Method for extracting and separating eurycomanone in eurycoma longifolia
By optimizing enzymatic hydrolysis parameters and dynamically adjusting enzyme dosage, temperature, and hydrolysis time, the problem of incomplete enzymatic hydrolysis in the extraction of tongkat aliquots was solved, achieving efficient extraction and improved purity, while reducing costs and risks.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHANGSHA HUAKANG BIOTECH DEV
- Filing Date
- 2026-03-27
- Publication Date
- 2026-04-28
AI Technical Summary
In the current technology for extracting ketone from Tongkat Ali, the enzymatic hydrolysis parameters are not optimized during the enzymatic hydrolysis process, resulting in incomplete cell wall destruction, which affects the extraction rate and purity. Furthermore, traditional methods may increase costs or introduce the risk of contamination by other microorganisms.
By using comprehensive evaluation methods and optimization algorithms, the enzyme dosage, temperature, and hydrolysis time are dynamically adjusted. Combined with multiple linear regression and decision tree algorithms, the hydrolysis parameters are optimized to achieve efficient cell wall breaking and improve the yield of broad ketone.
This improved the extraction yield and purity of fenestrated ketone, reduced the risk of enzyme inactivation and microbial contamination, and optimized the stability and economy of the process.
Smart Images
Figure CN121930243A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of ketone extraction technology, specifically to a method for extracting and separating ketone from Tongkat Ali. Background Technology
[0002] Enzymatic hydrolysis is a crucial pretreatment technique in the extraction and separation of tongkat aliquots. Its application stems from the limited effectiveness of traditional solvent extraction methods in disrupting the dense plant cell walls, resulting in low tongkat aliquot dissolution rates and numerous impurities. With advancements in bio-enzyme technology, this technique draws upon the modernization approach of traditional Chinese medicine, specifically hydrolyzing structural polysaccharides within the cell walls using cellulase and pectinase. It efficiently and gently disrupts cell walls, significantly improving the extraction efficiency and product purity of tongkat aliquots, while avoiding the damage to heat-sensitive components caused by high temperatures and reducing the use of organic solvents. This embodies a green and environmentally friendly extraction philosophy, laying a solid foundation for subsequent high-precision separation and purification.
[0003] During enzymatic hydrolysis, if the amount of cellulase used is relatively small compared to the raw material, it cannot effectively break down the tough cell wall structure of Tongkat Ali, thus affecting the degradation of the cellulose skeleton. Furthermore, if the temperature and reaction time of the enzymatic hydrolysis are not systematically optimized, the speed and efficiency of the hydrolysis may not meet expectations. This can lead to a reduced extraction rate when the target product, fenestrated ketone, is released from Tongkat Ali due to insufficient cell wall disintegration, resulting in a significant decrease in extraction efficiency in subsequent extraction steps and ultimately affecting the overall yield of the process. Current technologies typically improve the destruction effect by extending the hydrolysis time, increasing the amount of enzyme, or raising the reaction temperature. However, extending the reaction time may adversely affect subsequent production processes and increase the risk of microbial contamination; arbitrarily increasing the amount of enzyme will raise the production cost of the product; and excessively high temperatures may cause denaturation and inactivation of enzyme proteins. Therefore, there is a lack of an effective method that can balance efficiency and cost to achieve efficient and controllable destruction of the cell wall. Summary of the Invention
[0004] In view of the above, it is necessary to provide a method for the extraction and separation of ketones from Tongkat Ali to solve the above problems.
[0005] One embodiment of this application provides a method for extracting and separating ketone from Tongkat Ali, the method comprising: For each production batch of fenestrate extraction and separation, obtain the enzymatic hydrolysis parameters and the actual yield of fenestrate after the process is completed; A comprehensive evaluation method was used to evaluate the enzymatic hydrolysis parameters and actual yield of each production batch and all previous production batches to obtain an evaluation score for each production batch. Based on the evaluation score, all production batches were classified, and the difference in enzymatic hydrolysis effect of each production batch was determined based on the dispersion of the number of data in all categories. Combined with the overall distribution of the evaluation scores, the enzymatic hydrolysis condition discrimination of each production batch was determined. The linear relationship between the discrimination of enzymatic hydrolysis conditions and the actual yield was analyzed for each production batch and all previous production batches to obtain the predicted yield. The confidence factor for the validity of the enzymatic hydrolysis parameters for each production batch was determined by combining the homogeneity of all enzymatic hydrolysis parameters in each production batch. Based on all enzymatic hydrolysis parameters for each production batch, an optimization algorithm is used to maximize the validation confidence factor to obtain the optimal enzyme dosage. Combining the validation confidence factor and the cellulase dosage in the current enzymatic hydrolysis parameters, the cellulase dosage for the next production batch is updated.
[0006] Preferably, the specific process flow for each production batch is as follows: Tongkat Ali raw material was crushed and soaked in water. Cellulase preparation was added to obtain enzymatically hydrolyzed Tongkat Ali. Water was added and the extraction was carried out continuously in a countercurrent manner to obtain Tongkat Ali extract. After cooling, the extract was centrifuged to obtain defatted liquid. The defatted liquid was subjected to ultrafiltration and nanofiltration to obtain ketone nanofiltration retentate. The ketone nanofiltration retentate was loaded with macroporous resin and eluted and concentrated to obtain concentrated liquid. The concentrated liquid was subjected to alcohol phase crystallization and drying to obtain ketone product.
