Ficus juice production and processing numerical control system based on PLC automatic control technology

Abstract

The application discloses a fig juice production and processing numerical control system based on PLC automatic control technology and relates to the technical field of fig juice processing control. The system comprises a data correlation unit, a steady state determination unit, a correction binding unit and a batch scheduling unit. The data correlation unit is used for matching a standard processing parameter table according to initial sugar content and initial acidity of fig raw materials, and generating a processing parameter mapping table. The steady state determination unit is used for collecting actual output variables of a PLC controller and determining whether a processing stage enters a non-steady state period according to a time sequence deviation. The correction binding unit is used for extracting a dynamic correction coefficient and binding the dynamic correction coefficient to an integral term of a PID control loop after triggering of the non-steady state period. The batch scheduling unit is used for reordering a batch execution sequence of multiple production lines according to a processing time length weight and an actual energy consumption rate, and generating a scheduling instruction. The system can adapt to quality differences of multiple batches of raw materials and dynamically respond to non-steady state disturbances in a processing process.

Description

Technical Field

[0001] This invention relates to the field of fig juice processing control technology, specifically a numerical control system for fig juice production and processing based on PLC automatic control technology. Background Technology

[0002] The production and processing of fig juice requires high precision in controlling process parameters such as temperature, pressure, and flow rate. Traditional methods often rely on manual experience to set fixed processing parameters, which is insufficient to address the natural differences in ripeness, sugar content, and acid content between different batches of raw materials. When the quality of upstream raw materials fluctuates, fixed processing parameters often lead to unstable juice yield and poor consistency in flavor and quality, increasing the frequency of production line adjustments and material losses.

[0003] Existing automation control schemes typically focus on the logic interlocking and basic loop regulation of individual equipment. Effective automation methods are lacking for systematically mapping the biochemical indicators of raw materials into a complete set of processing parameters covering multiple stages such as washing, crushing, enzymatic hydrolysis, enzyme inactivation, and filtration. The conventional approach involves process engineers manually adjusting a few parameters based on sampling results and experience. This discrete correction method cannot form a continuous processing standard indexed by raw material batches. Transitions between different batches often involve long steady-state reconstruction times, and the scheduling of parallel tasks on the production line is independent of real-time material properties and energy consumption status. Task allocation between different processing lines still follows a first-in-first-out (FIFO) or fixed-priority strategy. When multiple batches of raw materials enter the production system simultaneously, it easily leads to some production lines being overloaded while others are idle, resulting in poor overall processing efficiency and energy consumption.

[0004] On the other hand, even when a PID controller is configured for a single device, its integral term, when faced with unsteady processes caused by raw material switching or external disturbances, often suffers from accumulated historical deviations that lead to overshoot or response lag. Conventional integral separation or anti-saturation strategies only perform on / off processing based on the magnitude of the deviation, ignoring the differentiated impact of the duration and specific stage of the unsteady process on the system's dynamic characteristics. For batch processing processes like fig juice processing, which have significant multi-stage and time-varying characteristics, it is necessary to identify the unsteady period and, based on the specific process stage and duration of the disturbance, adaptively adjust the integral term in relation to the stage characteristics to match the dynamic demands of the current operating condition. Batch sequencing for parallel multi-line production also requires incorporating the energy consumption characteristics that change due to unsteady adjustments and the expected processing times of different batches into the scheduling decision to improve the overall coordination of concurrent multi-batch production. Summary of the Invention

[0005] The technical problem to be solved by this invention is how to automatically generate multi-stage processing parameters in fig juice production, taking into account the quality differences of multiple batches of raw materials, and after identifying the non-steady-state periods of each processing stage, to modify the PID integral term according to the stage characteristics and duration of the period, and to dynamically schedule the batch execution sequence of parallel production lines by combining the expected processing time of different batches with the real-time energy consumption rate.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a numerical control system for fig juice production and processing based on PLC automatic control technology, comprising: a data association unit, used to acquire the initial sugar content and initial acidity value of fig raw materials, and to perform association matching on multiple processing parameter intervals in a pre-stored standard processing parameter table based on the initial sugar content and initial acidity value, generating a processing parameter mapping table indexed by raw material batch; and a steady-state determination unit, used to collect the actual output variables of the PLC controller in each processing stage corresponding to the processing parameter mapping table, and to determine the relationship between the actual output variables and the corresponding values ​​in the processing parameter mapping table. The timing deviation between the theoretical output variables of each stage determines whether the current processing stage has entered a non-steady-state period. The correction binding unit is used to read the dynamic correction coefficient of the corresponding stage in the processing parameter mapping table according to the duration and start time of the non-steady-state period after the non-steady-state period is triggered, and bind the dynamic correction coefficient to the integral term of the PID control loop of the current processing stage. The batch scheduling unit is used to reorder the batch execution sequence of multiple parallel processing production lines according to the processing time weight of each raw material batch in the processing parameter mapping table and the actual energy consumption rate of the current processing stage, and generate a scheduling instruction containing priority identifier.

[0007] The data association unit performs a first-level match between the initial sugar content value and the sugar content feature intervals in the standard processing parameter table, filtering out all candidate processing parameter intervals containing the initial sugar content value. Then, it selects candidate intervals from all candidate processing parameter intervals where the initial acidity value falls within an acidity feature sub-interval as the final matching result, and generates the processing parameter mapping table based on the processing parameters corresponding to this final matching result. This two-level sugar-acid matching ensures a high degree of correspondence between processing parameters and the fruit's intrinsic quality, effectively avoiding flavor fluctuations caused by single-batch differences and guaranteeing batch-to-batch consistency in product taste.

