Controllable fermentation device for monitoring ham flavor based on electronic nose and application thereof

Abstract

This invention relates to a controllable fermentation device for monitoring ham flavor using an electronic nose and its application, belonging to the field of food fermentation and processing technology. It includes a fermentation chamber, a detection chamber, a motor, a PID controller, and a human-machine interface digital display platform. The electronic nose flavor detection chamber includes one or more electronic nose flavor detectors, each of which includes a ring-shaped circulation tube and six odor sensors. The human-machine interface digital display platform incorporates a BP artificial neural network (BP-ANN) intelligent analysis system, which can intelligently control the PID controller. This invention, with its PID controller and human-machine interface digital display platform, can receive real-time feedback from monitoring information for control. The human-machine interface digital display platform, equipped with a BP artificial neural network (BP-ANN) intelligent analysis system, enables intelligent analysis and control, is easy to use, and is more precise than manual control.

Description

Technical Field

[0001] This invention relates to a controllable fermentation device based on electronic nose monitoring of ham flavor and its application, belonging to the field of food fermentation and processing technology. Background Technology

[0002] Fermented meat products refer to meat products produced under natural or artificially controlled conditions using microbial fermentation, possessing unique flavor, color, and texture, and capable of long-term preservation. Among them, ham, as a type of fermented meat product, has existed for a long time. Its production process is not complicated and is widely known, but the key points lie in the control of temperature, humidity, and ventilation during curing and fermentation.

[0003] Nowadays, there are very few regions where ham can be produced naturally, such as Jinhua in Zhejiang, Xuanwei in Yunnan, Panxian in Guizhou, and Rugao in Jiangsu. Other regions are not suitable for ham production, so many machines have been developed for ham production. The most common are fermentation boxes and fermentation tanks, but these are limited to the difficult-to-adjust control of temperature, humidity, and airflow in an attempt to achieve a flavor similar to naturally fermented ham. However, in practice, the results are not ideal, and the flavor is slightly inferior to that of ham produced in a natural environment. This is because fermentation conditions in a natural environment are fluctuating, and different types of ham have their own unique microbial communities.

[0004] Research has found that artificial neural networks do not require pre-determined mathematical equations to map the relationship between input and output. They learn certain rules through training and obtain the closest output value to the expected value given an input value. This is very helpful for comparing the input and expected values ​​of odors. As an intelligent information processing system, the core of artificial neural networks is the algorithm.

[0005] Among them, the BP neural network is a multilayer feedforward network trained by backpropagation of error (abbreviated as backpropagation). Its algorithm is called the BP algorithm. Its basic idea is the gradient descent method, which uses gradient search technology to minimize the mean square error between the actual output value and the expected output value of the network, thereby making it easier to obtain more accurate data to achieve precise parameter value control.

[0006] Therefore, it is necessary to design a device that can control the fermentation process in real time based on simulated natural fermentation conditions and changes in flavor using an artificial neural network system. Summary of the Invention

[0007] To address the aforementioned problems, this invention provides a controllable fermentation device for monitoring ham flavor based on an electronic nose, comprising a fermentation chamber, a detection chamber, a motor, a PID controller, and a human-machine interface digital display platform. The motor and the PID controller are connected. The fermentation chamber is equipped with a humidity controller, a temperature controller, and a fan. The detection chamber and the PID controller are both connected to the fermentation chamber. The human-machine interface digital display platform is connected to the PID controller and the detection chamber. The PID controller can control the fermentation chamber based on information fed back from the detection chamber to the human-machine interface digital display platform.

[0008] The detection chamber includes an electronic nose flavor detection chamber, a temperature detector, and a humidity detector. The electronic nose flavor detection chamber is not connected to the temperature detector and the humidity detector. The electronic nose flavor detection chamber includes one or more electronic nose flavor detectors. Each electronic nose flavor detector includes an annular circulation tube and six odor sensors. Each odor sensor has an independent air chamber and is connected to the others by the annular circulation tube. The six odor sensors include:

[0009] Two odor sensors, including an S1 sulfide sensor and an S2 organic amine sensor;

[0010] Four characteristic aroma sensors, including S3 alcohols, aldehydes, and short-chain alkanes; S4 organic gases, benzophenones, alcohols and aldehydes, and aromatic compounds; S5 aromatic compounds and alcohols and aldehydes; and S6 ketones and alcohols.

