Daqu fermentation metabolic compound prediction method and device and electronic equipment

Abstract

The invention relates to the technical field of wine making, discloses a method and device for predicting metabolic compounds of yeast fermentation and electronic equipment, and aims to solve the problems of low efficiency, high cost and poor accuracy of an existing method for determining metabolic compounds of yeast fermentation. Establishing a yeast fermentation data set according to fermentation data, wherein the fermentation data comprises environmental data, fermentation stage data and metabolic compound data; training a time sequence neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model; and according to the metabolic compound data of the to-be-predicted yeast at the historical time point and the corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model, obtaining the metabolic compound content of the to-be-predicted yeast at the future time point. The method improves the determination efficiency and accuracy of the metabolic compound, reduces the cost, and is suitable for monitoring and optimization of the yeast fermentation process.

Description

Technical Field

[0001] The present invention relates to the technical field of winemaking, and in particular to a method and device for predicting metabolic compounds in Daqu fermentation, and electronic equipment. Background Art

[0002] Daqu is a type of koji that has been fermented for a long time. It is usually made from wheat, yellow rice, peas and other raw materials through a series of fermentation, koji culture and cooking processes. Daqu contains a variety of microorganisms, including molds, yeasts and bacteria. Daqu is one of the main fermentation agents used in the production of traditional Chinese liquor, and is widely used in the brewing process of Luzhou-flavor and Maotai-flavor liquors. Daqu fermentation refers to the process of brewing liquor using Daqu as a saccharification fermentation agent.

[0003] During the fermentation process of Daqu, the content of metabolic compounds changes over time, and the change in the content of metabolic compounds directly affects the quality, flavor and flavor characteristics of Daqu and the final liquor. Therefore, accurately predicting the changes in metabolic compounds during Daqu fermentation is of great significance for improving the quality of liquor production and optimizing the fermentation process. The traditional method of monitoring the content of metabolic compounds in the fermentation process mainly relies on manual observation and laboratory testing. The manual observation method relies on experience, and the consistency and accuracy are poor; while the laboratory testing method usually requires multiple sampling and chemical analysis, which is cumbersome and time-consuming, with low detection efficiency, high labor cost and time cost, and cannot achieve advance prediction of the content of metabolic compounds. In addition, the environmental factors of Daqu fermentation are closely related to the effect of Daqu fermentation. However, most methods in the prior art lack effective modeling for the temporal changes of the content of metabolic compounds (such as esters, alcohols, acids, ketones, etc.) during the fermentation process, and lack comprehensive consideration of the impact of environmental factors on metabolic compounds during Daqu fermentation, resulting in poor prediction accuracy. Summary of the invention

[0004] The present invention aims to solve the problems of low efficiency, high cost and poor accuracy in the existing method for determining metabolic compounds of Daqu fermentation, and proposes a method and device for predicting metabolic compounds of Daqu fermentation, and an electronic device.

[0005] The technical solution adopted by the present invention to solve the above technical problems is:

[0006] In a first aspect, the present invention provides a method for predicting metabolic compounds in Daqu fermentation, the method comprising:

[0007] Acquire fermentation data at each time point during the Daqu fermentation process, and establish a Daqu fermentation data set according to the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data, and metabolic compound data;

[0008] Training a time series neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model;

[0009] According to the metabolic compound data of the Daqu to be predicted at historical time points and its corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model, the metabolic compound content of the Daqu to be predicted at future time points is obtained.

[0010] Furthermore, the environmental data includes temperature data, humidity data and air oxygen content data; the fermentation stage data is the time node in the Daqu fermentation cycle, and the unit of the time node is day or hour.

[0011] Furthermore, the metabolic compound data includes one or more compound content data of esters, alcohols, aldehydes, acids, ketones, pyrazines, furans and aromatics, the esters include at least ethyl hexadecanoate and ethyl hexanoate, the alcohols include at least phenylethanol and 2,3-butanediol, the acids include at least acetic acid and hexanoic acid, and the pyrazines include at least 2,5-dimethylpyrazine, trimethylpyrazine and 2,6-dimethylpyrazine.