[0008] The steady-state determination unit calculates the slope of the actual output variable over multiple consecutive sampling periods and compares this slope with the theoretical slope of the theoretical output variable for the corresponding stage to obtain the difference. When the difference exceeds a preset difference threshold for three consecutive sampling periods, the first sampling period exceeding the preset difference threshold is taken as the start time of the non-steady-state period. This method can sensitively capture the moment when the processing deviates from the set steady state, providing an accurate triggering opportunity for subsequent control correction and avoiding process quality degradation caused by lag response.

[0009] The correction binding unit uses the start time of the non-steady-state period as the search key to query the processing parameter mapping table for the stage boundary time closest to the start time. Based on the stage boundary time, it determines the stage number of the current processing stage. Then, using the stage number and the duration as an index, it extracts the corresponding dynamic correction coefficient from the pre-stored correction coefficient matrix. Preferably, when binding the dynamic correction coefficient to the integral term of the PID control loop of the current processing stage, the dynamic correction coefficient is multiplied by the current cumulative value of the integral term and then rewritten into the integral term. By dynamically adjusting the integral action strength, process disturbances can be quickly compensated without causing overshoot, allowing the system to quickly recover to steady state and making the control output more accurate.

[0010] The batch scheduling unit accumulates the processing time weights of all batches waiting on each processing line to obtain the total weighted load of each processing line. Then, it divides the total weighted load of each processing line by its actual energy consumption rate to obtain the scheduling priority index for each processing line. The batch execution sequences of the multiple processing lines are then reordered according to the scheduling priority index from high to low. This scheduling method dynamically schedules production based on the ratio of processing time to real-time energy consumption, allowing high-load, low-energy-consumption production lines to be executed first, thereby improving the overall production cycle time and increasing equipment utilization.

[0011] As a technical solution of the present invention, the system further includes an interface generation unit, used to receive the scheduling instruction and generate a visual execution progress matrix in the human-machine interface, with batch numbers as rows and processing stages as columns, and each cell in the visual execution progress matrix is ​​bound to the numerical identifier of the dynamic correction coefficient of the corresponding stage. The interface generation unit draws an M-row, N-column table in the first display area of ​​the human-machine interface, where M equals the total number of batches currently in the execution and waiting states, and N equals the total number of processing stages in the standard processing parameter table. The batch number of each batch is filled into the first column of the corresponding row, and the rows are sorted according to the priority identifier in the scheduling instruction. The interface generation unit also displays the numerical identifier of the dynamic correction coefficient of each cell in the visual execution progress matrix as a clickable interactive control. When the interactive control is triggered, the real-time curve of the integral term of the PID control loop of the corresponding processing stage of the cell is displayed in the second display area of ​​the human-machine interface. The interface generation unit performs hierarchical coloring of the background color of the corresponding cell in the visual execution progress matrix according to the absolute value of the numerical identifier of the dynamic correction coefficient, with cells having larger absolute values ​​corresponding to darker background colors. This visualization method allows operators to intuitively monitor the correction intensity and quality control of each batch at each processing stage, and quickly trace process details through interactive operation, promptly detect abnormalities and intervene, and reduce the risk of defective products.

[0012] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0013] Using the initial sugar content of fig raw materials as the first-level matching condition and the initial acidity content as the second-level matching condition, the system filters out processing parameter combinations that simultaneously meet both dimensions from a pre-stored standard processing parameter table, generating a processing parameter mapping table indexed by raw material batch. This two-level correlation matching method directly establishes a systematic mapping between raw material biochemical indicators and multi-stage processing parameters, avoiding the problem of processing specification breaks between different batches caused by traditional manual adjustment of single parameters based on experience. When different batches of raw materials enter the production line, the system can automatically load an appropriate set of processing parameters for them, ensuring a consistent reference benchmark for parameter transitions during batch switching and reducing juice yield fluctuations and flavor deviations caused by parameter settings deviating from the actual quality of the raw materials.

[0014] In each processing stage corresponding to the processing parameter mapping table, the actual output variables of the PLC controller are continuously collected and their change slopes are calculated. The actual slope is compared with the theoretical slope by difference. When the difference exceeds the preset threshold for three consecutive sampling periods, it is determined that a non-steady-state period has been entered. The start time of this period is used as the search key to determine the stage number of the current processing stage from the processing parameter mapping table. Then, the stage number and the duration of the non-steady-state period are used as the index to extract the corresponding dynamic correction coefficient from the correction coefficient matrix. This coefficient is multiplied by the current cumulative value of the integral term of the PID control loop and rewritten into the integral term. This method binds the correction amount of the non-steady-state response to the process stage where the disturbance occurs and the duration of the disturbance, so that the adjustment strength of the integral term is adapted to the tolerance of the deviation in different stages, rather than simply performing a uniform on / off integral processing based on the magnitude of the deviation. Therefore, when disturbances occur in different stages such as enzymatic hydrolysis and enzyme inactivation, the control loop can absorb historical deviations with a correction strength that matches the dynamic characteristics of the stage, reduce the overshoot and steady-state recovery time, and make the actual output variable approach the theoretical output expectation more quickly.

[0015] The total weighted load of each processing line is calculated by accumulating the processing time weights of each raw material batch. Then, the total weighted load is divided by the actual energy consumption rate of that line to obtain a scheduling priority index. The batch execution sequences of parallel production lines are then reordered according to the scheduling priority index from high to low, generating scheduling instructions containing priority identifiers. This sorting mechanism incorporates both the expected processing time of the processing task and the current energy consumption status of the production line into a unified scheduling decision. This ensures that batches with larger processing time weights are preferentially allocated to production lines with lower energy consumption rates, avoiding the situation where high-load production lines are continuously overloaded while low-load production lines are idle when simply allocating tasks based on first-in-first-out priority. This promotes load balancing among multiple production lines and efficient utilization of unit energy consumption. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a schematic diagram of a numerical control system for fig juice production and processing based on PLC automatic control technology.

