Search

CN-121742409-B - Intelligent discharging monitoring system and method for medicine production

CN121742409BCN 121742409 BCN121742409 BCN 121742409BCN-121742409-B

Abstract

The application discloses a discharging intelligent monitoring system and method for medicine production, which belongs to the field of medicine discharging monitoring, the application obtains discharging characteristic parameters in the powder discharging process, extracts discharging state characteristics based on the obtained discharging characteristic parameters, generates a powder discharging state characteristic vector, obtain reation kettle production parameter, carry out powder ejection of compact risk assessment based on reation kettle production parameter and ejection of compact state characteristic, carry out risk influence degree judgement based on powder ejection of compact risk to carry out early warning and regulation and control, improve ejection of compact stability, promote production efficiency.

Inventors

  • TAO QIAN

Assignees

  • 江苏长江药业有限公司
  • 江苏恒拓医药股份有限公司

Dates

Publication Date
20260512
Application Date
20260227

Claims (7)

  1. 1. The intelligent discharge monitoring method for the medicine production is characterized by comprising the following specific steps of: Acquiring a discharging characteristic parameter in the powder discharging process; extracting discharging state characteristics based on the obtained discharging characteristic parameters, wherein the discharging state characteristics comprise weight change slope and fluctuation amplitude, vibration main frequency and energy distribution characteristics, discharging resistance change characteristics and sound effective frequency band energy characteristics, and generating a powder discharging state characteristic vector; the method comprises the following specific steps of obtaining production parameters of the reaction kettle, and carrying out powder discharging risk assessment based on the production parameters and discharging state characteristics of the reaction kettle: Carrying out normalization processing on each parameter in the powder discharging state feature vector, and carrying out weighted combination on the normalized parameters to obtain a real-time stability evaluation value; Acquiring temperature data and pressure data recorded in the production process of the reaction kettle, inputting the characteristic vector of the powder discharging state and the temperature data and pressure data recorded in the production process of the reaction kettle into a pre-trained deep learning neural network model for powder discharging risk assessment, and outputting a discharging flow risk index and a caking induction risk index; and judging the risk influence degree based on the powder discharging risk, and carrying out early warning and regulation.
  2. 2. The method for intelligently monitoring the discharge of medicines according to claim 1, wherein the step of obtaining the discharge characteristic parameters in the powder discharge process comprises the following specific steps: s11, acquiring an accumulated discharging weight value and corresponding time information in real time through a weighing device in the powder discharging process, and acquiring a mechanical vibration signal generated by a discharging channel in real time through a vibration sensor of the powder discharging channel; s12, reading a real-time motor current value through a motor drive controller; S13, collecting sound signals in the powder discharging process in real time through a pickup device in a discharging area.
  3. 3. The method for intelligently monitoring the discharge of medicines according to claim 2, wherein the step of extracting the discharge state characteristic based on the obtained discharge characteristic parameter to generate the powder discharge state characteristic vector comprises the following specific steps: s21, calculating weight change quantity through discharging weight between two adjacent sampling moments, dividing the weight change quantity by a corresponding sampling time interval to obtain weight change slopes in a corresponding time period, counting a plurality of weight change slopes in a preset time window, calculating deviation degree between each weight change slope and a corresponding average slope in the time window, and taking the deviation degree as fluctuation amplitude of weight change; s22, performing frequency domain conversion on the vibration signal through fast Fourier transform in a preset time window, selecting a frequency component with the largest energy ratio as a vibration main frequency characteristic, counting the energy of each frequency band of the vibration signal in a set frequency range, and calculating the proportion of the signal energy in each frequency band to the total energy to obtain a vibration energy distribution characteristic; s23, calculating the variation of the motor current between adjacent sampling moments based on the motor current data obtained by continuous sampling, and taking the variation as a resistance variation characteristic; s24, performing frequency domain conversion on the acquired sound signals, and counting sound energy in a frequency band based on a frequency interval of powder flow behaviors to serve as a sound effective frequency band energy characteristic; s25, generating a powder discharging state characteristic vector by using the weight change slope and fluctuation amplitude, the vibration main frequency and energy distribution characteristic, the discharging resistance change characteristic and the sound effective frequency band energy characteristic which are obtained in the same time window.
  4. 4. The intelligent monitoring method for discharging of medicine production according to claim 3, wherein the risk influence degree determination based on the powder discharging risk and the early warning and the regulation comprise the following specific steps: S41, comparing the real-time stability evaluation value with a preset comprehensive early warning threshold, wherein when the real-time stability evaluation value is smaller than the preset comprehensive early warning threshold, the discharging process is stable, the current discharging state is continued, and when the real-time stability evaluation value is larger than or equal to the preset comprehensive early warning threshold, the discharging process is unstable, and the primary early warning is triggered; S42, when the powder discharge flow risk index and the caking induction risk index are in a primary early warning state, respectively comparing the powder discharge flow risk index and the caking induction risk index with a preset flow threshold value, judging that the powder discharge flow risk index is normally discharged when the powder discharge flow risk index is smaller than the preset flow threshold value, judging that the powder discharge flow is abnormal when the powder discharge flow risk index is larger than or equal to the preset flow threshold value, reducing the rotating speed of a discharge mechanism driving motor within a preset range, simultaneously improving the vibration amplitude of a vibration device within the preset range, continuously evaluating the powder discharge flow risk index in an adjustment process, judging that the powder discharge flow risk index is normally discharged when the caking induction risk index is smaller than the preset caking threshold value, judging that the powder discharge caking is abnormal when the caking induction risk index is larger than or equal to the preset caking threshold value, gradually improving the vibration amplitude of the vibration device within the preset range, keeping the rotating speed of the discharge mechanism driving motor unchanged, and continuously evaluating the powder caking induction risk index in the adjustment process.
  5. 5. Intelligent monitoring system for the outfeed of pharmaceutical products, realized on the basis of the intelligent monitoring method for outfeed of pharmaceutical products according to any one of claims 1 to 4, characterized in that it comprises in particular: the data acquisition module is used for acquiring discharge characteristic parameters in the powder discharge process; the powder discharging state extraction module is used for extracting discharging state characteristics through the obtained discharging characteristic parameters to generate a powder discharging state characteristic vector; The powder discharging risk assessment module is used for acquiring production parameters of the reaction kettle and carrying out powder discharging risk assessment through the production parameters and discharging state characteristics of the reaction kettle; And the discharging risk early warning module is used for judging the risk influence degree through powder discharging risk and carrying out early warning and regulation.
  6. 6. An electronic device comprises a processor and a memory, wherein the memory stores a computer program which can be called by the processor; The method for intelligent monitoring of the discharge of pharmaceutical products according to any one of claims 1 to 4, characterized in that said processor executes the method for intelligent monitoring of the discharge of pharmaceutical products according to any one of claims 1 to 4 by calling a computer program stored in said memory.
  7. 7. A computer readable storage medium storing instructions which, when run on a computer, cause the computer to perform the intelligent monitoring method for outfeed for pharmaceutical production according to any one of claims 1-4.

