Multi-process roller optimization control method and system for traditional Chinese medicine decoction piece processing

Abstract

The invention provides a multi-process roller optimization control method and system for traditional Chinese medicine decoction piece processing, and belongs to the field of intelligent control. The method comprises the following steps: collecting healthy samples based on a roller processing mode, specifications and varieties of decoction pieces and a roller model; then performing correlation analysis on each evaluation index to obtain an index correlation attribute and correlation degree; the machining deviation parameters are disassembled according to the association attributes, and association deviation parameters of all the indexes are obtained; carrying out weighted calculation by taking the correlation degree as a weight to obtain a deviation characteristic value; dynamically configuring an evaluation index weight by using the deviation characteristic value, and calculating the sample fitness; and then the sample and the fitness are stored in an optimized particle space, and optimization is executed to obtain expected control parameters. By dynamically distributing index weights, enhancing disadvantage index performance and weakening advantage index performance, the problem of optimization deviation caused by existing fixed weight evaluation is solved, balanced optimization of multi-process roller control parameters is achieved, and the processing quality of traditional Chinese medicine decoction pieces is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control, and in particular to a multi-process drum optimization control method and system for processing Chinese herbal medicine slices. Background Art

[0002] In the industrial production of Chinese herbal medicine slices, the roller processing of Chinese herbal medicine slices usually involves multiple processes such as cleaning, screening and frying. Each process requires precise control of the corresponding processing parameters to ensure that the quality of the slices meets the standards. In the actual production process, the quality assessment of Chinese herbal medicine slices involves multiple indicators, such as water content, active ingredient content, appearance color, breakage rate, etc. In order to obtain high-quality slice products, it is necessary to optimize the processing parameters of the roller. However, the existing optimization control methods usually use a fixed weight method to comprehensively evaluate various evaluation indicators. In the optimization process, this fixed weight evaluation method easily makes the optimization algorithm biased towards those indicators with better performance, while paying insufficient attention to indicators with poor performance. As a result, the final control parameters may perform well in some indicators, but may have obvious shortcomings in other indicators, and it is impossible to achieve balanced optimization of various indicators. Summary of the Invention

[0003] The present invention aims to solve the technical problem that in the prior art, when optimizing the control parameters of multi-process roller processing of Chinese herbal medicine slices, fixed weight evaluation is used, which leads to the optimization results being biased towards indicators that perform advantages while ignoring indicators that perform disadvantages. The present invention provides a multi-process roller optimization control method and system for Chinese herbal medicine slice processing to solve the problem.

[0004] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a multi-process roller optimization control method for processing Chinese herbal medicine slices, comprising: based on the roller processing mode, the specifications of Chinese herbal medicine slices and the variety of Chinese herbal medicine slices, in combination with the roller model, collecting a first healthy sample, wherein the first healthy sample includes a first processing deviation parameter and a preset evaluation index detection value; traversing the preset evaluation index, performing correlation analysis on the processing control attribute, obtaining a first evaluation index association attribute and a first evaluation index correlation degree, up to the Nth evaluation index association attribute and the Nth evaluation index correlation degree; according to the first evaluation index association attribute up to the Nth evaluation index association attribute, disassembling the first processing deviation parameter, obtaining a first evaluation index association deviation parameter up to the Nth evaluation index association deviation parameter; and converting the first evaluation index association attribute into a first processing deviation parameter. The indicator correlation degree up to the Nth evaluation indicator correlation degree is used as the normalized deviation weight, and the weighted mean calculation is performed on the first evaluation indicator association deviation parameter up to the Nth evaluation indicator association deviation parameter to obtain the first evaluation indicator association attribute deviation characteristic value up to the Nth evaluation indicator association attribute deviation characteristic value; the weight of the preset evaluation indicator detection value is configured with the first evaluation indicator association attribute deviation characteristic value up to the Nth evaluation indicator association attribute deviation characteristic value, and the first fitness of the first healthy sample is calculated; the first healthy sample and the first fitness are associated and stored in the optimized particle space, and when the number of healthy samples in the optimized particle space meets the preset number, the optimization is performed to obtain the expected processing parameters for initializing the drum processing mode.

[0005] In the second aspect, the present invention provides a multi-process roller optimization control system for processing Chinese herbal medicine slices, including: a healthy sample collection module, which is used to collect a first healthy sample based on the roller processing mode, Chinese herbal medicine slice specifications and Chinese herbal medicine slice varieties, combined with the roller model, wherein the first healthy sample includes a first processing deviation parameter and a preset evaluation index detection value; a correlation analysis module, which is used to traverse the preset evaluation index, perform correlation analysis on the processing control attribute, obtain the first evaluation index association attribute and the first evaluation index correlation degree, until the Nth evaluation index association attribute and the Nth evaluation index correlation degree; a deviation parameter disassembly module, which is used to disassemble the first processing deviation parameter according to the first evaluation index association attribute to the Nth evaluation index association attribute, and obtain the first evaluation index association deviation parameter to the Nth evaluation index association deviation parameter; a deviation feature calculation module , used to use the first evaluation indicator correlation degree up to the Nth evaluation indicator correlation degree as the normalized deviation weight, perform weighted mean calculation on the first evaluation indicator correlation deviation parameter up to the Nth evaluation indicator correlation deviation parameter, and obtain the first evaluation indicator correlation attribute deviation characteristic value up to the Nth evaluation indicator correlation attribute deviation characteristic value; a fitness calculation module, used to configure the weight of the preset evaluation indicator detection value with the first evaluation indicator correlation attribute deviation characteristic value up to the Nth evaluation indicator correlation attribute deviation characteristic value, and calculate the first fitness of the first healthy sample; a processing mode initialization module, used to store the first healthy sample and the first fitness association into the optimized particle space, and when the number of healthy samples in the optimized particle space meets the preset number, perform optimization to obtain the expected processing parameters for initializing the drum processing mode.

[0006] The beneficial effects of the present invention are: Based on the roller processing mode, the specifications and varieties of Chinese herbal medicines, and in combination with the roller model, the first healthy sample is collected, wherein the first healthy sample includes the first processing deviation parameter and the preset evaluation index detection value, thereby establishing the basic data for optimization control and ensuring that the collected samples can reflect the processing characteristics under specific working conditions. Traverse the preset evaluation index, perform correlation analysis on the processing control attributes, obtain the first evaluation index correlation attribute and the first evaluation index correlation degree, until the Nth evaluation index correlation attribute and the Nth evaluation index correlation degree, thereby identifying which control attributes each evaluation index is related to and the degree of correlation through correlation analysis, providing a basis for subsequent parameter decomposition. According to the first evaluation index correlation attribute to the Nth evaluation index correlation attribute, decompose the first processing deviation parameter, obtain the first evaluation index correlation deviation parameter to the Nth evaluation index correlation deviation parameter, thereby decomposing the overall processing deviation parameter into sub-parameters related to each evaluation index, and realizing refined parameter management.

