Dehydration process for ganoderma lucidum spore oil production

Abstract

The invention relates to a dehydration process for ganoderma lucidum spore oil production, and belongs to the field of ganoderma lucidum spore oil production. The dehydration process comprises the following steps: S1, soaking picked ganoderma lucidum entities in clear water, and brushing away impurities and microbial colonies on the surfaces of ganoderma lucidum; s2, drying the cleaned ganoderma lucidum entity, and grinding the dried ganoderma lucidum entity into fine powder; and S3, drying and dehydrating the ganoderma lucidum spore powder in a drying device according to preset optimal process parameters. On the basis of multi-round experimental data, a BP neural network and a genetic algorithm are combined, a ganoderma lucidum spore powder drying process free radical scavenging rate prediction model and a process parameter optimization model are constructed, and the characteristic of rapid random search of the genetic algorithm is utilized, so that the active ingredients of the dried ganoderma lucidum spore powder are maintained at a preset value or above; optimal solution is carried out on the drying process parameters of the ganoderma lucidum spore powder, optimization of the process parameters in the drying process is achieved, and the quality of the dried ganoderma lucidum spore powder is further improved.

Description

Technical Field

[0001] The invention relates to the field of ganoderma lucidum spore oil production, in particular to a dehydration process for producing ganoderma lucidum spore oil. Background Art

[0002] Ganoderma lucidum is known as the "magic grass" in the treasure house of traditional Chinese medicine and has important medicinal research value. Ganoderma lucidum spore oil is one of the main products of the Ganoderma lucidum series. It concentrates a variety of effective active ingredients such as Ganoderma lucidum triterpenes, unsaturated fatty acids, organic germanium and Ganoderma lucidum sterols in the protoplasm of Ganoderma lucidum spores, and has a significant effect on immune regulation. Due to its significant immune regulation, liver protection and anti-tumor effects, it is widely used in the health food industry and has a high economic value. Ganoderma lucidum spore oil is easily oxidized and rancid by the effects of oxygen, light, enzymes and metal ions in the air, which greatly reduces the storage time and efficacy. In order to improve the stability of Ganoderma lucidum spore oil, the moisture content of Ganoderma lucidum spore powder is especially controlled during its production process. Appropriate drying treatment can quickly reduce moisture, maintain structural stability and extend the shelf life.

[0003] At present, the main drying method for Ganoderma lucidum spore powder is hot air drying. Hot air drying is a relatively economical drying method, but high temperature or high humidity hot air and excessive negative pressure will cause serious damage to the internal structure of the material and rapid loss of active ingredients. Therefore, in order to improve the drying quality and drying efficiency of Ganoderma lucidum spore powder, it is necessary to optimize the process parameters. At present, orthogonal experiments are often used to optimize drying process parameters, but orthogonal experiments appear to have slow convergence speed and difficulty in determining data change rules during preliminary screening. Increasing the number of experiments cannot provide clear directionality for further experiments, making the experiment still very exploratory and low in accuracy. Some researchers predict the moisture content of grains by constructing mathematical models of the drying process. Typical drying models include single-particle drying models, thin-layer drying models, and deep-bed drying models.

[0004] Although these models have good fitting effects, they ignore some key factors in the modeling process, such as the influence of process parameters on the content of active ingredients in the drying process of Ganoderma lucidum spore powder. Therefore, most of their structures are relatively simple, and their applicability and accuracy are insufficient. The change of active ingredient content in the drying process of Ganoderma lucidum spore powder is nonlinear, hysteretic, and multivariable coupled, and it is very challenging to establish an accurate drying model. At present, with the booming development of the health care products industry, the development prospects of Ganoderma lucidum spore oil are very broad. Many companies engaged in the production of Ganoderma lucidum spore oil focus on the prediction of moisture content, and do not involve the optimization of process parameters of the drying process and the prediction of active ingredients. Summary of the invention

[0005] Based on this, it is necessary to provide a dehydration process for the production of Ganoderma lucidum spore oil in view of the nonlinearity, hysteresis and multivariate coupling of the changes in the content of active ingredients in the drying process of Ganoderma lucidum spore powder, which is a great challenge to establish an accurate drying model.

