Intelligent injection molding process control method and system for plastic mold

Abstract

This invention discloses an intelligent injection molding process control method and system for plastic molds, relating to the field of plastic mold injection molding technology. The method includes collecting temperature data from an environmentally sensitive reference area, temperature data from a critical area of ​​local degradation, pressure data, and ultrasonic echo signals; calculating the slope of each parameter within a sliding window to extract an acoustic fingerprint; calculating the Euclidean distance between the acoustic fingerprint and a healthy baseline, and generating a material fatigue indicator if the distance continuously exceeds a threshold; otherwise, determining environmental drift or local degradation based on the temperature slope ratio and correcting it using a pressure-temperature correlation; narrowing or widening the tolerance window based on the indicator type, triggering an early warning or generating a feedforward fine-tuning command. This invention can distinguish between process deviations caused by environmental disturbances and equipment degradation, reducing unnecessary compensation adjustments, protecting equipment lifespan, and maintaining product quality stability.

Description

Technical Field

[0001] This invention relates to the field of plastic mold injection molding technology, and in particular to a method and system for intelligent injection molding process control of plastic molds. Background Technology

[0002] In plastic injection molding production, existing intelligent control systems monitor the physical states inside the mold, such as temperature and pressure, in real time and automatically adjust according to preset process parameters. However, during long-term, high-intensity production, subtle changes in the external environment (such as fluctuations in workshop ambient temperature and cooling water temperature drift) and the gradual wear of the injection molding machine and mold components (such as wear on the injection screw, decreased performance of the check ring seal, and scaling in the cooling water channels) can all affect the actual behavior of the molten plastic. Existing systems lack the ability to perceive and differentiate these unmonitored sources of disturbance, often attributing all detected deviations uniformly to deviations in process parameters and triggering compensatory adjustments.

[0003] This compensation mechanism makes it difficult for the system to accurately determine the true source of the deviation, thus trapping it in a cycle of frequent adjustments that can accelerate equipment wear and tear. For example, when the mold temperature rises slightly due to environmental factors, the system may automatically extend the cooling time or adjust the barrel temperature. While these adjustments may bring the data back to the target range in the short term, they increase the additional wear and tear on the core mechanical components of the injection molding machine. As the equipment continues to degrade, the actual melt behavior deviates from the system model, the adjustment effect gradually worsens, and product quality exhibits unexplained dimensional deviations or surface defects. Ultimately, this leads to a state of "controlled but uncontrolled" production, affecting long-term quality consistency and production efficiency. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent injection molding process control method and system for plastic molds, aiming to solve the problem that existing injection molding process control cannot distinguish between process deviations caused by external environmental disturbances and internal equipment degradation, thereby leading to frequent compensatory adjustments and accelerated equipment wear.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent injection molding process control of plastic molds, comprising the following steps: S1. Collect the reference temperature of the pre-divided environmentally sensitive reference area, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal in the local degradation critical area within the mold. S2. Calculate the first slope of the reference temperature, the second slope of the key temperature, and the third slope of the key pressure within the predetermined sliding time window; extract the sound wave propagation time, peak amplitude, and center frequency of the ultrasonic echo signal to construct a real-time acoustic fingerprint; S3. Calculate the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. If the Euclidean distance exceeds the preset fatigue threshold within the preset fatigue judgment time window, generate a material fatigue indicator; otherwise, do not generate one. When no material fatigue indicator is generated, it is determined whether the ratio of the second slope to the first slope falls within a preset ratio range: if yes, it is determined that the process drift originates from overall environmental factors and an environmental drift indicator is generated; if no, it is determined that the process drift originates from local physical degradation and a local degradation indicator is generated. When generating an environmental drift marker, it is determined whether the third slope and the first slope satisfy a preset physical correlation condition: if they satisfy the condition, the environmental drift marker is maintained; if they do not satisfy the condition, the environmental drift marker is corrected to a local degradation marker. S4. If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window of the local degradation critical area is narrowed and an early warning is triggered. The initial tolerance window is used to set a fluctuation reference range for the temperature and pressure of the local degradation critical area at the beginning of the injection molding process. If an environmental drift indicator is maintained, the initial tolerance window is widened and a feedforward fine-tuning command is generated.

[0006] Furthermore, the acquisition of the reference temperature of the pre-divided environmentally sensitive reference zone within the mold, the critical temperature and critical pressure of the pre-divided local degradation critical zone, and the ultrasonic echo signal within the local degradation critical zone includes: The raw resistance signal is collected by a PT100 platinum resistance temperature sensor deployed in the environmentally sensitive reference area. The raw resistance signal is then converted to temperature to generate a reference temperature for the environmentally sensitive reference area. The original thermoelectric potential signal is collected by K-type miniature thermocouples deployed in the local degradation critical area. Cold junction compensation and nonlinear correction are performed on the original thermoelectric potential signal to generate the critical temperature of the local degradation critical area. A piezoelectric pressure sensor deployed in a critical area of ​​local degradation is used to collect the original charge signal, and a voltage signal is generated by conditioning the original charge signal through a charge amplifier. The voltage signal is converted from analog to digital and the pressure calibration curve is calculated to generate the key pressure in the critical area of ​​local degradation. By emitting high-frequency electrical pulses through miniature ultrasonic transducers deployed in critical areas of local degradation, the echo electrical signals, after being reflected and scattered by the internal microstructure of the mold material and converted by the miniature ultrasonic transducers, are collected. Bandpass filtering and analog-to-digital conversion are performed on the echo electrical signals to generate ultrasonic echo signals.