[0007] Preferably, the specific operation of adding cellulase preparation to obtain enzymatically hydrolyzed Tongkat Ali is as follows: add cellulase preparation, the amount of enzyme is 1~1.5g, and enzymatically hydrolyze at 35~45°C for 90~150min to obtain enzymatically hydrolyzed Tongkat Ali.
[0008] Preferably, the ultrafiltration and nanofiltration of the degreasing solution to obtain the ketone nanofiltration retentate is specifically as follows: The degreasing liquid was subjected to ultrafiltration using an ultrafiltration membrane with a molecular weight cutoff of 40,000–60,000 Da at an operating pressure of 1.0–2.0 MPa and a material temperature of 25–35°C to obtain an ultrafiltrate. The ultrafiltrate was then subjected to nanofiltration using a nanofiltration membrane with a molecular weight cutoff of 5,000–7,000 Da at an operating pressure of 1.5–2.5 MPa and a material temperature of 25–35°C until the conductivity of the filtrate was 250–350 μS / cm, yielding a broad-flavonoid nanofiltration retentate.
[0009] Preferably, the specific steps for applying macroporous resin to the nanofiltration retentate of broad ketone and then eluting and concentrating it to obtain a concentrated solution are as follows: the macroporous resin is HP20 type, and the resin volume is 4% to 6.7% of the nanofiltration retentate volume; first, elute with an ethanol aqueous solution with a volume concentration of 5% to 15% for 1 to 3 BV, discard the eluent, then elute with an ethanol aqueous solution with a volume concentration of 70% to 90% for 1 to 3 BV, collect the eluent, and concentrate it under vacuum at a vacuum degree of -0.10 to -0.06 MPa and a temperature of 60 to 70°C to obtain a concentrated solution.
[0010] Preferably, the specific steps for obtaining the ketone product by performing alcohol phase crystallization and drying on the concentrated solution are as follows: 1.0~1.5L of anhydrous ethanol is added to the concentrated solution, stirred thoroughly, filtered, and cooled and crystallized at 0~5°C for 18~30h to obtain crystals; the crystals are washed 1~2 times with ice water at 0~5°C, each time using 5~15 times the volume of the crystals, and the cooling and crystallization are repeated 1~2 times and washed, and finally vacuum dried to obtain the ketone product.
[0011] Preferably, determining the distinguishability of enzymatic hydrolysis conditions for each production batch specifically involves: The ratio of the overall distribution characteristics of the evaluation scores obtained from all historical batches after normalization to the difference value of the enzymatic hydrolysis effect is used to obtain the enzymatic hydrolysis condition discrimination degree of each production batch.
[0012] Preferably, the predicted yield is obtained by performing a multiple linear fit on the enzymatic hydrolysis condition discrimination and actual yield of each production batch and all previous production batches.
[0013] Preferably, the determination of the validation confidence factor for the enzymatic hydrolysis parameters of each production batch specifically involves: A decision tree algorithm was used to obtain the average purity value of all nodes for all enzymatic hydrolysis parameters of each production batch. Obtain the difference between the predicted yield and the actual yield, and calculate the ratio to the actual maximum yield. Then, positively fuse the difference between the natural number 1 and the obtained ratio with the mean purity of all nodes to obtain the validation confidence factor of the enzymatic hydrolysis parameters for each production batch.
[0014] Preferably, the formula for updating the cellulase dosage for the next production batch is as follows: In the formula, This indicates an update in enzyme dosage; E represents the optimal enzyme dosage; B represents the current enzyme dosage; and B represents the confidence factor for the test. This indicates the preset learning rate.
[0015] This application has at least the following beneficial effects: First, to address the issue of mutual interference among enzymatic hydrolysis parameters during the enzymatic hydrolysis process and the inability to independently determine their degree of influence, this study comprehensively utilizes TOPSIS and K-means clustering to reflect the synergistic effect of the balance factors of enzyme activity and hydrolysis time-temperature. This eliminates the one-sidedness caused by analyzing a single parameter alone, effectively distinguishing the combination of conditions with high, medium, and low enzymatic hydrolysis effects, and providing a reference for subsequent process optimization.
[0016] Secondly, to address the problem that traditional regression methods struggle to achieve high prediction accuracy when predicting nonlinear enzymatic hydrolysis processes, a fusion algorithm combining multiple linear regression and decision tree is employed to represent the consistency and prediction accuracy characteristics of parameter combinations, eliminating the influence of noisy data and overfitting, and improving the reliability of yield condition identification.
