Processing method of radix paeoniae rubra decoction pieces

Abstract

The invention provides a processing method of radix paeoniae rubra decoction pieces, and relates to the technical field of traditional Chinese medicine processing. The processing method of the radix paeoniae rubra decoction pieces comprises the following steps: S1, based on a machine vision sorting technology, carrying out impurity identification on a radix paeoniae rubra raw material by adopting a YOLOv5 target detection algorithm, removing immature rhizomes with the diameter of less than 1cm through a morphological filtering algorithm, reserving qualified medicinal materials with the skin shrinkage degree of 0.2-0.4 mm, and generating a primary screening data set of the radix paeoniae rubra raw material; a machine vision sorting technology is combined with a morphological filtering algorithm to realize accurate identification and elimination of raw material impurities, the raw material screening precision is improved to a submillimeter level, component fluctuation caused by mixing of immature rhizomes is effectively avoided, a near infrared spectrum rapid detection technology is fused with a partial least squares regression algorithm to construct a component prediction model, and the accuracy of component prediction is improved. The method realizes lossless rapid screening of paeoniflorin content, improves screening efficiency by more than 20 times compared with a traditional chemical analysis method, and avoids environmental pollution caused by solvent consumption.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine processing technology, specifically a method for processing slices of red peony root. Background Technology

[0002] Traditional Chinese medicine processing is a comprehensive discipline that studies the processing methods of Chinese medicinal materials. It focuses on changing the physicochemical properties of medicinal materials through physical, chemical, or biological means to achieve purposes such as enhancing efficacy, reducing toxicity, improving taste, and facilitating storage. Among them, a processing method for red peony root slices refers to a systematic processing flow designed specifically for the characteristics of red peony root. Its purpose is to maximize the retention of active ingredients such as paeoniflorin and reduce the content of irritating substances such as tannins by controlling the degree of softening, cutting specifications, and processing parameters. This allows the finished slices to not only meet the clinical needs for promoting blood circulation, removing blood stasis, clearing heat and cooling blood, but also have the advantages of high dissolution rate and good gastrointestinal tolerance.

[0003] Traditional processing of red peony root relies on manual experience for raw material sorting. Operator fatigue can easily lead to the omission of immature rhizomes with diameters approaching 1 cm, resulting in an imbalance in the proportion of bitter and cold components in the processed slices. The softening process uses a fixed-time control mode, failing to dynamically adjust parameters according to differences in the moisture content of the medicinal materials. In some batches, fluctuations in environmental temperature and humidity result in the outer skin becoming soft and mushy while the interior remains unpenetrated, leading to an increased fragmentation rate in subsequent slicing processes. The slicing stage relies on the technician's feel and experience; errors in identifying fiber orientation often cause slice thickness variations exceeding 0.3 mm, affecting the uniformity of component dissolution rates during decoction. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for processing red peony root slices, which solves the problem that traditional red peony root processing relies on manual experience for raw material sorting. Operators' visual fatigue can easily lead to the omission of immature rhizomes with a diameter of nearly 1 cm, resulting in an imbalance in the proportion of bitter and cold components in the slices.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for processing slices of Paeonia lactiflora, comprising the following steps: S1: Based on machine vision sorting technology, the YOLOv5 target detection algorithm is used to identify impurities in the raw material of Paeonia lactiflora. Immature rhizomes with a diameter of <1cm are removed by morphological filtering algorithm, and qualified medicinal materials with an epidermal wrinkling degree of 0.2-0.4mm are retained to generate a preliminary screening dataset of Paeonia lactiflora raw materials. S2: Based on the initial screening dataset of Paeonia lactiflora raw materials, near-infrared spectroscopy rapid detection technology was used, and a paeoniflorin content prediction model was established through partial least squares regression algorithm. Batches with a content ≥3.5% were screened, and spectral fingerprints of Paeonia lactiflora components were generated. S3: Based on the spectral fingerprint of Paeonia lactiflora components, metabolomics dynamic monitoring technology is adopted, and the humidity of the softening machine is adjusted through fuzzy PID control algorithm. When the dissolution rate of paeoniflorin is detected to be >0.8mg / min, the softening is terminated, and the softening regulation parameters of Paeonia lactiflora are generated. S4: Based on the softening control parameters of Paeonia lactiflora, a deep residual network is used to analyze the fiber orientation of the medicinal material, and a genetic algorithm is used to optimize the cutting angle, outputting oblique slices with a thickness of 2.0±0.1mm, thus generating an intelligent cutting scheme for Paeonia lactiflora. S5: Based on the intelligent slicing scheme of red peony, a deep reinforcement learning algorithm is used to construct a stir-frying model, define the state space, and iteratively optimize the parameters through the reward function to generate a dynamic processing strategy for red peony; S6: Based on the dynamic processing strategy of red peony, low-oxygen drying technology is adopted. The oxygen concentration of the drying chamber is adjusted by fuzzy control algorithm. When the moisture content detected by near-infrared spectroscopy is ≤8%, a shutdown command is triggered to generate a stable drying curve for red peony. S7: Based on the stabilization drying curve of Paeonia lactiflora, blockchain smart contract technology is used to write the key parameters of each stage into the Ethereum consortium chain through the SHA-256 encryption algorithm to generate a digital twin of Paeonia lactiflora processing.

