Intelligent control system for PVC shrink film raw material mixing

Abstract

The present application relates to the technical field of high polymer material processing equipment control, and discloses an intelligent control system for PVC shrink film raw material mixing, comprising a data acquisition module, a state solving module, a feedforward control module, a disturbance execution module, a discharge determination module and a main control unit. The collected motor active power and temperature signals are converted into specific mechanical energy and temperature rise rate, a two-dimensional phase plane is constructed, and the tangent slope is calculated to feedforward adjust the servo proportional valve opening degree to match the material rheological state. When the temperature rise rate appears a minimum value, the system actively injects a frequency increment disturbance, and extracts a thermodynamic response characteristic quantity after a set temperature measurement time lag. A two-dimensional response vector is constructed using the characteristic quantity and the input quota threshold. The system determines the real flow state according to the deflection angle of the response vector and outputs a discharge instruction. The present application overcomes the thermal lag error of the temperature sensor, prevents material clumping during the liquid adding process, and improves the accuracy of the discharge end point identification.

Description

Technical Field

[0001] This invention relates to the field of polymer material processing equipment control technology, specifically to an intelligent control system for mixing PVC shrink film raw materials. Background Technology

[0002] Before production, PVC shrink film requires mixing PVC resin powder with plasticizers and other auxiliary materials in a high-speed mixer. During the operation of the mixing equipment, the material undergoes a physical rheological transformation from a dry powder state to a viscous phase transition period under the combined action of mechanical stirring and frictional heat generation, and finally transforms into a dry liquid state after absorption and maturation.

[0003] In existing mixing process control, plasticizers are mostly controlled by constant-rate addition or open-loop timed addition. When the mixture enters the viscous phase transition period, the pores on the particle surface are gradually filled, and the rate of plasticizer absorption decreases accordingly. If the initial addition rate is maintained at this time, the plasticizer that cannot be absorbed in time will accumulate locally inside the mixing tank, leading to the adhesion and clumping of powder materials and reducing the uniformity of mixing.

[0004] Furthermore, due to differences in the initial state of different batches of materials, the control system typically determines the discharge endpoint by monitoring the temperature of the materials within the mixing cylinder. However, in actual production, temperature sensing elements such as thermocouples installed inside the mixing cylinder have inherent heat transfer delays, causing the temperature signals acquired by the system to lag behind the actual thermodynamic state of the materials on the time axis. This sensor temperature measurement lag makes it difficult for the system to accurately pinpoint the process window when the material reaches its optimal dry state when relying solely on a set absolute temperature threshold for discharge determination. Affected by this lag, the discharge action often deviates; discharging too early leads to insufficient material maturation, while discharging too late causes continuous excessive friction, increasing ineffective energy consumption and easily causing the material to stick to the walls or undergo localized degradation due to heat.

[0005] Therefore, this invention proposes an intelligent control system for mixing PVC shrink film raw materials to address the shortcomings of existing technologies. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an intelligent control system for mixing PVC shrink film raw materials. This system solves the problems in existing PVC shrink film raw material mixing processes, such as the mismatch between the plasticizer addition rate and the material rheological state, which easily leads to material agglomeration, and the thermodynamic lag of the temperature sensing element causing inaccurate identification of the discharge endpoint.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control system for mixing PVC shrink film raw materials, comprising:

[0008] The data acquisition module is used to acquire the motor's three-phase active power sequence, operating frequency, and raw temperature signal output by the frequency converter;

[0009] The state calculation module is used to convert the three-phase active power sequence of the motor into accumulated specific mechanical energy and process the original temperature signal to output the temperature rise rate.

[0010] The feedforward control module is used to calculate the tangent slope based on the specific mechanical energy and the temperature rise rate, and adjust the opening command of the servo proportional valve according to the tangent slope.

[0011] The disturbance execution module is used to issue a frequency increment command when the temperature rise rate exhibits a minimum value characteristic, and to cancel the frequency increment command when the additional injected specific mechanical energy reaches a quota threshold.

[0012] The material discharge determination module is used to extract thermodynamic response characteristic quantities after canceling the frequency increment command and delaying by a set time lag, and output a material discharge command or a reset signal based on the deflection angle of the response vector constructed by the quota threshold and the thermodynamic response characteristic quantities.

[0013] The main control unit is used to maintain the batch control state machine to coordinate the execution priorities among the feedforward control module, the disturbance execution module and the material discharge judgment module, and to lock the servo proportional valve during disturbance verification.

[0014] By constructing a two-dimensional phase plane representing specific mechanical energy and temperature rise rate, the system transforms the physical state changes caused by the absorption of plasticizers during mixing into a numerical characteristic of the tangent slope of the phase plane trajectory. The opening of the servo proportional valve is adjusted using this tangent slope, allowing the liquid addition rate to follow the changes in the material's physical state. Furthermore, when the trajectory curve reaches a minimum, the system inputs a set mechanical energy disturbance to the main stirring motor. After crossing the time delay constant of the temperature sensing element, the system extracts the temperature response increment and determines the outcome based on the deflection angle of the response vector constructed from the quota threshold and the temperature response increment. When the material mixing reaches its endpoint, its internal friction efficiency decreases, and the input mechanical energy is mainly converted into material kinetic energy, resulting in a corresponding decrease in the deflection angle. Outputting a discharge command based on this deflection angle overcomes the endpoint identification error caused by the inherent delay in temperature measurement.

[0015] Furthermore, the data acquisition module also performs no-load calibration in the absence of material feeding: within a preset observation time window, the three-phase active power sequence of the motor is collected according to the set discrete sampling period, and its average value is calculated as the basic frictional active power constant; the basic frictional active power constant is used by the state calculation module to eliminate the mechanical transmission background loss during the subsequent batch mixing process.

[0016] Furthermore, the state calculation module is specifically used to: subtract the basic frictional active power constant from the values ​​in the discretely sampled three-phase active power sequence of the motor to obtain the net input active power; integrate the net input active power in the time domain and divide it by the total mass of the input materials to obtain the accumulated specific mechanical energy.

[0017] Furthermore, the feedforward control module is specifically used to: calculate the ratio of the difference in temperature rise rate between the current update point and the previous update point to the difference in specific mechanical energy, and obtain the tangent slope; when the tangent slope is greater than zero, it is determined that the material is in the pure mechanical friction period, and the reference maximum opening command is output as the opening command of the servo proportional valve; when the tangent slope is less than zero, it is determined that the material has entered the viscous phase transition period, and the preset slope proportional gain coefficient is used in combination with the tangent slope to calculate and attenuate the opening command of the servo proportional valve.

[0018] Furthermore, the feedforward control module also records the cumulative liquid addition amount of the servo proportional valve and imposes a total amount constraint on the liquid addition process according to the target addition amount set in the formula.

[0019] Furthermore, the minimum value characteristic is: the tangent slope changes from negative to positive and stably exceeds the set dead zone tolerance; when the disturbance execution module detects the minimum value characteristic, it marks the current state point as a suspected endpoint and starts an independent integrator to calculate the additional injected specific mechanical energy accumulated after the frequency increment command is issued.

[0020] Furthermore, the thermodynamic response characteristic quantity extracted by the discharge determination module is the true difference in temperature rise rate. The specific extraction process is as follows: after the set time delay ends, within the preset observation window, the difference between the current temperature rise rate and the temperature rise rate at the moment when the frequency increment command is canceled is extracted as the true difference in temperature rise rate.

[0021] Furthermore, when the discharge determination module outputs a discharge command or reset signal based on the deflection angle of the response vector constructed from the quota threshold and the thermodynamic response characteristic quantity, it is specifically used to: multiply the set horizontal axis normalization coefficient by the quota threshold to obtain the vector horizontal coordinate, multiply the vertical axis normalization coefficient by the true difference of the temperature rise rate to obtain the vector vertical coordinate; and calculate the arctangent function of the ratio of the vector vertical coordinate to the vector horizontal coordinate to obtain the deflection angle.

