Intelligent electric heating control system for traditional Chinese medicine steam distillation

Abstract

The invention discloses an intelligent electric heating control system for traditional Chinese medicine steam distillation, relates to the technical field of traditional Chinese medicine steam distillation, and can solve the problems of low extraction precision, low resource utilization rate and poor product quality batch stability in the traditional Chinese medicine steam distillation process. Collecting real-time spectral data of the distillate; wherein the real-time spectral data comprises absorbance information on a plurality of continuous wavelengths; determining the specific weight of each wavelength according to the real-time spectral data; wherein the specific weight is used for representing the specificity of the wavelength on the spectral signal of the target traditional Chinese medicine effective component; for each wavelength, determining the wavelength end point probability of each wavelength according to the absorbance change trend of the wavelength; according to the specific weight and the wavelength end point probability, determining the comprehensive probability that the current moment is the distillation end point; and dynamically adjusting the electric heating power according to the comprehensive probability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traditional Chinese medicine steam distillation, and particularly relates to an intelligent electric heating control system for traditional Chinese medicine steam distillation. BACKGROUND

[0002] Traditional Chinese medicine steam distillation is a key process for extracting volatile active ingredients from traditional Chinese medicinal materials, and the accuracy of process control is directly related to the yield of active ingredients, energy consumption and the quality of the final product. The traditional distillation endpoint determination relies on fixed volume collection or fixed time control, which assumes that the distillation rate of active ingredients is uniform. However, in actual production, the differences in medicinal material types, batches and process fluctuations will lead to dynamic changes in the distillation of active ingredients, and the fixed endpoint standard is difficult to adapt to the actual extraction process.

[0003] In the prior art, in order to improve the control automation level, some schemes use intelligent control systems based on temperature, pressure or simple flow accumulation, but they are still essentially within the category of preset fixed thresholds. Although these methods realize the automatic start and stop of heating, they cannot perceive the quality changes of the internal components of the distillate, and their judgment basis is disconnected from the actual chemical component extraction process. SUMMARY

[0004] In order to solve the technical problems of low extraction precision, low resource utilization and poor batch stability of product quality in traditional Chinese medicine steam distillation process due to the reliance on fixed parameters irrelevant to component changes as the distillation endpoint judgment basis, the purpose of the present application is to provide an intelligent electric heating control system for traditional Chinese medicine steam distillation, and the technical scheme adopted is as follows:

[0005] In a first aspect, the present application provides an intelligent electric heating control system for traditional Chinese medicine steam distillation, comprising: a spectrum acquisition module, a specificity evaluation module, an endpoint evaluation module, and a power control module; the spectrum acquisition module is used to acquire real-time spectrum data of the distillate during the distillation process; wherein the real-time spectrum data includes absorbance information at a plurality of consecutive wavelengths; the specificity evaluation module is used to determine the specificity weight of each wavelength according to the real-time spectrum data; wherein the specificity weight is used to represent the specificity of the wavelength to the spectrum signal of the target traditional Chinese medicine active ingredient; the endpoint evaluation module is used to determine the wavelength endpoint probability of each wavelength according to the absorbance change trend of the wavelength for each wavelength; wherein the wavelength endpoint probability is used to represent the probability of the completion of the distillation process indicated based on the signal change of the wavelength at the current time; the endpoint evaluation module is also used to determine the comprehensive probability of the current time being the distillation endpoint according to the specificity weight and the wavelength endpoint probability; and the power control module is used to dynamically adjust the electric heating power according to the comprehensive probability.

[0006] In a possible implementation, when determining the specificity weight of each wavelength according to the real-time spectral data, the specificity evaluation module specifically performs the following steps: the specificity evaluation module is further configured to determine the signal response strength of each wavelength according to the average absorbance difference of each wavelength in the distillation heating period and the distillation period; the specificity evaluation module is further configured to determine the signal identity weight of each wavelength according to the similarity between each wavelength and the characteristic absorption peak wavelength of the target traditional Chinese medicine effective component; and the specificity evaluation module is further configured to determine the specificity weight of each wavelength according to the signal response strength and the signal identity weight.

[0007] In a possible implementation, when determining the signal response strength of each wavelength according to the average absorbance difference of each wavelength in the distillation heating period and the distillation period, the specificity evaluation module specifically performs the following steps: the specificity evaluation module is further configured to determine the end time of the distillation heating period as the time point at which the distillation process first reaches and stabilizes at the boiling point; the specificity evaluation module is further configured to determine, for each wavelength, the end time of the distillation period as the time point corresponding to the starting point of the first continuous downward trend of the absorbance of the wavelength after the end time of the distillation heating period; and the specificity evaluation module is further configured to determine the signal response strength of each wavelength according to the average absorbance of each wavelength in the distillation heating period and the average absorbance in the distillation period.

[0008] In a possible implementation, when determining the signal identity weight of each wavelength according to the similarity between each wavelength and the characteristic absorption peak wavelength of the target traditional Chinese medicine effective component, the specificity evaluation module specifically performs the following steps: the specificity evaluation module is further configured to calculate, for each wavelength, the wavelength difference absolute value between the wavelength and the characteristic absorption peak wavelength of each target traditional Chinese medicine effective component; and the specificity evaluation module is further configured to determine the signal identity weight according to the minimum value in the wavelength difference absolute value and a preset distance decay function.

[0009] In a possible implementation, when determining, for each wavelength, the wavelength end point probability of each wavelength according to the absorbance change trend of the wavelength, the end point evaluation module specifically performs the following steps: the end point evaluation module is further configured to determine, for each wavelength, the signal interference degree of each wavelength at each historical time according to the change trend of the absorbance ratio of each wavelength to the adjacent wavelength in the time period from the distillation transition period to the current time; the end point evaluation module is further configured to determine the signal availability of each wavelength at each historical time according to the signal interference degree, and construct an effective absorbance change sequence according to the signal availability of each wavelength at each historical time; the effective absorbance change sequence is an ordered set composed of the absorbance change values at each historical time filtered based on the signal availability; and the end point evaluation module is further configured to determine the wavelength end point probability of each wavelength according to the absorbance change value of each wavelength at the current time and the effective absorbance change sequence.

[0010] In a possible implementation, when determining the signal interference degree of each wavelength at each historical time according to the trend of the absorbance ratio of each wavelength to adjacent wavelengths from the distillation transition period to the current time, the endpoint evaluation module specifically performs the following steps: the endpoint evaluation module is further configured to calculate, for each wavelength and each historical time, a first absorbance ratio of the wavelength to the left adjacent wavelength and a second absorbance ratio of the wavelength to the right adjacent wavelength; and the endpoint evaluation module is further configured to determine the signal interference degree according to the first absorbance ratio and the second absorbance ratio.

[0011] In a possible implementation, when determining the comprehensive probability that the current time is the distillation endpoint according to the specificity weight and the wavelength endpoint probability, the endpoint evaluation module specifically performs the following steps: the endpoint evaluation module is further configured to determine, for each wavelength, a weighted probability contribution value of each wavelength according to the wavelength endpoint probability and the specificity weight; the weighted probability contribution value is used to represent the contribution degree of the wavelength to the comprehensive probability; and the endpoint evaluation module is further configured to determine the comprehensive probability that the current time is the distillation endpoint according to the weighted probability contribution values of all wavelengths and the specificity weights of all wavelengths.