[0012] Furthermore, fermentation data at each time point during the Daqu fermentation process is obtained by sampling according to a preset sampling frequency, and the preset sampling frequency is half a day, one day, two days or three days.

[0013] Furthermore, the method further comprises:

[0014] The environmental data, fermentation stage data and metabolic compound data in the Daqu fermentation data set were standardized, and the metabolic compound data of the predicted Daqu at historical time points and their corresponding environmental data and fermentation stage data were standardized using the same processing method.

[0015] Further, training a time series neural network model based on the Daqu fermentation data set includes:

[0016] The metabolic compound data at historical time points and their corresponding environmental data and fermentation stage data are used as input features, and the metabolic compound data at corresponding future time points are used as output labels of the time series neural network model to construct training sets and validation sets;

[0017] Based on the training set and using a supervised learning method to train a time series neural network model, compare the output features of the time series neural network model with the corresponding output labels, calculate the corresponding loss function, and optimize the model through back propagation;

[0018] The time series neural network model is verified based on the validation set and the cross-validation test set. When the loss function of the time series neural network model is minimized and the prediction error of the validation set or the cross-validation test set is less than the error threshold, the corresponding time series neural network model is used as the metabolic compound prediction model.

[0019] Furthermore, the method further comprises:

[0020] The fermentation data at new time points during the Daqu fermentation process are continuously acquired, and the metabolic compound prediction model is regularly updated based on the fermentation data at new time points.

[0021] Furthermore, the temporal neural network model is a recurrent neural network model, a long short-term memory network model or a gated recurrent unit model for learning the temporal variation law of the content of metabolic compounds.

[0022] In a second aspect, the present invention provides a device for predicting metabolic compounds of Daqu fermentation, the device comprising:

[0023] An acquisition unit, used to acquire fermentation data at each time point during the fermentation process of Daqu, and establish a Daqu fermentation data set according to the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data and metabolic compound data;

[0024] A training unit, used for training a time series neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model;

[0025] The prediction unit is used to obtain the metabolic compound content of the Daqu to be predicted at a future time point based on the metabolic compound data of the Daqu to be predicted at a historical time point and its corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model.

[0026] In a third aspect, the present invention provides an electronic device, the electronic device comprising a processor, a memory and a communication bus;

[0027] The communication bus is used to realize the connection and communication between the processor and the memory;

[0028] The processor is used to execute one or more programs in the memory to implement the steps of the method for predicting metabolic compounds in Daqu fermentation as described in the first aspect.

[0029] The beneficial effects of the present invention are as follows: the metabolic compound prediction method, device, and electronic device of Daqu fermentation provided by the present invention can realize the prediction of the metabolic compound content at future time points during the Daqu fermentation process by constructing a metabolic compound prediction model based on the metabolic compound data at historical time points and their corresponding environmental data and fermentation stage data, thereby realizing the advance prediction of the metabolic compound content. Compared with the manual observation method, the present invention improves the consistency and accuracy, and compared with the laboratory detection method, improves the efficiency of determining the metabolic compound content and reduces the cost. In addition, the present invention improves the accuracy of the prediction of the metabolic compound content by comprehensively considering the impact of environmental factors on the metabolic compounds in the Daqu fermentation process. By predicting the metabolic compound content of Daqu fermentation, it is possible to realize dynamic real-time monitoring of the metabolic compound content during the Daqu fermentation cycle, providing a scientific basis for the optimization of the Daqu fermentation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of a process for predicting metabolic compounds in Daqu fermentation provided in an embodiment;

[0031] Figure 2 A schematic diagram for comparing output features and output labels for predicting metabolic compounds on the first day in the future provided in the embodiment;

[0032] Figure 3 A schematic diagram for comparing output features and output labels for predicting metabolic compounds on the second day in the future provided in the embodiment;

[0033] Figure 4 A schematic diagram of the structure of a device for predicting metabolic compounds in Daqu fermentation provided in an embodiment;

[0034] Figure 5 A schematic structural diagram of an electronic device provided in an embodiment. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solution in this embodiment will be clearly and completely described below in conjunction with the drawings in this embodiment.