[0007] Furthermore, the calculation of the first slope of the reference temperature, the second slope of the critical temperature, and the third slope of the critical pressure within the predetermined sliding time window includes: Within a preset sampling time, the reference temperature, critical temperature, and critical pressure that fall within a predetermined sliding time window are sequentially captured. Linear regression was performed with the time of data collection as the independent variable and the reference temperature as the dependent variable. The regression line of the reference temperature over time was calculated using the least squares method. The slope of the regression line was calculated and used as the first slope. Linear regression was performed with the acquisition time as the independent variable and the key temperature value as the dependent variable. The regression line of the key temperature over time was calculated using the least squares method. The slope of the regression line was calculated and used as the second slope. Linear regression was performed with the data collection time as the independent variable and the critical pressure value as the dependent variable. The regression line of critical pressure over time was calculated using the least squares method. The slope of the regression line was calculated and used as the third slope.

[0008] Furthermore, the extraction of the ultrasonic echo signal's propagation time, peak amplitude, and center frequency to construct a real-time acoustic fingerprint includes: Perform Hilbert transform on the ultrasonic echo signal to generate the envelope curve of the ultrasonic echo signal; On the envelope curve of the ultrasonic echo signal, the first peak point whose amplitude exceeds the preset amplitude threshold is identified, and the time corresponding to the peak point is taken as the arrival time of the sound wave. The difference between the arrival time of the sound wave and the preset emission reference time is calculated to generate the sound wave propagation time. The maximum amplitude within the pre-sampling interval is extracted from the envelope curve of the ultrasonic echo signal to generate the peak amplitude. Perform a fast Fourier transform on the ultrasonic echo signal, extract the frequency of the maximum amplitude from the transform result, and generate the center frequency. The sound wave propagation time, peak amplitude, and center frequency are combined into a feature vector, which is then used as a real-time acoustic fingerprint.

[0009] Furthermore, the preset health baseline fingerprint is a standard acoustic fingerprint obtained by collecting a large amount of acoustic fingerprint data and performing statistical processing, used as a comparison benchmark for the real-time acoustic fingerprint; the calculation of the Euclidean distance between the real-time acoustic fingerprint and the preset health baseline fingerprint, and the generation of a material fatigue indicator when the Euclidean distance exceeds a preset fatigue threshold within a preset fatigue judgment time window; otherwise, no indicator is generated, including: Within a preset fatigue assessment time window, the following process is periodically repeated: acquire ultrasonic echo signals, extract sound wave propagation time, peak amplitude, and center frequency, construct a real-time acoustic fingerprint, calculate the differences between the sound wave propagation time, peak amplitude, and center frequency of the real-time acoustic fingerprint and the baseline sound wave propagation time, baseline peak amplitude, and baseline center frequency of the preset healthy baseline fingerprint, construct a difference vector from the differences, calculate the Euclidean norm of the difference vector, and obtain the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. When every Euclidean distance calculated consecutively within the preset fatigue judgment time window exceeds the preset fatigue threshold, a material fatigue indicator is generated. When the Euclidean distance of any sampling point within the preset fatigue judgment time window is less than or equal to the preset fatigue threshold, no material fatigue identifier is generated.

[0010] Furthermore, the preset ratio range is determined through process verification based on the mold material, injection molding process parameters, and product precision requirements, and is used to characterize the consistency of the change trend between the key temperature and the reference temperature.

[0011] Furthermore, the preset physical correlation conditions include: when the reference temperature rises, the critical pressure should change in the opposite direction, and the negative ratio of the third slope to the first slope should fall within the pressure-temperature gradient coupling range calibrated based on historical health operation data; when generating an environmental drift indicator, it is determined whether the third slope and the first slope satisfy the preset physical correlation conditions: if satisfied, the environmental drift indicator is maintained; if not satisfied, the environmental drift indicator is corrected to a local degradation indicator, including: Obtain the sign directions of the first slope and the third slope, and determine whether the sign direction of the third slope is opposite to that of the first slope; When the directions are opposite, calculate the opposite of the ratio of the third slope to the first slope, and determine whether the opposite number is within the pressure-temperature gradient coupling range; When the opposite number is within the pressure-temperature gradient coupling range, it is determined that the preset physical correlation condition is met, and the environmental drift indicator is maintained. When the sign direction of the third slope is the same as that of the first slope, or the opposite number is outside the pressure-temperature gradient coupling range, it is determined that the physical correlation condition is not met, and the environmental drift indicator is corrected to a local degradation indicator.

[0012] Furthermore, if a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window for the local degradation critical area is narrowed, and an early warning is triggered, including: Based on the generated material fatigue markers or local degradation markers, the initial tolerance window for the critical local degradation area is narrowed to a smaller preset strict tolerance limit; the strict tolerance limit is the maximum allowable fluctuation half-width that is pre-calibrated for different fault types and severity based on previous process verification and material fatigue experiments to ensure that no defective products are produced. When a material fatigue identifier is generated, a first warning message is generated. The first warning message includes a material fatigue fault type code and a prompt text for checking the material status of the mold micro-area. When a local degradation flag is generated, a second warning message is generated. The second warning message includes a code for the local physical degradation fault type and a prompt text for checking for scale buildup in the cooling water passages.

[0013] Furthermore, if the environmental drift flag is maintained, the initial tolerance window is widened, and a feedforward fine-tuning instruction is generated, including: If the environmental drift indicator is maintained, when the second slope is positive, the upper limit boundary of the initial tolerance window is moved outward by a preset temperature relaxation margin; when the second slope is negative, the lower limit boundary is moved outward by a preset temperature relaxation margin to obtain the relaxed tolerance window; the preset temperature relaxation margin is a unilateral tolerance margin pre-calibrated based on the measured data of the initial dimensional accuracy requirements of the product and the thermal inertia response of the mold under historical environmental temperature fluctuations, ensuring that environmental drift does not cause quality defects. The direction of cooling water flow adjustment is determined based on the sign of the first slope, and the amount of cooling water flow adjustment is determined based on the absolute value of the first slope. A feedforward fine-tuning command is then generated, which is used to control the cooling water flow adjustment.

