Chinese patent medicine intelligent production process control management method and system

Abstract

The invention relates to the technical field of Chinese patent medicine production process control management, in particular to a Chinese patent medicine intelligent production process control management method and system. The method comprises the following steps: acquiring suppository state data and stirring parameter data; constructing a particle retention area identification strategy and a shear resistance fluctuation detection method, and monitoring whether a pre-aggregation characteristic formed by suppository drug particle retention and aperiodic stirring torque sudden change appears or not during suppository mixing in real time; a suppository state vector and a frozen branch track vector are analyzed, and a freezing potential energy function is constructed based on an unbalanced state minimum dissipation track principle and a configuration entropy flow coupling mechanism to identify a solidification freezing risk dead angle area before suppository mold filling; and executing directional unlocking intervention on the freezing risk dead angle area by using a local control strategy. According to the invention, a particle retention area identification strategy, a shear resistance fluctuation detection method and a freezing potential energy function are constructed, and the whole-process dynamic monitoring management of the particle retention and solidification trend in the suppository mixing process is realized.

Description

Technical Field

[0001] This invention relates to the field of process control and management technology for the production of traditional Chinese medicine, specifically to an intelligent process control and management method and system for the production of traditional Chinese medicine. Background Technology

[0002] Traditional Chinese medicine (TCM) preparations, as a pharmaceutical system developed under the guidance of TCM theory, have complex production processes characterized by multi-dosage form synergy, multi-stage coupling, and multi-physical state linkage. Especially in key stages such as mixing, molding, and drying, higher demands are placed on the accuracy and stability of process control. TCM manufacturing is rapidly moving towards an intelligent production system characterized by data-driven approaches, process modeling, intelligent feedback, and production process control. Under this trend, conventional dosage forms such as pills, tablets, and granules have been relatively maturely integrated into MES management, industrial big data analysis, and intelligent scheduling systems, forming a controllable and traceable process control model. However, suppositories, due to their high material thermosensitivity, complex morphological rheology, and strong interphase interface interactions, are relatively marginalized in current intelligent systems, still heavily reliant on manual judgment and experience-based control, and have a very high dependence on the molding process, resulting in significant gaps in process control.

[0003] Currently, there are still many technical problems in the intelligent production of traditional Chinese medicine suppositories. Specifically, in the suppository mixing process of traditional Chinese medicine production, drug powder and lipid matrix may thicken prematurely or undergo micro-phase separation in local areas at low to medium speeds. There is also a temperature difference between the mixing tank wall and the internal drug mixture. Due to the micro-cooling effect, the drug near the wall or in the dead corner of the equipment may freeze or degenerate prematurely, resulting in uneven drug diffusion, re-agglomeration of micro-clusters, and blockage of the mold opening during subsequent filling. Traditional technical solutions only control process disturbances when agglomeration has already occurred or is approaching, mainly relying on torque fluctuation or stirring current feedback control. However, these signals are lagging and cannot identify local unstable areas, making it difficult to disperse agglomerates once they have formed. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for intelligent production process control and management of traditional Chinese medicine, in order to solve the problems mentioned in the background art, such as premature thickening or micro-phase separation in local areas of drug powder and lipid matrix at low and medium speeds, and the temperature difference between the wall of the mixing tank and the internal drug mixture, which leads to uneven drug diffusion, re-coagulation of micro-clusters, and blockage of the mold opening during subsequent filling process.

[0005] To achieve the above objectives, the technical solution of the present invention is: a method for intelligent production process control and management of traditional Chinese medicine preparations, comprising:

[0006] S1. Use sensors to collect real-time suppository status data and stirring parameter data throughout the entire suppository mixing process;

[0007] S2. In the early and middle stages of suppository mixing, a particle retention zone identification strategy and a shear resistance fluctuation detection method are constructed based on suppository state data to jointly monitor in real time whether suppository drug particle retention and pre-aggregation characteristics constituted by abrupt changes in non-periodic stirring torque occur during suppository mixing.

[0008] S3. In the later stage of suppository mixing, based on suppository state data, analyze the suppository state vector and freezing support vector, and construct a freezing potential energy function based on the non-equilibrium minimum dissipation orbit principle and configuration entropy flow coupling mechanism to identify the solidification and freezing risk dead zone area before suppository filling.

[0009] S4. Use local control strategies to perform targeted unlocking interventions in areas with freezing risk blind spots.

