of the invention is as follows: ;in The final Euclidean distance is obtained. The natural angular frequency of the average oscillatory response waveform in the initial stage. The natural angular frequency of the average oscillatory response waveform in the final stage. The preset natural angular frequency reference value is used for normalization. The equivalent damping ratio of the average oscillatory response waveform in the initial stage. The equivalent damping ratio of the average oscillatory response waveform in the final stage. This is the preset equivalent damping ratio reference value used for normalization. That is, the above formula normalizes the differences by adding a benchmark value.

[0069] It should be noted that the oscillation event screening thresholds used in the oscillation event identification process described above in the embodiments of the present invention, as well as the reference values ​​used for normalization in the above formulas, were obtained during a one-time process characterization stage performed during the initial deployment or when a major change occurred in the welding process. The purpose of this stage is to collect and analyze the welding data of a batch of representative samples offline to generate a set of solidified parameters that do not require manual setting for subsequent online automated operation. The thresholds used in the subsequent process can be obtained at this stage as a reference and will not be elaborated further.

[0070] In this embodiment of the invention, the method for obtaining the temperature rise response sensitivity includes:

[0071] The ratio of the difference in response characteristics to the temperature rise within the interval is used as the temperature rise response sensitivity. A higher temperature rise response sensitivity indicates a drastic change in the intrinsic response characteristics of the molten pool (determined by equivalent viscosity and surface tension) under the accumulated heat experienced by the product. This drastic change in response characteristics signifies a loss of dynamic equilibrium in the molten pool fluid, a direct precursor to the inability of metal vapor to escape stably, leading to porosity, or uneven stress distribution during solidification, resulting in microcracks. Conversely, a lower sensitivity indicates a more normal welding quality for the product.

[0072] Step S5: Evaluate whether the welding of the perfume bottle nozzle is qualified based on the temperature rise response sensitivity; adjust the laser power according to the change in temperature rise response sensitivity corresponding to qualified products.

[0073] The welding quality of perfume bottle nozzles can be assessed based on temperature rise response sensitivity; that is, the higher the temperature rise response sensitivity, the worse the current welding quality. Therefore, in this embodiment of the invention, after normalizing the temperature rise response sensitivity, a quality judgment threshold of 0.7 is set. If the temperature rise response sensitivity of a product is greater than the quality judgment threshold, it is considered unqualified; otherwise, it is considered qualified.

[0074] Since the high temperature rise response sensitivity of a single product may be caused by isolated, accidental factors, a single product cannot represent changes in operating conditions, and a laser power adjustment command cannot be executed for each product. Therefore, this embodiment of the invention selects qualified products and adjusts the laser power based on the changing characteristics of the temperature rise response sensitivity corresponding to multiple qualified products. If there is a significant increasing trend, it means that the current laser power is relatively excessive in terms of energy input for the changing operating conditions, leading to increased sensitivity of the molten pool response characteristics to heat input. Therefore, negative feedback adjustment of the laser power can be performed based on the deviation caused by the change to compensate for the slow deterioration trend of the process state.

[0075] Preferably, in this embodiment of the invention, adjusting the laser power based on the change in temperature rise response sensitivity corresponding to a qualified product includes:

[0076] The average temperature rise response sensitivity of a preset number of newly produced qualified perfume bottle nozzles is statistically analyzed, and the difference between the average temperature rise response sensitivity and a preset sensitivity threshold is used as the first risk enhancement coefficient. In this embodiment of the invention, the preset number is set to 50, and the method for setting the sensitivity threshold has been explained in the above steps and will not be repeated here.

[0077] The sum of the differences between the temperature rise response sensitivity of a preset number of newly produced qualified perfume bottle nozzles and the preset sensitivity threshold is used as the second risk enhancement coefficient.

[0078] The first risk enhancement coefficient is an average deviation characterized by an average value, while the second risk enhancement coefficient is a historical deviation accumulation feature. Therefore, the two are further integrated by weighted summation of the first and second risk enhancement coefficients to obtain the control parameters of the PI controller. In this embodiment of the invention, the weight of the first risk enhancement coefficient is set to 0.6, and the weight of the second risk enhancement coefficient is set to 0.4. The control parameters are obtained by weighted summation.

[0079] The laser power is attenuated based on the control parameters. In this embodiment of the invention, the control parameters are normalized, limiting their value range to between 0 and 1. The normalized control parameter is subtracted from the positive integer 1 to obtain the adjustment ratio. This adjustment ratio is then multiplied by the initial laser power to obtain the attenuated laser power. That is, a larger control parameter indicates a stronger thermal sensitivity to the current operating condition, requiring a lower laser power for welding. The final attenuated laser power is downloaded to the laser controller for subsequent workpiece welding. In this way, the system can automatically adjust the welding energy input to counteract the effects of operating condition drift, keeping the inherent stability of the process consistently near the target level.

[0080] It should be noted that the embodiments of the present invention can set a laser power adjustment cycle, and analyze the changes in the working conditions at fixed intervals to maintain a healthy welding process. It should also be noted that in specific implementations, a lower limit for attenuation needs to be defined, meaning the laser power can only attenuate within a certain range. When the calculated attenuated laser power is less than the lower limit, the laser power corresponding to the lower limit is used as the final control command. Furthermore, if attenuation is still deemed necessary for a relatively long period after the laser power reaches the lower limit, welding needs to be suspended, and personnel should be notified to inspect the process equipment. Specific strategies can be set according to the specific implementation scenario, and the embodiments of the present invention will not elaborate or limit them.

