Intelligent control method and system for injection molding machine capable of recycling plastic

Abstract

The invention relates to the technical field of plastic forming processing, in particular to an intelligent control method and system for an injection molding machine for recycling plastic. The technical problem that the existing method adopts a fixed or simple feedback proportion and is difficult to continuously maintain the stability of the quality of the injection molding product is solved. The method comprises the following steps: acquiring real-time viscosity data acquired by a first sensor and a second sensor in real time; on the basis of the real-time viscosity data, preset reference viscosity data and the material flowing duration, determining an abnormal judgment result of the recycled material quality; performing primary regulation and control on the mixing proportion of the new material and the recycled material of the injection molding machine based on the primary regulation and control direction and the primary regulation and control quantity; on the basis of the approaching degree of the corrected viscosity data after the first regulation and control and the reference viscosity data, determining a secondary regulation and control amount; and performing secondary regulation and control on the mixing proportion of the new material and the recycled material of the injection molding machine based on the secondary regulation and control quantity. The method is used for intelligent control scenes of the injection molding machine.

Description

Technical Field

[0001] This invention relates to the field of plastic molding and processing technology, and specifically to an intelligent control method and system for injection molding machines that utilize recycled plastics. Background Technology

[0002] In the plastics manufacturing industry, adding a certain proportion of recycled plastics during injection molding has become a common practice to reduce raw material costs and implement green manufacturing principles. As the core equipment for this process, the injection molding machine's control system typically needs to manage the mixing ratio of virgin and recycled materials to balance product quality and production costs. Currently, common control methods rely on preset fixed mixing ratios or simple feedback adjustments based on limited process parameters. However, recycled materials, due to their complex origins and multiple heat treatments, often exhibit significant and unpredictable fluctuations in their molecular chain structure and mechanical properties. This makes it difficult to maintain consistent injection molded product quality when dealing with batch-to-batch variations in recycled materials, and hinders the ability to maximize recycled material utilization while ensuring quality. Summary of the Invention

[0003] To address the technical problem that existing methods, which use fixed or simple feedback ratios, struggle to maintain consistent injection molded product quality when dealing with recycled materials exhibiting significant batch variations, the present invention aims to provide an intelligent control method and system for injection molding machines using recycled plastics. The specific technical solution adopted is as follows: In a first aspect, the present invention provides an intelligent control method for an injection molding machine using recycled plastics. The method includes: acquiring real-time viscosity data collected by a first sensor and a second sensor; arranging the first and second sensors sequentially along the material flow direction in the melt flow channel of the injection molding machine; the material being a mixture of virgin material and recycled material; determining an anomaly judgment result for the quality of the recycled material based on the real-time viscosity data, preset reference viscosity data, and the material flow duration; the material flow duration being the time it takes for the material to flow from the position of the first sensor to the position of the second sensor; when the anomaly judgment result indicates an anomaly in the quality of the recycled material, performing a first adjustment on the mixing ratio of virgin material and recycled material in the injection molding machine based on a preliminary adjustment direction and a preliminary adjustment amount; the preliminary adjustment direction and preliminary adjustment amount being determined based on anomaly characteristic information from the anomaly judgment result; the anomaly characteristic information being used to characterize the attributes and degree of the anomaly in the quality of the recycled material; determining a second adjustment amount based on the degree of convergence between the corrected viscosity data after the first adjustment and the reference viscosity data; and performing a second adjustment on the mixing ratio of virgin material and recycled material in the injection molding machine based on the second adjustment amount.

[0004] In conjunction with the first aspect above, in one possible implementation, the real-time viscosity data includes: first real-time viscosity data and second real-time viscosity data; the first real-time viscosity data is real-time viscosity data collected by a first sensor, and the second real-time viscosity data is real-time viscosity data collected by a second sensor; the reference viscosity data includes: first reference viscosity data and second reference viscosity data; the first reference viscosity data is reference viscosity data of the first sensor, and the second reference viscosity data is reference viscosity data of the second sensor.

[0005] In conjunction with the first aspect described above, in one possible implementation, the method further includes: determining a first mass offset intensity value and a first control tendency based on the numerical change trend of a first difference amplitude during the monitoring period of the first sensor; the first control tendency is used to indicate whether the proportion of new material needs to be increased or decreased; the first difference amplitude is the difference amplitude between the first real-time viscosity data and the first reference viscosity data; determining a second mass offset intensity value and a second control tendency based on the numerical change trend of a second difference amplitude during the monitoring period of the second sensor; the second control tendency is used to indicate whether the proportion of new material needs to be increased or decreased; the second difference amplitude is the difference amplitude between the second real-time viscosity data and the second reference viscosity data; and forming a preliminary anomaly judgment based on the first mass offset intensity value and the second mass offset intensity value.

[0006] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: determining whether a first control tendency and a second control tendency are consistent; determining the numerical change trend of the first difference amplitude of the first sensor during a historical monitoring period; the historical monitoring period and the current monitoring period of the second sensor are separated by a material flow time; determining whether the numerical change trend of the second difference amplitude of the second sensor during the current monitoring period is consistent with the numerical change trend of the first difference amplitude of the first sensor during the historical monitoring period; when the first control tendency and the second control tendency are consistent and the numerical change trends are consistent, confirming the anomaly judgment result as an anomaly in the quality of the recycled material; determining a collaborative evaluation value based on the degree of matching between the numerical change trends of the first difference amplitude and the second difference amplitude; the collaborative evaluation value is used to characterize the consistency of the change trends.

[0007] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: determining a control necessity coefficient based on a first mass offset intensity value and a collaborative evaluation value; determining a basic control amount based on a preset benchmark adjustment step size and the control necessity coefficient; and determining a preliminary control amount based on the basic control amount and the preliminary control direction indicated by the first control tendency and / or the second control tendency.

[0008] In conjunction with the first aspect described above, in one possible implementation, the method specifically includes: acquiring the first real-time difference amplitude of the first sensor at the start of the first regulation; the first real-time difference amplitude being the difference amplitude between the first real-time viscosity data and the first reference viscosity data at the start of the first regulation; acquiring the second real-time difference amplitude of the first sensor at the end of the process response cycle; the second real-time difference amplitude being the difference amplitude between the first real-time viscosity data and the first reference viscosity data at the end of the process response cycle; the process response cycle being greater than or equal to the theoretical residence time of the material from the injection molding machine feed port to the position of the first sensor; determining the regulation effect characterization value based on the change of the second real-time difference amplitude relative to the first real-time difference amplitude; and determining the secondary regulation amount based on the initial regulation amount, the regulation effect characterization value, and the regulation effect information reflected by the second real-time difference amplitude.

