Intelligent traditional Chinese medicine ingredient blending system

Abstract

The invention relates to the field of traditional Chinese medicine intelligent manufacturing, in particular to an intelligent traditional Chinese medicine ingredient blending system which comprises a data acquisition module, a feature construction module, an ingredient knowledge modeling module, an intelligent blending decision module, a blending execution module and a monitoring feedback module. By constructing a component association relationship model, a component cooperation relationship model and an influence relationship model between components and blending process parameters, the compatibility relationship between the traditional Chinese medicine components and blending process factors are incorporated into a unified modeling framework, so that the component compatibility relationship is converted into a modeling and analyzable component knowledge system from experience judgment; according to the invention, in the intelligent blending decision-making module, the component knowledge model is used as a core basis of blending decision-making, and the traditional Chinese medicine component blending scheme is systematically reasoned and screened under the constraint of the blending target and the constraint condition, so that the intelligent generation of the blending scheme is realized, and the scientificity and controllability of the traditional Chinese medicine component blending process are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing of traditional Chinese medicine, specifically to an intelligent system for dispensing traditional Chinese medicine ingredients. Background Technology

[0002] The formulation of traditional Chinese medicine (TCM) preparations is a crucial step in the TCM production process, directly impacting the quality stability and efficacy consistency of the products. Current TCM formulation methods largely rely on manual experience or fixed formulas, making it difficult to fully consider the compatibility relationships between different TCM components and the influence of formulation process parameters on component properties during actual production. Furthermore, existing TCM formulation process management typically focuses on controlling single components or single process parameters, lacking systematic modeling and comprehensive analysis of the synergistic effects of multiple components. This results in poor consistency and controllability of formulation results across different production batches or under different formulation conditions. Simultaneously, existing TCM formulation schemes are mostly statically set during implementation, making it difficult to adjust formulation strategies promptly based on results. The lack of an effective feedback mechanism limits the intelligence and precision of the TCM formulation process. Summary of the Invention

[0003] To address the above issues and overcome the shortcomings of existing technologies, this invention provides an intelligent traditional Chinese medicine (TCM) ingredient blending system. In its ingredient knowledge modeling module, this invention constructs ingredient correlation models, ingredient synergy models, and influence relationship models between ingredients and blending process parameters. This integrates the compatibility relationships between TCM ingredients and blending process factors into a unified modeling framework, transforming ingredient compatibility relationships from empirical judgments into a modelable and analyzable ingredient knowledge system. Simultaneously, in its intelligent blending decision-making module, this invention uses the aforementioned ingredient knowledge model as the core basis for blending decisions. Under the constraints of blending objectives and conditions, it systematically reasones and filters TCM ingredient blending schemes, achieving intelligent generation of blending schemes and improving the scientific rigor and controllability of the TCM ingredient blending process.

[0004] The technical solution adopted in this invention is as follows: This invention provides an intelligent traditional Chinese medicine ingredient dispensing system, comprising a data acquisition module, a feature construction module, an ingredient knowledge modeling module, an intelligent dispensing decision module, a dispensing execution module, and a monitoring and feedback module, specifically including the following:

[0005] The data acquisition module collects multi-source data involved in the formulation of traditional Chinese medicine ingredients. The multi-source data includes attribute data of Chinese medicinal materials, component detection data, historical compatibility data, and process parameter data during the formulation process, forming the original dataset of traditional Chinese medicine ingredient formulation.

[0006] The feature construction module preprocesses and constructs features from the original dataset. The preprocessing includes outlier removal and data standardization. Based on this, multidimensional feature information reflecting the characteristics of Chinese medicine components, component compatibility relationships, and the influence of process parameter data is extracted to obtain the feature dataset.

[0007] The component knowledge modeling module constructs a traditional Chinese medicine component compatibility model based on the feature dataset, including a component correlation model, a component synergy model, and an influence relationship model between components and process parameter data;

[0008] The intelligent dispensing decision module sets dispensing targets and constraints, and combines them with a traditional Chinese medicine component compatibility model to perform intelligent reasoning on the dispensing scheme of traditional Chinese medicine components, generating a dispensing scheme of traditional Chinese medicine components and dispensing ratio parameters of each traditional Chinese medicine component.