[0006] The present invention is achieved through the following technical scheme: A dehydration process for producing ganoderma lucidum spore oil, characterized in that it comprises the following steps:

[0007] S1: Soak the harvested Ganoderma lucidum in clean water and brush away impurities and microbial colonies on the surface of the Ganoderma lucidum;

[0008] S2: drying the cleaned Ganoderma lucidum body by natural air-drying method, and grinding the dried Ganoderma lucidum body into fine powder;

[0009] S3: drying and dehydrating the Ganoderma lucidum spore powder in a drying device according to preset optimal process parameters, and breaking the dried Ganoderma lucidum spore powder by a high-speed wall-breaking machine;

[0010] In step S3, the preset optimal process parameters are obtained as follows:

[0011] S31: After pre-treating the Ganoderma lucidum spore powder, the powder is divided into multiple groups according to a preset mass, and the multiple groups of Ganoderma lucidum spore powder are placed in a drying test bench in turn to perform multiple rounds of independent drying experiments; the free radical scavenging rate of each group of samples is obtained at preset intervals and the drying process parameters of each group of samples are recorded, and the experiment ends when the moisture content of each group of samples reaches a uniform preset value;

[0012] S32: Constructing a free radical scavenging rate prediction model for the drying process of Ganoderma lucidum spore powder based on BP neural network;

[0013] S33: Determine the value range of the process parameters for drying the Ganoderma lucidum spore powder, construct a process parameter optimization model based on a genetic algorithm, and solve the input initial population of process parameters through the optimization model to obtain the optimal solution of the process parameters for drying the Ganoderma lucidum spore powder.

[0014] The invention is based on multiple rounds of experimental data, combines BP neural network and genetic algorithm, and optimizes the process parameters in the drying process by constructing a free radical scavenging rate prediction model and a process parameter optimization model for the drying process of ganoderma lucidum spore powder, thereby further improving the quality of the dried ganoderma lucidum spore powder.

[0015] In one embodiment, in step S31, the moisture content of the Ganoderma lucidum spore powder is G t The calculation formula is as follows:

[0016]

[0017] In the formula, g trepresents the mass of Ganoderma lucidum spore powder at any drying time t, and g0 represents the mass of dry matter of Ganoderma lucidum spore powder;

[0018] The calculation formula of the free radical scavenging rate ε of Ganoderma lucidum spore powder at any drying time t is as follows:

[0019]

[0020] Wherein, A1 represents the absorbance measured after the hot water extraction solution of the sample is fully mixed with the DPPH solution, A2 represents the absorbance measured after the hot water extraction solution of the sample is fully mixed with anhydrous ethanol, and A0 represents the absorbance measured after the distilled water and the DPPH solution are fully mixed.

[0021] The present invention adopts the free radical scavenging rate of the ganoderma spore powder as an indicator to measure the quality of the active substances in the ganoderma spore powder. A high free radical scavenging rate represents a stronger antioxidant capacity and a stronger scavenging capacity.

[0022] In one embodiment, in step S32, the steps of constructing a prediction model are as follows:

[0023] S321: Initialize BP neural network parameters and determine the number of neurons in each layer;

[0024] S322: Use principal component analysis to reduce the dimension of variables and calculate the cumulative contribution rate γ of each process parameter n Determine the main components and calculate the main components C by determining the coefficient matrix C of the main components n expression.

[0025] S323: Setting parameters such as the number of BP neural network trainings and the learning rate, and inputting the data filtered by the principal component into the training to obtain a prediction model to output a prediction result.

[0026] Based on the powerful nonlinear mapping capability of BP neural network, the present invention can better process the local relationship between multiple input data and extract the features in the data, and is suitable for solving the free radical scavenging rate with complex internal mechanisms. In addition, BP neural network has a high degree of self-learning and self-adaptive ability, and the trained model improves the accuracy of the prediction results, while improving the efficiency and stability of the prediction process.

[0027] In one embodiment, in step S321, the calculation formula for the number of neurons in the hidden layer is as follows:

[0028]

[0029] Where i represents the number of neurons in the input layer, j represents the number of neurons in the output layer, and s represents a random number between [0,10].

[0030] In one embodiment, in step S322, the cumulative contribution rate γ n The calculation formula is as follows:

[0031]

[0032] In the formula, λ n represents the nth largest eigenvalue, and δ represents a random number between (0,1).