[0010] Preferably, the suppository state data includes particle concentration data, mixed viscosity data, and temperature distribution data;

[0011] The mixing parameter data includes the physical characteristics of the mixing tank, the geometry of the impeller, the mixing trajectory, and the output torque data of the mixing motor.

[0012] Preferably, the particle retention area identification strategy is based on the correlation judgment of the positional structural parameters of the suppository particle concentration deviation area and the weak disturbance area, and is used to identify the suppository drug particle retention area in real time.

[0013] The specific strategy for identifying particle retention areas is as follows:

[0014] S2.1.1 Calculate the average particle concentration of the suppository within one cycle based on the particle concentration data collected in real time at each time and location;

[0015] S2.1.2 Set a concentration deviation threshold. Monitor the concentration of particles at a certain location within a time window. If the concentration at a certain location is always greater than the concentration deviation threshold, then that location is the concentration deviation point. The set of all concentration deviation points is the concentration deviation zone.

[0016] S2.1.3. Based on the physical characteristics of the mixing tank, the geometric structure of the impeller, and the mixing trajectory, calculate the set of disturbance blind zone locations of the mixing tank;

[0017] S2.1.4 If the concentration deviation point in the concentration deviation zone is located in the set of disturbance blind zone locations, then the point is the suppository drug particle retention point, and the set of all high-risk particle retention points is the suppository drug particle retention area.

[0018] Preferably, the shear resistance fluctuation detection method is based on the statistical analysis of the real-time torque change rate of the stirring motor and the historical disturbance power fluctuation characteristics, and is used to identify non-periodic shear resistance abrupt events caused by local particle aggregation during suppository mixing.

[0019] The specific method for detecting shear resistance fluctuations is as follows:

[0020] S2.2.1 Construct a torque time series based on the output torque data of the stirring motor, and calculate the first-order difference average rate of change of torque within a time window;

[0021] S2.2.2 If the average rate of change of the first-order torque difference is greater than the sum of the historical mean of the torque difference and the standard deviation of the torque difference, then the shear resistance fluctuation is judged to be abnormal.

[0022] Preferably, in step S2, the pre-aggregation characteristics caused by suppository drug particle retention and abrupt changes in non-periodic stirring torque are jointly monitored in real time during suppository mixing, as follows:

[0023] Under abnormal shear resistance fluctuations and in the presence of suppository drug particle retention areas as described in S2.1.4, the suppository mixing process will exhibit a state of structural stability degradation caused by suppository drug particle retention and abrupt changes in non-periodic stirring torque.

[0024] Among them, the structural stability degradation state refers to the state of drug dispersion, continuous disturbance, and heat loss balance during the suppository mixing process.

[0025] Preferably, in step S3, the suppository state vector and freezing support vector are analyzed based on suppository state data, and a freezing potential energy function is constructed based on the non-equilibrium minimum dissipation trajectory principle and the configuration entropy flow coupling mechanism to identify the solidification and freezing risk dead zone region before suppository molding. The specific method steps are as follows:

[0026] S3.1. Define the suppository state evolution vector S(x,t) and the freezing trajectory vector S based on suppository state data. f And calculate the freezing path offset angle cosθ freeze (x,t) and frozen trajectory offset curvature κ θ (x,t), based on the frozen trajectory offset curvature κ θ (x,t) identifies whether the offset trend is accelerating;

[0027] S3.2 Constructing the freezing potential energy function E based on the non-equilibrium minimum dissipation orbit principle and configuration entropy flow coupling mechanism freeze (x,t);

[0028] S3.3, Freezing potential energy function E based on spatial point x of the mixing tank freeze (x,t) and combined with the frozen trajectory offset curvature κθ (x,t) identifies the dead zone area of ​​solidification and freezing risk before suppository filling;

[0029] Where x is a spatial point in the mixing tank; t is time.

[0030] Preferably, in step S3.2, the freezing potential energy function E is constructed based on the non-equilibrium minimum dissipation orbital principle and the configuration entropy flow coupling mechanism. freeze (x,t), specifically as follows:

[0031] S3.2.1. Based on the spatial point x in the mixing tank, define the local evolution state path dissipation functional J required for the freezing suppository. diss (x):

[0032]

[0033] Among them, J diss (x) represents the dissipative functional of the local evolution state path; x is a spatial point in the stirred tank; t is time; T(x,t) represents the temperature distribution data; η(x,t) represents the mixing viscosity data; v shear (x,t) represents the local shear rate; ξ(x,t) represents the flow resistance loss; α1 represents the temperature distribution weight; α2 represents the viscous drag weight; α3 represents the flow resistance loss weight; t2 represents the freezing start time; t1 represents the freezing end time.