[0009] The dispensing execution module generates corresponding dispensing control instructions based on the dispensing scheme of traditional Chinese medicine ingredients and the dispensing ratio parameters of each traditional Chinese medicine ingredient, and automatically controls the dosage, dispensing order and process parameters of the traditional Chinese medicine ingredients during the dispensing process.

[0010] The monitoring and feedback module monitors the blending process and the blending results, acquires the blending result data, and feeds the blending result data back to the component knowledge modeling module and the intelligent blending decision module.

[0011] Furthermore, the component knowledge modeling module constructs a compatibility model of traditional Chinese medicine components, including a component correlation model, a component synergy model, and a model of the influence relationship between components and preparation process parameters. Specifically, this includes the following steps:

[0012] Step S1: Construction of multi-scale feature representation of components. Based on the feature dataset, multi-scale modeling is performed on the content variation characteristics, stability characteristics, and response characteristics of traditional Chinese medicine components under different blending process conditions. The component feature representation vector is constructed as follows:

[0013] ;

[0014] in, Indicates the first Each Chinese medicine ingredient The component feature representation vector, Indicates the first Traditional Chinese medicine components at various scales eigencomponents, The number of feature scales;

[0015] Step S2: Adaptive learning of component associations: Based on the component feature representation vector, calculate the association between any two Chinese herbal medicine components. and The association strength in a given allocation scenario is defined as follows:

[0016] ;

[0017] in, Indicates Chinese medicine ingredients With Chinese medicine ingredients The strength of the correlation between them Indicates the first Each component feature representation vector;

[0018] Step S3: Nonlinear modeling of component synergistic relationship model. Based on the component correlation model, this step is applied to the model of synergistic relationships among multiple traditional Chinese medicine components. The constituent set is constructed, and the proportions of each Chinese herbal medicine component are used as input variables. A nonlinear model of the synergistic relationship between the constituents under different proportions is performed, and the synergistic effect function is used to represent it, as defined below:

[0019] ;

[0020] in, Indicates ingredients The proportions of ingredients in the combination Indicates ingredients The single-component contribution weights, They represent the components determined based on the component association model. and ingredients The collaborative weights between them It is a nonlinear mapping function;

[0021] Step S4: Modeling the coupling influence between component and process parameter data. Based on the component feature representation vector and process parameter data, a model of the influence relationship between component and process parameter data is constructed. The coupling influence function is used to represent the influence relationship model, defined as follows:

[0022] ;

[0023] in, Indicates the effect of process parameter data on composition The component features represent the degree of influence of at least one feature component in the vector. For coupling mapping functions, This represents process parameter data;

[0024] Step S5: Dynamic update of the compatibility relationship model. Based on the blending result data obtained by the monitoring and feedback module, the component correlation relationship model is iteratively updated.

[0025] Furthermore, step S3 specifically includes the following steps:

[0026] Step S31: Determine the collaborative modeling object based on the component association model, targeting multiple traditional Chinese medicine components. The component set is constructed, and a threshold condition for component association strength is set. Multiple Chinese medicine components whose association strength meets the preset component association strength threshold condition are selected to form a component set for collaborative modeling.

[0027] Step S32: Construction of component synergy input variables. For the component set used for synergy modeling, extract the mixing ratio parameters of each Chinese herbal medicine component in the current mixing scheme, and construct the component synergy input vector, as shown below:

[0028] ;

[0029] in, Indicates ingredients The proportions of the ingredients in the set of components;

[0030] Step S33: Determining synergistic weights based on association constraints. Based on the strength of component associations, constraints are set on the synergistic weights of the components to ensure that the synergistic weights of the traditional Chinese medicine components are determined. With Chinese medicine ingredients Collaborative weights between The adjusted collaborative weights are obtained by adaptively adjusting the weights according to changes in the association strength.

[0031] Step S34: Nonlinear mapping modeling of component synergy effect. Based on the component synergy input vector and the adjusted synergy weights, construct the component synergy effect function and obtain the synergy effect evaluation result as the output result.

[0032] Step S35: Model output of synergistic effect results. The output results of the component synergistic effect function are used as the representation results of the component synergistic relationship model and sent to the intelligent allocation decision module.