[0033] The invention adopts the principal component analysis method to reduce the dimension of the input ganoderma lucidum spore powder drying process parameters, so as to improve the running speed and prediction accuracy of the BP neural network.

[0034] In one embodiment, in step S33, the steps of constructing a process parameter optimization model are as follows:

[0035] S331: Initializing genetic algorithm parameters and algebraic counters, and randomly generating an initial population of Ganoderma lucidum spore powder drying process parameters;

[0036] S332: The algebraic calculator starts counting, inputs the initial population of process parameters into the free radical scavenging rate prediction model to calculate the fitness value of the free radical scavenging rate of each population; uses the tournament selection method to select the individual with the best fitness to join the parent population, and encodes the process parameters into a chromosome in a binary encoding manner;

[0037] S333: performing crossover and mutation operations on the encoded process parameter population to generate new individuals; decoding the binary string generated by the encoding into decimal process parameter data, and merging the parent individual and the child individual to generate a new population;

[0038] S334: Input the new population of process parameters into the free radical scavenging rate prediction model to calculate the free radical scavenging rate values ​​of various populations. If the output free radical scavenging rate value meets the requirements, the optimal solution is output; if the output free radical scavenging rate value does not meet the requirements, the new population of process parameters is returned to S332 for loop processing until the model training results meet the requirements or the model training reaches the maximum number of iterations.

[0039] The invention utilizes the fast random search characteristics of the genetic algorithm to optimally solve the drying process parameters of the Ganoderma lucidum spore powder while ensuring that the active components of the Ganoderma lucidum spore powder after drying are maintained above a preset value.

[0040] In one embodiment, in step S331, parameters such as the number of algorithm iterations and the mutation rate need to be determined according to the actual situation of the silk making process, and the initial population size of the process parameters is set according to the value range of the process parameters. The specific calculation formula is as follows:

[0041] R n =a min +(amax -a min )×Z

[0042] In the formula, R n represents the size of the nth process parameter population, a min Indicates the minimum value of the nth process parameter, a max represents the maximum value of the nth process parameter, and Z represents a random number between (0,1).

[0043] In one embodiment, in step S332, the process parameter encoding step is as follows:

[0044] S3321: Determine the process parameter code length, encode the first row and the first column of the process parameter population, and convert the decimal process parameter code into a binary string using a binary encoding method;

[0045] S3322: The insufficient binary string is made to reach the encoding length by adding “0” at the front end, and the binary string of the process parameter is written into the corresponding position of the binary population;

[0046] S3323: Repeat the above operation in a loop to perform binary encoding on each row and column of the process parameter population.

[0047] In one embodiment, in step S333, crossover and mutation operations are performed on the encoded process parameter population, and the specific steps are as follows:

[0048] S3331: Determine the crossover rate K cro and mutation rate K var , randomly select the parent and mother generations for crossover operation and the crossover operation position in the first row of the encoded process parameter population;

[0049] S3332: Determine whether to perform a crossover operation; if so, generate a new individual of the process parameter offspring population by replacing the binary code of the parent generation with the parent generation; if not, directly use the process parameter parent generation individual as the process parameter offspring population individual;

[0050] S3333: Repeat the above operation for the individuals in the next column of the row until all individuals in the row complete the crossover and mutation operations;

[0051] S3334: Perform the above operations on all rows of the population until all rows and columns of the population have completed the crossover and mutation operations.

[0052] In one embodiment, in step S33, the experimental data of each group of sample drying experiments are collected as the original data set and standardized; one round of experimental data is randomly selected as the test set, and the experimental data of other rounds are used as the training set; a K-fold cross-validation strategy is adopted, and a part of the training set is taken as the validation set for each training, and the rest is taken as the training set for the training; the relative error RE is used to quantitatively evaluate the prediction performance of the trained model, and the specific calculation formula of the relative error RE is as follows:

[0053]

[0054] In the formula, It indicates the predicted value of free radical scavenging rate of Ganoderma lucidum spore powder. represents the actual value of the free radical scavenging rate of Ganoderma lucidum spore powder, n represents the number of test data groups, and i represents a random number between (1, n).