[0009] In conjunction with the first aspect described above, in one possible implementation, the method further includes: instructing the injection molding machine to operate with a standard mixing ratio of virgin material and recycled material and standard process parameters; collecting a first viscosity measurement value from a first sensor and a second viscosity measurement value from a second sensor; setting the statistical average of the first viscosity measurement values ​​collected by the first sensor as a first reference viscosity data; and setting the statistical average of the second viscosity measurement values ​​collected by the second sensor as a second reference viscosity data.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: determining the material flow duration based on the real-time rotational speed of the injection molding machine screw and a preset rotational speed-duration mapping relationship.

[0011] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: after completing the second adjustment, re-executing the step of acquiring real-time viscosity data collected in real time by the first and second sensors to form closed-loop control.

[0012] Secondly, the present invention provides an intelligent control system for an injection molding machine that utilizes recycled plastics. The system includes: a data acquisition unit for acquiring real-time viscosity data collected by a first sensor and a second sensor; the first and second sensors are arranged sequentially along the material flow direction in the melt flow channel of the injection molding machine; the material is a mixture of virgin material and recycled material; a quality assessment unit for determining an anomaly judgment result of the recycled material quality based on real-time viscosity data, preset benchmark viscosity data, and material flow duration; the material flow duration is the time it takes for the material to flow from the position of the first sensor to the position of the second sensor; a decision unit for performing a first adjustment on the mixing ratio of virgin material and recycled material in the injection molding machine based on a preliminary adjustment direction and a preliminary adjustment amount when the anomaly judgment result indicates an anomaly in the recycled material quality; the preliminary adjustment direction and preliminary adjustment amount are determined based on anomaly characteristic information from the anomaly judgment result; the anomaly characteristic information is used to characterize the attributes and degree of the recycled material quality anomaly; an adjustment amount determination unit for determining a second adjustment amount based on the degree of convergence between the corrected viscosity data after the first adjustment and the benchmark viscosity data; and an adjustment unit for performing a second adjustment on the mixing ratio of virgin material and recycled material in the injection molding machine based on the second adjustment amount.

[0013] The present invention has the following beneficial effects: This invention achieves real-time sensing of recycled material quality fluctuations by acquiring real-time viscosity data from sequentially arranged dual sensors and judging anomalies based on comparisons with benchmark values ​​and material flow duration. When an anomaly is detected, the direction and magnitude of adjustment are determined based on characteristic information representing the attributes and degree of the anomaly, and a first proportional adjustment is performed. Then, based on the degree to which the adjusted viscosity data approaches the benchmark, a second adjustment is determined and a final adjustment is performed. This invention can proactively respond to dynamic changes in recycled material quality. Through two-stage fine adjustment, it significantly reduces the need for a constant high proportion of virgin material addition while ensuring stable injection melt performance and final product quality. This achieves a unified optimization of quality stability and production economy, thus solving the technical problem of existing methods that use fixed or simple feedback ratios, making it difficult to maintain the stability of injection molded product quality when dealing with recycled materials with significant batch variations. Attached Figure Description

[0014] 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.

[0015] Figure 1 This is a schematic flowchart illustrating an intelligent control method for a recycled plastic injection molding machine according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the system architecture of an intelligent control system for a plastic injection molding machine, provided as an embodiment of the present invention. Detailed Implementation

[0016] 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 an intelligent control method and system for recycled plastic injection molding machines 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.

[0017] 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.

[0018] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent control method and system for reusing plastic injection molding machines provided by the present invention.

[0019] Please see Figure 1 The diagram illustrates a flow chart of an intelligent control method for an injection molding machine that utilizes recycled plastics, according to an embodiment of the present invention. The method includes the following steps S101-S105, which will be described in detail below.

[0020] S101. Obtain real-time viscosity data collected by the first and second sensors.

[0021] The first and second sensors are arranged sequentially along the material flow direction on the melt flow channel of the injection molding machine; the material is composed of a mixture of virgin material and recycled material.

[0022] In one possible implementation, a first sensor and a second sensor are sequentially installed on the melt flow channel of the injection molding machine along the flow direction of the material from the feed port to the mold cavity. The analog or digital signals of the two sensors are collected and converted into corresponding first real-time viscosity data and second real-time viscosity data sequences, respectively.

[0023] S102. Based on real-time viscosity data, preset benchmark viscosity data, and material flow time, determine the abnormal judgment result of the quality of recycled material.

[0024] In one possible implementation, the real-time viscosity data includes: first real-time viscosity data and second real-time viscosity data; the first real-time viscosity data is real-time viscosity data collected by a first sensor, and the second real-time viscosity data is real-time viscosity data collected by a second sensor; the reference viscosity data includes: first reference viscosity data and second reference viscosity data; the first reference viscosity data is reference viscosity data of the first sensor, and the second reference viscosity data is reference viscosity data of the second sensor.

[0025] In one possible implementation, firstly, for the first and second real-time viscosity data, the real-time difference amplitudes with the corresponding reference viscosity data are calculated. The numerical change trends of the first and second difference amplitudes over a continuous monitoring period are analyzed. Based on these trends, the first and second mass offset intensity values, reflecting the severity of the anomaly, are determined, along with the first and second control tendencies indicating the direction of control requirements. Then, a key verification stage is entered: on one hand, verifying whether the first and second control tendencies are consistent; on the other hand, comparing the change trend of the second difference amplitude observed by the second sensor in the current monitoring period with the change trend of the first difference amplitude observed by the first sensor in a historical monitoring period one material flow duration prior, to verify whether the two are consistent. When the control tendencies are consistent and the temporal change trends are consistent, an anomaly in the quality of the recovered material is confirmed, and a synergistic evaluation value is generated based on the degree of matching between the two trends. Finally, the anomaly judgment result is reflected as a series of anomaly characteristic information used to guide control, which includes at least the first mass offset intensity value, the second mass offset intensity value, and the synergistic evaluation value.

[0026] S103. When the abnormality judgment result indicates that the quality of the recycled material is abnormal, the mixing ratio of the new material and the recycled material in the injection molding machine is adjusted for the first time based on the initial adjustment direction and the initial adjustment amount.

[0027] Among them, the initial control direction and the initial control amount are determined based on the abnormal feature information of the abnormal judgment results; the abnormal feature information is used to characterize the attributes and degree of abnormality in the quality of recycled materials.