[0033] Furthermore, the intelligent dispensing decision module sets dispensing targets and constraints, and combines a traditional Chinese medicine (TCM) component compatibility model to intelligently reason about the dispensing scheme of TCM components, generating a TCM component dispensing scheme and dispensing ratio parameters for each TCM component. Specifically, this includes the following steps:

[0034] Step D1: Setting the blending objectives and constraints. Based on the blending requirements of traditional Chinese medicine components, set the blending objectives and simultaneously set the corresponding constraints. The blending objectives include the objectives of enhancing the synergistic effect of components, ensuring the stability of key components, and controlling the blending process. The constraints include the blending ratio range constraints of each traditional Chinese medicine component, the total amount of components constraints, and the feasible range constraints of the blending process parameters.

[0035] Step D2: Construction of the candidate set of dispensing schemes. Under the premise of meeting the constraints, based on the Chinese herbal medicine ingredients participating in the current dispensing task, the dispensing ratio of each Chinese herbal medicine ingredient is combined to form multiple feasible dispensing schemes of Chinese herbal medicine ingredients, which serve as the candidate dispensing scheme set.

[0036] Step D3: Scheme evaluation based on the component matching relationship model. For the set of candidate allocation schemes, the component matching relationship model is called to evaluate the candidate allocation schemes and obtain evaluation results that reflect the characteristics of different allocation schemes.

[0037] Step D4: Scheme screening based on process parameter constraints. In combination with the constraints, the feasibility of the candidate allocation scheme set is judged, and candidate allocation schemes that do not meet the constraints are eliminated.

[0038] Step D5: Determination of the blending scheme. After step D4, the remaining candidate blending schemes are obtained. From the remaining candidate blending schemes, the blending scheme of Chinese medicine components that meets the blending target is determined, and the corresponding blending ratio parameters of each Chinese medicine component are output.

[0039] Step D6: Output the blending scheme. Output the blending scheme of Chinese herbal ingredients that meets the blending target and the corresponding blending ratio parameters of each Chinese herbal ingredient to the blending execution module to generate blending control instructions.

[0040] The beneficial effects achieved by the present invention using the above solution are as follows:

[0041] (1) In the component knowledge modeling module, this invention constructs a component association model, a component synergy model, and an influence relationship model between components and preparation process parameters, thereby incorporating the compatibility relationship between Chinese medicine components and preparation process factors into a unified modeling framework, so that the component compatibility relationship is transformed from empirical judgment into a modelable and analyzable component knowledge system.

[0042] (2) In the intelligent dispensing decision module, the present invention uses the component knowledge model as the core basis for dispensing decision. Under the constraints of dispensing objectives and constraints, it systematically reasons and screens the dispensing schemes of traditional Chinese medicine components, realizes the intelligent generation of dispensing schemes, and improves the scientificity and controllability of the dispensing process of traditional Chinese medicine components. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of an intelligent traditional Chinese medicine ingredient dispensing system proposed in this invention.

[0044] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0045] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0046] Example 1, see Figure 1 The present invention provides an intelligent traditional Chinese medicine ingredient dispensing system, comprising a data acquisition module, a feature construction module, an ingredient knowledge modeling module, an intelligent dispensing decision module, a dispensing execution module, and a monitoring and feedback module, specifically including the following:

[0047] In this embodiment, the traditional Chinese medicine preparation includes ingredients A, B, and C, and the above three traditional Chinese medicine ingredients participate in the preparation process of the same traditional Chinese medicine preparation.

[0048] Before and during the blending process, the data acquisition module collects the medicinal herb attribute data, component detection data, and historical compatibility data corresponding to component A, component B, and component C. At the same time, it collects the process parameter data involved in the blending process to form the original dataset of medicinal herb component blending for this blending.

[0049] The feature construction module receives the original dataset, performs abnormal data removal and data standardization on the collected data, and extracts multi-dimensional feature information reflecting the component characteristics of component A, component B and component C, the compatibility relationship between components and the influence of process parameters to construct a feature dataset.

[0050] Based on the feature dataset, the component knowledge modeling module constructs a component association model, a component synergy model, and an influence model between components and process parameter data for components A, B, and C. These models are used to characterize the compatibility characteristics of the three traditional Chinese medicine components under different blending ratios and different process parameter conditions.

[0051] The intelligent dispensing decision module sets dispensing targets and constraints based on the dispensing requirements of this traditional Chinese medicine preparation. Under the constraints of the dispensing targets and constraints, and in conjunction with the traditional Chinese medicine component compatibility relationship model constructed by the component knowledge modeling module, it performs intelligent reasoning on the dispensing ratio of component A, component B and component C, generates a traditional Chinese medicine component dispensing scheme for this dispensing, and outputs the dispensing ratio parameters corresponding to component A, component B and component C.

