Control method and system of continuous mixer model pre-control algorithm

Abstract

The present invention provides a control method and system for a continuous mixer model pre-control algorithm, which learns the equipment operating condition information in advance, obtains the variables of near-infrared numerical filtering, and collects the data of time, servo motor speed and near-infrared filtering value under different mixing degrees, and feeds the information back to the processor. The controller establishes a time-mixing uniformity data model and a time-speed data model according to the learned operating condition information, and establishes a digital model of time-mixing uniformity-speed after processing the two models. The controller calculates the speed based on the established digital model and the mixing uniformity measured by near-infrared and transmits it to the servo driver to change the motor running speed, thereby achieving the purpose of effective control. The beneficial effects of the present invention are to solve the hysteresis, delay and volatility of the PID algorithm in the prior art. The accuracy, stability and reliability of the existing equipment are greatly improved. The service life of the equipment is guaranteed and the power consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the field of continuous mixer control systems, and more particularly to a control method and system for a continuous mixer model pre-control algorithm. Background Art

[0002] Continuous mixers are crucial equipment for uniformly mixing drug particles in solid dosage forms. To ensure stable operation and precise control of continuous mixers, a control system is required to monitor and control various mixer parameters. However, existing continuous mixer control systems often suffer from technical issues such as low control accuracy and poor stability. Summary of the Invention

[0003] The present invention overcomes the deficiencies in the prior art and provides a control method and system for a continuous mixer model pre-control algorithm.

[0004] The purpose of the present invention is achieved through the following technical solutions.

[0005] The control method of the continuous mixer model pre-control algorithm comprises the following steps:

[0006] S1, near-infrared numerical filtering,

[0007] Collect multiple sets of real-time data of near-infrared detection and mixing uniformity readings within a fixed time to form an array. After sorting the data in the array in a bubble manner, obtain the median value of the sorted data, write the median value into the average filter algorithm array as the average filter array value, and calculate the average result of the average filter array value as the near-infrared filter value and write it into the filter result variable;

[0008] S2, learning device data,

[0009] Set the initial mixing uniformity, operate the device to obtain the servo motor speed when the equipment has the highest mixing uniformity, gradually increase the servo motor speed according to a fixed time until the servo motor speed at the highest mixing uniformity is reached, record the near-infrared filter value, motor speed value and corresponding time value of each increase, write the data value into the corresponding data array, and obtain the average operation curve model data;

[0010] S3, restore the data model, plot the data obtained in S2, obtain a time-mixing uniformity data model and a time-velocity data model, restore and smooth the time-mixing uniformity data model and the time-velocity data model to obtain a smooth data curve, and overlap the two curve models with time as zero point to obtain a time-mixing uniformity-velocity model;

[0011] S4. Control the operation of the equipment according to the time-mixing uniformity-speed model in S3.

[0012] The specific steps of S1 include:

[0013] S11, data acquisition, reads the real-time data of mixing uniformity through near-infrared detection, collects data N times within a fixed time, and uploads the collected raw data to the array of the processor;

[0014] S12, debounce sorting, compare two adjacent data using the bubble method with the collected N results. If the condition that the value of the previous data is greater than the value of the next data is met, the positions of the previous and next data are swapped. If the condition is not met, the swap is not performed until the overall sorting of the data array is finally completed and the result is written into the array;

[0015] S13, de-jittering and obtaining the median value, dividing the array data after bubble sort by 2 N times or N+1 times to find the median value, and writing the median value into the average filter algorithm array;

[0016] S14, average filtering, after filling the values ​​of the average filter array, add the array values ​​to obtain the total value, divide the total value by the number of average arrays, calculate the average result of the average filter array, and write the result into the filter result variable.

[0017] The specific steps of S13 are: determine whether the value of N is odd or even; if N is odd, then S=(N+1) / 2, and the Sth data in the array is written into the average filter algorithm array as the intermediate value; if N is even, then S=N / 2, and the average of the Sth and S+1th data in the array is written into the average filter algorithm array as the intermediate value.