[0012] In a possible implementation, when dynamically adjusting the electric heating power according to the comprehensive probability, the power control module specifically performs the following steps: the power control module is further configured to control the heater to operate in a first power mode when the comprehensive probability is less than a first preset threshold; the power control module is further configured to control the heater to switch to a second power mode in a smooth transition manner according to the comprehensive probability and a preset power adjustment algorithm when the comprehensive probability is greater than or equal to the first preset threshold and less than a second preset threshold; the second preset threshold is greater than the first preset threshold; and the power control module is further configured to control the heater to stop heating when the comprehensive probability is greater than or equal to the second preset threshold.

[0013] In a possible implementation, when controlling the heater to operate in the second power mode, the power control module specifically performs the following steps: the power control module is further configured to determine a real-time power adjustment factor according to the comprehensive probability and a preset fuzzy control rule base; the power control module is further configured to determine a real-time target heating power according to the power adjustment factor and a reference power corresponding to the first power mode; and the power control module is further configured to generate a control signal according to the real-time target heating power to drive the heater to perform power adjustment.

[0014] In one possible implementation, when the spectral acquisition module acquires real-time spectral data of the distillate, it specifically performs the following steps: the spectral acquisition module is also used to measure the distillate flow path at a fixed sampling frequency using an online infrared spectrometer; the spectral acquisition module is also used to continuously acquire and output real-time spectral data containing absorbance information within a continuous wavelength range during the distillation process.

[0015] Secondly, the present invention provides an intelligent electric heating control method for steam distillation of traditional Chinese medicine, comprising: acquiring real-time spectral data of the distillate during the distillation process; wherein the real-time spectral data includes absorbance information at multiple consecutive wavelengths; determining a specific weight for each wavelength based on the real-time spectral data; wherein the specific weight is used to characterize the specificity of the wavelength to the spectral signal of the target effective component of the traditional Chinese medicine; determining the wavelength endpoint probability for each wavelength based on the absorbance change trend; wherein the wavelength endpoint probability is used to characterize the probability that the distillation process is completed based on the signal change of the wavelength at the current moment; determining the comprehensive probability that the current moment is the distillation endpoint based on the specific weight and the wavelength endpoint probability; and dynamically adjusting the electric heating power based on the comprehensive probability.

[0016] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer-executable instructions, and when the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform the intelligent electric heating control method for steam distillation of traditional Chinese medicine as described in the first aspect and any possible implementation thereof.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by an electronic device of the present invention, cause the electronic device to perform the intelligent electric heating control method for steam distillation of traditional Chinese medicine as described in the first aspect and any possible implementation thereof.

[0018] Fifthly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the electronic device of the present invention to perform the intelligent electric heating control method for steam distillation of traditional Chinese medicine as described in the first aspect and any possible implementation thereof.

[0019] In a sixth aspect, the present invention provides a chip system applied to an intelligent electric heating control device for steam distillation of traditional Chinese medicine; the chip system includes one or more interface circuits and one or more processors. The interface circuits and the processors are interconnected via lines; the interface circuits are used to receive signals from the memory of the intelligent electric heating control device for steam distillation of traditional Chinese medicine and to send the signals to the processors, the signals including computer instructions stored in the memory. When the processor executes the computer instructions, the intelligent electric heating control device for steam distillation of traditional Chinese medicine performs the intelligent electric heating control method for steam distillation of traditional Chinese medicine as described in the first aspect and any possible design of the present invention.

[0020] This invention has the following beneficial effects: By monitoring the dynamic changes of distillate composition in real time using online spectral technology, and based on the specificity of each wavelength signal to the target effective component and its absorbance change trend, it intelligently calculates the comprehensive probability characterizing the possibility of the distillation endpoint. This replaces the traditional experience-based judgment of fixed volume or time, achieving accurate and dynamic determination of the distillation endpoint and adaptive adjustment of heating power. This effectively avoids energy waste and time loss caused by over-distillation, while preventing water-soluble impurities from dissolving into the distillate later. It ensures the quality purity and batch stability of the Chinese herbal extract from the source of the process, and improves the intelligence level and product quality control capabilities of the entire steam distillation process. Attached Figure Description

[0021] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the architecture of an intelligent electric heating control system for steam distillation of traditional Chinese medicine, provided in one embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the architecture of a specificity evaluation module provided in one embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram of the architecture of an endpoint evaluation module provided in one embodiment of the present invention;

[0025] Figure 4 This is one of the flowcharts illustrating an intelligent electric heating control method for steam distillation of traditional Chinese medicine, provided in an embodiment of the present invention.

[0026] Figure 5 This is a second schematic flowchart of an intelligent electric heating control method for steam distillation of traditional Chinese medicine, provided as an embodiment of the present invention.

[0027] Figure 6 This is a third schematic flowchart of an intelligent electric heating control method for steam distillation of traditional Chinese medicine, provided as an embodiment of the present invention.

[0028] Figure 7 This is the fourth flowchart illustrating an intelligent electric heating control method for steam distillation of traditional Chinese medicine, provided as an embodiment of the present invention. Detailed Implementation

[0029] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0031] The following description, in conjunction with the accompanying drawings, details a specific solution for an intelligent electric heating control system for steam distillation of traditional Chinese medicine provided by the present invention.

[0032] For example, such as Figure 1 The diagram shown is a schematic representation of the architecture of an intelligent electric heating control system (hereinafter referred to as the electric heating control system) for steam distillation of traditional Chinese medicine, according to an embodiment of the present invention. The electric heating control system 10 includes: a spectrum acquisition module 11, a specificity evaluation module 12, an endpoint evaluation module 13, and a power control module 14. The modules are described below in sequence:

[0033] (1) Spectral acquisition module 11.

[0034] The spectral acquisition module 11 is responsible for acquiring real-time spectral data of the distillate during the distillation process, providing the most original information on component changes for subsequent specificity assessment and endpoint probability assessment.

[0035] Optionally, the spectral acquisition module 11 is used to acquire real-time spectral data. The real-time spectral data is used to characterize the absorbance information of the distillate at multiple consecutive wavelengths.

[0036] Specifically, the spectral acquisition module 11 uses an integrated online infrared spectrometer, with its detection probe installed in the condensate outlet pipeline of the distillation apparatus. This module continuously measures the distillate flowing through the probe at a fixed sampling frequency of once per second, and simultaneously outputs spectral data containing the correspondence between wavelength and absorbance. This data is transmitted in real time to the specificity evaluation module 12 and the endpoint evaluation module 13, forming the data foundation for the entire intelligent judgment and control system.

[0037] (2) Specificity assessment module 12.

[0038] The specificity assessment module 12 is responsible for analyzing and determining the specific weight of each wavelength to the spectral signal of the target traditional Chinese medicine active ingredient based on the real-time spectral data output by the spectral acquisition module 11. This weight will be used in the endpoint assessment module 13 to measure the reliability and contribution of different wavelength signals.

