Wall breaking, drying and subpackaging integrated production device for traditional Chinese medicine decoction pieces

Abstract

The invention discloses a traditional Chinese medicine decoction piece wall breaking, drying and sub-packaging integrated production device, and relates to the technical field of traditional Chinese medicine decoction piece processing, the traditional Chinese medicine decoction piece wall breaking, drying and sub-packaging integrated production device comprises an adjustable conveying mechanism, a wall breaking mechanism, a discharging mechanism and an intelligent cooperation system matched with each mechanism; the intelligent cooperative system comprises a sensing acquisition module, an intelligent analysis module and a cooperative control module; and the intelligent analysis module processes and analyzes real-time data, including multi-dimensional analysis and establishment of a dynamic association model to evaluate the operation state, identification of potential faults and early warning in combination with a historical fault database, fusion of decoction piece characteristics and real-time data to generate a self-adaptive control strategy, and evaluation and feedback of the quality grade of a finished product. Through combination of a mechanical structure and an intelligent cooperative system, the core pain point of traditional equipment is solved from the three aspects of parameter matching, production efficiency and quality stability, integrated cooperative operation of wall breaking, drying and subpackaging of traditional Chinese medicine decoction pieces and parameter self-adaptive optimization are achieved, and the production efficiency and the finished product quality are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of processing Chinese herbal medicine slices, and in particular to a wall-breaking, drying and packaging integrated production device for Chinese herbal medicine slices. Background Art

[0002] As an important form of clinical TCM medication, the quality of Chinese herbal medicine slices is directly related to their efficacy. In the production process of Chinese herbal medicine slices, cell wall breaking, drying, and packaging are key steps to ensure the preservation of active ingredients, extend shelf life, and facilitate clinical use. Traditional Chinese herbal medicine slice production often uses a staged process, where cells are first broken down using a separate cell wall breaking device, then transferred to a drying device for water removal, and finally metered and packaged manually or using simple packaging equipment.

[0003] During use, the existing integrated production equipment for breaking, drying and packaging of Chinese herbal medicine slices is prone to over-drying or insufficient drying due to poor parameter matching between different equipment (for example, the humidity of the slices after breaking is difficult to accurately coordinate with the parameters of the drying equipment), which in turn affects the stability of the slices and the content of active ingredients. Summary of the Invention

[0004] The present invention aims to solve the problems of poor matching of equipment parameters and poor drying effect in the existing integrated production device for breaking, drying and packaging of Chinese herbal medicine slices, and to provide a production device that can realize integrated and coordinated operation of breaking, drying and packaging, and has adaptive parameter optimization.

[0005] To achieve the above-mentioned object, the present invention adopts the following technical solution: an integrated production device for breaking, drying and packaging Chinese herbal medicine slices, comprising a device housing, characterized in that first guard plates are hingedly connected to both inner walls above one end of the device housing, an adjustable conveying mechanism is installed between the first guard plates, a wall breaking mechanism is vertically installed in the middle of the device housing at one end of the adjustable conveying mechanism, and a discharge mechanism is installed on one side of the lower end of the wall breaking mechanism;

[0006] It also includes an intelligent collaborative system adapted to each mechanism in the device, and the intelligent collaborative system includes a sensing acquisition module, an intelligent analysis module, and a collaborative control module;

[0007] The sensor collection module is used to collect the physical parameters and material status of each link of the device, including the delivery volume and pressure of the adjustable conveying mechanism, the broken wall particle size, the temperature and humidity in the cylinder of the wall breaking mechanism, the discharge speed and material accumulation status of the discharge mechanism;

[0008] The intelligent analysis module is used to process and analyze the real-time data obtained by the sensor acquisition module, including multi-dimensional analysis to establish a dynamic correlation model to evaluate the operating status of each link, identify potential faults and determine whether to issue an early warning based on the historical fault database, integrate the characteristics of the medicinal materials with real-time data to generate an adaptive control strategy, and evaluate the quality level of the finished product and provide feedback;

[0009] The collaborative control module is used to drive the corresponding execution equipment according to the real-time control parameters output by the intelligent analysis module; and dynamically revise the control strategy of the next batch based on the quality assessment report.

[0010] Preferably, the intelligent analysis module specifically includes a status monitoring and analysis unit, a fault diagnosis and early warning unit, a parameter optimization and decision-making unit, and a quality assessment and feedback unit;

[0011] The status monitoring and analysis unit is used to perform multi-dimensional analysis on the real-time data of the sensor acquisition module, establish a dynamic correlation model of conveying volume, wall breaking strength, drying temperature, and discharge speed, and output quantitative evaluation results of the operating status of each link;

[0012] The fault diagnosis and early warning unit identifies potential fault types based on the abnormal fluctuation characteristics of status data and combines it with the historical fault database, and determines whether to generate an early warning signal;

[0013] Parameter optimization and decision-making unit, used to integrate the characteristic parameters of Chinese herbal medicine slices with real-time operation data, and generate adaptive control strategies through reinforcement learning algorithms;

[0014] The quality assessment and feedback unit evaluates the quality grade of the finished product based on the particle size distribution after wall breaking and the moisture content after drying, and feeds the results back to the parameter optimization unit.