[0052] The dispensing execution module receives the dispensing scheme of the traditional Chinese medicine ingredients and the dispensing ratio parameters of each traditional Chinese medicine ingredient, generates corresponding dispensing control instructions, and automatically controls the dosage, dispensing order and process parameters of ingredient A, ingredient B and ingredient C and the dispensing process according to the dispensing control instructions to complete the dispensing operation of the traditional Chinese medicine preparation.

[0053] After the formulation is completed, the monitoring and feedback module monitors the formulation results of this traditional Chinese medicine component formulation, obtains the formulation result data, and feeds the formulation result data back to the component knowledge modeling module and the intelligent formulation decision module for subsequent updates of the traditional Chinese medicine component compatibility relationship model and optimization of the formulation decision strategy.

[0054] Example 2, based on the above examples, further describes how the component knowledge modeling module constructs a model of the compatibility relationship between traditional Chinese medicine components, including a component correlation model, a component synergistic relationship model, and a model of the influence relationship between components and formulation process parameters. Specifically, this includes the following steps:

[0055] Step S1: Based on the component detection data, historical compatibility data, and blending process parameter data obtained by the data acquisition module, the feature construction module analyzes the content changes, stability changes, and response to changes in process parameters of components A, B, and C under different blending process conditions, and models them at different feature scales. The feature components of each component are extracted at multiple scales. On this basis, the component knowledge modeling module constructs component feature representation vectors for components A, B, and C respectively. Each component feature representation vector is composed of feature components at multiple scales, which are used to characterize the comprehensive characteristics of the corresponding Chinese medicine components under different blending process conditions.

[0056] Step S2: By calculating the similarity of the component feature representation vectors of component A and component B, component A and component C, and component B and component C, the corresponding correlation strength is obtained, which reflects the degree of compatibility between different Chinese medicine components under the current preparation conditions, thus forming a component correlation model.

[0057] Step S3: Using the proportions of ingredients A, B, and C in this formulation as input, and combining the synergistic weights between ingredients determined by the ingredient correlation model, construct an ingredient synergistic relationship model to describe the nonlinear synergistic effect generated when multiple Chinese medicine ingredients participate in the formulation together, thereby obtaining the corresponding ingredient synergistic effect results.

[0058] Step S4: Based on the component feature representation vectors of components A, B, and C, and the process parameter data used in this formulation process, construct an influence relationship model on the characteristic components of each traditional Chinese medicine component to characterize the changes in the properties of each traditional Chinese medicine component under different process parameter conditions.

[0059] Step S5: Iteratively update the component association model among component A, component B, and component C, so that the component compatibility model can gradually adapt to the actual changes in different batches during the blending process, thereby improving the accuracy of subsequent modeling and decision-making for the blending of traditional Chinese medicine components.

[0060] Example 3, based on the above examples, specifically includes the following steps:

[0061] Step S31: Set a threshold condition for component association strength. When the association strength between any two Chinese medicine components meets the threshold condition, it is considered that there is a significant potential for synergistic relationship between the two Chinese medicine components. The association strength between component A and component B, and between component A and component C both meet the threshold condition, and the association strength between component B and component C also meets the threshold condition. In this embodiment, component A, component B, and component C together constitute the component set used for synergistic modeling.

[0062] Step S32: The blending ratio parameters correspond to the blending ratios of component A, component B, and component C, respectively. The component knowledge modeling module uses the blending ratio parameters as input to construct a component collaborative input vector, which represents the proportional distribution of components A, B, and C in the component set under the current blending scheme.

[0063] Step S33: Adjust the corresponding synergy weights according to the relative magnitude of the correlation strength between different component pairs, so that component pairs with higher correlation strength have higher synergy weights in the component synergy relationship model, while component pairs with relatively lower correlation strength have lower synergy weights, thus obtaining component synergy weights that match the current allocation scenario.

[0064] Step S34: Based on the component synergy input vector constructed in step S32 and the synergy weight determined in step S33, the component knowledge modeling module performs nonlinear mapping modeling on the synergy effect of component A, component B and component C under the current blending ratio. Through the component synergy effect function, the comprehensive synergy effect formed when multiple Chinese medicine components participate in the blending is calculated, and the synergy effect evaluation result used to characterize the component synergy level under the current blending scheme is obtained.