[0007] The correlations of the first through the Nth evaluation indicators are used as normalized deviation weights, and a weighted mean calculation is performed on the deviation parameters associated with the first through the Nth evaluation indicators to obtain the deviation eigenvalues ​​of the first through the Nth evaluation indicators. Through this weighted calculation, deviation eigenvalues ​​that reflect the characteristics of each evaluation indicator are obtained. The first fitness of the first healthy sample is calculated using the deviation eigenvalues ​​of the first through the Nth evaluation indicators, along with the weights assigned to the preset evaluation indicator detection values. This implements dynamic weight allocation, giving higher weights to indicators with weaker performance and lower weights to indicators with stronger performance, thereby achieving balanced optimization. The first healthy sample and the first fitness are associated and stored in an optimized particle space. When the number of healthy samples in the optimized particle space meets a preset number, an optimization search is performed to obtain the desired processing parameters for initializing the roller processing mode. The optimal solution is then found within the constructed optimized particle space, resulting in control parameters that achieve relative balance among the evaluation indicators.

[0008] Through the above technical solution, dynamic balanced optimization of the control parameters of the multi-process roller processing of Chinese herbal medicine slices is achieved, so that the weights of the performance disadvantage indicators can be automatically enhanced and the weights of the performance advantage indicators can be weakened during the optimization process, thereby avoiding the optimization bias problem caused by the traditional fixed weight method, and finding the optimal expected processing parameters with relatively balanced evaluation indicators in the constructed optimization particle space, thereby improving the overall processing quality of Chinese herbal medicine slices. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic flow chart of a multi-process drum optimization control method for processing Chinese herbal medicine slices provided by the present invention; Figure 2 This is a structural schematic diagram of a multi-process drum optimization control system for processing Chinese herbal medicine slices provided by the present invention.

[0010] In the accompanying drawings, the components represented by the reference numerals are as follows: Healthy sample collection module 11, correlation analysis module 12, deviation parameter decomposition module 13, deviation feature calculation module 14, fitness calculation module 15, processing mode initialization module 16. DETAILED DESCRIPTION

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0012] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0013] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0014] Example 1, as Figure 1 As shown, an embodiment of the present invention provides a multi-process drum optimization control method for processing Chinese herbal medicine pieces, which is applied to a Chinese herbal medicine piece processing drum, wherein the Chinese herbal medicine piece processing drum includes a cleaning processing mode, a screening processing mode, and a frying processing mode.

[0015] Specifically, the method provided in the embodiment of the present application is applied to a Chinese herbal medicine processing drum, which is an integrated multifunctional processing equipment that can realize multiple processes in the processing of Chinese herbal medicine. Specifically, the Chinese herbal medicine processing drum has three main processing modes, namely, a cleaning processing mode, a screening processing mode, and a frying processing mode. Among them, the cleaning processing mode is used to perform pre-treatment operations such as cleaning, removing impurities, and dust removal on the raw materials of Chinese herbal medicine slices to remove mud, sand, impurities and other pollutants on the surface of the medicinal materials, and ensure that the cleanliness of the Chinese herbal medicine slices meets the pharmaceutical standards; the screening processing mode is used to grade and screen the cleaned Chinese herbal medicine slices, and classify them according to the size, shape, quality and other characteristics of the slices to meet the production needs of Chinese herbal medicine slices with different specifications; the frying processing mode is used to fry the screened Chinese herbal medicine slices, and by controlling parameters such as temperature, time, and speed, different processing degrees such as frying yellow, frying burnt, and charred frying of the Chinese herbal medicine slices can be achieved to change the medicinal properties of the medicinal materials or enhance their efficacy.

[0016] By integrating the above three processing modes on the Chinese herbal medicine processing drum, continuous and automated production of Chinese herbal medicine processing can be achieved, production efficiency can be improved, and the stability and consistency of product quality can be ensured.

[0017] The multi-process drum optimization control method includes: S1. Based on the roller processing mode, the specifications and varieties of Chinese herbal medicines, and the roller model, a first healthy sample is collected, wherein the first healthy sample includes a first processing deviation parameter and a preset evaluation index detection value.

[0018] Specifically, the drum processing mode refers to the current working state of the Chinese herbal medicine processing drum, which can be one of the cleaning processing mode, screening processing mode or frying processing mode; the specifications of Chinese herbal medicines correspond to the size parameter requirements of the medicines; the varieties of Chinese herbal medicines correspond to different types of Chinese medicinal materials; the drum model corresponds to the equipment model of the specific Chinese herbal medicine processing drum used.

[0019] According to the roller processing mode, Chinese herbal medicine slice specifications, Chinese herbal medicine slice varieties and roller models, historical processing samples are retrieved in the historical processing database of Chinese herbal medicine slices to obtain the first healthy sample that meets the conditions. Among them, meeting the conditions means that the roller processing mode, Chinese herbal medicine slice specifications, Chinese herbal medicine slice varieties and roller models of the historical sample match the current process conditions to be optimized, and the processing results corresponding to the historical sample meet the quality standard requirements, that is, the healthy sample. Among them, the first processing deviation parameter refers to the difference between the actual control parameters and the standard control parameters during the processing process, including but not limited to temperature deviation, speed deviation, time deviation, pressure deviation, etc.; the preset evaluation index detection value refers to the actual measurement value of each indicator used to evaluate the processing quality of Chinese herbal medicine slices, including but not limited to moisture content, breakage rate, color uniformity, effective ingredient content, appearance morphology score, etc.

[0020] Through the sample collection method based on multidimensional conditions, it is ensured that the first healthy sample can accurately reflect the processing status and quality performance under specific process conditions, laying a data foundation for subsequent optimization control analysis.

[0021] S2. Traverse the preset evaluation indicators, perform correlation analysis on the processing control attributes, obtain the first evaluation indicator correlation attribute and the first evaluation indicator correlation degree, and so on until the Nth evaluation indicator correlation attribute and the Nth evaluation indicator correlation degree.

[0022] Specifically, the preset evaluation indicators refer to the various indicators used to evaluate the processing quality of Chinese herbal medicines, such as moisture content, breakage rate, color uniformity, active ingredient content, appearance morphology score, etc.; processing control attributes refer to the various parameter attributes that can be adjusted and controlled during the processing of Chinese herbal medicines, including temperature, speed, time, pressure, humidity, etc.

[0023] By traversing the set evaluation indicators, each one is analyzed in turn. Each time, one set evaluation indicator is extracted as the analysis object, which is recorded as the first evaluation indicator. For the first evaluation indicator, the correlation relationship between it and each processing control attribute is analyzed to determine which control attributes have a significant impact on the set evaluation indicator. These control attributes with a significant impact are the first evaluation indicator association attributes. At the same time, the degree of correlation between these first evaluation indicator association attributes and the first evaluation indicator is calculated to obtain the first evaluation indicator association degree. Following the same method, the correlation analysis is continued for the second, third, and Nth set evaluation indicators, respectively, to obtain the second evaluation indicator association attribute and the second evaluation indicator association degree, the third evaluation indicator association attribute and the third evaluation indicator association degree, until the Nth evaluation indicator association attribute and the Nth evaluation indicator association degree are obtained. Where N represents the total number of set evaluation indicators.