[0039] Optionally, the specificity assessment module 12 is used to determine the specificity weight of each wavelength based on real-time spectral data; wherein the specificity weight is used to characterize the specificity of the wavelength to the spectral signal of the target Chinese medicine active ingredient.

[0040] For example, such as Figure 2 As shown, the specificity assessment module 12 may include three sub-modules: response intensity analysis sub-module 121, identity weight analysis sub-module 122, and weight fusion sub-module 123. These three sub-modules are described below:

[0041] (2.1) Response intensity analysis submodule 121.

[0042] Optionally, the response intensity analysis submodule 121 is used to determine the signal response intensity of each wavelength based on the difference in average absorbance during the distillation heating period (the stage from the start of heating until the system temperature first reaches and stabilizes at the boiling point, during which very little distillate is produced) and the main distillation period (the stage after boiling stabilizes, during which a large amount of the target active ingredient is distilled out and its concentration reaches its peak).

[0043] Specifically, this submodule first uses the moment when the system temperature first reaches and stabilizes at the boiling point as the common end point of the "distillation heating period" for all wavelengths. Then, for each wavelength, it scans its absorbance data to find the starting point where a continuous downward trend first appears after this marker moment, as the end point of the main distillation period specific to that wavelength. Next, it calculates the average absorbance of each wavelength during the initial heating phase (representing the background noise level) and its average absorbance during the main distillation period (representing the signal peak level). Finally, it determines the signal response intensity for that wavelength based on the normalized difference between these two average absorbances. A higher response intensity indicates a more significant change in absorbance during the distillation process for that wavelength.

[0044] (2.2) Identity weight analysis submodule 122.

[0045] Optionally, the identity weight analysis submodule 122 is used to determine the signal identity weight of each wavelength based on the similarity between each wavelength and the characteristic absorption peak wavelength of the target Chinese medicine active ingredient.

[0046] Specifically, this submodule pre-stores the standard infrared characteristic absorption peak wavelengths of the target traditional Chinese medicine's active ingredients. For each current wavelength in the real-time spectrum, it calculates the absolute value of the difference between it and each standard characteristic absorption peak wavelength, and selects the minimum value. Then, it calculates the signal identity weight based on this minimum value using a preset distance decay function (such as an exponential decay function). The closer the wavelength is to the standard peak, the higher the weight value, indicating that the signal at that wavelength originates from the chemical identity of the target active ingredient is more reliable.

[0047] (2.3) Weight fusion submodule 123.

[0048] Optionally, the weight fusion submodule 123 is used to perform fusion calculations based on the signal response intensity and signal identity weights to ultimately determine the specific weights for each wavelength.

[0049] Specifically, this submodule multiplies the signal response intensity by the signal identity weight to obtain a comprehensive specificity weight. This weight takes into account both the significance of the signal change and the accuracy of the signal source; the higher the value, the greater the value of the wavelength signal in determining the distillation endpoint of the target effective component. The specificity weights of each wavelength output by the specificity evaluation module 12 are transmitted to the endpoint evaluation module 13 for subsequent weighted fusion.

[0050] (3) Endpoint assessment module 13.

[0051] The endpoint evaluation module 13 is responsible for comprehensively utilizing spectral data and specific weights to dynamically evaluate whether the current moment is the distillation endpoint and output a comprehensive probability value.

[0052] Optionally, the endpoint evaluation module 13 is used to determine the wavelength endpoint probability for each wavelength based on the absorbance change trend of the wavelength. The wavelength endpoint probability is used to characterize the probability that the distillation process is completed based on the signal change of the wavelength at the current moment. It is also used to determine the comprehensive probability that the current moment is the distillation endpoint based on the specific weight and the wavelength endpoint probability.

[0053] For example, such as Figure 3 As shown, the endpoint evaluation module 13 may include two sub-modules: a wavelength endpoint probability calculation sub-module 131 and a comprehensive probability fusion sub-module 132. These sub-modules are described below:

[0054] (3.1) Wavelength endpoint probability calculation submodule 131.

[0055] Optionally, the wavelength endpoint probability calculation submodule 131 is used to determine the wavelength endpoint probability of each wavelength based on the absorbance variation trend of the wavelength.

[0056] Specifically, the wavelength endpoint probability calculation submodule 131 is the first stage of endpoint assessment, responsible for independently analyzing and judging each wavelength. First, it analyzes the stability of each wavelength signal from the distillation transition period to the current time (assessing the degree of interference by the change in the absorbance ratio of its adjacent wavelengths), thereby determining the reliability (signal availability) of the historical data at each time point. Second, it constructs an effective absorbance change sequence for each wavelength based on the reliable data. Finally, by analyzing the comparison between the absorbance change value at the current time and this historical sequence, it calculates the probability of the distillation process completion indicated by the wavelength signal itself, i.e., the wavelength endpoint probability. The wavelength endpoint probability calculation submodule 131 outputs an independent probability value for each wavelength as the basis for subsequent fusion decisions.

[0057] (3.2) Integrated probability fusion submodule 132.

[0058] Optionally, the comprehensive probability fusion submodule 132 is used to determine the comprehensive probability that the current time is the distillation endpoint based on the specific weight and the wavelength endpoint probability.

[0059] Specifically, the comprehensive probability fusion submodule 132 is the second stage of endpoint evaluation, responsible for integrating the evaluation results of all wavelengths to make a final decision. The comprehensive probability fusion submodule 132 receives the specific weights of each wavelength from the specificity evaluation module 12, and the wavelength endpoint probabilities of each wavelength from the wavelength endpoint probability calculation submodule 131. It first multiplies the wavelength endpoint probability of each wavelength by its specific weight to obtain the weighted probability contribution value of that wavelength, reflecting the principle that "wavelengths with high specificity weights have more important opinions." Subsequently, it sums the weighted probability contribution values ​​of all wavelengths and divides them by the sum of the specific weights of all wavelengths, calculating the final, unified comprehensive probability through a weighted average. This comprehensive probability is the system's final quantitative judgment on "whether the current moment is the distillation endpoint," and is directly output to the power control module 14 to trigger the corresponding heating control action.

[0060] (4) Power control module 14.

[0061] The power control module 14 is responsible for intelligently adjusting the power of the electric heater based on the comprehensive probability calculated in real time by the endpoint evaluation module 13, and automatically stopping heating when the distillation is determined to be complete, thereby achieving precise endpoint control and energy saving.

[0062] Optionally, the power control module 14 is used to dynamically adjust the electric heating power according to the overall probability, and to stop heating when the overall probability reaches a preset threshold.

[0063] Specifically, the power control module 14 has two preset probability thresholds. When the overall probability is below the first preset threshold, it indicates that the active ingredient is distilling out in large quantities, and the module controls the heater to maintain a high power output to ensure extraction efficiency. When the overall probability is between the first preset threshold and a higher second preset threshold, it indicates that distillation is entering its final stage, and the module initiates a power reduction strategy, such as calling a fuzzy PID control algorithm to smoothly reduce the heating power based on the probability value, achieving a smooth transition. When the overall probability reaches or exceeds the second preset threshold, the module determines that distillation is complete and immediately cuts off the heating power. This dynamic control based on process quality signals fundamentally avoids over-distillation caused by the fixed-volume method, saving energy while ensuring the quality of the distillate.