[0011] The human-computer interaction digital display platform carries an intelligent analysis system based on a BP artificial neural network (BP-ANN), which can intelligently control the PID controller.

[0012] The fermentation chamber is also equipped with a water outlet, and the electronic nose air inlet pipe is also equipped with a filter screen.

[0013] In one embodiment of the present invention, the electronic nose flavor detection chamber is connected to the fermentation chamber via an electronic nose air inlet pipe, the humidity controller includes a water tank and a humidifier, a foam partition is provided between the PID controller and the electronic nose flavor detection chamber, the detection chamber is also provided with a heat insulation layer, a hook is provided on the top of the fermentation chamber, and casters are provided on the bottom of the device.

[0014] In one embodiment of the present invention, the fan is a 0-100W adjustable power fan.

[0015] This invention also provides an application of a controllable fermentation device based on electronic nose monitoring of ham flavor, wherein the specific steps for regulating the ham fermentation process are as follows:

[0016] Step 1: Raw material processing: Select pork hind legs from the market as raw materials for ham, trim them, remove scraps, and keep the meat surface flat;

[0017] Step 2: Marinating: Use a marinating agent to marinate the raw materials in three additions, rubbing them with your hands after each addition to ensure the salt is fully absorbed into the meat. Then marinate at 4°C for 42 days.

[0018] Step 3: Air drying control: Wash the marinated meat with clean water to remove the surface salt, wipe it dry, and then put it into the fermentation chamber for air drying. After the temperature and humidity of the fermentation chamber are preset, the PID controller controls the fan to air dry the ham meat, and controls the air drying temperature and humidity according to the preset temperature and humidity values.

[0019] Step 4: Fermentation Control: After air drying, set the temperature and humidity values ​​in the fermentation chamber and turn on the fan to ferment the ham. During the process, the aroma of the ham will enter the electronic nose flavor detection chamber. One or more electronic nose flavor detectors will detect the aroma information of the ham and feed it back to the human-computer interaction digital display platform. The intelligent analysis system of BP artificial neural network (BP-ANN) will analyze the data and feed it back to the PID controller to adjust the fermentation parameters of the fermentation chamber, thereby realizing intelligent control of fermentation parameters until the ham is fully fermented and matured.

[0020] In one embodiment of the present invention, during step four, the temperature detector and the humidity detector also feed back temperature and humidity information to the human-computer interactive digital display platform. The data is analyzed by the intelligent analysis system of a BP artificial neural network (BP-ANN) and fed back to the PID controller for further adjustment of fermentation parameters in the fermentation chamber. The weight of the ham raw material is selected as 15-20 kg, and the length is selected as 89-110 cm.

[0021] The beneficial effects of this invention are:

[0022] 1. This invention is equipped with a PID controller and a human-machine interactive digital display platform, which can receive real-time feedback of monitoring information for regulation;

[0023] The human-computer interaction digital display platform is equipped with a BP artificial neural network (BP-ANN) intelligent analysis system, which can realize intelligent analysis and control, is easy to use, and is more accurate than manual control.

[0024] 2. The present invention is equipped with an electronic nose flavor detector, which can monitor various types of odors in real time during the fermentation process and feed the monitoring information back to the PID controller, enabling precise control of the fermentation process and fermentation that is closer to natural conditions; the electronic nose flavor detector is also equipped with an annular circulation tube and 6 independent air chambers, which can ensure that various odors are fully absorbed and detected, reducing omissions.

[0025] 3. The electronic nose flavor detection chamber of the present invention is directly connected to the fermentation chamber by the electronic nose air inlet pipe, which is conducive to the more concentrated and effective entry of gas into the detection area during the fermentation process; the electronic nose air inlet pipe is also equipped with a filter screen, which can filter the effective detection gas.

[0026] 4. A foam partition is provided between the PID controller and the electronic nose flavor detection chamber of the present invention, which can better isolate the components and prevent mutual interference.

[0027] 5. The device of the present invention is also equipped with casters at the bottom, which makes it easy for users to move it to any position.