[0055] The present invention adopts a K-fold cross-validation strategy when inputting experimental data into the process parameter optimization model for training. Through multiple cross-validations of the training set, the possibility of overfitting can be reduced, and the situation where the model is overfitted on the training set can be avoided. The generalization ability of the model is effectively improved, and the optimal parameter combination is helped to select, so that the model achieves the best performance on the validation set.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The present invention is based on multiple rounds of experimental data, combines BP neural network and genetic algorithm, constructs a free radical scavenging rate prediction model and a process parameter optimization model for the drying process of Ganoderma lucidum spore powder, utilizes the fast random search characteristics of the genetic algorithm, ensures that the active ingredients of the Ganoderma lucidum spore powder are maintained above the preset value after drying, optimizes the process parameters in the drying process, and further improves the quality of the dried Ganoderma lucidum spore powder.

[0058] 2. The present invention adopts the free radical scavenging rate of Ganoderma lucidum spore powder as an indicator to measure the quality of active substances in Ganoderma lucidum spore powder. A high free radical scavenging rate represents a stronger antioxidant capacity and a stronger scavenging ability. Based on the powerful nonlinear mapping ability of the BP neural network, the present invention adopts the principal component analysis method to reduce the dimension of the input Ganoderma lucidum spore powder drying process parameters, thereby improving the speed and prediction accuracy of the BP neural network operation. It can better handle the local relationship between multiple input data and extract features from the data, and is suitable for predicting and solving free radical scavenging rates with complex internal mechanisms. In addition, the BP neural network has a high degree of self-learning and self-adaptive capabilities. The trained model improves the accuracy of the prediction results, while improving the efficiency and stability of the prediction process.

[0059] 3. The present invention adopts a K-fold cross-validation strategy when inputting experimental data into the process parameter optimization model for training. Through multiple cross-validations of the training set, the possibility of overfitting can be reduced, and the situation where the model is overfitted on the training set can be avoided. The generalization ability of the model is effectively improved, and the optimal parameter combination is helped to select, so that the model can achieve the best performance on the validation set. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a step diagram of a dehydration process for producing Ganoderma lucidum spore oil according to Example 1 of the present invention;

[0061] Figure 2 for Figure 1 A step diagram for obtaining the preset optimal process parameters;

[0062] Figure 3 for Figure 1 Diagram of the steps for constructing a free radical scavenging rate prediction model;

[0063] Figure 4 for Figure 1 Diagram of the steps for constructing a process parameter optimization model. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] It should be noted that when a component is referred to as being "mounted on" another component, it may be directly on the other component or there may be a central component. When a component is considered to be "set on" another component, it may be directly set on the other component or there may be a central component at the same time. When a component is considered to be "fixed to" another component, it may be directly fixed on the other component or there may be a central component at the same time.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "or / and" used herein includes any and all combinations of one or more of the related listed items.

[0067] Example 1: Please refer to Figure 1-Figure 4 This embodiment provides a dehydration process for producing Ganoderma lucidum spore oil, and the dehydration process steps are as follows:

[0068] S1: Soak the harvested Ganoderma lucidum in clean water and brush away impurities and microbial colonies on the surface of the Ganoderma lucidum.

[0069] S2: Dry the cleaned Ganoderma lucidum body by natural airing to avoid moisture absorption during grinding, which will affect the quality of spore powder. Grind the dried Ganoderma lucidum body into fine powder by high-speed centrifugal grinder. Put the dried Ganoderma lucidum fruiting body into a freezer for 30 minutes before grinding to improve the grinding efficiency of Ganoderma lucidum body.

[0070] S3: drying and dehydrating the Ganoderma lucidum spore powder in a drying device according to preset optimal process parameters, and breaking the dried Ganoderma lucidum spore powder by a high-speed wall breaking machine.

[0071] The steps for obtaining the preset optimal process parameters are as follows:

[0072] S31: After the Ganoderma lucidum spore powder is pre-treated, it is divided into multiple groups according to the preset mass. In this embodiment, the pre-treatment includes adjusting the moisture content of the Ganoderma lucidum spore powder to about 26% by artificial humidity control, sealing the Ganoderma lucidum spore powder after humidity control and storing it at 2-4°C for 3 days, and taking out the sealed Ganoderma lucidum spore powder before the test and placing it at ambient temperature for 3 hours before performing a drying test. Multiple groups of Ganoderma lucidum spore powder are placed in a drying test bench in turn to perform multiple rounds of independent drying experiments. The drying test bench in this embodiment includes: an electric drying oven, an electronic balance, a pipeline wind speed transmitter, a temperature transmitter, and a temperature and humidity transmitter.