[0018] Figure 2 This is a flowchart of production line scheduling priority calculation based on weighted load and energy consumption rate;

[0019] Figure 3 This is a schematic diagram illustrating the clustering of initial sugar and acidity values ​​of fig raw materials to define processing parameter ranges.

[0020] Figure 4 It is a scatter plot of the dynamic correction coefficients and durations at different processing stages. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0022] See Figure 1 This invention provides a numerical control system for fig juice production and processing based on PLC automatic control technology, including a data association unit, a steady-state determination unit, a correction binding unit, and a batch scheduling unit. The data association unit acquires the initial sugar and acidity values ​​of the fig raw materials and matches them with multiple processing parameter ranges in a pre-stored standard processing parameter table to generate a processing parameter mapping table indexed by the raw material batch. The steady-state determination unit collects the actual output variables of the PLC controller in each processing stage corresponding to the processing parameter mapping table and determines whether the current processing stage has entered a non-steady-state period based on the timing deviation between the actual output variables and the theoretical output variables of the corresponding stage in the processing parameter mapping table. The correction binding unit, after triggering a non-steady-state period, reads the dynamic correction coefficient of the corresponding stage in the processing parameter mapping table based on the duration and start time of the non-steady-state period and binds the dynamic correction coefficient to the integral term of the PID control loop of the current processing stage. The batch scheduling unit is used to reorder the batch execution sequences of multiple parallel processing production lines based on the processing time weight in the processing parameter mapping table of each raw material batch and the actual energy consumption rate of the current processing stage, and generate scheduling instructions containing priority identifiers.

[0023] Example 1

[0024] In practice, the data association unit obtains the initial sugar and acidity values ​​of the fig raw materials. A standard processing parameter table is pre-stored in the CNC system's storage module. This table contains multiple processing parameter ranges, each associated with a set of processing parameters, and each range is configured with a sugar content characteristic range and an acidity characteristic sub-range. The sugar content characteristic range represents the applicable range of raw material sugar content values, and the acidity characteristic sub-range represents the applicable range of raw material acidity values.

[0025] The data association unit performs a first-level matching between the initial sugar content value and the sugar content feature interval corresponding to each processing parameter interval in the standard processing parameter table. During this first-level matching, the data association unit iterates through all processing parameter intervals. For each interval, it determines whether the initial sugar content value falls within the corresponding sugar content feature interval. If the initial sugar content value does fall within the sugar content feature interval, the corresponding processing parameter interval is marked as a candidate processing parameter interval, thus obtaining a set of candidate processing parameter intervals.

[0026] After obtaining the set of candidate processing parameter intervals, the data association unit selects the final matching result from the set. The selection process is as follows: for each candidate processing parameter interval in the set, the corresponding acidity feature sub-interval is obtained, and it is determined whether the initial acidity value is within the acidity feature sub-interval; the candidate processing parameter interval whose initial acidity value is within the acidity feature sub-interval is determined as the final matching result. Optionally, if multiple candidate processing parameter intervals satisfy the condition that the initial acidity value is within the acidity feature sub-interval, a unique final matching result is determined according to a preset selection rule. The preset selection rule is to select the candidate processing parameter interval with the smallest absolute value of the difference between the center value of the acidity feature sub-interval and the initial acidity value, or to select the candidate processing parameter interval with the smallest storage sequence number.

[0027] After determining the final matching result, the data association unit extracts a set of processing parameters associated with the final matching result. These processing parameters include temperature setpoints, pressure setpoints, speed setpoints, time setpoints, and initial parameters for the proportional-integral-derivative (PID) control loop at each processing stage. Using the raw material batch as an index, the data association unit writes the extracted processing parameters into a processing parameter mapping table. Each record in the processing parameter mapping table corresponds to a raw material batch and stores the processing parameter values ​​required for each processing stage of the raw material batch.

[0028] In some embodiments, the sugar content characteristic intervals and acidity characteristic sub-intervals in the standard processing parameter table are generated by clustering historical batch data. The specific clustering method is as follows: Measured sugar content and acidity values ​​of multiple batches of fig raw materials, along with their corresponding optimal processing parameters, are collected. The raw materials are divided into multiple clusters using the K-means clustering algorithm. The minimum and maximum measured sugar content values ​​within each cluster constitute the sugar content characteristic interval, and the minimum and maximum measured acidity values ​​within each cluster constitute the acidity characteristic sub-interval. The optimal processing parameters corresponding to each cluster are then used as the processing parameters for the processing parameter intervals.

[0029] Optionally, the initial sugar content and initial acidity value are collected in real time by online sugar content detectors and online acidity detectors set at the raw material inlet, and then converted into digital signals by the analog input module of the programmable logic controller and transmitted to the data association unit.

[0030] See Figure 3 The figure illustrates the two-dimensional distribution of initial sugar and initial acid values ​​for multiple processing parameter intervals as described in the standard processing parameter table of Embodiment 1 of this invention, along with the division of processing parameter intervals. The horizontal axis represents the initial sugar value of the fig raw material, ranging from approximately 9 to 22; the vertical axis represents the initial acid value, ranging from approximately 0.5 to 1.5. The figure uses scatter points of different shapes and colors to correspond to four processing parameter intervals: processing parameter interval 1 (blue dots), processing parameter interval 2 (orange squares), processing parameter interval 3 (green triangles), and processing parameter interval 4 (red rhombuses). The scatter points within each processing parameter interval represent the distribution of initial sugar and initial acid values ​​for the corresponding raw material batch in historical batch data.