[0022] Furthermore, the material discharge determination module compares the calculated deflection angle with a reference threshold: if the deflection angle is greater than or equal to the reference threshold, it sends a reset signal to the disturbance execution module to end the verification state and simultaneously releases the lock on the servo proportional valve during the disturbance verification; if the deflection angle is less than the reference threshold, it outputs the material discharge command to control the pneumatic material discharge valve to open.

[0023] Furthermore, the batch control state machine of the main control unit is specifically used to: upon receiving the reset signal, control the frequency converter to restore the reference operating frequency and reactivate the feedforward control module to monitor the tangent slope.

[0024] This invention provides an intelligent control system for mixing PVC shrink film raw materials. It has the following beneficial effects:

[0025] 1. This invention constructs a two-dimensional phase plane representing the specific mechanical energy and temperature rise rate, and extracts the tangent slope to adjust the opening command of the servo proportional valve. This allows the plasticizer addition rate to follow the physical state changes of the material from the pure mechanical friction stage to the viscous phase transition stage in real time. This method solves the problem of localized liquid accumulation that easily occurs when the material's pore absorption capacity decreases with conventional constant-rate addition, effectively reducing material agglomeration and improving the uniformity of raw material mixing.

[0026] 2. When the temperature rise rate reaches a minimum, this invention actively injects a frequency increment into the motor and accumulates additional fixed-rate mechanical energy, while simultaneously extracting the thermodynamic response characteristic quantity after a set time delay. This asynchronous observation mechanism avoids the inherent thermal response delay of temperature sensors, enabling the acquisition of the true temperature change difference after the material receives a specific energy input, overcoming the errors caused by traditional methods that rely solely on real-time temperature measurement data for state feedback.

[0027] 3. This invention constructs a two-dimensional response vector using a fixed threshold and thermodynamic response characteristics, and determines the discharge timing based on the comparison between the deflection angle and the benchmark threshold. This determination logic utilizes the physical mechanism that the efficiency of internal friction work decreases when material mixing reaches its endpoint. By accurately identifying the true flow state of the material through the numerical change of the deflection angle, it avoids premature discharge of uncooked material or wall-sticking carbonization caused by excessive friction, thus reducing ineffective mechanical energy consumption while ensuring consistent batch production quality. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the system framework of the present invention;

[0029] Figure 2 This is a schematic diagram of the method flow of the present invention;

[0030] Figure 3This is a schematic diagram of the underlying state data acquisition and basic parameter calibration process of the present invention;

[0031] Figure 4 This is a schematic diagram of the phase plane basic state variable solution process of the present invention;

[0032] Figure 5 This is a schematic diagram of the phase plane trajectory optimization and feedforward liquid addition control process of the present invention;

[0033] Figure 6 This is a schematic diagram of the feature initial screening and energy quota disturbance execution process of the present invention;

[0034] Figure 7 This is a schematic diagram of the asynchronous slip observation and discharge closed-loop determination process of the present invention.

[0035] Figure 8 This is a logic mapping diagram of the phase plane trajectory and liquid addition feedforward control of the present invention;

[0036] Figure 9 This is a thermodynamic response comparison curve for the endpoint identification stage of this invention.

[0037] Among them, 10 is the main control unit; 20 is the main stirring motor; 30 is the communication connection acquisition frequency converter; 40 is the thermocouple sensor; 50 is the servo proportional valve; 60 is the pneumatic discharge valve; 100 is the data acquisition module; 200 is the state calculation module; 300 is the feedforward control module; 400 is the disturbance execution module; and 500 is the discharge judgment module. Detailed Implementation

[0038] The technical solutions in 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.

[0039] Please see the appendix Figure 1 This invention provides an intelligent control system for mixing PVC shrink film raw materials, comprising:

[0040] The data acquisition module 100 is used to acquire the three-phase active power and operating frequency of the motor output by the frequency converter 30 and the raw temperature signal obtained by the thermocouple sensor 40 through the communication connection.

[0041] The state calculation module 200 is used to convert the three-phase active power sequence of the motor into the accumulated specific mechanical energy, and to perform polynomial filtering and differential processing on the original temperature signal to output a smooth temperature rise rate.

[0042] The feedforward control module 300 is used to construct a two-dimensional phase plane coordinate system based on specific mechanical energy and temperature rise rate, calculate the tangent slope of the phase plane trajectory, and adjust the opening command of the servo proportional valve 50 according to the changing characteristics of the tangent slope.

[0043] The disturbance execution module 400 is used to send a frequency increment command to the frequency converter 30 when the temperature rise rate curve shows a minimum value characteristic, and to calculate the additional specific mechanical energy injected during the disturbance stage. When the additional specific mechanical energy injected reaches the quota threshold, the frequency increment command is cancelled.

[0044] The discharge judgment module 500 is used to start a timer after canceling the frequency increment command, delay the set time delay constant to extract thermodynamic response characteristics, construct a two-dimensional response vector using the quota threshold and thermodynamic response characteristics, and control the pneumatic discharge valve 60 to open or send a reset signal to the disturbance execution module 400 according to the comparison result of the deflection angle of the response vector and the reference threshold. The main control unit 10 is also used to maintain the batch control state machine to coordinate the execution priority among the feedforward control module 300, the disturbance execution module 400 and the discharge judgment module 500. When the system is in the disturbance verification stage or the delayed observation stage, the liquid addition control of the servo proportional valve 50 is locked to avoid re-triggering the maximum liquid addition opening due to the tangent slope returning to normal during the endpoint verification process.

[0045] See attached document Figure 2 This invention provides an intelligent control method for mixing PVC shrink film raw materials, comprising the following steps:

[0046] S10, the data acquisition module 100 acquires the three-phase active power, operating frequency and original temperature signal inside the mixing cylinder during the operation of the main stirring motor;

[0047] S20 converts the three-phase active power sequence into specific mechanical energy through the state solution module 200, and performs smoothing, noise reduction and derivative calculation on the original temperature signal to extract the temperature rise rate.

[0048] S30, through the feedforward control module 300, calculates the tangent slope of the state trajectory in a two-dimensional phase plane coordinate system constructed with specific mechanical energy and temperature rise rate, performs feedforward attenuation control on the liquid addition opening of the servo proportional valve based on the change of the tangent slope, and performs total amount constraint on the cumulative liquid addition process of the servo proportional valve according to the target addition amount set in the formula.

[0049] S40, when the temperature rise rate sequence shows a minimum value characteristic, the perturbation execution module 400 injects a frequency increment into the main stirring motor, locks the liquid addition state of the servo proportional valve, and restores the original operating frequency when the additional accumulated specific mechanical energy in this frequency band reaches the quota threshold.

[0050] S50: After the set observation time delay ends, the material discharge judgment module 500 extracts the thermodynamic response characteristic quantity, constructs the response vector using the quota threshold and the thermodynamic response characteristic quantity, and calculates the deflection angle. Based on the comparison result between the deflection angle and the benchmark threshold, the material discharge command is output or the verification status is reset.

[0051] To further clarify the implementation of each technical aspect of the present invention, the following will provide a detailed description of the implementation of each functional module involved above and its internal processing flow.

[0052] See attached document Figure 3 Step S10 is used to complete the acquisition of the basic state variables of the mixing process and the calibration of the no-load friction power, so as to provide discrete data under a unified time base for subsequent specific mechanical energy calculation and temperature rise rate extraction. Step S10 specifically includes the following steps:

[0053] S101, the data acquisition module 100 establishes a hardware communication mapping and performs discrete sampling to obtain the time series of basic physical quantities.