[0073] The drying temperature and humidity were different for each round of the experiment. During the experiment, a small amount of Ganoderma lucidum spore powder was taken out from the electric drying oven every 3 minutes. The moisture content of the sample was determined by the 105°C standard drying oven method and the free radical scavenging rate of the sample was determined by the phenol-sulfuric acid method. The average value was taken after each sample was measured 3 times. The sample measurement data and the drying process parameters at that moment were recorded. The drying process parameters included: hot air temperature T1, hot air humidity H1, hot air speed V1, exhaust temperature T2, exhaust humidity H2, exhaust speed V3, negative pressure P, sample temperature T3 and sample moisture content G t The test ends when the moisture content of each group of samples reaches a uniform preset value.

[0074] Moisture content of Ganoderma lucidum spore powder at any drying time tG t The calculation formula is as follows:

[0075]

[0076] In the formula, g t represents the mass of Ganoderma lucidum spore powder at any drying time t, and g0 represents the mass of dry matter of Ganoderma lucidum spore powder;

[0077] The calculation formula of the free radical scavenging rate ε of Ganoderma lucidum spore powder at any drying time t is as follows:

[0078]

[0079] Wherein, A1 represents the absorbance measured after the hot water extraction solution of the sample is fully mixed with the DPPH solution, A2 represents the absorbance measured after the hot water extraction solution of the sample is fully mixed with anhydrous ethanol, and A0 represents the absorbance measured after the distilled water and the DPPH solution are fully mixed.

[0080] S32: A free radical scavenging rate prediction model for the drying process of Ganoderma lucidum spore powder is constructed based on BP neural network. The free radical scavenging rate prediction model for the drying process of Ganoderma lucidum spore powder includes an input layer, a hidden layer and an output layer. The steps of constructing the prediction model are as follows:

[0081] S321: Initialize BP neural network parameters and determine the number of neurons in each layer; the number of neurons in the input layer is determined by the number of input variables, the number of neurons in the output layer is determined by the output results, and the calculation formula for the number of neurons in the hidden layer is as follows:

[0082]

[0083] Where i represents the number of neurons in the input layer, j represents the number of neurons in the output layer, and s represents a random number between [0,10].

[0084] S322: Use the principal component analysis method to reduce the dimension of the variables, and use the process parameters with cumulative contribution rates exceeding 0.5 as the principal components to reduce the number of input parameters of the BP neural network. The calculation formula of the principal component analysis method is as follows:

[0085]

[0086] In the formula, C n represents the principal component, a nw represents the coefficient, w represents the number of initial variables, and n represents the number of principal components.

[0087] Calculate the cumulative contribution rate γ of each process parameter n , if γ n When the requirements are met, the coefficient matrix C of the principal component is determined to be [a1, a2, …, a m ] T , calculate the principal component C n The cumulative contribution rate γ n The calculation formula is as follows:

[0088]

[0089] In the formula, λ n represents the nth largest eigenvalue.

[0090] Since there are many process parameters, if all of them are input into the BP neural network, the training speed and prediction accuracy of the BP model will be reduced. By using the principal component analysis method to reduce the dimension of the drying process parameters of the tofu drying machine, the speed of the BP model operation can be improved.

[0091] S323: Set the BP neural network training times to 1000, the training target to 0.0001, input the data filtered by the principal component into the training to obtain the prediction model, and the prediction model expression is as follows:

[0092] U=U net (C1,C2,…,C n )

[0093] This embodiment uses the free radical scavenging rate of Ganoderma lucidum spore powder as an indicator to measure the quality of active substances in Ganoderma lucidum spore powder. A high free radical scavenging rate represents a stronger antioxidant capacity and a stronger scavenging ability. Based on the powerful nonlinear mapping ability of the BP neural network, the principal component analysis method is used to reduce the dimension of the input Ganoderma lucidum spore powder drying process parameters to improve the speed and prediction accuracy of the BP neural network operation. It can better handle the local relationship between multiple input data and extract features from the data, and is suitable for predicting and solving free radical scavenging rates with complex internal mechanisms. In addition, the BP neural network has a high degree of self-learning and self-adaptive capabilities. The trained model improves the accuracy of the prediction results, while improving the efficiency and stability of the prediction process.