[0036] In some of the processes described in the specification of the present invention and the above-mentioned figures, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The sequence numbers of the operations are only used to distinguish different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0037] At present, the monitoring scheme for determining the content of metabolic compounds in the fermentation process of Daqu is usually manual observation and laboratory analysis. Manual observation means that the staff judges or predicts the content of metabolic compounds based on experience, and the laboratory detection method is to sample Daqu and detect the content of metabolic compounds therein. The inventors have found through research that the manual observation method relies on experience, and the consistency and accuracy are poor; while the laboratory detection method usually requires multiple sampling and chemical analysis, which is cumbersome and time-consuming, with low detection efficiency, high labor and time costs, and cannot achieve advance prediction of the content of metabolic compounds.

[0038] Based on this, the technical scheme of the present invention is proposed. In the present invention, fermentation data at each time point in the Daqu fermentation process are obtained, and a Daqu fermentation data set is established according to the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data and metabolic compound data; a time series neural network model is trained based on the Daqu fermentation data set to obtain a metabolic compound prediction model; according to the metabolic compound data of the Daqu to be predicted at historical time points and their corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model, the metabolic compound content of the Daqu to be predicted at a future time point is obtained.

[0039] Specifically, the present invention first constructs a Daqu fermentation data set including environmental data, fermentation stage data and metabolic compound data at each time point during the Daqu fermentation process, and then trains a metabolic compound prediction model based on the Daqu fermentation data set to characterize the temporal variation law of the metabolic compound content, and finally inputs the metabolic compound data of the Daqu to be predicted at a historical time point and its corresponding environmental data and fermentation stage data into the metabolic compound prediction model, and the metabolic compound content of the Daqu to be predicted at a future time point can be obtained. The present invention does not rely on personnel experience, improves the consistency and accuracy of determining the content of metabolic compounds, and the present invention realizes the advance prediction of the content of metabolic compounds, and also improves the efficiency of determining the content of metabolic compounds, reducing costs. And the present invention improves the accuracy of predicting the content of metabolic compounds by comprehensively considering the impact of environmental factors on metabolic compounds in the Daqu fermentation process.

[0040] The technical solution in this embodiment will be clearly and completely described below in conjunction with the drawings in this embodiment. Obviously, the described embodiment is only a part of the embodiments of the present invention, rather than all the embodiments.

[0041] Figure 1 A schematic diagram showing a method for predicting metabolic compounds in Daqu fermentation is shown in FIG. Figure 1 , the method comprises the following steps:

[0042] Step 1: Obtain fermentation data at each time point during the Daqu fermentation process, and establish a Daqu fermentation data set based on the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data, and metabolic compound data.

[0043] In this embodiment, the environmental data includes temperature, humidity data and air oxygen content data. After obtaining the environmental data, the minimum and maximum normalization method can be used for standardization, and the maximum and minimum values ​​are obtained based on all the data in the data set. For example, the temperature and humidity data at a certain time point are [35,32]. Assuming that the maximum value of the temperature data is 60°C and the minimum value is 30°C, the maximum value of the humidity is 36 and the minimum value is 9, the temperature and humidity data [35,32] are normalized to [0.14,0.85].