[0006] Preferably, the generation of the initial screening dataset for Paeonia lactiflora raw materials based on S1 includes the following steps: S101: Based on an industrial-grade linear CCD camera, it uses a Gaussian filtering algorithm to acquire images of the raw material surface, and enhances texture features through histogram equalization to generate a set of original images of peony raw materials; S102: Based on the original image set, an improved YOLOv5s model is used for impurity detection. Targets with a confidence level > 0.9 are screened using a non-maximum suppression algorithm, and a heatmap of peony impurity coordinates is generated. S103: Based on the coordinate heat map, a three-axis robotic arm is driven by a PID position control algorithm to remove foreign objects with a diameter difference greater than 15% through a vacuum suction cup, generating a preliminary screening dataset of peony raw materials.

[0007] Preferably, the generation of Paeonia lactiflora component spectral fingerprints based on S2 includes the following steps: S201: Based on fiber optic probe diffuse reflection technology, the 900-1700nm band spectrum is acquired using the Savitzky-Golay smoothing algorithm, and baseline drift is eliminated by first-order derivative processing to generate the peony near-infrared spectral matrix. S202: Based on the spectral matrix, a paeoniflorin quantitative model was established using the partial least squares regression algorithm. The model was evaluated by leave-one-out cross-validation with R² > 0.93, and the predicted values ​​of Paeonia lactiflora components were generated. S203: Based on the predicted component values, the K-means clustering algorithm (k=3) is used to classify the quality grades, and batches with paeoniflorin ≥3.8% as the class center are screened to generate the spectral fingerprint of Paeonia lactiflora components.

[0008] Preferably, the S3-based parameters for regulating the softening of Paeonia lactiflora include the following steps: S301: Based on a multi-channel temperature and humidity sensor, a Kalman filter algorithm is used to collect real-time environmental data inside the medicated medicine machine and generate a monitoring stream for the softening environment of peony root. S302: Based on environmental monitoring flow, a fuzzy PID control algorithm is used to adjust the spray volume, maintain humidity at 45±1.5%, and generate red peony softening control parameters; S303: Based on the control parameters, the dissolution rate of paeoniflorin is monitored online using ultraviolet-visible spectrophotometry. When the rate drops to <0.5mg / min, a termination signal is triggered, generating a red peony softening termination command.

[0009] Preferably, the intelligent peony slicing solution based on S4 includes the following steps: S401: Based on a structured light 3D scanner, a point cloud registration algorithm is used to reconstruct the 3D model of medicinal materials. The fiber orientation features are extracted through principal component analysis to generate a vector diagram of the fiber orientation of Paeonia lactiflora. S402: Based on vector graphics, a genetic algorithm is used to optimize the cutting angle. The objective function is a slice integrity index ≥ 0.85, generating the optimal cutting angle for Paeonia lactiflora. S403: Based on the optimal angle, adaptive fuzzy control is used to adjust the slicer feed speed, outputting oblique slices with a thickness of 2.0±0.08mm, generating a smart peony slicing scheme.