[0017] Finally, by employing a particle swarm optimization-based dynamic adjustment strategy for enzyme dosage, and utilizing validation confidence factors, the optimal combination of enzyme dosage, temperature, and hydrolysis time was achieved in real time. This allowed for accurate and efficient control of the cell wall decomposition process. While significantly improving the yield and purity of ketone extraction, this approach avoided enzyme inactivation, microbial contamination, and cost waste, resulting in a substantial improvement in both process stability and economic efficiency. Attached Figure Description
[0018] Figure 1 A process flow diagram of a method for extracting and separating quinone from Tongkat Ali provided in this application; Figure 2 A flowchart illustrating the method for improving the yield of broad-flavor ketone provided in this application. Detailed Implementation
[0019] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.
[0020] 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 application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
[0021] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.
[0022] 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 application pertains.
[0023] Example 1 This application proposes a method for the extraction and separation of fenestrate from Tongkat Ali, the process flow diagram of which is attached. Figure 1 The process includes the following steps: (1) Enzymatic hydrolysis: 5 kg of Tongkat Ali raw material is crushed to 10 mesh, then soaked in 7 L of water, and cellulase preparation is added. The amount of enzyme is 1~1.5 g. Enzymatic hydrolysis is carried out at 35~45°C for 90~150 min to obtain enzymatic hydrolyzed Tongkat Ali. (2) Heating extraction: Add 30 kg of water to the enzymatically hydrolyzed Tongkat Ali obtained in step (1) and continuously extract countercurrently at 80°C for 120 min to obtain Tongkat Ali extract; (3) Centrifugation and degreasing: After cooling the Tongkat Ali extract obtained in step (2) to 5°C, first centrifuge it at a rate of 3500 r / min in a horizontal screw, and then centrifuge it at a rate of 8000 r / min in a tube to obtain the degreased liquid. (4) Ultrafiltration and nanofiltration: The degreasing liquid obtained in step (3) is subjected to ultrafiltration using an ultrafiltration membrane with a molecular weight cutoff of 40,000 Da at an operating pressure of 1.0 MPa and a material temperature of 25°C to obtain ultrafiltrate; the ultrafiltrate is then subjected to nanofiltration using a nanofiltration membrane with a molecular weight cutoff of 5,000 Da at an operating pressure of 1.5 MPa and a material temperature of 25°C until the conductivity of the filtrate is 250 μS / cm to obtain the ketone nanofiltration retentate; (5) Resin chromatography: The ketone nanofiltration retentate obtained in step (4) is loaded onto a macroporous resin, the macroporous resin being HP20 type, with a resin volume of 4% of the nanofiltration retentate volume. First, 1 BV is eluted with a 5% ethanol aqueous solution, and the eluent is discarded. Then, 1 BV is eluted with a 70% ethanol aqueous solution, and the eluent is collected. The eluent is then concentrated under vacuum at a vacuum degree of -0.10 MPa and a temperature of 70°C to obtain a concentrated solution. (6) Alcohol phase crystallization and drying: Add 1.0L of anhydrous ethanol to the concentrated solution obtained in step (5), stir thoroughly and filter, cool and crystallize at 0°C for 18h to obtain crystals; wash the crystals twice with ice water at 0°C, each time using 15 times the volume of the crystals, repeat the cooling and crystallization twice and washing, and finally vacuum dry to obtain the ketone product.
[0024] The obtained fenestrated ketone product was a white powder. The peak time, purity, and final yield of fenestrated ketone were obtained by high performance liquid chromatography.
[0025] The flowchart of the method for improving the yield of broadleaf ketone in this application is as follows: Figure 2 As shown, the method includes the following steps: Step 1: For each production batch of fenestrate extraction and separation, obtain the enzymatic hydrolysis parameters and the actual yield of fenestrate after the process is completed.
[0026] During the enzymatic hydrolysis process, enzymatic hydrolysis parameters and yield data for each batch of production are collected and recorded. Specifically, this includes: adding a high-precision temperature sensor to the enzymatic hydrolysis reactor to collect hydrolysis temperature data, ensuring that the cellulase reacts within its optimal temperature range during the hydrolysis process; equipping the enzyme dosing point with an electronic scale to collect the amount of cellulase added, ensuring sufficient cellulase for decomposing the cellulose skeleton during the hydrolysis reaction; integrating a digital timer into the automated control system to collect the hydrolysis time, ensuring accurate recording of the hydrolysis time with allowance; and finally, using high-performance liquid chromatography (HPLC) to collect the actual yield of fenestrated ketone after the process is completed. By comparing the purity of the hydrolyzed product with the yield of this particular hydrolysis, the overall yield of the enzymatic hydrolysis process is calculated, ensuring the overall yield and the effectiveness of subsequent hydrolysis processes.
[0027] Enzymatic hydrolysis parameters (enzyme dosage, temperature, and hydrolysis time) and actual yield data were collected for each production batch and all previous production batches. All collected data (enzyme dosage, temperature, and hydrolysis time) were normalized to remove the influence of dimensions. In this method, the enzyme dosage, temperature, hydrolysis time, and actual yield data are all process data collected and organized batch by batch, corresponding to batch numbers, thus providing a reasonable and reliable data foundation for subsequent data analysis and process optimization.
[0028] For each production batch, the data obtained from all previous historical batches are sorted in ascending order of collection time. It should be noted that the enzymatic hydrolysis parameters of the initial production batch are set to random values within the range of process parameters. The enzyme dosage is adjusted only after the initial number of production batches has been accumulated. In this embodiment, the enzyme dosage adjustment is started after accumulating 20 initial production batches.