[0015] Preferably, the specific analysis steps of the condition monitoring and analysis unit include:

[0016] Data preprocessing: noise reduction is performed on the conveying volume, pressure, particle size, temperature, humidity, discharge speed and stacking status data obtained by the sensor acquisition module, abnormal jump values ​​are eliminated, and the data are converted into unified dimension parameters through data standardization;

[0017] Feature extraction: extract the time series fluctuation characteristics of the conveying volume, the distribution characteristics of the broken particle size, the gradient characteristics of the drying temperature field, and the coupling characteristics of the discharge speed and stacking state from the preprocessed data;

[0018] Dynamic correlation modeling: Based on the extracted features, the mapping relationship between conveying volume and wall breaking strength, the response relationship between drying temperature and discharge speed, and the cross-influence function between various parameters are constructed to form a multi-dimensional dynamic correlation model;

[0019] Quantitative evaluation is generated by outputting the operation stability index, parameter matching coefficient and efficiency loss rate of each link through a multi-dimensional dynamic correlation model, and integrating them to form a quantitative evaluation result of the operation status of each link.

[0020] Preferably, the specific analysis steps of the fault diagnosis and early warning unit include:

[0021] Abnormal feature extraction: From the quantitative evaluation results output by the condition monitoring and analysis unit, identify characteristic parameters that exceed the normal fluctuation range, and extract the abnormal parameter change rate, duration, and coordinated variation characteristics of related parameters;

[0022] Fault pattern matching: calculate the similarity between the extracted abnormal features and the typical fault pattern feature vectors stored in the historical fault database to determine the fault type with the highest matching degree;

[0023] Early warning judgment calculates the probability of fault occurrence based on the matching results and combines it with the fault impact level. When the coupled value of the probability and level reaches the preset trigger condition, an early warning signal is generated containing the fault location, type, and recommended handling method; the fault impact level includes minor impact, performance degradation, and equipment damage.

[0024] Preferably, the specific analysis steps of the parameter optimization and decision-making unit include:

[0025] Feature fusion: The characteristic parameters of Chinese herbal medicine slices are integrated with the real-time operation parameters output by the status monitoring and analysis unit at the feature level to construct a comprehensive feature matrix;

[0026] Strategy generation, taking the comprehensive feature matrix as input, calls the historical optimal control strategy library through the reinforcement learning algorithm, iteratively calculates the efficiency benefits under different control parameter combinations, including wall breaking rate, drying uniformity, and energy consumption, and generates the initial adaptive control strategy;

[0027] Strategy verification: Substitute the initial strategy into the dynamic association model for simulation verification, evaluate the parameter matching and goal achievement after the strategy is implemented, correct the deviation items, and form the final control strategy.

[0028] Preferably, the specific analysis steps of the quality assessment and feedback unit include:

[0029] Quality parameter extraction: statistical analysis of particle size distribution data after wall breaking to extract particle size distribution uniformity and target particle size ratio; spatial distribution analysis of moisture content data after drying to extract average moisture content and moisture content range;

[0030] Grading: Based on the particle size distribution index and moisture content index, the grade is determined by comparing it with the preset quality standard value range group, generating a quality grade label and detailed list of non-compliant items; the preset quality standard value range group includes excellent, qualified, and rework value ranges;

[0031] Feedback correction: The quality grade label, details of non-compliant items, and corresponding real-time operating parameters are fed back to the parameter optimization and decision-making unit as penalty or reward items when generating the next batch of strategies, adjusting the weight coefficient of the reinforcement learning algorithm to achieve closed-loop correction.

[0032] Preferably, the adjustable conveying mechanism includes rollers rotatably mounted at both ends between the first guard plates on both sides, a first transmission belt is sleeved on the outer side of the roller, the upper end of the first transmission belt is fixedly connected to a blocking plate at equal distances, and the first guard plates on both sides of the first transmission belt are hinged with a hinged rod, the other end of the hinged rod is connected to the telescopic end of the electric telescopic rod by a thread, and the other end of the electric telescopic rod is hinged to one side of the device housing;

[0033] A second servo motor is installed on one side of the device housing, and an output end of the second servo motor passes through the device housing and is fixedly connected to one end of the inner roller of the first transmission belt.

[0034] Preferably, the wall-breaking mechanism includes a square funnel fixedly installed below one end of an adjustable conveying mechanism inside the device casing, a wall-breaking cylinder is installed at the lower end of the square funnel, the lower end of the wall-breaking cylinder is located between the device casing and is horizontally fixedly connected with a fixing plate, a wall-breaking motor is vertically installed inside the device casing at the lower end of the fixing plate, the output end of the wall-breaking motor passes through the fixing plate and is fixedly connected to the bottom of the rotating rod, the rotating rod is rotatably installed in the vertical direction of the middle of the wall-breaking cylinder, and a plurality of wall-breaking knives are fixedly connected vertically and equidistantly on the circumference of the rotating rod;

[0035] A heating rod is installed on the inner side of the wall of the wall-breaking cylinder, and a discharge port is equidistantly opened on one side of the lower end of the wall-breaking cylinder. The lower end of the discharge port is located between the device shell and is fixedly connected with a guide plate at an angle.

[0036] Preferably, the discharging mechanism includes a second transmission belt installed at the lower end of the guide plate, the second transmission belt is located at the outer end of the device housing and is installed with a second guard plate, and the second transmission belt is sleeved on a roller rotatably installed between the device housing and the second guard plate, and a first servo motor is installed on one side of the device housing, and the output end of the first servo motor passes through the device housing and is fixedly clamped at one end of the roller.

[0037] Preferably, a triangular drainage plate is installed on the upper end of the square funnel of the device shell, and drainage ports are provided on the device shell at both ends of the triangular drainage plate. The outside of the drainage ports are fixedly connected to a fixed frame, a storage box is inserted into the fixed frame, and through-holes are provided at equal intervals on the bottom plate of the fixed frame.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. This solution reduces the transfer process of medicinal slices and avoids material loss and contamination through the integrated design of adjustable conveying mechanism, wall breaking cylinder and discharge mechanism. At the same time, it physically links the conveying, wall breaking and discharge parameters to solve the problem of low efficiency caused by poor connection of traditional equipment.