[0010] Preferably, the dynamic processing strategy for Paeonia lactiflora based on S5 includes the following steps: S501: Based on a digital twin system, a processing parameter space is constructed using Latin hypercube sampling to generate a virtual scene for processing Paeonia lactiflora. S502: Based on a virtual scenario, a deep deterministic policy gradient algorithm is used for policy optimization. The reward function R = 0.6 × bioavailability + 0.4 × energy efficiency is defined to generate a red peony processing policy model. S503: Based on the strategy model, the model predictive control is used to dynamically adjust the temperature of the wok to maintain the pyrolysis rate of paeoniflorin ≤0.2% / min, thus generating a dynamic processing strategy for red peony.

[0011] Preferably, the generation of the Paeonia lactiflora stabilization and drying curve based on S6 includes the following steps: S601: Based on a zirconia oxygen sensor, a proportional-integral-differential algorithm is used to control the nitrogen injection volume, maintain the oxygen concentration in the drying chamber at 3.5±0.3%, and generate low-oxygen environment parameters for peony. S602: Based on low oxygen parameters, microwave-hot air coupled drying technology is used, and the drying rate is calculated through a moisture migration model to generate a combined drying curve for Paeonia lactiflora. S603: Based on the drying curve, near-infrared online detection is used to monitor the moisture content in real time. When the detected value is ≤7.5%, the machine is triggered to stop and generate a stable drying curve for peony root.

[0012] Preferably, the generation of a digital twin of processed Paeonia lactiflora based on S7 includes the following steps: S701: Based on the Internet of Things edge computing gateway, it uses a time series database to collect sensor data for each process and generate a dataset of the red peony processing process. S702: Based on the process dataset, data blocks are constructed using a Merkle tree structure, and hash values ​​are generated through a proof-of-work algorithm to generate data blocks for the peony process. S703: Based on data blocks, it uses smart contract technology to deploy quality verification logic on the Ethereum private chain to generate a digital twin.

[0013] This invention provides a method for processing slices of Paeonia lactiflora. It has the following beneficial effects: This invention utilizes machine vision sorting technology combined with morphological filtering algorithms to achieve precise identification and removal of impurities in raw materials, improving the screening accuracy to sub-millimeter level and effectively avoiding component fluctuations caused by the contamination of immature rhizomes. Near-infrared spectroscopy rapid detection technology, integrated with partial least squares regression algorithms, constructs a component prediction model to achieve non-destructive and rapid screening of paeoniflorin content, increasing screening efficiency by more than 20 times compared to traditional chemical analysis methods, while avoiding environmental pollution caused by solvent consumption. Metabolomics dynamic monitoring technology, combined with fuzzy PID control algorithms, establishes a dynamic mapping relationship between the dissolution rate of active ingredients and humidity parameters during softening, improving the accuracy of softening termination point judgment to 95%, avoiding over-softening or under-softening caused by traditional experience-based judgment. Deep residual networks and genetic algorithms collaboratively optimize the cutting angle, generating the optimal cutting scheme by analyzing the fiber orientation characteristics of medicinal materials, improving the slice integrity index to 0.92, and reducing the loss of effective components by 38% compared to traditional manual cutting. A deep reinforcement learning algorithm-based roasting model, by defining a multi-dimensional state space and a composite reward function, dynamically optimizes roasting parameters, improving the thermal stability of paeoniflorin by 42% and controlling the degradation rate of tannins to below 12%. Low-oxygen drying technology, combined with a fuzzy control algorithm, constructs a nonlinear regulation model of oxygen concentration and moisture content, reducing the oxidation loss rate of active ingredients during drying to below 1.5%. Blockchain smart contract technology, through encrypted storage of process parameters and quality data, enables full lifecycle traceability of the roasting process, reducing the risk of data tampering to the order of 10^-6. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the main steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a detailed schematic diagram of S6 of the present invention; Figure 8 This is a detailed schematic diagram of S7 of the present invention. Detailed Implementation

[0015] To address the shortcomings of existing technologies, this invention provides a method for processing red peony root slices, which solves the problem that traditional red peony root processing relies on manual experience for raw material sorting. Operators' visual fatigue can easily lead to the omission of immature rhizomes with a diameter of nearly 1 cm, resulting in an imbalance in the proportion of bitter and cold components in the slices.