[0029] At this point, the enzymatic hydrolysis parameter sequences (enzyme dosage sequence, temperature sequence, and hydrolysis time sequence) and the actual yield data sequence for each production batch were obtained.
[0030] Step 2: A comprehensive evaluation method is used to evaluate the enzymatic hydrolysis parameters and actual yield of each production batch and all previous production batches to obtain an evaluation score for each production batch; based on the evaluation score, all production batches are classified, and the difference in enzymatic hydrolysis effect of each production batch is determined based on the dispersion of the number of data in all categories. Combined with the overall distribution of the evaluation scores, the enzymatic hydrolysis condition discrimination of each production batch is determined.
[0031] Because Tongkat Ali cell walls are relatively robust and primarily composed of cellulose, if the cellulase is not effective during enzymatic hydrolysis, the cell walls cannot be fully opened, and the quinone cannot be completely released. Therefore, in the production process, if the enzymatic hydrolysis parameters are not properly optimized, the degree of hydrolysis may be insufficient, or the amount of enzyme may be inadequate, resulting in some cellulose skeletons not being completely hydrolyzed. Furthermore, if the temperature is not within the optimal range, enzyme activity will decrease, and the hydrolysis time may be insufficient, leading to inadequate hydrolysis. Consequently, the cell walls cannot be fully opened, preventing the complete release of quinone and affecting the extraction efficiency during subsequent leaching, resulting in a low final yield.
[0032] Using the enzymatic hydrolysis parameters (enzyme dosage, temperature, and hydrolysis time) and actual yield data of all previous historical production batches as inputs to the Top-Order Ideal Solution Approximation Method (TOPSIS), the enzymatic hydrolysis parameters (enzyme dosage, temperature, and hydrolysis time) and actual yield data are used as evaluation objects. Each enzymatic hydrolysis parameter (enzyme dosage, temperature, and hydrolysis time) is evaluated using an intermediate index. Specifically, the median value of the three process parameter ranges (enzyme dosage of 1~1.5g, hydrolysis at 35~45°C for 90~150min) in the above process flow is used as the ideal value of the median index of the three enzymatic hydrolysis parameters. This balances extraction efficiency and cost, avoids excessively large parameters leading to enzyme inactivation, increased impurities, or cost waste, and also prevents excessively small parameters from causing insufficient cell wall breakdown, ensuring a stable and reliable process. The actual yield data is evaluated using a very large index. The evaluation score S for each production batch is output by using the Top-to-Ideal Solution Ranking Method (TOPSIS). The evaluation score S reflects the comprehensive performance of each production batch in terms of the degree to which the enzymatic hydrolysis parameters fall within the optimal range and the actual yield. The higher the score, the closer the enzymatic hydrolysis conditions of the batch are to the ideal state and the stronger the overall enzymatic hydrolysis effect.
[0033] However, while a single evaluation score S can rank the overall performance of a single batch, it cannot automatically identify batch groups with similar performance levels from historical data, making it difficult to grasp the overall distribution pattern of process conditions macroscopically. Therefore, the evaluation scores S of all historical production batches before each production batch are used as input, and the K-means clustering algorithm is used. The number of clusters is set to K=3, and 3 clusters are output. According to the average of all evaluation scores in the clusters, low, medium and high enzymatic hydrolysis effect groups are obtained from low to high. Then, the dispersion of the number of data in all clusters is calculated as the difference value of enzymatic hydrolysis effect for each production batch. In this embodiment, the standard deviation is used to measure the dispersion of the number of data in all clusters. In other embodiments, the range value can also be used to measure the dispersion.
[0034] Based on the above analysis, a distinguishing factor A for each production batch's enzymatic hydrolysis conditions is constructed. Specifically, the ratio of the overall distribution characteristics of the evaluation scores obtained from all historical batches after normalization to the difference in enzymatic hydrolysis effect is calculated as the distinguishing factor for each production batch's enzymatic hydrolysis conditions. In this embodiment, the overall distribution characteristics of the data are calculated using the average value; in other embodiments, the median value can also be used.
[0035] It should be understood that the Top-Approximation-to-Ideal-Solution Ranking (TOPSIS) method is used for multi-index decision analysis, which comprehensively evaluates each scheme by calculating its relative closeness to the ideal solution; the K-means clustering algorithm is used for unsupervised pattern recognition, which reveals the data distribution structure by grouping data points into different clusters, and the standard deviation of the number of data points within each cluster characterizes the consistency of cluster size. In this application, the mean of the normalized evaluation score indicates that the closer the enzymatic hydrolysis conditions of the batch are to the ideal state, the more it reflects the comprehensive efficiency of parameter settings and yield output; the consistency of cluster size is characterized by the difference in enzymatic hydrolysis effect value, and the larger the value, the more uneven the distribution of enzymatic hydrolysis effect level of historical batches, reflecting the dispersion of process condition efficiency; by calculating the product of these two values, the synergistic effect of the dispersion of efficiency distribution and the overall efficiency level is captured, thereby achieving a quantitative evaluation of the degree of differentiation of enzymatic hydrolysis conditions of historical batches, providing macro-distribution insights for process optimization.