[0024] Through traversal analysis, it is possible to fully identify which processing control attributes are associated with each preset evaluation indicator, as well as the strength of the association, providing an accurate attribute mapping relationship for subsequent parameter optimization.

[0025] S3. Decompose the first processing deviation parameter according to the first evaluation indicator association attribute to the Nth evaluation indicator association attribute to obtain first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters.

[0026] Specifically, the first processing deviation parameter is a collection of control parameter deviations contained in the first healthy sample, including deviation values ​​for all processing control attributes, such as temperature deviation, speed deviation, time deviation, and pressure deviation. The first through Nth evaluation indicator-associated attributes represent sets of control attributes associated with each preset evaluation indicator.

[0027] Based on the obtained first evaluation indicator association attributes, through the Nth evaluation indicator association attributes, deviation parameters associated with each preset evaluation indicator are extracted from the first processing deviation parameters. For example, if the associated attributes of the first evaluation indicator (such as moisture content) include temperature and time, the temperature deviation and time deviation are extracted from the first processing deviation parameters to form the first evaluation indicator association deviation parameters. If the associated attributes of the second evaluation indicator (such as crushing rate) include speed and pressure, the speed deviation and pressure deviation are extracted from the first processing deviation parameters to form the second evaluation indicator association deviation parameters. Following this decomposition method, all preset evaluation indicators are processed sequentially, ultimately obtaining the first evaluation indicator association deviation parameters, the second evaluation indicator association deviation parameters, and so on. Each evaluation indicator association deviation parameter only contains the control attribute deviation values ​​that affect the preset evaluation indicator, achieving precise classification and targeted extraction of deviation parameters.

[0028] Through the above decomposition method, the original comprehensive deviation parameters are grouped according to their correlation with each preset evaluation indicator, providing a structured data basis for subsequent targeted deviation analysis and weight calculation.

[0029] S4. Use the first evaluation indicator correlation degree to the Nth evaluation indicator correlation degree as normalized deviation weights, perform weighted mean calculation on the first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters, and obtain the first evaluation indicator association attribute deviation characteristic value to the Nth evaluation indicator association attribute deviation characteristic value.

[0030] Specifically, the first evaluation indicator correlation degree to the Nth evaluation indicator correlation degree represents the correlation strength between each processing control attribute and the corresponding preset evaluation indicator; the first evaluation indicator correlation deviation parameter to the Nth evaluation indicator correlation deviation parameter include the control attribute deviation value related to each preset evaluation indicator.

[0031] First, the correlation of the first evaluation indicator to the Nth evaluation indicator is normalized. Specifically, taking the first evaluation indicator as an example, its correlation includes the correlation values ​​of multiple correlation attributes. These correlation values ​​are added together to obtain the sum, and then each correlation value is divided by the sum to obtain the normalized correlation as the normalized deviation weight. For example, if the first evaluation indicator has three correlation attributes, and the correlations are 0.6, 0.4, and 0.3, respectively, the sum of the correlations is 1.3, and the normalized weights are 0.6 / 1.3≈0.46, 0.4 / 1.3≈0.31, and 0.3 / 1.3≈0.23, respectively. Normalization ensures that the sum of the weight coefficients is 1, so that the control attributes with high correlation have a greater influence weight in subsequent calculations.

[0032] Then, the normalized deviation weight is used to calculate the weighted mean of the corresponding associated deviation parameters. Specifically, for the first evaluation index, each of its associated deviation parameters is multiplied by the corresponding normalized deviation weight, and then all the products are added together to obtain the first evaluation index associated attribute deviation characteristic value. This characteristic value comprehensively reflects the weighted influence of all relevant control attribute deviations on the evaluation index. According to the same processing method, the correlation degrees of the second evaluation index to the Nth evaluation index are normalized in turn, and the corresponding associated attribute deviation characteristic values ​​are calculated, thereby obtaining the first evaluation index associated attribute deviation characteristic value to the Nth evaluation index associated attribute deviation characteristic value.

[0033] Through normalization and weighted mean calculation, the deviation values ​​of multiple associated attributes are weighted and integrated according to their correlation strength to obtain characteristic values ​​that can accurately reflect the degree to which each evaluation indicator is affected by the deviation of the control attribute, providing a standardized data basis for subsequent fitness calculations.

[0034] S5. Using the first evaluation indicator-associated attribute deviation characteristic value up to the Nth evaluation indicator-associated attribute deviation characteristic value, configure the weight of the preset evaluation indicator detection value and calculate the first fitness of the first healthy sample.

[0035] Specifically, the first evaluation indicator associated attribute deviation characteristic value to the Nth evaluation indicator associated attribute deviation characteristic value reflects the comprehensive influence degree of each evaluation indicator by the control attribute deviation; the preset evaluation indicator detection value is the actual measurement value of each quality indicator contained in the first healthy sample.

[0036] First, weights are assigned based on the associated attribute deviation characteristic values ​​of each evaluation indicator. Specifically, the weights for the preset evaluation indicator detection values ​​are assigned based on the associated attribute deviation characteristic values ​​of the first evaluation indicator through the Nth evaluation indicator. The principle of weight assignment is as follows: evaluation indicators with larger associated attribute deviation characteristic values ​​are more affected by controlled attribute deviations and are at a relative disadvantage in quality control. Therefore, higher weights are assigned to the corresponding preset evaluation indicator detection values. Evaluation indicators with smaller associated attribute deviation characteristic values ​​are less affected by controlled attribute deviations and are at a relative advantage. Therefore, lower weights are assigned to the corresponding preset evaluation indicator detection values. This weight assignment approach prioritizes disadvantaged indicators to achieve dynamic balanced optimization.

[0037] The configured weights are then used to calculate the fitness of the pre-set evaluation indicator values. Specifically, each pre-set evaluation indicator value is compared with its expected value (or standard value), the degree of deviation is calculated, and then multiplied by the corresponding weight. Finally, all weighted deviations are combined to obtain the first fitness of the first healthy sample. This first fitness reflects the overall quality performance of the first healthy sample after considering the dynamic weights of each indicator.

[0038] Through dynamic weight configuration based on deviation eigenvalues, the importance of each evaluation indicator in the fitness calculation is automatically adjusted, so that indicators with poor performance receive more attention and the weight of indicators with good performance is relatively reduced, thereby promoting the balanced optimization of various quality indicators and avoiding the problem that some indicators may be over-optimized while other indicators are neglected due to traditional fixed weight methods.

[0039] S6. Associate the first healthy sample with the first fitness and store it in an optimized particle space. When the number of healthy samples in the optimized particle space meets a preset number, perform optimization to obtain desired processing parameters to initialize the drum processing mode.