[0054] In this embodiment, the main control unit 10 establishes a communication connection with the frequency converter 30 through an industrial communication bus interface. The data acquisition module 100 accesses the internal communication register of the frequency converter 30 according to a preset communication baud rate and message period to read the three-phase active power of the main stirring motor 20 during operation. and real-time operating frequency .in, The total three-phase active power calculated and output by inverter 30 is not the active power of any single-phase branch.

[0055] Thermocouple sensor 40 is installed in the mixing cylinder to output an analog signal corresponding to the local temperature of the material. The main control unit 10 contains an analog input module that receives the analog signal output from the thermocouple sensor 40 and sequentially performs analog-to-digital conversion, cold junction compensation, linearization conversion, and engineering quantity calibration to obtain discrete raw temperature signals. .

[0056] To ensure that electrical and thermodynamic measurements are consistent across time, the data acquisition module 100 uses a hardware timer from the main control unit 10 to trigger a sampling interrupt, which serves as the unified sampling reference for power, frequency, and temperature data. The system's discrete sampling period is denoted as... As an optional implementation, The time interval is set to 0.1 to 1 second to balance the temperature change resolution, the inverter communication refresh cycle, and the real-time computing load of the main control unit 10.

[0057] Within each sampling period, the data acquisition module 100 acquires the three-phase active power of the motor. Operating frequency and the original temperature signal The data is then written with the same timestamp. The collected data is stored in the buffer of the main control unit 10 in chronological order for use by the status calculation module 200. The network topology of the industrial communication bus, the isolation method of the analog input modules, and the field anti-interference wiring method can be conventionally configured according to the control platform and the field electrical environment, and are not limited here.

[0058] S102, the data acquisition module 100 performs no-load calibration calculation of frictional active power in the absence of material feeding.

[0059] When the main stirring motor 20 drives the mixer's main shaft to rotate, inherent mechanical losses occur in the bearings, couplings, and transmission mechanism. Only the power output by the motor, after deducting these inherent losses, can be considered the effective input power acting on the material's shearing, friction, and tumbling processes. Therefore, before calculating the material's cumulative specific mechanical energy, it is necessary to obtain the basic frictional active power constant used for power deduction. .

[0060] After the equipment is put into operation for the first time, after maintenance and cleaning, or when it is necessary to recalibrate the no-load loss, the main control unit 10 confirms that the pneumatic discharge valve 60 is in the closed state and that no PVC resin or plasticizer has been added to the mixing cylinder, and then issues a no-load operation command to the frequency converter 30. The main stirring motor 20 speeds up to the process reference operating frequency and operates stably at that frequency.

[0061] The length of the preset observation time window is determined based on the equipment power level and startup stabilization time. If the observation time is too short, transient impact power during the acceleration phase may be included; if the observation time is too long, the temperature rise of the transmission components may cause changes in lubrication conditions, resulting in friction loss drift. As a preferred embodiment, the preset observation time window is between 10 and 30 seconds.

[0062] Basic frictional active power It can be calculated using the following formula:

[0063] ;

[0064] In the formula, This represents the basic frictional active power constant; This indicates the total number of discrete sampling points within the preset observation time window; Indicates the first The instantaneous value of the three-phase active power of the motor read at each sampling period.

[0065] After completing the above calibration calculations, the data acquisition module 100 will use the basic frictional active power constant. The constant is written to the non-volatile storage area of ​​the main control unit 10. In subsequent normal material mixing batches, this constant serves as a fixed parameter for the system to eliminate the background loss of the mechanical transmission system and participates in the integral calculation of the state calculation module 200.

[0066] See attached document Figure 4 Step S20 is used to convert the electrical and temperature quantities acquired by the data acquisition module 100 into state variables that can be used for phase plane analysis. The state variables include specific mechanical energy characterizing the material's mechanical energy absorption process and temperature rise rate characterizing the material's thermal response process. Step S20 specifically includes the following steps:

[0067] S201, calculates specific mechanical energy based on the integral of effective input power.

[0068] In the mixing process of PVC shrink film raw materials, if only mixing time is used as the control criterion, variations in the quality of different batches of feed, power grid fluctuations, and changes in stirring load can all lead to inconsistencies in the actual mechanical shear energy experienced by the materials. To reduce the impact of these factors on process judgment, this embodiment uses specific mechanical energy... As a lateral characterization of the energy absorption state of materials, it is used to reflect the actual amount of mechanical work done per unit mass of material during the mixing process.

[0069] In this embodiment, the operator inputs the total mass of materials fed into the mixing cylinder for the current batch through the human-machine interface of the main control unit 10. Before initiating the integration operation, the state solution module 200 performs the following checks: Perform a numerical validity check. If detected... If the material mass is less than or equal to the set minimum material mass limit, the main control unit 10 will issue an alarm for abnormal feeding quality parameters and suspend the specific mechanical energy integration process. After the operator confirms that the current batch uses the standard feeding amount, the system will call the preset default standard batch mass parameters to participate in the subsequent calculations.

[0070] After the material feeding quality parameters are verified, the status calculation module 200 reads the real-time three-phase total active power of the motor output by the data acquisition module 100. And call the pre-calibrated basic frictional active power constant. After deducting the background losses, the net input active power acting on the material shearing, friction and tumbling processes is obtained.

[0071] Before entering the disturbance verification phase, the main stirring motor 20 maintains the reference operating frequency. The running state calculation module 200 is calibrated using the reference operating frequency. Deductions are made. If the operating frequency of the main stirring motor 20 is allowed to change during the normal mixing phase, the state calculation module 200 calculates the frequency according to the pre-established friction power frequency correction model. Alternatively, dynamic compensation can be performed on the basic frictional active power using a frequency ratio correction method.

[0072] Since the main control unit 10 operates with discrete sampling periods, the state calculation module 200 uses a discrete accumulation method to approximate the net input active power through integration. (Cumulative specific mechanical energy) It can be calculated using the following formula:

[0073] ;

[0074] In the formula, This indicates the time elapsed since the start of the hybrid procedure until the current moment. The cumulative mechanical energy; This indicates the total mass of the current batch of materials fed. This represents the total number of sampling points at the current moment; Indicates the first The instantaneous value of the total three-phase active power of the motor read in each sampling cycle; This represents the basic frictional active power constant; This represents the discrete sampling period. Through the above processing, the state calculation module 200 converts the instantaneous power quantity into a cumulative energy state quantity that reflects the material rheological process.

[0075] S202, the state solution module 200 performs a temperature rise rate fidelity derivative based on Savitzky-Golay polynomial filtering.

[0076] The material temperature rise rate reflects the trend of mechanical energy conversion into heat during mixing and is an important longitudinal state quantity for judging the absorption, swelling, and phase change processes of plasticizers. Temperature signals in industrial settings are easily affected by high-frequency interference from frequency converters, electromagnetic noise from motors, and local material disturbances. Directly differentiating the original temperature signal will result in significant fluctuations in the temperature rise rate; using a standard low-pass filter may introduce substantial phase lag and weaken the local extrema of the temperature rise rate curve.

[0077] To balance noise reduction and preservation of extreme value characteristics, in this embodiment, the state calculation module 200 employs the Savitzky-Golay sliding window polynomial fitting algorithm to process the original temperature sequence. This algorithm fits the temperature change trend with a polynomial within a local data window and calculates the first derivative based on the fitted polynomial, thereby suppressing random noise while preserving the temperature rise rate transition characteristics caused by changes in material state.

[0078] Specifically, the state resolution module 200 establishes a memory structure with a length of [length missing]. A sliding data window with multiple sampling points is used to store the raw temperature signal through a first-in-first-out circular buffer. When a new temperature sample value is written to the buffer, if the amount of data in the buffer reaches... If there are 1 sampling point, the state calculation module 200 will use the historical time corresponding to the center point of the window. The rate of temperature rise at the center point is calculated as the time of calculation.