[0028] 6. The curing process of this invention is carried out in three stages, which allows the salt to be absorbed more fully by the ham, thus improving and ensuring its quality.

[0029] 7. This invention can simultaneously monitor the odor, temperature, and humidity during the fermentation and drying processes, and adjust the fermentation parameters in real time based on intelligent analysis. It is highly efficient and can also adjust multiple parameters simultaneously. Attached Figure Description

[0030] Figure 1 These are front and side views of the device structure in one embodiment of the present invention.

[0031] Figure 2 This is a schematic diagram of the structure of an electronic nose flavor detector in one embodiment of the present invention.

[0032] Figure 3 This is a flowchart illustrating the workflow of a BP artificial neural network in one embodiment of the present invention.

[0033] Figure 4 This is a performance diagram of the optimal neural network model established based on the feedback neural network temperature and humidity prediction model of an electronic nose flavor detector in one embodiment of the present invention.

[0034] Figure 5 This is a flowchart illustrating the intelligent control of the fermentation process in one embodiment of the present invention.

[0035] Figure 1 In the middle, 1: casters, 2: motor, 3: foam partition, 4: temperature sensor, 5: water tank, 6: electronic nose flavor detection chamber, 7: heat insulation layer, 8: humidity detector, 9: filter screen, 10: electronic nose air inlet pipe, 11: hook, 12: humidifier, 13: water outlet, 14: human-machine interactive digital display platform, 15: PID controller, 16: fermentation chamber.

[0036] Figure 2In the diagram, 6.1: S1 sulfide sensor, 6.2: S2 organic amine sensor, 6.3: S3 alcohol, aldehyde, and short-chain alkane sensor, 6.4: S4 organic gas, benzophenone, alcohol, aldehyde, and aromatic compound sensor, 6.5: S5 aromatic compound and alcohol sensor, 6.6: S6 ketone and alcohol sensor, 6.7: annular circulation tube, 6.8: independent gas chamber, and 6.9: air inlet.

[0037] Figure 4 In the diagram, (a) is the training set, (b) is the validation set, (c) is the test set, and (d) is the overall dataset. Detailed Implementation

[0038] Example 1

[0039] like Figure 1 As shown, the present invention provides a controllable fermentation device for monitoring ham flavor based on an electronic nose, including a fermentation chamber 16, a detection chamber, a motor 2, a PID controller 15, and a human-machine interactive digital display platform 14.

[0040] The motor 2 and the PID controller 15 are connected and placed at the bottom of the entire device.

[0041] The fermentation chamber 16 is placed on top of the motor 2. The fermentation chamber 16 is equipped with a hook 11, a humidifier 12, a temperature controller, and a water outlet 13. The humidifier 12 is also connected to a water tank 5. The detection chamber is also equipped with a heat insulation layer 7. The bottom of the device is equipped with casters 1.

[0042] The detection chamber is located directly in front of the fermentation chamber 16 and on the foam partition 3 on top of the PID controller 15. The detection chamber includes a temperature detector 4, a humidity detector 8 and an electronic nose flavor detection chamber 6. As can be seen from the front view, the chamber formed by the temperature detector 4 and the humidity detector 8 is not connected to the electronic nose flavor detection chamber 6 and works independently without interference.

[0043] The electronic nose flavor detection chamber 6 includes one or more electronic nose flavor detectors. The electronic nose flavor detection chamber 6 is connected to the fermentation chamber 16 by an electronic nose air inlet pipe 10. The electronic nose air inlet pipe 10 is equipped with a filter screen 9.

[0044] like Figure 2 As shown, each of the electronic nose flavor detectors includes an annular circulation tube 6.7 and six odor sensors. Each odor sensor has an independent air chamber 6.8 and is connected to each other by the annular circulation tube 6.7. The six odor sensors include:

[0045] Two odor sensors, including S1 sulfide sensor 6.1 and S2 organic amine sensor 6.2;

[0046] Four characteristic aroma sensors are included: S3 (alcohols, aldehydes, and short-chain alkanes) sensor 6.3; S4 (organic gases, benzophenones, alcohols and aldehydes, and aromatic compounds) sensor 6.4; S5 (aromatic compounds and alcohols and aldehydes) sensor 6.5; and S6 (ketones and alcohols) sensor 6.6.

