Multi-mechanism linkage whole-process automatic preparation method and system for medicinal food

Abstract

This application relates to a fully automated method and system for preparing medicinal cuisine involving multiple mechanisms, comprising: acquiring target formula information and matching a first preparation strategy; generating a second preparation strategy based on the matching results of effective ingredient content; pre-treating medicinal materials and ingredients according to the second preparation strategy and dividing them into first and second feeding groups; feeding the first feeding group and ingredients into a decoction mechanism, and collecting temperature curves, liquid level change curves, and heat switching time sequences; generating a predicted concentration ratio and comparing it with the target concentration ratio; identifying the boiling stability range when the deviation exceeds a threshold, and generating a third preparation strategy based on the duration and direction of the deviation. This application achieves batch-adaptive parameter matching by dynamically adjusting the preparation strategy through the effective ingredient content; and achieves closed-loop optimization of the preparation strategy through temperature curve slope feature identification and dynamic calculation of the predicted concentration ratio; thus improving the automation level of the entire process, enhancing batch-to-batch quality consistency, and reducing the loss rate of medicinal components.

Description

Technical Field

[0001] This application relates to the technical field of automated preparation of traditional Chinese medicine, and in particular to a fully automated method and system for the preparation of medicinal diets involving multiple mechanisms. Background Technology

[0002] As a traditional health preservation system that deeply integrates TCM theory and dietary therapy practice, medicinal cuisine has irreplaceable application value in modern health management and disease prevention. However, the traditional preparation process of medicinal cuisine usually relies heavily on the subjective experience of the operators, which can easily lead to defects such as deviations in formula ratios, ambiguities in process parameters, and poor batch-to-batch quality consistency during the execution of the formula, making it difficult to meet the requirements of standardization and traceability for industrial production.

[0003] Existing automated medicinal cuisine preparation equipment is mainly limited to mechanical replacement of single processes. For example, automatic decoction machines only execute fixed time and temperature programs, and intelligent stewing equipment only performs basic heating control. These solutions typically use preset static process parameters, making it difficult to adjust in real time based on dynamic factors such as batch differences in medicinal materials and changes in ingredient freshness. Although some systems are equipped with temperature or liquid level sensors, the sensor monitoring data may only be used for simple threshold alarms or post-event recording, lacking dynamic modeling and prediction of the preparation process, and failing to achieve closed-loop optimization of process parameters. In key aspects such as controlling the soaking time of medicinal materials, precise particle size cutting, feeding sequence planning, and real-time adjustment of the concentration ratio, existing automated preparation systems still require manual intervention, resulting in low automation levels, high labor costs, and significant fluctuations in product quality. At the same time, for medicinal materials with high content of volatile active ingredients, existing automated preparation systems still lack differentiated feeding strategies based on component characteristics. For example, the failure to dynamically adjust the feeding timing before the boiling stabilization stage causes a large amount of medicinal components to be lost. Summary of the Invention

[0004] To address the aforementioned shortcomings, this application provides a fully automated method and system for preparing medicinal cuisine involving multiple interconnected mechanisms.

[0005] The above-mentioned objective of this application is achieved through the following technical solution: A fully automated method for preparing medicinal cuisine involving multiple mechanisms, comprising the following steps: Obtain target formula information, match target medicinal materials and target food ingredients based on target formula information, and match the first preparation strategy from the preset preparation strategy library based on target formula information; Based on the first preparation strategy, the effective component content of the target medicinal material and the target food is matched, and the first and second parameters of the first preparation strategy are adjusted according to the matching results to generate the second preparation strategy. The first parameter includes the wetting parameter and the particle size parameter, and the second parameter includes the feeding parameter and the firepower parameter. Based on the second preparation strategy, a processing instruction is generated and sent to the processing mechanism to preprocess the target medicinal materials and target food ingredients, and the preprocessed target medicinal materials are divided into a first feeding group and a second feeding group. According to the second preparation strategy, the first feeding group and the target ingredients are put into the frying mechanism for frying, and the temperature curve, liquid level change curve and fire switching time sequence are collected at the preset sampling period. The predicted concentration ratio is generated based on the temperature curve and the liquid level change curve, and the predicted concentration ratio is compared with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation. When the comparison deviation exceeds the preset deviation threshold, the boiling stability range is identified based on the slope change characteristics of the temperature curve, and the second preparation strategy is adjusted in real time according to the deviation direction between the duration of the boiling stability range and the target concentration ratio, thereby generating the third preparation strategy. When the decoction time is reached, the second feeding group is added, and the actual concentration ratio is calculated after the decoction is completed. The actual concentration ratio, the second preparation strategy, and the third preparation strategy are then updated to the preparation strategy library.

[0006] The second objective of this invention is achieved through the following technical solution: A fully automated medicinal diet preparation system with multi-mechanism linkage, comprising: The strategy matching module is used to obtain target formula information, match target medicinal materials and target food ingredients based on the target formula information, and match the first preparation strategy from the preset preparation strategy library based on the target formula information. The parameter adjustment module is used to match the effective component content of the target medicinal material and the target food based on the first preparation strategy, and adjust the first parameter and the second parameter of the first preparation strategy according to the matching result to generate the second preparation strategy. The first parameter includes the wetting parameter and the particle size parameter, and the second parameter includes the feeding parameter and the firepower parameter. The instruction generation module is used to generate processing instructions based on the second preparation strategy, send the processing instructions to the processing mechanism to preprocess the target medicinal materials and target food ingredients, and divide the preprocessed target medicinal materials into a first feeding group and a second feeding group. The sampling module is used to put the first feeding group and the target ingredients into the frying mechanism for frying according to the second preparation strategy, and to collect temperature curves, liquid level change curves and fire switching time sequences at a preset sampling period. The comparison module is used to generate a predicted concentration ratio based on the temperature curve and the liquid level change curve, and compare the predicted concentration ratio with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation. The strategy adjustment module is used to identify the boiling stability range based on the slope change characteristics of the temperature curve when the comparison deviation exceeds the preset deviation threshold, and adjust the second preparation strategy in real time according to the deviation direction between the duration of the boiling stability range and the target concentration ratio to generate the third preparation strategy. The strategy update module is used to add the second feeding group when the decoction time is reached, and to calculate the actual concentration ratio after the decoction is completed, and update the actual concentration ratio, the second preparation strategy and the third preparation strategy to the preparation strategy library.

[0007] In summary, this application achieves batch-adaptive matching of preparation process parameters by dynamically adjusting the preparation strategy based on the effective component content of the target medicinal materials and target food ingredients; it achieves dynamic calculation of the predicted concentration ratio and closed-loop optimization of the preparation strategy by modeling the evaporation rate in stages and identifying the slope characteristics of the temperature curve; it achieves continuous accumulation of preparation experience by writing the actual concentration ratio and preparation strategy back to the preparation strategy library; and finally, through the synergy of the above processes, it can improve the degree of process automation and batch-to-batch quality consistency, and reduce the loss rate of active ingredients. Attached Figure Description

[0008] Figure 1 This is a flowchart of an embodiment of a fully automated preparation method for medicinal cuisine involving multiple institutions, as described in this application. Figure 2 This is a flowchart of step S10 in an embodiment of a fully automated medicinal diet preparation method involving multiple institutions in this application; Figure 3 This is a flowchart of step S20 in an embodiment of a fully automated medicinal diet preparation method involving multiple institutions, as described in this application. Detailed Implementation

[0009] The following is in conjunction with the appendix Figures 1-3 This application will be described in further detail.

[0010] In one embodiment, such as Figure 1 As shown, this application discloses a fully automated method for preparing medicinal cuisine involving multiple institutions, specifically including the following steps: S10: Obtain target formula information, match target medicinal materials and target food ingredients based on target formula information, and match the first preparation strategy from the preset preparation strategy library based on target formula information; In this embodiment, the target formula information refers to a detailed set of data used to guide the preparation of medicinal cuisine, typically including the name of the medicinal materials, the dosage of the medicinal materials, the name and dosage of the ingredients, and efficacy labels. The preparation strategy library refers to a pre-set database that stores various process schemes for preparing medicinal cuisine, each scheme being associated with a specific formula feature vector and corresponding preparation parameters. The first preparation strategy refers to the initial preparation scheme obtained by preliminary matching from the preparation strategy library based on the target formula information, which typically includes pre-set wetting parameters, particle size parameters, feeding parameters, and heat parameters.

[0011] Specifically, target formula information can be manually entered by the operator through a human-computer interaction interface, or selected from a preset formula list. When matching target medicinal materials and target food ingredients, a search can be conducted in a preset raw material inventory database based on the names listed in the formula. For matching the first preparation strategy, a direct association method can be used, that is, each formula is pre-bound with a fixed preparation strategy, or a preliminary preparation scheme can be selected by keyword matching of the formula name.