[0028] In one possible implementation, when an abnormality in the quality of recycled material is confirmed, the output of the abnormality judgment result and the associated abnormality characteristic information (which characterizes the attributes and degree of the quality abnormality) are transmitted to the control decision module. Based on the control requirements implied by the abnormality characteristic information, a specific preliminary control direction (indicating whether the proportion of virgin material should be increased or decreased) and a specific preliminary control amount (indicating the magnitude of the proportion adjustment) are analyzed and determined. Subsequently, a corresponding control signal is generated to drive the virgin material and recycled material feeding mechanism (such as a loss-in-weight feeder or a proportional valve) of the injection molding machine to perform the first adjustment, changing the instantaneous supply rate of the two materials according to the preliminary control direction and preliminary control amount, thereby changing the mixing ratio entering the hopper in real time.

[0029] S104. Based on the degree of similarity between the corrected viscosity data after the first adjustment and the reference viscosity data, determine the amount of the second adjustment.

[0030] In one possible implementation, after the initial control is executed, a preset process response cycle is entered for state monitoring and effect evaluation. During this cycle, viscosity monitoring data from the first sensor is continuously acquired as a basis for evaluating the control effect. At the end of the process response cycle, the acquired viscosity data (i.e., the corrected viscosity data reflecting the state after the initial control) is compared with the baseline viscosity data, and the degree of convergence between the two is calculated and evaluated. Based on the control effect information reflected by this degree of convergence, a secondary control amount for subsequent precise adjustments is calculated and determined through an internally preset optimization decision logic.

[0031] S105. The mixing ratio of new material and recycled material in the injection molding machine is adjusted a second time based on the secondary adjustment amount.

[0032] In one possible implementation, once the secondary control amount is determined, a decision command is immediately issued. A corresponding control signal is generated to directly drive the new material and recycled material feeding mechanism of the injection molding machine. Based on the precise adjustment range and direction indicated by the secondary control amount, the supply ratio of the two materials is adjusted a second time. This adjustment is a direct optimization and correction of the result of the first control.

[0033] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment acquires real-time viscosity data from sequentially arranged dual sensors and makes anomaly judgments based on the comparison with the benchmark value and the material flow time, thereby realizing real-time perception of the quality fluctuations of recycled materials. When an anomaly is judged, the control direction and amount are determined based on the characteristic information representing the abnormal attributes and degree, and the first proportional adjustment is performed; then, based on the degree to which the adjusted viscosity data approaches the benchmark, the second control amount is determined and the final adjustment is performed. It can actively respond to the dynamic changes in the quality of recycled materials. Through two-level fine adjustment, while ensuring the stability of injection molding melt performance and final product quality, it significantly reduces the need for a constant high proportion of virgin material, thereby achieving a unified optimization of quality stability and production economy. This solves the technical problem that existing methods using fixed or simple feedback ratios are unable to continuously maintain the stability of injection molded product quality when facing recycled materials with significant batch differences.

[0034] In one possible implementation, a preliminary anomaly judgment needs to be formed based on the first mass offset intensity value and the second mass offset intensity value. This process can be implemented in detail below through the following S201-S203.

[0035] S201. Based on the numerical change trend of the first difference amplitude during the monitoring period of the first sensor, determine the first mass offset intensity value and the first control tendency.

[0036] Among them, the first control tendency is used to indicate whether the proportion of new material needs to be increased or decreased; the first difference range is the difference range between the first real-time viscosity data and the first reference viscosity data.

[0037] In one possible implementation, for the first sensor, the first difference amplitude between the first real-time viscosity data and the first reference viscosity data at each sampling moment is continuously calculated, forming a data sequence arranged in chronological order. A recent continuous monitoring period (e.g., including the last 5 sampling points) is selected, and the numerical change trend of the first difference amplitude sequence within this period is analyzed. Based on this analysis, two key outputs are simultaneously determined: one is a first mass offset intensity value, which is a comprehensive quantitative index used to describe the intensity of abnormal fluctuations and the level of persistence risk exhibited by the first difference amplitude within the monitoring period; the other is a first control tendency, which is a directional indicator used to clarify whether to adopt a control strategy of increasing or decreasing the proportion of new material in response to the currently monitored trend.

[0038] For example, the analysis of numerical trend changes may include assessing the monotonicity of the sequence (continuous increase, continuous decrease, or fluctuation) and the rate of change. Typically, if the first difference magnitude shows a continuous decreasing trend, it may indicate a decreasing melt viscosity and a potential deterioration in the quality of recycled material; in this case, the first regulatory tendency is to increase the proportion of virgin material. Conversely, if it shows a continuous increasing trend, the tendency may be to reduce the proportion of virgin material. The first quality deviation intensity value is then comprehensively assessed based on the significance and magnitude of the trend; the more significant the trend and the faster the change, the higher the risk value.

[0039] For example, the difference between the value monitored by the sensor at time i and the reference value. Satisfy the following formula 1: in, This is the sequence number of the sampling time, used to distinguish data from different time points; Let be the real-time viscosity value measured by the sensor at the i-th sampling time; This is the reference viscosity value for the sensor, a constant calibrated under standard process conditions; The absolute deviation between the current measured viscosity and the ideal reference viscosity was calculated, directly reflecting the magnitude of the instantaneous deviation; divided by the denominator... The purpose is to normalize, converting absolute deviation into relative deviation relative to a reference value, so that process data from different sensors and different batches are comparable; It characterizes the relative degree of deviation of the melt viscosity from the standard state (qualified mixture) at a specific moment.

[0040] It can be understood that in the injection molding process, the viscosity of the plastic melt cannot be zero, therefore It is a constant greater than zero, and there is no case where the denominator is zero.

[0041] For example, the mass offset intensity value detected by the sensor at time i. The following formula 2 is satisfied: in, For time from time in to time i-1 The arithmetic mean of the series represents the recent average level of difference; n is the length of the selected historical data, for example, n=5 represents the trend of the last 5 historical points; Let the difference magnitude be at time j; The sum of the first-order differences of the magnitude of the difference between adjacent time points is calculated. Its absolute value represents the cumulative strength and direction of the change in the magnitude of the difference over the most recent n time periods. A positive sum indicates a continuous upward trend, while a negative sum indicates a continuous downward trend. This is a normalization function that maps the calculation results to the interval [0, 1] through linear normalization.

[0042] This represents the severity of the deviation; the higher the average value, the higher the overall level of deviation in the recent period. This represents the degree of drastic change; the larger the absolute value, the more pronounced the rapid increase or decrease in recent deviation. Multiplying the two means that the risk value is affected by both the level of deviation and the trend of change. Only when both are high will the risk value increase significantly, which aligns with engineering intuition: the situation of both deviation from the benchmark and rapid deterioration is the most dangerous.

[0043] S202. Based on the numerical change trend of the second difference amplitude during the monitoring period of the second sensor, determine the second mass offset intensity value and the second control tendency.

[0044] Among them, the second control tendency is used to indicate whether the proportion of new material needs to be increased or decreased; the second difference range is the difference range between the second real-time viscosity data and the second reference viscosity data.