[0094] S33: Determine the value range of the process parameters for drying the spore powder of Ganoderma lucidum. This embodiment refers to the boundary values ​​of the process parameters for drying the spore powder of Ganoderma lucidum of a certain enterprise: 20.31°C≤T1≤20.51°C, 49.98%≤H1≤51.84%, 1.36m*s -1 ≤V1≤1.44m*s -1 , 19.78℃≤T2≤20.01℃, 61.01≤H2≤62.54, 1.74m*s -1 ≤V3≤1.77m*s -1 , -33.38kPa≤P≤-27.74kPa, 24.94℃≤T3≤25.21℃, 13.59%≤G t ≤13.94%.

[0095] The process parameter optimization model is constructed by genetic algorithm to achieve the optimal solution of the process parameters for drying the Ganoderma lucidum spore powder. The steps of constructing the process parameter optimization model are as follows:

[0096] S331: Initialize the genetic algorithm parameters and algebraic counter, and randomly generate the initial population of the process parameters for drying the Ganoderma lucidum spore powder; the size of the initial population of the process parameters is set according to the value range of the process parameters, and the specific calculation formula is as follows:

[0097] R n =a min +(a max -a min )×Z

[0098] In the formula, R n represents the size of the nth process parameter population, a min Indicates the minimum value of the nth process parameter, a max represents the maximum value of the nth process parameter, and Z represents a random number between (0,1).

[0099] S332: The algebraic calculator starts counting, inputs the initial population of process parameters into the free radical scavenging rate prediction model to calculate the fitness values ​​of the free radical scavenging rates of various populations; uses the tournament selection method to select the individual with the best fitness to join the parent population, and encodes the process parameters into a chromosome in a binary encoding manner. The process parameter encoding steps are as follows:

[0100] S3321: Determine the process parameter code length, encode the first row and the first column of the process parameter population, and convert the decimal process parameter code into a binary string using a binary encoding method;

[0101] S3322: The insufficient binary string is made to reach the encoding length by adding “0” at the front end, and the binary string of the process parameter is written into the corresponding position of the binary population;

[0102] S3323: Repeat the above operation in a loop to perform binary encoding on each row and column of the process parameter population.

[0103] S333: Perform crossover and mutation operations on the encoded process parameter population to generate new individuals; decode the binary string generated by the encoding into decimal process parameter data, and merge the parent individual and the child individual to generate a new population. The specific steps are as follows:

[0104] S3331: Determine the crossover rate K cro and mutation rate K var , randomly select the parent and mother generations for crossover operation and the crossover operation position in the first row of the encoded process parameter population;

[0105] S3332: Determine whether to perform a crossover operation; if so, generate a new individual of the process parameter offspring population by replacing the binary code of the parent generation with the parent generation; if not, directly use the process parameter parent generation individual as the process parameter offspring population individual;

[0106] S3333: Repeat the above operation for the individuals in the next column of the row until all individuals in the row complete the crossover and mutation operations;

[0107] S3334: Perform the above operations on all rows of the population until all rows and columns of the population have completed the crossover and mutation operations.

[0108] S334: Input the new population of process parameters into the free radical scavenging rate prediction model to calculate the free radical scavenging rate values ​​of various populations. If the output free radical scavenging rate value meets the requirements, the optimal solution is output; if the output free radical scavenging rate value does not meet the requirements, the new population of process parameters is returned to S332 for loop processing until the model training results meet the requirements or the model training reaches the maximum number of iterations.

[0109] The experimental data of each group of sample drying experiments were collected as the original data set and standardized; one round of experimental data was randomly selected as the test set, and the experimental data of other rounds were used as the training set; the K-fold cross-validation strategy was adopted, and a part of the training set was taken as the validation set for each training, and the rest was used as the training set for that training; the Z-Score method was used to standardize the experimental data of all parameter items, and the standardization calculation formula for any parameter was as follows:

[0110]

[0111] Where y is the standard value of the parameter, x is the experimental value of the parameter, μ is the mean of the experimental values ​​of the parameter, and σ is the standard deviation of the experimental values ​​of the parameter.