[0079] The above online calculation method will introduce While it has a fixed computational lag, it can avoid extreme point shifts and curve distortions caused by unilateral difference or simple filtering, making it more suitable for subsequent endpoint identification based on the minimum value of the temperature rise rate.

[0080] To avoid phase plane state point misalignment caused by hysteresis in the output of the temperature rise rate, the state calculation module 200 outputs the temperature rise rate... At the same time, the center timestamp corresponding to the temperature rise rate is output synchronously. ,in The state calculation module 200 reads the same central timestamp from the specific mechanical energy cache sequence. corresponding This causes the state points in the subsequent phase plane to be aligned with the time sequence. and constitute.

[0081] Regarding parameter selection, if the sliding window is too short, the temperature rise rate is easily affected by random noise; if the sliding window is too long, it may weaken the local characteristics caused by abrupt changes in material state. If the polynomial order is too low, it is difficult to describe the local curve shape; if the order is too high, overfitting may occur. In this embodiment, the window half-width parameter... Preferably an integer between 5 and 15, the polynomial order. The preferred values ​​are 2 or 3, and the specific values ​​can be adjusted based on the on-site sampling period, the temperature signal signal-to-noise ratio, and the inverter carrier interference.

[0082] Solving for the first derivative of a local polynomial can be equivalent to performing a linear convolution on the original temperature sequence within a sliding window. The first derivative convolution weight coefficients are then used. and , It is related to the sampling period and does not change with real-time temperature data. The main control unit 10 can, during the initialization phase, adjust the settings according to the parameters. and The set of weight coefficients can be calculated and fixed, or the weight coefficients can be written into the storage area in advance for the state calculation module 200 to call.

[0083] Smoothed temperature rise rate It can be calculated using the following formula:

[0084] ;

[0085] In the formula, This represents the current temperature rise rate after smoothing and retaining its true characteristics; The parameter representing the half-width of the sliding data window; This represents the first-order derivative convolution weight coefficients, which are pre-calculated using the least-squares fitting matrix. Indicates the first [number]th ... The original temperature discrete values ​​at each relative offset position.

[0086] Through this smoothing differentiation operation, the state calculation module 200 outputs a temperature rise rate sequence with white noise interference removed, providing reliable data support for the feedforward control module 300 to construct a two-dimensional phase plane.

[0087] See attached document Figure 5 Step S30 is used to feedforward adjust the plasticizer addition process based on the energy absorption state and thermal response changes of the material during the mixing process, so that the addition rhythm no longer depends solely on a fixed time, but corresponds to the actual rheological state of the material. Step S30 specifically includes the following sub-steps:

[0088] S301, the feedforward control module 300 constructs a two-dimensional phase plane coordinate system and solves for the slope of the trajectory tangent.

[0089] During the mixing process of PVC shrink film raw materials, the porosity, initial moisture content, and powder packing state of different batches of PVC resin may vary. If the injection rate of plasticizer is controlled only according to a preset time, problems such as insufficient absorption in some batches and excessive local addition of liquid in other batches may easily occur. To address this, the feedforward control module 300 uses a two-dimensional phase plane composed of specific mechanical energy and temperature rise rate to describe the material state.

[0090] Specifically, the feedforward control module 300 establishes a two-dimensional phase plane coordinate system in the memory area of ​​the main control unit 10. The horizontal axis represents the cumulative specific mechanical energy output by the state calculation module 200. The vertical axis represents the smoothed temperature rise rate. When the state calculation module 200 uses a smooth differentiation algorithm with center point delay output, the state points in the two-dimensional phase plane are timestamped. and ,in The output center timestamp corresponds to the temperature rise rate. As the main stirring motor 20 continues to run, the material state points form a continuous state trajectory in this two-dimensional phase plane.

[0091] The feedforward control module 300 determines the energy conversion trend of the material based on the local tangent slope of the state trajectory. To avoid the specific mechanical energy increment being too small due to excessively short sampling intervals, which would cause instability in the slope calculation, the feedforward control module 300 sets a specific mechanical energy calculation step size threshold. Only when the difference in specific mechanical energy between the current state point and the previous slope update point is greater than or equal to... A tangent slope update is triggered only once. As a preferred implementation, The value can be set from 0.5 kJ / kg to 2.0 kJ / kg, and the specific value can be set according to the rated power of the main stirring motor 20, the sampling period and the batch size of the material.

[0092] Tangent slope of phase plane trajectory It can be calculated using the following formula:

[0093] ;

[0094] In the formula, This represents the slope of the tangent line to the phase plane trajectory at the current update time; and These represent the rate of temperature rise and specific mechanical energy at the current update moment, respectively. This represents the last recorded time when the step size threshold update condition was met. Through the above calculations, the system obtains the phase plane slope characteristic quantity, which directly characterizes the energy absorption and conversion properties of the material.

[0095] S302, performs baseline liquid addition control when the material is in the pure mechanical friction period.

[0096] In the initial stage of mixing, the plasticizer has just come into contact with the PVC resin powder and has not yet fully penetrated into the microporous structure of the resin particles. During this stage, the mechanical energy of the input material is mainly converted into sensible heat through dry friction between particles. The rate of temperature rise usually increases with the increase of specific mechanical energy, and the phase plane trajectory extends to the upper right.

[0097] The feedforward control module 300 continuously monitors the tangent slope. When detected When the judgment result meets the preset anti-shake confirmation conditions, the feedforward control module 300 determines that the material is in the pure mechanical heating stage. At this time, the surface of the resin powder still has a high liquid absorption capacity, and the feedforward control module 300 outputs a reference maximum opening command to the servo proportional valve 50, so that the plasticizer enters the mixing cylinder at a higher flow rate.

[0098] To avoid frequent slope changes near zero due to measurement noise, the feedforward control module has 300 pairs of... The positive / negative judgment is set with a hysteresis dead zone, or the judgment result is required to be held continuously for several sampling periods before updating the valve state. The opening command in this stage can be expressed as:

[0099] ;

[0100] In the formula, This indicates the opening command issued to the servo proportional valve 50 at the current moment; This indicates the pre-set maximum foundation opening limit parameter for the equipment.

[0101] Meanwhile, the feedforward control module 300 does not only control the valve opening based on slope changes, but also imposes total quantity constraints on the liquid addition process based on the target dosage of plasticizer set in the formula. Specifically, the main control unit 10 converts the opening command of the servo proportional valve 50 into the instantaneous liquid addition flow rate according to the pre-calibrated opening-flow mapping relationship, and accumulates the cumulative dosage of the current batch according to the discrete sampling period. When the cumulative dosage reaches the target dosage set in the formula, the feedforward control module 300 outputs a closing command or a preset hold command to the servo proportional valve 50, and marks the liquid addition status of the current batch as complete.

[0102] In other embodiments, if the system is equipped with a flow meter or metering pump, the main control unit 10 preferentially uses the flow meter feedback value or the metering pump pulse feedback value to calculate the cumulative dosage in order to improve the accuracy of total liquid addition control.

[0103] S303 performs feedforward decay control when the material enters the viscous phase transition period.

[0104] As the mixing process progresses, the plasticizer gradually penetrates the microporous structure of the PVC resin particles, causing the material to swell, wet, and gel. This process absorbs some of the mechanical input energy, reducing the proportion of mechanical energy converted into sensible heat, and the rate of temperature rise begins to decline. At this point, the phase plane trajectory changes from an upward trend to a downward trend.