[0045] In one possible implementation, for the second sensor, the second difference amplitude between the second real-time viscosity data and the second reference viscosity data at each sampling time is continuously calculated, forming a corresponding time-series data sequence. A continuous monitoring period synchronized with or corresponding to the analysis of the first sensor is selected, and the numerical change trend of the second difference amplitude sequence within this period is independently analyzed. Based on this analysis, two key outputs are simultaneously determined: one is the second mass offset intensity value, which is a quantitative indicator used to characterize the intensity and risk level of abnormal fluctuations reflected by the second difference amplitude at the monitoring location of the second sensor; the other is the second control tendency, which is a directional indicator used to clarify whether the control direction should be increased or decreased based on the monitoring trend of the second sensor itself.

[0046] For example, the analysis method for numerical change trends is consistent with the principles used at the first sensor. If the second difference amplitude shows a continuous decreasing trend, it usually indicates that the melt viscosity at that monitoring point is also decreasing, and the determined second control tendency is to increase the proportion of new material; if it shows a continuous increasing trend, the tendency may be to reduce the proportion of new material. The second mass offset intensity value is given after a comprehensive evaluation based on the significance and magnitude of the trend at that location.

[0047] S203. A preliminary anomaly judgment is made based on the first mass offset intensity value and the second mass offset intensity value.

[0048] In one possible implementation, the two risk values ​​are combined and compared, and a comprehensive preliminary anomaly judgment result is generated based on preset logical rules. This result is not the final conclusion, but a trigger signal or status flag indicating that a higher-level, more rigorous verification process is needed.

[0049] For example, the preset logic rule could be a threshold comparison and logical "OR" / "AND" operation between two risk values. For instance, when the mass offset intensity value of either sensor exceeds its independently set threshold (e.g., 0.7), an anomaly is determined to exist, triggering a preliminary anomaly judgment; or, to reduce false alarms, it can be set to trigger only when both risk values ​​exceed their respective thresholds.

[0050] The technical solution provided by the above embodiments can bring at least the following beneficial effects: Based on data independence, this embodiment determines the intensity value of mass deviation and the control tendency by analyzing the numerical change trend of the difference amplitude of each sensor. It transforms instantaneous viscosity deviation into a trend-based risk quantification indicator and a clear direction for control needs. It not only focuses on the deviation of the current state but also senses the trend and intensity of mass changes, thereby achieving an upgrade from static threshold alarm to dynamic trend early warning, significantly improving the predictive judgment ability for potential mass degradation or fluctuations.

[0051] In one possible implementation, it is also necessary to determine the collaborative evaluation value based on the degree of matching between the numerical change trend of the first difference magnitude and the numerical change trend of the second difference magnitude. This process can be specifically implemented through the following S301-S305, which will be explained in detail below.

[0052] S301. Determine whether the first regulatory tendency and the second regulatory tendency are consistent.

[0053] In one possible implementation, a first regulatory tendency (based on a first sensor trend) and a second regulatory tendency (based on a second sensor trend) determined by the upstream analysis process are obtained. The regulatory tendency is a binary or multi-logic value with a clear direction, such as whether the proportion of new material needs to be increased or can be reduced. Consistency is determined by directly comparing whether these two logic values ​​are the same.

[0054] For example, if the first regulatory tendency is to increase the proportion of new materials and the second regulatory tendency is also to increase the proportion of new materials, then they are considered to be consistent; if the first regulatory tendency is to increase the proportion of new materials and the second regulatory tendency is to reduce the proportion of new materials, then they are considered to be inconsistent.

[0055] S302. Determine the numerical trend of the first difference amplitude of the first sensor during the historical monitoring period.

[0056] In one possible implementation, the specific time range of the historical monitoring period is calculated based on the material flow duration. This period is separated from the current monitoring period of the second sensor by one material flow duration on the time axis. Multiple first difference amplitude data points continuously recorded by the first sensor within the historical monitoring period are retrieved to form a historical sequence of first difference amplitude within the period. Then, the numerical change trend of the historical sequence is identified and characterized.

[0057] For example, suppose the current time is T, and the material flow duration is... The historical monitoring period is [T- - T- ],in The window length required for trend analysis (e.g., 5 seconds). Analyze the curve formed by the first difference magnitude data points within this time period to determine its trend, such as whether it is continuously rising, continuously falling, or basically stable, and may extract the morphological characteristics of the trend (e.g., rate of change).

[0058] S303. Determine whether the numerical trend of the second difference amplitude of the second sensor during the current monitoring period is consistent with the numerical trend of the first difference amplitude of the first sensor during historical monitoring periods.

[0059] In one possible implementation, based on clearly defining the historical monitoring period and the current monitoring period (separated by a material flow duration), two analyses are performed in parallel: first, feature extraction of the numerical change trend of the first difference amplitude sequence recorded by the first sensor during the historical monitoring period; and second, simultaneous trend feature extraction of the second difference amplitude sequence recorded by the second sensor during the current monitoring period. Subsequently, the two sets of extracted trend features are compared, and a quantitative matching score is calculated or a qualitative logical judgment is made using a preset matching degree evaluation algorithm to determine whether the two meet the consistency criteria.

[0060] For example, trend feature extraction may include determining the main monotonic direction of the sequence (such as rising or falling), the rate of change, and the fluctuation pattern. Consistency determination does not require the two curves to completely overlap, but rather focuses on the similarity of key features. For instance, if the historical trend of the first sensor shows a rapid, linear decline, and the current trend of the second sensor also shows a rapid, linear decline, even if the absolute values ​​are different, the trends can be determined to be consistent. Conversely, if the historical trend is downward while the current trend is upward or has no obvious trend, then inconsistency is determined.

[0061] S304. When the first control tendency and the second control tendency are consistent and the numerical change trend is consistent, the abnormal judgment result is confirmed as an abnormality in the quality of the recycled material.

[0062] In one possible implementation, two key judgment results are received from the upstream verification process: one is a logical value regarding whether the first and second control tendencies are consistent; the other is a logical value regarding whether the historical trend of the first sensor and the current trend of the second sensor are consistent. These two logical values ​​are used as inputs, and a comprehensive judgment is made according to preset decision rules. The anomaly judgment result is confirmed as an anomaly in the quality of the recycled material if and only if both input values ​​are true (i.e., the control tendencies are consistent, and the numerical change trends are consistent).

[0063] S305. Determine the collaborative evaluation value based on the degree of matching between the numerical change trend of the first difference magnitude and the numerical change trend of the second difference magnitude.