[0018] The specific steps of S2 include:

[0019] S21, specifying the mixing uniformity and setting the initial setting value of the mixing uniformity;

[0020] S22, obtaining the motor speed when the device has the highest mixing uniformity, gradually increasing the servo motor speed according to the time interval, and recording the near-infrared filter value, motor speed value, and corresponding time value obtained in S14 for each increase;

[0021] S23, repeat S22 multiple times to obtain data, and record the data values ​​obtained N times;

[0022] S24. Calculate the average operation curve. Arrange the near-infrared filter numerical results in the N times of data obtained in S23 according to the time interval column, remove the maximum and minimum values, divide the total value of the remaining data values ​​by 2 / N-2 to obtain the result and store it in the average filter algorithm array. Write the near-infrared filter value into the near-infrared array, write the motor speed value into the speed array, and write the time value into the time array to obtain the complete average operation curve model data.

[0023] The specific steps of S3 include:

[0024] S31, restore the two-dimensional data points,

[0025] According to the average filter curve data, the X-axis direction is used as the time axis and the Y-axis direction is used as the mixing uniformity data axis to obtain the time-mixing uniformity data model;

[0026] According to the average filter curve data, the X-axis direction is used as the time axis and the Y-axis direction is used as the speed data axis to obtain the time-speed data model;

[0027] S32, restoring the basic model, directly connecting the two data points obtained in S31 to obtain the original non-smooth data curve;

[0028] S33, processing the basic model to obtain a smoothing model, performing smoothing filling processing on the original data curve data using a quintic polynomial curve to obtain a smooth data curve;

[0029] S34. Overlap the two curve models obtained in S33 based on the time zero point to obtain a time-mixing uniformity-velocity model, and obtain two smooth model curves centered on the time axis.

[0030] The specific steps of S4 include:

[0031] S41, pre-operating the device, first operating the device at a set fixed speed value until a near-infrared value appears;

[0032] S42, pre-modulating the operating speed, comparing the near-infrared data value with the time-mixing uniformity-speed model curve in S3 to find the position in the time-mixing uniformity-speed model corresponding to the current data, adjusting the motor speed according to the speed value at that position, and approximating the corresponding speed in the model by sending instructions to the servo driver;

[0033] S43, coefficient control adjustment time, according to the value collected within a fixed time, compare the data of the time-mixing uniformity-velocity model to obtain the corresponding time adjustment coefficient k, and calculate the corresponding time position after adjustment by adding the time adjustment coefficient k to the current time of the time-mixing uniformity-velocity model position;

[0034] S44, fine adjustment, compare the near-infrared value in real time with the data in the time-mixing uniformity-speed model according to the adjusted corresponding time position obtained in S43, traverse the time-mixing uniformity-speed model according to the value difference result to obtain the speed value at the next moment corresponding to the value, and output the value to the servo according to the corresponding speed value at the next moment during pre-control.

[0035] The control system of the continuous mixer model pre-control algorithm includes:

[0036] A near-infrared detection module is used to detect mixing uniformity data and feed the mixing uniformity data back to the mixing uniformity model pre-control algorithm module;

[0037] A mixing uniformity model pre-control algorithm module is used to control the near-infrared detection module and the servo drive module through a digital model to control the operation of the continuous mixer device;

[0038] Servo drive module for rotating the mixing screw.

[0039] The beneficial effects of the present invention are:

[0040] This solution pre-learns equipment operating condition information and feeds it back to the processor. The controller automatically builds a digital model based on this learned condition information. Based on this digital model and the mixing uniformity measured by near-infrared sensors, the controller automatically calculates the rotational speed and transmits this information to the servo driver to adjust the motor's operating speed, achieving effective control. This effectively addresses the hysteresis, delay, and volatility inherent in existing PID algorithms, significantly improving the accuracy, stability, and reliability of existing equipment. This ensures longer equipment life and reduces energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the working flow chart of the control system of the continuous mixer model pre-control algorithm. DETAILED DESCRIPTION

[0042] The technical solution of the present invention is further described below through specific embodiments. Example

[0043] The control method of the continuous mixer model pre-control algorithm comprises the following steps:

[0044] S1, near-infrared numerical filtering,

[0045] Collect multiple sets of real-time data of near-infrared detection and mixing uniformity readings within a fixed time to form an array. After sorting the data in the array in a bubble manner, obtain the median value of the sorted data, write the median value into the average filter algorithm array as the average filter array value, and calculate the average result of the average filter array value as the near-infrared filter value and write it into the filter result variable;

[0046] S2, learning device data,

[0047] Set the initial mixing uniformity, operate the device to obtain the servo motor speed when the equipment has the highest mixing uniformity, gradually increase the servo motor speed according to a fixed time until the servo motor speed at the highest mixing uniformity is reached, record the near-infrared filter value, motor speed value and corresponding time value of each increase, write the data value into the corresponding data array, and obtain the average operation curve model data;

[0048] S3, restore the data model, plot the data obtained in S2, obtain a time-mixing uniformity data model and a time-velocity data model, restore and smooth the time-mixing uniformity data model and the time-velocity data model to obtain a smooth data curve, and overlap the two curve models with time as zero point to obtain a time-mixing uniformity-velocity model;

[0049] S4. Control the operation of the equipment according to the time-mixing uniformity-speed model in S3.