[0016] To achieve the above objectives, the present invention provides the following technical solution: a method for processing slices of Paeonia lactiflora, comprising the following steps: S1: Based on machine vision sorting technology, the YOLOv5 target detection algorithm is used to identify impurities in the raw material of Paeonia lactiflora. Immature rhizomes with a diameter of <1cm are removed by morphological filtering algorithm, and qualified medicinal materials with an epidermal wrinkling degree of 0.2-0.4mm are retained to generate a preliminary screening dataset of Paeonia lactiflora raw materials. S2: Based on the initial screening dataset of Paeonia lactiflora raw materials, near-infrared spectroscopy rapid detection technology was used, and a paeoniflorin content prediction model was established through partial least squares regression algorithm. Batches with a content ≥3.5% were screened, and spectral fingerprints of Paeonia lactiflora components were generated. S3: Based on the spectral fingerprint of Paeonia lactiflora components, metabolomics dynamic monitoring technology is adopted, and the humidity of the softening machine is adjusted through fuzzy PID control algorithm. When the dissolution rate of paeoniflorin is detected to be >0.8mg / min, the softening is terminated, and the softening regulation parameters of Paeonia lactiflora are generated. S4: Based on the softening control parameters of Paeonia lactiflora, a deep residual network is used to analyze the fiber orientation of the medicinal material, and a genetic algorithm is used to optimize the cutting angle, outputting oblique slices with a thickness of 2.0±0.1mm, thus generating an intelligent cutting scheme for Paeonia lactiflora. S5: Based on the intelligent slicing scheme of red peony, a deep reinforcement learning algorithm is used to build a stir-frying model, define the state space, and iteratively optimize the parameters through the reward function to generate a dynamic roasting strategy for red peony; S6: Based on the dynamic processing strategy of red peony, low-oxygen drying technology is adopted. The oxygen concentration of the drying chamber is adjusted by fuzzy control algorithm. When the moisture content detected by near-infrared spectroscopy is ≤8%, a shutdown command is triggered to generate a stable drying curve for red peony. S7: Based on the stabilization drying curve of Paeonia lactiflora, blockchain smart contract technology is used to write the key parameters of each stage into the Ethereum consortium chain through the SHA-256 encryption algorithm to generate a digital twin of Paeonia lactiflora processing.

[0017] Preferably, the generation of the initial screening dataset for Paeonia lactiflora raw materials based on S1 includes the following steps: S101: Based on an industrial-grade linear CCD camera, it uses a Gaussian filtering algorithm to acquire images of the raw material surface, and enhances texture features through histogram equalization to generate a set of original images of peony raw materials; S102: Based on the original image set, an improved YOLOv5s model is used for impurity detection. Targets with a confidence level > 0.9 are screened using a non-maximum suppression algorithm, and a heatmap of peony impurity coordinates is generated. S103: Based on the coordinate heatmap, a PID position control algorithm is used to drive a three-axis robotic arm to remove foreign objects with a diameter difference greater than 15% through a vacuum suction cup, generating a preliminary screening dataset of peony raw materials.

[0018] Preferably, the generation of Paeonia lactiflora component spectral fingerprints based on S2 includes the following steps: S201: Based on fiber optic probe diffuse reflection technology, the 900-1700nm band spectrum is acquired using the Savitzky-Golay smoothing algorithm, and baseline drift is eliminated by first-order derivative processing to generate the peony near-infrared spectral matrix. S202: Based on the spectral matrix, a paeoniflorin quantitative model was established using the partial least squares regression algorithm. The model was evaluated by leave-one-out cross-validation with R² > 0.93, and the predicted values ​​of Paeonia lactiflora components were generated. S203: Based on the predicted component values, the quality grades are divided using the K-means clustering algorithm (k=3), and batches with paeoniflorin ≥3.8% are screened to generate the spectral fingerprint of Paeonia lactiflora components.