[0064] One possible implementation involves obtaining two consistent difference amplitude sequences: a first difference amplitude sequence from the first sensor during a historical monitoring period, and a second difference amplitude sequence from the second sensor during the current monitoring period. Using a pre-defined similarity or correlation evaluation algorithm, the overall shape, direction of change, and fluctuation rhythm of these two sequences are quantitatively compared to calculate a co-evaluation value characterizing their matching degree. This value is a numerical value within a pre-defined range (e.g., between 0 and 1), and its magnitude directly and uniquely reflects the degree of agreement or similarity between the two trend sequences after considering time delay.

[0065] For example, the collaborative evaluation value of sensor q (front) and sensor h (back). The following formula 3 is satisfied: Where t is the flow time required for the material to flow from sensor position q to sensor position h; Let q be the viscosity measurement value of sensor q at time jt (i.e., t time in advance); Here, is the viscosity measurement value of sensor h at the current time j; m is the length of the sliding time window for comparison (e.g., m=5). Calculate the cumulative absolute error between the aligned readings of the two sensors within the window period; The normalization function maps the cumulative absolute error to the interval [0, 1] by normalizing the maximum and minimum values.

[0066] The total difference between the two signal sequences after alignment was calculated. The smaller the value, the closer the two signals are in shape and value. Then the degree of difference is converted into the degree of consistency. The higher the value (the closer to 1), the higher the timing consistency between the two signals; The quantification determines the extent to which the signal received by the subsequent sensor is a reproduction of the signal from the preceding sensor after a time delay of t.

[0067] The technical solution provided by the above embodiments can bring at least the following beneficial effects: Based on trend analysis, this embodiment introduces the verification of the consistency of the control tendency of the two sensors, and creatively requires that the current trend of the second sensor be consistent with the historical trend of the first sensor one flow time in advance. A verification logic based on the physical flow sequence of materials is constructed. This logic can effectively distinguish between real abnormal signals transmitted along the flow channel caused by batch quality fluctuations of recycled materials, and instantaneous and synchronous interference signals caused by equipment vibration, temperature disturbances, etc., which greatly improves the specificity and reliability of anomaly judgment.

[0068] In one possible implementation, it is also necessary to determine the initial control amount, which can be achieved through the following steps S401-S403, which will be explained in detail below.

[0069] S401. Based on the first mass offset intensity value and the collaborative evaluation value, determine the control necessity coefficient.

[0070] In one possible implementation, a first mass offset intensity value and a co-evaluation value are used. The first mass offset intensity value quantifies the strength of the abnormal trend detected at the upstream monitoring point (at the first sensor); the co-evaluation value quantifies the matching reliability of the abnormal signal transmitted from upstream to downstream (at the second sensor). These two input values ​​are comprehensively processed through a preset fusion rule or decision function to output a single regulation necessity coefficient. This coefficient is a continuous scalar value, and its magnitude directly reflects the urgency and strength of proportional regulation determined based on all current evidence.

[0071] For example, the regulation necessity coefficient at time i. The following formula 4 is satisfied: in, The collaborative evaluation value ranges from [0, 1] and represents the temporal consistency (matching degree) of the sensor signals before and after. The average difference amplitude of the upstream sensor (sensor q) within the time window from time it to time i is the value of the value. It represents the strength of the abnormal signal. The larger the value, the more serious the deviation of the upstream detected quality from the benchmark, and the greater the need for regulation. This is used to modulate the anomaly intensity based on the signal's credibility. When the co-evaluation value z is high (close to 1), the value of 1 / (1+z) is close to 1 / 2. This means that for highly credible signals, the anomaly intensity r participates in the final decision with approximately 50% weight. When the co-evaluation value z is low (close to 0), the value of 1 / (1+z) is close to 1. Since physically significant anomalies are usually accompanied by credible transmission signals, the value of r itself tends to decrease, making modulation necessary. Maintaining a low level naturally suppresses erroneous regulation based on low-reliability signals, reflecting the control principle of prioritizing safety.

[0072] S402. Determine the basic control quantity based on the preset benchmark adjustment step size and control necessity coefficient.

[0073] In one possible implementation, a pre-set baseline adjustment step size is invoked. This step size is a constant designed and determined based on process experience, representing the unit change in a single proportional adjustment (e.g., a 5% change in the proportion of new material). Simultaneously, a control necessity coefficient is obtained. This coefficient is a continuous dimensionless scalar; a larger value indicates a stronger control requirement. The basic control amount is calculated by multiplying the baseline adjustment step size by the control necessity coefficient. Numerically, this basic control amount is equal to the baseline adjustment step size scaled by the control necessity coefficient. Physically, it represents the initial adjustment magnitude that should be applied based on the current decision, ignoring the control direction.

[0074] S403. Determine the initial control amount based on the basic control amount and the initial control direction indicated by the first control tendency and / or the second control tendency.

[0075] In one possible implementation, a basic control quantity (a scalar value representing the control magnitude) and a preliminary control direction (a logical direction determined by a first control tendency and / or a second control tendency, such as increasing or decreasing new material) are obtained. Based on the preliminary control direction, the basic control quantity is assigned a corresponding directional attribute, thereby synthesizing the final preliminary control quantity. This preliminary control quantity is a complete instruction with a sign (positive or negative) or explicit semantics (such as increasing by X% or decreasing by X%).

[0076] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment uses the first mass offset intensity value (characterizing the intensity of the anomaly) and the collaborative evaluation value (characterizing the credibility of the anomaly) to determine the necessity coefficient of regulation, and then combines it with the benchmark step size to generate the basic regulation amount, and finally determines the preliminary regulation amount by the regulation tendency. By integrating the two key dimensions of anomaly judgment, severity and true credibility, the magnitude of the preliminary regulation amount not only reflects the severity of the problem, but also considers the certainty of the judgment, thereby avoiding over-regulation or under-regulation caused by misjudgment, and making each proportional adjustment more prudent and well-founded.

[0077] In one possible implementation, the process of determining the secondary control amount based on the degree of similarity between the corrected viscosity data after the first control and the reference viscosity data can be specifically implemented through the following S501-S504, which will be explained in detail below.

[0078] S501. Obtain the first real-time difference amplitude of the first sensor at the start of the first regulation.

[0079] The first real-time difference amplitude is the difference amplitude between the first real-time viscosity data and the first reference viscosity data at the start of the first adjustment.

[0080] In one possible implementation, at the instant when a preliminary control command is generated based on the anomaly judgment result and the actuator begins the first control, the first real-time difference amplitude, measured and calculated in real time by the first sensor at that moment (i.e., at the start of the first control), is captured and saved as a specific state variable. The first real-time difference amplitude is the difference between the first real-time viscosity data from the first sensor and the first reference viscosity data at that precise moment. Acquiring and fixing this value establishes a clear and unique comparison origin for subsequent calculations of the control effect.