[0050] The specific steps of S1 include:

[0051] S11, data acquisition, reads the real-time data of mixing uniformity through near-infrared detection, collects data N times within a fixed time, and uploads the collected raw data to the array of the processor;

[0052] S12, debounce sorting, compare two adjacent data using the bubble method with the collected N results. If the condition that the value of the previous data is greater than the value of the next data is met, the positions of the previous and next data are swapped. If the condition is not met, the swap is not performed until the overall sorting of the data array is finally completed and the result is written into the array;

[0053] In S12, there are n numbers in an array that need to be sorted from smallest to largest. The specific steps are as follows: In the first round of sorting, starting from the left, compare two adjacent numbers. If they are in reverse order, swap their positions until the Nth number is compared. The first number is compared with the second, the second with the third, and so on until the N-1th number is compared with the Nth number, achieving ascending sorting.

[0054] S13, de-jittering and obtaining the median value, dividing the array data after bubble sort by 2 N times or N+1 times to find the median value, and writing the median value into the average filter algorithm array;

[0055] S14, average filtering, after filling the values ​​of the average filter array, add the array values ​​to obtain the total value, divide the total value by the number of average arrays, calculate the average result of the average filter array, and write the result into the filter result variable.

[0056] The specific steps of S13 are: determine whether the value of N is odd or even; if N is odd, then S=(N+1) / 2, and the Sth data in the array is written into the average filter algorithm array as the intermediate value; if N is even, then S=N / 2, and the average of the Sth and S+1th data in the array is written into the average filter algorithm array as the intermediate value.

[0057] The specific steps of S2 include:

[0058] S21, specifying the mixing uniformity and setting the initial setting value of the mixing uniformity;

[0059] S22, obtaining the motor speed when the device has the highest mixing uniformity, gradually increasing the servo motor speed according to the time interval, and recording the near-infrared filter value, motor speed value, and corresponding time value obtained in S14 for each increase;

[0060] S23, repeat S22 multiple times to obtain data, and record the data values ​​obtained N times;

[0061] S24. Calculate the average operation curve. Arrange the near-infrared filter numerical results in the N times of data obtained in S23 according to the time interval column, remove the maximum and minimum values, divide the total value of the remaining data values ​​by 2 / N-2 to obtain the result and store it in the average filter algorithm array. Write the near-infrared filter value into the near-infrared array, write the motor speed value into the speed array, and write the time value into the time array to obtain the complete average operation curve model data.

[0062] The specific steps of S3 include:

[0063] S31, restore the two-dimensional data points,

[0064] According to the average filter curve data, the X-axis direction is used as the time axis and the Y-axis direction is used as the mixing uniformity data axis to obtain the time-mixing uniformity data model;

[0065] According to the average filter curve data, the X-axis direction is used as the time axis and the Y-axis direction is used as the speed data axis to obtain the time-speed data model;

[0066] S32, restoring the basic model, directly connecting the two data points obtained in S31 to obtain the original non-smooth data curve;

[0067] S33, processing the basic model to obtain a smoothing model, performing smoothing filling processing on the original data curve data using a quintic polynomial curve to obtain a smooth data curve;

[0068] S34. Overlap the two curve models obtained in S33 based on the time zero point to obtain a time-mixing uniformity-velocity model, and obtain two smooth model curves centered on the time axis.