[0081] In summary, this invention decomposes the original molten pool temperature sequence into a molten pool temperature oscillation sequence through filtering. Oscillation events are identified and smoothed within each stable segment to extract a dynamic response pattern that represents the common and deterministic characteristics of the molten pool under specific thermal conditions. The response characteristics between two stable segments are then compared, and the temperature rise interval is used to obtain the temperature rise response sensitivity. The weld qualification is then assessed based on the temperature rise response sensitivity. Furthermore, the degree of power attenuation is determined by analyzing qualified weld results. This invention makes laser power control an adaptive process based on actual welding results, avoiding strategy failure caused by unreasonable attenuation levels.

[0082] Based on the same inventive concept, this invention also proposes a sensor-based automatic control system for a perfume bottle nozzle assembly line, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the steps of the sensor-based automatic control method for a perfume bottle nozzle assembly line.

[0083] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A sensor-based automatic control method for a perfume bottle nozzle assembly line, characterized in that, The method includes: During a single welding cycle of the perfume bottle nozzle assembly, the angular position sequence of the laser head controller and the corresponding raw sequence of the molten pool temperature on the perfume bottle nozzle assembly are acquired in real time. The original sequence of molten pool temperature is filtered to different degrees to obtain the molten pool temperature oscillation sequence and the weld reference temperature rise sequence; The angular position sequence is divided into an initial stable segment and a final stable segment. In each stable segment, smoothing is performed based on the oscillation event to obtain the average oscillation response waveform of the initial stage and the average oscillation response waveform of the final stage corresponding to the molten pool temperature oscillation sequence. On the weld reference temperature rise sequence, the interval temperature rise between two stable segments is obtained; the response characteristic difference between the average oscillation response waveform in the initial stage and the average oscillation response waveform in the final stage is obtained; the temperature rise response sensitivity is obtained based on the response characteristic difference and the interval temperature rise. The welding of the perfume bottle nozzle is evaluated based on the temperature rise response sensitivity; the laser power is adjusted based on the change in temperature rise response sensitivity corresponding to qualified products. The method for obtaining the response feature differences includes: The average oscillating response waveforms of the initial stage and the final stage are processed using the logarithmic decay method to obtain the molten pool response characteristics of the two waveforms. The molten pool response characteristics include the equivalent damping ratio and the natural angular frequency. The Euclidean distance between the molten pool response characteristics is taken as the difference between the response characteristics.

2. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 1, characterized in that, The method for obtaining the original sequence of molten pool temperature includes: The original light intensity sequence of the molten pool region is collected by a dual-color spot detector, and the original temperature sequence of the molten pool is obtained based on the dual-color temperature measurement principle.

3. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 1, characterized in that, The method for obtaining the molten pool temperature oscillation sequence and the weld reference temperature rise sequence includes: The original sequence of molten pool temperature is input into a preset high-pass filter, which is a fourth-order Butterworth high-pass filter. The filtering result is the oscillation sequence of molten pool temperature. The original sequence of molten pool temperature is input into a preset low-pass filter, which is a second-order Butterworth low-pass filter. The filtering result is the reference temperature rise sequence of the weld bead.

4. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 1, characterized in that, The method for identifying the oscillation event includes: A peak detection algorithm is used to identify the bands of the molten pool temperature oscillation sequence in each stable segment, and the bands are used as the oscillation events.

5. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 4, characterized in that, The smoothing based on oscillation events includes: The time axis is aligned with the main peak of the band as a reference, and a point-by-point arithmetic average is performed to obtain a smooth result.

6. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 1, characterized in that, The method for obtaining the temperature rise in the specified interval includes: The average value of the elements of the weld reference temperature rise sequence in the initial stable section is taken as the initial temperature reference value; the average value of the elements of the weld reference temperature rise sequence in the final stable section is taken as the final temperature reference value; and the difference between the final temperature reference value and the initial temperature reference value is taken as the interval temperature rise amount.

7. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 1, characterized in that, The method for obtaining the temperature rise response sensitivity includes: The ratio of the difference in response characteristics to the temperature rise in the interval is used as the temperature rise response sensitivity.

8. The sensor-based automatic control method for a perfume bottle nozzle assembly line according to claim 1, characterized in that, The adjustment of laser power based on the change in temperature rise response sensitivity corresponding to qualified products includes: The average temperature rise response sensitivity of a preset number of newly produced qualified perfume bottle nozzles is statistically analyzed, and the difference between the average temperature rise response sensitivity and a preset sensitivity threshold is used as a first risk amplification coefficient. The sum of the differences between the temperature rise response sensitivity of a preset number of newly produced qualified perfume bottle nozzles and the preset sensitivity threshold is used as a second risk amplification coefficient. The first risk amplification coefficient and the second risk amplification coefficient are weighted and fused to obtain the control parameters of the PI controller. The laser power is attenuated based on the control parameters.

9. A sensor-based automatic control system for a perfume bottle nozzle assembly line, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the sensor-based automatic control method for perfume bottle nozzle assembly line as described in any one of claims 1 to 8.

Images