[0105] The current feed control module 300 detected When the judgment result meets the preset anti-shaking confirmation conditions, the material is determined to have entered the viscous phase change absorption period. During this stage, the resin particles' ability to continue absorbing plasticizer decreases. If the maximum liquid addition flow rate is maintained, unabsorbed free plasticizer can easily cause the material surface to become wet and sticky, thereby increasing the risk of agglomeration, clumping, and sudden changes in stirring load.

[0106] To mitigate the aforementioned risks, the feedforward control module 300 performs feedforward attenuation of the opening of the servo proportional valve 50 based on the intensity of the negative slope. Specifically, a negative slope intensity parameter is defined. and with As the input for calculating the opening attenuation, the opening command can be expressed as:

[0107] ;

[0108] In the formula, The preset slope proportional gain coefficient is used to adjust the valve's sensitivity to the phase change state. This is the absolute value of the current negative tangent slope; This is the minimum safe opening limit set to prevent the valve from completely locking up and causing pressure buildup in the pipeline; This is a function to find the maximum value. In actual debugging, Typically set to 10% to 15%; slope proportional gain coefficient The value is determined based on the dynamic viscosity of the plasticizer, and is usually between 5 and 20.

[0109] Through this feedforward attenuation control mechanism, the system achieves adaptive following of the liquid addition rate to the real-time absorption capacity of the material, reducing the sudden change in stirring load caused by excessive local liquid addition.

[0110] When the cumulative dosage reaches the target dosage, or when the batch control state machine enters the disturbance verification state, the feedforward control module 300 stops according to... recover The servo proportional valve 50 remains closed or maintains a preset safe opening.

[0111] See attached document Figure 6 Step S40 is used to actively verify the material state after the temperature rise rate shows suspected endpoint characteristics. Unlike the method of waiting for material discharge according to a fixed time, this embodiment applies a controlled frequency disturbance to the main stirring motor 20 and uses a fixed mechanical energy ratio as the disturbance cut-off condition, so that the subsequent thermal response judgment has a unified energy benchmark.

[0112] In this embodiment, the main control unit 10 internally includes a batch control state machine. The batch control state machine includes at least the following states: liquid addition and mixing state, extreme value candidate state, disturbance verification state, delayed observation state, discharge state, and reset scan state. When the system enters the disturbance verification state or the delayed observation state, the valve opening update logic of the feedforward control module 300 is disabled, and the liquid addition state of the servo proportional valve 50 is locked. The disturbance execution module 400 no longer responds to new extreme points in these states until the discharge determination module 500 outputs a discharge command or a reset signal.

[0113] Step S40 specifically includes the following sub-steps:

[0114] S401, the disturbance execution module 400 performs the initial screening of topological features of the concave extreme points of the temperature rise rate.

[0115] As the absorption of the viscous phase transition nears completion, the free plasticizer is gradually absorbed by the PVC resin, reducing the wetness and stickiness of the particle surface and causing the material to tend towards a dry flow state. At this point, mechanical friction becomes the primary source of heat again, and the rate of temperature rise reverses from decreasing to increasing. Reflected on the phase plane trajectory, this process is characterized by the appearance of a concave extreme point.

[0116] The disturbance execution module 400 continuously scans the tangent slope sequence output by the feedforward control module 300 to capture the aforementioned extreme value characteristics. The tangent slope sequence is calculated based on the phase plane state points after timestamp alignment, thereby avoiding the inconsistency between the extreme value identification time and the specific mechanical energy state caused by temperature filtering hysteresis.

[0117] Specifically, the disturbance execution module 400 detects the zero-crossing process where the tangent slope changes from negative to positive. To reduce false triggering caused by sensor noise, local material inhomogeneity, or short-term disturbances, the disturbance execution module 400 adopts a sliding window confirmation method: when the tangent slope remains positive for multiple consecutive calculation cycles, and the cumulative amplitude of crossing the zero point exceeds the preset dead zone tolerance, the disturbance execution module 400 identifies this state point as a suspected rheological absorption endpoint. As an optional implementation, the continuous calculation cycle is 3 to 5 cycles, and the dead zone tolerance is 0.05 to 0.15.

[0118] Once the suspected rheological absorption endpoint is confirmed, the main control unit 10 locks the current extreme point record and switches the batch control state machine to the disturbance verification state. Simultaneously, the main control unit 10 switches the servo proportional valve 50 to the liquid addition prohibition state, controlling the servo proportional valve 50 to close or maintain a preset safe holding opening. During the disturbance verification state and subsequent delayed observation state, the feedforward control module 300 no longer restores the maximum liquid addition opening based on changes in the tangent slope.

[0119] S402, the disturbance execution module 400 sends a micro-step frequency increment to the frequency converter 30 and triggers the independent integrator.

[0120] A suspected rheological absorption endpoint only indicates a change in the apparent state of the material and cannot be directly equated to the completion of plasticization inside the material. If local liquid bridges or agglomerates still exist inside the material, surface drying may cause a false endpoint. To distinguish between a true dry state and a pseudo-dry powder state, the disturbance execution module 400 sends a frequency increment command to the frequency converter 30 after confirming the extreme point, applying a short-term energy disturbance to the main stirring motor 20.

[0121] The frequency increment command is preferably output via a ramp function to avoid sudden frequency changes that could cause overcurrent protection of the inverter 30 or impact on the transmission system. The frequency increment amplitude can be set to 5% to 10% of the current reference operating frequency. This amplitude produces an observable thermodynamic response without causing excessive impact on the main stirring motor 20 and the transmission mechanism.

[0122] As the frequency converter 30 responds to the frequency increment and drives the main stirring motor 20 to increase speed, the disturbance execution module 400 is reset and the independent integrator is activated. The independent integrator is used to calculate the test phase specific mechanical energy of the injected material during the disturbance verification phase. The integral result is independent of the cumulative mechanical energy throughout the mixing process and is used by the subsequent material discharge judgment module 500 to evaluate the thermal response.

[0123] Considering that after the main mixing motor speed increases by 20, bearing friction, wind resistance, and transmission losses will increase with frequency, if the basic frictional active power at the reference frequency is still used... Subtracting from the effective input power can easily lead to an overestimation of the calculated effective input power. Therefore, the disturbance execution module 400 operates at the test frequency. With reference operating frequency The ratio of the basic frictional active power Perform a first-order correction.

[0124] Test phase compared to mechanical energy It can be calculated using the following formula:

[0125] ;

[0126] In the formula, This represents the cumulative mechanical energy accumulated during the test phase from the triggering of the disturbance to the current moment. This indicates the trigger moment when the suspected rheological absorption endpoint is confirmed and the frequency increment is issued; Indicates the current moment; Indicates the discrete sampling period; This represents the three-phase active power of the motor at the corresponding moment. The basic frictional active power constant; This refers to the test running frequency after the frequency increment is issued; This is the baseline operating frequency for a typical hybrid program; This refers to the total mass of the materials fed.

[0127] Through the aforementioned independent integration, the disturbance execution module 400 separates the energy input of the verification phase from the total energy accumulation of the mixed process, so that each endpoint verification is based on a measurable mechanical energy input as the excitation condition.

[0128] S403, the disturbance execution module 400 cuts off the frequency increment instruction based on the quota threshold mechanism.

[0129] If a fixed-time acceleration method is used for disturbance verification, the mechanical energy of the injected material will not be consistent within the same acceleration time due to differences in material load, friction state, and motor output power among different batches, making it difficult to directly compare the subsequent temperature rise response. Therefore, this embodiment does not use the disturbance duration as the cutoff condition, but uses whether the specific mechanical energy reaches the quota threshold during the test phase as the basis for frequency recovery.

[0130] The disturbance execution module 400 continuously monitors the output of the independent integrator. .when When the preset threshold is reached, the disturbance execution module 400 sends a frequency recovery command to the frequency converter 30 to cancel the previously issued frequency increment, so that the main mixing motor 20 drops back to the reference operating frequency.