[0064] The above describes the intelligent electric heating control system 10 used for steam distillation of traditional Chinese medicine and its included modules.

[0065] For example, such as Figure 4 The diagram shown is a schematic flow chart of an intelligent electric heating control method for steam distillation of traditional Chinese medicine according to an embodiment of the present invention, including the following steps:

[0066] S401. During the distillation process, real-time spectral data of the distillate is collected. This real-time spectral data includes absorbance information at multiple consecutive wavelengths.

[0067] For example, this step can be performed by the spectral acquisition module 11 in the electric heating control system 10 described above, specifically including: measuring the distillate flow path with an online infrared spectrometer at a fixed sampling frequency, and continuously acquiring and outputting real-time spectral data containing absorbance information in a continuous wavelength range during the distillation process.

[0068] Specifically, the spectral acquisition module 11 uses an integrated online infrared spectrometer, with its detection probe installed in the condensate outlet line of the distillation apparatus. This module continuously measures the distillate flowing through the probe at a fixed sampling frequency of once per second, and simultaneously outputs spectral data containing the correspondence between wavelength and absorbance.

[0069] S402. Based on real-time spectral data, determine the specificity weight for each wavelength. The specificity weight characterizes the specificity of the wavelength to the spectral signal of the target herbal active ingredient.

[0070] For example, this step can be performed by the specificity evaluation module 12 in the electrically heated control system 10 described above. Specifically, it includes: First, analyzing the difference in average absorbance for each wavelength during the distillation heating period (the stage from the start of heating until the system temperature first reaches and stabilizes at the boiling point, during which very little distillate is produced) and the main distillation period (the stage after boiling stabilizes, when a large amount of the target active ingredient is distilled off and its concentration reaches its peak) to evaluate its signal response intensity. Second, evaluating the proximity of each wavelength to the pre-stored characteristic absorption peaks of the target traditional Chinese medicine active ingredient to determine its signal identity weight. Finally, fusing the signal response intensity and signal identity weight to determine the specificity weight of each wavelength. It should be noted that the specific process for determining the specificity weight of each wavelength is described in S501-S503 below and will not be repeated here.

[0071] In another possible implementation, the specificity evaluation module 12 may also employ an evaluation method based on the slope of signal change when determining the specificity weight of each wavelength. This method does not directly rely on the precise division between the initial heating phase and the distillation phase, but rather measures the response sensitivity of a wavelength to the distillation process by calculating the maximum rising slope of the absorbance curve for each wavelength within a preset time window; the greater the slope, the higher the response intensity.

[0072] Alternatively, the specificity assessment module 12 can also use a standard spectral profile similarity matching method when determining the signal identity weight for each wavelength. This method not only considers the distance between the wavelength and a single absorption peak, but also calculates the similarity between the spectral profile (absorbance distribution shape) of a small band containing the current wavelength and the spectral profile of the standard effective component (e.g., calculating the correlation coefficient), and uses the similarity calculation result as the signal identity weight.

[0073] Therefore, the specificity assessment module 12 comprehensively considers the intensity of the dynamic change of the wavelength signal during the distillation process and its degree of conformity with the spectral characteristics of the target component, and assigns a quantified specificity weight to each wavelength, thereby identifying the characteristic wavelength that is most indicative of the distillation endpoint.

[0074] S403. For each wavelength, determine the wavelength endpoint probability based on the absorbance change trend. The wavelength endpoint probability characterizes the probability that the distillation process is complete, indicated by the signal change at the current moment based on the wavelength.

[0075] For example, this step can be performed by the endpoint evaluation module 13 in the electric heating control system 10 described above, specifically including: First, analyzing the stability of each wavelength signal since the transition period, assessing the degree of interference from sudden impurities, and determining the reliability (availability) of the historical data at each time point. Second, constructing an effective sequence reflecting the steady change trend of the absorbance of that wavelength based on the reliable historical data. Finally, by comparing the current absorbance change value with the statistical relationship of the effective sequence, assessing the probability that the wavelength signal indicates the end of distillation. It should be noted that the specific process for determining the wavelength endpoint probability of each wavelength is described in S601-S603 below, and will not be repeated here.

[0076] In another possible implementation, the endpoint evaluation module 13 may also employ a wavelet transform-based noise separation method when determining the specific weight for each wavelength. This method separates the low-frequency components representing signal trends and the high-frequency components representing noise and sudden interference by performing wavelet decomposition on the wavelength absorbance time series data. The degree of interference to the signal is quantified by the energy magnitude of the high-frequency components.

[0077] In another possible implementation, the endpoint evaluation module 13 can also employ a robust statistical method based on a moving time window when constructing an effective absorbance change sequence. This method does not pre-select time points, but instead uses the median rather than the average to calculate the absorbance change value within a moving time window, thus naturally resisting the influence of a few outlier data points (interference) and directly constructing a robust change sequence.

[0078] Therefore, the endpoint assessment module 13 calculates a probability value for each wavelength that represents the likelihood of its own signal indicating the distillation endpoint by filtering out interference, extracting reliable trends, and analyzing the anomalies of the current changes.

[0079] S404. Determine the overall probability that the current time is the distillation endpoint based on the specificity weight and the wavelength endpoint probability.

[0080] For example, this step can be performed by the endpoint evaluation module 13 in the electric heating control system 10 described above, specifically including: First, multiplying the wavelength endpoint probability of each wavelength by its specific weight to obtain the weighted probability contribution value of that wavelength. Second, calculating the sum of the weighted probability contribution values ​​of all wavelengths and dividing it by the sum of the specific weights of all wavelengths, and obtaining the final comprehensive probability by weighted averaging. It should be noted that the specific process for determining the wavelength endpoint probability of each wavelength is described in S701-S702 below, and will not be repeated here.

[0081] In another possible implementation, the endpoint evaluation module 13 can also use a voting mechanism to determine the overall probability that the current moment is the distillation endpoint. This mechanism sets a voting threshold based on specific weights for each wavelength. When the endpoint probability of a wavelength exceeds its own threshold, it is counted as a "agree to end" vote. The final overall probability is determined by the ratio of the sum of the specific weights corresponding to the "agree to end" votes to the total sum of the specific weights of all wavelengths.

[0082] Therefore, the endpoint evaluation module 13 integrates the independent judgments of all characteristic wavelengths and fuses them with the specific weight of each wavelength as the basis for its judgment reliability, and finally outputs a unified, robust comprehensive probability value that characterizes whether the entire distillation process has reached the endpoint.

[0083] S405. Dynamically adjust the electric heating power based on the overall probability.

[0084] For example, this step can be performed by the power control module 14 in the electric heating control system 10 described above, and specifically includes the following steps:

[0085] (1) When the overall probability is less than the first preset threshold, control the heater to operate in the first power mode.