[0012] S20: Based on the first preparation strategy, the effective component content of the target medicinal material and the target food is matched, and the first and second parameters of the first preparation strategy are adjusted according to the matching results to generate the second preparation strategy. The first parameter includes the wetting parameter and the particle size parameter, and the second parameter includes the feeding parameter and the firepower parameter. In this embodiment, effective ingredient content matching refers to detecting or evaluating the effective ingredient content of the current batch of target medicinal materials and target food ingredients, and comparing the detection results with the preset effective ingredient content benchmark value in the first preparation strategy to quantify the difference between the current batch and the benchmark batch. The first parameter refers to the key parameters affecting the physical processing of medicinal materials and food ingredients, including wetting parameters and particle size parameters. Wetting parameters refer to the wetting process parameters during the pretreatment of target medicinal materials, including wetting time, wetting temperature, and the amount of wetting solution, used to control the degree of water absorption and expansion of the medicinal materials before decoction, thereby affecting the dissolution efficiency of the effective ingredients. Particle size parameters refer to the particle size control parameters during the cutting or pulverizing of target medicinal materials and target food ingredients, including the target particle size range and cutting accuracy, etc. Particle size directly affects the specific surface area of ​​the medicinal materials, thereby affecting the dissolution rate of the effective ingredients. The second parameter refers to the key parameters affecting the decoction process of medicinal cuisine, including feeding parameters and heat parameters. Among them, the feeding parameters refer to the process parameters of the target medicinal materials and ingredients being fed into the decoction mechanism, including the feeding sequence, feeding amount, and feeding timing. For medicinal materials with a high content of volatile active ingredients, precise control of the feeding timing can greatly reduce the loss of medicinal components. The heat parameters refer to the heating control parameters of the decoction mechanism during the decoction process, including the heat intensity of the high heat stage, the heat intensity of the low heat stage, the duration of the high heat stage, and the duration of the low heat stage. These directly affect the trend of the decoction temperature curve and the evaporation rate. The second preparation strategy refers to the optimized preparation scheme generated by dynamically adjusting the target medicinal materials and ingredients according to their actual characteristics (such as the content of active ingredients and freshness) based on the first preparation strategy.

[0013] Specifically, the matching of active ingredient content can be judged based on the general standard values ​​of medicinal materials, that is, all batches of a certain medicinal material are assumed to have an effective ingredient content within an average range. When a difference is found between the medicinal material or food and the general standard, the first parameter (such as soaking time, cutting particle size) and the second parameter (such as the order of adding ingredients, decoction heat) can be adjusted. For example, if the medicinal material is hard, the soaking time can be extended; if the food has a high moisture content, the decoction heat can be increased.

[0014] S30: Generate processing instructions based on the second preparation strategy, send the processing instructions to the processing mechanism to preprocess the target medicinal materials and target food ingredients, and divide the preprocessed target medicinal materials into the first feeding group and the second feeding group; In this embodiment, the processing mechanism refers to a set of automated equipment for pre-processing medicinal materials and ingredients. It may include a soaking mechanism, a cutting mechanism, a crushing mechanism, a washing mechanism, a peeling mechanism, and a dicing mechanism, etc., used to complete the corresponding physical processing according to the processing instructions. The first feeding group and the second feeding group are two groups of pre-processed target medicinal materials according to the feeding sequence. The first feeding group is added at the initial stage of decoction, and the second feeding group is added when the decoction time is reached. This grouping is mainly for medicinal materials with high content of volatile effective components. By delaying the feeding, the large amount of medicinal components lost due to prolonged boiling is avoided.

[0015] Specifically, processing instructions can be generated based on the preset general processing flow in the second preparation strategy. For example, all medicinal materials can be washed and sliced, and all food ingredients can be washed and cut into pieces. After receiving the processing instructions, the processing mechanism performs the corresponding physical processing. The pre-treated target medicinal materials can be divided into a first feeding group and a second feeding group according to their general properties (such as volatility), i.e., feeding first and feeding later.

[0016] S40: According to the second preparation strategy, the first feeding group and the target ingredients are put into the frying mechanism for frying, and the temperature curve, liquid level change curve and fire switching time sequence are collected at the preset sampling period. In this embodiment, the decoction mechanism refers to an automated device used for decocting medicinal herbs and ingredients. It typically possesses precise temperature control and liquid level monitoring capabilities, and can execute the decoction process according to set heat parameters. The temperature curve refers to the time-series temperature data continuously collected by the decoction mechanism at a preset sampling period during the decoction process. It reflects the temperature change trend throughout the decoction process and is used for slope change feature extraction and boiling stability interval identification. The liquid level change curve refers to the time-series liquid level data continuously collected by the decoction mechanism at a preset sampling period during the decoction process. It reflects the liquid evaporation throughout the decoction process and is used for calculating the staged evaporation rate and predicting the concentration ratio. The heat switching time sequence refers to the time-series data set recording the switching moments between high heat and low heat during the decoction process. It is used to divide the actual duration of the high heat stage and the low heat stage, providing a time reference for calculating the staged evaporation rate.

[0017] Specifically, the decocting mechanism can perform decocting according to a fixed time schedule and heat program preset in the second preparation strategy, for example, first high heat for 30 minutes, then low heat for 60 minutes. Meanwhile, the feeding mechanism adds the first batch of ingredients and the target food into the decocting mechanism at preset fixed times. During the decocting process, preset sensors can collect real-time temperature and liquid level data of the decocting liquid at fixed time intervals, and record the moment when the heat switches from high to low.

[0018] S50: Generate a predicted concentration ratio based on the temperature curve and liquid level change curve, and compare the predicted concentration ratio with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation; In this embodiment, the predicted concentration ratio refers to the predicted value of the ratio of the solid solution concentration to the initial liquid volume at the end of decoction, based on the current temperature curve and liquid level change curve, combined with a staged evaporation rate model, etc. This is used to assess in advance whether the final concentration result meets the target requirements during the decoction process. The target concentration ratio refers to the desired final concentration level of the medicinal food preset in the second preparation strategy, serving as a reference benchmark for real-time control and adjustment of the decoction process. The comparison deviation refers to the difference between the predicted concentration ratio and the target concentration ratio associated with the second preparation strategy, used to quantify the degree to which the current decoction process deviates from the target concentration ratio, and to guide the adjustment direction of the preparation strategy through the direction of the deviation.

[0019] Specifically, the predicted concentration ratio can be estimated using empirical formulas, such as estimating the degree of concentration based on the percentage drop in liquid level. The comparison deviation can be obtained by directly subtracting the predicted concentration ratio from the target concentration ratio preset in the second preparation strategy.

[0020] S60: When the comparison deviation exceeds the preset deviation threshold, the boiling stability range is identified based on the slope change characteristics of the temperature curve, and the second preparation strategy is adjusted in real time according to the deviation direction between the duration of the boiling stability range and the target concentration ratio, and a third preparation strategy is generated. In this embodiment, the boiling stability interval refers to the continuous period in which the absolute value of the slope of the temperature curve is consistently lower than a preset slope threshold. This corresponds to the time interval during which the liquid maintains a stable boiling state, and its duration reflects the stability of the boiling state during the current cooking process, serving as a basis for judging the cooking state. The third preparation strategy refers to a dynamically optimized process parameter scheme generated by real-time correction of the heat parameters and cooking time based on the direction of the comparison deviation and the duration of the boiling stability interval, building upon the second preparation strategy.

[0021] Specifically, when the absolute value of the comparison deviation exceeds a preset deviation threshold, it is determined that adjustment is necessary. The boiling stability range can be identified by observing whether the temperature curve fluctuates within a certain small range. The adjustment strategy can be based on preset corrections according to the direction of the deviation; for example, if the predicted concentration ratio is too low, the cooking time is extended; if the predicted concentration ratio is too high, the heat is reduced.

[0022] S70: When the decoction time is reached, the second feeding group is added, and the actual concentration ratio is calculated after the decoction is completed. The actual concentration ratio, the second preparation strategy, and the third preparation strategy are then updated to the preparation strategy library.

[0023] In this embodiment, the decoction time refers to the total duration of the decoction process preset in the second preparation strategy. Starting from the beginning of decoction, it encompasses the complete decoction cycle, including both the high-heat and low-heat stages, and is used to determine the timing of the second feeding group's feeding and the criterion for judging the end of decoction. It should be noted that when the third preparation strategy is generated, the decoction time may dynamically change with adjustments to the extension of the high-heat time or the reduction of the low-heat time. The actual concentration ratio refers to the measured value of the concentration ratio calculated by actually measuring the volume of the decoction liquid and the solid solution content after decoction, corresponding to the predicted concentration ratio generated during the decoction process. The actual concentration ratio serves as the final quality evaluation index for this preparation process. Together with the second and third preparation strategies, it is written back to the preparation strategy library, forming a closed-loop optimization mechanism to correct prediction deviations of historical strategies and improve the accuracy of matching preparation strategies for subsequent batches.

[0024] Specifically, the second batch of feed can be added when the total cooking time reaches the preset value. After cooking, the actual concentration ratio can be calculated by measuring the final liquid level and the initial liquid level, combined with the initial solid content. Subsequently, the actual concentration ratio of this preparation, as well as the second preparation strategy used and the possible third preparation strategy, are recorded as historical data.

[0025] This application achieves intelligent control of the pretreatment of medicinal materials and ingredients, as well as the decoction process, by acquiring formula information and matching it with an initial preparation strategy, and dynamically adjusting parameters based on the actual characteristics of the medicinal materials. By monitoring the decoction state in real time and predicting the concentration ratio, it can promptly identify the boiling stability range and adjust the preparation strategy when deviations occur, thereby improving the accuracy and quality consistency of the medicinal cuisine preparation process. Furthermore, by feeding the actual results back to the strategy library after decoction, continuous learning and optimization of the preparation strategy can be achieved, effectively solving the problems of unstable formula execution, difficulty in quantifying process parameters, and large batch-to-batch quality fluctuations in traditional medicinal cuisine preparation. This has the effect of improving the efficiency and standardization level of automated medicinal cuisine preparation.

[0026] In one embodiment, such as Figure 2 As shown, step S10 includes: S11: Obtain target formula information, which includes the name of medicinal materials, dosage of medicinal materials, name of ingredients, dosage of ingredients, and efficacy label; In this embodiment, the target formula information is the starting input of the automated preparation process. It can be obtained through user interface input, scanning QR codes, reading from databases, etc. The target formula information not only includes the types and amounts of medicinal materials and ingredients, but can also be further refined to efficacy labels, such as "tonifying qi", "nourishing blood", "strengthening spleen", etc., which can more comprehensively reflect the expected effects and intrinsic properties of medicinal diet.