[0034] S3.2.2 Constructing the local configuration entropy S based on the probability estimation model of multi-configuration particle distribution conf (x,t):

[0035]

[0036] Among them, S conf (x,t) is the local configuration entropy; k B N is the Boltzmann constant; p P represents the total number of configuration categories; k is the configuration type index; k (x,t) represents the configuration probability distribution;

[0037] S3.2.3, Fusion Local Evolution State Path Dissipation Function J diss (x) and local configuration entropy S conf (x,t), define the freezing potential function E freeze (x,t):

[0038]

[0039] Among them, E freeze (x,t) is the freezing potential energy function; λ is the fusion amplification coefficient.

[0040] Preferably, in S3.3, the freezing potential energy function E is based on the spatial point x of the mixing tank. freeze (x,t) and combined with the frozen trajectory offset curvature κ θ (x,t) is used to identify the potential dead zone for solidification and freezing risk before suppository filling. The specific method is as follows:

[0041] If the freezing potential energy function of spatial point x in the mixing tank at time t is E freeze (x,t) is greater than the set freezing potential energy critical threshold, and the freezing trajectory offset curvature κ θ If the derivative of (x,t) with respect to time t is greater than the set freezing offset curvature growth criterion, then the spatial point x of the mixing tank is a dead zone point of solidification and freezing risk.

[0042] The set of all points at risk of freezing and solidification is called the freezing and solidification risk dead zone.

[0043] Preferably, the local control strategy is as follows:

[0044] S4.1 Activate directional heating in areas with risk of solidification and freezing;

[0045] S4.2 Change the operating mode of the stirring blades to disturb the dead zone area at risk of solidification and freezing;

[0046] S4.3 Increase the vibration frequency in the dead zone area where solidification and freezing risks occur.

[0047] On the other hand, the present invention provides an intelligent production process control and management system for traditional Chinese medicine, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the aforementioned intelligent production process control and management method for traditional Chinese medicine.

[0048] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects:

[0049] 1. In this invention, by constructing a freezing trend precursor identification mechanism based on a particle retention zone identification model and a shear resistance fluctuation detection model, dynamic identification and early warning of local particle retention, shear weakening and structural stability degradation states are achieved at the process level in real-time process control during the early and middle stages of suppository mixing.

[0050] 2. In this invention, a freezing potential energy function is constructed based on the principle of minimum dissipation trajectory in non-equilibrium state and the coupling mechanism of configuration entropy flow. This function reflects the nonlinear coupling relationship between the organization rate of drug particle microstructure and the dissipation cost of evolution path, thereby realizing the spatial positioning, identification and control of the irreversible bifurcation trajectory of local freezing trend in the mixing tank during the later stage of suppository mixing. Attached Figure Description

[0051] Figure 1 This is a flowchart of one embodiment of the present invention. Detailed Implementation

[0052] Example 1, as Figure 1 As shown, the specific implementation steps of the intelligent production process control and management method for traditional Chinese medicine proposed in this invention are as follows:

[0053] S1. Use sensors to collect real-time suppository status data and stirring parameter data throughout the entire suppository mixing process;

[0054] The suppository state data includes particle concentration data, mixing viscosity data, and temperature distribution data; the stirring parameter data includes the physical properties of the stirring tank, the blade geometry, the stirring trajectory, and the output torque data of the stirring motor.

[0055] In this embodiment, after the drug particles are dispersed in the suppository matrix such as lipids or polyethylene glycol, they exhibit absorption characteristics for near-infrared light of specific wavelengths. An embedded NIR fiber optic probe is installed, and near-infrared spectroscopy combined with a point probe array method is used to monitor the characteristic absorption peaks representing drug concentration in real time. A multi-band modeling method combined with principal component regression is used to convert the spectral signals into concentration values. All probe data are aggregated to form the particle concentration data of the mixing chamber. During the rotation of the mixer spindle, viscosity changes directly affect the required driving torque. A method combining indirect estimation of motor torque, intelligent soft measurement models, and local magnetoresistive sensing is used to collect motor power and torque signals in real time, and the global mixing viscosity data is estimated by combining the rotational speed and blade structure parameters. A thermocouple array combined with an infrared temperature measurement module is used to collect temperature distribution data for various areas of the mixing tank.