[0040] Specifically, first, the first healthy sample and the first fitness are stored in association, that is, the two are stored as a whole data pair in the optimized particle space. This associative storage ensures that each healthy sample has a corresponding fitness value, which is convenient for subsequent optimization calculations. As data collection progresses, more healthy samples and their fitness will be stored in the optimized particle space. Then, the number of healthy samples in the optimized particle space is continuously monitored. When the number of healthy samples reaches a preset number, it indicates that a sufficient data base has been accumulated, and the optimization process is triggered. Among them, the setting of the preset number needs to balance the optimization accuracy and computational efficiency to ensure that there are enough samples to support the optimization calculation, and that the computational burden is not too heavy due to too many samples.

[0041] During the optimization process, all healthy samples in the optimization particle space and their fitness are used to search for the optimal solution using an optimization algorithm (such as the particle swarm algorithm) to obtain the desired processing parameters. These parameters are defined as the set of control parameter values ​​that optimize fitness, including temperature, speed, time, and pressure. The obtained desired processing parameters are then used to initialize the roller processing mode. These parameters are set as the initial control parameters for the corresponding processing mode, providing an optimized process parameter benchmark for actual production.

[0042] Through a method based on sample accumulation and optimization, the transformation from historical data to optimized parameters is realized, and a data-driven parameter optimization solution is provided for the roller processing mode, so that the control parameters can be adaptively adjusted according to different varieties, specifications and processing modes of Chinese herbal medicines, thereby improving the stability and consistency of Chinese herbal medicine processing. At the same time, the problem of over-optimization of a single indicator is avoided through the dynamic weight mechanism, and a balanced improvement of various quality indicators is achieved, thereby improving the overall processing quality and production efficiency of Chinese herbal medicines.

[0043] Furthermore, traversing the preset evaluation indicators, performing correlation analysis on the processing control attributes, obtaining a first evaluation indicator correlation attribute and a first evaluation indicator correlation degree, and so on, until the Nth evaluation indicator correlation attribute and the Nth evaluation indicator correlation degree, including: S21, extracting a first evaluation indicator of the preset evaluation indicators and extracting a first processing control attribute of the processing control attributes; S22. Based on the drum processing mode, the specifications of the Chinese herbal medicine slices, and the varieties of the Chinese herbal medicine slices, and in combination with the drum model, a processing abnormality sample set is collected, wherein any processing abnormality sample in the processing abnormality sample set includes an abnormal processing control attribute set and an abnormality evaluation index set; S23, calculating the ratio of the triggering frequency of the first evaluation indicator and the first processing control attribute in the same processing abnormality sample to the total number of processing abnormality samples; S24. When the ratio is greater than or equal to a preset ratio, the first processing control attribute is added to the first evaluation indicator associated attribute. After the processing control attribute traversal is completed, a correlation analysis is performed on the first evaluation indicator associated attribute based on the first evaluation indicator to obtain the first evaluation indicator correlation degree. S25 , updating the evaluation indicators and performing cyclic analysis until the Nth evaluation indicator association attribute and the Nth evaluation indicator association degree are obtained.

[0044] In one feasible implementation, one evaluation indicator, such as moisture content, is extracted from the pre-set evaluation indicators as the current analysis target, recorded as the first evaluation indicator. Simultaneously, one control attribute, such as temperature, is extracted from the processing control attributes as the first processing control attribute. Then, under four-dimensional conditions (drum processing mode, TCM decoction piece specifications, TCM decoction piece varieties, and drum model), sample data of historical quality anomalies is collected to generate a processing anomaly sample set. Each processing anomaly sample in this processing anomaly sample set records the abnormal processing control attribute set (i.e., which control parameters were abnormal) and the abnormal evaluation indicator set (i.e., which quality indicators did not meet the standards) at the time of the anomaly.

[0045] Subsequently, all processing anomaly samples in the processing anomaly sample set are traversed, and the number of times the first evaluation indicator and the first processing control attribute are simultaneously abnormal is counted. This means that within the same processing anomaly sample, the evaluation indicator fails to meet the standard and the control attribute is abnormal. This number of simultaneous triggering events is divided by the total number of processing anomaly samples in the processing anomaly sample set to obtain a trigger frequency ratio. If this ratio reaches a preset value (e.g., 0.6), it indicates a strong correlation between the first processing control attribute and the first evaluation indicator, and the first processing control attribute is added to the first evaluation indicator associated attributes. The traversal of other control attributes continues, repeating the judgment process from S21 to S24. Once all processing control attributes have been traversed, all control attributes associated with the first evaluation indicator are obtained. These associated attributes are then subjected to in-depth correlation analysis to calculate the first evaluation indicator correlation degree.

[0046] After completing the analysis of the first evaluation indicator, the second evaluation indicator is updated and the process of S21-S24 is repeated. This process is repeated until the identification of associated attributes and the calculation of association degrees for all N preset evaluation indicators are completed, and a complete result is obtained from the associated attributes and association degrees of the first evaluation indicator to the associated attributes and association degrees of the Nth evaluation indicator.

[0047] Through the correlation analysis method based on abnormal sample statistics, the intrinsic relationship between each evaluation indicator and the control attribute can be accurately identified, providing accurate dependency mapping for subsequent parameter optimization.

[0048] Furthermore, based on the first evaluation indicator, performing a correlation analysis on the first evaluation indicator correlation attribute to obtain the first evaluation indicator correlation degree further includes: S241, based on the drum processing mode, the specifications of the Chinese herbal medicine pieces and the variety of the Chinese herbal medicine pieces, combined with the drum model, collecting a plurality of first evaluation index associated attribute record values ​​and a plurality of first evaluation index record values; S242: Calculate the same attribute deviation of the plurality of first evaluation index associated attribute record values ​​to obtain a plurality of first evaluation index associated attribute deviation moduli; S243, performing deviation calculation on a plurality of first evaluation index recorded values ​​to obtain a plurality of first evaluation index recorded value deviation moduli; S244. Extract the first associated attribute deviation modulus value sequence up to the Qth associated attribute deviation modulus value sequence from the several first evaluation indicator associated attribute deviation modulus values, construct a grey correlation matrix in combination with the several first evaluation indicator record value deviation modulus values, perform correlation analysis, obtain the first associated attribute correlation degree up to the Qth associated attribute correlation degree, and add it into the first evaluation indicator correlation degree.

[0049] In a preferred embodiment, multiple sets of data are first collected from a historical processing database under the same four-dimensional conditions as described above (drum processing mode, TCM decoction piece specifications, TCM decoction piece varieties, and drum model). These data include several first evaluation indicator-associated attribute record values ​​and several first evaluation indicator record values. The first evaluation indicator-associated attribute record values ​​refer to the actual measured values ​​of each processing control attribute associated with the first evaluation indicator at different times; the first evaluation indicator record values ​​refer to the actual measured values ​​of the first evaluation indicator at corresponding times. These data constitute the original dataset for correlation analysis.