[0131] The quota threshold can be set based on the material formula, batch feed amount, material specific heat capacity, and on-site process test results. As an optional implementation, the quota threshold is set between 1.0 kJ / kg and 3.0 kJ / kg.

[0132] By adopting the above-mentioned fixed energy removal method, even if the real-time frictional load of different batches of materials differs, the test mechanical energy of the injected material during the disturbance verification stage can still be limited to a preset range. Therefore, the thermodynamic response characteristics extracted by the discharge determination module 500 in subsequent steps are comparable, providing stable input conditions for determining whether the material has reached the discharge state.

[0133] See attached document Figure 7 Step S50 is used to determine whether the material has reached the discharge state based on the delayed thermal response of the material after the disturbance execution module 400 completes the quota energy disturbance.

[0134] Because the thermocouple sensor 40 has thermal inertia, changes in the temperature signal typically lag behind changes in the actual thermal state of the material. In this embodiment, the discharge determination module 500 does not directly take the value at the moment the frequency disturbance ends, but extracts the temperature rise rate response at the time point of the matching sensor's thermal hysteresis, and further constructs a response vector to distinguish between the true dry flow state and the pseudo dry powder state. Step S50 specifically includes the following steps:

[0135] S501, the material discharge determination module 500 executes an asynchronous time-shift observation mechanism that matches thermal inertia.

[0136] After the disturbance execution module 400 cancels the frequency increment command, the material has completed one quota mechanical energy disturbance, but the thermocouple sensor 40 cannot immediately output the actual temperature change corresponding to the disturbance. The reason is that the thermocouple sensor 40 in industrial settings is usually equipped with a metal protective sleeve. The heat capacity and thermal resistance of the protective sleeve itself will create thermal inertia, causing the sensor output signal to lag behind the actual temperature change of the material.

[0137] If the temperature rise rate is extracted immediately after the disturbance ends, the discharge determination module 500 may still obtain the residual response before or at the beginning of the disturbance, which cannot accurately reflect the material state after the injection of the quota mechanical energy. Therefore, after receiving the frequency increment cancellation signal, the discharge determination module 500 starts a software waiting timer. The duration of the software waiting timer is set according to the pre-calibrated hot jacket time delay constant.

[0138] When the state calculation module 200 adopts the Savitzky-Golay filtering algorithm with center point delay output, the material discharge judgment module 500 adjusts the hot jacket time delay constant and filter fixed hysteresis in its software implementation. A unified compensation can be performed; alternatively, the material discharge judgment module 500 can use the center timestamp corresponding to the filtered output as a reference to extract the temperature rise rate response value at the corresponding moment. This avoids the observation time shift caused by the superposition of temperature filtering delay and sensor thermal hysteresis.

[0139] As an optional implementation, the time delay constant of the heat jacket can be obtained through offline calibration by step heating test under no-load conditions, and the preferred value range is 2 seconds to 8 seconds.

[0140] After the software wait timer expires, the material discharge judgment module 500 opens a preset time observation window and extracts the peak value of the smooth temperature rise rate sequence within this window. The length of the observation window is preferably 3 to 5 sampling periods, which can cover the main response peak value after the disturbance and reduce the impact of subsequent substrate temperature rise drift on the judgment result.

[0141] Subsequently, the material discharge determination module 500 calculates the difference between the peak value and the reference temperature rise rate before the disturbance is triggered, thus obtaining the temperature rise rate difference caused by the rated mechanical energy during this test phase. The reference temperature rise rate before the disturbance trigger can be the average or median value of the smooth temperature rise rate within a preset reference window before the disturbance trigger time. As an optional implementation, the preset reference window covers 3 to 5 slope update cycles to reduce the impact of single-point noise on the difference results.

[0142] S502, the material discharge judgment module 500 constructs the test response vector and calculates the deflection angle.

[0143] Through the standard disturbance in step S40 and the asynchronous observation in step S501, the material discharge determination module 500 obtains two quantities for state determination: one is the standard specific mechanical energy representing the input excitation intensity, and the other is the temperature rise rate difference representing the output thermal response. The material discharge determination module 500 uses the above two quantities to construct a two-dimensional test response vector to characterize the thermal response characteristics of the current material to the disturbance of the quota mechanical energy.

[0144] Since the difference between specific mechanical energy and temperature rise rate belongs to different physical dimensions, they cannot be directly compared as geometric vector components. Therefore, the material discharge judgment module 500 introduces a dimensional normalization coefficient before calculation to unify the scaling of the transverse and longitudinal components respectively.

[0145] in, This is the reciprocal of the specific mechanical energy reference scale. The specific mechanical energy reference scale can be selected from the upper limit of the quota threshold, the standard batch test energy, or the maximum test specific mechanical energy allowed by the process. This is the reciprocal of the reference scale for the rate of temperature rise. The reference scale can be selected as the maximum perturbed rate of temperature rise calibrated from a standard formulation, or the characteristic upper limit of the rate of temperature rise during normal mixing. After normalization, input energy and output thermal response of different dimensions are mapped to a comparable dimensionless range.

[0146] The material discharge judgment module 500 then calculates the deflection angle of the normalized test response vector relative to the horizontal axis. The deflection angle It can be calculated using the following formula:

[0147] ;

[0148] In the formula, This indicates the angle of deflection of the test response vector relative to the horizontal axis; This represents the true difference in temperature rise rate extracted by the asynchronous time-shift observation mechanism; This represents the fixed threshold parameter set in the disturbance execution module 400, i.e., the fixed injected energy; and These represent the dimensionless normalization coefficients for the vertical and horizontal axes, respectively.

[0149] By measuring this angle, the material discharge judgment module 500 converts the strength of the material's thermal response to the disturbance of the quota mechanical energy into a single judgment index. The larger the deflection angle, the more obvious the change in the rate of temperature rise caused by the unit test energy; the smaller the deflection angle, the less significant the viscous shear heat generation caused by the test energy.

[0150] S503, the discharge judgment module 500 performs physical verification of the main flow state and valve discharge logic based on the deflection angle.

[0151] The material discharge determination module 500 will calculate the deflection angle. The angle is compared with a preset benchmark threshold. The benchmark threshold represents the upper limit of the allowable deflection angle of a standard dry-state material under a fixed disturbance. As an optional implementation, the benchmark threshold is between 15 degrees and 35 degrees.

[0152] The baseline threshold can be obtained through standard formula calibration during the trial production stage. Specifically, after the standard formula materials reach the qualified discharge state, the control system repeatedly performs quota disturbance tests, calculates the deflection angle corresponding to each test, and obtains the statistical distribution of the deflection angle; then, the mean of the statistical distribution is added with a safety margin, or the preset upper quantile value of the statistical distribution is set as the baseline threshold.

[0153] When the material discharge determination module 500 detects... When the value is less than the baseline threshold, it indicates that the current test response is within the allowable range of the standard main flow state. At this time, the rated mechanical energy is mainly used for the collision, tumbling, and macroscopic motion between material particles, without generating significant viscous shear heat. Based on this, the discharge judgment module 500 determines that the surface viscosity of the material has decreased, the plasticizer absorption process is basically complete, and the material meets the discharge conditions.

[0154] Under the above circumstances, the discharge determination module 500 outputs a discharge command to the main control unit 10. The main control unit 10 drives the pneumatic discharge valve 60 to open, discharging the mixed material to the downstream cooling section.

[0155] When the material discharge determination module 500 detects... A reading greater than or equal to the baseline threshold indicates that the frequency disturbance has caused a strong rate of temperature rise response. This response typically corresponds to the presence of microscopic liquid bridges, localized agglomerations, or insufficiently absorbed plasticizers within the material. Under the influence of a frequency step, the agglomerates are sheared and torn apart, generating internal frictional heat, which leads to an increase in the deflection of the test response vector.