[0040] 2. The intelligent collaborative system of this solution uses dynamic association models and reinforcement learning to adapt parameters such as conveying volume, heating power, and discharge speed in real time to avoid excessive or insufficient drying and improve the retention rate and stability of the effective ingredients in the medicinal slices.

[0041] 3. This solution's quality assessment unit accurately detects particle size distribution and moisture content, and dynamically adjusts parameters through a feedback mechanism. This increases the rate of high-quality products, reduces the amount of rework, lowers manual sorting costs, and addresses the pain point of poor quality stability.

[0042] In summary, this solution solves the core pain points of traditional equipment in terms of parameter matching, production efficiency, and quality stability through the combination of mechanical structure and intelligent collaborative system. It realizes the integrated collaborative operation and parameter adaptive optimization of Chinese herbal medicine wall breaking, drying, and packaging, significantly improving production efficiency and finished product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0044] Figure 1 This is a schematic structural diagram of Example 1 of an integrated production device for breaking, drying and packaging Chinese herbal medicine slices provided by the present invention;

[0045] Figure 2 This is a schematic diagram of the overall three-dimensional structure of the other side proposed by the present invention;

[0046] Figure 3 This is a schematic diagram of the overall three-dimensional structure proposed by the present invention when viewed from above;

[0047] Figure 4 This is a schematic diagram of the overall three-dimensional structure of the present invention when viewed from the bottom from another side;

[0048] Figure 5This is a schematic diagram of the partial and overall three-dimensional structure proposed by the present invention;

[0049] Figure 6 This is a schematic diagram of the overall three-dimensional structure of the wall-breaking device proposed in the present invention.

[0050] Figure 7 This is a schematic diagram of the overall three-dimensional structure of the wall-breaking device proposed by the present invention when viewed from above.

[0051] Figure 8 This is a schematic diagram of the cross-sectional structure of the wall breaking device proposed in the present invention;

[0052] Figure 9 This is a connection relationship diagram of the intelligent collaborative system of Example 2 of an integrated production device for breaking, drying and packaging Chinese herbal medicine slices provided by the present invention.

[0053] Serial numbers in the figure: 1-device housing; 2-first guard plate; 3-electric telescopic rod; 4-hinge rod; 5-first transmission belt; 6-fixed frame; 7-first servo motor; 8-second guard plate; 9-storage box; 10-square funnel; 11-second transmission belt; 12-triangular drainage plate; 13-wall-breaking cylinder; 14-material guide plate; 15-wall-breaking motor; 16-fixed plate; 17-discharge port; 18-wall-breaking knife; 19-second servo motor. DETAILED DESCRIPTION

[0054] The following is a further description of specific embodiments of the present invention in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is intended to facilitate understanding of the present invention and does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0055] Example: See Figures 1-8The present invention provides an integrated production device for breaking, drying and packaging Chinese herbal medicine slices, comprising a device shell 1, first guard plates 2 hingedly connected to both inner walls above one end of the device shell 1, an adjustable conveying mechanism installed between the first guard plates 2, a wall breaking mechanism installed vertically in the middle of the device shell 1 at one end of the adjustable conveying mechanism, and a discharge mechanism installed on one side of the lower end of the wall breaking mechanism. A basic frame is constructed by building the device shell 1 and the first guard plate 2, and in conjunction with the adjustable conveying mechanism, the wall breaking mechanism and the discharge mechanism, it is convenient to realize the integration of conveying, wall breaking and packaging of Chinese herbal medicine slices. Production; The adjustable conveying mechanism includes rollers rotatably installed at both ends between the first guard plates 2 on both sides, and a first transmission belt 5 is sleeved on the outside of the roller. The upper end of the first transmission belt 5 is fixedly connected to a barrier plate at equal distances, and the first guard plates 2 on both sides of the first transmission belt 5 are hinged with a hinged rod 4, and the other end of the hinged rod 4 is connected to the telescopic end of the electric telescopic rod 3 through a threaded connection. The other end of the electric telescopic rod 3 is hinged to one side of the device casing 1, and the rollers and the first transmission belt 5 between the first guard plates 2 cooperate with the hinged rod 4 and the electric telescopic rod 3 to facilitate flexible adjustment of the conveying angle.

[0056] In the present invention, a second servo motor 19 is installed on one side of the device housing 1, and the output end of the second servo motor 19 passes through the device housing 1 and is fixedly clamped on one end of the inner roller of the first transmission belt 5. It is connected to the inner roller of the first transmission belt 5 through the second servo motor 19 on one side of the device housing 1, so as to provide stable power for the conveying mechanism; the wall breaking mechanism includes a square funnel 10 fixedly installed below one end of the adjustable conveying mechanism inside the device housing 1, and a wall breaking cylinder 13 is installed at the lower end of the square funnel 10. The lower end of the wall breaking cylinder 13 is located between the device housing 1 and is fixedly connected and installed with a fixing plate 16. The lower end of the fixing plate 16 is vertically installed with a wall breaking motor 15 inside the device housing 1, and the output end of the wall breaking motor 15 passes through The fixing plate 16 is fixedly connected to the wall-breaking cylinder 13 at the bottom of the rotating rod, and the rotating rod is rotatably installed in the vertical direction of the middle part of the wall-breaking cylinder 13, and multiple wall-breaking knives 18 are fixedly connected vertically and equidistantly around the rotating rod. Through the cooperation of the square funnel 10, the wall-breaking cylinder 13, the wall-breaking motor 15, the rotating rod and the wall-breaking knife 18, it is convenient to achieve efficient wall breaking of Chinese herbal medicine slices; a heating rod is installed on the inner side of the wall of the wall-breaking cylinder 13, and a discharge port 17 is equidistantly opened on one side of the lower end of the wall-breaking cylinder 13. The lower end of the discharge port 17 is located between the device casing 1 and is obliquely fixedly connected to a guide plate 14. Through the heating rod, the discharge port 17 and the guide plate 14 on the inner side of the wall-breaking cylinder 13, it is convenient to achieve drying synchronously during the wall-breaking process and guide the discharge of materials.