[0036] To capture the temporal variation trend of the enzymatic hydrolysis condition discrimination, reduce the interference of transient fluctuations, and provide stable sequence data for subsequent models to optimize enzymatic hydrolysis parameters, for each production batch, the enzymatic hydrolysis condition discrimination calculated from all previous production batches is sorted according to the batch sorting order to obtain the enzymatic hydrolysis condition discrimination sequence for each production batch.
[0037] Step 3: Analyze the linear relationship between the differentiation of enzymatic hydrolysis conditions and the actual yield for each production batch and all previous production batches to obtain the predicted yield. Combine the homogeneity of all enzymatic hydrolysis parameters in each production batch to determine the validation confidence factor of the enzymatic hydrolysis parameters for each production batch.
[0038] In the enzymatic hydrolysis of Tongkat Ali extract, key parameters such as cellulase dosage, reaction temperature, and hydrolysis time lack data-driven systematic optimization and rely solely on empirical settings, resulting in low hydrolysis efficiency. Specifically, Tongkat Ali raw material has a tough cell wall structure, with cellulose as its main component. Insufficient enzyme dosage cannot fully hydrolyze the cellulose skeleton, while too short a hydrolysis time leads to incomplete reaction, and too long a hydrolysis time increases the risk of contamination by other microorganisms and costs. These factors combined result in incomplete cell wall breakdown, hindered release of ketones from the cell walls, and consequently, low extraction efficiency.
[0039] Therefore, using the enzyme hydrolysis condition discrimination sequence and actual yield data sequence for each production batch as input, a model is built based on a multiple linear regression algorithm. The regularization parameter is set to 0 to prevent overfitting and ensure the model's generalization ability. The output is the fitting slope k and the predicted yield. The fitting slope k represents the strength of the positive correlation between index A and the actual yield of broad ketone, and the predicted yield. This represents the theoretically highest yield that can be achieved with current enzymatic hydrolysis.
[0040] Multiple linear regression assumes linearity, but enzymatic hydrolysis is non-linear. For example, enzyme activity is temperature-dependent, leading to inaccurate predictions under different conditions and an inability to identify high-yield combinations. Using the discriminant sequence of enzymatic hydrolysis conditions and the sequence of enzymatic hydrolysis parameters (enzyme dosage, temperature, and hydrolysis duration) for each production batch, a decision tree algorithm was employed. The maximum depth of the decision tree was set to 5 to prevent overfitting, balancing complexity and interpretability; the minimum number of sample splits was set to 10 to ensure a sufficiently large number of nodes after branching, facilitating sample generalization; and noise was avoided. Finally, the mean purity P of all nodes was output. Based on classification rules, the decision path for high-yield conditions was identified. The larger the mean purity P of all nodes (closer to 1), the more consistent and reliable the conditions, and the more beneficial they are for cell wall disruption.
[0041] Based on the above analysis, a validation confidence factor B for the enzymatic hydrolysis parameters of each production batch is constructed. The specific calculation process is as follows: the difference between the predicted yield and the actual yield is obtained, and the ratio of this difference to the actual maximum yield is calculated. The difference between the natural number 1 and the obtained ratio is positively fused with the purity mean of all nodes to obtain the validation confidence factor for the enzymatic hydrolysis parameters of each production batch. In this embodiment, the difference between the predicted yield and the actual yield is calculated using the absolute value of the difference, and the absolute value of the difference between the predicted yield and the actual yield is recorded as the prediction error. The division method is used to compare one variable with another. The multiplication method is used to positively fuse multiple variables.
[0042] It should be understood that the mean purity P of all nodes measures the homogeneity of sample categories within a node. The larger the P value, the more uniform the sample categories within that node. The comparison results are used as a relative evaluation index of the model's prediction accuracy. The smaller the value, the closer the prediction results are to the true values.
[0043] It's further important to understand that the mean purity can be used to represent the true reliability of high-yield combinations of conditions; the resulting comparisons are used to quantitatively represent the gap between model predictions and actual conditions; in enzymatic hydrolysis process optimization, the key objective is to maximize the prediction error. Whether the predicted value is higher or lower than the actual value, a large error indicates inaccurate model prediction. The prediction error directly reflects the magnitude of the error, without distinguishing between positive and negative directions, thus avoiding the possibility that positive errors (over-prediction) and negative errors (under-prediction) might cancel each other out, leading to incorrect judgments about model performance. Therefore, the validation confidence factor comprehensively considers condition consistency, influence strength, and prediction accuracy. A larger validation confidence factor indicates that the current combination of enzymatic hydrolysis parameters has strong credibility and good consistency, while also having a significant impact on yield and accurate model prediction, indicating high confidence in optimization and the ability to effectively guide process optimization.
[0044] Step 4: Based on all enzymatic hydrolysis parameters for each production batch, an optimization algorithm is used to maximize the validation confidence factor to obtain the optimal enzyme dosage. Combining the validation confidence factor and the cellulase dosage in the current enzymatic hydrolysis parameters, the cellulase dosage for the next production batch is updated.