[0014] A smart injection molding process control system for plastic molds, comprising: Signal acquisition module: used to acquire the reference temperature of the pre-divided environmentally sensitive reference area, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal in the local degradation critical area within the mold; Acoustic fingerprint construction module: used to calculate the first slope of reference temperature, the second slope of key temperature, and the third slope of key pressure within a predetermined sliding time window; extract the sound wave propagation time, peak amplitude, and center frequency of ultrasonic echo signals to construct a real-time acoustic fingerprint; State recognition module: Used to calculate the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. When the Euclidean distance exceeds the preset fatigue threshold within the preset fatigue judgment time window, a material fatigue label is generated; otherwise, no label is generated. When no material fatigue label is generated, it is determined whether the ratio of the second slope to the first slope falls within a preset ratio range: if yes, it is determined that the process drift originates from overall environmental factors, and an environmental drift label is generated; if no, it is determined that the process drift originates from local physical degradation, and a local degradation label is generated. When an environmental drift label is generated, it is determined whether the third slope and the first slope meet the preset physical correlation conditions: if yes, the environmental drift label is maintained; if no, the environmental drift label is corrected to a local degradation label. Window adjustment module: If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window of the local degradation critical area is narrowed and an early warning is triggered. The initial tolerance window is used to set a fluctuation reference range for the temperature and pressure of the local degradation critical area at the beginning of the injection molding process. If the environmental drift indicator is maintained, the initial tolerance window is widened and a feedforward fine-tuning command is generated.

[0015] The present invention provides an intelligent injection molding process control method and system for plastic molds, the beneficial effects of which are mainly reflected in the following aspects: 1. This invention collects reference temperatures in environmentally sensitive reference areas, critical temperatures and pressures in locally degraded critical areas, and ultrasonic echo signals. It calculates the slope of each parameter within a sliding time window and combines this with the Euclidean distance between the acoustic fingerprint and the healthy baseline to determine the material fatigue state. This method can accurately distinguish the source of process drift: when the Euclidean distance continuously exceeds a threshold, it is identified as material fatigue; otherwise, the temperature slope ratio is used to determine whether it is environmental drift or local degradation, and further verification and correction are performed through the physical correlation between pressure and temperature. This multi-parameter fusion discrimination mechanism enables the system to perceive the internal degradation state of equipment, avoiding confusion between environmental disturbances and equipment wear.

[0016] 2. This invention dynamically adjusts the initial tolerance window of critical areas of local degradation based on the identification results: the window is narrowed and an early warning is triggered for material fatigue or local degradation indicators, while the window is widened and a feedforward fine-tuning command is generated for confirmed environmental drift indicators. This strategic control reduces unnecessary frequent compensation adjustments, thereby reducing the additional wear load on core components such as the injection molding machine screw and check ring. By responding to environmental changes in advance through feedforward fine-tuning, process stability is maintained, which helps to maintain the quality consistency of high-precision products over long production cycles. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the intelligent injection molding process control method for plastic molds according to the present invention. Figure 2 This is a functional block diagram of an intelligent injection molding process control system for plastic molds according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides a method for intelligent injection molding process control of plastic molds, comprising the following steps: S1. Collect the reference temperature of the pre-divided environmentally sensitive reference area, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal in the local degradation critical area within the mold. In this embodiment, the acquisition of the reference temperature of the pre-divided environmentally sensitive reference area within the mold, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal within the local degradation critical area includes: The raw resistance signal is collected by a PT100 platinum resistance temperature sensor deployed in the environmentally sensitive reference area. The raw resistance signal is then converted to temperature to generate a reference temperature for the environmentally sensitive reference area.

[0020] Specifically, the environmentally sensitive reference area refers to the region on the mold that responds quickly to changes in external ambient temperature, such as near the mold surface or cooling channels. The resistance of the PT100 platinum resistance thermometer increases with temperature. The original resistance signal is converted to a reference temperature (based on the PT100 standard calibration table) to characterize the overall thermal impact of environmental factors on the mold.

[0021] The original thermoelectric potential signal is collected by K-type miniature thermocouples deployed in the local degradation critical region. Cold junction compensation and nonlinear correction are performed on the original thermoelectric potential signal to generate the critical temperature of the local degradation critical region.

[0022] Specifically, the critical area of ​​local degradation refers to the high-stress area within the mold that is prone to fatigue or degradation, such as the corner of the cavity or near the gate. The K-type thermocouple outputs a weak thermoelectric potential, which, after cold junction compensation (eliminating the error of the reference junction temperature change) and nonlinear correction (correcting the nonlinear characteristics of the thermocouple), yields the critical temperature, which is used to monitor the local thermal state of this area.

[0023] A piezoelectric pressure sensor deployed in a critical area of ​​local degradation is used to collect the original charge signal, and a voltage signal is generated by conditioning the original charge signal through a charge amplifier.

[0024] Specifically, piezoelectric pressure sensors use the piezoelectric effect to convert pressure into electrical charge. The original charge signal is conditioned by a charge amplifier (high input impedance circuit) and becomes a voltage signal, the amplitude of which is linearly related to the applied pressure.

[0025] The voltage signal is converted from analog to digital and then converted to a pressure calibration curve to generate the critical pressure in the local degradation critical zone.

[0026] Specifically, analog-to-digital conversion discretizes the voltage signal, and pressure calibration curve conversion refers to converting the voltage value into a key pressure using a pre-calibrated mapping relationship (obtained by calibrating the sensor point by point using a pressure standard), reflecting the actual pressure at that location within the mold cavity.