[0086] Specifically, in this case, the overall probability is low, indicating that the target active ingredient is still being distilled out in large quantities. The power control module 14 controls the heater to maintain a high and stable heating power (first power mode) to ensure extraction efficiency.

[0087] For example, the first preset threshold is set to 0.6, corresponding to a probability threshold of 85% distillation yield of the effective component. Alternatively, in practical applications, it can be determined based on statistical analysis of multiple batches of experimental data conducted in the region where the distillation process takes place.

[0088] (2) When the overall probability is greater than or equal to the first preset threshold and less than the second preset threshold, the heater is controlled to switch to the second power mode in a smooth transition manner according to the overall probability and the preset power adjustment algorithm. Wherein, the second preset threshold is greater than the first preset threshold.

[0089] Specifically, in this case, the overall probability enters the middle range, indicating that distillation is entering its final stage. The power control module 14 calls a preset fuzzy PID control algorithm, using the overall probability as input, to dynamically calculate a power adjustment factor. Based on this factor and the reference power in the first power mode, the real-time target heating power is calculated, and a control signal is generated to drive the heater to smoothly reduce its power, achieving a transition from the first power mode to a lower-power second power mode.

[0090] For example, the second preset threshold is set to 0.9, corresponding to a probability threshold of 99% distillation yield of the effective component. Alternatively, in practical applications, it can be determined based on statistical analysis of multiple batches of experimental data conducted in the region where the distillation process takes place.

[0091] (3) When the overall probability is greater than or equal to the second preset threshold, control the heater to stop heating.

[0092] Specifically, in this case, if the overall probability reaches or exceeds the second preset threshold, the system determines that the distillation of the effective component is complete. The power control module 14 immediately issues a command to cut off the power supply to the heater, automatically terminating the distillation process.

[0093] Based on the above technical solution, this invention uses online spectroscopy to monitor the dynamic changes of distillate composition in real time. Based on the specificity of each wavelength signal to the target effective component and the trend of its absorbance change, it intelligently calculates the comprehensive probability characterizing the possibility of the distillation endpoint. This replaces the traditional experience-based judgment of fixed volume or time, achieving accurate and dynamic determination of the distillation endpoint and adaptive adjustment of heating power. This effectively avoids energy waste and time loss caused by over-distillation, while preventing water-soluble impurities from dissolving into the distillate later. It ensures the quality purity and batch stability of the Chinese herbal extract from the source of the process, and improves the intelligence level and product quality control capabilities of the entire steam distillation process.

[0094] For example, in combination Figure 4 ,like Figure 5 The diagram shown is a flowchart illustrating another intelligent electric heating control method for steam distillation of traditional Chinese medicine according to an embodiment of the present invention. In this method, the specific weight of each wavelength is determined based on real-time spectral data, specifically including the following steps:

[0095] S501. Determine the signal response intensity of each wavelength based on the difference in average absorbance between the heating and distillation periods.

[0096] For example, this step is performed by the response intensity analysis submodule 121 in the specificity assessment module 12 described above, and specifically includes the following steps:

[0097] (1) The moment when the boiling point is first reached and stabilized during the distillation process is determined as the end time of the distillation heating period.

[0098] Specifically, the response intensity analysis submodule 121 uses the initial heating phase ending when the system temperature first reaches and stabilizes at the boiling point as a common marker for all wavelengths. T At this point, the time period corresponding to the distillation heating period is [0, T ].

[0099] (2) For each wavelength, the time corresponding to the starting point when the absorbance of the wavelength first shows a continuous downward trend after the end of the distillation heating period is determined as the end time of the distillation period.

[0100] In this sub-step, the response intensity analysis submodule 121 targets each wavelength. Lambda The absorbance was scanned in the distillate data. If a continuous decreasing trend in absorbance was observed over 10 consecutive monitoring periods, the starting point of this trend was defined as the end of the distillation period specific to that wavelength. At this point, the time period corresponding to the distillation period of a single wavelength λ is [ , If a wavelength does not meet the above conditions, then the signal response intensity of that wavelength to the effective Chinese medicinal material component signal is determined to be 0.

[0101] (3) Determine the signal response intensity of each wavelength based on the average absorbance during the distillation heating period and the average absorbance during the distillation period.

[0102] For example, the response intensity analysis submodule 121 calculates the signal response intensity for each wavelength using the following formula: ;

[0103] In the above formula, This represents the signal response intensity at wavelength λ. This represents the maximum and minimum value normalization function, used to map the difference to the [0, 1] interval, facilitating cross-wavelength comparison; This indicates that the wavelength λ is during the distillation period [ , Average absorbance within ]; This represents the average absorbance of wavelength λ during the distillation heating period [0, T].

[0104] Understandably, according to the Lambert-Beer Law, the absorbance of a solution is directly proportional to the concentration of its active ingredient in the above formula. During the heating phase of distillation, the concentration of the active ingredient in the distillate is extremely low, and the absorbance is close to the baseline level; after entering the distillation phase, the concentration of the active ingredient increases sharply. Therefore, the greater the difference in average absorbance between the distillation phase and the heating phase, the stronger the response of that wavelength to changes in component concentration, i.e., the higher the signal response intensity. By normalization, the response intensity value of that wavelength relative to other wavelengths is obtained.

[0105] S502. Determine the signal identity weight of each wavelength based on the similarity between each wavelength and the characteristic absorption peak wavelength of the target Chinese medicine active ingredient.

[0106] For example, this step is performed by the identity weight analysis submodule 122 in the specificity assessment module 12 described above, and specifically includes the following steps:

[0107] (1) For each wavelength, calculate the absolute value of the wavelength difference between the wavelength and the characteristic absorption peak wavelength of each target Chinese medicine active ingredient.

[0108] For example, the identity weight analysis submodule 122 targets the first [aspect] on the spectrum. Lambda The wavelengths were calculated to correlate with the effective components of traditional Chinese medicine in the spectrum of the first wavelength. i wavelength corresponding to each absorption peak absolute value of wavelength difference .

[0109] (2) Determine the signal identity weight based on the minimum value of the absolute value of the wavelength difference and the preset distance attenuation function.

[0110] Furthermore, the identity weight analysis submodule 122 calculates the signal identity weight for each wavelength λ using the following formula: ;

[0111] In the above formula, The signal identity weight represents the wavelength λ; This represents an exponential function with the natural constant e as its base, used to achieve distance decay; This represents the minimum absolute value of the differences between wavelength λ and the characteristic absorption peak wavelengths of all target effective components; The wavelength of the characteristic absorption peak of the i-th target Chinese herbal medicine active ingredient is represented.

[0112] It should be noted that the above formula is intended to assess the correctness of the wavelength signal source. Even if a wavelength has a high response intensity, it may be responding to impurities rather than the target effective component. By calculating the minimum wavelength distance between this wavelength and the standard absorption peak of the known target component, and inputting it into an exponential decay function, a weight reflecting its reliability can be obtained: the closer to the standard peak, the higher the weight (approaching 1); the farther the distance, the lower the weight (approaching 0). This weight serves as a measure of confidence in whether the wavelength signal source is the target component.