[0045] The cell walls of Tongkat Ali raw materials are very compact, and the cellulose skeleton is difficult to break down using ordinary enzymatic hydrolysis methods. The parameters such as enzyme dosage, temperature, and hydrolysis time in traditional processes have not been systematically optimized, resulting in the cellulase not playing a full hydrolytic role. This leads to low release efficiency of fenestrated ketone and unstable extraction yield. Blindly adjusting parameters can easily cause problems such as enzyme inactivation, increased costs, and contamination by other microorganisms, which in turn affect industrial production control and economic benefits.
[0046] Therefore, it is necessary to design improvement strategies for the technical problems based on the validation confidence factor B: The algorithm uses a combination of enzyme dosage, temperature, and hydrolysis time as a particle, with 20 particles iterating 100 times. A particle swarm optimization algorithm is employed, randomly initializing the parameter range and setting dynamic inertia weights to maximize the validation confidence factor B. This strategy dynamically adjusts the parameters based on the B value to find the optimal combination of enzyme dosage, temperature, and hydrolysis time. Then, based on the process confidence represented by B, the hydrolysis parameters are dynamically adjusted: a small B value indicates that the current hydrolysis conditions are not ideal for cell wall breakdown and the system needs to move closer to the optimal parameters; a large B value indicates that the current hydrolysis conditions are relatively stable and the system can move closer to the target. The final output maximizes the validation confidence factor B and the optimal combination of enzyme dosage, temperature, and hydrolysis time.
[0047] It should be understood that the particle swarm optimization algorithm systematically searches for the global optimum in the parameter space through swarm intelligence, and outputs... This represents the reference value for the amount of enzyme used to most effectively break down cellulose cell walls, taking into account factors such as temperature and enzymatic hydrolysis time. However, what we get is the optimal amount of enzyme used to maximize the confidence factor of the assay. Since there may be differences in enzyme raw materials and enzyme storage environment, this application combines the confidence factor of the current batch to make local adjustments to the amount of enzyme used in the next batch, so as to better adapt to the changes between different batches.
[0048] Based on this, an improved formula for enzyme dosage is constructed: In the formula, This indicates an update in enzyme dosage; E represents the optimal enzyme dosage; B represents the current enzyme dosage; and B represents the confidence factor for the test. This represents the preset learning rate, which ranges from 0.2 to 0.4, allowing the enzyme dosage to achieve a good balance between convergence speed and stability. In this embodiment, the specific value is 0.3.
[0049] During the dynamic optimization process, the improved formula for constraining enzyme dosage is used to ensure the updated enzyme dosage. The enzyme dosage should always be kept within the range of 1 to 1.5g. If the amount of enzyme used after constraint exceeds this range, the amount of enzyme used after constraint should be set to 1.5g. If the amount of enzyme used after constraint is less than this range, the amount of enzyme used after constraint should be set to 1g. This can effectively break down the cell wall of Tongkat Ali and avoid cost waste or enzyme inactivation caused by excessive enzyme dosage. This ensures the stability and economy of the process while optimizing extraction efficiency.
[0050] This application constructs an improved enzyme dosage formula based on global optimal search and local dynamic adjustment, wherein... This can be viewed as an updated weight, quantifying the degree of instability of the current conditions. Approaching 1 (model reliability), update the weights to the maximum, and make full use of the optimization results. ;when When the weights approach 0 (the model is unreliable), update the weights until they approach 0, then maintain the current parameters. The formula remains unchanged to ensure safety. By controlling the step size through the α coefficient, it achieves precise improvement in the release efficiency of broadleaf ketone while ensuring effective cell wall breakdown and avoiding the risk of increased costs or enzyme inactivation due to excessive fluctuations in enzyme dosage.
[0051] Example 2 This application proposes a method for the extraction and separation of fenestrate from Tongkat Ali, the process flow diagram of which is attached. Figure 1 The process includes the following steps: (1) Enzymatic hydrolysis: 5 kg of Tongkat Ali raw material is crushed to 30 mesh, then soaked in 15 L of water, and cellulase preparation is added. The amount of enzyme is 1~1.5 g. Enzymatic hydrolysis is carried out at 35~45°C for 90~150 min to obtain enzymatic hydrolyzed Tongkat Ali.
[0052] (2) Heating extraction: Add 45 kg of water to the enzymatically hydrolyzed Tongkat Ali obtained in step (1) and continuously extract countercurrently at 90°C for 60 min to obtain Tongkat Ali extract.
[0053] (3) Centrifugation and degreasing: After cooling the Tongkat Ali extract obtained in step (2) to 15°C, first centrifuge it at a rate of 2500 r / min in a horizontal screw, and then centrifuge it at a rate of 12000 r / min in a tube to obtain the degreased liquid.
[0054] (4) Ultrafiltration and nanofiltration: The degreasing liquid obtained in step (3) is subjected to ultrafiltration using an ultrafiltration membrane with a molecular weight cutoff of 60,000 Da at an operating pressure of 2.0 MPa and a material temperature of 35°C to obtain ultrafiltrate; the ultrafiltrate is then subjected to nanofiltration using a nanofiltration membrane with a molecular weight cutoff of 7,000 Da at an operating pressure of 2.5 MPa and a material temperature of 35°C until the conductivity of the filtrate is 350 μS / cm to obtain ketone nanofiltration retentate.