[0027] By emitting high-frequency electrical pulses through miniature ultrasonic transducers deployed in critical areas of local degradation, the echo electrical signals, after being reflected and scattered by the internal microstructure of the mold material and converted by the miniature ultrasonic transducers, are collected. Bandpass filtering and analog-to-digital conversion are performed on the echo electrical signals to generate ultrasonic echo signals.

[0028] Specifically, the miniature ultrasonic transducer combines transmitting and receiving functions, and is excited by high-frequency electrical pulses to generate ultrasonic waves. The echo electrical signal carries information about the internal microstructure of the material, and after bandpass filtering (preserving the effective frequency band) and analog-to-digital conversion, the ultrasonic echo signal is obtained, which is used to analyze the degree of material fatigue or degradation.

[0029] S2. Calculate the first slope of the reference temperature, the second slope of the key temperature, and the third slope of the key pressure within the predetermined sliding time window; extract the sound wave propagation time, peak amplitude, and center frequency of the ultrasonic echo signal to construct a real-time acoustic fingerprint; In this embodiment, the calculation of the first slope of the reference temperature, the second slope of the critical temperature, and the third slope of the critical pressure within a predetermined sliding time window includes: Within a preset sampling time, the reference temperature, critical temperature, and critical pressure that fall within a predetermined sliding time window are captured sequentially.

[0030] Specifically, the preset sampling time refers to the total data collection duration set in advance according to the process cycle, such as 30 minutes of continuous production; the predetermined sliding time window refers to the sub-interval that slides at a fixed step size (such as 5 minutes) within this duration, and its length is determined by process verification based on the mold thermal response time; the three sets of data after extraction are aligned in time and used as samples for slope calculation.

[0031] Linear regression was performed with the time of data collection as the independent variable and the reference temperature as the dependent variable. The regression line of the reference temperature over time was calculated using the least squares method. The slope of the regression line was calculated and used as the first slope.

[0032] Specifically, the least squares method is a well-known fitting method, and the formula for calculating the slope of the regression line is: ; In the formula, This represents the number of sampling points within the window. This refers to the i-th data acquisition time. The corresponding reference temperature value; the calculated first slope It characterizes the average rate of change of the reference temperature within the window.

[0033] Linear regression was performed with the data collection time as the independent variable and the key temperature value as the dependent variable. The regression line of the key temperature over time was calculated using the least squares method. The slope of the regression line was calculated and used as the second slope.

[0034] Specifically, using the same least squares formula, the dependent variable is replaced with the critical temperature value, and the second slope is calculated to characterize the rate of temperature change in the local degradation critical zone.

[0035] Linear regression was performed with the data collection time as the independent variable and the critical pressure value as the dependent variable. The regression line of critical pressure over time was calculated using the least squares method. The slope of the regression line was calculated and used as the third slope.

[0036] Specifically, the dependent variable is replaced with the key pressure value, and the third slope is calculated using the same formula to characterize the rate of change of pressure in the local degradation key area. The first, second, and third slopes are all obtained within the same sliding time window and are used for subsequent proportional comparison and physical correlation judgment.

[0037] In this embodiment, the step of extracting the acoustic propagation time, peak amplitude, and center frequency of the ultrasonic echo signal to construct a real-time acoustic fingerprint includes: Perform a Hilbert transform on the ultrasonic echo signal to generate the envelope curve of the ultrasonic echo signal.

[0038] Specifically, the Hilbert transform is used to extract the signal envelope. After converting the real signal into an analytic signal, its magnitude is taken to obtain the envelope curve. The envelope curve eliminates high-frequency oscillations and makes it easier to identify the echo arrival time and peak amplitude.

[0039] On the envelope curve of the ultrasonic echo signal, the first peak point whose amplitude exceeds the preset amplitude threshold is identified. The time corresponding to the peak point is taken as the arrival time of the sound wave. The difference between the arrival time of the sound wave and the preset emission reference time is calculated to generate the sound wave propagation time.

[0040] Specifically, the preset amplitude threshold is set to 3 to 5 times the average background noise level to distinguish between effective echoes and noise; the preset emission reference time is the start time of the ultrasonic transducer emitting high-frequency electrical pulses; the sound wave propagation time is the difference between the two, reflecting the round-trip propagation time of the ultrasonic wave in the mold material.

[0041] The maximum amplitude within the presampling interval is extracted from the envelope curve of the ultrasonic echo signal to generate the peak amplitude.

[0042] Specifically, the presampling interval refers to the time range centered on the arrival time of the sound wave and covering the width of the echo pulse. Its length is pre-calibrated experimentally based on the transducer bandwidth. Within this interval, the maximum value of the envelope curve, i.e., the peak amplitude, is searched to characterize the echo intensity, which is related to the scattering ability of the material's internal microstructure.

[0043] Perform a fast Fourier transform on the ultrasonic echo signal, extract the frequency of the maximum amplitude from the transform result, and generate the center frequency.

[0044] Specifically, the Fast Fourier Transform converts the time-domain echo signal into a frequency-domain amplitude spectrum; the frequency corresponding to the maximum amplitude is the center frequency, which shifts with the change in ultrasonic attenuation characteristics caused by material fatigue or defects.

[0045] The sound wave propagation time, peak amplitude, and center frequency are combined into a feature vector, which is then used as a real-time acoustic fingerprint.

[0046] Specifically, the feature vector is an ordered triplet of [sound wave propagation time, peak amplitude, center frequency]; the real-time acoustic fingerprint is the feature vector extracted from the local degradation critical area in the current injection molding cycle, which is used to compare with the healthy baseline fingerprint to determine the degree of material fatigue.