[0027] S12: Based on the target formula information, match the corresponding target medicinal materials and target ingredients from the preset raw material library; In this embodiment, after obtaining the target formula information, the raw materials are matched against a pre-defined raw material library based on the names of medicinal herbs and ingredients listed in the target formula information. Upon successful matching, the detailed attributes of the matched raw materials are further extracted. For target medicinal herbs, batch information, origin information, and storage time can be obtained to assess the quality, effective component content, and potential changes in physicochemical properties. For target ingredients, their freshness level (e.g., obtained through visual recognition, sensor detection, etc.) and specification information (such as size, shape, weight, etc.) can be obtained. These attributes directly affect the pretreatment methods and cooking results of the ingredients.

[0028] S13: Generate a combination of medicinal herb names and the proportion of medicinal herb dosage based on the target medicinal herbs and target food ingredients, and combine the efficacy tags as a strategy matching feature vector; In this embodiment, to achieve more accurate strategy matching, key features are further extracted from the acquired target medicinal materials and target food ingredients information, and a strategy matching feature vector is constructed. Specifically, based on the target medicinal materials and target food ingredients, key features including the combination of medicinal material names, the proportion of medicinal materials used, and the efficacy tags contained in the target formula information are generated. The generated key features are then integrated into a multi-dimensional vector that can comprehensively reflect the core components, compatibility relationships, and expected efficacy of the formula, i.e., the strategy matching feature vector.

[0029] S14: Calculate the similarity between the strategy matching feature vector and the formulation feature vector corresponding to each preparation strategy in the preparation strategy library, and select the preparation strategy with the highest similarity as the first preparation strategy.

[0030] In this embodiment, after constructing the strategy matching feature vector, it is further compared with the pre-stored formula feature vectors associated with each preparation strategy in the preparation strategy library for similarity calculation. The similarity calculation can employ various algorithms, such as cosine similarity or the reciprocal of Euclidean distance, to quantify the degree of matching between the current target formula and existing strategies in the library. The preparation strategy with the highest similarity in the calculation results will be selected as the first preparation strategy, serving as the initial guiding scheme for this medicinal diet preparation.

[0031] Furthermore, to address the possibility that highly matching formula feature vectors may not exist in the preparation strategy library, a fault-tolerance mechanism can be pre-defined. Specifically, if the calculated highest similarity is still lower than the pre-defined similarity threshold, it indicates that the current target formula differs significantly from existing strategies. In this case, based on the medicinal material attributes (such as texture and medicinal properties) and food attributes (such as hardness and moisture content) in the target formula information, several default preparation parameters can be extracted from the pre-defined process parameter template library. These include default wetting parameters, default particle size parameters, default feeding parameters, and default heat parameters. These default parameters can constitute an initial preparation strategy as the first preparation strategy to ensure that even when facing entirely new or rare formulas, a relatively reasonable initial preparation scheme can be provided.

[0032] By employing the aforementioned technical solutions, target formula information is obtained, and a strategy matching feature vector generated based on the acquired detailed information can more accurately characterize the intrinsic attributes and expected efficacy of the target formula. Furthermore, by calculating the similarity between the strategy matching feature vector and the formula feature vectors in the preparation strategy library, the first preparation strategy that best matches the current formula can be accurately identified and selected, thereby ensuring the rationality and effectiveness of the initial preparation plan and avoiding deviations caused by simple matching. In addition, when no highly similar strategies exist in the preparation strategy library, default parameters can be extracted from the process parameter template library based on the properties of medicinal materials and ingredients to generate an initial preparation strategy, enhancing the process's adaptability and practicality. This allows it to handle diverse medicinal cuisine formulas, even novel or rare formulas, starting with a reasonable initial preparation plan, thereby improving the automation level and final quality of medicinal cuisine preparation.

[0033] In one embodiment, such as Figure 3 As shown, step S20 includes: S21: Obtain batch information of the target medicinal material and extract effective ingredient content data, wherein the effective ingredient content data includes volatile oil content, polysaccharide content, alkaloid content and flavonoid content; In this embodiment, to ensure the traceability and accuracy of the medicinal material quality, the batch information of the target medicinal material is first obtained, such as production batch number, supplier, place of origin, and harvest date. Simultaneously, the content data of active ingredients associated with that batch of medicinal material is extracted. This content data can typically be obtained through professional quality testing methods, such as high-performance liquid chromatography (HPLC) for determining the content of flavonoids and alkaloids, gas chromatography-mass spectrometry (GC-MS) for determining the content of volatile oils, and the phenol-sulfuric acid method for determining the content of polysaccharides.

[0034] S22: Compare the effective ingredient content data with the standard effective ingredient content range associated with the first preparation strategy, and calculate the effective ingredient content deviation rate; In this embodiment, after obtaining the actual effective component content data of the target medicinal material, it is compared with the preset standard effective component content range in the first preparation strategy. The standard effective component content range is typically an ideal range set based on the pharmacological activity of the medicinal material, optimal extraction efficiency, and the efficacy requirements of the medicinal diet. It can be set according to actual needs and the knowledge available in the field. Through comparison, the content deviation rate of each effective component can be calculated. For example, if the actual content is lower than the lower limit of the standard range, the deviation rate is negative; if it is higher than the upper limit of the standard range, the deviation rate is positive.

[0035] S23: When the content deviation rate exceeds the preset deviation threshold, select the corresponding adjustment parameter from the first parameter and the second parameter according to the physicochemical properties of the effective ingredient, and determine the adjustment range according to the content deviation rate; In this embodiment, when the content deviation rate of any active ingredient exceeds a preset allowable deviation threshold, an intelligent adjustment process can be initiated. Specifically, based on the physicochemical properties of the active ingredient, such as volatility, water solubility, and heat sensitivity, the most suitable parameter is selected from the first and second parameters for adjustment. For example, if the content of volatile components is low, it may be necessary to adjust the wetting parameter to reduce volatilization, or adjust the feeding parameter to optimize the extraction timing; if the content of water-soluble components is low, it may be necessary to adjust the particle size parameter to increase the contact area, or adjust the heat parameter to enhance the dissolution efficiency. Furthermore, the adjustment range can be determined proportionally based on the content deviation rate or through a preset mapping table to ensure the accuracy of the adjustment.

[0036] S24: Obtain the freshness grade and specifications of the target ingredient, extract the moisture content and fiber hardness information, and adjust the first parameter and the second parameter based on the moisture content and fiber hardness information. In this embodiment, since the quality of ingredients, in addition to medicinal herbs, also significantly impacts the taste and nutritional value of medicinal cuisine, the freshness level of the target ingredients (e.g., through visual recognition, odor sensors, or data provided by the supplier) and their specifications (e.g., size, shape, weight) are obtained. Based on this, the moisture content and fiber hardness information of the ingredients are extracted. Moisture content directly affects the liquid level changes and concentration effect during the decoction process, while fiber hardness affects the tenderness of the ingredients and the release of active ingredients. The first and second parameters are adjusted accordingly based on this information. For example, ingredients with high moisture content may require a reduced initial water amount or a longer high-heat stage to evaporate excess moisture; ingredients with high fiber hardness may require a longer soaking time or a smaller particle size to promote softening and ingredient dissolution.

[0037] S25: Generate a second preparation strategy based on the adjusted first and second parameters, and associate it with the batch information of the target formulation information.

[0038] In this embodiment, after the aforementioned adjustments to the effective components of the medicinal materials and the quality of the ingredients, all adjusted wetting parameters, particle size parameters, feeding parameters, and heat parameters are integrated to form a second preparation strategy. This second preparation strategy is a customized optimization scheme adjusted for the current batch of medicinal materials and ingredients. To achieve traceability of the preparation process and subsequent strategy optimization, the generated second preparation strategy is linked to the specific batch information of the target formula information, so as to analyze the impact of different batches of raw materials on the preparation effect and continuously improve the preparation strategy library.

[0039] By employing the aforementioned technical solution, batch information and effective component content data of the target medicinal materials are obtained and compared with standard ranges. This embodiment's solution can quantify the actual quality differences of the medicinal materials. Based on this, and combined with the physicochemical properties of the effective components, intelligent adjustments to the wetting parameters, particle size parameters, feeding parameters, and heat parameters can further ensure the full extraction of effective components from the medicinal materials. Simultaneously, by obtaining the freshness grade, specifications, moisture content, and fiber hardness information of the target ingredients, and adjusting the preparation parameters accordingly, the impact of ingredient quality differences on taste and nutrition can be effectively addressed, ensuring the ingredients achieve the ideal tenderness and flavor. The resulting second preparation strategy is a personalized optimization scheme for the current batch of raw materials, which improves the accuracy, stability, and quality of the final product in medicinal cuisine preparation, thereby ensuring the consistency of the medicinal cuisine's efficacy and taste.

[0040] In one embodiment, step S23 includes: S231: Extract the physicochemical property labels of the active ingredients, wherein the physicochemical property labels include volatile labels, water-soluble labels and heat-sensitive labels; In this embodiment, the physicochemical property label of an active ingredient refers to a characteristic identifier used to describe the physical and chemical properties of a specific active ingredient in a medicinal material, which directly affects its stability and extraction efficiency during the preparation process. The physicochemical property label can be pre-stored in a database and associated with the names or IDs of various active ingredients. For example, for volatile oils, the physicochemical property label might include "volatility"; for polysaccharides or alkaloids, it might include "water solubility"; and for flavonoids, it might include "heat sensitivity". After obtaining the active ingredient content data, the corresponding physicochemical property label can be automatically extracted according to the ingredient type.