[0069] The specific steps of S4 include:

[0070] S41, pre-operating the device, first operating the device at a set fixed speed value until a near-infrared value appears;

[0071] S42, pre-modulating the operating speed, comparing the near-infrared data value with the time-mixing uniformity-speed model curve in S3 to find the position in the time-mixing uniformity-speed model corresponding to the current data, adjusting the motor speed according to the speed value at that position, and approximating the corresponding speed in the model by sending instructions to the servo driver;

[0072] S43, coefficient control adjustment time, according to the value collected within a fixed time, compare the data of the time-mixing uniformity-velocity model to obtain the corresponding time adjustment coefficient k, and calculate the corresponding time position after adjustment by adding the time adjustment coefficient k to the current time of the time-mixing uniformity-velocity model position;

[0073] S44, fine adjustment, compare the near-infrared value in real time with the data in the time-mixing uniformity-speed model according to the adjusted corresponding time position obtained in S43, traverse the time-mixing uniformity-speed model according to the value difference result to obtain the speed value at the next moment corresponding to the value, and output the value to the servo according to the corresponding speed value at the next moment during pre-control.

[0074] The time difference between the motor's acceleration and deceleration processes creates a lag when adjusting the speed at that position. Therefore, the speed at the corresponding time position after adjustment needs to be calculated in advance by adding the current time value of the time-mixing uniformity-speed model to the adjustment coefficient k. This eliminates the effects of the acceleration and deceleration processes. The adjustment coefficient k can be a preset value and can be adjusted in practice based on the servo motor model.

[0075] The control system of the continuous mixer model pre-control algorithm includes:

[0076] A near-infrared detection module is used to detect mixing uniformity data and feed the mixing uniformity data back to the mixing uniformity model pre-control algorithm module;

[0077] A mixing uniformity model pre-control algorithm module is used to control the near-infrared detection module and the servo drive module through a digital model to control the operation of the continuous mixer device;

[0078] Servo drive module for rotating the mixing screw.

[0079] like Figure 1 As shown, the working steps of the control system of the continuous mixer model pre-control algorithm are as follows.

[0080] A1. Initialize the system.

[0081] A2. Check the status of each linked device. If there is a fault, the system will be locked and an alarm will be issued. If it is normal, the system will enter the patrol mode.

[0082] A3. Enter and select patrol mode, and judge the system control according to the learning equipment working condition / automatic operation / manual operation mode.

[0083] A4. Select the automatic operation mode. The servo speed is calculated in real time based on the digital model and near-infrared feedback data. The servo driver adjusts the speed according to the given value. If the servo system or near-infrared detection fails, the system locks and issues a system alarm.

[0084] A5. Select the learning equipment working mode, run the mixer at a fixed speed, perform near-infrared detection and real-time data feedback, and the processor records data in a fixed collection cycle, calculates the data model based on the collected data, and then enters the automatic operation mode.

[0085] A6. Select manual operation mode and manually set the servo speed. The servo driver adjusts the speed according to the given value. If there is a fault in the servo system or near-infrared detection, the system will lock and issue a system alarm as a prompt.

[0086] The embodiments of the present invention are described in detail above, but the contents described are only preferred embodiments of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A control method of a continuous mixer model pre-control algorithm, characterized in that the steps include: S1, near infrared numerical filtering, Collect multiple sets of real-time data of near-infrared detection and mixing uniformity reading within a fixed time to form an array, sort the data in the array in a bubble manner, obtain the middle value of the sorted data, write the middle value into the average filter algorithm array as the average filter array value, obtain the average result of the average filter array value as the near-infrared filter value and write it into the filter result variable; S2, learning device data, Set the initial mixing uniformity, operate the device to obtain the speed of the servo motor when the equipment has the highest mixing uniformity, gradually increase the speed of the servo motor according to a fixed time until the speed of the servo motor reaches the highest mixing uniformity, record the near-infrared filter value, motor speed value and corresponding time value each time, write the data value into the corresponding data array, and obtain the average operation curve model data; S3, restore the data model, plot the data obtained in S2 respectively, obtain the time-mixing uniformity data model and the time-speed data model, restore and smooth the time-mixing uniformity data model and the time-speed data model to obtain a smooth data curve, overlap the two curve models with time as the zero point, and obtain the time-mixing uniformity-speed model; S4. Control the operation of the equipment according to the time-mixing uniformity-speed model in S3.