[0031] Each processing parameter range is marked with a dashed rectangle to indicate its sugar content characteristic range and acidity characteristic sub-range. The sugar content characteristic range of processing parameter range 1 is approximately 9 to 12, and the acidity characteristic sub-range is approximately 0.5 to 0.7; the sugar content characteristic range of processing parameter range 2 is approximately 13 to 15, and the acidity characteristic sub-range is approximately 0.8 to 1.05; the sugar content characteristic range of processing parameter range 3 is approximately 15 to 19, and the acidity characteristic sub-range is approximately 1.15 to 1.45; the sugar content characteristic range of processing parameter range 4 is approximately 19 to 21.5, and the acidity characteristic sub-range is approximately 0.6 to 0.85.

[0032] Example 2

[0033] In practice, the steady-state determination unit periodically reads the actual output variables from the output register of the programmable logic controller (PLC). The processing parameter mapping table records the theoretical output variables for the corresponding stages. These theoretical output variables are pre-set reference output value sequences, which contain the theoretical values ​​corresponding to each sampling period. The sampling period is generated by a timer interrupt of the PLC, and the sampling period duration is set to a fixed value, ranging from 50 milliseconds to 200 milliseconds.

[0034] The steady-state determination unit calculates the slope of the actual output variable over multiple consecutive sampling periods. The slope represents the rate of change of the actual output variable per unit time. In one optional implementation, the slope is calculated using a first-order difference form. For the... The slope of the actual output variable in each sampling period Calculated using the following formula:

[0035]

[0036] in, Indicates the first The slope of the actual output variable change corresponding to each sampling period Indicates the first The actual output variable values ​​collected in each sampling period Indicates the first The actual output variable values ​​collected in each sampling period This indicates the sampling period duration, which is 100 milliseconds.

[0037] The steady-state determination unit extracts the theoretical slope of the theoretical output variable for the corresponding stage from the processing parameter mapping table. Theoretical slope The first one stored in the processing parameter mapping table The theoretical rate of change of the output variable corresponding to each sampling period, and the theoretical slope value, are determined by the stage process curve. The stage process curve consists of multiple linear segments, each corresponding to a constant theoretical slope value. The steady-state determination unit reads the theoretical slope value of the corresponding time period according to the segment index of the stage process curve.

[0038] The steady-state determination unit will change the slope Compared with the theoretical slope Perform the difference to obtain the first... Difference value of each sampling period Difference value Calculated using the following formula:

[0039]

[0040] The steady-state determination unit will use the difference value Differential threshold with preset Compare the results. Preset difference threshold. The preset differential threshold is determined based on the allowable fluctuation range of fig juice processing technology. Set to 20% of the upper limit of the theoretical slope absolute value. If the upper limit of the theoretical slope absolute value is 5.0 units / second, then preset the difference threshold. Set to 1.0 units / second.

[0041] The steady-state determination unit is equipped with a continuous exceedance counter, which is initially set to zero. When the... Difference value of each sampling period Exceeding the preset difference threshold When the steady-state determination unit continuously exceeds the counter value by one, it records the first occurrence. The moment of the sampling period. When the first sampling period... Difference value of each sampling period Not exceeding the preset difference threshold When this happens, the steady-state determination unit will continuously exceed the counter value and reset it to zero.

[0042] The steady-state determination unit detects values ​​that continuously exceed the counter. When the number of consecutive values ​​exceeding the counter reaches three, it determines that the current processing stage has entered a non-steady-state period. The steady-state determination unit extracts the first sampling period from the records of three consecutive sampling periods that exceed the preset difference threshold, takes the first sampling period exceeding the preset difference threshold as the start time of the non-steady-state period, and sends the start time of the non-steady-state period to the correction binding unit.

[0043] In some embodiments, the actual output variables include temperature variables in the enzymatic hydrolysis stage, vacuum variables in the concentration stage, or flow rate variables in the sterilization stage. Different processing stages correspond to the reading of the corresponding actual output variables. The processing parameter mapping table stores the theoretical output value sequence and theoretical slope sequence of the corresponding output variables for each stage. The steady-state determination unit automatically switches the data address to be read according to the stage number of the current processing stage, thereby obtaining the theoretical output variables and theoretical slopes that match the type of actual output variables.

[0044] Example 3

[0045] In practice, after the correction binding unit is triggered during a non-steady-state period, it obtains the start time and duration of the non-steady-state period. The start time of the non-steady-state period is output by the steady-state determination unit, and the duration is obtained by accumulating the timer inside the correction binding unit from the start time of the non-steady-state period. The timer increments once in each sampling period, and the duration is obtained by multiplying the timer value by the sampling period duration.

[0046] The correction binding unit uses the start time of the non-steady-state period as the search key to query the processing parameter mapping table for the stage boundary time closest to the start time. The processing parameter mapping table stores the start and end times of each processing stage; these are collectively referred to as stage boundary times. The query process is as follows: the correction binding unit iterates through the processing parameter mapping table for all processing stages corresponding to the current raw material batch, calculates the absolute value of the time difference between each stage boundary time and the start time of the non-steady-state period, and selects the stage boundary time with the smallest absolute time difference as the query result.

[0047] In some embodiments, when two stage boundary moments have the same absolute time difference from the start moment and both are minimum values, the correction binding unit selects the earlier stage boundary moment as the query result. Optionally, the correction binding unit uses a binary search method to accelerate the query process. The binary search method sorts all stage boundary moments of the current raw material batch in the processing parameter mapping table according to time sequence to construct an ordered list. The correction binding unit performs a binary search on the ordered list to locate the insertion position, compares the absolute time difference between the two stage boundary moments before and after the insertion position and the start moment, and selects the stage boundary moment with the smaller absolute time difference.