[0041] S232: When the active ingredient has a volatile label, determine whether the content deviation rate of the volatile oil content exceeds the preset first deviation threshold. If it does, calculate the first difference between the content deviation rate and the first deviation threshold, and query the corresponding wetting parameter adjustment amount and feeding parameter adjustment amount from the preset adjustment mapping table based on the first difference, and classify the corresponding medicinal material into the second feeding group. In this embodiment, when an active ingredient has a volatility label, it indicates that the active ingredient is easily volatile. When the detected deviation rate of volatile oil content exceeds a preset first deviation threshold, it indicates that the actual content of volatile oil differs significantly from the standard. To make precise adjustments, the first difference between the content deviation rate and the first deviation threshold is calculated, and the adjustment amounts of the soaking parameters and the feeding parameters are retrieved from a preset adjustment mapping table based on the first difference. The adjustment amount of the soaking parameters may involve shortening the soaking time or lowering the soaking temperature to reduce the loss of volatile oil; the adjustment amount of the feeding parameters may involve classifying such medicinal materials into a second feeding group, i.e., adding them in the later stage of decoction, to retain their volatile components to the greatest extent.

[0042] S233: When the active ingredient has a water-soluble label, determine whether the content deviation rate of polysaccharide content or the content deviation rate of alkaloid content exceeds the preset second deviation threshold. If it exceeds the threshold, calculate the second difference between the content deviation rate and the second deviation threshold, and query the corresponding particle size parameter adjustment amount and firepower parameter adjustment amount from the preset adjustment mapping table based on the second difference. In this embodiment, when an active ingredient has a water-soluble label, it indicates that the active ingredient is readily soluble in water. When the content deviation rate of polysaccharide content or alkaloid content exceeds a preset second deviation threshold, it indicates that its solubility or extraction efficiency may be affected. At this time, a second difference between the content deviation rate and the second deviation threshold is calculated, and based on the second difference, the particle size parameter adjustment amount and the heat parameter adjustment amount are queried from a preset adjustment mapping table. The particle size parameter adjustment amount may involve reducing the particle size of the medicinal material to increase the surface area and promote the dissolution of water-soluble components; the heat parameter adjustment amount may involve adjusting the heat intensity or duration during the decoction process to optimize the extraction efficiency of water-soluble components.

[0043] S234: When the active ingredient has a heat-sensitive label, determine whether the content deviation rate of flavonoids exceeds the preset third deviation threshold. If it does, calculate the third difference between the content deviation rate and the third deviation threshold, and query the corresponding heat parameter adjustment amount and total decoction time adjustment amount from the preset adjustment mapping table based on the third difference.

[0044] In this embodiment, when the active ingredient has a heat-sensitive label, it indicates that the active ingredient is heat-sensitive, and high temperatures may cause it to decompose or become inactive. When the content deviation rate of flavonoid compounds exceeds a preset third deviation threshold, it indicates that there may be a risk of heat damage during the preparation process. At this time, the third difference between the content deviation rate and the third deviation threshold is calculated, and the adjustment amount of the heat parameter and the total decoction time are queried from a preset adjustment mapping table based on the third difference. The adjustment amount of the heat parameter may involve reducing the decoction temperature or shortening the duration of the high-temperature stage; the adjustment amount of the total decoction time may involve appropriately shortening the overall decoction time while ensuring the extraction effect, so as to reduce the exposure time of heat-sensitive components. The adjustment mapping table is a pre-set database or lookup table that stores the correspondence between different content deviation rates or their differences and the corresponding parameter adjustment amounts. The adjustment mapping table can be constructed and optimized based on a large amount of experimental data, pharmacological knowledge and experience. For example, when the volatile oil content deviation rate is too high, the larger the first difference, the larger the queried immersion parameter adjustment amount (such as shortening the immersion time) and feeding parameter adjustment amount (such as feeding later) will be.

[0045] Through the above technical solution, this application can make targeted adjustments to the preparation parameters based on the physicochemical properties of different active ingredients in the target medicinal materials. Specifically, for active ingredients with a volatile label, adjusting the wetting parameters and feeding parameters can effectively reduce the loss of volatile components and optimize their retention during decoction; for active ingredients with a water-soluble label, adjusting the particle size parameters and heat parameters can improve the dissolution efficiency of water-soluble components; for active ingredients with a heat-sensitive label, adjusting the heat parameters and total decoction time can effectively prevent heat-sensitive components from decomposing or becoming inactive due to high temperatures. Through this dynamic adjustment based on the physicochemical properties of active ingredients, it can be ensured that the preparation process of medicinal cuisine can retain and extract the active ingredients in the medicinal materials as much as possible, thereby improving the efficacy and overall quality of the medicinal cuisine.

[0046] In one embodiment, step S24 includes: S241: Extract moisture content data from the freshness grade of the target ingredient, compare the moisture content data with the preset standard moisture content range, and calculate the moisture deviation rate; In this embodiment, the freshness grade of the target ingredient is an important indicator reflecting its quality, and it is usually closely related to the ingredient's moisture content. By extracting moisture content data from the freshness grade, the moisture content of the ingredient can be quantified. For example, freshness grades can be divided into "premium," "grade 1," etc., with each grade corresponding to a moisture content range or typical value. Simultaneously, a standard moisture content range can be preset; for example, for a certain leafy vegetable, the standard moisture content range might be 85%-90%. By comparing the actual extracted moisture content data with this standard range, the moisture deviation rate can be calculated, i.e., the degree of deviation between the actual moisture content and the median or boundary value of the standard range.

[0047] S242: When the moisture deviation rate is positive and exceeds the preset upper limit threshold of moisture, it is determined that the moisture content of the target ingredient is too high. The initial liquid level setting value in the fire power parameters is adjusted to reduce the amount of water added and extend the duration of the high heat stage. In this embodiment, when the moisture content of the ingredients is detected to be too high, it indicates that the ingredients themselves will release a significant amount of water. If the same amount of water is added according to the standard formula, the medicinal soup may become too diluted, affecting the concentration effect and the dissolution of active ingredients. Therefore, it is necessary to adjust the initial liquid level setting in the heat parameters, i.e., reduce the amount of water added at the beginning of simmering to compensate for the moisture released by the ingredients themselves. At the same time, extending the duration of the high-heat stage allows the high temperature to quickly evaporate excess water, accelerating the concentration process and ensuring that the expected concentration ratio is achieved within the total simmering time.

[0048] S243: When the moisture deviation rate is negative and its absolute value exceeds the preset lower limit threshold of moisture, it is determined that the moisture content of the target ingredient is too low. The initial liquid level setting value in the fire power parameters is adjusted to increase the amount of water added and shorten the duration of the high heat stage.

[0049] In this embodiment, when the moisture content of the ingredients is too low, the ingredients themselves release less water. If the amount of water added is not increased, the broth may concentrate too early or even dry out, affecting the full dissolution of the effective components of the medicinal herbs. Therefore, it is necessary to adjust the initial liquid level setting in the heat parameters and increase the amount of water added to ensure sufficient solvent for decoction. Shortening the duration of the high-heat stage can avoid excessive concentration or scorching due to insufficient moisture, while allowing more time for dissolution in the subsequent low-heat stage.

[0050] Furthermore, in addition to calculating the moisture deviation rate, the hardness deviation rate can also be calculated. Specifically, this includes: extracting fiber hardness data from the specification information of the target ingredient, comparing the fiber hardness data with the preset standard fiber hardness range, and calculating the hardness deviation rate. When the hardness deviation rate exceeds the preset hardness upper limit threshold, it is determined that the fiber hardness of the target food is too high, and the wetting parameters and particle size parameters are adjusted. When the hardness deviation rate is lower than the preset lower limit threshold, it is determined that the fiber hardness of the target ingredient is too low. The soaking parameters are adjusted to shorten the pre-soaking time of the ingredient, and the heat parameters are adjusted to reduce the maximum temperature setting of the high heat stage to avoid the ingredient becoming overly soft.

[0051] The specifications of the target ingredients, such as size, shape, and origin, are often related to the fiber hardness. Fiber hardness is a key factor affecting the softening degree of ingredients and the dissolution of active ingredients. By extracting fiber hardness data from the specifications, the physical structural characteristics of the ingredients can be quantified. For example, for root vegetables, the specifications may include diameter and length, and these specifications are related to fiber hardness. In this case, a standard fiber hardness range can be preset, and the actual extracted hardness data can be compared with this standard range to calculate the hardness deviation rate.

[0052] When the fiber of food is too hard, its structure is dense, making it difficult for active ingredients to dissolve and soften. In this case, it is necessary to adjust the soaking and particle size parameters. Extending the pre-soaking time allows the food to fully absorb water and soften. Reducing the particle size of the food, such as by slicing, chopping, or crushing, increases the contact area between the food and the decoction liquid, thereby accelerating the dissolution of active ingredients and improving decoction efficiency.

[0053] When the fiber hardness of ingredients is too low, their structure is relatively loose, making them prone to over-softening or even mushy, affecting the taste and quality of medicinal cuisine. In this case, it's necessary to adjust the soaking parameters, shortening the pre-soaking time to prevent the ingredients from absorbing too much water and softening before cooking. Simultaneously, adjusting the heat parameters and lowering the maximum temperature setting during the high-heat stage can slow down the softening process, preventing rapid decomposition at high temperatures, thus maintaining the shape and texture of the ingredients and ensuring the stable dissolution of active ingredients.

[0054] Through the above technical solution, this application can dynamically adjust the initial liquid level setting, high-heat stage duration, pre-soaking time, cutting particle size, and maximum high-heat stage setting during the decoction process based on the actual moisture content and fiber hardness of the target ingredients. This effectively solves the impact of uneven moisture and hardness caused by individual differences in ingredients on the preparation effect of medicinal cuisine. This adaptive adjustment mechanism based on the physical characteristics of ingredients in this embodiment can improve the accuracy and automation level of medicinal cuisine preparation, thereby enhancing the quality stability and effectiveness of the medicinal cuisine.