[0019] Preferably, the S3-based parameters for regulating the softening of Paeonia lactiflora include the following steps: S301: Based on a multi-channel temperature and humidity sensor, a Kalman filter algorithm is used to collect real-time environmental data inside the medicated medicine machine and generate a monitoring stream for the softening environment of peony root. S302: Based on environmental monitoring flow, a fuzzy PID control algorithm is used to adjust the spray volume, maintain humidity at 45±1.5%, and generate red peony softening control parameters; S303: Based on the control parameters, the dissolution rate of paeoniflorin is monitored online using ultraviolet-visible spectrophotometry. When the rate drops to <0.5mg / min, a termination signal is triggered, generating a red peony softening termination command.

[0020] Preferably, the intelligent peony slicing solution based on S4 includes the following steps: S401: Based on a structured light 3D scanner, a point cloud registration algorithm is used to reconstruct the 3D model of medicinal materials. The fiber orientation features are extracted through principal component analysis to generate a vector diagram of the fiber orientation of Paeonia lactiflora. S402: Based on vector graphics, a genetic algorithm is used to optimize the cutting angle. The objective function is a slice integrity index ≥ 0.85, generating the optimal cutting angle for Paeonia lactiflora. S403: Based on the optimal angle, adaptive fuzzy control is used to adjust the slicer feed speed, outputting oblique slices with a thickness of 2.0±0.08mm, generating a smart peony slicing scheme.

[0021] Preferably, the dynamic processing strategy for Paeonia lactiflora based on S5 includes the following steps: S501: Based on a digital twin system, a processing parameter space is constructed using Latin hypercube sampling to generate a virtual scene for processing Paeonia lactiflora. S502: Based on a virtual scenario, a deep deterministic policy gradient algorithm is used for policy optimization. The reward function R = 0.6 × bioavailability + 0.4 × energy efficiency is defined to generate a red peony processing policy model. S503: Based on the strategy model, the model predictive control is used to dynamically adjust the temperature of the wok to maintain the pyrolysis rate of paeoniflorin ≤0.2% / min, thus generating a dynamic processing strategy for red peony.

[0022] Preferably, the generation of the Paeonia lactiflora stabilization and drying curve based on S6 includes the following steps: S601: Based on a zirconia oxygen sensor, a proportional-integral-differential algorithm is used to control the nitrogen injection volume, maintain the oxygen concentration in the drying chamber at 3.5±0.3%, and generate low-oxygen environment parameters for peony. S602: Based on low oxygen parameters, microwave-hot air coupled drying technology is used, and the drying rate is calculated through a moisture migration model to generate a combined drying curve for Paeonia lactiflora. S603: Based on the drying curve, near-infrared online detection is used to monitor the moisture content in real time. When the detected value is ≤7.5%, the machine is triggered to stop and generate a stable drying curve for peony root.

[0023] Preferably, the generation of a digital twin of processed Paeonia lactiflora based on S7 includes the following steps: S701: Based on the Internet of Things edge computing gateway, it uses a time series database to collect sensor data for each process and generate a dataset of the red peony processing process. S702: Based on the process dataset, data blocks are constructed using a Merkle tree structure, and hash values ​​are generated through a proof-of-work algorithm to generate data blocks for the peony process. S703: Based on data blocks, it uses smart contract technology to deploy quality verification logic on the Ethereum private chain to generate a digital twin.

[0024] This invention provides a method for processing slices of Paeonia lactiflora. It has the following beneficial effects: This invention utilizes machine vision sorting technology combined with morphological filtering algorithms to achieve precise identification and removal of impurities in raw materials, improving the screening accuracy to sub-millimeter level and effectively avoiding component fluctuations caused by the contamination of immature rhizomes. Near-infrared spectroscopy rapid detection technology, integrated with partial least squares regression algorithms, constructs a component prediction model to achieve non-destructive and rapid screening of paeoniflorin content, increasing screening efficiency by more than 20 times compared to traditional chemical analysis methods, while avoiding environmental pollution caused by solvent consumption. Metabolomics dynamic monitoring technology, combined with fuzzy PID control algorithms, establishes a dynamic mapping relationship between the dissolution rate of active ingredients and humidity parameters during softening, improving the accuracy of softening termination point judgment to 95%, avoiding over-softening or under-softening caused by traditional experience-based judgment. Deep residual networks and genetic algorithms collaboratively optimize the cutting angle, generating the optimal cutting scheme by analyzing the fiber orientation characteristics of medicinal materials, improving the slice integrity index to 0.92, and reducing the loss of effective components by 38% compared to traditional manual cutting. A deep reinforcement learning algorithm-based roasting model, by defining a multi-dimensional state space and a composite reward function, dynamically optimizes roasting parameters, improving the thermal stability of paeoniflorin by 42% and controlling the degradation rate of tannins to below 12%. Low-oxygen drying technology, combined with a fuzzy control algorithm, constructs a nonlinear regulation model of oxygen concentration and moisture content, reducing the oxidation loss rate of active ingredients during drying to below 1.5%. Blockchain smart contract technology, through encrypted storage of process parameters and quality data, enables full lifecycle traceability of the roasting process, reducing the risk of data tampering to the order of 10^-6.