[0050] Then, for the first evaluation index, deviation calculations are performed on the recorded values ​​of each processing-related attribute. Specifically, for a processing-related attribute (e.g., temperature), the deviations of each recorded value from the standard value or mean of the processing-related attribute are calculated, and the modulus values ​​are taken to obtain a sequence of deviation modulus values ​​for the processing-related attribute. This process is repeated for all associated attributes to obtain a number of deviation modulus values ​​of the attributes associated with the first evaluation index. Similarly, deviation calculations are performed on the sequence of recorded values ​​of the first evaluation index, calculating the deviations of each recorded value from the standard value or mean of the index, and the modulus values ​​are taken to obtain a number of deviation modulus values ​​of the recorded values ​​of the first evaluation index.

[0051] Next, the deviation moduli of the associated attributes of the first evaluation indicators are grouped by attribute to obtain a sequence of deviation moduli of the first to the Qth associated attributes, where Q is the total number of associated attributes in the associated attributes of the first evaluation indicators. Subsequently, a grey correlation matrix is ​​constructed using the deviation moduli of the first evaluation indicator record values ​​as a reference sequence and the deviation moduli of the first to the Qth associated attributes as a comparison sequence. The rows of the matrix represent different moments, and the columns represent the reference sequence and each comparison sequence. Then, based on this grey correlation matrix, a correlation analysis is performed to obtain the correlations of the first to the Qth associated attributes, which are then added to the first evaluation indicator correlation. Specifically, the correlation is calculated for each correlation attribute deviation moduli and the deviation moduli of the first evaluation indicator record values: first, the absolute difference between the two sequences at each moment is calculated to find the minimum and maximum difference; then, the correlation coefficient is calculated for each moment; and finally, the correlation coefficients at all moments are averaged to obtain the correlations of the associated attributes, thereby obtaining the correlations of the first to the Qth associated attributes.

[0052] Through the grey correlation analysis method based on deviation sequence, the dynamic correlation relationship between the correlation attribute of the first evaluation index and the first evaluation index can be deeply explored, providing more accurate correlation quantification results.

[0053] Furthermore, taking the first evaluation indicator association degree to the Nth evaluation indicator association degree as the normalized deviation weight, performing weighted mean calculation on the first evaluation indicator association deviation parameter to the Nth evaluation indicator association deviation parameter, and obtaining the first evaluation indicator association attribute deviation characteristic value to the Nth evaluation indicator association attribute deviation characteristic value, including: S41, extracting a plurality of associated attribute deviation parameters from the first evaluation indicator associated deviation parameter; S42, traversing the plurality of associated attribute deviation parameters and performing normalization processing to obtain a plurality of associated attribute normalized deviations; S43, extracting a plurality of correlation attribute correlations from the first evaluation indicator correlation; S44. Perform weighted mean calculation on the normalized deviations of the plurality of associated attributes according to the association degrees of the plurality of associated attributes to obtain the first evaluation indicator associated attribute deviation characteristic value.

[0054] In a preferred embodiment, the first evaluation indicator associated deviation parameter first includes the deviation values ​​of all control attributes associated with the first evaluation indicator. Several associated attribute deviation parameters are then extracted from this, specifically the deviation parameter values ​​corresponding to each specific associated attribute (such as temperature, speed, time, etc.). These extracted associated attribute deviation parameters are then normalized one by one. Normalization converts deviation parameters of different dimensions and numerical ranges into a unified interval of [0, 1] to ensure comparability in subsequent calculations. After normalization, several normalized deviations of associated attributes are obtained.

[0055] The first evaluation indicator correlation degree includes the correlation strength value between each associated attribute and the first evaluation indicator. A number of associated attribute correlation degrees are extracted therefrom, and these correlation degrees correspond one to one with the associated attribute deviation parameters extracted in step S41. Subsequently, the several associated attribute correlation degrees are used as weight coefficients to calculate the weighted mean of the corresponding several associated attribute normalized deviations. Specifically, each associated attribute normalized deviation is multiplied by its corresponding associated attribute correlation degree, and then all the products are added and divided by the sum of the correlation degrees to obtain the first evaluation indicator associated attribute deviation characteristic value. This characteristic value comprehensively reflects the weighted influence of all associated attribute deviations on the first evaluation indicator.

[0056] Through the above processing steps S41-S44, the deviation feature extraction of the first evaluation indicator is achieved. The second evaluation indicator to the Nth evaluation indicator is processed in the same way to obtain the deviation feature value of the attribute associated with the first evaluation indicator through the Nth evaluation indicator, providing feature data for the subsequent dynamic weight configuration.

[0057] Furthermore, configuring the weight of the preset evaluation indicator detection value based on the first evaluation indicator associated attribute deviation characteristic value to the Nth evaluation indicator associated attribute deviation characteristic value includes: S51, summing the first evaluation index associated attribute deviation characteristic value up to the Nth evaluation index associated attribute deviation characteristic value to obtain a sum of the deviation characteristic values; S52 , traverse the first evaluation indicator associated attribute deviation characteristic value until the Nth evaluation indicator associated attribute deviation characteristic value, compare with the sum of the deviation characteristic values, and obtain the first evaluation indicator weight until the Nth evaluation indicator weight.

[0058] In a preferred embodiment, the first evaluation indicator-associated attribute deviation characteristic value through the Nth evaluation indicator-associated attribute deviation characteristic value each reflect the degree to which each evaluation indicator is comprehensively affected by the control attribute deviation. These N deviation characteristic values ​​are summed (i.e., the first evaluation indicator-associated attribute deviation characteristic value + the second evaluation indicator-associated attribute deviation characteristic value + ... + the Nth evaluation indicator-associated attribute deviation characteristic value) to obtain a sum of the deviation characteristic values. This sum represents the total of all evaluation indicator deviation characteristic values.

[0059] The associated attribute deviation characteristic values ​​of each evaluation indicator are traversed. Specifically, the first evaluation indicator's associated attribute deviation characteristic value is divided by the sum of the deviation characteristic values ​​to obtain the first evaluation indicator weight; the second evaluation indicator's associated attribute deviation characteristic value is divided by the sum of the deviation characteristic values ​​to obtain the second evaluation indicator weight; and so on, until the Nth evaluation indicator's associated attribute deviation characteristic value is divided by the sum of the deviation characteristic values ​​to obtain the Nth evaluation indicator weight.

[0060] Through ratio calculations, weights are normalized, ensuring that the sum of all evaluation indicator weights is 1. Indicators with larger deviation eigenvalues ​​receive higher weights, indicating that they are significantly affected by deviation under the current control conditions and require more attention in fitness calculations. Conversely, indicators with smaller deviation eigenvalues ​​receive lower weights, indicating relative stability. This dynamic weighting method, based on deviation eigenvalues, automatically prioritizes weaker indicators while moderately devaluing stronger ones, contributing to the balanced optimization of various quality indicators.