[0055] In one embodiment, the processing mechanism includes an impregnation mechanism, a cutting mechanism, and a crushing mechanism, and step S30 includes: S31: Obtain the texture attributes of the target medicinal material, and determine the type of medicinal material based on the texture attributes. The types of medicinal materials include rhizomes, minerals, leaves, flowers, and fruits. In this embodiment, the soaking mechanism is mainly used to soak the medicinal materials to soften and expand them, which is beneficial for subsequent cutting or the dissolution of active ingredients. This can be achieved using automated equipment such as soaking tanks with temperature and time control functions and circulating pumps. The cutting mechanism is used to cut the medicinal materials into specific shapes and sizes to increase the surface area and improve decoction efficiency. This can be achieved using automated cutting equipment composed of various knives, conveyor belts, and size sensors, such as slicers and block cutters. The grinding mechanism is used to grind the medicinal materials into powder or fine particles to further increase the surface area and accelerate the dissolution of active ingredients. This can be achieved using grinders, pulverizers, and sieving devices. Texture attributes refer to the physical properties of the medicinal materials, such as hardness, toughness, and brittleness, which can be obtained through preset database queries, sensor detection (such as hardness testers and spectrometers), or manual input. Medicinal material types are classifications based on these attributes, including rhizomes, minerals, leaves, flowers, and fruits.

[0056] S32: When the medicinal material is a root or mineral, a soaking instruction is generated and sent to the soaking mechanism. The soaking instruction includes the soaking duration and the soaking temperature. In this embodiment, root and mineral medicinal materials are typically hard and dense, making direct decoction ineffective in releasing their components. Soaking, however, softens the tissue, removes impurities, and promotes the dissolution of active ingredients. The soaking command includes soaking time and temperature. The soaking time can be set based on factors such as the hardness, density, and solubility characteristics of the active ingredients. For example, for hard medicinal materials, the soaking time can be set to several hours; for mineral medicinal materials, a longer soaking time may be needed to remove surface impurities. The soaking temperature can be set according to the properties of the medicinal material. For example, for some heat-sensitive medicinal materials, room temperature soaking can be used; for medicinal materials requiring accelerated softening, the soaking temperature can be appropriately increased. After receiving the soaking command, the soaking mechanism can make corresponding adjustments by controlling the water level, temperature, and soaking time in the soaking tank.

[0057] S33: When the medicinal material is a leaf, flower or fruit, determine whether the particle size of the target medicinal material is greater than the preset cutting threshold. If the particle size is greater than the preset cutting threshold, generate a cutting instruction and send the cutting instruction to the cutting mechanism. The cutting instruction includes the slice thickness and the cut size. In this embodiment, leafy, flower, and fruit-type medicinal materials are relatively soft, but still require appropriate cutting to increase surface area and improve decoction efficiency. Unnecessary cutting can be avoided by judging particle size. The particle size value can be obtained through a visual recognition system (such as image processing combined with size measurement) or a weighing volume method. If the particle size value is greater than a preset cutting threshold, a cutting instruction is generated and sent to the cutting mechanism. The cutting instruction includes slice thickness and block size. The slice thickness or block size in the cutting instruction is determined based on the tissue structure of the medicinal material, the distribution of active ingredients, and decoction requirements (such as decoction time and heat intensity). For example, a smaller slice thickness can be set for thin-sliced ​​medicinal materials; a suitable block size can be set for block-shaped fruits. After receiving the cutting instruction, the cutting mechanism can perform precise cutting by automatically adjusting the blade spacing or cutting mode.

[0058] S34: When the medicinal material is a leaf, flower or fruit, if the particle size is less than or equal to the preset cutting threshold and greater than the preset pulverizing threshold, a pulverizing instruction is generated and sent to the pulverizing mechanism. The pulverizing instruction includes the pulverizing mesh size.

[0059] In this embodiment, for medicinal materials with small particle sizes that still require further processing to improve dissolution rate, further processing can be performed by a pulverizing mechanism. The pulverization mesh size determines the fineness of the pulverization, while the pulverization threshold can be preset according to the characteristics of the medicinal materials and the dissolution requirements of the active ingredients. The pulverization instruction includes the pulverization mesh size, which can usually be set according to the dissolution characteristics of the active ingredients of the medicinal materials, the decoction time, and the taste requirements of the final medicinal diet. For example, for medicinal materials that require rapid dissolution of active ingredients, a higher pulverization mesh size can be set; for medicinal materials that need to retain a certain particle texture, a lower pulverization mesh size can be set. After receiving the pulverization instruction, the pulverizing mechanism can automatically adjust the grinding parameters or the sieve aperture to pulverize the medicinal materials to the specified fineness.

[0060] Through the above technical solutions, this application can determine the type of medicinal material based on its texture properties and select appropriate pretreatment methods such as soaking, cutting, or pulverizing to ensure the full release of the effective components of different medicinal materials and avoid efficacy loss or resource waste caused by improper processing. For example, for hard root or mineral medicinal materials, the soaking time and temperature can be precisely controlled to soften them fully, improving the dissolution rate of effective components in subsequent decoctions; for leaf, flower, or fruit medicinal materials, appropriate slice thickness, cut size, or pulverization mesh can be selected based on their particle size to maximize their surface area and accelerate the dissolution of effective components. Through the above treatments, not only can the automation level of medicinal food preparation be improved, but also the quality and efficacy of medicinal foods can be enhanced.

[0061] In one embodiment, step S40 includes: S41: Based on the heat parameters and the characteristics of the target medicinal materials, the decoction process is divided into a high-heat stage and a low-heat stage. The high-heat stage is associated with a high-heat temperature setpoint and a high-heat duration, and the low-heat stage is associated with a low-heat temperature setpoint. In this embodiment, the stage division of the decoction process is a crucial step in the preparation of medicinal cuisine. It aims to maximize the dissolution of active ingredients and optimize the taste by using different heating intensities and times based on the physicochemical properties of the medicinal materials and ingredients. The decoction process is divided into a high-heat stage and a low-heat stage. The high-heat stage typically refers to a rapid heating process at high temperature and high power, designed to quickly reach boiling and promote the initial dissolution of active ingredients. Its characteristics include a high high-heat temperature setting and a preset high-heat duration. The low-heat stage refers to a slow simmering process at a lower temperature and low power, designed to maintain a stable boil, promote the slow dissolution of deeper active ingredients, and reduce the loss of volatile components. Its characteristics include a lower low-heat temperature setting.

[0062] S42: Generate feeding timing instructions based on feeding parameters and send the feeding timing instructions to the feeding mechanism, so that the feeding mechanism feeds the first feeding group and the target ingredients into the pre-set multi-temperature zone frying mechanism in the order associated with the feeding timing instructions. In this embodiment, different medicinal herbs and ingredients have different decoction characteristics. For example, some medicinal herbs need to be decocted first, while others need to be added later; some ingredients require long-term stewing, while others are easy to cook. By generating a feeding sequence instruction, and having the feeding mechanism add the first feeding group and the target ingredients into the decoction mechanism according to the instruction, it can be ensured that all kinds of raw materials enter the decoction process at the optimal time, thereby avoiding over-decoction or under-decoction, and improving the quality of the final medicinal diet and the integrity of the effective components. The feeding sequence instruction includes the feeding time and amount of the target medicinal herbs and target ingredients. The decoction mechanism has multiple preset temperature zones, meaning that its internal temperature can be controlled in different areas as needed to adapt to the temperature requirements of different feeding stages.

[0063] S43: During the high-temperature stage, the high-temperature setpoint is used as the target value of the PID controller. The PID controller calculates the first power adjustment amount based on the deviation between the current temperature and the high-temperature setpoint, the integral of the deviation, and the derivative of the deviation, and sends the first adjustment command to the simmering mechanism based on the first power adjustment amount. In this embodiment, the PID controller refers to a feedback controller used in industrial control to precisely control the system output to make it as close as possible to the target value. During the high-temperature stage, the decocting mechanism needs to quickly reach and maintain a high temperature. The PID controller uses the high-temperature setpoint as the target and monitors the current temperature within the decocting mechanism in real time. When there is a deviation between the current temperature and the setpoint, the PID controller calculates a first power adjustment amount based on the proportion, accumulation, and rate of change of the deviation. This first power adjustment amount is converted into a first adjustment command and sent to the heating unit of the decocting mechanism to dynamically adjust the heating power, ensuring that the temperature of the decocting liquid can quickly and accurately reach and stabilize at the high-temperature setpoint, thereby achieving efficient initial decocting.

[0064] S44: When the actual duration of the high-heat stage reaches the duration of the high-heat stage, record the moment of power switching, use the low-heat temperature setpoint as the target value of the PID controller, and switch to the low-heat stage. In this embodiment, the duration of the high-heat stage can be preset based on actual needs and historical data. When the actual simmering time reaches the high-heat duration, the moment of heat switching is recorded. At this time, the simmering process transitions from the high-heat stage to the low-heat stage. During the switch, the target value of the PID controller is updated from the high-heat temperature setpoint to the low-heat temperature setpoint, preparing for the subsequent low-heat simmering process.

[0065] S45: The PID controller calculates the second power adjustment amount based on the deviation between the current temperature and the simmering temperature setpoint, the integral of the deviation, and the derivative of the deviation, and sends the second adjustment command to the simmering mechanism based on the second power adjustment amount; In this embodiment, after entering the simmering stage, the decoction mechanism needs to maintain a relatively low and stable temperature to promote the full dissolution of active ingredients and reduce volatilization. At this time, the PID controller continues to function, but its target value has become the simmering temperature setpoint. Simultaneously, the PID controller also calculates a second power adjustment based on the deviation between the current temperature and the simmering temperature setpoint, the integral of the deviation, and the derivative of the deviation. This second power adjustment is converted into a second adjustment command and sent to the decoction mechanism to dynamically control the heating power, ensuring that the decoction temperature is stably maintained near the simmering temperature setpoint. This ensures that the medicinal materials and ingredients fully release their active ingredients in a gentle environment, while avoiding component destruction or loss due to excessive boiling.