[0047] All six odor sensors can be selected from commonly available sensor units on the market, such as the six metal oxide gas sensor units contained in Shanghai Baosheng Technology's portable electronic nose product:

[0048] Table 1 shows the six types of metal oxide gas sensor units:

[0049] Table 1. Introduction to Shanghai Baosheng Technology's Portable Electronic Nose Sensor

[0050]

[0051] The detection chamber and the PID controller 15 are both connected to the fermentation chamber 16, and the human-machine interactive digital display platform 14 is connected to the PID controller 15 and the flavor detection chamber.

[0052] like Figure 3 As shown, the human-computer interaction digital display platform 14 carries an intelligent analysis system with a BP artificial neural network (BP-ANN) that can intelligently control the PID controller.

[0053] In the following embodiments, the specific method for establishing the BP artificial neural network model is as follows:

[0054] 1. Database establishment: Electronic nose data of the ham during the fermentation process in the ham factory was recorded, along with the corresponding temperature and humidity parameters, resulting in a total of 120 sets of data.

[0055] 2. Referring to the device and method for intelligent control of the drying endpoint of high starch fruit based on low-field nuclear magnetic resonance disclosed in CN109769922A, the

[0041] -

[0051] sections contain the method for establishing a neural network prediction model, including "optimizing the number of neurons in 8 hidden layers (1, 5, 10, 15, 20, 25, 30, 35), 3 transfer functions (tansig, logsig, purelin), and 7 training functions (SDBP, MOBP, VLBP, RPROP, CGBP, QN, LM), and finally determining a 4-25-1 neural network topology, where logsig and purelin are the transfer functions of the hidden and output layers respectively, and LM is the optimal model parameter of the training function." Figure 3 The evaluation results of the model are presented. Under optimal conditions, the training set, validation set, test set, and overall R-value are all measured. 2The values ​​were 0.9992, 0.9915, 0.9964, and 0.9975, respectively, and the MSE was 0.9414 × 10⁻⁶. -5 9.9756×10 -4 5.4353×10 -4 and 3.0227×10 -4 The establishment scheme of the flavor BP artificial neural network temperature and humidity prediction model based on electronic nose in this invention is as follows:

[0056] Using electronic nose signals A11, A22, A33, and A44 as neural network input parameters and temperature and humidity parameters as output parameters, optimization was performed on 10 different numbers of hidden layer neurons (1, 5, 10, 15, 20, 25, 30, 35, 40, 45), 3 transfer functions (tansig, logsig, purelin), and 7 training functions (SDBP, MOBP, VLBP, RPROP, CGBP, QN, LM). Ultimately, a 4-25-1 neural network topology was determined (the topology of a neural network includes the number of network layers, the number of neurons in each layer, and the way neurons are interconnected). From a topological perspective, artificial neural network models can be categorized into hierarchical and interconnected types. Hierarchical models divide the neural network into input layers, hidden layers, and output layers, with each layer sequentially connected. Input layer neurons receive input information from the outside world and transmit it to hidden layer neurons. Hidden layers are responsible for information processing and transformation within the neural network. Typically, one or more hidden layers are designed depending on the transformation requirements. Artificial neural network models primarily consider the network topology, neuron characteristics, learning rules, etc. `logsig` and `purelin` are the transfer functions of the hidden and output layers, respectively, and `LM` represents the optimal model parameters for the training function. Figure 3 The evaluation results of the model are given, under optimal conditions, such as... Figure 4 As shown, the training set (a), validation set (b), and test set (c) are shown; R 2 The values ​​were 0.9099, 0.9757, and 0.8911, respectively; the MSE values ​​of the datasets were 5.8793, 1.7599, and 6.9110, respectively.

[0057] It should be noted that the subset of data used to learn the parameters is usually still called the training set (do not confuse it with the larger dataset used throughout the training process); the subset of data used to select hyperparameters is called the validation set.

[0058] Typically, 80% of the training data is used for training, and 20% is used for validation.

[0059] Since the electronic nose is mainly composed of three parts: gas-sensitive sensor array, signal preprocessing, and pattern recognition, when a certain odor is presented in front of a sensor of an active material, the sensor will convert the chemical signal input of the odor into an electrical signal of the odor. The response of multiple sensors to the electrical signal of an odor constitutes the response spectrum of the sensor array to that odor, and the flavor intensity value of the electronic nose reflects the magnitude of the odor signal received by each sensor.