Claims

1. A method for processing slices of Paeonia lactiflora, characterized in that... Includes the following steps: S1: Based on machine vision sorting technology, the YOLOv5 target detection algorithm is used to identify impurities in the raw material of Paeonia lactiflora. Immature rhizomes with a diameter of <1cm are removed by morphological filtering algorithm, and qualified medicinal materials with an epidermal wrinkling degree of 0.2-0.4mm are retained to generate a preliminary screening dataset of Paeonia lactiflora raw materials. S2: Based on the initial screening dataset of Paeonia lactiflora raw materials, near-infrared spectroscopy rapid detection technology was used, and a paeoniflorin content prediction model was established through partial least squares regression algorithm. Batches with a content ≥3.5% were screened, and spectral fingerprints of Paeonia lactiflora components were generated. S3: Based on the spectral fingerprint of Paeonia lactiflora components, metabolomics dynamic monitoring technology is adopted, and the humidity of the softening machine is adjusted through fuzzy PID control algorithm. When the dissolution rate of paeoniflorin is detected to be >0.8mg / min, the softening is terminated, and the softening regulation parameters of Paeonia lactiflora are generated. S4: Based on the softening control parameters of Paeonia lactiflora, a deep residual network is used to analyze the fiber orientation of the medicinal material, and a genetic algorithm is used to optimize the cutting angle, outputting oblique slices with a thickness of 2.0±0.1mm, thus generating an intelligent cutting scheme for Paeonia lactiflora. S5: Based on the intelligent slicing scheme of red peony, a deep reinforcement learning algorithm is used to construct a stir-frying model, define the state space, and iteratively optimize the parameters through the reward function to generate a dynamic processing strategy for red peony; S6: Based on the dynamic processing strategy of red peony, low-oxygen drying technology is adopted. The oxygen concentration of the drying chamber is adjusted by fuzzy control algorithm. When the moisture content detected by near-infrared spectroscopy is ≤8%, a shutdown command is triggered to generate a stable drying curve for red peony. S7: Based on the stabilization drying curve of Paeonia lactiflora, blockchain smart contract technology is used to write the key parameters of each stage into the Ethereum consortium chain through the SHA-256 encryption algorithm to generate a digital twin of Paeonia lactiflora processing.

2. The method for processing red peony root slices according to claim 1, characterized in that: The generation of the initial screening dataset for Paeonia lactiflora raw materials based on S1 includes the following steps: S101: Based on an industrial-grade linear CCD camera, it uses a Gaussian filtering algorithm to acquire images of the raw material surface, and enhances texture features through histogram equalization to generate a set of original images of peony raw materials; S102: Based on the original image set, an improved YOLOv5s model is used for impurity detection. Targets with a confidence level > 0.9 are screened using a non-maximum suppression algorithm, and a heatmap of peony impurity coordinates is generated. S103: Based on the coordinate heat map, a three-axis robotic arm is driven by a PID position control algorithm to remove foreign objects with a diameter difference greater than 15% through a vacuum suction cup, generating a preliminary screening dataset of peony raw materials.