[0065] Step S35: Save the synergistic effect evaluation result obtained in step S34 as the output result of the component synergistic relationship model, and send the output result to the intelligent allocation decision module.

[0066] Example 4, based on the above examples, describes an intelligent dispensing decision module that sets dispensing targets and constraints, combines a traditional Chinese medicine (TCM) component compatibility model, and performs intelligent reasoning on the dispensing scheme of TCM components to generate a TCM component dispensing scheme and dispensing ratio parameters for each TCM component. Specifically, this includes the following steps:

[0067] In this embodiment, the core code used is as follows:

[0068] import itertools

[0069] import numpy as np

[0070] # =========================

[0071] # Basic Data Definitions (Components A, B, C)

[0072] # =========================

[0073] components = ["A", "B", "C"]

[0074] # Component Correlation Strength Matrix

[0075] association_strength = {

[0076] ("A", "B"): 0.8,

[0077] ("A", "C"): 0.6,

[0078] ("B", "C"): 0.7

[0079] }

[0080] # Symmetrical Completion

[0081] def get_assoc(i, j):

[0082] if i == j:

[0083] return 0.0

[0084] return association_strength.get((i, j)) or association_strength.get((j, i))

[0085] # =========================

[0086] # Step S31: Determining Collaborative Modeling Objects

[0087] # =========================

[0088] ASSOC_THRESHOLD = 0.5

[0089] collab_components = []

[0090] for i in range(len(components)):

[0091] for j in range(i + 1, len(components)):

[0092] if get_assoc(components[i], components[j]) >= ASSOC_THRESHOLD:

[0093] collab_components.append((components[i], components[j]))

[0094] # =========================

[0095] # Step S32: Construction of component co-input vector

[0096] # =========================

[0097] def build_input_vector(ratio_dict):

[0098] return np.array([ratio_dict[c] for c in components])

[0099] # =========================

[0100] # Step S33: Determine the collaborative weights

[0101] # =========================

[0102] def build_weight_matrix():

[0103] W = np.zeros((len(components), len(components)))

[0104] for i, ci in enumerate(components):

[0105] for j, cj in enumerate(components):

[0106] W[i][j] = get_assoc(ci, cj)

[0107] return W

[0108] W = build_weight_matrix()

[0109] # =========================

[0110] # Step S34: Nonlinear Modeling of Synergistic Effects

[0111] # =========================

[0112] def synergy_function(x, W):

[0113] # Nonlinear Synergistic Effect Function

[0114] return float(xT @ W @ x)

[0115] # =========================

[0116] # Step S35: Output of Collaborative Results

[0117] # =========================

[0118] def compute_synergy_score(ratio_dict):

[0119] x = build_input_vector(ratio_dict)

[0120] return synergy_function(x, W)

[0121] # =====================================================

[0122] # Step D1: Adjusting Objectives and Constraints

[0123] # =====================================================

[0124] RANGE = {

[0125] "A": (0.30, 0.50),

[0126] "B": (0.20, 0.40),

[0127] "C": (0.10, 0.30)

[0128] }

[0129] TOTAL_SUM = 1.0

[0130] # =====================================================

[0131] # Step D2: Generation of candidate allocation schemes

[0132] # =====================================================

[0133] def generate_candidates(step=0.05):

[0134] candidates = []

[0135] for a, b, c in itertools.product(

[0136] np.arange(0.30, 0.51, step),

[0137] np.arange(0.20, 0.41, step),

[0138] np.arange(0.10, 0.31, step) ):

[0140] if abs(a + b + c - TOTAL_SUM) < 1e-6:

[0141] candidates.append({"A": a, "B": b, "C": c})

[0142] return candidates

[0143] candidates = generate_candidates()

[0144] # =====================================================

[0145] # Step D3: Evaluation of Candidate Solutions

[0146] # =====================================================

[0147] def evaluate_candidates(candidates):

[0148] results = []

[0149] For scheme in candidates:

[0150] score = compute_synergy_score(scheme)

[0151] results.append((scheme, score))

[0152] return results

[0153] evaluated = evaluate_candidates(candidates)

[0154] # =====================================================

[0155] # Step D4: Screening of process parameter constraints

[0156] # =====================================================

[0157] def process_constraint_filter(scheme):

[0158] An excessively high A ratio is considered a process risk.