[0060] The above-mentioned device and the method of establishing a model using a BP artificial neural network are then applied to the fermentation control process of ham with four different curing methods, while keeping the fermentation conditions consistent.

[0061] Example 2

[0062] Dry-curing ham with 100% NaCl:

[0063] Step 1: Raw material processing: Collect 3 hind legs of Xuanhe pig from Yunnan from the market. The weights of the three hind legs are 16.27kg, 17.65kg and 18.44kg respectively, and the lengths are 96cm, 92cm and 104cm respectively. Trim the meat of the three hind legs, remove the edges and corners, and keep the meat surface flat; prepare for marinating.

[0064] Step 2: Marinating: Use 100% NaCl, which accounts for 13% of the weight of the pork hind leg, and marinate for 42 days at 4℃ and 80%–90% relative humidity. Add salt in three batches, and rub the meat with your hands after each addition to ensure that the salt is fully absorbed into the meat.

[0065] Step 3: Air drying control: Wash the marinated meat with clean water to remove the surface salt, wipe it dry, and then put it into the fermentation chamber 16 for air drying. After the temperature and humidity of the fermentation chamber 16 are preset (12℃, relative humidity 50%), the air drying temperature and humidity are controlled by the PID controller according to the preset temperature and humidity values, and the ham is air dried for 60 days.

[0066] Step 4: Fermentation control: After air drying, set the temperature of fermentation chamber 16 to 20℃ and the humidity to 60% in the early stage of fermentation, and turn on the fan to ferment the ham.

[0067] During the mid-fermentation period after 151 days, the S3 alcohol, aldehyde, and short-chain alkanes sensor 6.3, the S4 organic gas, benzophenone, alcohol aldehyde, and aromatic compound sensor 6.4, the S5 aromatic compound and alcohol aldehyde sensor 6.5, and the S6 ketone and alcohol sensor 6.6 in the electronic nose flavor detection chamber 6 were able to detect excessively high ham flavor intensity values, as shown in Table 2. The four odor sensors fed back to the human-computer interactive digital display platform 14, and the data was analyzed by the intelligent analysis system of BP artificial neural network (BP-ANN). When the values ​​deviated from the reference values, the data was fed back to the PID controller to adjust the fermentation parameters of the fermentation chamber, which was adjusted to a temperature of 28°C and a humidity of 75%, 11 days earlier than the manual adjustment of temperature and humidity parameters.

[0068] During the later stages of fermentation, the temperature was adjusted to 20℃ and the humidity to 65% until the ham was fully fermented and matured, a total of 361 days.

[0069] Finally, none of the three hams were spoiled, and the electronic nose flavor intensity values ​​of the three hams remained stable at around that of premium hams, with a yield of 100%.

[0070] Table 2. Electronic nose flavor intensity values ​​of 100% NaCl dry-cured ham during fermentation.

[0071]

[0072] Example 3

[0073] Dry-cured ham using 59.375% NaCl + 28% KCl + 12% MgCl2 + 0.625% amino acids:

[0074] Step 1: Raw material processing: Collect 3 hind legs of Xuanhe pig from Yunnan from the market. The weights of the three hind legs are 17.74kg, 18.66kg and 17.23kg respectively, and the lengths are 89cm, 92cm and 98cm respectively. Trim the meat of the three hind legs, remove the edges and corners, and keep the meat surface flat; prepare for marinating.

[0075] Step 2: Marinating: Use a mixture of 59.375% NaCl + 28% KCl + 12% MgCl2 + 0.625% amino acids, which accounts for 13% of the weight of the pork hind leg, and marinate for 42 days at 4℃ and 80%–90% relative humidity. Add salt in three batches, and rub the meat with your hands after each addition to ensure that the salt is fully absorbed into the meat.

[0076] Step 3: Air drying control: Wash the marinated meat with clean water to remove the surface salt, wipe it dry, and then put it into the fermentation chamber 16 for air drying. After the temperature and humidity of the fermentation chamber 16 are preset (12℃, relative humidity 50%), the air drying temperature and humidity are controlled by the PID controller according to the preset temperature and humidity values, and the ham is air dried for 60 days.