2. The control method of the continuous mixer model pre-control algorithm according to claim 1 is characterized in that: The specific steps of S1 include: S11, data acquisition, reading the real-time data of mixing uniformity through near infrared detection, collecting data N times within a fixed time, and uploading the collected raw data to the array of the processor; S12, de-jittering sorting, comparing two adjacent data by bubbling the collected N results, if the condition that the value of the previous data is greater than the value of the following data is met, the positions of the previous and following data are swapped, if the condition is not met, no swap is performed, until the overall sorting of the data array is finally completed, and the result is written into the array; S13, de-jitter and obtain the median value, divide the array data after bubble sort by 2 N times or N+1 times to find the middle value, and write the middle value into the average filtering algorithm array; S14, average filtering, after filling the values ​​of the average filter array, add the array values ​​to obtain the total value, divide the total value by the number of average arrays, calculate the average result of the average filter array, and write the result as the near-infrared filter value into the filter result variable.

3. The control method of the continuous mixer model pre-control algorithm according to claim 2 is characterized in that: The specific steps of S13 are: determine whether the value of N is odd or even. If N is odd, then S=(N+1) / 2, and the Sth data in the array is written into the average filtering algorithm array as the intermediate value; if N is an even number, then S=N / 2, and the average of the Sth and S+1th data in the array is written into the average filtering algorithm array as the intermediate value.

4. The control method of the continuous mixer model pre-control algorithm according to claim 2 is characterized in that: The specific steps of S2 include: S21, specifying the mixing uniformity and setting the initial setting value of the mixing uniformity; S22, obtaining the motor speed when the device has the highest mixing uniformity, gradually increasing the servo motor speed according to the time interval, and recording the near-infrared filter value, motor speed value and corresponding time value written into the S14 filter result variable each time; S23, repeat S22 for multiple times to obtain data, and record the data values ​​obtained N times; S24, calculate the average operation curve, arrange the near-infrared filtering numerical results in the N times of data obtained in S23 according to the time interval column, remove the maximum and minimum values, divide the total value of the remaining data values ​​by 2 / N-2 to obtain the result and store it in the average filtering algorithm array, write the near-infrared filtering value into the near-infrared array, write the motor speed value into the speed array, write the time value into the time array, and obtain the complete average operation curve model data.

5. The control method of the continuous mixer model pre-control algorithm according to claim 4 is characterized in that: The specific steps of S3 include: S31, restore the two-dimensional data points, According to the average filtering curve data, point drawing is performed with the X-axis direction as the time axis and the Y-axis direction as the mixing uniformity data axis to obtain a time-mixing uniformity data model; According to the average filter curve data, the points are plotted with the X-axis direction as the time axis and the Y-axis direction as the speed data axis to obtain a time-speed data model; S32, restoring the basic model, directly connecting the two data points obtained in S31 to obtain the original non-smooth data curve; S33, processing the basic model to obtain a smoothing model, performing smoothing filling processing on the original data curve data with a quintic polynomial curve to obtain a smooth data curve; S34, superimposing the two curve models obtained in S33 based on the time zero point, obtaining a time-mixing uniformity-velocity model, and obtaining two smooth model curves centered on the time axis.

6. The control method of the continuous mixer model pre-control algorithm according to claim 5, characterized in that: The specific steps of S4 include: S41, pre-operating the equipment, operating the equipment at a set fixed speed value until a near-infrared value appears; S42, pre-modulate the running speed, find the position in the time-mixing uniformity-speed model corresponding to the current data according to the near-infrared data value and the time-mixing uniformity-speed model curve in S3, adjust the motor speed according to the speed value of the position, and approach the corresponding speed in the model by sending instructions to the servo driver; S43, coefficient control adjustment time, according to the value collected within a fixed time, compare the data of the time-mixing uniformity-velocity model to obtain the corresponding time adjustment coefficient k, and calculate the corresponding time position after adjustment by adding the time adjustment coefficient k to the current time of the time-mixing uniformity-velocity model position; S44, fine adjustment, compare the near-infrared value in real time with the data in the time-mixing uniformity-speed model according to the adjusted corresponding time position obtained in S43, traverse the time-mixing uniformity-speed model according to the value difference result to obtain the speed value at the next moment corresponding to the value, and output the value to the servo according to the corresponding speed value at the next moment during pre-control.

7. A control system for a continuous mixer model pre-control algorithm, characterized in that the control method for implementing the continuous mixer model pre-control algorithm according to any one of claims 1 to 6 comprises: A near infrared detection module is used to detect the mixing uniformity data and feed the mixing uniformity data back to the mixing uniformity model pre-control algorithm module; A mixing uniformity model pre-control algorithm module is used to control the near infrared detection module and the servo drive module through a digital model to control the operation of the continuous mixer device; Servo drive module for mixing screw rotation drive.

Images