[0156] In this case, the extreme point identified by the discharge determination module 500 in step S40 is a suspected endpoint, not the actual discharge endpoint. The discharge determination module 500 discards the current extreme point record, clears the current response vector calculation result, and sends a reset signal to the disturbance execution module 400. The disturbance execution module 400 clears the independent integrator; the batch control state machine switches from the delayed observation state to the reset scan state; the main stirring motor 20 maintains the reference operating frequency, and the system re-enters the phase plane feature scan process.

[0157] After reset, the main control unit 10 determines the state of the servo proportional valve 50 based on the cumulative dosage of the current batch. If the cumulative dosage has not yet reached the target dosage set in the formula, the feedforward control module 300 is allowed to resume the liquid addition opening control; if the cumulative dosage has reached the target dosage, the servo proportional valve 50 continues to remain closed or maintains a preset safe opening to avoid repeated liquid addition.

[0158] To prevent repeated system verifications from failing due to sensor drift, formula anomalies, or material batch anomalies, the discharge judgment module 500 is equipped with a retry counter. If the number of consecutively triggered and discarded extreme points in the same batch reaches the maximum retry limit, or if the current total mixing time exceeds the safety timeout threshold, the main control unit 10 triggers an abnormal alarm and switches to a backup protection discharge mode based on a fixed temperature. As an optional implementation, the maximum retry limit is 3 times.

[0159] Before entering the backup protection discharge mode, the main control unit 10 first confirms whether the cumulative dosage of the current batch has reached the target dosage. If the cumulative dosage has not reached the target dosage, the system maintains an alarm state and waits for manual confirmation to avoid direct discharge when the plasticizer dosage is insufficient. If the cumulative dosage has reached the target dosage, the system can continue to execute according to the backup protection discharge logic.

[0160] Through the above-mentioned asynchronous observation, response vector determination and reset protection mechanism, the discharge determination module 500 can identify the true rheological state of the material when there is thermal hysteresis in the sensor, thereby reducing the risk of mis-discharge caused by false dry powder state.

[0161] To further illustrate the technical details and practical effects of the present invention, a complete application example is provided below, taking into account a specific PVC shrink film raw material mixing production scenario, and corresponding experimental verification data is given.

[0162] Specific application examples:

[0163] This embodiment uses a high-speed mixer with a total volume of 500L and a main stirring motor with a rated power of 75kW. The current batch of PVC resin powder and auxiliary materials added totals... =200kg. The target dosage of plasticizer is set at 50kg. Discrete sampling period. Set to 0.5s.

[0164] Before system startup, the discharge valve is closed and there is no material in the cylinder. The main control unit 10 drives the main stirring motor 20 to the reference operating frequency. =30Hz operation. After sampling and calculation within a preset observation time window of 20 seconds, the state calculation module 200 measured the basic frictional active power constant. =12.5kW. Resin powder was then added, and the main stirring motor 20 resumed operation at 30Hz. The system began real-time acquisition of the original temperature signal. and the total active power of the motor At this time, the reading is... =52.5kW, after deducting the background loss, the system recorded a net input active power of 40.0kW at that moment. The state calculation module 200 integrates the discrete net power according to the formula and divides it by the total mass, continuously outputting the cumulative specific mechanical energy. .

[0165] When the system runs to When the value is 15.0 kJ / kg, the state calculation module 200 uses the half-width parameter. A polynomial filter with a value of 10 outputs a smooth temperature rise rate. =0.18℃ / s. The feedforward control module uses the previous update point (14.0 kJ / kg, 0.12℃ / s) as a reference to calculate the current tangent slope. =(0.18−0.12) / (15.0−14.0)=0.06(℃ / s) / (kJ / kg).

[0166] because The system determines that the material is in a purely mechanical friction phase. The feedforward control module 300 issues a reference opening command. =85%, servo proportional valve 50 injects plasticizer at a high flow rate. As mixing continues, due to the large-scale penetration of plasticizer into the powder pores, part of the mechanical work is converted into internal energy rather than sensible heat. When the value is 35.0 kJ / kg, it is measured that =0.11℃ / s, at which point the calculated tangent slope is... =−0.04(℃ / s) / (kJ / kg). Upon detecting a negative slope and meeting the anti-jitter condition, the system determines that the viscous phase transition has begun. Based on the pre-calibrated dynamic viscosity characteristics of this type of plasticizer, the slope proportional gain coefficient is set to... =200. The opening of the servo proportional valve 50 is attenuated by feedforward to: This effectively reduces the risk of localized clumping caused by excessive free droplets.

[0167] As the plasticizer absorption nears completion, the material in the cylinder gradually transitions to a dry flow state. When the concentration reaches 55.0 kJ / kg, the tangent slope changes from negative to positive and stably exceeds the dead zone tolerance of 0.05. The disturbance execution module 400 marks this state point as a suspected endpoint and simultaneously locks the servo proportional valve 50 (at this point, the cumulative dosage has reached 48.5 kg, and liquid addition is temporarily suspended). The main control unit 10 sends a step frequency increment to the frequency converter, and the test operating frequency rises to... =32Hz. Independent integrator starts when the additional injected test phase exceeds the mechanical energy threshold. When the temperature reaches 2.0 kJ / kg, the frequency increment command is cancelled. After the frequency is restored, the discharge judgment module 500 starts a software wait timer with a duration of 5.0 seconds to compensate for the lag of the heat jacket. After the timer expires, the true difference in temperature rise rate is extracted within the subsequent 2.0-second observation window. =0.03℃ / s. The system-defined horizontal axis normalization coefficient. =0.2 (corresponding to the reciprocal of the maximum test energy of 5.0 kJ / kg), normalization coefficient on the vertical axis. =10.0 (the reciprocal of the maximum temperature rise range of 0.1℃ / s). The deflection angle is calculated as follows:

[0168] ;

[0169] The system's baseline threshold is set to 25°. Since 36.87° ≥ 25°, the discharge judgment module 500 determines that micro-agglomeration still exists inside the material at this time, and the previously observed extreme point is a false endpoint. The system clears the independent integrator, discards the extreme point, resumes 30Hz baseline operation, and releases the valve lock to allow the remaining 1.5kg of plasticizer to continue to be added.

[0170] After the system continued operating for 35 seconds, the tangent slope again met the positive over-limit condition. This triggered a second quota disturbance. =2.0 kJ / kg). In this delayed observation, the measured true difference in the rate of temperature rise significantly decreased to =0.012℃ / s. The second deflection angle is calculated as follows:

[0171] ;

[0172] Since 16.70° < 25°, it was determined that the energy quota was completely converted into macroscopic kinetic energy without causing internal friction heat generation. The system confirmed that the material entered the true dry flow state, and the main control unit 10 immediately output a discharge command. The pneumatic discharge valve 60 opened, and the batch was successfully mixed and discharged, effectively avoiding the process accident of premature discharge of uncooked material.

[0173] Experimental verification and effect comparison:

[0174] To verify the effectiveness of the feedforward liquid addition and phase plane asynchronous observation control scheme of the present invention, comparative experiments were carried out under the same hardware equipment and formulation system.

[0175] The experiment was divided into two groups:

[0176] Control group: adopts the traditional open-loop control method based on constant liquid addition rate and a single temperature threshold (discharge when reaching 115℃).

[0177] Implementation Group: The feedforward liquid addition control and asynchronous slip observation verification method provided by this invention were adopted.

[0178] Each group continuously produced 50 standard batches, collecting key process indicators. Experimental statistical results show that the implementation plan demonstrates significant advantages in multiple core process indicators.