[0077] Step 4: Fermentation control: After air drying, set the temperature of fermentation chamber 16 to 20℃ and the humidity to 60% in the early stage of fermentation, and turn on the fan to ferment the ham.

[0078] After 176 days, the S1 sulfide sensor 6.1 and S2 sulfide sensor 6.2 in the electronic nose flavor detection chamber 6 were able to detect excessively high ham flavor intensity values, indicating that the fermentation chamber 16 had a putrid smell, as shown in Table 3. The two odor sensors fed back to the human-machine interactive digital display platform 14, and the intelligent analysis system of BP artificial neural network (BP-ANN) identified and analyzed the data. It was found that it was still in the early stage of fermentation. When it deviated from the reference value, it fed back to the PID controller to adjust the fermentation parameters of the fermentation chamber, adjusting the temperature to 20°C, the humidity to 60%, and increasing the fan power.

[0079] After 332 days, the fermentation reached the middle stage, and the temperature was adjusted to 28℃ and the humidity to 75%.

[0080] Once the electronic nose response value dropped below the safe range, the fermentation process entered the later stage. The temperature was adjusted to 20℃ and the humidity to 65%. The fan was also returned to normal power. Then, fermentation continued in fermentation chamber 16 until the ham was fully fermented. In the end, the ham matured 11 days later than expected, but this ensured that the ham did not spoil (a total of 383 days of fermentation).

[0081] Finally, none of the three hams were spoiled, and the electronic nose flavor intensity values ​​of the three hams remained stable at around that of premium hams, with a yield of 100%.

[0082] Table 359.37 Electronic nose flavor intensity values ​​of dry-cured ham containing 5% NaCl, 28% KCl, 12% MgCl2, and 0.625% amino acids during fermentation.

[0083]

[0084] Example 4

[0085] Dry-curing ham with 60% NaCl + 28% KCl + 12% MgCl2:

[0086] Step 1: Raw material processing: Collect 3 pig hind legs from Xuanhe, Yunnan, from the market. The weights of the three pig hind legs are 17.45kg, 18.02kg and 17.77kg respectively, and the lengths are 95cm, 93cm and 101cm respectively. Trim the meat of the three pig hind legs, remove the edges and corners, and keep the meat surface flat; prepare for marinating.

[0087] Step 2: Marinating: Use a mixture of 60% NaCl + 28% KCl + 12% MgCl2, which accounts for 13% of the weight of the pork hind leg, and marinate for 42 days at 4℃ and 80%–90% relative humidity. Add salt in three batches, and rub the meat with your hands after each addition to ensure that the salt is fully absorbed into the meat.

[0088] Step 3: Air drying control: Wash the marinated meat with clean water to remove the surface salt, wipe it dry, and then put it into the fermentation chamber 16 for air drying. After the temperature and humidity of the fermentation chamber 16 are preset (12℃, relative humidity 50%), the air drying temperature and humidity are controlled by the PID controller according to the preset temperature and humidity values, and the ham is air dried for 60 days.

[0089] Step 4: Fermentation control: After air drying, set the temperature of fermentation chamber 16 to 20℃ and the humidity to 60% in the early stage of fermentation, and turn on the fan to ferment the ham.

[0090] After 170 days, the S3 alcohol, aldehyde, and short-chain alkanes sensor 6.3, the S4 organic gas, benzophenone, alcohol, aldehyde, and aromatic compound sensor 6.4, the S5 aromatic compound and alcohol sensor 6.5, and the S6 ketone and alcohol sensor 6.6 in the electronic nose flavor detection chamber 6 were able to detect excessively high ham flavor intensity values, as shown in Table 4. At this point, the fermentation reached the middle stage, and the temperature was adjusted to 28°C and the humidity to 75%.

[0091] After 236 days, the S3 alcohol, aldehyde, and short-chain alkanes sensor 6.3, the S4 organic gas, benzophenone, alcohol, aldehyde, and aromatic compound sensor 6.4, the S5 aromatic compound and alcohol sensor 6.5, and the S6 ketone and alcohol sensor 6.6 in the electronic nose flavor detection chamber 6 were able to detect excessively high ham flavor intensity values, as shown in Table 5. Entering the later stage of fermentation, when the values ​​deviated from the reference values, feedback was sent to the PID controller to adjust the fermentation parameters of the fermentation chamber. The temperature was adjusted to 20℃, the humidity to 65%, and the fan returned to normal power. Then, the fermentation chamber 16 continued fermentation until the ham was fully fermented, for a total of 357 days.