[0057] In the present invention, the discharging mechanism includes a second transmission belt 11 installed at the lower end of the guide plate 14, the second transmission belt 11 is located at the outer end of the device shell 1 and is installed with a second guard plate 8, and the second transmission belt 11 is sleeved on the roller rotatably installed between the device shell 1 and the second guard plate 8, and a first servo motor 7 is installed on one side of the device shell 1, and the output end of the first servo motor 7 passes through the device shell 1 and is fixedly connected to one end of the roller, through the second transmission belt 11, the second guard plate 8 and the first servo motor 7 at the lower end of the guide plate 14, it is convenient to realize the packaging and transportation of the medicinal slices after wall breaking; the device shell 1 is located at the upper end of the square funnel 10 and is installed with a triangular drainage plate 12, and the device shell 1 at both ends of the triangular drainage plate 12 is provided with a drainage port, and the outside of the drainage port is fixedly connected with a fixed frame 6, and a storage box 9 is inserted and connected in the fixed frame 6, and the bottom plate of the fixed frame 6 is equidistantly provided with through ports, which are convenient for collecting waste liquid or impurities generated in the production process through the triangular drainage plate 12, the drainage port, the fixed frame 6 and the storage box 9.

[0058] Working principle of the present invention:

[0059] Device operation process:

[0060] Conveying adjustment: according to the demand for conveying the medicinal slices, start the electric telescopic rod 3 and adjust the inclination angle of the first transmission belt 5 through the hinged rod 4; start the second servo motor 19 to drive the first transmission belt 5 to run and convey the medicinal slices to the square funnel 10;

[0061] Wall breaking and drying: The pieces of medicine enter the wall breaking cylinder 13 through the square funnel 10, and the wall breaking motor 15 is started to drive the rotating rod and the wall breaking knife 18 to rotate to achieve wall breaking; at the same time, the heating rod inside the wall breaking cylinder 13 works to dry the pieces of medicine, and the generated water vapor enters the storage box 9 through the triangular drainage plate 12 and the drainage port for collection;

[0062] Discharging and packaging: The slices after wall breaking and drying slide to the second transmission belt 11 through the discharge port 17 and the guide plate 14. The first servo motor 7 is started to drive the second transmission belt 11 to run and transport the slices to the end for packaging.

[0063] Example 2

[0064] Please refer to Figure 9 As shown, based on the integrated production device for breaking, drying and packaging of Chinese herbal medicine slices provided in Example 1 of the present application, Example 2 of the present application proposes another integrated production device for breaking, drying and packaging of Chinese herbal medicine slices. Example 2 is only a preferred embodiment of Example 1, and the implementation of Example 2 will not affect the independent implementation of Example 1.

[0065] Specifically, the difference of the integrated production device for breaking, drying and packaging Chinese herbal medicine slices provided in Example 2 of the present application is that:

[0066] Also included is an intelligent collaborative system adapted to the device in Example 1, the intelligent collaborative system including a sensing acquisition module, an intelligent analysis module, and a collaborative control module;

[0067] The sensor collection module is used to collect the physical parameters and material status of each link of the device, including the delivery volume and pressure of the adjustable conveying mechanism, the broken wall particle size, the temperature and humidity in the cylinder of the wall breaking mechanism, the discharge speed and material accumulation status of the discharge mechanism;

[0068] The intelligent analysis module includes a condition monitoring and analysis unit, a fault diagnosis and early warning unit, a parameter optimization and decision-making unit, and a quality assessment and feedback unit;

[0069] The status monitoring and analysis unit is used to perform multi-dimensional analysis on the real-time data of the sensor acquisition module, establish a dynamic correlation model of conveying volume, wall breaking strength, drying temperature, and discharge speed, and output quantitative evaluation results of the operating status of each link;

[0070] The fault diagnosis and early warning unit identifies potential fault types based on the abnormal fluctuation characteristics of status data and combines it with the historical fault database, and determines whether to generate an early warning signal;

[0071] Parameter optimization and decision-making unit, used to integrate the characteristic parameters of Chinese herbal medicine slices with real-time operation data, and generate adaptive control strategies through reinforcement learning algorithms;

[0072] The quality assessment and feedback unit evaluates the quality grade of the finished product based on the particle size distribution after wall breaking and the moisture content after drying, and feeds the results back to the parameter optimization unit;

[0073] The collaborative control module is used to drive the corresponding execution equipment according to the real-time control parameters output by the intelligent analysis module; and dynamically revise the control strategy of the next batch based on the quality assessment report.