[0047] S3. Calculate the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. If the Euclidean distance exceeds the preset fatigue threshold within the preset fatigue judgment time window, generate a material fatigue indicator; otherwise, do not generate one. In this embodiment, the preset health baseline fingerprint is a standard acoustic fingerprint obtained by collecting a large amount of acoustic fingerprint data and statistically processing it, and is used as a comparison benchmark for the real-time acoustic fingerprint; the calculation of the Euclidean distance between the real-time acoustic fingerprint and the preset health baseline fingerprint, and the generation of a material fatigue indicator when the Euclidean distance exceeds a preset fatigue threshold within a preset fatigue judgment time window; otherwise, no indicator is generated, including: Within a preset fatigue assessment time window, the following process is periodically repeated: ultrasonic echo signals are acquired, sound wave propagation time, peak amplitude, and center frequency are extracted, a real-time acoustic fingerprint is constructed, and the differences between the sound wave propagation time, peak amplitude, and center frequency of the real-time acoustic fingerprint and the baseline sound wave propagation time, baseline peak amplitude, and baseline center frequency of the preset healthy baseline fingerprint are calculated. The differences are then used to construct a difference vector, and the Euclidean norm of the difference vector is calculated to obtain the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint.

[0048] Specifically, the preset healthy baseline fingerprint involves continuously collecting over 100 sets of acoustic fingerprints in a brand-new mold state, and using the average value of each dimension as the baseline value; the preset fatigue judgment time window is, for example, 10 injection cycles, and its length is calibrated through accelerated testing based on the fatigue evolution rate of the mold material; the Euclidean distance calculation formula is: ; In the formula, These are the acoustic wave propagation time, peak amplitude, and center frequency of the real-time acoustic fingerprint, respectively. These are the baseline acoustic propagation time, baseline peak amplitude, and baseline center frequency of the preset healthy baseline fingerprint; the Euclidean norm of the difference vector, i.e., the Euclidean distance. It is used to quantify the degree of deviation between the current acoustic fingerprint and the health status.

[0049] When every Euclidean distance calculated continuously within the preset fatigue judgment time window exceeds the preset fatigue threshold, a material fatigue indicator is generated.

[0050] Specifically, the preset fatigue threshold is determined through material fatigue experiments, taking the critical value of the Euclidean distance when the mold material begins to show microcracks; if the Euclidean distance of each sampling point within the window is greater than this threshold, it indicates that the acoustic characteristics are continuously deviating from the healthy state, and the mold material is judged to have undergone fatigue, generating a material fatigue indicator.

[0051] When the Euclidean distance of any sampling point within the preset fatigue judgment time window is less than or equal to the preset fatigue threshold, no material fatigue identifier is generated.

[0052] Specifically, if the Euclidean distance of at least one sampling point within the window does not exceed the threshold, it indicates that the material condition is acceptable or only fluctuates momentarily, and does not meet the stringent conditions for fatigue judgment. Therefore, no material fatigue indicator is generated.

[0053] When no material fatigue indicator is generated, it is determined whether the ratio of the second slope to the first slope falls within a preset ratio range: if yes, it is determined that the process drift originates from overall environmental factors and an environmental drift indicator is generated; if no, it is determined that the process drift originates from local physical degradation and a local degradation indicator is generated. In this embodiment, the preset ratio range is determined through process verification based on the mold material, injection molding process parameters, and product precision requirements, and is used to characterize the consistency of the change trend between the key temperature and the reference temperature.

[0054] Specifically, the preset ratio range refers to the allowable range of the ratio of the second slope to the first slope. Its upper and lower limits are determined through process verification: for the current mold material, injection molding process parameters and product precision requirements, the ratio range corresponding to qualified products is tested; this range is used to determine whether the temperature change in the local degradation critical area is due to the overall environmental factors. If the ratio falls into the range, it indicates that the two temperature change trends are consistent.

[0055] When generating an environmental drift marker, it is determined whether the third slope and the first slope satisfy a preset physical correlation condition: if they satisfy the condition, the environmental drift marker is maintained; if they do not satisfy the condition, the environmental drift marker is corrected to a local degradation marker. In this embodiment, the preset physical correlation conditions include: when the reference temperature rises, the critical pressure should change in the opposite direction, and the negative ratio of the third slope to the first slope should fall within the pressure-temperature gradient coupling range calibrated based on historical health operation data; when generating an environmental drift indicator, it is determined whether the third slope and the first slope satisfy the preset physical correlation conditions: if satisfied, the environmental drift indicator is maintained; if not satisfied, the environmental drift indicator is corrected to a local degradation indicator, including: Obtain the sign directions of the first slope and the third slope, and determine whether the sign direction of the third slope is opposite to that of the first slope.

[0056] Specifically, the first slope represents the rate of change of the reference temperature, and the third slope represents the rate of change of the critical pressure. Based on the physical law of thermal expansion and contraction, the cavity volume increases when the reference temperature rises, leading to a decrease in pressure, so the signs should be opposite. This judgment is used to preliminarily verify whether the drift originates from changes in the ambient temperature.

[0057] When the directions are opposite, calculate the opposite of the ratio of the third slope to the first slope, and determine whether the opposite number is within the pressure-temperature gradient coupling range.

[0058] Specifically, the pressure-temperature gradient coupling range is calibrated based on the mold's historical healthy operation data, specifically the measured pressure and temperature changes during the stable production phase of the mold's first use. This reflects the degree of temperature influence on pressure under normal operating conditions; the formula for calculating the negative of the ratio is... ; In the formula, The third slope, The first slope is represented by the negative number, which characterizes the pressure response amplitude caused by a unit temperature change. If the value falls within the range, it indicates that the pressure change conforms to the expected thermodynamic coupling characteristics.