[0044] In this embodiment, the fermentation stage data is a time node in the Daqu fermentation cycle, and the unit of the time node is day or hour. That is, the fermentation stage data is used to identify the specific moment in the current Daqu fermentation cycle, which can be expressed in days or hours. For example, if the fermentation time is the 1st day, it is expressed as 1, and if the fermentation time is the 16th day, it is expressed as 16. In this embodiment, the Daqu stage data does not need to be normalized, because the consideration form of the data in the neural network model in this embodiment is not defined in the form of a numerical value, but in the form of a data structure; for example, if the time series length of the input data is preset to 28 days in this embodiment, when it is less than 28 days, the previous data is filled with 0. For example, when the input is 5 days of data, then in the time series data with a length of 28, the last 5 days of data have valid values, and the data of the first 23 days are recorded as 0.

[0045] In the present embodiment, the metabolite compound data represents compound data related to Daqu fermentation, and the compound data may be compound content data. After obtaining the metabolite compound data, the metabolite compound data may be standardized using a minimum-maximum normalization method. For example, the metabolite compound data include one or more compound content data in esters, alcohols, aldehydes, acids, ketones, pyrazines, furans, and aromatics, wherein the esters include at least ethyl hexadecanoate and ethyl hexanoate, the alcohols include at least phenylethyl alcohol and 2,3-butanediol, the acids include at least acetic acid and hexanoic acid, and the pyrazines include at least 2,5-dimethylpyrazine, trimethylpyrazine, and 2,6-dimethylpyrazine. The unit of compound content data is ppm, and for example, the metabolite compound data at a certain moment is shown as: [1.017500007, 0.360192202, 0.28742875, 0.282392042, 0, 0.071789635, ...]. If the maximum value of the metabolite compound data is 4ppm and the minimum value is 0ppm, the metabolite compound data after normalization processing is shown as: [0.254375002, 0.090048051, 0.071857188, 0.070598011, 0, 0.017947409, …].

[0046] In this embodiment, the fermentation data at each time point during the Daqu fermentation process is obtained by sampling according to a preset sampling frequency, and the preset sampling frequency is half a day, one day, two days or three days.

[0047] Step 2: training a time series neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model.

[0048] In this embodiment, training a temporal neural network model based on the Daqu fermentation data set includes:

[0049] Step 21: Use the metabolic compound data at historical time points and their corresponding environmental data and fermentation stage data as input features, and use the corresponding metabolic compound data at future time points as output labels of the time series neural network model to construct a training set and a validation set.

[0050] In this embodiment, the temporal neural network model may be, but is not limited to, a recurrent neural network model, a long short-term memory network model, or a gated recurrent unit model for learning the temporal variation pattern of the content of metabolic compounds.

[0051] In this step, the time point of the output label is the future time point of the input feature, that is, the input feature of the time series neural network model is the metabolic compound data at the historical time point and its corresponding environmental data and fermentation stage data, the output label is the actual metabolic compound data at the future time point, and the output feature of the time series neural network model is the predicted metabolic compound data at the future time point.

[0052] In practical applications, the metabolic compound data at historical time points and their corresponding environmental data and fermentation stage data constitute time series data. Assuming that the Daqu fermentation cycle is 30 days, the maximum time series length for each set of fermentation data is 30. If you need to predict the data for the next two days, the maximum time series length is 28 days. When it is less than 28 days, the previous data can be set to 0. For example, when the input is 5 days of data, then in the time series data with a length of 28, only the last 5 days of data have valid values, and the data for the first 23 days are set to 0.

[0053] Step 22: Based on the training set, the time series neural network model is trained using a supervised learning method, the output features of the time series neural network model are compared with the corresponding output labels, the corresponding loss function is calculated, and the model is optimized through back propagation.

[0054] Step 23, verify the time series neural network model based on the validation set and the cross-validation test set. When the loss function of the time series neural network model is minimized and the prediction error of the validation set or the cross-validation test set is less than the error threshold, the corresponding time series neural network model is used as the metabolic compound prediction model.