Description

Intelligent discharging monitoring system and method for medicine production Technical Field The application belongs to the field of medicine discharging monitoring, and particularly relates to an intelligent discharging monitoring system and method for medicine production. Background In the pharmaceutical industry, the transportation and discharge of powder materials is one of the key links in the production process. In the process of conveying powder materials from a reaction kettle, a storage tank or a hopper and other containers to downstream equipment, the powder materials are often influenced by multiple factors such as material characteristics, a discharging equipment structure, vibration, temperature, process parameters and the like, and abnormal conditions such as uneven discharging, intermittent flow, bridging or caking and the like are easy to occur. Such anomalies not only lead to a decrease in production efficiency, but also can cause excessive equipment loads, pipe blockage, and even affect the quality stability of the final product. For example, in the production of pharmaceutical solid preparations, powder agglomeration can cause non-uniformity in downstream tabletting or mixing, affecting the accuracy of the drug dosage. At present, a monitoring method of a powder discharging process mainly depends on manual inspection or simple physical quantity measurement, such as a weighing device, motor current monitoring or a local vibration sensor, and the like, and single physical quantity monitoring can only reflect local states, so that the flowing condition and agglomeration risk of powder in the whole discharging channel are difficult to comprehensively evaluate. In addition, the existing automatic control usually depends on empirical parameter setting, such as setting a fixed discharge mechanism rotating speed or vibration amplitude, but the empirical control method cannot cope with dynamic changes under different material characteristics and production process conditions, and lacks intelligent analysis capability on the instantaneous discharge state of powder. Meanwhile, the powder discharging process is obviously influenced by production process parameters such as the temperature, the pressure and the like of the reaction kettle, and a complex nonlinear relation exists between the parameters and the discharging state. According to the application, the powder discharging state and the production parameters are acquired in real time through the multisource sensor, the flow and agglomeration risk prediction is carried out by combining the deep learning model, the intelligent monitoring and the regulation of the discharging process are realized, the discharging stability is improved, the agglomeration incidence rate is reduced, and the production efficiency is improved. Disclosure of Invention Aiming at the defects of the prior art, the application provides an intelligent discharging monitoring system and method for medicine production In order to achieve the above purpose, the present application provides the following technical solutions: A discharging intelligent monitoring system and method for medicine production comprises the following specific steps: Acquiring a discharging characteristic parameter in the powder discharging process; Extracting a discharging state characteristic based on the obtained discharging characteristic parameter to generate a powder discharging state characteristic vector; Acquiring production parameters of a reaction kettle, and performing powder discharge risk assessment based on the production parameters and discharge state characteristics of the reaction kettle; and judging the risk influence degree based on the powder discharging risk, and carrying out early warning and regulation. Preferably, the step of obtaining the discharge characteristic parameters in the powder discharge process comprises the following specific steps: S11, acquiring an accumulated discharging weight value and corresponding time information in real time through a weighing device in the powder discharging process, and acquiring mechanical vibration signals generated by a discharging channel in real time through a vibration sensor of the powder discharging channel, wherein the mechanical vibration signals comprise vibration acceleration and vibration amplitude signals and are used for reflecting the flowing state of powder in the discharging channel; S12, reading a real-time motor current value through a motor drive controller, and reflecting the mechanical obstruction degree of the powder on a discharging mechanism; S13, collecting sound signals in the powder discharging process in real time through a pickup device in a discharging area. Preferably, the extracting the discharging state feature based on the obtained discharging feature parameter, and generating the powder discharging state feature vector comprises the following specific steps: S21, calculating weight change quantity through