[0074] In this application, the specific analysis steps of the condition monitoring and analysis unit include:

[0075] Data preprocessing: noise reduction is performed on the conveying volume, pressure, particle size, temperature, humidity, discharge speed and stacking status data obtained by the sensor acquisition module, abnormal jump values ​​are eliminated, and the data are converted into unified dimension parameters through data standardization;

[0076] Feature extraction: extract four core features from preprocessed data and construct a multi-dimensional feature set The four core characteristics include the time series fluctuation characteristics of the conveying volume, the distribution characteristics of the broken particle size, the gradient characteristics of the drying temperature field, and the coupling characteristics of the discharge speed and the stacking state:

[0077] Among them, the fluctuation characteristics of the delivery time series : Calculate the standard deviation of the delivery volume within the set number of sampling periods and volatility coefficient ,in represents the mean value of the transport volume during this period, where is the standardized value of the delivery volume, is the mean value of the transport volume; As the characteristic vector of the conveying link;

[0078] Distribution characteristics of broken particle size : Detect the particle size data after wall breaking and divide the particle size range , if k1≥5, set according to the particle size requirements of the quality standard and calculate:

[0079] Particle ratio in each interval , represents the number of particles in the jth interval, and N is the total number of particles;

[0080] Particle distribution entropy , the lower the entropy value, the more concentrated the particle distribution;

[0081] by As the characteristic vector of the wall-breaking link;

[0082] Drying temperature field gradient characteristics : Collect temperature sensor data at different locations in the drying cylinder (such as top, middle, and bottom) ,calculate:

[0083] Temperature gradient , where h represents the sensor spacing and is the normalized value;

[0084] Temperature uniformity ,in, 、 Represent the maximum and minimum values ​​of temperature respectively;

[0085] by As the characteristic vector of the drying stage;

[0086] Coupling characteristics of discharge speed and accumulation state : Combine the discharge speed v and the stacking area S (or height) data detected by the stacking state to calculate:

[0087] Coupling coefficient , represents the normalized value of the stacking area, so that dimensionless;

[0088] Accumulation change rate , represents the accumulation area at the rth moment;

[0089] by As the characteristic vector of the nesting link;

[0090] Dynamic association modeling, based on the extracted multi-dimensional feature set Construct a multi-dimensional dynamic association model M, specifically including:

[0091] Mapping relationship between transport volume and wall breaking strength: based on transport volume characteristics As input, the wall-breaking granularity feature The index that is strongly correlated with the wall breaking strength is used as the output, and the mapping model is constructed using multiple linear regression: ;in 、 、 is the regression coefficient (determined by fitting historical data), is the residual, realizing the quantitative correlation between the transport parameters and the wall breaking effect;

[0092] Drying temperature-discharge speed response relationship: based on the drying temperature characteristics As input, discharge speed v is output, and the response relationship is constructed using the transfer function model: , where K is the gain coefficient (the influence of temperature change on discharge speed), is the pure lag time (temperature conduction delay), T is the time constant (response rate to temperature change), and the model parameters are fitted through system identification (such as step response method) to quantify the dynamic effect of drying temperature on discharge speed;

[0093] Multi-parameter cross-influence function: Four types of features Concatenate into high-dimensional feature vectors , Gaussian process regression is used to construct the cross-influence function: ,in is the mean function (usually set to zero mean), is the covariance function (such as the squared exponential function is the signal variance, is the length scale), used to measure the two input and The similarities between Represents the predicted output of the equipment operating status, express The distribution of accords with the statistical law of the Gaussian process on the right. Represents a Gaussian process; by maximizing the log-likelihood to estimate hyperparameters, the nonlinear correlation between multi-link parameters is quantitatively expressed, forming a multi-dimensional dynamic correlation model M;

[0094] The quantitative evaluation results are output through the multi-dimensional dynamic correlation model M, which outputs three types of quantitative evaluation indicators and integrates them to form the quantitative evaluation results of the operating status of each link, including motion stability index, parameter matching coefficient, and efficiency loss rate. Specifically:

[0095] Calculate the coefficient of variation of characteristic parameters of each link , the stability index is constructed using the weighted average method , the formula is: , is the weight coefficient of the s-th feature and ,Specifically set according to the degree of quality impact and process complexity, the operation stability index range is [0, 1]. The closer the value is to 1, the higher the stability;

[0096] Compare the actual correlation relationship of each parameter in the dynamic correlation model M with the historical optimal correlation relationship, and use cosine similarity to calculate the matching coefficient , the formula is: ,in is the current parameter associated feature, is the historical optimal correlation feature, , the closer the value is to 1, the higher the parameter matching degree is;

[0097] Based on device energy consumption data Theoretical minimum energy consumption , calculate the efficiency loss rate , the formula is: , , the closer the value is to 1, the lower the performance loss;

[0098] Will 、 、 The characteristic vectors of each link and the output values ​​of the correlation model are integrated into the structured evaluation results;

[0099] The specific analysis steps of the fault diagnosis and early warning unit described in this application include:

[0100] Abnormal feature extraction: from the quantitative evaluation results output by the condition monitoring and analysis unit, identify the characteristic parameters that exceed the normal fluctuation range, extract the abnormal parameter change rate, duration and the coordinated variation characteristics of the associated parameters, and form the current abnormal feature vector , where n is the dimension of abnormal feature parameters, It represents the normalized value of the ith abnormal characteristic parameter, which is in the range of [0, 1] after data preprocessing. The larger the value, the higher the degree of abnormality.