[0059] When the negative number is within the pressure-temperature gradient coupling range, it is determined that the preset physical correlation condition is met, and the environmental drift indicator is maintained.

[0060] Specifically, if the calculated negative number falls within the range, it indicates that the direction and magnitude of the key pressure change are in normal coupling with the reference temperature, verifying that the drift does indeed originate from overall environmental factors. Therefore, the environmental drift indicator should be kept unchanged.

[0061] When the sign direction of the third slope is the same as that of the first slope, or the opposite number is outside the pressure-temperature gradient coupling range, it is determined that the physical correlation condition is not met, and the environmental drift indicator is corrected to a local degradation indicator.

[0062] Specifically, if the same direction of the sign violates the law of thermal expansion and contraction, or if the opposite number exceeds the range, it indicates that there is interference from non-temperature factors, such as mold wear or cooling water channel blockage. In this case, the original environmental drift indicator should be corrected to a local degradation indicator.

[0063] S4. If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window of the local degradation critical area is narrowed and an early warning is triggered. The initial tolerance window is used to set a fluctuation reference range for the temperature and pressure of the local degradation critical area at the beginning of the injection molding process. If an environmental drift indicator is maintained, the initial tolerance window is widened and a feedforward fine-tuning command is generated.

[0064] In this embodiment, if a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window for the local degradation critical area is narrowed, and an early warning is triggered, including: Based on the generated material fatigue markers or local degradation markers, the initial tolerance window for the critical local degradation area is narrowed to a smaller preset strict tolerance limit; the strict tolerance limit is the maximum permissible fluctuation half-width that is pre-calibrated for different fault types and severity based on previous process verification and material fatigue experiments to ensure that no defective products are produced.

[0065] Specifically, the initial tolerance window refers to the fluctuation reference range of temperature and pressure set for the critical area of ​​local degradation during the initial stage of injection molding; the strict tolerance limit is calibrated through previous process verification and material fatigue experiments to ensure that defective products are not produced even if there is a fault within this limit; the narrowing operation is used to improve the monitoring sensitivity of material fatigue or local degradation state.

[0066] When a material fatigue identifier is generated, a first warning message is generated. The first warning message includes a material fatigue failure type code and a prompt text for checking the material status of the mold micro-area.

[0067] Specifically, the material fatigue failure type code is a predefined code, indicating that the mold material has suffered fatigue damage; the prompt text for checking the material condition of the mold micro-area suggests that operators confirm the degree of material deterioration through metallographic analysis or hardness testing in order to arrange maintenance.

[0068] When a local degradation flag is generated, a second warning message is generated. The second warning message includes a code for the local physical degradation fault type and a prompt text for checking for scale buildup in the cooling water passages.

[0069] Specifically, the local physical degradation fault type code represents a performance degradation category such as cooling channel scaling or cavity wear; the prompt text for checking cooling channel scaling suggests that operators clean the channels or check the seals to restore the mold's heat exchange efficiency.

[0070] In this embodiment, if the environmental drift flag is maintained, the initial tolerance window is widened and a feedforward fine-tuning instruction is generated, including: If the environmental drift indicator is maintained, when the second slope is positive, the upper limit boundary of the initial tolerance window is moved outward by a preset temperature relaxation margin; when the second slope is negative, the lower limit boundary is moved outward by a preset temperature relaxation margin to obtain the relaxed tolerance window; the preset temperature relaxation margin is a unilateral tolerance allowance pre-calibrated based on the measured data of the initial dimensional accuracy requirements of the product and the thermal inertia response of the mold under historical environmental temperature fluctuations, ensuring that environmental drift does not cause quality defects.

[0071] Specifically, the initial tolerance window is the fluctuation reference range set for the temperature of the critical area of ​​local degradation during the initial stage of injection molding; the second slope is the rate of change of the critical temperature, and its sign indicates the trend of temperature rise and fall; the preset temperature allowance is pre-calibrated based on the product dimensional accuracy requirements and historical measured data (mold thermal inertia response under ambient temperature fluctuations), and is used to unidirectionally expand the window boundary to allow the process parameters to fluctuate naturally in order to reduce unnecessary compensation adjustments.

[0072] The direction of cooling water flow adjustment is determined based on the sign of the first slope, and the amount of cooling water flow adjustment is determined based on the absolute value of the first slope. A feedforward fine-tuning command is then generated, which is used to control the cooling water flow adjustment.

[0073] Specifically, the first slope represents the reference temperature change rate, and its sign indicates the adjustment direction (positive for increasing cooling water flow, negative for decreasing it); its absolute value characterizes the rate of temperature change, and the adjustment amount is calculated through a pre-set proportional coefficient; the feedforward fine-tuning command is sent to the cooling water flow control valve in advance to achieve proactive adjustment to maintain process stability.

[0074] Please see Figure 2 This invention provides an intelligent injection molding process control system for plastic molds, comprising: Signal acquisition module: used to acquire the reference temperature of the pre-divided environmentally sensitive reference area, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal in the local degradation critical area within the mold; Acoustic fingerprint construction module: used to calculate the first slope of reference temperature, the second slope of key temperature, and the third slope of key pressure within a predetermined sliding time window; extract the sound wave propagation time, peak amplitude, and center frequency of ultrasonic echo signals to construct a real-time acoustic fingerprint; State recognition module: Used to calculate the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. When the Euclidean distance exceeds the preset fatigue threshold within the preset fatigue judgment time window, a material fatigue label is generated; otherwise, no label is generated. When no material fatigue label is generated, it is determined whether the ratio of the second slope to the first slope falls within a preset ratio range: if yes, it is determined that the process drift originates from overall environmental factors, and an environmental drift label is generated; if no, it is determined that the process drift originates from local physical degradation, and a local degradation label is generated. When an environmental drift label is generated, it is determined whether the third slope and the first slope meet the preset physical correlation conditions: if yes, the environmental drift label is maintained; if no, the environmental drift label is corrected to a local degradation label. Window adjustment module: If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window of the local degradation critical area is narrowed and an early warning is triggered. The initial tolerance window is used to set a fluctuation reference range for the temperature and pressure of the local degradation critical area at the beginning of the injection molding process. If the environmental drift indicator is maintained, the initial tolerance window is widened and a feedforward fine-tuning command is generated.