[0048] The correction binding unit determines the stage number of the current processing stage based on the stage boundary time. Each processing stage in the processing parameter mapping table corresponds to a unique stage number, which increments sequentially according to the processing stage's time order. There is a one-to-one correspondence between the stage boundary time and the stage number. The correction binding unit identifies which processing stage's start or end time the stage boundary time belongs to and obtains the corresponding stage number as the stage number of the current processing stage.

[0049] The correction binding unit uses both the stage number and duration as indexes to extract the corresponding dynamic correction coefficients from a pre-stored correction coefficient matrix. The pre-stored correction coefficient matrix is ​​a two-dimensional data table. The first dimension index of the correction coefficient matrix is ​​the stage number, and the second dimension index is the duration interval number. The duration interval number is determined by the duration range into which the duration falls. The correction binding unit compares the duration with multiple predefined duration intervals one by one to determine the duration interval to which the duration belongs, thus obtaining the duration interval number. Each element in the correction coefficient matrix stores a dynamic correction coefficient value. The correction binding unit reads the corresponding dynamic correction coefficient from the correction coefficient matrix using the stage number as the row index and the duration interval number as the column index.

[0050] In some embodiments, the correction coefficient matrix is ​​generated through offline experimental calibration. The offline experimental calibration process includes: applying integral term correction coefficients of different sizes to each processing stage under different duration intervals, measuring the quality indicators of the finished juice, and selecting the integral term correction coefficient that makes the quality indicators reach the target range as the dynamic correction coefficient under the corresponding stage number and duration interval number. The quality indicators include sugar retention rate and acidity retention rate.

[0051] The correction binding unit binds the dynamic correction coefficient to the integral term of the proportional-integral-derivative (PID) control loop in the current processing stage. The integral term is stored in the programmable logic controller's data register and contains the current accumulated value. The binding process is as follows: the correction binding unit reads the current accumulated value of the integral term, multiplies the dynamic correction coefficient by the current accumulated value to obtain the corrected integral accumulated value. The correction calculation is expressed by the following formula:

[0052]

[0053] in, This represents the corrected cumulative integral value. Indicates the dynamic correction factor. The value range is from 0.5 to 1.5. This represents the current accumulated value of the integral term.

[0054] The corrected binding unit will adjust the accumulated points value. The integral term data register of the proportional-integral-derivative (PID) control loop is written, overwriting the current accumulated value of the original integral term, thus completing the rewrite operation of the integral term. In the next control cycle, the PID control loop uses the corrected accumulated integral value for output calculation, thereby adjusting the control output.

[0055] Optionally, the method for determining the value of the dynamic correction coefficient also considers the direction of deviation between the actual output variable and the theoretical output variable during the non-steady-state period. When the actual output variable is consistently higher than the theoretical output variable, the dynamic correction coefficient is multiplied by a decay factor based on the value read from the correction coefficient matrix, and the decay factor value is 0.9. When the actual output variable is consistently lower than the theoretical output variable, the dynamic correction coefficient is multiplied by an enhancement factor based on the value read from the correction coefficient matrix, and the enhancement factor value is 1.1.

[0056] See Figure 4 In the graph, the vertical axis represents the dynamic correction coefficient, ranging from approximately 0.6 to 1.4, indicating the magnitude of the integral term correction coefficient calculated after the correction binding unit is triggered during the non-steady-state period. The horizontal axis represents the duration in seconds, ranging from 0 to 60 seconds, indicating the length of the non-steady-state period. The graph uses scatter dots of different shapes and colors to represent the five processing stages in fig juice production: enzymatic hydrolysis (blue dots), filtration (orange squares), concentration (green triangles), blending (red diamonds), and sterilization (purple plus signs), reflecting the distribution of the dynamic correction coefficient for each stage as the duration of the non-steady-state period changes.

[0057] As shown in the figure, the dynamic correction coefficient in the enzymatic hydrolysis stage generally decreases with increasing duration. Initially, the coefficient is approximately 1.0 to 1.1, but drops to about 0.8 as the duration reaches 60 seconds, indicating a weakening of the integral term correction during prolonged non-steady-state periods. The dynamic correction coefficient in the filtration stage shows an increasing trend, starting at about 1.0 and gradually increasing with longer duration, reaching a maximum of about 1.35, indicating a significant strengthening effect on the integral term correction as the non-steady-state period lengthens. The dynamic correction coefficient in the concentration stage changes relatively smoothly with a slight decreasing trend, fluctuating between 0.8 and 1.05, indicating relatively stable integral term correction during this stage. The dynamic correction coefficient in the blending stage generally remains between 1.05 and 1.2, with a slight increasing trend, indicating a moderate strengthening of the integral term correction during the non-steady-state duration of the blending stage. The dynamic correction coefficient for the sterilization stage is generally low and shows a decreasing trend as the duration increases, reaching a minimum of close to 0.6. This reflects that the correction strength of the integral term in this stage is relatively small and further weakens as the non-steady-state time increases.

[0058] The scatter plots of dynamic correction coefficients for different processing stages show a clear distinction in the vertical direction, indicating that the correction binding unit adopted different dynamic correction strategies for the duration of unsteady-state conditions in different processing stages, and that the numerical range and trend of the correction coefficients match the process characteristics of each stage. The distribution of duration on the horizontal axis is relatively uniform, indicating that the start and duration of the unsteady-state period are covered within the range of 0 to 60 seconds, reflecting the impact of different durations on the dynamic correction coefficients.