[0159] return scheme["A"] <= 0.48

[0160] filtered = [

[0161] (scheme, score)

[0162] for scheme, score in evaluated

[0163] if process_constraint_filter(scheme) ]

[0165] # =====================================================

[0166] # Step D5: Determining the optimal allocation plan

[0167] # =====================================================

[0168] best_scheme, best_score = max(filtered, key=lambda x: x[1])

[0169] # =====================================================

[0170] # Step D6: Output of the allocation plan

[0171] # =====================================================

[0172] def output_control_instruction(scheme):

[0173] return {

[0174] "component_A_ratio": scheme["A"],

[0175] "component_B_ratio": scheme["B"],

[0176] "component_C_ratio": scheme["C"],

[0177] "instruction": "AUTO_BLEND_EXECUTE"

[0178] }

[0179] control_instruction = output_control_instruction(best_scheme)

[0180] # =========================

[0181] # Final Output

[0182] # =========================

[0183] print("Optimal allocation scheme:", best_scheme)

[0184] print("Synergy score:", best_score)

[0185] print("Control command:", control_instruction).

[0186] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0187] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0188] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. An intelligent traditional Chinese medicine ingredient dispensing system, characterized in that: It includes a data acquisition module, a feature construction module, a component knowledge modeling module, an intelligent allocation decision-making module, an allocation execution module, and a monitoring and feedback module, specifically including the following: The data acquisition module collects multi-source data involved in the formulation of traditional Chinese medicine ingredients. The multi-source data includes attribute data of Chinese medicinal materials, component detection data, historical compatibility data, and process parameter data during the formulation process, forming the original dataset of traditional Chinese medicine ingredient formulation. The feature construction module preprocesses and constructs features from the original dataset. The preprocessing includes outlier removal and data standardization. Based on this, multidimensional feature information reflecting the characteristics of Chinese medicine components, component compatibility relationships, and the influence of process parameter data is extracted to obtain the feature dataset. The component knowledge modeling module constructs a traditional Chinese medicine component compatibility model based on the feature dataset, including a component correlation model, a component synergy model, and an influence relationship model between components and process parameter data; The intelligent dispensing decision module sets dispensing targets and constraints, and combines them with a traditional Chinese medicine component compatibility model to perform intelligent reasoning on the dispensing scheme of traditional Chinese medicine components, generating a dispensing scheme of traditional Chinese medicine components and dispensing ratio parameters of each traditional Chinese medicine component. The dispensing execution module generates corresponding dispensing control instructions based on the dispensing scheme of traditional Chinese medicine ingredients and the dispensing ratio parameters of each traditional Chinese medicine ingredient, and automatically controls the dosage, dispensing order and process parameters of the traditional Chinese medicine ingredients during the dispensing process. The monitoring and feedback module monitors the blending process and the blending results, acquires the blending result data, and feeds the blending result data back to the component knowledge modeling module and the intelligent blending decision module.

2. The intelligent traditional Chinese medicine ingredient dispensing system according to claim 1, characterized in that: The component knowledge modeling module constructs a compatibility model of traditional Chinese medicine components, including a component correlation model, a component synergistic model, and a model of the influence relationship between components and preparation process parameters. Specifically, it includes the following steps: Step S1: Construction of multi-scale feature representation of components. Based on the feature dataset, multi-scale modeling is performed on the content variation characteristics, stability characteristics, and response characteristics of traditional Chinese medicine components under different blending process conditions. The component feature representation vector is constructed as follows: ； in, Indicates the first Each Chinese medicine ingredient The component feature representation vector, Indicates the first Traditional Chinese medicine components at various scales eigencomponents, The number of feature scales; Step S2: Adaptive learning of component associations: Based on the component feature representation vector, calculate the association between any two Chinese herbal medicine components. and The association strength in a given allocation scenario is defined as follows: ； in, Indicates Chinese medicine ingredients With Chinese medicine ingredients The strength of the correlation between them Indicates the first Each component feature representation vector; Step S3: Nonlinear modeling of component synergistic relationship model. Based on the component correlation model, this step is applied to the model of synergistic relationships among multiple traditional Chinese medicine components. The constituent set is constructed, and the proportions of each Chinese herbal medicine component are used as input variables. A nonlinear model of the synergistic relationship between the constituents under different proportions is performed, and the synergistic effect function is used to represent it, as defined below: ； in, Indicates ingredients The proportions of ingredients in the combination Indicates ingredients The single-component contribution weights, They represent the components determined based on the component association model. and ingredients The collaborative weights between them It is a nonlinear mapping function; Step S4: Modeling the coupling influence between component and process parameter data. Based on the component feature representation vector and process parameter data, a model of the influence relationship between component and process parameter data is constructed. The coupling influence function is used to represent the influence relationship model, defined as follows: ； in, Indicates the effect of process parameter data on composition The component features represent the degree of influence of at least one feature component in the vector. For coupling mapping functions, This represents process parameter data; Step S5: Dynamic update of the compatibility relationship model. Based on the blending result data obtained by the monitoring and feedback module, the component correlation relationship model is iteratively updated.