[0056] Particle concentration data represents the concentration distribution of drug particles in the matrix at different locations and at different times, used to determine whether the particles are uniformly distributed and whether there are agglomerated areas; mixing viscosity data represents the dynamic viscosity of the suppository mixture at different locations and changes over time, used to determine whether the system thickens / agglomerates and whether the fluidity decreases; temperature distribution data represents the temperature distribution at various locations in the mixing cavity, used to determine whether freezing may occur in local thermal imbalance areas.

[0057] S2. In the early and middle stages of suppository mixing, a particle retention zone identification strategy and a shear resistance fluctuation detection method are constructed based on suppository state data to jointly monitor in real time whether suppository drug particle retention and pre-aggregation characteristics constituted by abrupt changes in non-periodic stirring torque occur during suppository mixing.

[0058] The particle retention area identification strategy is based on the correlation judgment of the positional and structural parameters of the suppository particle concentration deviation area and the weak disturbance area, and is used to identify the suppository drug particle retention area in real time.

[0059] In this embodiment, the particle retention area identification strategy is as follows:

[0060] S2.1.1 Calculate the average particle concentration of the suppository within one cycle based on the particle concentration data collected in real time at each time and location;

[0061] S2.1.2 Set a concentration deviation threshold. Monitor the concentration of particles at a certain location within a time window. If the concentration at a certain location is always greater than the concentration deviation threshold, then that location is the concentration deviation point. The set of all concentration deviation points is the concentration deviation zone.

[0062] S2.1.3. Based on the physical characteristics of the mixing tank, the geometric structure of the impeller, and the mixing trajectory, calculate the set of disturbance blind zone locations of the mixing tank;

[0063] S2.1.4 If the concentration deviation point in the concentration deviation zone is located in the set of disturbance blind zone locations, then the point is the suppository drug particle retention point, and the set of all high-risk particle retention points is the suppository drug particle retention area.

[0064] The shear resistance fluctuation detection method is based on the statistical analysis of the real-time torque change rate of the stirring motor and the historical disturbance power fluctuation characteristics. It is used to identify abrupt changes in shear resistance caused by local particle aggregation during suppository mixing.

[0065] In this embodiment, the shear resistance fluctuation detection method is as follows:

[0066] S2.2.1 Construct a torque time series based on the output torque data of the stirring motor, and calculate the first-order difference average rate of change of torque within a time window;

[0067] S2.2.2 If the average rate of change of the first-order torque difference is greater than the sum of the historical mean of the torque difference and the standard deviation of the torque difference, then the shear resistance fluctuation is judged to be abnormal.

[0068] In this embodiment, shear resistance fluctuation refers to the small, continuous change in the shear fluid resistance of the mixing medium on the time axis during the stirring process. During the suppository mixing process, the drug particles are suspended in the molten matrix. As the stirring paddle rotates, the slurry is subjected to fluid viscous resistance. This resistance is not completely constant but fluctuates with time, mainly affected by the following factors: particle distribution uniformity, viscosity change, and stirring flow field disturbance structure.

[0069] Abnormal shear resistance fluctuations refer to a phenomenon where the rate of torque change during stirring significantly exceeds its historical or normal fluctuation range, resulting in non-periodic and sudden torque increases. Specific manifestations include: a sudden increase in the real-time torque curve of the motor; this fluctuation does not occur periodically with stirring, but rather occurs locally and suddenly; when it occurs, there may be local aggregation of drug particles, micro-coagulated matrix clumps, or a viscosity dissimilar zone suddenly passing through the impeller leading edge in the local space; the data characteristic is an abnormally high first-order difference average rate of change of torque.

[0070] In S2, the pre-aggregation characteristics caused by suppository particle retention and abrupt changes in non-periodic stirring torque are monitored in real time during suppository mixing, as follows:

[0071] Under abnormal shear resistance fluctuations and in the presence of suppository drug particle retention areas as described in S2.1.4, the suppository mixing process will exhibit a state of structural stability degradation caused by suppository drug particle retention and abrupt changes in non-periodic stirring torque.

[0072] Among them, the structural stability degradation state refers to the state of drug dispersion, continuous disturbance, and heat loss balance during the suppository mixing process.

[0073] In this embodiment, the structural stability degradation state specifically includes: drug particles begin to irreversibly aggregate in certain areas, form a stagnant layer in the disturbance coverage blind zone, have an unbalanced flow field structure, uneven spatial distribution of viscosity, decreased heat exchange efficiency, decreased mold flowability, and increased risk of filling failure.