[0059] Example 4

[0060] In specific implementation, please refer to Figure 2 The batch scheduling unit obtains the status information of multiple parallel-running processing lines from the data storage area of ​​the programmable logic controller (PLC). The status information includes a list of raw material batches in a waiting state on each processing line. The waiting state indicates that the raw material batch has completed the previous processing stage but has not yet entered the current processing stage. The waiting state is determined by the PLC based on the production line sensor signals and the processing stage completion flag.

[0061] For each processing line, the batch scheduling unit iterates through all raw material batches in a waiting state. For each batch encountered, the batch scheduling unit reads the corresponding processing time weight from the processing parameter mapping table. The processing time weight is a numerical field in the processing parameter mapping table, representing the relative processing time required for the raw material batch in the current processing stage. The processing time weight is calculated by the data association unit when generating the processing parameter mapping table, based on the processing stage duration benchmark in the standard processing parameter table and the deviation of the initial sugar content and initial acidity value of the raw material. The value of the processing time weight ranges from 0.8 to 1.2, and the specific value is stored in the data row corresponding to the raw material batch index in the processing parameter mapping table.

[0062] The batch scheduling unit accumulates the processing time weights of all raw material batches waiting on each processing line to obtain the total weighted load for each processing line. The accumulation process is as follows: For the q-th processing line, initialize the total weighted load. If the value is zero, sequentially read the processing time weight of each waiting batch of raw materials on the q-th processing line. ,implement The total weighted load of the q-th processing line is obtained after all waiting batches of raw materials for that processing line have been accumulated. . This represents the total weighted load of the q-th processing line. This indicates the weight of processing time for a single batch of raw materials.

[0063] The batch scheduling unit obtains the actual energy consumption rate of each processing line at the current processing stage. The actual energy consumption rate is measured in real time by a power transmitter installed on the power circuit of each processing line. The analog signal output by the power transmitter is converted into a digital signal by the analog input module of the programmable logic controller (PLC). The digital signal represents the instantaneous power value, and the actual energy consumption rate is obtained by accumulating these instantaneous power values ​​over a unit of time. The actual energy consumption rate represents the energy consumption per unit of time, expressed in kilowatt-hours per hour. The batch scheduling unit obtains the actual energy consumption rate of each processing line by reading the energy consumption rate data register of the corresponding line in the PLC. , This represents the actual energy consumption rate of the q-th processing line.

[0064] The batch scheduling unit divides the total weighted load of each processing line by the actual energy consumption rate of that processing line to obtain the scheduling priority index for each processing line. The formula for calculating the scheduling priority index is:

[0065]

[0066] in, This represents the scheduling priority index of the q-th processing line. This represents the total weighted load of the q-th processing line. This represents the actual energy consumption rate of the q-th processing line. Scheduling priority index. The larger the value, the heavier the processing load on the production line under the same energy consumption, and the more priority it needs to be scheduled.

[0067] After obtaining the scheduling priority indices of all processing lines, the batch scheduling unit reorders the batch execution sequences of multiple processing lines according to their scheduling priority indices from high to low. The reordering process is as follows: Each processing line and its corresponding scheduling priority index... Create a record, and sort all the records by The values ​​are sorted in descending order, and the resulting order is the scheduling sequence. The batch scheduling unit generates a corresponding priority identifier for each processing line based on the sorted order. The priority identifier is an integer, and the integer value is equal to the sorted order. The processing line with the first priority has a priority identifier of 1, the second priority has a priority identifier of 2, and so on.

[0068] The batch scheduling unit sends scheduling instructions containing priority identifiers to the production line control bus. The production line control bus schedules the allocation order of shared equipment and the response order of material conveying devices according to the priority identifiers. The scheduling instructions also include the batch number of the first raw material batch in the batch execution sequence of each processing production line, so that the actuator will first perform processing operations on the corresponding raw material batch on the processing production line with priority identifier 1 at the beginning of the next scheduling cycle.

[0069] In some embodiments, when two or more processing lines have the same scheduling priority index, the batch scheduling unit compares the total weight load of the processing lines with the same index, and the processing line with the larger total weight load is assigned a higher priority identifier. When the total weight load is also equal, the batch scheduling unit assigns priority identifiers according to the physical number of the processing lines in ascending order.

[0070] Example 5

[0071] In practical implementation, the interface generation unit receives scheduling instructions from the batch scheduling unit. These instructions contain priority identifiers for multiple processing lines and the batch execution sequence for each processing line. The interface generation unit also reads the total number of batches currently in execution and waiting states from the programmable logic controller's data storage area. The total number of batches is represented by a symbol... express. The value is equal to the sum of the number of batches in the executing state and the number of batches in the waiting state on all processing production lines. The batches in the executing state refer to the raw material batches that are being processed in a certain stage on the processing production line.

[0072] The interface generation unit obtains the total number of processing stages from the standard processing parameter table. The total number of processing stages is represented by a symbol. express. The value represents the number of processing stages predefined in the standard processing parameter table. This table divides the fig juice production process into enzymatic hydrolysis, filtration, concentration, blending, and sterilization stages. The value is 5.

[0073] The interface generation unit draws in the first display area of ​​the human-computer interaction interface. OK The table is drawn as follows: the interface generation unit calls the table control from the graphical user interface control library and sets the row property value of the table control to... The column attribute value is Table controls are generated automatically. OK The column is a grid of cells. The first display area is located on the left side of the human-computer interaction interface. The width of the first display area accounts for 60% of the total width of the human-computer interaction interface, and the height of the first display area accounts for 80% of the total height of the human-computer interaction interface.