3. The intelligent traditional Chinese medicine ingredient dispensing system according to claim 2, characterized in that: Step S3 specifically includes the following steps: Step S31: Determine the collaborative modeling object based on the component association model, targeting multiple traditional Chinese medicine components. The component set is constructed, and a threshold condition for component association strength is set. Multiple Chinese medicine components whose association strength meets the preset component association strength threshold condition are selected to form a component set for collaborative modeling. Step S32: Construction of component synergy input variables. For the component set used for synergy modeling, extract the mixing ratio parameters of each Chinese herbal medicine component in the current mixing scheme, and construct the component synergy input vector, as shown below: ； in, Indicates ingredients The proportions of the ingredients in the set of components; Step S33: Determining synergistic weights based on association constraints. Based on the strength of component associations, constraints are set on the synergistic weights of the components to ensure that the synergistic weights of the traditional Chinese medicine components are determined. With Chinese medicine ingredients Collaborative weights between The adjusted collaborative weights are obtained by adaptively adjusting the weights according to changes in the association strength. Step S34: Nonlinear mapping modeling of component synergy effect. Based on the component synergy input vector and the adjusted synergy weights, construct the component synergy effect function and obtain the synergy effect evaluation result as the output result. Step S35: Model output of synergistic effect results. The output results of the component synergistic effect function are used as the representation results of the component synergistic relationship model and sent to the intelligent allocation decision module.

4. The intelligent traditional Chinese medicine ingredient dispensing system according to claim 1, characterized in that: The intelligent dispensing decision module sets dispensing targets and constraints, and combines a traditional Chinese medicine (TCM) component compatibility model to intelligently reason about the dispensing scheme of TCM components, generating a TCM component dispensing scheme and dispensing ratio parameters for each TCM component. Specifically, it includes the following steps: Step D1: Setting the blending objectives and constraints. Based on the blending requirements of traditional Chinese medicine components, set the blending objectives and simultaneously set the corresponding constraints. The blending objectives include the objectives of enhancing the synergistic effect of components, ensuring the stability of key components, and controlling the blending process. The constraints include the blending ratio range constraints of each traditional Chinese medicine component, the total amount of components constraints, and the feasible range constraints of the blending process parameters. Step D2: Construction of the candidate set of dispensing schemes. Under the premise of meeting the constraints, based on the Chinese herbal medicine ingredients participating in the current dispensing task, the dispensing ratio of each Chinese herbal medicine ingredient is combined to form multiple feasible dispensing schemes of Chinese herbal medicine ingredients, which serve as the candidate dispensing scheme set. Step D3: Scheme evaluation based on the component matching relationship model. For the set of candidate allocation schemes, the component matching relationship model is called to evaluate the candidate allocation schemes and obtain evaluation results that reflect the characteristics of different allocation schemes. Step D4: Scheme screening based on process parameter constraints. In combination with the constraints, the feasibility of the candidate allocation scheme set is judged, and candidate allocation schemes that do not meet the constraints are eliminated. Step D5: Determination of the blending scheme. After step D4, the remaining candidate blending schemes are obtained. From the remaining candidate blending schemes, the blending scheme of Chinese medicine components that meets the blending target is determined, and the corresponding blending ratio parameters of each Chinese medicine component are output. Step D6: Output the blending scheme. Output the blending scheme of Chinese herbal ingredients that meets the blending target and the corresponding blending ratio parameters of each Chinese herbal ingredient to the blending execution module to generate blending control instructions.

Images