[0113] S503. Determine the specific weight of each wavelength based on the signal response strength and signal identity weight.

[0114] For example, this step is performed by the weighted fusion submodule 123 in the specificity evaluation module 12 described above, specifically including: the weighted fusion submodule 123 converting the signal response intensity of wavelength λ. Its signal identity weight Multiplying these values ​​yields the specific weight of the wavelength for the signal of the target traditional Chinese medicine's active ingredient. : ;

[0115] It should be noted that the above formula combines the wavelength response intensity (reflecting its sensitivity to changes in composition) with "identity weight" (reflecting whether the signal source is the target component). Through multiplicative fusion, a comprehensive specificity weight is obtained. This weight considers both the amplitude of signal changes and the reliability of chemical identity, thus enabling more accurate screening of characteristic wavelengths that are both sensitive to the target component and actually originate from it, providing a reliable basis for the subsequent weighted calculation of endpoint probabilities.

[0116] Based on the above technical solution, this embodiment of the invention analyzes real-time spectral data of the distillate, sequentially calculates the signal response intensity and signal identity weight for each wavelength, and finally fuses them to obtain the specificity weight. This method can accurately identify those signals that are sensitive and reliable to the effective components of the target traditional Chinese medicine from a large number of spectral wavelengths, laying a scientific and objective data foundation for subsequent calculation of the distillation endpoint probability based on multi-wavelength weighted fusion, thereby achieving intelligent and precise control of the distillation process.

[0117] For example, in combination Figure 4 ,like Figure 6 The diagram shown is a flowchart illustrating another intelligent electric heating control method for steam distillation of traditional Chinese medicine according to an embodiment of the present invention. In this method, for each wavelength, the wavelength endpoint probability is determined based on the absorbance variation trend, specifically including the following steps:

[0118] S601. For each wavelength, based on the changing trend of the absorbance ratio of each wavelength to the adjacent wavelength during the time period from the start of the distillation transition period to the current time, determine the degree of signal interference at each historical moment for each wavelength.

[0119] For example, this step can be performed by the wavelength endpoint probability calculation submodule 131 in the endpoint evaluation module 13 described above, specifically including the following steps:

[0120] (1) For each wavelength and each historical moment, calculate the first absorbance ratio of the wavelength to the wavelength adjacent to the left and the second absorbance ratio of the wavelength to the wavelength adjacent to the right.

[0121] Specifically, the wavelength endpoint probability calculation submodule 131, for the λ-th wavelength in the spectrum, obtains the following absorbance data at the t-th time (t≥1) from the start of the distillation transition period to the current time: , representing the absorbance at the current wavelength λ at time t; , representing the absorbance of the adjacent wavelength λ-1 on the left at time t; This represents the absorbance of the adjacent wavelength λ+1 at time t. Simultaneously, the absorbance data for the previous time t-1 is also obtained. , representing the absorbance data of wavelength λ, the left adjacent wavelength λ-1, and the right adjacent wavelength λ+1 at time t-1, respectively.

[0122] It should be noted that the "adjacent wavelength to the left" mentioned above refers to the wavelength in the spectral sequence whose index is one less than the current wavelength (i.e., the λ-th wavelength). The index of the adjacent wavelength to the left is denoted as λ-1. Similarly, the adjacent wavelength to the right refers to the wavelength in the spectral sequence whose index is one greater than the current wavelength (i.e., the λ-th wavelength). The index of the adjacent wavelength to the right is denoted as λ+1. The spectral sequence refers to a discrete sequence arranged in ascending order of wavelength values ​​based on real-time spectral data. Each wavelength has a unique index; for example, the wavelength index currently being calculated is λ.

[0123] (2) Determine the degree of interference to the signal based on the first absorbance ratio and the second absorbance ratio.

[0124] Specifically, the wavelength endpoint probability calculation submodule 131 calculates the degree of interference to the signal at time t using the following formula. : ;

[0125] In the above formula, It represents the degree of interference of the signal with wavelength λ at historical time t. The larger the value, the more likely the signal with wavelength λ is to be interfered with by impurities at that time. This indicates the fluctuation in the ratio of the absorbance of wavelength λ to its left-adjacent wavelength λ-1 from time t-1 to time t. This indicates the fluctuation in the ratio of absorbance of wavelength λ to its adjacent wavelength λ+1 to the right of wavelength λ from time t-1 to time t. This represents the maximum and minimum value normalization function, which normalizes the sequence of B values ​​for all times with the same wavelength λ from the start of the distillation transition period to the current time, mapping the values ​​to the interval [0,1] for easy comparison across time periods; The first absorbance ratio at time t (current wavelength and the adjacent wavelength to the left); This represents the first absorbance ratio at time t-1; The second absorbance ratio at time t (the ratio of the adjacent wavelength on the right to the current wavelength); This represents the second absorbance ratio at time t-1. For the term in the denominator of the formula, the absorbance at each wavelength is positive within the effective spectral measurement range, therefore the denominator is not zero.

[0126] It is understandable that the absorbance changes of the absorption peak of the target active ingredient between adjacent wavelengths are usually continuous and correlated, and their ratio should remain relatively stable over a short period of time. If a wavelength is interfered with by impurities at a specific moment, its absorbance will undergo a specific abrupt change, causing a drastic jump in the absorbance ratio between that wavelength and adjacent wavelengths. The above formula measures the severity of this ratio jump by calculating the absolute difference between the first and second absorbance ratios at the current moment and the previous moment, respectively. After normalization, the ratio jumps in the two directions are averaged to obtain the overall signal interference level. The higher this value, the less reliable the signal at that wavelength at the current moment is, and the more likely it is to be affected by impurities.

[0127] S602. Based on the degree of signal interference, determine the signal availability of each wavelength at each historical moment, and construct an effective absorbance change sequence based on the signal availability of each wavelength at each historical moment. The effective absorbance change sequence is an ordered set of absorbance change values ​​at each historical moment, selected based on signal availability.

[0128] In this step, the wavelength endpoint probability calculation submodule 131 first calculates the signal interference level based on S601. Determine the wavelength λ for each wavelength from the start of the distillation transition period to the current time. (The current time is a selected time to be evaluated as the end of distillation.) Signal availability at each historical time t within this period. : ;

[0129] In the above formula, This indicates the signal availability of wavelength λ at historical time t. The larger the value, the higher the overall reliability of the signal at that wavelength, accumulated from the start of the transition period to historical time t. The index represents the traversal index, indicating the l-th time from the start of the distillation transition period (l=1) to the historical time t, and its value ranges from 1 to t; The degree of interference to the signal with wavelength λ at time l is represented by S601; This represents an exponential function with the natural constant e as its base. Its function is to map the degree of disturbance to the contribution to availability; the higher the degree of disturbance, the lower the contribution. This indicates that the availability contribution of time l is time-weighted, and the closer the time is to the historical time t (the larger l is), the higher its weight.