[0101] Fault pattern matching: calculate the similarity between the extracted abnormal features and the typical fault pattern feature vectors stored in the historical fault database to determine the fault type with the highest matching degree;

[0102] The typical failure mode is constructed in the following way:

[0103] Feature extraction is performed on several batches of production data and several simulated fault test data accumulated from the device operation. Feature parameters in the scenarios of material blockage, blade wear, uneven heating, and conveying jam are collected (such as the discharge pressure change rate and speed attenuation coefficient corresponding to material blockage; the standard deviation of the current fluctuation of the wall-breaking motor 15 and the cutting force spectrum offset corresponding to blade wear); after dimensionality reduction through principal component analysis (PCA), a standardized feature vector library is formed. , where m is the total number of typical failure modes, The characteristic vector representing the k2th type of typical fault is expressed as , is the standard value of the g-th characteristic parameter in the k2-th type fault;

[0104] The cosine similarity algorithm is used to calculate the similarity between the current abnormal feature vector and each pattern in the feature vector library. The calculation method is: vector A and vector The dot product of the vector A and the vector The product of the modulus of The dot product is , the modulus of vector A is the square root of ,vector The modulus length is the square root of , and record the calculation result as the similarity value , the value range is [0, 1]. The closer the similarity value is to 1, the higher the matching degree is. The fault mode corresponding to the maximum similarity value is selected as the fault type with the highest matching degree.

[0105] Early warning judgment, calculating the probability of fault occurrence based on matching results , a probability model is constructed by integrating similarity and historical frequency. The formula is: ;in, represents the weight coefficient used to balance the real-time similarity and the historical fault frequency, Indicates the The actual occurrence frequency of the similar failure mode in historical data is given by ,in is the number of actual failures that occurred after matching the pattern in history, is the total number of matches for the pattern;

[0106] Combine the fault impact level to construct the coupling value Assess the failure risk, the formula is , where L represents the quantitative value of the fault impact level, which can be divided into slight impact (L=1), performance degradation (L=2), and equipment damage (L=3). Indicates the weight coefficient corresponding to the level, which can be set based on the expert scoring method and fault loss analysis. The fault impact levels include minor impact, performance degradation, and equipment damage.

[0107] When the coupling value C is greater than or equal to the preset trigger threshold, an early warning signal including the fault location, type and preset recommended processing method is generated.

[0108] Among them, the preset suggested processing methods are based on the fault type, location and historical solutions to build a strategy library. For example: for minor faults (such as short-term fluctuations in conveying volume), push notifications of inspections with status feedback and continuous observation at set times; for efficiency degradation faults (such as abnormal drying temperature gradients), automatically trigger dynamic fine-tuning of heating power or parameter adaptation solutions for switching redundant sensing channels; for equipment damage faults (such as blade wear of the wall-breaking knife 18), immediately generate maintenance work orders for shutdown inspections and replacement of designated spare parts, and simultaneously link them to spare parts inventory warnings. All suggestions are pushed to the central control system and operation and maintenance terminals in real time, achieving standardized fault handling and immediate response.

[0109] The dynamic fine-tuning of the heating power can be calculated through a heat balance formula based on the drying chamber volume, the density of the medicinal material, and the specific heat capacity.

[0110] The specific analysis steps of the parameter optimization and decision-making unit described in this application include:

[0111] The characteristic parameters of Chinese herbal medicine slices are integrated with the real-time operation parameters output by the status monitoring and analysis unit at the feature layer, specifically including:

[0112] Identify the characteristic parameters of Chinese herbal medicine slices, including initial moisture content , hardness H, component ratio vector , k3 is the number of characteristic components, is the mass proportion of the fth component, satisfying , forming the characteristic parameter vector of the decoction piece ;

[0113] Obtain real-time operating parameters, including delivery volume Q, wall breaking strength , drying temperature field gradient , discharge speed v, constitute the operating parameter vector ;

[0114] The characteristic parameters and operating parameters are Z-score standardized respectively, and then the comprehensive feature matrix X is constructed by feature splicing, that is: ,in Represents vector concatenation operation;

[0115] Taking the comprehensive feature matrix X as input, the initial adaptive control strategy is generated by the reinforcement learning algorithm, which includes:

[0116] Obtain the cumulative production data of several batches of equipment and extract the control parameter combination , and the corresponding performance benefit Y, the sample with the set ratio before screening the performance benefit is constructed into a strategy library ;in is the conveying motor speed, The wall breaking motor has a power of 15. is the drying heating power, is the opening of the discharge valve;

[0117] Use deep Q network algorithm to build a policy network Output control parameter probability distribution, value network Evaluate the strategy benefits, the objective function is to maximize the cumulative benefits: ,in , 2. 2. 2 are the weight coefficients corresponding to the wall breaking rate, drying uniformity and energy consumption respectively;

[0118] By using the strategy library Interactive iteration: when the cumulative profit J converges to a preset profit ratio above the historical optimal value, the initial control strategy is output;

[0119] Substitute the initial control strategy into the dynamic correlation model for simulation verification, including:

[0120] Parameter matching evaluation, calculating the matching degree between control parameters and physical constraints of the equipment: ,in To control the upper and lower limits of parameters, , , must meet ;

[0121] Target achievement assessment: Calculate the degree of fit between simulation results and quality goals: ,in For simulation performance gain, is the target value, is the penalty coefficient, , must meet ;

[0122] Bias correction, if or If any value in the is not up to standard, the control parameter out of limit is corrected by projected gradient descent method, or the reinforcement learning weight coefficient is adjusted. 、 2. Iterate again and finally output the final control strategy that meets the conditions.