[0074] The interface generation unit fills the batch number of each batch into the first column of the corresponding row. The filling process is as follows: the interface generation unit extracts all batch numbers from the batch execution sequence from the scheduling instructions, and writes each batch number into the first column cell of the corresponding row according to the string order of the batch numbers or a preset batch number sorting rule. The first column cell of each row displays the batch number as the row identifier.

[0075] The interface generation unit sorts the rows according to the priority identifiers in the scheduling instructions. The sorting process is as follows: The interface generation unit reads the priority identifier of each processing line from the scheduling instructions. The priority identifier is an integer value; the smaller the priority identifier, the higher the scheduling priority. The interface generation unit groups batches belonging to the same processing line together, arranging each group in ascending order of the processing line's priority identifier. Rows corresponding to processing lines with smaller priority identifier values ​​are placed at the top of the table, and rows corresponding to processing lines with larger priority identifier values ​​are placed at the bottom of the table. Within the same processing line, batches are arranged according to the order in the batch execution sequence.

[0076] After filling and sorting, the interface generation unit generates a visual execution progress matrix with batch numbers as rows and processing stages as columns. The first column of the visual execution progress matrix stores the batch number, and the second column stores the batch number, and so on. The columns correspond to the first to the second processing stages in the standard processing parameter table, respectively. Each processing stage is represented by a batch number at the beginning of each row and column, which identifies a raw material batch. The name of the processing stage is displayed at the beginning of each column.

[0077] After generating the visual execution progress matrix, the interface generation unit binds a numerical identifier of the dynamic correction coefficient corresponding to the relevant stage to each cell in the matrix. The binding process is as follows: the interface generation unit iterates through each cell in the visual execution progress matrix except for the first column, and extracts the dynamic correction coefficient value corresponding to that batch and processing stage from the correction binding unit or storage module based on the batch number corresponding to the cell's row and the processing stage corresponding to its column. The interface generation unit converts the extracted dynamic correction coefficient value into a string format numerical identifier and writes the numerical identifier into the cell as the cell's display content.

[0078] In one optional implementation, the interface generation unit obtains the dynamic correction coefficient by querying a mapping table. The mapping table uses the batch number and stage number as a joint index to store the dynamic correction coefficients applied to each batch in each processing stage. If a batch has not yet triggered a non-steady-state period in a certain processing stage, the dynamic correction coefficient value corresponding to that batch in that stage is displayed as the default value, which is "1.00".

[0079] The interface generation unit displays the dynamic correction coefficient values ​​of each cell in the visualized execution progress matrix as clickable interactive controls. The interactive controls are implemented as follows: the interface generation unit sets each cell containing a numerical identifier as a button control, the button control's surface text is equal to the numerical identifier string, and the button control's click event is bound to a callback function. The callback function receives the batch number of the cell's row and the stage number of its column as parameters.

[0080] When an interactive control is triggered—that is, when an operator clicks a button control in a cell on the human-machine interface (HMI)—the interface generation unit retrieves the batch number and stage number corresponding to the triggered cell in the callback function. Based on the batch number and stage number, the interface generation unit reads the real-time value of the integral term of the proportional-integral-derivative (PID) control loop corresponding to the processing stage from the real-time data area of ​​the programmable logic controller (PLC). The real-time value of the integral term is a dynamic value that changes over time. The interface generation unit plots the real-time curve of the integral term in the second display area of ​​the HMI, with time as the horizontal axis and the real-time value of the integral term as the vertical axis. The second display area is located on the right side of the HMI, with its width accounting for 35% of the total width and its height accounting for 40% of the total height.

[0081] When plotting the real-time curve of the integral term, the interface generation unit labels the batch number and processing stage name above the second display area, marks the time range below the horizontal axis of the curve (sixty seconds prior to the current moment), and labels the integral term value scale to the left of the vertical axis. The real-time curve of the integral term is updated in a scrolling manner, adding new integral term value data points to the end of the curve every sampling period, while removing old data points that are outside the time range at the beginning of the curve.

[0082] When the interface generation unit applies graded coloring to the background of cells, it obtains the absolute value of the dynamic correction coefficient identifier for each cell in the visual execution progress matrix. The dynamic correction coefficient identifier is in string format; the interface generation unit parses the string into a floating-point number and then takes the absolute value, displaying the absolute value using the sign... express, This represents the absolute value of the dynamic correction factor for a single cell.

[0083] The interface generation unit is based on absolute value The size determines the background color depth of the corresponding cell. The hierarchical coloring rule uses a linear mapping algorithm. In one implementation, the hierarchical coloring mapping relationship is expressed by the following formula:

[0084]

[0085] in, This represents the background color depth value assigned to the cell. The background color depth value ranges from 0 to 255, with larger values ​​resulting in a darker color. Indicates the lower limit of the background color depth value. The value is 30, which represents the depth value corresponding to the lightest color. Indicates the upper limit of the background color depth value. The value is 240, which represents the depth value corresponding to the deepest color. This represents the absolute value of the dynamic correction factor corresponding to the cell. This represents the minimum absolute value of the dynamic correction coefficient across all cells. It is calculated by traversing all cells in the interface generation unit. This represents the maximum absolute value of the dynamic correction coefficient across all cells. It is calculated by traversing all cells in the interface generation unit.

[0086] In some embodiments, when equal hour, The calculated value equals That is, 30, which corresponds to a lighter background color. When equal hour, The calculated value equals That is, 240, which corresponds to a darker background color. When Between and In between, The calculated values ​​are determined by linear interpolation. The interface generation unit will use the calculated values... The numerical value is a combination of the blue and green channels in the red-green-blue color model, while the red channel is fixed at 220 minus... The numerical value generates a background color gradient from light yellow to dark orange.