[0130] It should be noted that since impurity interference is often persistent, this formula assesses the overall reliability of the wavelength signal by calculating the cumulative availability from the start of the transition period to historical time t. For each time l between the start of the distillation transition period (l=1) and historical time t, the degree of interference is... After exponential decay, it transforms into a positive availability contribution value. Then, the contribution value is multiplied by a time-weighted coefficient. This emphasizes the impact of more recent moments. Finally, the weighted contribution values ​​of all moments are summed to obtain the cumulative signal availability at historical moment t. This value comprehensively reflects the continuous impact and temporal proximity of historical interference; the higher the availability, the more reliable the signal at that wavelength in the current and subsequent moments.

[0131] Furthermore, the wavelength endpoint probability calculation submodule 131 calculates the signal availability. For each wavelength λ, select the wavelengths from the start of the distillation transition period to the current time. Within the range of (excluding the current time), all historical moments t in which the signal availability is greater than the preset availability threshold constitute a set of high-availability historical moments, denoted as . For sets For each selected time t, calculate the change in absorbance. All of these Arranged in chronological order according to the corresponding time t, this constitutes the effective absorbance change sequence for wavelength λ. This sequence excludes data from severely disrupted or unreliable moments, and only includes absorbance changes calculated based on high-availability signals.

[0132] For example, the availability threshold can be set to 0.5 to filter out reliable data with less than 50% interference, thus ensuring the validity of the sequence; or it can be determined according to the needs in actual application.

[0133] S603. Determine the wavelength endpoint probability for each wavelength based on the absorbance change value and effective absorbance change sequence at the current moment.

[0134] Specifically, the wavelength endpoint probability calculation submodule 131 calculates the wavelength λ at the current time using the following formula. wavelength endpoint probability : ;

[0135] In the above formula, This indicates that the wavelength λ is at the current moment. The wavelength endpoint probability, the closer its value is to 1, the more likely the component corresponding to that wavelength has been completely distilled; The set of highly available historical moments selected from S602 corresponding to the effective absorbance variation sequence at wavelength λ is represented by t, where t represents the set to be traversed. A historical moment in history; The overall normalized weight represents the proportion of importance of the signal at that time in the overall high-confidence historical signal. In practical applications, since the spectrum includes multiple complex wavelengths, the denominator of this term will not be zero. This indicates that the minimum value within the parentheses is taken; Indicates the current time The absorbance change value (forward difference); This represents the change in absorbance at historical time t. These are parameter tuning coefficients, and should be extremely small positive numbers, for example, 10 to the power of -10, to prevent the denominator from being too large. A value of zero renders the calculation meaningless.

[0136] It should be noted that towards the end of distillation, the active ingredient is about to be depleted, and its distillation rate undergoes a qualitative change, manifested as an unusually small change in absorbance at the current moment, becoming a significant "valley" in the historical value sequence. This formula modifies this by combining the current absorbance change with a set of high-availability historical values. The values ​​are compared with each historical change value of high confidence. For each historical time t, the ratio is calculated. The smaller this ratio (the smaller the current change value relative to the historical change value), the better. The larger the value, the more likely the current moment is to be the endpoint. The function restricts the ratio to the interval [0, 1], ensuring a reasonable calculation result even if the current change is unexpectedly larger than the historical change. This probability is then multiplied by the normalized signal availability weight for that historical moment, indicating that opinions from more reliable historical moments have higher weights. Finally, the weighted results for all high-confidence historical moments are summed to obtain the comprehensive wavelength endpoint probability. This probability value effectively integrates the trend information of historically reliable signals, enabling sensitive capture of signals where the distillation process has essentially stopped.

[0137] Based on the above technical solution, this invention achieves independent and robust endpoint probability assessment for each wavelength by sequentially analyzing the interference level of each wavelength signal, determining its historical availability, constructing an effective absorbance change sequence based on a set of high-availability moments, and finally weighting and comparing the current signal change with historical reliable changes. This method effectively suppresses the influence of impurity interference on endpoint judgment and provides accurate and reliable basic probability input for subsequent fusion of multi-wavelength information for global endpoint decision-making.

[0138] For example, in combination Figure 4 ,like Figure 7The diagram shown is a flowchart illustrating another intelligent electric heating control method for steam distillation of traditional Chinese medicine according to an embodiment of the present invention. In this method, the comprehensive probability of the current moment being the distillation endpoint is determined based on specificity weights and wavelength endpoint probabilities, specifically including the following steps:

[0139] S701. For each wavelength, determine the weighted probability contribution value of each wavelength based on the wavelength endpoint probability and specificity weight. The weighted probability contribution value characterizes the contribution of the wavelength to the overall probability.

[0140] For example, this step can be performed by the comprehensive probability fusion submodule 132 in the endpoint evaluation module 13 described above. Specifically, the comprehensive probability fusion submodule 132 calculates the weighted probability contribution value for each wavelength λ: ;

[0141] In the above formula, This represents the weighted probability contribution value of wavelength λ, used to quantitatively characterize the magnitude of the contribution of this wavelength to the overall probability of the final distillation endpoint. This represents the specific weight of wavelength λ on the signal of the target Chinese herbal medicine's effective components; This indicates that the wavelength λ is at the current moment. The probability of the wavelength endpoint.

[0142] It should be noted that the above formula combines the endpoint probability of each wavelength with its credit rating (specificity weight). Even if a wavelength has a high endpoint probability, its contribution should be suppressed if its signal source is unreliable (low specificity weight); conversely, a wavelength with a reliable signal source (high specificity weight) should receive a greater weight for its endpoint probability. Through multiplication, the weighted probability contribution value of each wavelength is obtained, which comprehensively reflects the credibility and importance of the endpoint determination for that wavelength.

[0143] S702. Determine the overall probability that the current time is the distillation endpoint based on the weighted probability contribution value of all wavelengths and the specific weight of all wavelengths.

[0144] In this step, the integrated probability fusion submodule 132 calculates the current time using the following formula. The combined probability of the distillation endpoint : ;

[0145] In the above formula, Indicates the current moment of the wave The overall probability of the distillation endpoint is given; the closer its value is to 1, the more certain the system is that the distillation process has been completed. N represents the total number of spectral wavelengths involved in the calculation. For the summation index, represents the first... One wavelength; This represents the sum of the weighted probability contributions of all wavelengths; This represents the summation of the specific weights for all N wavelengths (in practical applications, this term will not be zero due to the complexity of the spectrum).

[0146] Therefore, the above formula assigns a weighted probability contribution value to each wavelength. Summing the values ​​and then dividing by the sum of the specific weights for all wavelengths yields the overall probability. This calculation method is mathematically equivalent to using specific weights... The weights represent the wavelength endpoint probabilities for each wavelength. A weighted average is applied. In this way, wavelengths with high specificity weights (i.e., wavelengths that strongly respond to the target component and have reliable identification) have a higher weight in the final decision, while the influence of wavelengths with low specificity weights (which may be affected by noise or impurities) is effectively suppressed. This method integrates the independent judgments of all wavelength signals to form a comprehensive and robust overall probability value, thereby achieving a scientific and reliable judgment of the distillation endpoint.