[0081] S502, Obtain the second real-time difference amplitude of the first sensor at the end of the process response cycle.

[0082] The second real-time difference amplitude is the difference amplitude between the first real-time viscosity data and the first reference viscosity data at the end of the process response cycle; the process response cycle is greater than or equal to the theoretical residence time of the material from the injection molding machine feed port to the position of the first sensor.

[0083] In one possible implementation, after the first adjustment is executed, a preset process response cycle is initiated and timing begins. This cycle is a pre-set waiting time based on process knowledge, designed to allow the new mixture to be conveyed from the feed port, plasticized, and flow to the monitoring position of the first sensor, thereby enabling the sensor to stably reflect the melt state under the adjusted new material ratio. When this process response cycle ends, the real-time difference amplitude measured and calculated by the first sensor at this moment is immediately acquired and recorded as the second real-time difference amplitude. The second real-time difference amplitude is the difference between the first real-time viscosity data of the first sensor and the first reference viscosity data at that specific evaluation moment.

[0084] Understandably, the duration of the process response cycle must be greater than or equal to the theoretical residence time required for the material to be conveyed from the injection molding machine feed port to the installation position of the first sensor. This setting ensures that the obtained second real-time difference amplitude is indeed the effect of the new material ratio changed by the first adjustment, rather than the old material state before the adjustment.

[0085] For example, if the theoretical residence time is 40 seconds, the process response cycle can be set to 45 to 60 seconds.

[0086] S503. Determine the characterization value of the adjustment effect based on the change in the second real-time difference amplitude relative to the first real-time difference amplitude.

[0087] In one possible implementation, a first real-time difference magnitude recorded at the start of regulation (as the initial state baseline) and a second real-time difference magnitude recorded at the end of the process response cycle (as the post-response state) are obtained. The change between these two values ​​is calculated, and its absolute value is determined as the characterization value of the regulation effect. This value is a non-negative scalar, and its magnitude directly characterizes the magnitude or intensity of the change or influence of the first regulation action on the upstream melt viscosity state.

[0088] S504. Based on the information on the control effect reflected by the initial control amount, the control effect characterization value, and the second real-time difference amplitude, determine the secondary control amount.

[0089] In one possible implementation, the initial control amount, the control effect representation value, and the second real-time difference magnitude are used as input parameters and fed into a pre-defined optimization decision algorithm. The initial control amount represents the intensity of the previously taken control action; the control effect representation value quantifies the actual impact of the action; and the second real-time difference magnitude represents the residual deviation that still exists after the action has been executed and the response has occurred. By analyzing the intrinsic relationship between these three parameters, the control efficiency and the current state are comprehensively evaluated, and an optimized secondary control amount is finally calculated and output.

[0090] For example, secondary regulation amount The following formula 5 is satisfied: in, This is for initial regulation. To adjust the effect characterization value, that is, the change in the difference amplitude of the first sensor before and after the first adjustment, The larger the value, the more significant the initial adjustment effect; This represents the current difference magnitude of the first sensor at the end of the process response cycle after the first adjustment. These are parameter tuning coefficients, and their values ​​should be extremely small positive numbers (e.g., 0.01) to avoid denominators of 0.

[0091] This constitutes a feedback mechanism based on results. In the denominator, it means that the more significant the initial adjustment effect ( The larger the value, the more secondary adjustments are needed. The smaller the relative size; This constitutes a feedback based on the current state, the current residual bias. The larger the value, the further the distance from the target is, thus requiring a secondary adjustment. The larger; It is a corrective adjustment amount derived through optimized calculations after observing the effects of the first adjustment.

[0092] In one possible implementation, to ensure the robustness of the control system and prevent [further issues], When the value is extremely small, unrealistically large regulatory commands are generated, requiring adjustments to the calculated values. Impose a maximum limit. Specifically, set a maximum permissible amount of control variation. This threshold can be preset based on process knowledge and system safety requirements. For example, it can be set as a multiple of the initial control amount ∆r (e.g., 2 times), or directly set as a fixed upper limit of proportional change (e.g., 10%). Not exceeding If the theoretical calculated value exceeds the given value, then the calculation will be applied; if the theoretical calculated value exceeds the given value, then the calculation will be applied. The final amount of regulation is then limited to .

[0093] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment quantifies the adjustment effect characterization value by obtaining the real-time difference amplitude change before and after the first adjustment, and determines the second adjustment amount by comprehensively considering the initial adjustment amount, the effect characterization value, and the current residual difference amplitude. A feedback optimization mechanism based on the actual effect of the initial adjustment is established. It no longer relies on a single open-loop adjustment, but can correct and optimize subsequent adjustment commands based on the response effect of the previous adjustment action, thereby achieving a more refined and adaptive progressive approximation adjustment of the mixing ratio, effectively improving the accuracy and stability of the final control result.

[0094] In one possible implementation, it is also necessary to determine the first reference viscosity data and the second reference viscosity data. This process can be specifically implemented through the following S601-S604, which will be explained in detail below.

[0095] S601 indicates that the injection molding machine should operate with the standard mixing ratio of virgin and recycled materials and the standard process parameters.

[0096] In one possible implementation, upon initial operation or periodic calibration, an instruction is sent to the injection molding machine to operate according to a pre-set standard mixing ratio of virgin and recycled materials, and standard process parameters (including but not limited to barrel temperature, screw speed, and back pressure), representing ideal or typical production conditions. Once the injection molding machine operates under these settings and reaches a stable production state, the data acquisition unit begins to synchronously acquire and record the viscosity measurement signals output by the first and second sensors over a continuous period. Subsequently, the acquired viscosity measurement value sequences from each sensor are statistically analyzed, and their statistical average (e.g., arithmetic mean) is calculated and set as the reference viscosity data for the corresponding sensor.

[0097] S602, Collect the first viscosity measurement value from the first sensor and the second viscosity measurement value from the second sensor.

[0098] In one possible implementation, after the injection molding machine operates under the command of the control unit with the aforementioned standard mixing ratio and standard process parameters and confirms that a stable state has been reached, the data acquisition unit is triggered to begin synchronous or quasi-synchronous continuous acquisition of the output signals from the first and second sensors. The acquisition process lasts for a sufficiently long period to cover multiple injection cycles, thereby obtaining a series of data points that reflect steady-state statistical characteristics. During this process, the raw signal from the first sensor is recorded as a first viscosity measurement value sequence, and the raw signal from the second sensor is recorded as a second viscosity measurement value sequence.

[0099] S603. Set the statistical average value of the first viscosity measurement value collected by the first sensor as the first reference viscosity data.