[0074] S3. In the later stage of suppository mixing, based on suppository state data, analyze the suppository state vector and freezing support vector, and construct a freezing potential energy function based on the non-equilibrium minimum dissipation orbit principle and configuration entropy flow coupling mechanism to identify the solidification and freezing risk dead zone area before suppository filling.

[0075] In this embodiment, the principle of the minimum dissipation trajectory in non-equilibrium states that in an open dissipation system far from thermodynamic equilibrium, the system will preferentially evolve along the trajectory with the minimum path dissipation during self-organization evolution or structural transition.

[0076] The configuration entropy-flow coupling mechanism refers to the dynamic reflection of the organization rate of the microstructure of a suppository mixture from high to low degrees of freedom by calculating the time change rate of the distribution of multiconfigurational particles in the suppository mixture, i.e., the first derivative of the configuration information entropy, within the theoretical framework of non-equilibrium thermodynamics and statistical mechanics. By embedding the configuration entropy flow as the driving force of freezing trend evolution into the freezing potential energy function, continuous monitoring and trend response modeling of the degradation process of suppository structural stability can be achieved. This mechanism is used to quantify the organization rate of particle microstructure in suppositories in real time, serving as a dynamic characterization index of the driving force of freezing trend evolution.

[0077] In step S3, the suppository state vector and freezing support vector are analyzed based on suppository state data. A freezing potential energy function is constructed based on the non-equilibrium minimum dissipation trajectory principle and the configuration entropy flow coupling mechanism to identify the solidification and freezing risk dead zone region before suppository molding. The specific method steps are as follows:

[0078] S3.1. Define the suppository state evolution vector S(x,y) and the freezing trajectory vector S based on suppository state data. f And calculate the freezing path offset angle cosθ freeze (x,t) and frozen trajectory offset curvature κ θ (x,t), based on the frozen trajectory offset curvature κ θ (x,t) identifies whether the offset trend is accelerating;

[0079] In this embodiment S3.1, the suppository state evolution vector S(x,t) describes the actual multiphysics evolution state of the suppository at point x in the mixing tank at time t. It is the basic vector for determining the freezing trend direction, specifically:

[0080]

[0081] in, The rate of temperature change is used to determine whether a cooling and freezing trend has occurred. It is the second derivative of temperature diffusion, used to determine whether freezing is caused by a break in heat flow; The viscosity increase rate is a precursor to local freezing; The perturbation contraction rate;

[0082] Freeze the support vector S f Indicates the direction of change of each physical quantity of the system under the freezing trend. If:

[0083]

[0084] Among them, the frozen support vector S f The first and second terms are -1, indicating that the temperature is decreasing and thermal diffusion is weakening, and the system is entering a state of insufficient thermal energy; the frozen lateral vector S f The third term is +1, indicating a rapid increase in viscosity, a precursor to solidification; the frozen support vector S f The fourth term is -1, indicating that the contraction shear energy of the disturbed flow field decreases and an ineffective stirring region appears;

[0085] Freeze path offset angle cosθ freeze (x,t) represents the suppository state evolution vector S(x,t) and the freezing trajectory vector S. f The cosine of the included angle is a quantitative structural indicator of whether the current evolution trend is consistent with the direction of freezing, and is used for the initial judgment of the possibility of freezing.

[0086] Freeze trajectory offset curvature κ θ (x,t) is used to describe the rate of change of the offset angle, and is calculated as follows:

[0087]

[0088] S3.2 Constructing the freezing potential energy function E based on the non-equilibrium minimum dissipation orbit principle and configuration entropy flow coupling mechanism freeze (x,t);

[0089] S3.3, Freezing potential energy function E based on spatial point x of the mixing tank freeze (x,t) and combined with the frozen trajectory offset curvature k θ (x,t) identifies the dead zone area of ​​solidification and freezing risk before suppository filling;

[0090] Where x is a spatial point in the mixing tank; t is time.

