Study On Modeling And Quality Control Method Of Pharmaceutical Granulation Tableting Process Based On Changes Of Raw Material Properties

The pharmaceutical industry is an important industry that affects the national economy and the people’s livelihood.In the drug development process,the customization requirements of the drug will make the drug formulation often adjust,which will change the raw material properties of the pharmaceutical process,which leads to difficulties in the quality control method based on the process model.The data-driven control method is to directly design the controller by using the input and output data of the process.Therefore,it is of great theoretical and practical significance to study the data-driven pharmaceutical process quality control method under the condition that the material properties are frequently changed.This thesis first introduces the basic principles of the granulation and tableting process including fluid bed granulation and tableting based on a high-speed rotary tablet press,the process flow and the working principles of related production equipment,and analyzes the main influencing factors of the process(raw materials attributes and process operating variables)and their quality indicators.Then,through a more in-depth analysis and research on the process mechanism of the drug granulation and tableting process,a mechanism model was established to simulate the actual pharmaceutical process.First,for the fluidized bed granulation process,the drying mechanism model of the granules is introduced to calculate the moisture content of the granulation terminal granules,and it is coupled with the granulation mechanism model established by the predecessors,so as to establish the ability to simultaneously reflect the properties of raw materials,operating variables and multi-input and multi-output mechanism model of particle quality indicators such as the correlation between particle size and moisture content;Second,for the tableting process,based on the mechanism model established by the predecessors to characterize the relationship between particle quality,tableting machine operating variables and tablet porosity,introduce the mechanism model of tablet quality indicators such as tensile strength,hardness,dissolution,disintegration time and porosity,so as to establish a mechanism model that reflects the correlation between the quality of granules,the operating variables of the tablet press and the tablet quality indicators;Third,the mechanism model of fluidized bed granulation process and tableting process were linked to establish the process mechanism model of drug granulation tabletting process,and the validity of the model was verified by simulation experiments.This model is used to simulate the actual drug granulation compression process and use it to generate process data.Then,according to the variable coupling and quality range constraints in the granulation and tableting process,different data-driven control methods are applied to control the quality of the variables.Aiming at the problem of variable coupling,this thesis proposes a tablet quality control method based on feed-forward compensation data to drive adaptive decoupling for granulation and tableting processes;and for range-constrained quality indicators,a data-driven predictive iterative learning control for tableting and tableting process tablet quality control method.Based on the idea of fuzzy control,a data-driven control method based on fuzzy adaptive adjustment of algorithm parameters was proposed to improve the convergence speed of the tablet quality control method.Finally,in view of the influence of the disturbance of particle quality in the actual fluidized bed granulation process,this topic proposes the control of the quality range of the pharmaceutical tablets with the disturbance,and the validity of the method is verified by simulation experiments.