[0055] In practical applications, the training set and the validation set can be divided into a ratio of 4:1. The training set can be used to train the time series neural network model, the validation set can be used to verify the prediction error of the time series neural network model, and the cross-validated test set can be used to verify the prediction error of the time series neural network model. When the loss function of the validation set is the smallest and a certain prediction error is less than the error threshold, the metabolic compound prediction model can be obtained. Among them, the loss function can be an error metric function suitable for regression tasks, such as the mean square error, mean absolute error, etc. For example, the loss function is the MSE loss, which is established by the nn.MSELoss() function in the PyTorch library. The Adam optimizer can be used for model optimization.

[0056] After the training is completed, the validation set is verified using the time series neural network model. Figure 2 A schematic diagram showing a comparison of the output features (predicted values) and output labels (true values) for predicting metabolic compounds on the first day in the future is shown. Figure 3 A schematic diagram showing a comparison of the output features (predicted values) and output labels (true values) for predicting metabolic compounds on the second day in the future is shown. Figure 2and Figure 3 In the figure, the horizontal axis represents the metabolic compound number, each metabolic compound number corresponds to a metabolic compound, the vertical axis represents the content of the metabolic compound, and the two colors represent the predicted value and the true value, respectively.

[0057] Step 3: According to the metabolic compound data of the Daqu to be predicted at historical time points and its corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model, the metabolic compound content of the Daqu to be predicted at a future time point is obtained.

[0058] In this embodiment, if it is necessary to determine the metabolic compounds of the Daqu to be predicted at a future time point, firstly, the metabolic compound data of the Daqu to be predicted at multiple historical time points and its corresponding environmental data and fermentation stage data are obtained, and standardized using the same processing method, and then input into the metabolic compound prediction model to obtain the metabolic compound content of the Daqu to be predicted at the corresponding future time point.

[0059] In this embodiment, the method further includes: continuously acquiring fermentation data at new time points during the Daqu fermentation process, and regularly updating the metabolic compound prediction model according to the fermentation data at the new time points.

[0060] Specifically, this embodiment constructs a dynamic update mechanism for the metabolic compound prediction model, and updates the prediction model regularly or in real time based on the latest acquired fermentation data to improve the accuracy of the prediction.

[0061] In summary, the method for predicting metabolic compounds in Daqu fermentation provided in this embodiment, by constructing a metabolic compound prediction model, can realize the prediction of the content of metabolic compounds at future time points in the Daqu fermentation process based on the metabolic compound data at historical time points and their corresponding environmental data and fermentation stage data, thereby realizing the advance prediction of the content of metabolic compounds. Compared with the method of manual observation, the present invention improves the consistency and accuracy, and compared with the method of laboratory detection, improves the efficiency of determining the content of metabolic compounds and reduces the cost. And the present invention improves the accuracy of predicting the content of metabolic compounds by comprehensively considering the influence of environmental factors on metabolic compounds in the Daqu fermentation process. By predicting the content of metabolic compounds in Daqu fermentation, it is possible to realize dynamic real-time monitoring of the content of metabolic compounds during the Daqu fermentation cycle, providing a scientific basis for the optimization of the Daqu fermentation process.

[0062] Figure 4 A schematic diagram of a device for predicting metabolic compounds of Daqu fermentation is shown. Figure 4 , the device comprises:

[0063] An acquisition unit, used to acquire fermentation data at each time point during the fermentation process of Daqu, and establish a Daqu fermentation data set according to the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data and metabolic compound data;

[0064] A training unit, used for training a time series neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model;

[0065] The prediction unit is used to obtain the metabolic compound content of the Daqu to be predicted at a future time point based on the metabolic compound data of the Daqu to be predicted at a historical time point and its corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model.

[0066] Figure 5 A schematic diagram of the structure of an electronic device is shown. Figure 5 , the electronic device includes a processor, a memory and a communication bus;

[0067] The communication bus is used to realize the connection and communication between the processor and the memory;

[0068] The processor is used to execute one or more programs in the memory to implement the steps of the method for predicting metabolic compounds in Daqu fermentation as described in this embodiment.