[0087] In some embodiments, the interface generation unit recalculates all cells every preset refresh cycle. The system updates the numerical value and the background color of the corresponding cell, with a preset refresh cycle of 1 second. When the correction binding unit writes a new dynamic correction coefficient for a certain processing stage, the interface generation unit automatically updates the numerical label display and background color of the corresponding cell in the next refresh cycle.

[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A numerical control system for fig juice production and processing based on PLC automatic control technology, characterized in that, The system includes: The data association unit is used to obtain the initial sugar content and initial acidity value of fig raw materials, and to perform association matching on multiple processing parameter ranges in the pre-stored standard processing parameter table based on the initial sugar content and initial acidity value, so as to generate a processing parameter mapping table indexed by raw material batch; The steady-state determination unit is used to collect the actual output variables of the PLC controller in each processing stage corresponding to the processing parameter mapping table, and determine whether the current processing stage has entered a non-steady-state period based on the timing deviation between the actual output variables and the theoretical output variables of the corresponding stage in the processing parameter mapping table. The correction binding unit is used to read the dynamic correction coefficient of the corresponding stage in the processing parameter mapping table according to the duration and start time of the non-steady-state period after the non-steady-state period is triggered, and bind the dynamic correction coefficient to the integral term of the PID control loop of the current processing stage; The batch scheduling unit is used to reorder the batch execution sequences of multiple parallel processing production lines based on the processing time weight in the processing parameter mapping table of each raw material batch and the actual energy consumption rate of the current processing stage, and generate scheduling instructions containing priority identifiers.

2. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 1, characterized in that, The data association unit performs association and matching on multiple processing parameter ranges in the pre-stored standard processing parameter table based on the initial sugar content value and the initial acidity value, specifically as follows: The initial sugar content value is matched with the sugar content feature interval in the standard processing parameter table at the first level to filter out all candidate processing parameter intervals containing the initial sugar content value. Then, the candidate interval in which the initial acidity value is located within the acidity feature sub-interval is selected from all candidate processing parameter intervals as the final matching result, and the processing parameter mapping table is generated with the processing parameters corresponding to the final matching result.

3. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 1, characterized in that, The steady-state determination unit determines whether the current processing stage has entered a non-steady-state period based on the time-series deviation between the actual output variable and the theoretical output variable of the corresponding stage in the processing parameter mapping table. Specifically: Calculate the slope of the actual output variable change within multiple consecutive sampling periods, and differentiate the slope of the change with the theoretical slope of the theoretical output variable for the corresponding stage in the processing parameter mapping table. When the difference exceeds a preset difference threshold within three consecutive sampling periods, the first sampling period exceeding the preset difference threshold is taken as the start time of the non-steady-state period.

4. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 1, characterized in that, After the non-steady-state period is triggered, the correction binding unit reads the dynamic correction coefficient of the corresponding stage in the processing parameter mapping table according to the duration and start time of the non-steady-state period, specifically: Using the start time of the non-steady-state period as the search key, the process parameter mapping table is used to query the stage boundary time that is closest to the start time. The stage number of the current processing stage is determined based on the stage boundary time. Then, the corresponding dynamic correction coefficient is extracted from the pre-stored correction coefficient matrix using the stage number and the duration as the index.

5. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 1, characterized in that, The batch scheduling unit reorders the batch execution sequences of multiple parallel processing lines based on the processing time weights in the processing parameter mapping table for each raw material batch and the actual energy consumption rate of the current processing stage. Specifically: The processing time weights of all batches in the waiting state on each processing line are summed to obtain the total weighted load of each processing line. Then, the total weighted load of each processing line is divided by the actual energy consumption rate of that processing line to obtain the scheduling priority index of each processing line. The batch execution sequences of the multiple processing lines are reordered in descending order of the scheduling priority index.

6. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 1, characterized in that, The system also includes: The interface generation unit is used to receive the scheduling instruction and generate a visual execution progress matrix in the human-computer interaction interface with batch number as the row and processing stage as the column, and each cell in the visual execution progress matrix is ​​bound to the numerical identifier of the dynamic correction coefficient of the corresponding stage.

7. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 6, characterized in that, The interface generation unit receives the scheduling instruction and generates a visual execution progress matrix in the human-computer interaction interface, with batch number as the row and processing stage as the column, specifically: A table with M rows and N columns is drawn in the first display area of ​​the human-computer interaction interface, where M equals the total number of batches currently in the execution and waiting states, and N equals the total number of processing stages in the standard processing parameter table. The batch number of each batch is filled into the first column of the corresponding row, and the rows are sorted according to the priority identifier in the scheduling instruction.

8. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 7, characterized in that, The interface generation unit also displays the numerical identifier of the dynamic correction coefficient of each cell in the visualized execution progress matrix as a clickable interactive control. When the interactive control is triggered, the real-time curve of the integral term of the PID control loop corresponding to the processing stage of the cell is displayed in the second display area of ​​the human-computer interaction interface.

9. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 8, characterized in that, The interface generation unit assigns graded background colors to the corresponding cells in the visualized execution progress matrix based on the absolute value of the dynamic correction coefficient. Cells with larger absolute values ​​correspond to darker background colors.

10. The CNC system for fig juice production and processing based on PLC automatic control technology according to claim 4, characterized in that, When the correction binding unit binds the dynamic correction coefficient to the integral term of the PID control loop in the current processing stage, it multiplies the dynamic correction coefficient by the current cumulative value of the integral term and rewrites it into the integral term.

Images