[0100] In one possible implementation, after fully acquiring the first viscosity measurement value sequence output by the first sensor, a specific statistical average calculation (usually an arithmetic average) is performed on the data sequence, that is, the sum of all validly acquired first viscosity measurements divided by the number of data points. After the calculation is completed, the obtained value is set as the first reference viscosity data corresponding to the first sensor and stored in the system's non-volatile memory as a fixed reference value for all subsequent real-time monitoring and calculations.

[0101] S604. Set the statistical average value of the second viscosity measurement value collected by the second sensor as the second reference viscosity data.

[0102] In one possible implementation, after fully acquiring the second viscosity measurement value sequence output by the second sensor, the same statistical average calculation (usually an arithmetic average) is performed on the data sequence as on the upstream sensor. After the calculation, the resulting value is set as the second reference viscosity data corresponding to the second sensor, and it is stored together with the first reference viscosity data as a complete set of reference parameters.

[0103] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment specifies a method for calibrating the reference viscosity data of each sensor through stable operation measurement under standard operating conditions. This provides a stable and reliable performance reference origin for the entire intelligent control system. This reference value, based on actual standard production conditions, fully integrates the characteristics of specific equipment and standard materials, ensuring that all subsequent real-time comparisons and anomaly judgments have a consistent and reproducible reference standard, fundamentally guaranteeing the long-term consistency and accuracy of monitoring and control logic.

[0104] In one possible implementation, it is also necessary to determine the material flow duration. This process can be specifically implemented through the following S701, which will be described in detail below.

[0105] S701. Based on the real-time rotational speed of the injection molding machine screw and the preset rotational speed-duration mapping relationship, the material flow duration is determined.

[0106] One possible implementation involves acquiring the screw speed signal of the injection molding machine in real time. Simultaneously, the system stores a pre-established speed-time mapping relationship, calibrated under standard conditions (exemplarily 5-15 bar) and a standard temperature range (exemplarily 200°C to 230°C). This relationship exists in the form of data tables, fitted curves, or empirical formulas, explicitly describing the correspondence between the screw speed and the time required for material to flow from the first sensor position to the second sensor position (i.e., material flow duration) under the current injection molding machine and material system. The default operating condition is assumed when determining the duration. By querying or calculating this mapping relationship based on the current real-time screw speed, the corresponding material flow duration can be obtained in real time.

[0107] In one possible implementation, the rotational speed-time mapping relationship can be established through experimental calibration: The injection molding machine is run at different stable screw speeds. The average time required for the material to flow through the channel between the two sensors is actually measured using methods such as tracers, special pulse signals, or direct measurement of the material's arrival time. This establishes a set of data points corresponding to the rotational speed and flow time, which are then fitted to form a continuous mapping relationship that can be used for online querying. This relationship typically exhibits a non-linear inverse relationship where higher rotational speeds result in shorter flow times.

[0108] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment determines key time parameters by establishing a mapping relationship between screw speed and material flow duration. This enables the core logic timing verification of the system to adapt to changes in process parameters (such as different screw speeds) during the actual operation of the injection molding machine. This method of dynamically determining the flow duration ensures that the prediction of signal transmission time is accurate regardless of changes in production rate, thereby guaranteeing the universality and reliability of the timing verification logic under different operating conditions.

[0109] In one possible implementation, a closed-loop control needs to be formed after the second regulation is completed. This process can be specifically implemented through the following S801, which will be explained in detail below.

[0110] S801. After completing the second adjustment, the step of acquiring real-time viscosity data collected by the first and second sensors is repeated to form closed-loop control.

[0111] In one possible implementation, once the second control command is executed and the control action is confirmed to be complete, the control logic will proactively reset its internal state. It will immediately exit the dedicated processing flow of anomaly assessment-control-optimization and return control to the upstream continuous monitoring module. This means the system ignores the history of all previous control actions and starts again from the initial state, repeatedly executing the steps of acquiring real-time viscosity data collected by the first and second sensors. Based on the latest data stream, it re-enters the complete decision-making chain of anomaly assessment-preliminary control-effect evaluation-secondary control.

[0112] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment clarifies that after the second adjustment is completed, the monitoring and judgment process will restart to form a closed-loop control. The entire intelligent control method is constructed as a continuously running, self-iterable loop, capable of uninterrupted monitoring and adjustment of the quality fluctuations of recycled materials between batches or even within batches in the continuous production process. This achieves continuous optimization and self-adaptation of the entire production process, transforming quality control from discrete, fragmented intervention to continuous, embedded process management.

[0113] Please see Figure 2 This diagram illustrates a system architecture of an intelligent control system 200 for a recycled plastic injection molding machine according to an embodiment of the present invention. The system includes: a data acquisition unit 201 for acquiring real-time viscosity data collected by a first sensor and a second sensor; the first and second sensors are arranged sequentially along the material flow direction in the melt flow channel of the injection molding machine; the material is composed of a mixture of virgin and recycled material; and a quality assessment unit 202 for determining anomalies in the quality of the recycled material based on the real-time viscosity data, preset benchmark viscosity data, and the material flow time; the material flow time is the time from the position of the first sensor to the position of the second sensor. The duration of the process; Decision unit 203, used to perform the first adjustment on the mixing ratio of new material and recycled material in the injection molding machine based on the initial adjustment direction and initial adjustment amount when the anomaly judgment result is that the quality of recycled material is abnormal; The initial adjustment direction and initial adjustment amount are determined based on the abnormality characteristic information of the anomaly judgment result; The abnormality characteristic information is used to characterize the attribute and degree of the abnormality of the quality of recycled material; Adjustment amount determination unit 204, used to determine the second adjustment amount based on the degree of convergence between the corrected viscosity data after the first adjustment and the reference viscosity data; Adjustment unit 205, used to perform the second adjustment on the mixing ratio of new material and recycled material in the injection molding machine based on the second adjustment amount.

[0114] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment, through the synergistic effect of units such as data acquisition, quality assessment, decision-making, determination of control quantities, and control, concretizes advanced control methods into modular and integrated hardware and software entities. The system's units have clear division of labor and tight logical connections, enabling efficient and reliable automatic execution of complex monitoring, analysis, and control tasks. This significantly reduces the threshold and complexity of implementing this intelligent control technology, facilitating its rapid promotion and application in industry.

[0115] 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.