[0092] Finally, none of the three hams were spoiled, and the electronic nose flavor intensity values ​​of the three hams remained stable at around that of premium hams, with a yield of 100%.

[0093] Table 4. Electronic nose flavor intensity values ​​of 60% NaCl + 28% KCl + 12% MgCl2 dry-cured ham during fermentation.

[0094]

[0095] Table 5. Electronic nose flavor intensity values ​​of dry-cured ham with 60% NaCl + 28% KCl + 12% MgCl2 during fermentation.

[0096]

[0097] Example 5

[0098] Dry-curing ham with 60% NaCl + 28% KCl + 12% MgCl2:

[0099] Step 1: Raw material processing: Collect 3 hind legs of Xuanhe pig from Yunnan from the market. The weights of the three hind legs are 16.89kg, 17.47kg and 16.78kg respectively, and the lengths are 92cm, 101cm and 98cm respectively. Trim the meat of the three hind legs, remove the edges and corners, and keep the meat surface flat; prepare for marinating.

[0100] Step 2: Marinating: Use a mixture of 60% NaCl + 28% KCl + 12% MgCl2, which accounts for 13% of the weight of the pork hind leg, and marinate for 42 days at 4℃ and 80%–90% relative humidity. Add salt in three batches, and rub the meat with your hands after each addition to ensure that the salt is fully absorbed into the meat.

[0101] Step 3: Air drying control: Wash the marinated meat with clean water to remove the surface salt, wipe it dry, and then put it into the fermentation chamber 16 for air drying. After the temperature and humidity of the fermentation chamber 16 are preset (12℃, relative humidity 50%), the air drying temperature and humidity are controlled by the PID controller according to the preset temperature and humidity values, and the ham is air dried for 60 days.

[0102] Step 4: Fermentation control: After air drying, set the temperature of fermentation chamber 16 to 20℃ and the humidity to 60% in the early stage of fermentation, and turn on the fan to ferment the ham.

[0103] After 155 days, the S3 alcohol, aldehyde, and short-chain alkane sensor 6.3, the S4 organic gas, benzophenone, alcohol, aldehyde, and aromatic compound sensor 6.4, the S5 aromatic compound and alcohol sensor 6.5, and the S6 ketone and alcohol sensor 6.6 of the electronic nose flavor detection chamber 6 were able to detect excessively high ham flavor intensity values, as shown in Table 6. At this point, the fermentation has reached the middle stage. When the value deviates from the reference value, the feedback is sent to the PID controller to adjust the fermentation parameters of the fermentation chamber to 28°C and 75% humidity.

[0104] Finally, in the later stage of fermentation, the temperature was adjusted to 20℃ and the humidity to 65%, and the fan was also returned to normal power. Then, fermentation continued in fermentation box 16 until the ham was fully fermented, for a total of 365 days.

[0105] Finally, none of the three hams were spoiled, and the electronic nose flavor intensity values ​​of the three hams remained stable at around that of premium hams, with a yield of 100%.

[0106] Table 6. Electronic nose flavor intensity values ​​of 60% NaCl + 28% KCl + 12% MgCl2 dry-cured ham during fermentation.

[0107]

[0108] Comparative Example

[0109] This comparative example uses a traditional fermentation process and is compared with the intelligent fermentation processes of Examples 2, 3, 4 and 5 of this invention.

[0110] In this comparative example, the pickling, air-drying, and fermentation times, temperatures, and humidity conditions were the same as those in Examples 2, 3, 4, and 5, as shown in Table 7.

[0111] Table 7 Preparation conditions

[0112]

[0113] After experimental comparison, the following experimental results were obtained, as shown in Table 8:

[0114] Table 8. Fermentation results of the intelligent fermentation process of this invention compared to the traditional fermentation process.

[0115]

[0116]

[0117] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the claims.