[0061] Furthermore, calculating the first fitness of the first healthy sample includes: S53, obtaining a preset evaluation indicator expected value, and calculating a deviation modulus between the preset evaluation indicator detection value and the preset evaluation indicator deviation modulus value; S54: Based on the first evaluation indicator weight up to the Nth evaluation indicator weight, perform weighted summation on the preset evaluation indicator deviation modulus values ​​to obtain the first fitness.

[0062] In a preferred embodiment, the preset evaluation indicator expected values ​​are the ideal target values ​​or standard values ​​for each evaluation indicator, representing the quality objectives for TCM decoction piece processing; the preset evaluation indicator test values ​​are the actual measured values ​​of each quality indicator included in the first healthy sample. For each preset evaluation indicator, the difference between the test value and the expected value is calculated, and the absolute value (i.e., the deviation modulus) is taken to obtain the preset evaluation indicator deviation modulus value. This preset evaluation indicator deviation modulus value reflects the gap between the actual processing quality and the target quality.

[0063] Then, using the obtained first to Nth evaluation indicator weights, a weighted sum is performed on the corresponding preset evaluation indicator deviation moduli. Specifically, the first preset evaluation indicator deviation moduli is multiplied by the first evaluation indicator weight, the second preset evaluation indicator deviation moduli is multiplied by the second evaluation indicator weight, and so on, until the Nth preset evaluation indicator deviation moduli is multiplied by the Nth evaluation indicator weight, and then all weighted deviation moduli are added together to obtain the first fitness.

[0064] The smaller the first fitness value, the closer the overall quality of the first healthy sample is to the desired target; conversely, the larger the first fitness value, the further it deviates from the target. By introducing dynamic weights, indicators with larger deviations are given a higher weight in the fitness calculation, prompting the optimization process to focus more on weak links and achieve a balanced improvement in various quality indicators.

[0065] Furthermore, the first healthy sample and the first fitness are associated and stored in an optimized particle space. When the number of healthy samples in the optimized particle space meets a preset number, an optimization search is performed to obtain desired processing parameters, including: S61, performing particle swarm optimization based on the healthy sample set and the fitness set to obtain an expanded sample; S62, retrieving a sample set of the same modality as the expanded sample; S63: When the proportion of abnormal samples in the homomodal sample set is greater than or equal to the abnormal number proportion threshold, deleting the expanded sample; S64. When the proportion of abnormal samples in the homomodal sample set is less than the abnormal number proportion threshold, setting the mean value of the processing deviation parameter of the homomodal sample set as the expanded sample processing deviation parameter, and setting the mean value of the preset evaluation index detection value of the homomodal sample set as the expanded sample preset evaluation index detection value; S65 , performing cyclic optimization based on the expanded sample processing deviation parameter and the expanded sample preset evaluation index detection value.

[0066] In a preferred embodiment, the healthy sample set includes all stored healthy samples in the optimization particle space. Each healthy sample includes processing deviation parameters and a preset evaluation index test value. The fitness set contains the fitness value corresponding to each healthy sample. In the particle swarm optimization algorithm, each healthy sample is treated as a particle, its position defined by the processing deviation parameters, and the fitness value serves as the evaluation criterion for the quality of that position. During algorithm execution, each particle updates its velocity and position based on its own historical optimal position and the optimal position of the swarm. Specifically, particles learn from individual and swarm experience to move through the parameter space, exploring new processing parameter combinations. After a certain number of iterations, the algorithm generates new positions, known as the expanded sample. The expanded sample contains new processing deviation parameter combinations and represents potential optimization solutions discovered during the algorithm's optimization process. The expanded sample is then further analyzed, and historical samples with the same control state as the expanded sample are retrieved from the historical processing database. Specific search criteria include the same drum processing mode (cleaning, screening, or frying), the same TCM decoction piece specifications, the same TCM decoction piece variety, and the same drum model. All historical samples that meet these conditions are grouped into a homomodal sample set. These samples share the same process background as the expanded sample, and their historical performance can provide a reference for reliability assessment of the expanded sample.

[0067] Subsequently, a quality analysis is performed on the homomodal sample set, and the number of abnormal samples is counted. Abnormal samples are historical samples that failed to meet processing quality standards or exhibited quality issues. The ratio of the number of abnormal samples to the total number of homomodal samples is calculated to determine the abnormal sample percentage. If the abnormal sample percentage reaches or exceeds a preset abnormality percentage threshold (e.g., 0.7), this indicates a high historical incidence of quality issues under similar control conditions, indicating that the parameter combination corresponding to the expanded sample presents a high risk. Therefore, the expanded sample is removed from the candidate solution to avoid outputting high-risk parameters as optimization results.

[0068] If the proportion of abnormal samples is lower than the threshold, this indicates that the control condition has been relatively stable and reliable in historical production. At this point, the expanded sample is empirically corrected using statistical information from the homomodal sample set. The specific correction method involves calculating the arithmetic mean of the processing deviation parameters (such as temperature deviation and speed deviation) for all samples in the homomodal sample set and using this mean as the new processing deviation parameter for the expanded sample. Simultaneously, the arithmetic mean of the preset evaluation index test values ​​(such as moisture content and crushing rate) for all samples is calculated and used as the preset evaluation index test value for the expanded sample. This correction method, based on historical data means, makes the expanded sample more consistent with actual production experience, improving its reliability and practicality. Subsequently, the dynamic weighted fitness calculation process is re-executed using the corrected expanded sample parameters from step S64—namely, the expanded sample processing deviation parameters and the expanded sample preset evaluation index test values—to obtain the fitness value for the expanded sample. The calculated expanded sample and its fitness are added to the optimized particle space, and the particle swarm state is updated. Continue to execute the next iteration of the particle swarm optimization algorithm and repeat the process from S61 to S65 until the algorithm converges or reaches the maximum number of iterations. Finally, the sample with the best fitness is selected from the optimized particle space, and its processing parameters are the desired processing parameters.

[0069] By combining historical sample verification, experience correction and cyclic optimization methods, not only the exploratory and innovative nature of the optimization process is guaranteed, but also the reliability and practicality of the optimization results are ensured, effectively improving the accuracy of the optimization of processing parameters of Chinese herbal medicines.

[0070] Example 2, as Figure 2 As shown, based on the same inventive concept as the multi-process drum optimization control method for processing Chinese herbal medicine pieces provided in Example 1, the embodiment of the present invention also provides a multi-process drum optimization control system for processing Chinese herbal medicine pieces, comprising: The health sample collection module 11 is used to collect a first health sample based on the roller processing mode, the specifications and types of Chinese herbal medicine pieces, and the roller model, wherein the first health sample includes a first processing deviation parameter and a preset evaluation index detection value; A correlation analysis module 12 is configured to traverse preset evaluation indicators, perform correlation analysis on the processing control attributes, and obtain a first evaluation indicator correlation attribute and a first evaluation indicator correlation degree, up to an Nth evaluation indicator correlation attribute and an Nth evaluation indicator correlation degree; a deviation parameter decomposition module 13, configured to decompose the first processing deviation parameter according to the first evaluation indicator association attribute to the Nth evaluation indicator association attribute, and obtain first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters; a deviation characteristic calculation module 14, configured to use the first evaluation indicator association degree to the Nth evaluation indicator association degree as normalized deviation weights, perform weighted mean calculation on the first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters, and obtain first evaluation indicator association attribute deviation characteristic values ​​to the Nth evaluation indicator association attribute deviation characteristic values; A fitness calculation module 15 is configured to calculate a first fitness of the first healthy sample by configuring a weight of the preset evaluation indicator detection value based on the first evaluation indicator associated attribute deviation characteristic value up to the Nth evaluation indicator associated attribute deviation characteristic value; The processing mode initialization module 16 is used to associate and store the first healthy sample and the first fitness into the optimized particle space. When the number of healthy samples in the optimized particle space meets the preset number, the optimization is performed to obtain the desired processing parameters to initialize the drum processing mode.