[0123] The specific analysis steps of the quality assessment and feedback unit described in this application include:

[0124] Extract quality parameters and perform statistical analysis on the particle size distribution data after wall breaking to extract the particle size distribution uniformity and target particle size ratio:

[0125] The number of particles identified within the preset target particle size range is recorded as , and then get the total number of particles , calculate the target particle size ratio by the number of particles and the total number of particles , the formula is ;

[0126] The coefficient of variation is used to characterize and calculate the target particle size ratio, and the formula is: ,in is the standard deviation of particle size distribution, is the mean particle size;

[0127] Identify the number of test points k4 and obtain the moisture content of the test points , d1 represents the number of the test point, and the average moisture content is calculated by the number of test points and the moisture content. The formula is: ;

[0128] Then obtain the maximum and minimum values ​​of the moisture content, and record the difference between the maximum and minimum values ​​of the moisture content as the moisture content range;

[0129] Perform spatial distribution analysis on moisture content data after drying to extract average moisture content and moisture content range;

[0130] Based on the preset quality standards, weighted comprehensive score The judgment level is: ,in The maximum allowable moisture content is The maximum allowable moisture content range;

[0131] Preset quality standard value range groups, including the corresponding value ranges for excellent, qualified, and to be reworked; match the comprehensive score with the quality standard value range group to output the corresponding quality grade label and details of non-compliant items;

[0132] The quality assessment results are converted into the reward and punishment coefficients of reinforcement learning. The formula is: ;in is the adjustment amount of the reinforcement learning weight coefficient in the parameter optimization and decision-making unit, is the correction factor, is the target quality score, when S> hour, A positive value indicates a reward, which enhances the corresponding strategy weight; when S≤ hour, A negative value indicates punishment, which reduces the weight of the corresponding strategy and achieves closed-loop correction.

[0133] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It is apparent to those skilled in the art that various changes, modifications, substitutions, and variations of these embodiments may be made without departing from the principles and spirit of the present invention, and these changes and modifications still fall within the scope of protection of the present invention.

Claims

1. A device for the integrated production of Chinese herbal medicine pieces by wall breaking, drying and packaging, comprising a device housing (1), characterized in that: First guard plates (2) are hingedly connected to the inner walls on both sides above one end of the device housing (1), an adjustable conveying mechanism is installed between the first guard plates (2), a wall breaking mechanism is vertically installed in the middle of the device housing (1) at one end of the adjustable conveying mechanism, and a discharge mechanism is installed on one side of the lower end of the wall breaking mechanism; It also includes an intelligent collaborative system adapted to each mechanism in the device, and the intelligent collaborative system includes a sensing acquisition module, an intelligent analysis module, and a collaborative control module; The sensor collection module is used to collect the physical parameters and material status of each link of the device, including the delivery volume and pressure of the adjustable conveying mechanism, the broken wall particle size, the temperature and humidity in the cylinder of the wall breaking mechanism, the discharge speed and material accumulation status of the discharge mechanism; The intelligent analysis module is used to process and analyze the real-time data obtained by the sensor acquisition module, including multi-dimensional analysis to establish a dynamic correlation model to evaluate the operating status of each link, identify potential faults and determine whether to issue an early warning based on the historical fault database, integrate the characteristics of the medicinal materials with real-time data to generate an adaptive control strategy, and evaluate the quality level of the finished product and provide feedback; The collaborative control module is used to drive the corresponding execution equipment according to the real-time control parameters output by the intelligent analysis module; and dynamically revise the control strategy of the next batch based on the quality assessment report.

2. The integrated production device for breaking, drying and packaging Chinese herbal medicine slices according to claim 1, characterized in that: The intelligent analysis module specifically includes a status monitoring and analysis unit, a fault diagnosis and early warning unit, a parameter optimization and decision-making unit, and a quality assessment and feedback unit; The status monitoring and analysis unit is used to perform multi-dimensional analysis on the real-time data of the sensor acquisition module, establish a dynamic correlation model of conveying volume, wall breaking strength, drying temperature, and discharge speed, and output quantitative evaluation results of the operating status of each link; The fault diagnosis and early warning unit identifies potential fault types based on the abnormal fluctuation characteristics of status data and combines it with the historical fault database, and determines whether to generate an early warning signal; Parameter optimization and decision-making unit, used to integrate the characteristic parameters of Chinese herbal medicine slices with real-time operation data, and generate adaptive control strategies through reinforcement learning algorithms; The quality assessment and feedback unit evaluates the quality grade of the finished product based on the particle size distribution after wall breaking and the moisture content after drying, and feeds the results back to the parameter optimization unit.

3. The integrated production device for wall-breaking, drying and packaging of Chinese herbal medicine slices according to claim 2, characterized in that: The specific analysis steps of the condition monitoring and analysis unit include: Data preprocessing: noise reduction is performed on the conveying volume, pressure, particle size, temperature, humidity, discharge speed and stacking status data obtained by the sensor acquisition module, abnormal jump values ​​are eliminated, and the data are converted into unified dimension parameters through data standardization; Feature extraction: extract the time series fluctuation characteristics of the conveying volume, the distribution characteristics of the broken particle size, the gradient characteristics of the drying temperature field, and the coupling characteristics of the discharge speed and stacking state from the preprocessed data; Dynamic correlation modeling: Based on the extracted features, the mapping relationship between conveying volume and wall breaking strength, the response relationship between drying temperature and discharge speed, and the cross-influence function between various parameters are constructed to form a multi-dimensional dynamic correlation model; Quantitative evaluation is generated by outputting the operation stability index, parameter matching coefficient and efficiency loss rate of each link through a multi-dimensional dynamic correlation model, and integrating them to form a quantitative evaluation result of the operation status of each link.