[0069] It can be understood that since the metabolic compound prediction device and electronic device for Daqu fermentation described in this embodiment are devices and electronic devices for implementing the metabolic compound prediction method for Daqu fermentation described in the embodiment, for the device and electronic device disclosed in the embodiment, since they correspond to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method, which will not be repeated here.

Claims

1. A method for predicting metabolic compounds of Daqu fermentation, characterized in that: The method comprises: Acquire fermentation data at each time point during the Daqu fermentation process, and establish a Daqu fermentation data set according to the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data, and metabolic compound data; Training a time series neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model; According to the metabolic compound data of the Daqu to be predicted at historical time points and its corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model, the metabolic compound content of the Daqu to be predicted at future time points is obtained.

2. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: The environmental data include temperature data, humidity data and air oxygen content data; the fermentation stage data are time nodes in the Daqu fermentation cycle, and the unit of the time node is day or hour.

3. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: The metabolic compound data include one or more compound content data of esters, alcohols, aldehydes, acids, ketones, pyrazines, furans and aromatics, the esters include at least ethyl hexadecanoate and ethyl hexanoate, the alcohols include at least phenylethanol and 2,3-butanediol, the acids include at least acetic acid and hexanoic acid, and the pyrazines include at least 2,5-dimethylpyrazine, trimethylpyrazine and 2,6-dimethylpyrazine.

4. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: The fermentation data at each time point during the Daqu fermentation process is obtained by sampling according to a preset sampling frequency, and the preset sampling frequency is half a day, one day, two days or three days.

5. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: The method further comprises: The environmental data, fermentation stage data and metabolic compound data in the Daqu fermentation data set were standardized, and the metabolic compound data of the predicted Daqu at historical time points and their corresponding environmental data and fermentation stage data were standardized using the same processing method.

6. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: Training a time series neural network model based on the Daqu fermentation data set includes: The metabolic compound data at historical time points and their corresponding environmental data and fermentation stage data are used as input features, and the metabolic compound data at corresponding future time points are used as output labels of the time series neural network model to construct training sets and validation sets; Based on the training set and using a supervised learning method to train a time series neural network model, compare the output features of the time series neural network model with the corresponding output labels, calculate the corresponding loss function, and optimize the model through back propagation; The time series neural network model is verified based on the validation set and the cross-validation test set. When the loss function of the time series neural network model is minimized and the prediction error of the validation set or the cross-validation test set is less than the error threshold, the corresponding time series neural network model is used as the metabolic compound prediction model.

7. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: The method further comprises: The fermentation data at new time points during the Daqu fermentation process are continuously acquired, and the metabolic compound prediction model is regularly updated based on the fermentation data at new time points.

8. The method for predicting metabolic compounds of Daqu fermentation according to claim 1, characterized in that: The temporal neural network model is a recurrent neural network model, a long short-term memory network model or a gated recurrent unit model for learning the temporal variation law of the content of metabolic compounds.

9. A device for predicting metabolic compounds of Daqu fermentation, characterized in that: The device comprises: An acquisition unit, used to acquire fermentation data at each time point during the fermentation process of Daqu, and establish a Daqu fermentation data set according to the fermentation data, wherein the fermentation data includes environmental data, fermentation stage data and metabolic compound data; A training unit, used for training a time series neural network model based on the Daqu fermentation data set to obtain a metabolic compound prediction model; The prediction unit is used to obtain the metabolic compound content of the Daqu to be predicted at a future time point based on the metabolic compound data of the Daqu to be predicted at a historical time point and its corresponding environmental data and fermentation stage data, and based on the metabolic compound prediction model.

10. An electronic device, characterized in that: The electronic device includes a processor, a memory and a communication bus; The communication bus is used to realize the connection and communication between the processor and the memory; The processor is used to execute one or more programs in the memory to implement the steps of the method for predicting metabolic compounds in Daqu fermentation as described in any one of claims 1 to 8.

Images