[0071] Furthermore, the correlation analysis module 12 includes the following execution steps: Extracting a first evaluation indicator of the preset evaluation indicators and extracting a first processing control attribute of the processing control attributes; Based on the drum processing mode, the specifications of the Chinese herbal medicine slices, and the varieties of the Chinese herbal medicine slices, combined with the drum model, a processing abnormality sample set is collected, wherein any processing abnormality sample in the processing abnormality sample set includes an abnormal processing control attribute set and an abnormality evaluation index set; Counting the ratio of the triggering frequency of the first evaluation indicator and the first processing control attribute in the same processing abnormality sample to the total number of processing abnormality samples; When the ratio is greater than or equal to a preset ratio, the first processing control attribute is added to the first evaluation index associated attribute; after the processing control attribute traversal is completed, a correlation analysis is performed on the first evaluation index associated attribute based on the first evaluation index to obtain the first evaluation index correlation degree; The evaluation indicators are updated and analyzed cyclically until the Nth evaluation indicator association attribute and the Nth evaluation indicator association degree are obtained.

[0072] Furthermore, the correlation analysis module 12 further includes the following execution steps: Based on the roller processing mode, the specifications of the Chinese herbal medicine slices and the varieties of the Chinese herbal medicine slices, and in combination with the roller model, a plurality of first evaluation index associated attribute record values ​​and a plurality of first evaluation index record values ​​are collected; Performing same-attribute deviation calculation on the plurality of first evaluation indicator associated attribute record values ​​to obtain a plurality of first evaluation indicator associated attribute deviation modulus values; Performing deviation calculation on a plurality of first evaluation index recorded values ​​to obtain a plurality of first evaluation index recorded value deviation moduli; From the deviation moduli of the associated attributes of the several first evaluation indicators, extract the first associated attribute deviation modulus sequence up to the Qth associated attribute deviation modulus sequence, combine the deviation moduli of the several first evaluation indicator record values, construct a grey correlation matrix, perform correlation analysis, obtain the first associated attribute correlation degree up to the Qth associated attribute correlation degree, and add it into the first evaluation indicator correlation degree.

[0073] Furthermore, the deviation feature calculation module 14 includes the following execution steps: Extracting a plurality of associated attribute deviation parameters from the first evaluation indicator associated deviation parameter; Traversing the plurality of associated attribute deviation parameters and performing normalization processing to obtain a plurality of associated attribute normalized deviations; Extracting a plurality of correlation attribute correlations from the first evaluation indicator correlation; According to the correlation degrees of the plurality of correlation attributes, a weighted mean calculation is performed on the normalized deviations of the plurality of correlation attributes to obtain the first evaluation indicator correlation attribute deviation characteristic value.

[0074] Furthermore, the fitness calculation module 15 includes the following execution steps: Adding the first evaluation index associated attribute deviation characteristic value up to the Nth evaluation index associated attribute deviation characteristic value to obtain a deviation characteristic value sum value; Traverse the first evaluation indicator associated attribute deviation characteristic value until the Nth evaluation indicator associated attribute deviation characteristic value, compare with the sum of the deviation characteristic values, and obtain the first evaluation indicator weight until the Nth evaluation indicator weight.

[0075] Furthermore, the fitness calculation module 15 further includes the following execution steps: Obtaining a preset evaluation indicator expected value, calculating a deviation modulus from the preset evaluation indicator detection value, and obtaining a preset evaluation indicator deviation modulus value; Based on the first evaluation indicator weight up to the Nth evaluation indicator weight, the preset evaluation indicator deviation modulus values ​​are weighted and summed to obtain the first fitness.

[0076] Furthermore, the processing mode initialization module 16 includes the following execution steps: Perform particle swarm optimization based on the healthy sample set and fitness set to obtain expanded samples; Retrieving a sample set of the same modality as the expanded sample; When the proportion of abnormal samples in the same modality sample set is greater than or equal to the abnormal number proportion threshold, the expanded samples are deleted; When the proportion of abnormal samples in the homomodal sample set is less than the abnormal number proportion threshold, the mean value of the processing deviation parameter of the homomodal sample set is set as the expanded sample processing deviation parameter, and the mean value of the preset evaluation index detection value of the homomodal sample set is set as the expanded sample preset evaluation index detection value; Execute a cyclic optimization based on the expanded sample processing deviation parameter and the expanded sample preset evaluation index detection value.

[0077] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0078] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0082] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0083] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A multi-process drum optimization control method for processing Chinese herbal medicine slices, characterized in that: Applied to the Chinese herbal medicine slice processing drum, the Chinese herbal medicine slice processing drum includes a cleaning processing mode, a screening processing mode and a frying processing mode, including: Based on the roller processing mode, the specifications and varieties of Chinese herbal medicine pieces, and the roller model, a first healthy sample is collected, wherein the first healthy sample includes a first processing deviation parameter and a preset evaluation index detection value; Traversing the preset evaluation indicators, performing correlation analysis on the processing control attributes, obtaining a first evaluation indicator correlation attribute and a first evaluation indicator correlation degree, and so on until an Nth evaluation indicator correlation attribute and an Nth evaluation indicator correlation degree; Decomposing the first processing deviation parameter according to the first evaluation indicator association attribute to the Nth evaluation indicator association attribute to obtain first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters; Using the first evaluation indicator association degree to the Nth evaluation indicator association degree as normalized deviation weights, performing weighted mean calculation on the first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters to obtain the first evaluation indicator association attribute deviation characteristic value to the Nth evaluation indicator association attribute deviation characteristic value; Using the first evaluation indicator associated attribute deviation characteristic value up to the Nth evaluation indicator associated attribute deviation characteristic value, configuring the weight of the preset evaluation indicator detection value, and calculating the first fitness of the first healthy sample; The first healthy sample and the first fitness are associated and stored in an optimized particle space. When the number of healthy samples in the optimized particle space meets a preset number, optimization is performed to obtain desired processing parameters to initialize the drum processing mode.