[0147] Based on the above technical solution, this embodiment of the invention constructs a comprehensive distillation endpoint probability that reflects the global signal state by weighting and fusing the specific weight of each wavelength with its independent endpoint probability. This method fully utilizes the complementary information of signals of different wavelengths and, through a credit weighting mechanism, ensures that the final decision is derived from all available data while assigning higher decisive weights to high-quality, high-reliability signals, thereby achieving accurate and adaptive endpoint determination in the complex distillation process.

[0148] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0149] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An intelligent electric heating control system for traditional Chinese medicine steam distillation, characterized in that, The system comprises a spectrum acquisition module, a specificity evaluation module, an endpoint evaluation module, and a power control module; The spectrum acquisition module is configured to acquire real-time spectrum data of distillate during distillation, wherein the real-time spectrum data comprises absorbance information at a plurality of continuous wavelengths; The specificity evaluation module is configured to determine a specificity weight of each wavelength based on the real-time spectrum data, wherein the specificity weight is used to represent specificity of the wavelength to a spectrum signal of a target traditional Chinese medicine effective component; The endpoint evaluation module is configured to determine a wavelength endpoint probability of each wavelength based on a trend of absorbance change of the wavelength, wherein the wavelength endpoint probability is used to represent a probability of completion of the distillation process indicated by a signal change of the wavelength at a current time point; The endpoint evaluation module is further configured to determine a comprehensive probability that the current time point is a distillation endpoint based on the specificity weight and the wavelength endpoint probability; The power control module is configured to dynamically adjust an electric heating power based on the comprehensive probability.

2. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 1, characterized in that, When the specificity evaluation module determines the specificity weight of each wavelength based on the real-time spectrum data, the following steps are specifically performed: The specificity evaluation module is further configured to determine a signal response intensity of each wavelength based on an average absorbance difference of each wavelength in a distillation heating period and a distillation period; The specificity evaluation module is further configured to determine a signal identity weight of each wavelength based on a similarity between the wavelength and a characteristic absorption peak wavelength of a target traditional Chinese medicine effective component; The specificity evaluation module is further configured to determine the specificity weight of each wavelength based on the signal response intensity and the signal identity weight.

3. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 2, characterized in that, When the specificity evaluation module determines the signal response intensity of each wavelength based on the average absorbance difference of each wavelength in the distillation heating period and the distillation period, the following steps are specifically performed: The specificity evaluation module is further configured to determine an ending time point of the distillation heating period as a time point at which the distillation process reaches the boiling point for the first time and stabilizes; The specificity evaluation module is further configured to determine an ending time point of the distillation period as a time point corresponding to a starting point of a continuous downward trend of absorbance of the wavelength after the ending time point of the distillation heating period; The specificity evaluation module is further configured to determine the signal response intensity of each wavelength based on an average absorbance of the wavelength in the distillation heating period and an average absorbance of the wavelength in the distillation period.

4. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 2, characterized in that, When the specificity evaluation module determines the signal identity weight of each wavelength based on the similarity between the wavelength and the characteristic absorption peak wavelength of the target traditional Chinese medicine effective component, the following steps are specifically performed: The specificity evaluation module is further configured to calculate, for each wavelength, a wavelength difference absolute value between the wavelength and a characteristic absorption peak wavelength of each target traditional Chinese medicine effective component; The specificity evaluation module is further configured to determine the signal identity weight based on a minimum value in the wavelength difference absolute value and a preset distance decay function.

5. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 1, characterized in that, The endpoint evaluation module specifically performs the following steps when determining the wavelength endpoint probability of each wavelength according to the absorbance change trend of the wavelength for each wavelength: The endpoint evaluation module is further configured to determine, for each wavelength, a signal interference degree of the each wavelength at each historical time according to a change trend of an absorbance ratio of the each wavelength to an adjacent wavelength within a time period from the distillation transition period to the current time; The endpoint evaluation module is further configured to determine a signal availability of the each wavelength at each historical time according to the signal interference degree, and construct an effective absorbance change sequence according to the signal availability of the each wavelength at each historical time; wherein the effective absorbance change sequence is an ordered set of absorbance change values at each historical time filtered based on the signal availability; The endpoint evaluation module is further configured to determine the wavelength endpoint probability of the each wavelength according to an absorbance change value of the each wavelength at the current time and the effective absorbance change sequence.

6. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 5, characterized in that, The endpoint evaluation module specifically performs the following steps when determining the signal interference degree of the each wavelength at each historical time according to a change trend of an absorbance ratio of the each wavelength to an adjacent wavelength within a time period from the distillation transition period to the current time for each wavelength: The endpoint evaluation module is further configured to calculate, for each wavelength and each historical time, a first absorbance ratio of the wavelength to a left adjacent wavelength, and a second absorbance ratio of the wavelength to a right adjacent wavelength; The endpoint evaluation module is further configured to determine the signal interference degree according to the first absorbance ratio and the second absorbance ratio.

7. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 1, characterized in that, The endpoint evaluation module specifically performs the following steps when determining the comprehensive probability that the current time is a distillation endpoint according to the specificity weight and the wavelength endpoint probability: The endpoint evaluation module is further configured to determine, for each wavelength, a weighted probability contribution value of the each wavelength according to the wavelength endpoint probability and the specificity weight; wherein the weighted probability contribution value is used to represent a contribution degree of the wavelength to the comprehensive probability; The endpoint evaluation module is further configured to determine the comprehensive probability that the current time is a distillation endpoint according to the weighted probability contribution values of all wavelengths and the specificity weights of all wavelengths. 8.The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 1, characterized in that, The power control module specifically performs the following steps when dynamically adjusting the electric heating power according to the comprehensive probability: The power control module is further configured to control the heater to operate in a first power mode when the comprehensive probability is less than a first preset threshold; The power control module is further configured to control the heater to switch to a second power mode in a smooth transition manner according to the comprehensive probability and a preset power adjustment algorithm when the comprehensive probability is greater than or equal to the first preset threshold and less than a second preset threshold; wherein the second preset threshold is greater than the first preset threshold; The power control module is further configured to control the heater to stop heating when the comprehensive probability is greater than or equal to the second preset threshold.

9. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to claim 8, characterized in that, The power control module specifically performs the following steps when controlling the heater to operate in the second power mode: The power control module is further configured to determine a real-time power adjustment factor according to the comprehensive probability and a preset fuzzy control rule base; The power control module is further configured to determine a real-time target heating power according to the power adjustment factor and a reference power corresponding to the first power mode; The power control module is further configured to generate a control signal according to the real-time target heating power to drive the heater to perform power adjustment.

10. The intelligent electric heating control system for traditional Chinese medicine steam distillation according to any one of claims 1-9, characterized in that, The spectrum acquisition module specifically performs the following steps when acquiring real-time spectrum data of the distillate: The spectrum acquisition module is further configured to measure the distillate flow path at a fixed sampling frequency through an online infrared spectrometer; The spectrum acquisition module is further configured to continuously acquire and output real-time spectrum data containing absorbance information in a continuous wavelength range during the distillation process.

Images