4. The integrated production device for wall-breaking, drying and packaging of Chinese herbal medicine slices according to claim 2, characterized in that: The specific analysis steps of the fault diagnosis and early warning unit include: Abnormal feature extraction: From the quantitative evaluation results output by the condition monitoring and analysis unit, identify characteristic parameters that exceed the normal fluctuation range, and extract the abnormal parameter change rate, duration, and coordinated variation characteristics of related parameters; Fault pattern matching: calculate the similarity between the extracted abnormal features and the typical fault pattern feature vectors stored in the historical fault database to determine the fault type with the highest matching degree; Early warning judgment calculates the probability of fault occurrence based on the matching results and combines it with the fault impact level. When the coupled value of the probability and level reaches the preset trigger condition, an early warning signal is generated containing the fault location, type, and recommended handling method; the fault impact level includes minor impact, performance degradation, and equipment damage.

5. The integrated production device for wall-breaking, drying and packaging of Chinese herbal medicine slices according to claim 2, characterized in that: The specific analysis steps of the parameter optimization and decision-making unit include: Feature fusion: The characteristic parameters of Chinese herbal medicine slices are integrated with the real-time operation parameters output by the status monitoring and analysis unit at the feature level to construct a comprehensive feature matrix; Strategy generation, taking the comprehensive feature matrix as input, calls the historical optimal control strategy library through the reinforcement learning algorithm, iteratively calculates the efficiency benefits under different control parameter combinations, including wall breaking rate, drying uniformity, and energy consumption, and generates the initial adaptive control strategy; Strategy verification: Substitute the initial strategy into the dynamic association model for simulation verification, evaluate the parameter matching and goal achievement after the strategy is implemented, correct the deviation items, and form the final control strategy.

6. The integrated production device for breaking, drying and packaging Chinese herbal medicine slices according to claim 2, characterized in that: The specific analysis steps of the quality assessment and feedback unit include: Quality parameter extraction: statistical analysis of particle size distribution data after wall breaking to extract particle size distribution uniformity and target particle size ratio; spatial distribution analysis of moisture content data after drying to extract average moisture content and moisture content range; Grading: Based on the particle size distribution index and moisture content index, the grade is determined by comparing it with the preset quality standard value range group, generating a quality grade label and detailed list of non-compliant items; the preset quality standard value range group includes excellent, qualified, and rework value ranges; Feedback correction: The quality grade label, details of non-compliant items, and corresponding real-time operating parameters are fed back to the parameter optimization and decision-making unit as penalty or reward items when generating the next batch of strategies, adjusting the weight coefficient of the reinforcement learning algorithm to achieve closed-loop correction.

7. The integrated production device for breaking, drying and packaging Chinese herbal medicine slices according to claim 1, characterized in that: The adjustable conveying mechanism comprises rollers rotatably mounted at both ends between the first guard plates (2) on both sides, a first transmission belt (5) is sleeved on the outer side of the roller, the upper end of the first transmission belt (5) is fixedly connected to a blocking plate at equal intervals, and a hinged rod (4) is hinged on the first guard plates (2) on both sides of the first transmission belt (5), the other end of the hinged rod (4) is connected to the telescopic end of the electric telescopic rod (3) by a thread, and the other end of the electric telescopic rod (3) is hinged to one side of the device housing (1); A second servo motor (19) is installed on one side of the device housing (1), and an output end of the second servo motor (19) passes through the device housing (1) and is fixedly connected to one end of the inner roller of the first transmission belt (5).

8. The integrated production device for breaking, drying and packaging Chinese herbal medicine slices according to claim 1, characterized in that: The wall breaking mechanism comprises a square funnel (10) fixedly mounted below one end of an adjustable conveying mechanism inside the device housing (1); a wall breaking cylinder (13) is mounted at the lower end of the square funnel (10); the lower end of the wall breaking cylinder (13) is located between the device housing (1) and is horizontally fixedly connected to a fixing plate (16); a wall breaking motor (15) is vertically mounted inside the device housing (1) at the lower end of the fixing plate (16); the output end of the wall breaking motor (15) passes through the fixing plate (16) and is fixedly connected to the wall breaking cylinder (13) at the bottom of a rotating rod; the rotating rod is rotatably mounted in the middle of the wall breaking cylinder (13) in a vertical direction, and a plurality of wall breaking knives (18) are fixedly connected vertically and equidistantly around the rotating rod; A heating rod is installed on the inner side of the wall of the wall-breaking cylinder (13), and a discharge port (17) is equidistantly provided on one side of the lower end of the wall-breaking cylinder (13). The lower end of the discharge port (17) is located between the device housing (1) and is fixedly connected to a guide plate (14).

9. The integrated production device for breaking, drying and packaging Chinese herbal medicine slices according to claim 1, characterized in that: The discharging mechanism includes a second transmission belt (11) installed at the lower end of the guide plate (14), the second transmission belt (11) is located at the outer end of the device housing (1) and is installed with a second guard plate (8), and the second transmission belt (11) is sleeved on a roller rotatably installed between the device housing (1) and the second guard plate (8), and a first servo motor (7) is installed on one side of the device housing (1), and the output end of the first servo motor (7) passes through the device housing (1) and is fixedly connected to one end of the roller.

10. The integrated production device for breaking, drying and packaging Chinese herbal medicine slices according to claim 8, characterized in that: The device housing (1) is located at the upper end of the square funnel (10) and is provided with a triangular drainage plate (12). Liquid discharge ports are provided on the device housing (1) at both ends of the triangular drainage plate (12). The outsides of the liquid discharge ports are fixedly connected to a fixed frame (6). A storage box (9) is inserted and connected to the fixed frame (6), and a bottom plate of the fixed frame (6) is provided with through-ports at equal intervals.

Images