[0091] In S3.2, the freezing potential energy function E is constructed based on the principle of minimum dissipation orbit in non-equilibrium state and the coupling mechanism of configuration entropy flow. freeze (x,t), specifically as follows:

[0092] S3.2.1. Based on the spatial point x in the mixing tank, define the local evolution state path dissipation functional J required for the freezing suppository. diss (x):

[0093]

[0094] Among them, J diss (x) represents the dissipative functional of the local evolution state path; x is a spatial point in the stirred tank; t is time; T(x,t) represents the temperature distribution data; η(x,t) represents the mixing viscosity data; v shear (x,t) represents the local shear rate; ξ(x,t) represents the flow resistance loss; α1 represents the temperature distribution weight; α2 represents the viscous drag weight; α3 represents the flow resistance loss weight; t2 represents the freezing start time; t1 represents the freezing end time.

[0095] S3.2.2 Constructing the local configuration entropy S based on the probability estimation model of multi-configuration particle distribution conf (x,t):

[0096]

[0097] Among them, S conf (x,t) is the local configuration entropy; k B N is the Boltzmann constant; p P represents the total number of configuration categories; k is the configuration type index; k (x,t) represents the configuration probability distribution;

[0098] In this embodiment S3.2.2, the multiconfiguration particle distribution probability estimation model is a configuration classification system based on image recognition algorithm, particle spectrum feature separation algorithm and disturbance response behavior analysis. It is used to identify the structural state of the particles in the suppository at spatial point x and time t in the mixing tank, and express its statistical probability distribution in the form of configuration proportion, providing particle size controllable input data for local configuration entropy calculation.

[0099] In this embodiment, a local configuration entropy S is constructed based on a multi-configuration particle distribution probability estimation model. conf (x,t), specifically as follows:

[0100] The space of the mixing tank is divided into several mixing tank space points x. At time t, real-time image sequences, spectral NIR signals, particle trajectory drift and vibration frequency response are collected from each mixing tank space point x.

[0101] Set the total number of configuration categories N p =4, k=1 indicates that the drug particles are in a free and dispersed state, with a uniform image, low peaks in the spectrum, and high perturbation response; k=2 indicates that the drug particles are in a weakly aggregated state, with concentrated density at the image edges and small peaks; k=3 indicates that the drug particles are in a clearly agglomerated state, with double peaks in the spectrum and local flocculent structure in the image; k=4 indicates that the drug particles are in a primary frozen state with high-frequency microcrystalline peaks and significantly reduced fluidity.

[0102] At time t at point x in the mixing tank, count the number N particles belonging to configuration k. k (x,t) and the total number of particles N total (x,t) yields the configuration probability distribution P. k (x,t):

[0103]

[0104] Based on configuration probability distribution P k Construct the local configuration entropy S for (x,t). conf (x,t), and the local configuration entropy S is updated every 10 seconds. conf (x,t);

[0105] When all P k (x,t) are all close to 1 / N p At a certain point, the local configurational entropy is at its maximum, and the drug particle state is at its most disordered; when a certain P k When (x,t) is close to 1, the local configurational entropy is the minimum, the drug particle state is the most ordered, and it is close to freezing.

[0106] S3.2.3, Fusion Local Evolution State Path Dissipation Function J diss (x) and local configuration entropy S conf (x,t), define the freezing potential function Efreeze (x,t):

[0107]

[0108] Among them, E freeze (x,t) is the freezing potential energy function; λ is the fusion amplification coefficient;

[0109] In this embodiment, the local evolution state path dissipation functional J diss (x) is a functional of the principle of minimum dissipation, expressing all energy consumption, including heat dissipation, during the process from the current state to the frozen state in the interval from time t2 to time t1. Shear loss And the mixed flow resistance dissipation, i.e., α3ξ(x,t);

[0110] S is the derivative of the local configuration entropy. conf (x,t) is the local configuration entropy describing whether the drug particles are multiconfigurative dispersions or have a simple aggregate structure. When the derivative of the local configuration entropy term... This indicates that the drug particles are becoming structurally ordered, i.e., the freezing tendency is increasing, and the derivative of the local configuration entropy term... Taking the absolute value indicates that the faster the structure organizes, the stronger the tendency to freeze; the local configuration entropy derivative term is a nonlinear amplification mechanism: if the entropy flow is fast, it indicates that the system has an extremely strong freezing tendency, and even if the local evolution state path dissipation functional is high, it may be frozen and solidified.

[0111] Freezing potential energy function E freeze In (x,t), the local evolution state path dissipation functional J diss The larger (x) is, the greater the resistance to the freezing path, and the greater the resistance to the freezing path, the more difficult it is to freeze. The value of λ represents the driving force of the freezing trend; the larger the value, the easier it is to freeze. λ represents the driving effect of the structural trend on the freezing rate, which is calibrated by experts.