[0075] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0076] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0077] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for intelligent injection molding process control of plastic molds, characterized in that, Includes the following steps: S1. Collect the reference temperature of the pre-divided environmentally sensitive reference area, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal in the local degradation critical area within the mold. S2. Calculate the first slope of the reference temperature, the second slope of the key temperature, and the third slope of the key pressure within the predetermined sliding time window; extract the sound wave propagation time, peak amplitude, and center frequency of the ultrasonic echo signal to construct a real-time acoustic fingerprint; S3. Calculate the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. If the Euclidean distance exceeds the preset fatigue threshold within the preset fatigue judgment time window, generate a material fatigue indicator; otherwise, do not generate one. When no material fatigue indicator is generated, it is determined whether the ratio of the second slope to the first slope falls within a preset ratio range: if yes, it is determined that the process drift originates from overall environmental factors and an environmental drift indicator is generated; if no, it is determined that the process drift originates from local physical degradation and a local degradation indicator is generated. When generating an environmental drift marker, it is determined whether the third slope and the first slope satisfy a preset physical correlation condition: if they satisfy the condition, the environmental drift marker is maintained; if they do not satisfy the condition, the environmental drift marker is corrected to a local degradation marker. S4. If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window of the local degradation critical area is narrowed and an early warning is triggered. The initial tolerance window is used to set a fluctuation reference range for the temperature and pressure of the local degradation critical area at the beginning of the injection molding process. If an environmental drift indicator is maintained, the initial tolerance window is widened and a feedforward fine-tuning command is generated.

2. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, The acquisition includes the reference temperature of the pre-divided environmentally sensitive reference zone within the mold, the critical temperature and critical pressure of the pre-divided locally degraded critical zone, and the ultrasonic echo signal within the locally degraded critical zone, including: The raw resistance signal is collected by a PT100 platinum resistance temperature sensor deployed in the environmentally sensitive reference area. The raw resistance signal is then converted to resistance-temperature to generate a reference temperature for the environmentally sensitive reference area. The original thermoelectric potential signal is collected by K-type miniature thermocouples deployed in the local degradation critical area. Cold junction compensation and nonlinear correction are performed on the original thermoelectric potential signal to generate the critical temperature of the local degradation critical area. A piezoelectric pressure sensor deployed in a critical area of ​​local degradation is used to collect the original charge signal, and a voltage signal is generated by conditioning the original charge signal through a charge amplifier. The voltage signal is converted from analog to digital and the pressure calibration curve is calculated to generate the key pressure in the critical area of ​​local degradation. By emitting high-frequency electrical pulses through miniature ultrasonic transducers deployed in critical areas of local degradation, the echo electrical signals, after being reflected and scattered by the internal microstructure of the mold material and converted by the miniature ultrasonic transducers, are collected. Bandpass filtering and analog-to-digital conversion are performed on the echo electrical signals to generate ultrasonic echo signals.

3. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, The calculation of the first slope of the reference temperature, the second slope of the critical temperature, and the third slope of the critical pressure within the predetermined sliding time window includes: Within a preset sampling time, the reference temperature, critical temperature, and critical pressure that fall within a predetermined sliding time window are sequentially captured. Linear regression was performed with the time of data collection as the independent variable and the reference temperature as the dependent variable. The regression line of the reference temperature over time was calculated using the least squares method. The slope of the regression line was calculated and used as the first slope. Linear regression was performed with the acquisition time as the independent variable and the key temperature value as the dependent variable. The regression line of the key temperature over time was calculated using the least squares method. The slope of the regression line was calculated and used as the second slope. Linear regression was performed with the data collection time as the independent variable and the critical pressure value as the dependent variable. The regression line of critical pressure over time was calculated using the least squares method. The slope of the regression line was calculated and used as the third slope.

4. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, The extraction of the ultrasonic echo signal's propagation time, peak amplitude, and center frequency to construct a real-time acoustic fingerprint includes: Perform Hilbert transform on the ultrasonic echo signal to generate the envelope curve of the ultrasonic echo signal; On the envelope curve of the ultrasonic echo signal, the first peak point whose amplitude exceeds the preset amplitude threshold is identified, and the time corresponding to the peak point is taken as the arrival time of the sound wave. The difference between the arrival time of the sound wave and the preset emission reference time is calculated to generate the sound wave propagation time. The maximum amplitude within the pre-sampling interval is extracted from the envelope curve of the ultrasonic echo signal to generate the peak amplitude. Perform a fast Fourier transform on the ultrasonic echo signal, extract the frequency of the maximum amplitude from the transform result, and generate the center frequency. The sound wave propagation time, peak amplitude, and center frequency are combined into a feature vector, which is then used as a real-time acoustic fingerprint.

5. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, The preset health baseline fingerprint is a standard acoustic fingerprint obtained by collecting a large amount of acoustic fingerprint data and statistically processing it, and is used as a comparison benchmark for real-time acoustic fingerprints; the calculation of the Euclidean distance between the real-time acoustic fingerprint and the preset health baseline fingerprint, and the generation of a material fatigue indicator when the Euclidean distance exceeds a preset fatigue threshold within a preset fatigue judgment time window; otherwise, no indicator is generated, including: Within a preset fatigue assessment time window, the following process is periodically repeated: acquire ultrasonic echo signals, extract sound wave propagation time, peak amplitude, and center frequency, construct a real-time acoustic fingerprint, calculate the differences between the sound wave propagation time, peak amplitude, and center frequency of the real-time acoustic fingerprint and the baseline sound wave propagation time, baseline peak amplitude, and baseline center frequency of the preset healthy baseline fingerprint, construct a difference vector from the differences, calculate the Euclidean norm of the difference vector, and obtain the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. When every Euclidean distance calculated consecutively within the preset fatigue judgment time window exceeds the preset fatigue threshold, a material fatigue indicator is generated. When the Euclidean distance of any sampling point within the preset fatigue judgment time window is less than or equal to the preset fatigue threshold, no material fatigue identifier is generated.

6. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, The preset ratio range is determined through process verification based on mold material, injection molding process parameters, and product precision requirements, and is used to characterize the consistency of the change trend between the critical temperature and the reference temperature.

7. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, The preset physical correlation conditions include: when the reference temperature rises, the critical pressure should change in the opposite direction, and the negative ratio of the third slope to the first slope should fall within the pressure-temperature gradient coupling range calibrated based on historical health operation data; when generating an environmental drift indicator, it is determined whether the third slope and the first slope satisfy the preset physical correlation conditions: if satisfied, the environmental drift indicator is maintained; if not satisfied, the environmental drift indicator is corrected to a local degradation indicator, including: Obtain the sign directions of the first slope and the third slope, and determine whether the sign direction of the third slope is opposite to that of the first slope; When the directions are opposite, calculate the opposite of the ratio of the third slope to the first slope, and determine whether the opposite number is within the pressure-temperature gradient coupling range; When the opposite number is within the pressure-temperature gradient coupling range, it is determined that the preset physical correlation condition is met, and the environmental drift indicator is maintained. When the sign direction of the third slope is the same as that of the first slope, or the opposite number is outside the pressure-temperature gradient coupling range, it is determined that the physical correlation condition is not met, and the environmental drift indicator is corrected to a local degradation indicator.

8. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window for the local degradation critical area will be narrowed, and an early warning will be triggered, including: Based on the generated material fatigue markers or local degradation markers, the initial tolerance window for the critical local degradation area is narrowed to a smaller preset strict tolerance limit; the strict tolerance limit is the maximum allowable fluctuation half-width that is pre-calibrated for different fault types and severity based on previous process verification and material fatigue experiments to ensure that no defective products are produced. When a material fatigue identifier is generated, a first warning message is generated. The first warning message includes a material fatigue failure type code and a prompt text for checking the material status of the mold micro-area. When a local degradation identifier is generated, a second warning message is generated. The second warning message includes a local physical degradation fault type code and a prompt text for checking for scale buildup in the cooling water passages.

9. The intelligent injection molding process control method for plastic molds according to claim 1, characterized in that, If the environmental drift indicator is maintained, the initial tolerance window is widened, and a feedforward fine-tuning instruction is generated, including: If the environmental drift indicator is maintained, when the second slope is positive, the upper limit boundary of the initial tolerance window is moved outward by a preset temperature relaxation margin; when the second slope is negative, the lower limit boundary is moved outward by a preset temperature relaxation margin to obtain the relaxed tolerance window; the preset temperature relaxation margin is a unilateral tolerance margin pre-calibrated based on the measured data of the initial dimensional accuracy requirements of the product and the thermal inertia response of the mold under historical environmental temperature fluctuations, ensuring that environmental drift does not cause quality defects. The direction of cooling water flow adjustment is determined based on the sign of the first slope, and the amount of cooling water flow adjustment is determined based on the absolute value of the first slope. A feedforward fine-tuning command is then generated, which is used to control the cooling water flow adjustment.

10. A smart injection molding process control system for plastic molds, used in accordance with the smart injection molding process control method for plastic molds as described in any one of claims 1 to 9, characterized in that, include: Signal acquisition module: used to acquire the reference temperature of the pre-divided environmentally sensitive reference area, the critical temperature and critical pressure of the pre-divided local degradation critical area, and the ultrasonic echo signal in the local degradation critical area within the mold; Acoustic fingerprint construction module: used to calculate the first slope of reference temperature, the second slope of key temperature, and the third slope of key pressure within a predetermined sliding time window; extract the sound wave propagation time, peak amplitude, and center frequency of ultrasonic echo signals to construct a real-time acoustic fingerprint; State recognition module: Used to calculate the Euclidean distance between the real-time acoustic fingerprint and the preset healthy baseline fingerprint. When the Euclidean distance exceeds the preset fatigue threshold within the preset fatigue judgment time window, a material fatigue label is generated; otherwise, no label is generated. When no material fatigue label is generated, it is determined whether the ratio of the second slope to the first slope falls within a preset ratio range: if yes, it is determined that the process drift originates from overall environmental factors, and an environmental drift label is generated; if no, it is determined that the process drift originates from local physical degradation, and a local degradation label is generated. When an environmental drift label is generated, it is determined whether the third slope and the first slope meet the preset physical correlation conditions: if yes, the environmental drift label is maintained; if no, the environmental drift label is corrected to a local degradation label. Window adjustment module: If a material fatigue indicator or a local degradation indicator is generated, the preset initial tolerance window of the local degradation critical area is narrowed and an early warning is triggered. The initial tolerance window is used to set a fluctuation reference range for the temperature and pressure of the local degradation critical area at the beginning of the injection molding process. If the environmental drift indicator is maintained, the initial tolerance window is widened and a feedforward fine-tuning command is generated.

Images