[0066] S46: Collect the real-time temperature value and real-time liquid level value of the cooking mechanism at a preset sampling period, generate a temperature curve based on the real-time temperature value, generate a liquid level change curve based on the real-time liquid level value, and record the recorded fire switching time to the fire switching time sequence.

[0067] In this embodiment, to achieve comprehensive monitoring and adjustment of the decoction process, real-time temperature and liquid level values ​​of the decoction mechanism can be continuously collected at a preset sampling period. The real-time temperature value is used to generate a temperature curve, which visually reflects the temperature change trend over time during decoction, and is used to determine the heating effect and boiling state. The real-time liquid level value is used to generate a liquid level change curve, which reflects the evaporation of the decoction liquid, and is used to assess the concentration ratio and determine whether water replenishment is needed. Simultaneously, the recorded heat switching times during the decoction process are also integrated into the heat switching time sequence. All of these data together constitute a complete picture of the decoction process, providing a data foundation for subsequent prediction, comparison, and strategy adjustment.

[0068] Through the above technical solution, this application precisely divides the decoction process into a high-heat stage and a low-heat stage, and sets clear temperature targets and durations for each stage, enabling automated control of traditional medicinal cuisine decoction experience. By introducing a PID controller to provide real-time feedback control of the heating power of the decoction mechanism, it can ensure that the temperature of the decoction liquid can quickly and accurately reach the high-heat temperature setpoint, and remain stably at the low-heat temperature setpoint during the low-heat stage. This reduces the risk of temperature fluctuations and inaccurate control that may be caused by manual operation, ensuring the full dissolution of the effective components of the medicinal materials and improving the stability of the decoction effect. In addition, by generating feeding sequence instructions based on the feeding parameters, and having the feeding mechanism feed the materials in sequence, medicinal materials and ingredients with different decoction characteristics can be added to the decoction at the optimal time, avoiding over- or under-decoction, which can further improve the quality of the medicinal cuisine. Meanwhile, by continuously collecting real-time temperature and liquid level values ​​at a preset sampling cycle, generating corresponding curves and recording the timing of fire switching, accurate data support can be provided for subsequent evaluation of decoction effects, prediction of concentration ratio, and real-time adjustment of preparation strategies. This has the effect of improving the intelligence level of automated preparation of medicinal cuisine and product consistency.

[0069] In one embodiment, step S50 includes: S51: Extract the average temperature and temperature fluctuation amplitude at the current sampling moment based on the temperature curve, and extract the actual duration of the intense fire phase and the actual duration of the simmering fire phase based on the fire switching time sequence. In this embodiment, the average temperature reflects the overall heat input level during the simmering process, while the temperature fluctuation range may indicate the intensity of boiling or the stability of heat transfer. The heat switching time sequence records the start and end times of different heat stages. Specifically, the average temperature can be obtained by calculating the arithmetic mean of the temperature curve within a preset time window near the current sampling time; the temperature fluctuation range can be obtained by calculating the difference between the maximum and minimum temperature values ​​within the time window, or by calculating its standard deviation; the heat switching time sequence can be automatically recorded and stored when the heat switching operation is performed, such as recording the start and end times of high heat.

[0070] S52: Extract the initial liquid level value and the current liquid level value based on the liquid level change curve, calculate the current liquid level drop, and calculate the evaporation rate of the high-heat stage and the evaporation rate of the low-heat stage based on the current liquid level drop and the actual duration of the high-heat stage and the actual duration of the low-heat stage, respectively. In this embodiment, the liquid level drop reflects the loss of water. By correlating it with the duration of different heat stages, the average evaporation rate of the high-heat and low-heat stages can be calculated separately. Since the evaporation rates of different heat stages differ significantly, calculating them separately improves prediction accuracy. Specifically, the initial liquid level value can be obtained by a liquid level sensor at the start of cooking, while the current liquid level value can be obtained in real time by the liquid level sensor during each sampling period. The current liquid level drop refers to the difference between the initial liquid level value and the current liquid level value. The evaporation rate of the high-heat stage can be obtained by dividing the total liquid level drop of the high-heat stage by the actual duration of the high-heat stage; similarly, the evaporation rate of the low-heat stage can be obtained by dividing the total liquid level drop of the low-heat stage by the actual duration of the low-heat stage.

[0071] S53: Based on the evaporation rate during the high-heat stage and the evaporation rate during the low-heat stage, combined with the remaining high-heat time and the remaining low-heat time in the second preparation strategy, predict the liquid level at the end of the decoction. In this embodiment, the predicted liquid level value can be calculated by subtracting the product of the evaporation rate during the high-heat stage and the remaining duration of the high-heat stage from the current liquid level value, and then subtracting the product of the evaporation rate during the low-heat stage and the remaining duration of the low-heat stage. The remaining durations of the high-heat and low-heat stages can be obtained from the second preparation strategy and are dynamically updated based on the currently performed cooking time.

[0072] S54: Extract the initial solid solution content of the target medicinal material from the second preparation strategy, and calculate the predicted concentration ratio based on the initial liquid level value, the predicted liquid level value and the initial solid solution content according to the preset concentration ratio calculation model. In this embodiment, the concentration ratio calculation model can be constructed based on the mapping relationship between the total liquid level drop and the amount of dissolved solids. That is, the concentration ratio calculation model considers the dissolution efficiency of the effective components during decoction, rather than just simple volume changes. Specifically, the initial solid content can be pre-stored in the second preparation strategy, or obtained from a database based on information such as the type and dosage of the target medicinal material. The concentration ratio calculation model can be a function trained based on empirical data, physicochemical principles, or machine learning. For example, it can be the predicted concentration ratio = (initial solid content + amount of dissolved solids) / predicted liquid level at the end of decoction. The amount of dissolved solids can be estimated by multiplying the total liquid level drop by a preset dissolution efficiency coefficient. This dissolution efficiency coefficient may be related to the characteristics of the medicinal material, decoction temperature, time, etc., and can be set by those skilled in the art based on existing knowledge and historical experience.

[0073] S55: Calculate the difference between the predicted concentration ratio and the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation.

[0074] In this embodiment, by calculating the difference between the predicted concentration ratio and the target concentration ratio, the degree of deviation between the current cooking state and the ideal state can be quantified, providing a basis for subsequent strategy adjustments. Specifically, the comparison deviation can be directly calculated using the difference between the predicted concentration ratio and the target concentration ratio, where the target concentration ratio is the preset ideal concentration ratio value in the second preparation strategy.

[0075] Through the above technical solution, this application can extract the average temperature, fluctuation amplitude, and actual duration of different heat stages from the real-time temperature curve and liquid level change curve during the decoction process, combined with the heat switching time sequence. Based on this, by calculating the evaporation rates of the high-heat and low-heat stages separately, and combining this with the remaining decoction time in the second preparation strategy, the liquid level value at the end of decoction can be predicted more accurately. Furthermore, by introducing the initial solids content of the target medicinal material and constructing a concentration ratio calculation model based on the mapping relationship between the total liquid level drop and the amount of solids dissolved, the calculation of the predicted concentration ratio not only considers water evaporation but also incorporates the dissolution efficiency of the effective components. This multi-dimensional prediction method improves the accuracy of the predicted concentration ratio, allowing the comparison deviation between the predicted and target concentration ratios to more accurately reflect the degree of deviation between the current decoction state and the preset target. This effectively avoids over- or under-adjustment due to inaccurate predictions, ultimately ensuring that the concentration and quality of the effective components in the medicinal diet reach their optimal state.

[0076] In one embodiment, step S60 includes: S61: When the absolute value of the comparison deviation exceeds the preset deviation threshold, calculate the temperature slope sequence between adjacent sampling times based on the temperature curve, and perform sliding window averaging on the temperature slope sequence to obtain a smooth slope sequence. In this embodiment, when the absolute value of the comparison deviation exceeds a preset deviation threshold, a temperature slope sequence between adjacent sampling times is calculated based on the real-time acquired temperature curve. This temperature slope sequence reflects the rate of temperature change over time. To eliminate the influence of instantaneous noise and fluctuations on the slope calculation, a sliding window averaging process is further applied to the temperature slope sequence to obtain a more stable and reliable smooth slope sequence. The sliding window averaging process can be achieved by setting a fixed-size window, averaging the slope values ​​within the window, and then sliding the window forward by one step. Repeating this process provides a more accurate temperature change trend.

[0077] S62: Identify continuous periods where the absolute value of the slope is consistently lower than a preset slope threshold based on the smooth slope sequence, determine the continuous periods as boiling stability intervals, and calculate the duration of the boiling stability intervals. In this embodiment, the boiling stability range is a crucial state during the decoction process. It is typically characterized by a relatively stable temperature change after reaching the boiling point, despite continued heating; that is, the absolute value of the temperature slope is very small. The preset slope threshold can be empirically set based on the characteristics of different medicinal materials and ingredients, as well as the performance of the decoction equipment, or optimized using experimental data.

[0078] S63: Compare the duration of the boiling stability zone with the preset duration threshold, and adjust the second preparation strategy based on the comparison result and the deviation direction of the target concentration ratio to generate the third preparation strategy.

[0079] In this embodiment, the preset duration threshold represents the stable boiling time required to achieve the target concentration ratio under ideal conditions. The comparison results, combined with the direction of deviation from the target concentration ratio (e.g., whether the predicted concentration ratio is higher or lower than the target), jointly determine how to adjust the current second preparation strategy.