[0055] (5) Resin chromatography: The ketone nanofiltration retentate obtained in step (4) is loaded onto a macroporous resin, the macroporous resin being HP20 type, with a resin volume of 6.7% of the nanofiltration retentate volume. First, 3BV is eluted with a 15% ethanol aqueous solution, and the eluent is discarded. Then, 3BV is eluted with a 90% ethanol aqueous solution, and the eluent is collected. The eluent is then concentrated under vacuum at a vacuum degree of -0.06 MPa and a temperature of 60°C to obtain a concentrated solution.
[0056] (6) Alcohol phase crystallization and drying: Add 1.5L of anhydrous ethanol to the concentrated solution obtained in step (5), stir thoroughly and filter, cool and crystallize at 5°C for 30h to obtain crystals; wash the crystals once with ice water at 5°C, each time using 5 times the volume of the crystals, repeat the cooling and crystallization twice and washing, and finally vacuum dry to obtain the ketone product.
[0057] The amount of cellulase used was adjusted according to steps 1 to 4 in Example 1 to improve the yield of broadleaf ketone.
[0058] Example 3 This application proposes a method for the extraction and separation of fenestrate from Tongkat Ali, the process flow diagram of which is attached. Figure 1 The process includes the following steps: (1) Enzymatic hydrolysis: 5 kg of Tongkat Ali raw material is crushed to 20 mesh, then soaked in 12 L of water, and cellulase preparation is added. The amount of enzyme is 1~1.5 g. Enzymatic hydrolysis is carried out at 35~45°C for 90~150 min to obtain enzymatic hydrolyzed Tongkat Ali.
[0059] (2) Heating extraction: Add 43 kg of water to the enzymatically hydrolyzed Tongkat Ali obtained in step (1) and continuously extract countercurrently at 82°C for 80 min to obtain Tongkat Ali extract.
[0060] (3) Centrifugation and degreasing: After cooling the Tongkat Ali extract obtained in step (2) to 9°C, first centrifuge it at a rate of 2800 r / min in a horizontal screw, and then centrifuge it at a rate of 10000 r / min in a tube to obtain the degreased liquid.
[0061] (4) Ultrafiltration and nanofiltration: The degreasing liquid obtained in step (3) is subjected to ultrafiltration using an ultrafiltration membrane with a molecular weight cutoff of 55000 Da at an operating pressure of 2.0 MPa and a material temperature of 35°C to obtain ultrafiltrate; the ultrafiltrate is then subjected to nanofiltration using a nanofiltration membrane with a molecular weight cutoff of 6000 Da at an operating pressure of 1.8 MPa and a material temperature of 30°C until the conductivity of the filtrate is 300 μS / cm to obtain ketone nanofiltration retentate.
[0062] (5) Resin chromatography: The ketone nanofiltration retentate obtained in step (4) is loaded onto a macroporous resin, the macroporous resin being HP20 type, with a resin volume of 5.2% of the nanofiltration retentate volume. First, 3 BV is eluted with a 10% ethanol aqueous solution, and the eluent is discarded. Then, 1 BV is eluted with a 90% ethanol aqueous solution, and the eluent is collected. The eluent is then concentrated under vacuum at a vacuum degree of -0.09 MPa and a temperature of 67°C to obtain a concentrated solution.
[0063] (6) Alcohol phase crystallization and drying: Add 1.2L of anhydrous ethanol to the concentrated solution obtained in step (5), stir thoroughly and filter, cool and crystallize at 2°C for 24h to obtain crystals; wash the crystals once with ice water at 0°C, each time using 9 times the volume of the crystals, repeat the cooling and crystallization twice and washing, and finally vacuum dry to obtain the ketone product.
[0064] The amount of cellulase used was adjusted according to steps 1 to 4 in Example 1 to improve the yield of broadleaf ketone.
[0065] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.
[0066] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for extracting and separating ketone from Tongkat Ali, characterized in that, The method includes: For each production batch of fenestrate extraction and separation, obtain the enzymatic hydrolysis parameters and the actual yield of fenestrate after the process is completed; A comprehensive evaluation method was used to evaluate the enzymatic hydrolysis parameters and actual yield of each production batch and all previous production batches to obtain an evaluation score for each production batch. Based on the evaluation score, all production batches were classified, and the difference in enzymatic hydrolysis effect of each production batch was determined based on the dispersion of the number of data in all categories. Combined with the overall distribution of the evaluation scores, the enzymatic hydrolysis condition discrimination of each production batch was determined. The linear relationship between the discrimination of enzymatic hydrolysis conditions and the actual yield was analyzed for each production batch and all previous production batches to obtain the predicted yield. The confidence factor for the validity of the enzymatic hydrolysis parameters for each production batch was determined by combining the homogeneity of all enzymatic hydrolysis parameters in each production batch. Based on all enzymatic hydrolysis parameters for each production batch, an optimization algorithm is used to maximize the validation confidence factor to obtain the optimal enzyme dosage. Combining the validation confidence factor and the cellulase dosage in the current enzymatic hydrolysis parameters, the cellulase dosage for the next production batch is updated.