3. The method for processing red peony root slices according to claim 1, characterized in that: The generation of spectral fingerprints of Paeonia lactiflora components based on S2 includes the following steps: S201: Based on fiber optic probe diffuse reflection technology, the 900-1700nm band spectrum is acquired using the Savitzky-Golay smoothing algorithm, and baseline drift is eliminated by first-order derivative processing to generate the peony near-infrared spectral matrix. S202: Based on the spectral matrix, a paeoniflorin quantitative model was established using the partial least squares regression algorithm. The model was evaluated by leave-one-out cross-validation with R² > 0.93, and the predicted values ​​of Paeonia lactiflora components were generated. S203: Based on the predicted component values, the K-means clustering algorithm (k=3) is used to classify the quality grades, and batches with paeoniflorin ≥3.8% as the class center are screened to generate the spectral fingerprint of Paeonia lactiflora components.

4. The method for processing red peony root slices according to claim 1, characterized in that: The generation of softening regulation parameters for Paeonia lactiflora based on S3 includes the following steps: S301: Based on a multi-channel temperature and humidity sensor, a Kalman filter algorithm is used to collect real-time environmental data inside the medicated medicine machine and generate a monitoring stream for the softening environment of peony root. S302: Based on environmental monitoring flow, a fuzzy PID control algorithm is used to adjust the spray volume, maintain humidity at 45±1.5%, and generate red peony softening control parameters; S303: Based on the control parameters, the dissolution rate of paeoniflorin is monitored online using ultraviolet-visible spectrophotometry. When the rate drops to <0.5mg / min, a termination signal is triggered, generating a red peony softening termination command.

5. The method for processing red peony root slices according to claim 1, characterized in that: The intelligent peony slicing solution based on S4 includes the following steps: S401: Based on a structured light 3D scanner, a point cloud registration algorithm is used to reconstruct the 3D model of medicinal materials. The fiber orientation features are extracted through principal component analysis to generate a vector diagram of the fiber orientation of Paeonia lactiflora. S402: Based on vector graphics, a genetic algorithm is used to optimize the cutting angle. The objective function is a slice integrity index ≥ 0.85, generating the optimal cutting angle for Paeonia lactiflora. S403: Based on the optimal angle, adaptive fuzzy control is used to adjust the slicer feed speed, outputting oblique slices with a thickness of 2.0±0.08mm, generating a smart peony slicing scheme.

6. The method for processing red peony root slices according to claim 1, characterized in that: The S5-based strategy for generating dynamic processing of Paeonia lactiflora includes the following steps: S501: Based on a digital twin system, a processing parameter space is constructed using Latin hypercube sampling to generate a virtual scene for processing Paeonia lactiflora. S502: Based on a virtual scenario, a deep deterministic policy gradient algorithm is used for policy optimization. The reward function R = 0.6 × bioavailability + 0.4 × energy efficiency is defined to generate a red peony processing policy model. S503: Based on the strategy model, the model predictive control is used to dynamically adjust the temperature of the wok to maintain the pyrolysis rate of paeoniflorin ≤0.2% / min, thus generating a dynamic processing strategy for red peony.

7. The method for processing red peony root slices according to claim 1, characterized in that; The process of generating a stable drying curve for Paeonia lactiflora based on S6 includes the following steps: S601: Based on a zirconia oxygen sensor, a proportional-integral-differential algorithm is used to control the nitrogen injection volume, maintain the oxygen concentration in the drying chamber at 3.5±0.3%, and generate low-oxygen environment parameters for peony. S602: Based on low oxygen parameters, microwave-hot air coupled drying technology is used, and the drying rate is calculated through a moisture migration model to generate a combined drying curve for Paeonia lactiflora. S603: Based on the drying curve, near-infrared online detection is used to monitor the moisture content in real time. When the detected value is ≤7.5%, the machine is triggered to stop and generate a stable drying curve for peony root.

8. The method for processing red peony root slices according to claim 1, characterized in that; The process of generating a digital twin of processed Paeonia lactiflora based on S7 includes the following steps: S701: Based on the Internet of Things edge computing gateway, it uses a time series database to collect sensor data for each process and generate a dataset of the red peony processing process. S702: Based on the process dataset, data blocks are constructed using a Merkle tree structure, and hash values ​​are generated through a proof-of-work algorithm to generate data blocks for the peony process. S703: Based on data blocks, it uses smart contract technology to deploy quality verification logic on the Ethereum private chain to generate a digital twin.

Images