2. The method according to claim 1, wherein Traversing the preset evaluation indicators, performing correlation analysis on the processing control attributes, obtaining a first evaluation indicator correlation attribute and a first evaluation indicator correlation degree, until an Nth evaluation indicator correlation attribute and an Nth evaluation indicator correlation degree, including: Extracting a first evaluation indicator of the preset evaluation indicators and extracting a first processing control attribute of the processing control attributes; Based on the drum processing mode, the specifications of the Chinese herbal medicine slices, and the varieties of the Chinese herbal medicine slices, combined with the drum model, a processing abnormality sample set is collected, wherein any processing abnormality sample in the processing abnormality sample set includes an abnormal processing control attribute set and an abnormality evaluation index set; Counting the ratio of the triggering frequency of the first evaluation indicator and the first processing control attribute in the same processing abnormality sample to the total number of processing abnormality samples; When the ratio is greater than or equal to a preset ratio, the first processing control attribute is added to the first evaluation index associated attribute; after the processing control attribute traversal is completed, a correlation analysis is performed on the first evaluation index associated attribute based on the first evaluation index to obtain the first evaluation index correlation degree; The evaluation indicators are updated and analyzed cyclically until the Nth evaluation indicator association attribute and the Nth evaluation indicator association degree are obtained.

3. The method according to claim 2, wherein Based on the first evaluation indicator, performing a correlation analysis on the first evaluation indicator correlation attribute to obtain the first evaluation indicator correlation degree, further comprising: Based on the roller processing mode, the specifications of the Chinese herbal medicine slices and the varieties of the Chinese herbal medicine slices, and in combination with the roller model, a plurality of first evaluation index associated attribute record values ​​and a plurality of first evaluation index record values ​​are collected; Performing same-attribute deviation calculation on the plurality of first evaluation indicator associated attribute record values ​​to obtain a plurality of first evaluation indicator associated attribute deviation modulus values; Performing deviation calculation on a plurality of first evaluation index recorded values ​​to obtain a plurality of first evaluation index recorded value deviation moduli; From the deviation moduli of the associated attributes of the several first evaluation indicators, extract the first associated attribute deviation modulus sequence up to the Qth associated attribute deviation modulus sequence, combine the deviation moduli of the several first evaluation indicator record values, construct a grey correlation matrix, perform correlation analysis, obtain the first associated attribute correlation degree up to the Qth associated attribute correlation degree, and add it into the first evaluation indicator correlation degree.

4. The method according to claim 1, wherein The method further comprises: using the first evaluation indicator association degree to the Nth evaluation indicator association degree as normalized deviation weights, performing weighted mean calculation on the first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters, and obtaining the first evaluation indicator association attribute deviation characteristic value to the Nth evaluation indicator association attribute deviation characteristic value, including: Extracting a plurality of associated attribute deviation parameters from the first evaluation indicator associated deviation parameter; Traversing the plurality of associated attribute deviation parameters and performing normalization processing to obtain a plurality of associated attribute normalized deviations; Extracting a plurality of correlation attribute correlations from the first evaluation indicator correlation; According to the correlation degrees of the plurality of correlation attributes, a weighted mean calculation is performed on the normalized deviations of the plurality of correlation attributes to obtain the first evaluation indicator correlation attribute deviation characteristic value.

5. The method according to claim 1, wherein Configuring the weight of the preset evaluation indicator detection value based on the first evaluation indicator associated attribute deviation characteristic value up to the Nth evaluation indicator associated attribute deviation characteristic value includes: Adding the first evaluation index associated attribute deviation characteristic value up to the Nth evaluation index associated attribute deviation characteristic value to obtain a deviation characteristic value sum value; Traverse the first evaluation indicator associated attribute deviation characteristic value until the Nth evaluation indicator associated attribute deviation characteristic value, compare with the sum of the deviation characteristic values, and obtain the first evaluation indicator weight until the Nth evaluation indicator weight.

6. The method according to claim 5, wherein Calculating a first fitness of the first healthy sample includes: Obtaining a preset evaluation indicator expected value, calculating a deviation modulus from the preset evaluation indicator detection value, and obtaining a preset evaluation indicator deviation modulus value; Based on the first evaluation indicator weight up to the Nth evaluation indicator weight, the preset evaluation indicator deviation modulus values ​​are weighted and summed to obtain the first fitness.

7. The method according to claim 1, wherein The first healthy sample and the first fitness are associated and stored in an optimized particle space. When the number of healthy samples in the optimized particle space meets a preset number, an optimization search is performed to obtain desired processing parameters, including: Perform particle swarm optimization based on the healthy sample set and fitness set to obtain expanded samples; Retrieving a sample set of the same modality as the expanded sample; When the proportion of abnormal samples in the same modality sample set is greater than or equal to the abnormal number proportion threshold, the expanded samples are deleted; When the proportion of abnormal samples in the homomodal sample set is less than the abnormal number proportion threshold, the mean value of the processing deviation parameter of the homomodal sample set is set as the expanded sample processing deviation parameter, and the mean value of the preset evaluation index detection value of the homomodal sample set is set as the expanded sample preset evaluation index detection value; Execute a cyclic optimization based on the expanded sample processing deviation parameter and the expanded sample preset evaluation index detection value.

8. A multi-process drum optimization control system for processing Chinese herbal medicine slices, characterized in that: For implementing the method according to any one of claims 1 to 7, the system is applied to a Chinese herbal medicine slice processing drum, the Chinese herbal medicine slice processing drum includes a cleaning processing mode, a screening processing mode and a frying processing mode, and the system includes: A health sample collection module is used to collect a first health sample based on the roller processing mode, the specifications and varieties of Chinese herbal medicine pieces, and the roller model, wherein the first health sample includes a first processing deviation parameter and a preset evaluation index detection value; a correlation analysis module, configured to traverse preset evaluation indicators, perform correlation analysis on the processing control attributes, obtain a first evaluation indicator correlation attribute and a first evaluation indicator correlation degree, and so on, until an Nth evaluation indicator correlation attribute and an Nth evaluation indicator correlation degree; a deviation parameter decomposition module, configured to decompose the first processing deviation parameter according to the first evaluation indicator association attribute to the Nth evaluation indicator association attribute, and obtain first evaluation indicator association deviation parameters to the Nth evaluation indicator association deviation parameters; a deviation characteristic calculation module, configured to use the first evaluation indicator association degree to the Nth evaluation indicator association degree as normalized deviation weights, perform weighted mean calculation on the first evaluation indicator association deviation parameter to the Nth evaluation indicator association deviation parameter, and obtain the first evaluation indicator association attribute deviation characteristic value to the Nth evaluation indicator association attribute deviation characteristic value; A fitness calculation module is configured to calculate a first fitness of the first healthy sample by configuring a weight of the preset evaluation indicator detection value based on the first evaluation indicator associated attribute deviation characteristic value up to the Nth evaluation indicator associated attribute deviation characteristic value; The processing mode initialization module is used to associate and store the first healthy sample and the first fitness into the optimized particle space. When the number of healthy samples in the optimized particle space meets the preset number, the optimization is performed to obtain the desired processing parameters to initialize the drum processing mode.

Images