[0112] The relative error is used to quantitatively evaluate the prediction performance of the model. The specific calculation formula of the relative error RE is as follows:

[0113]

[0114] In the formula, It indicates the predicted value of free radical scavenging rate of Ganoderma lucidum spore powder. represents the actual value of the free radical scavenging rate of Ganoderma lucidum spore powder, n represents the number of test data groups, and i represents a random number between (1, n).

[0115] The test set was used to verify the process parameter optimization model. When the relative error of the predicted value of free radical scavenging rate of Ganoderma lucidum spore powder was less than 2×10 -6 When the trained model meets the design requirements. This embodiment adopts the K-fold cross-validation strategy. Through multiple cross-validations of the training set, the possibility of overfitting can be reduced and the situation where the model is overfitted on the training set can be avoided. The generalization ability of the model is effectively improved, and the optimal parameter combination is helped to select so that the model can achieve the best performance on the validation set.

[0116] This embodiment combines BP neural network and genetic algorithm, by constructing a free radical scavenging rate prediction model and a process parameter optimization model for the Ganoderma lucidum spore powder drying process, and utilizing the fast random search characteristics of the genetic algorithm, while ensuring that the active ingredients of the Ganoderma lucidum spore powder after drying are maintained above the preset value, the optimal solution for the Ganoderma lucidum spore powder drying process parameters is obtained; the process parameters in the drying process are optimized, and the quality of the Ganoderma lucidum spore powder is further improved.

[0117] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0118] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A dehydration process for producing Ganoderma lucidum spore oil, characterized in that: The following steps are involved: S1: Soak the harvested Ganoderma lucidum in clean water and brush away impurities and microbial colonies on the surface of the Ganoderma lucidum; S2: drying the cleaned Ganoderma lucidum body by natural air-drying method, and grinding the dried Ganoderma lucidum body into fine powder; S3: drying and dehydrating the Ganoderma lucidum spore powder in a drying device according to preset optimal process parameters, and breaking the dried Ganoderma lucidum spore powder by a high-speed wall-breaking machine; In step S3, the steps of obtaining the preset optimal process parameters are as follows: S31: After pre-treating the Ganoderma lucidum spore powder, divide it into multiple groups according to preset weights, and put the multiple groups of Ganoderma lucidum spore powder into a drying test bench in turn to perform multiple rounds of independent drying experiments; obtain the free radical scavenging rate of each group of samples at preset intervals and record the drying process parameters of each group of samples, and end the experiment when the moisture content of each group of samples reaches a uniform preset value; S32: Constructing a free radical scavenging rate prediction model for the drying process of Ganoderma lucidum spore powder based on BP neural network; S33: Determine the value range of the process parameters for drying the ganoderma lucidum spore powder, construct a process parameter optimization model based on a genetic algorithm, and solve the input initial population of process parameters through the optimization model to obtain the optimal solution of the process parameters for drying the ganoderma lucidum spore powder.

2. The dehydration process for producing Ganoderma lucidum spore oil according to claim 1, characterized in that: In step S31, the moisture content of the Ganoderma lucidum spore powder is G t The calculation formula is as follows: In the formula, g t represents the mass of Ganoderma lucidum spore powder at any drying time t, and g0 represents the mass of dry matter of Ganoderma lucidum spore powder; The calculation formula of the free radical scavenging rate ε of Ganoderma lucidum spore powder at any drying time t is as follows: Wherein, A1 represents the absorbance measured after the hot water extraction solution of the sample is fully mixed with the DPPH solution, A2 represents the absorbance measured after the hot water extraction solution of the sample is fully mixed with anhydrous ethanol, and A0 represents the absorbance measured after the distilled water and the DPPH solution are fully mixed.

3. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 1, characterized in that: In step S32, the steps of constructing the prediction model are as follows: S321: Initialize BP neural network parameters and determine the number of neurons in each layer; S322: Use principal component analysis to reduce the dimension of variables and calculate the cumulative contribution rate γ of each process parameter n Determine the main components, calculate the main components C by determining the coefficient matrix C of the main components n expression. S323: Setting parameters such as the number of BP neural network trainings and the learning rate, and inputting the data screened by the principal component into the training to obtain the prediction model output prediction results.

4. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 3, characterized in that: In step S321, the calculation formula for the number of neurons in the hidden layer is as follows: Where i represents the number of neurons in the input layer, j represents the number of neurons in the output layer, and s represents a random number between [0,10].

5. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 3, characterized in that: In step S322, the cumulative contribution rate γ n The calculation formula is as follows: In the formula, λ n represents the nth largest eigenvalue, and δ represents a random number between (0,1).

6. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 1, characterized in that: In step S33, the steps of constructing the process parameter optimization model are as follows: S331: Initializing genetic algorithm parameters and algebraic counters, and randomly generating an initial population of Ganoderma lucidum spore powder drying process parameters; S332: The algebraic calculator starts counting, inputs the initial population of process parameters into the free radical scavenging rate prediction model to calculate the fitness value of the free radical scavenging rate of each population; uses the tournament selection method to select the individual with the best fitness to join the parent population, and encodes the process parameters into a chromosome in a binary encoding manner; S333: performing crossover and mutation operations on the encoded process parameter population to generate new individuals; decoding the binary string generated by the encoding into decimal process parameter data, and merging the parent individual and the child individual to generate a new population; S334: Input the new population of process parameters into the free radical scavenging rate prediction model to calculate the free radical scavenging rate values ​​of various populations. If the output free radical scavenging rate value meets the requirements, the optimal solution is output; if the output free radical scavenging rate value does not meet the requirements, the new population of process parameters is returned to S332 for loop processing until the model training results meet the requirements or the model training reaches the maximum number of iterations.

7. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 6, characterized in that: In step S331, parameters such as the number of algorithm iterations and the mutation rate need to be determined according to the actual situation of the silk-making process. The initial population size of the process parameters is set according to the value range of the process parameters. The specific calculation formula is as follows: R n =a min +(a max -a min )×Z In the formula, R n represents the size of the nth process parameter population, a min Indicates the minimum value of the nth process parameter, a max represents the maximum value of the nth process parameter, and Z represents a random number between (0,1).

8. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 6, characterized in that: In step S332, the process parameter encoding steps are as follows: S3321: Determine the process parameter code length, encode the first row and the first column of the process parameter population, and convert the decimal process parameter code into a binary string using a binary encoding method; S3322: The insufficient binary string is filled with "0" at the front end to reach the encoding length, and the binary string of the process parameter is written into the corresponding position of the binary population; S3323: Repeat the above operation in a loop to perform binary encoding on each row and column of the process parameter population.

9. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 6, characterized in that: In step S333, crossover and mutation operations are performed on the encoded process parameter population. The specific steps are as follows: S3331: Determine the crossover rate K cro and mutation rate K var , randomly select the parent and mother generations for crossover operation and the crossover operation position in the first row of the encoded process parameter population; S3332: Determine whether to perform a crossover operation; If it is done, the new individuals of the process parameter offspring population are generated by replacing the binary code of the parent generation with the parent generation; if it is not done, the process parameter parent generation individuals are directly used as the process parameter offspring population individuals; S3333: Repeat the above operation for the individuals in the next column of the row until all individuals in the row complete the crossover and mutation operations; S3334: Perform the above operations on all rows of the population until all rows and columns of the population have completed the crossover and mutation operations.

10. The dehydration process for producing Ganoderma lucidum spore powder oil according to claim 6, characterized in that: In step S334, the experimental data of each group of sample drying experiments are collected as the original data set and standardized; one round of experimental data is randomly selected as the test set, and the experimental data of other rounds are used as the training set; a K-fold cross-validation strategy is adopted, and a part of the training set is taken as the validation set for each training, and the rest is taken as the training set for this training; the relative error RE is used to quantitatively evaluate the prediction performance of the trained model, and the specific calculation formula of the relative error RE is as follows: In the formula, It indicates the predicted value of free radical scavenging rate of Ganoderma lucidum spore powder. represents the actual value of the free radical scavenging rate of Ganoderma lucidum spore powder, n represents the number of test data groups, and i represents a random number between (1, n).

Images