2. The method for extracting and separating ketone from Tongkat Ali as described in claim 1, characterized in that, The specific process flow for each production batch is as follows: Tongkat Ali raw material was crushed and soaked in water. Cellulase preparation was added to obtain enzymatically hydrolyzed Tongkat Ali. Water was added and the extraction was carried out continuously in a countercurrent manner to obtain Tongkat Ali extract. After cooling, the extract was centrifuged to obtain defatted liquid. The defatted liquid was subjected to ultrafiltration and nanofiltration to obtain ketone nanofiltration retentate. The ketone nanofiltration retentate was loaded with macroporous resin and eluted and concentrated to obtain concentrated liquid. The concentrated liquid was subjected to alcohol phase crystallization and drying to obtain ketone product.
3. The method for extracting and separating ketone from Tongkat Ali as described in claim 2, characterized in that, The specific operation of adding cellulase preparation to obtain enzymatically hydrolyzed Tongkat Ali is as follows: add cellulase preparation, the amount of enzyme is 1~1.5g, and enzymatically hydrolyze at 35~45°C for 90~150min to obtain enzymatically hydrolyzed Tongkat Ali.
4. The method for extracting and separating ketone from Tongkat Ali as described in claim 2, characterized in that, The degreasing solution is subjected to ultrafiltration and nanofiltration to obtain a broad-flavor ketone nanofiltration retentate, specifically: The degreasing liquid was subjected to ultrafiltration using an ultrafiltration membrane with a molecular weight cutoff of 40,000–60,000 Da at an operating pressure of 1.0–2.0 MPa and a material temperature of 25–35°C to obtain an ultrafiltrate. The ultrafiltrate was then subjected to nanofiltration using a nanofiltration membrane with a molecular weight cutoff of 5,000–7,000 Da at an operating pressure of 1.5–2.5 MPa and a material temperature of 25–35°C until the conductivity of the filtrate was 250–350 μS / cm, yielding a broad-flavonoid nanofiltration retentate.
5. The method for extracting and separating ketone from Tongkat Ali as described in claim 2, characterized in that, The specific steps for applying macroporous resin to the nanofiltration retentate of broad-leaved ketone and then eluting and concentrating it to obtain a concentrated solution are as follows: the macroporous resin is HP20 type, and the resin volume is 4% to 6.7% of the nanofiltration retentate volume; first, elute with an ethanol aqueous solution with a volume concentration of 5% to 15% for 1 to 3 BV, discard the eluent, then elute with an ethanol aqueous solution with a volume concentration of 70% to 90% for 1 to 3 BV, collect the eluent, and concentrate it under vacuum at a vacuum degree of -0.10 to -0.06 MPa and a temperature of 60 to 70°C to obtain a concentrated solution.
6. The method for extracting and separating ketone from Tongkat Ali as described in claim 2, characterized in that, The specific steps for obtaining the ketone product by alcohol-phase crystallization and drying of the concentrate are as follows: 1.0-1.5 L of anhydrous ethanol is added to the concentrate, stirred thoroughly, filtered, and cooled and crystallized at 0-5°C for 18-30 h to obtain crystals; the crystals are washed 1-2 times with ice water at 0-5°C, each time using 5-15 times the volume of the crystals, and the cooling and crystallization are repeated 1-2 times, followed by washing, and finally vacuum drying to obtain the ketone product.
7. The method for extracting and separating ketone from Tongkat Ali as described in claim 1, characterized in that, The determination of the enzymatic hydrolysis condition differentiation for each production batch specifically involves: The ratio of the overall distribution characteristics of the evaluation scores obtained from all historical batches after normalization to the difference value of the enzymatic hydrolysis effect is used to obtain the enzymatic hydrolysis condition discrimination degree of each production batch.
8. The method for extracting and separating ketone from Tongkat Ali as described in claim 1, characterized in that, The predicted yield is obtained by performing a multiple linear fit on the enzymatic hydrolysis condition discrimination and actual yield of each production batch and all previous production batches.
9. The method for extracting and separating ketone from Tongkat Ali as described in claim 1, characterized in that, The determination of the validation confidence factor for the enzymatic hydrolysis parameters of each production batch is specifically as follows: A decision tree algorithm was used to obtain the average purity value of all nodes for all enzymatic hydrolysis parameters of each production batch. Obtain the difference between the predicted yield and the actual yield, and calculate the ratio to the actual maximum yield. Then, positively fuse the difference between the natural number 1 and the obtained ratio with the mean purity of all nodes to obtain the validation confidence factor of the enzymatic hydrolysis parameters for each production batch.
10. The method for extracting and separating ketone from Tongkat Ali as described in claim 1, characterized in that, The dosage of cellulase for the next production batch will be updated using the following formula: In the formula, This indicates the amount of enzyme used has been updated; E represents the optimal enzyme dosage; B represents the current enzyme dosage; and B represents the confidence factor for the test. This indicates the preset learning rate.