[0116] 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 method for intelligent control of an injection molding machine using recycled plastics, characterized in that, The method includes: The system acquires real-time viscosity data collected by the first and second sensors; the first and second sensors are arranged sequentially along the material flow direction in the melt flow channel of the injection molding machine; the material is composed of a mixture of virgin material and recycled material. Based on the real-time viscosity data, the preset reference viscosity data, and the material flow time, the abnormal judgment result of the recycled material quality is determined; the material flow time is the time it takes for the material to flow from the first sensor position to the second sensor position. When the anomaly judgment result indicates that the quality of recycled material is abnormal, the mixing ratio of new material and recycled material in the injection molding machine is adjusted for the first time based on the preliminary adjustment direction and preliminary adjustment amount; the preliminary adjustment direction and preliminary adjustment amount are determined based on the anomaly characteristic information of the anomaly judgment result; the anomaly characteristic information is used to characterize the attributes and degree of the abnormality in the quality of recycled material. The amount of secondary adjustment is determined based on the degree of similarity between the corrected viscosity data after the first adjustment and the reference viscosity data. The mixing ratio of virgin material and recycled material in the injection molding machine is adjusted a second time based on the aforementioned secondary adjustment amount.

2. The intelligent control method for an injection molding machine using recycled plastics according to claim 1, characterized in that, The real-time viscosity data includes: first real-time viscosity data and second real-time viscosity data; the first real-time viscosity data is the real-time viscosity data collected by the first sensor, and the second real-time viscosity data is the real-time viscosity data collected by the second sensor; the reference viscosity data includes: first reference viscosity data and second reference viscosity data; the first reference viscosity data is the reference viscosity data of the first sensor, and the second reference viscosity data is the reference viscosity data of the second sensor.

3. The intelligent control method for an injection molding machine using recycled plastics according to claim 2, characterized in that, The method further includes: Based on the numerical change trend of the first difference amplitude during the monitoring period of the first sensor, a first mass offset intensity value and a first adjustment tendency are determined; the first adjustment tendency is used to indicate whether the proportion of new material needs to be increased or decreased; the first difference amplitude is the difference amplitude between the first real-time viscosity data and the first reference viscosity data; Based on the numerical change trend of the second difference amplitude during the monitoring period of the second sensor, the second mass offset intensity value and the second control tendency are determined; the second control tendency is used to indicate whether the proportion of new material needs to be increased or decreased; the second difference amplitude is the difference amplitude between the second real-time viscosity data and the second reference viscosity data; A preliminary anomaly judgment is made based on the first mass offset intensity value and the second mass offset intensity value.

4. The intelligent control method for an injection molding machine using recycled plastics according to claim 3, characterized in that, The method further includes: Determine whether the first regulatory tendency is consistent with the second regulatory tendency; Determine the numerical trend of the first difference amplitude of the first sensor during a historical monitoring period; the historical monitoring period is separated from the current monitoring period of the second sensor by one material flow duration; Determine whether the numerical change trend of the second difference amplitude of the second sensor during the current monitoring period is consistent with the numerical change trend of the first difference amplitude of the first sensor during the historical monitoring period. When the first control tendency is consistent with the second control tendency, and the numerical change trend is consistent, the abnormal judgment result is confirmed as an abnormality in the quality of the recycled material. A collaborative evaluation value is determined based on the degree of matching between the numerical change trend of the first difference magnitude and the numerical change trend of the second difference magnitude; the collaborative evaluation value is used to characterize the consistency of the change trend.

5. The intelligent control method for an injection molding machine using recycled plastics according to claim 4, characterized in that, The method further includes: Based on the first mass offset intensity value and the collaborative evaluation value, the control necessity coefficient is determined; The basic control amount is determined based on the preset benchmark adjustment step size and the control necessity coefficient. The preliminary control amount is determined based on the basic control amount and the preliminary control direction indicated by the first control tendency and / or the second control tendency.

6. The intelligent control method for an injection molding machine using recycled plastics according to claim 5, characterized in that, The determination of the secondary adjustment amount based on the degree of similarity between the corrected viscosity data after the first adjustment and the reference viscosity data includes: The first real-time difference amplitude of the first sensor is obtained at the start of the first regulation; the first real-time difference amplitude is the difference amplitude between the first real-time viscosity data and the first reference viscosity data at the start of the first regulation. The second real-time difference amplitude of the first sensor is obtained at the end of the process response cycle; the second real-time difference amplitude is the difference amplitude between the first real-time viscosity data and the first reference viscosity data at the end of the process response cycle; the process response cycle is greater than or equal to the theoretical residence time of the material from the injection molding machine feed port to the position of the first sensor; The adjustment effect characterization value is determined based on the change of the second real-time difference amplitude relative to the first real-time difference amplitude; The secondary control amount is determined based on the initial control amount, the control effect characterization value, and the control effect information reflected by the second real-time difference amplitude.

7. The intelligent control method for an injection molding machine using recycled plastics according to claim 1, characterized in that, The method further includes: Instruct the injection molding machine to operate with a standard mixing ratio of virgin and recycled materials and standard process parameters; Collect the first viscosity measurement value from the first sensor and the second viscosity measurement value from the second sensor; The statistical average value of the first viscosity measurement collected by the first sensor is set as the first reference viscosity data. The statistical average value of the second viscosity measurement collected by the second sensor is set as the second reference viscosity data.

8. The intelligent control method for an injection molding machine using recycled plastics according to claim 1, characterized in that, The method further includes: The material flow duration is determined based on the real-time rotational speed of the injection molding machine screw and a preset rotational speed-duration mapping relationship.

9. The intelligent control method for an injection molding machine using recycled plastics according to claim 1, characterized in that, The method further includes: After completing the second adjustment, the step of acquiring real-time viscosity data collected by the first and second sensors is repeated to form a closed-loop control.

10. An intelligent control system for an injection molding machine that utilizes recycled plastics, characterized in that, The system includes: The data acquisition unit is used to acquire real-time viscosity data collected by the first sensor and the second sensor; the first sensor and the second sensor are arranged sequentially along the material flow direction on the melt flow channel of the injection molding machine; the material is composed of a mixture of virgin material and recycled material; The quality assessment unit is used to determine the abnormal judgment result of the quality of the recycled material based on the real-time viscosity data, the preset benchmark viscosity data, and the material flow time; the material flow time is the time it takes for the material to flow from the first sensor position to the second sensor position. The decision unit is used to adjust the mixing ratio of new material and recycled material in the injection molding machine for the first time based on the preliminary adjustment direction and preliminary adjustment amount when the anomaly judgment result indicates that the quality of recycled material is abnormal. The preliminary adjustment direction and preliminary adjustment amount are determined based on the anomaly characteristic information of the anomaly judgment result. The anomaly characteristic information is used to characterize the attributes and degree of the abnormality in the quality of recycled material. The adjustment amount determination unit is used to determine the secondary adjustment amount based on the degree of similarity between the corrected viscosity data after the first adjustment and the reference viscosity data; The control unit is used to perform a second adjustment on the mixing ratio of virgin material and recycled material in the injection molding machine based on the second control amount.

Images