[0080] Furthermore, specifically, when the deviation direction is positive and there is an overboiling indicator, the amount of firepower reduction and the amount of reduction in the duration of intense boiling can be calculated based on the absolute value of the comparison deviation. The amount of firepower reduction and the amount of reduction in the duration of intense boiling are then subtracted from the firepower parameters and the duration of intense boiling in the second preparation strategy to generate the third preparation strategy. When the deviation direction is positive and there is an insufficient boiling indicator, the reduction in simmering time can be calculated based on the absolute value of the comparison deviation. The reduction in simmering time is then subtracted from the simmering duration in the second preparation strategy to generate the third preparation strategy. When the deviation direction is negative and there is an overboiling indicator, the extension of the simmering time can be calculated based on the absolute value of the comparison deviation. The extension of the simmering time is then added to the simmering duration in the second preparation strategy to generate the third preparation strategy. When the deviation direction is negative and there is an insufficient boiling indicator, the amount of firepower increase and the amount of fire duration extension can be calculated based on the absolute value of the comparison deviation. The amount of firepower increase and the amount of fire duration extension are then added to the firepower parameters and fire duration in the second preparation strategy to generate the third preparation strategy.

[0081] When the deviation is positive (i.e., the predicted concentration ratio is higher than the target concentration ratio) and an over-boiling indicator is present, the amount of heat reduction and the amount of high-heat duration reduction can be calculated based on the absolute value of the deviation. The calculated heat reduction and high-heat duration reduction are then subtracted from the heat parameters and high-heat duration in the second preparation strategy to generate the third preparation strategy. Since the over-boiling indicator usually indicates excessive energy input or excessively vigorous boiling during the cooking process, the heat parameters can be adjusted to reduce the heating power or the heating temperature setpoint, and the high-heat duration can be directly shortened.

[0082] When the deviation is positive and there is an indication of insufficient boiling, the reduction in simmering time can be calculated based on the absolute value of the deviation. This reduction can then be subtracted from the simmering duration in the second preparation strategy to generate the third preparation strategy. This may mean that although the concentration ratio is high, the dissolution efficiency of the effective components during the decoction process is not high, or the water evaporation rate is too fast during the simmering stage. In this case, shortening the simmering time can reduce the total decoction time, thereby reducing the total amount of water evaporation and making the final concentration ratio closer to the target value.

[0083] When the deviation is negative (i.e., the predicted concentration ratio is lower than the target concentration ratio) and there is an over-boiling indicator, the extension of the simmering time can be calculated based on the absolute value of the deviation. This extension is then added to the simmering duration in the second preparation strategy to generate the third preparation strategy. Since the over-boiling indicator may indicate that although boiling is vigorous, the concentration effect is poor due to other factors (such as excessive initial water volume or high water absorption of medicinal materials), extending the simmering time can increase the total amount of water evaporated, promote further dissolution of active ingredients, and thus improve the final concentration ratio.

[0084] When the deviation is negative and there is an indication of insufficient boiling, the amount of heat adjustment and the amount of high-heat duration extension can be calculated based on the absolute value of the deviation. The calculated heat adjustment and high-heat duration are then added to the heat parameters and high-heat duration in the second preparation strategy to generate the third preparation strategy. This clearly indicates insufficient energy input or inadequate boiling intensity during the decoction process, resulting in low efficiency in water evaporation and dissolution of active ingredients. In this case, the heat parameters can be adjusted to increase the heating power or raise the heating temperature setpoint, and the high-heat duration can be directly extended to ensure that the decoction can boil sufficiently and achieve the desired concentration effect.

[0085] Through the above technical solution, this application can monitor the temperature change trend during the decoction process in real time, and accurately identify the boiling stability range and duration of the decoction liquid using a temperature slope sequence and its smoothing process. By sensing the boiling state and further combining the duration of the boiling stability range with the deviation direction from the target concentration ratio, intelligent adjustment of the second preparation strategy can be achieved. For example, when the predicted concentration ratio is too high and there is excessive boiling, the heat can be reduced and the high-heat time shortened to avoid excessive concentration of the medicinal food; when the predicted concentration ratio is too low and there is insufficient boiling, the heat can be increased and the high-heat time extended to ensure sufficient dissolution of effective components and appropriate evaporation of water. Through the above real-time strategy adjustment based on the dynamic characteristics of the decoction process, the accuracy and adaptability of medicinal food preparation can be improved, effectively addressing the impact of uncertainties such as batch differences in medicinal materials and environmental fluctuations on the decoction effect, thereby ensuring that the final concentration ratio and effective component content of the medicinal food meet the preset standards, and improving the quality stability and automation level of medicinal food preparation.

[0086] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0087] In one embodiment, a multi-mechanism, fully automated medicinal diet preparation system is provided, which corresponds one-to-one with the multi-mechanism, fully automated medicinal diet preparation method described in the previous embodiment. This multi-mechanism, fully automated medicinal diet preparation system includes: The strategy matching module is used to obtain target formula information, match target medicinal materials and target food ingredients based on the target formula information, and match the first preparation strategy from the preset preparation strategy library based on the target formula information. The parameter adjustment module is used to match the effective component content of the target medicinal material and the target food based on the first preparation strategy, and adjust the first parameter and the second parameter of the first preparation strategy according to the matching result to generate the second preparation strategy. The first parameter includes the wetting parameter and the particle size parameter, and the second parameter includes the feeding parameter and the firepower parameter. The instruction generation module is used to generate processing instructions based on the second preparation strategy, send the processing instructions to the processing mechanism to preprocess the target medicinal materials and target food ingredients, and divide the preprocessed target medicinal materials into a first feeding group and a second feeding group. The sampling module is used to put the first feeding group and the target ingredients into the frying mechanism for frying according to the second preparation strategy, and to collect temperature curves, liquid level change curves and fire switching time sequences at a preset sampling period. The comparison module is used to generate a predicted concentration ratio based on the temperature curve and the liquid level change curve, and compare the predicted concentration ratio with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation. The strategy adjustment module is used to identify the boiling stability range based on the slope change characteristics of the temperature curve when the comparison deviation exceeds the preset deviation threshold, and adjust the second preparation strategy in real time according to the deviation direction between the duration of the boiling stability range and the target concentration ratio to generate the third preparation strategy. The strategy update module is used to add the second feeding group when the decoction time is reached, and to calculate the actual concentration ratio after the decoction is completed, and update the actual concentration ratio, the second preparation strategy and the third preparation strategy to the preparation strategy library.

[0088] Specific limitations regarding a multi-mechanism, fully automated medicinal cuisine preparation system can be found in the above description of a multi-mechanism, fully automated medicinal cuisine preparation method, and will not be repeated here. Each module in the aforementioned multi-mechanism, fully automated medicinal cuisine preparation system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0089] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A fully automated method for preparing medicinal cuisine involving multiple mechanisms, characterized in that, Including the following steps: Obtain target formula information, match target medicinal materials and target food ingredients based on target formula information, and match the first preparation strategy from the preset preparation strategy library based on target formula information; Based on the first preparation strategy, the effective component content of the target medicinal material and the target food is matched, and the first and second parameters of the first preparation strategy are adjusted according to the matching results to generate the second preparation strategy. The first parameters include the wetting parameters and the particle size parameters, and the second parameters include the feeding parameters and the heat parameters. Based on the second preparation strategy, a processing instruction is generated and sent to the processing mechanism to preprocess the target medicinal materials and target food ingredients, and the preprocessed target medicinal materials are divided into a first feeding group and a second feeding group. According to the second preparation strategy, the first feeding group and the target ingredients are put into the frying mechanism for frying, and the temperature curve, liquid level change curve and fire switching time sequence are collected at the preset sampling period. The predicted concentration ratio is generated based on the temperature curve and the liquid level change curve, and the predicted concentration ratio is compared with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation. When the comparison deviation exceeds the preset deviation threshold, the boiling stability range is identified based on the slope change characteristics of the temperature curve, and the second preparation strategy is adjusted in real time according to the deviation direction between the duration of the boiling stability range and the target concentration ratio, thus generating the third preparation strategy. When the decoction time is reached, the second feeding group is added, and the actual concentration ratio is calculated after the decoction is completed. The actual concentration ratio, the second preparation strategy, and the third preparation strategy are then updated to the preparation strategy library.

2. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 1, characterized in that: The steps of obtaining target formula information, matching target medicinal materials and target food ingredients based on the target formula information, and matching a first preparation strategy from a preset preparation strategy library based on the target formula information include: Obtain target formula information, which includes the name of medicinal material, dosage of medicinal material, name of food ingredient, dosage of food ingredient, and efficacy label; Based on the target formula information, match the corresponding target medicinal materials and target ingredients from the preset raw material library; Generate combinations of medicinal herb names and dosage ratios based on the target medicinal herbs and target food ingredients, and combine them with efficacy tags as strategy matching feature vectors; The similarity between the strategy matching feature vector and the formulation feature vector corresponding to each preparation strategy in the preparation strategy library is calculated, and the preparation strategy with the highest similarity is selected as the first preparation strategy.

3. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 1, characterized in that: The step of matching the effective component content of the target medicinal material and the target food ingredient based on the first preparation strategy, and adjusting the first and second parameters of the first preparation strategy according to the matching results to generate a second preparation strategy, wherein the first parameters include wetting parameters and particle size parameters, and the second parameters include feeding parameters and heat parameters, includes: Obtain batch information of the target medicinal material and extract effective component content data, including volatile oil content, polysaccharide content, alkaloid content and flavonoid content; The effective ingredient content data is compared with the standard effective ingredient content range associated with the first preparation strategy, and the content deviation rate of the effective ingredient is calculated. When the content deviation rate exceeds the preset deviation threshold, the corresponding adjustment parameter is selected from the first parameter and the second parameter according to the physicochemical properties of the active ingredient, and the adjustment range is determined according to the content deviation rate. Obtain the freshness grade and specifications of the target ingredient, extract the moisture content and fiber hardness information, and adjust the first parameter and the second parameter based on the moisture content and fiber hardness information; A second preparation strategy is generated based on the adjusted first and second parameters and linked to the batch information of the target formulation information.

4. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 3, characterized in that: The step of selecting a corresponding adjustment parameter from the first parameter and the second parameter based on the physicochemical properties of the active ingredient when the content deviation rate exceeds a preset deviation threshold, and determining the adjustment range based on the content deviation rate, includes: Extract the physicochemical property labels of the active ingredients, wherein the physicochemical property labels include volatile labels, water-soluble labels, and heat-sensitive labels; When the active ingredient has a volatile label, it is determined whether the content deviation rate of the volatile oil content exceeds the preset first deviation threshold. If it does, the first difference between the content deviation rate and the first deviation threshold is calculated, and the corresponding wetting parameter adjustment amount and feeding parameter adjustment amount are queried from the preset adjustment mapping table based on the first difference, and the corresponding medicinal material is assigned to the second feeding group. When the active ingredient has a water-soluble label, determine whether the content deviation rate of polysaccharide content or the content deviation rate of alkaloid content exceeds the preset second deviation threshold. If it does, calculate the second difference between the content deviation rate and the second deviation threshold, and query the corresponding particle size parameter adjustment amount and firepower parameter adjustment amount from the preset adjustment mapping table based on the second difference. When the active ingredient has a heat-sensitive label, it is determined whether the content deviation rate of flavonoids exceeds the preset third deviation threshold. If it does, the third difference between the content deviation rate and the third deviation threshold is calculated, and the corresponding adjustment amount of the heat parameter and the total decoction time are queried from the preset adjustment mapping table based on the third difference.

5. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 3, characterized in that: The steps of obtaining the freshness grade and specification information of the target ingredient, extracting moisture content and fiber hardness information, and adjusting the first parameter and the second parameter based on the moisture content and fiber hardness information include: Moisture content data is extracted from the freshness grade of the target ingredient, and the moisture content data is compared with the preset standard moisture content range to calculate the moisture deviation rate. When the moisture deviation rate is positive and exceeds the preset upper limit threshold of moisture, it is determined that the moisture content of the target ingredient is too high. The initial liquid level setting value in the fire power parameters is adjusted to reduce the amount of water added and extend the duration of the high heat stage. When the moisture deviation rate is negative and its absolute value exceeds the preset lower moisture threshold, it is determined that the moisture content of the target ingredient is too low. The initial liquid level setting in the fire parameters is adjusted to increase the amount of water added and shorten the duration of the high-heat stage.

6. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 1, characterized in that: The processing mechanism includes an immersion mechanism, a cutting mechanism, and a pulverizing mechanism. The step of generating processing instructions based on the second preparation strategy, sending the processing instructions to the processing mechanism to pre-treat the target medicinal materials and target ingredients, and dividing the pre-treated target medicinal materials into a first feeding group and a second feeding group includes: The texture attributes of the target medicinal material are obtained, and the type of medicinal material is determined based on the texture attributes. The types of medicinal materials include rhizomes, minerals, leaves, flowers, and fruits. When the medicinal material is a root or mineral, a soaking instruction is generated and sent to the soaking mechanism. The soaking instruction includes the soaking duration and the soaking temperature. When the medicinal material is a leaf, flower or fruit, it is determined whether the particle size of the target medicinal material is greater than the preset cutting threshold. If the particle size is greater than the preset cutting threshold, a cutting instruction is generated and sent to the cutting mechanism. The cutting instruction includes the slice thickness and the cut size. When the medicinal material is a leaf, flower, or fruit, if the particle size is less than or equal to the preset cutting threshold and greater than the preset pulverizing threshold, a pulverizing instruction is generated and sent to the pulverizing mechanism. The pulverizing instruction includes the pulverization mesh size.

7. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 1, characterized in that: The steps of feeding the first batch of ingredients and the target ingredients into the frying mechanism according to the second preparation strategy, and collecting temperature curves, liquid level change curves, and heat switching time sequences at a preset sampling period, include: Based on the heat parameters and the characteristics of the target medicinal materials, the decoction process is divided into a high-heat stage and a low-heat stage. The high-heat stage is associated with a high-heat temperature setpoint and a high-heat duration, while the low-heat stage is associated with a low-heat temperature setpoint. Based on the feeding parameters, a feeding sequence instruction is generated and sent to the feeding mechanism, so that the feeding mechanism feeds the first feeding group and the target ingredients into the pre-set frying mechanism with multiple temperature zones in the order associated with the feeding sequence instruction. During the high-temperature stage, the high-temperature setpoint is used as the target value of the PID controller. The PID controller calculates the first power adjustment based on the deviation between the current temperature and the high-temperature setpoint, the integral of the deviation, and the derivative of the deviation, and sends the first adjustment command to the cooking mechanism based on the first power adjustment. When the actual duration of the high-heat phase reaches the duration of the high-heat phase, record the moment of fire switching, use the low-heat temperature setpoint as the target value of the PID controller, and switch to the low-heat phase. The PID controller calculates the second power adjustment based on the deviation between the current temperature and the simmering temperature setpoint, the integral of the deviation, and the derivative of the deviation, and sends a second adjustment command to the simmering mechanism based on the second power adjustment. The real-time temperature and liquid level values ​​of the cooking mechanism are collected at a preset sampling period. A temperature curve is generated based on the real-time temperature value, and a liquid level change curve is generated based on the real-time liquid level value. The recorded fire switching time is also recorded in the fire switching time sequence.

8. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 1, characterized in that: The step of generating a predicted concentration ratio based on temperature and liquid level change curves, and comparing the predicted concentration ratio with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation, includes: The average temperature and temperature fluctuation range at the current sampling time are extracted based on the temperature curve, and the actual duration of the intense fire phase and the actual duration of the simmering fire phase are extracted based on the fire switching time sequence. The initial liquid level value and the current liquid level value are extracted based on the liquid level change curve. The current liquid level drop is calculated. Based on the current liquid level drop and the actual duration of the high heat stage and the actual duration of the low heat stage, the evaporation rate of the high heat stage and the evaporation rate of the low heat stage are calculated respectively. Based on the evaporation rates during the high-heat and low-heat stages, and combined with the remaining high-heat and low-heat times in the second preparation strategy, the predicted liquid level at the end of the decoction is calculated. The initial solid solution content of the target medicinal material is extracted from the second preparation strategy. Based on the initial liquid level value, the predicted liquid level value, and the initial solid solution content, the predicted concentration ratio is calculated according to the preset concentration ratio calculation model. The difference between the predicted concentration ratio and the target concentration ratio associated with the second preparation strategy is calculated to obtain the comparison deviation.

9. The fully automated preparation method for medicinal cuisine with multi-mechanism linkage according to claim 1, characterized in that: The step of identifying the boiling stability zone based on the slope change characteristics of the temperature curve when the comparison deviation exceeds a preset deviation threshold, and adjusting the second preparation strategy in real time according to the deviation direction between the duration of the boiling stability zone and the target concentration ratio to generate the third preparation strategy includes: When the absolute value of the comparison deviation exceeds the preset deviation threshold, the temperature slope sequence between adjacent sampling times is calculated based on the temperature curve, and the temperature slope sequence is averaged by a sliding window to obtain a smooth slope sequence. Based on the smooth slope sequence, a continuous period of time in which the absolute value of the slope is continuously lower than a preset slope threshold is identified. This continuous period of time is determined as the boiling stable interval, and the duration of the boiling stable interval is calculated. The duration of the boiling stability zone is compared with a preset duration threshold. Based on the comparison results and the deviation direction of the target concentration ratio, the second preparation strategy is adjusted to generate the third preparation strategy.

10. A fully automated medicinal diet preparation system with multi-mechanism linkage, characterized in that, include: The strategy matching module is used to obtain target formula information, match target medicinal materials and target food ingredients based on the target formula information, and match the first preparation strategy from the preset preparation strategy library based on the target formula information. The parameter adjustment module is used to match the effective component content of the target medicinal material and the target food based on the first preparation strategy, and adjust the first parameter and the second parameter of the first preparation strategy according to the matching result to generate the second preparation strategy. The first parameter includes the wetting parameter and the particle size parameter, and the second parameter includes the feeding parameter and the firepower parameter. The instruction generation module is used to generate processing instructions based on the second preparation strategy, send the processing instructions to the processing mechanism to preprocess the target medicinal materials and target food ingredients, and divide the preprocessed target medicinal materials into a first feeding group and a second feeding group. The sampling module is used to put the first feeding group and the target ingredients into the frying mechanism for frying according to the second preparation strategy, and to collect temperature curves, liquid level change curves and fire switching time sequences at a preset sampling period. The comparison module is used to generate a predicted concentration ratio based on the temperature curve and the liquid level change curve, and compare the predicted concentration ratio with the target concentration ratio associated with the second preparation strategy to obtain the comparison deviation. The strategy adjustment module is used to identify the boiling stability range based on the slope change characteristics of the temperature curve when the comparison deviation exceeds the preset deviation threshold, and adjust the second preparation strategy in real time according to the deviation direction between the duration of the boiling stability range and the target concentration ratio to generate the third preparation strategy. The strategy update module is used to add the second feeding group when the decoction time is reached, and to calculate the actual concentration ratio after the decoction is completed, and update the actual concentration ratio, the second preparation strategy and the third preparation strategy to the preparation strategy library.

Images