Claims

1. A controllable fermentation device based on electronic nose monitoring of ham flavor, characterized in that, The system includes a fermentation chamber, a detection chamber, a motor, a PID controller, and a human-machine interface digital display platform. The motor and the PID controller are connected. The fermentation chamber is equipped with a humidity controller, a temperature controller, and a fan. The detection chamber and the PID controller are both connected to the fermentation chamber. The human-machine interface digital display platform is connected to the PID controller and the detection chamber. The PID controller can control the fermentation chamber based on the information fed back from the detection chamber to the human-machine interface digital display platform. The detection chamber includes an electronic nose flavor detection chamber, a temperature detector, and a humidity detector. The electronic nose flavor detection chamber is not connected to the temperature detector and the humidity detector. The electronic nose flavor detection chamber includes one or more electronic nose flavor detectors. Each electronic nose flavor detector includes an annular circulation tube and six odor sensors. Each odor sensor has an independent air chamber and is connected to each other by the annular circulation tube. The six odor sensors include: Two odor sensors, including an S1 sulfide sensor and an S2 organic amine sensor; Four characteristic aroma sensors, including S3 alcohols, aldehydes, and short-chain alkanes; S4 organic gases, benzophenones, alcohols and aldehydes, and aromatic compounds; S5 aromatic compounds and alcohols and aldehydes; and S6 ketones and alcohols. The human-computer interaction digital display platform incorporates an intelligent analysis system based on a BP artificial neural network (BP-ANN), which can intelligently control the PID controller. The fermentation chamber is also equipped with a water outlet; The electronic nose flavor detection chamber is connected to the fermentation chamber by an electronic nose air inlet pipe, and the electronic nose air inlet pipe is also equipped with a filter.

2. The controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 1, characterized in that, The humidity controller includes a water tank and a humidifier.

3. The controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 1, characterized in that, A foam partition is provided between the PID controller and the electronic nose flavor detection chamber, and the detection chamber is also provided with a heat insulation layer.

4. The controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 1, characterized in that, The fermentation chamber is equipped with hooks at the top.

5. The controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 1, characterized in that, The controllable fermentation device based on electronic nose monitoring of ham flavor is equipped with casters at the bottom.

6. The controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 1, characterized in that, The fan is a 0-100W adjustable power fan.

7. The application of the controllable fermentation device for monitoring ham flavor based on an electronic nose as described in claim 1, characterized in that, The specific steps for controlling the fermentation process of ham are as follows: Step 1: Raw material processing: Select pork hind legs from the market as raw materials for ham, trim them, remove scraps, and keep the meat surface flat; Step 2: Marinating: Use a marinating agent to marinate the raw materials, adding it in three batches, and rubbing it with your hands after each addition to ensure that the salt is fully absorbed into the meat. Then marinate at 4°C for 42 days. Step 3: Air drying control: Wash the marinated meat with clean water to remove the surface salt, wipe it dry, and then put it into the fermentation chamber for air drying. After the temperature and humidity of the fermentation chamber are preset, the PID controller controls the fan to air dry the ham meat, and controls the air drying temperature and humidity according to the preset temperature and humidity values. Step 4: Fermentation Control: After air drying, set the temperature and humidity values ​​in the fermentation chamber and turn on the fan to ferment the ham. During the process, the aroma of the ham will enter the electronic nose flavor detection chamber. One or more electronic nose flavor detectors will detect the aroma information of the ham and feed it back to the human-computer interaction digital display platform. The intelligent analysis system of artificial neural network (BP-ANN) will analyze the data and feed it back to the PID controller to adjust the fermentation parameters of the fermentation chamber, thereby realizing intelligent control of fermentation parameters until the ham is fully fermented and matured.

8. The application of the controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 7, characterized in that, During step four, the temperature detector and the humidity detector will also feed back temperature and humidity information to the human-computer interaction digital display platform. The intelligent analysis system of BP artificial neural network (BP-ANN) will perform data analysis and feed it back to the PID controller to further regulate the fermentation parameters of the fermentation chamber.

9. The application of the controllable fermentation device for monitoring ham flavor based on an electronic nose according to claim 7, characterized in that, The weight of the ham raw material is selected as 15-20kg, and the length is selected as 89-110cm.

Images