[0112] In S3.3, the freezing potential energy function E is based on the spatial point x of the mixing tank. freeze (x,y) and combined with the frozen trajectory offset curvature κ θ (x,t) is used to identify the potential dead zone for solidification and freezing risk before suppository filling. The specific method is as follows:

[0113] If the freezing potential energy function of spatial point x in the mixing tank at time t is E freeze (x,t) is greater than the set freezing potential energy critical threshold, and the freezing trajectory offset curvature κ θ If the derivative of (x,t) with respect to time t is greater than the set freezing offset curvature growth criterion, then the spatial point x of the mixing tank is a dead zone point of solidification and freezing risk.

[0114] The set of all points at risk of freezing and solidification is called the freezing and solidification risk dead zone.

[0115] S4. Use local control strategies to perform targeted unlocking interventions in areas with freezing risk blind spots;

[0116] In this embodiment, the local control strategy is as follows:

[0117] S4.1 Activate directional heating in areas with risk of solidification and freezing;

[0118] S4.2 Change the operating mode of the stirring blades to disturb the dead zone area at risk of solidification and freezing;

[0119] S4.3 Increase the vibration frequency in the dead zone area where solidification and freezing risks occur.

[0120] Example 2: The present invention proposes an intelligent production process control and management system for traditional Chinese medicine, which is applied to the intelligent production process control and management method for traditional Chinese medicine proposed in Example 1. It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the intelligent production process control and management method for traditional Chinese medicine in Example 1.

[0121] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for intelligent production process control and management of traditional Chinese medicine preparations, characterized in that, Includes the following steps: S1. Use sensors to collect real-time suppository status data and stirring parameter data throughout the entire suppository mixing process; S2. In the early and middle stages of suppository mixing, a particle retention zone identification strategy and a shear resistance fluctuation detection method are constructed based on suppository state data to jointly monitor in real time whether suppository drug particle retention and pre-aggregation characteristics constituted by abrupt changes in non-periodic stirring torque occur during suppository mixing. S3. In the later stage of suppository mixing, based on suppository state data, analyze the suppository state vector and freezing support vector, and construct a freezing potential energy function based on the non-equilibrium minimum dissipation orbit principle and configuration entropy flow coupling mechanism to identify the solidification and freezing risk dead zone area before suppository filling. S4. Use local control strategies to perform targeted unlocking interventions in areas with freezing risk blind spots.

2. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 1, characterized in that: The suppository state data includes particle concentration data, mixed viscosity data, and temperature distribution data; The mixing parameter data includes the physical characteristics of the mixing tank, the geometry of the impeller, the mixing trajectory, and the output torque data of the mixing motor.

3. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 2, characterized in that: The particle retention area identification strategy is based on the correlation judgment of the positional and structural parameters of the suppository particle concentration deviation area and the weak disturbance area, and is used to identify the suppository drug particle retention area in real time. The specific strategy for identifying particle retention areas is as follows: S2.1.1 Calculate the average particle concentration of the suppository within one cycle based on the particle concentration data collected in real time at each time and location; S2.1.2 Set a concentration deviation threshold. Monitor the concentration of particles at a certain location within a time window. If the concentration at a certain location is always greater than the concentration deviation threshold, then that location is the concentration deviation point. The set of all concentration deviation points is the concentration deviation zone. S2.1.

3. Based on the physical characteristics of the mixing tank, the geometric structure of the impeller, and the mixing trajectory, calculate the set of disturbance blind zone locations of the mixing tank; S2.1.4 If the concentration deviation point in the concentration deviation zone is located in the set of disturbance blind zone locations, then the point is the suppository drug particle retention point, and the set of all high-risk particle retention points is the suppository drug particle retention area.

4. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 3, characterized in that: The shear resistance fluctuation detection method is based on the statistical analysis of the real-time torque change rate of the stirring motor and the historical disturbance power fluctuation characteristics. It is used to identify abrupt non-periodic shear resistance changes caused by local particle aggregation during suppository mixing. The specific method for detecting shear resistance fluctuations is as follows: S2.2.1 Construct a torque time series based on the output torque data of the stirring motor, and calculate the first-order difference average rate of change of torque within a time window; S2.2.2 If the average rate of change of the first-order torque difference is greater than the sum of the historical mean of the torque difference and the standard deviation of the torque difference, then the shear resistance fluctuation is judged to be abnormal.

5. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 4, characterized in that: In S2, the pre-aggregation characteristics caused by suppository particle retention and abrupt changes in non-periodic stirring torque are monitored in real time during suppository mixing, as follows: Under abnormal shear resistance fluctuations and in the presence of suppository drug particle retention areas as described in S2.1.4, the suppository mixing process will exhibit a state of structural stability degradation caused by suppository drug particle retention and abrupt changes in non-periodic stirring torque. Among them, the structural stability degradation state refers to the state of drug dispersion, continuous disturbance, and heat loss balance during the suppository mixing process.

6. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 5, characterized in that: In step S3, the suppository state vector and freezing support vector are analyzed based on suppository state data. A freezing potential energy function is constructed based on the non-equilibrium minimum dissipation trajectory principle and the configuration entropy flow coupling mechanism to identify the solidification and freezing risk dead zone region before suppository molding. The specific method steps are as follows: S3.

1. Define the suppository state evolution vector S(x,t) and the freezing trajectory vector S based on suppository state data. f And calculate the freezing path offset angle cosθ freeze (x,t) and frozen trajectory offset curvature κ θ (x,t), based on the frozen trajectory offset curvature κ θ (x,t) identifies whether the offset trend is accelerating; S3.2 Constructing the freezing potential energy function E based on the non-equilibrium minimum dissipation orbit principle and configuration entropy flow coupling mechanism freeze (x,t); S3.3, Freezing potential energy function E based on spatial point x of the mixing tank freeze (x,t) and combined with the frozen trajectory offset curvature κ θ (x,t) identifies the dead zone area of ​​solidification and freezing risk before suppository filling; Where x is a spatial point in the mixing tank; t is time.

7. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 6, characterized in that: In S3.2, the freezing potential energy function E is constructed based on the principle of minimum dissipation orbit in non-equilibrium state and the coupling mechanism of configuration entropy flow. freeze (x,t), specifically as follows: S3.2.

1. Based on the spatial point x in the mixing tank, define the local evolution state path dissipation functional J required for the freezing suppository. diss (x): Among them, J diss (x) represents the dissipative functional of the local evolution state path; x is a spatial point in the stirred tank; t is time; T(x,t) represents the temperature distribution data; η(x,t) represents the mixing viscosity data; v shear (x,t) represents the local shear rate; ξ(x,t) represents the flow resistance loss; α1 represents the temperature distribution weight; α2 represents the viscous drag weight; α3 represents the flow resistance loss weight; t2 represents the freezing start time; t1 represents the freezing end time. S3.2.2 Constructing the local configuration entropy S based on the probability estimation model of multi-configuration particle distribution conf (x,t): Among them, S conf (x,t) is the local configuration entropy; k B N is the Boltzmann constant; p P represents the total number of configuration categories; k is the configuration type index; k (x,t) represents the configuration probability distribution; S3.2.3, Fusion Local Evolution State Path Dissipation Function J diss (x) and local configuration entropy S conf (x,t), define the freezing potential function E freeze (x,t): Among them, E freeze (x,t) is the freezing potential energy function; λ is the fusion amplification coefficient.

8. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 7, characterized in that: In S3.3, the freezing potential energy function E is based on the spatial point x of the mixing tank. freeze (x,t) and combined with the frozen trajectory offset curvature κ θ (x,t) is used to identify the potential dead zone for solidification and freezing risk before suppository filling. The specific method is as follows: If the freezing potential energy function of spatial point x in the mixing tank at time t is E freeze (x,t) is greater than the set freezing potential energy critical threshold, and the freezing trajectory offset curvature κ θ If the derivative of (x,t) with respect to time t is greater than the set freezing offset curvature growth criterion, then the spatial point x of the mixing tank is a dead zone point of solidification and freezing risk. The set of all points at risk of freezing and solidification is called the freezing and solidification risk dead zone.

9. The intelligent production process control and management method for traditional Chinese medicine preparations according to claim 8, characterized in that: The local control strategy is as follows: S4.1 Activate directional heating in areas with risk of solidification and freezing; S4.2 Change the operating mode of the stirring blades to disturb the dead zone area at risk of solidification and freezing; S4.3 Increase the vibration frequency in the dead zone area where solidification and freezing risks occur.

10. A smart production process control and management system for traditional Chinese medicine preparations, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes a computer program to implement the intelligent production process control and management method for traditional Chinese medicine as described in any one of claims 1-9.

Images