[0179] Regarding the consistency of material quality, the cooled and discharged material was screened using a 100-mesh standard sieve for verification. The control group, due to the constant-rate addition of the plasticizer during the viscous phase transition, was prone to localized overwetting, resulting in an average residue on the sieve (agglomeration rate) as high as 4.2%, and a batch-to-batch standard deviation of 1.1%. In contrast, the implementation group, benefiting from two-dimensional phase plane slope feedforward attenuation control, ensured that the plasticizer addition rate strictly followed the material's real-time absorption capacity, reducing the average agglomeration rate to 0.6% and the batch-to-batch standard deviation to 0.15%. This comparison indicates that the particle dispersion state of the implementation group was more ideal.

[0180] Regarding the accuracy of material discharge timing determination, the traditional single temperature threshold method is highly susceptible to the influence of ambient temperature, sensor cold junction drift, and the initial hot state of the equipment. In the control group's 50 batches, six instances of false pulverization leading to mis-discharge occurred (materials severely adhered to the cooling section walls). The implementation group constructed a two-dimensional response vector using micro-step disturbances and delayed response characteristics, utilizing the dimensionless deflection angle... Static thermal hysteresis interference was completely eliminated. The discharge qualification rate of the 50 batches implemented in the group reached 100%. Not only did no wall adhesion accidents occur, but the included angle data fed back under the test conditions also stably fell within the main flow confirmation range of 15° to 22°.

[0181] Regarding energy consumption, the control group often artificially extended the mixing time as a safety margin to ensure the material was not undercooked, resulting in an average specific mechanical energy consumption of 88.5 kJ / kg per batch. The implementation group, however, accurately captured and actively verified the rheological absorption endpoint, eliminating the ineffective excessive friction stage. This reduced the average specific mechanical energy consumption per batch to 76.2 kJ / kg, and the inter-batch energy consumption distribution was extremely convergent, demonstrating a very high adaptive optimization capability.

[0182] See attached document Figure 8 , Figure 8 The diagram illustrates the logic mapping of phase plane trajectory and liquid addition feedforward control in an application embodiment of the present invention. The horizontal axis in the diagram represents specific mechanical energy. The vertical axis is defined as the smooth temperature rise rate. The solid lines with arrows shown from left to right in the figure represent the true phase plane evolution trajectories of a typical batch. The figure clearly marks the viscous phase transition initiation point where the tangent slope changes from positive to negative, as well as the corresponding step distribution region of the servo proportional valve opening decay; it also marks the suspected extreme points where the trajectory ends with a concave inflection and triggers the quota verification disturbance, intuitively reflecting the state calculation and feature optimization process of this invention.

[0183] See attached document Figure 9 , Figure 9The graph shows a comparison of the thermodynamic responses of the traditional control method and the control method of this invention at the endpoint identification stage, based on experimental verification data extraction. The graph includes a main axis (temperature response) and a secondary axis (active power), with a finely detailed engineering grid baseline in the background. The contrast between the two solid lines in the graph reveals that the temperature rise curve of the traditional control group exhibits an inertial ramp-up overshoot phenomenon lasting several seconds after the main motor is unloaded, masking the true internal energy change of the material. In contrast, the curve of the implementation group, with its locally magnified observation window, clearly shows the minute difference in temperature rise rate accurately extracted by the smoothing derivative operator within the preset observation window after the timer intervenes. The characteristic peaks demonstrate the accuracy of the asynchronous observation mechanism of this invention in stripping away the thermal hysteresis effect.

[0184] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent control system for mixing PVC shrink film raw materials, characterized in that, include: The data acquisition module is used to acquire the motor's three-phase active power sequence, operating frequency, and raw temperature signal output by the frequency converter; The state calculation module is used to convert the three-phase active power sequence of the motor into accumulated specific mechanical energy and process the original temperature signal to output the temperature rise rate. The feedforward control module is used to calculate the tangent slope based on the specific mechanical energy and the temperature rise rate, and adjust the opening command of the servo proportional valve according to the tangent slope. The disturbance execution module is used to issue a frequency increment command when the temperature rise rate exhibits a minimum value characteristic, and to cancel the frequency increment command when the additional injected specific mechanical energy reaches a quota threshold. The material discharge determination module is used to extract thermodynamic response characteristic quantities after canceling the frequency increment command and delaying by a set time lag, and output a material discharge command or a reset signal based on the deflection angle of the response vector constructed by the quota threshold and the thermodynamic response characteristic quantities. The main control unit is used to maintain the batch control state machine to coordinate the execution priorities among the feedforward control module, the disturbance execution module and the material discharge judgment module, and to lock the servo proportional valve during disturbance verification.

2. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, The data acquisition module is also used to perform no-load calibration in the absence of material feeding: Within a preset observation time window, the three-phase active power sequence of the motor is collected according to the set discrete sampling period, and its average value is calculated as the basic frictional active power constant. The basic triboelectric active power constant is used by the state calculation module to eliminate the mechanical transmission background loss during subsequent batch mixing processes.

3. The intelligent control system for mixing PVC shrink film raw materials according to claim 3, characterized in that, The state calculation module is specifically used for: The net input active power is obtained by subtracting the basic frictional active power constant from the values ​​in the discretely sampled three-phase active power sequence of the motor. The net input active power is then integrated in the time domain and divided by the total mass of the input materials to obtain the accumulated specific mechanical energy.

4. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, The feedforward control module is specifically used for: The slope of the tangent is obtained by calculating the ratio of the difference in the temperature rise rate between the current update point and the previous update point to the difference in the specific mechanical energy. When the slope of the tangent is greater than zero, it is determined that the material is in a period of pure mechanical friction, and the maximum opening command of the reference is output as the opening command of the servo proportional valve. When the tangent slope is less than zero, it is determined that the material has entered the viscous phase transition period. The preset slope proportional gain coefficient is used in combination with the tangent slope to calculate and attenuate the opening command of the servo proportional valve.

5. The intelligent control system for mixing PVC shrink film raw materials according to claim 4, characterized in that, The feedforward control module is also used to record the cumulative liquid addition amount of the servo proportional valve and to impose a total amount constraint on the liquid addition process according to the target addition amount set in the formula.

6. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, The minimum value characteristic is: The tangent slope changes from negative to positive and stably exceeds the set dead zone tolerance; When the disturbance execution module detects the minimum value feature, it marks the current state point as a suspected endpoint and starts an independent integrator to calculate the additional injected specific mechanical energy accumulated after the frequency increment command is issued.

7. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, The thermodynamic response feature extracted by the discharge determination module is the true difference in temperature rise rate. The specific extraction process is as follows: After the set time delay ends, within a preset observation window, the difference between the current temperature rise rate and the temperature rise rate at the time the frequency increment command is cancelled is extracted as the true temperature rise rate difference.

8. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, When the material discharge determination module outputs a material discharge command or a reset signal based on the deflection angle of the response vector constructed from the quota threshold and the thermodynamic response characteristic quantity, it is specifically used for: The vector abscissa is obtained by multiplying the set horizontal axis normalization coefficient by the quota threshold, and the vector ordinate is obtained by multiplying the vertical axis normalization coefficient by the actual difference in the temperature rise rate. The deflection angle is obtained by taking the arctangent function of the ratio of the ordinate to the abscissa of the vector.

9. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, The material discharge determination module compares the calculated deflection angle with a benchmark threshold: If the deflection angle is greater than or equal to the reference threshold, a reset signal is sent to the disturbance execution module to end the verification state, and the lock on the servo proportional valve during the disturbance verification is released. If the deflection angle is less than the reference threshold, the discharge command is output to control the pneumatic discharge valve to open.

10. The intelligent control system for mixing PVC shrink film raw materials according to claim 1, characterized in that, The batch control state machine of the main control unit is specifically used for: Upon receiving the reset signal, the inverter is controlled to restore the reference operating frequency, and the feedforward control module is reactivated to monitor the tangent slope.

Images