Fat and oil degradation prediction device, degradation prediction system, degradation prediction method, fat and oil replacement system, and fryer system

Abstract

The present invention provides a degradation prediction device that easily and accurately predicts the degradation of edible fats and oils. A deterioration prediction device

Description

[Technical Field]

[0001] This invention relates to a deterioration prediction device, a deterioration prediction system, a deterioration prediction method, a grease replacement system, and a fryer system for predicting the degree of deterioration of grease. [Previous Technology]

[0002] The cooking oil used for frying food will gradually deteriorate after being used to cook the food several times, so it must be replaced at the appropriate time. In order to objectively determine when to replace such oil, there are known devices that detect the color, viscosity, odor, etc. of the oil.

[0003] For example, the detection device in Patent Document 1 described below has a sensing unit installed on the exhaust fan above the oil tank. This sensing unit includes a sensing membrane that adsorbs gas molecules from the source of the odor, and a converter that converts the gas molecules attached to the sensing membrane into electrical signals, and detects the odor produced by the cooking oil. Then, the control unit of the detection device determines the degree of deterioration of the cooking oil based on the odor information detected by the sensing unit during frying and the type of food cooked with the cooking oil (paragraphs 0017, 0021, Figure 1).

[0003] [Previous Technical Documents]

[0003] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent No. 6448811

[0005] However, in the case of the detection device in Patent Document 1, there are various odors in the cooking place in addition to the ingredients being cooked (odors other than the aroma of fried food, burnt odors, etc.), so it is difficult to accurately predict the deterioration of cooking oil based solely on odors.

[0006] The present invention is made in view of such circumstances, and aims to provide a deterioration prediction device that can easily and accurately predict the deterioration of oils.

[0007] The deterioration prediction device of the present invention predicts the degree of deterioration of edible oils. The deterioration prediction device comprises: an audio data acquisition unit that acquires audio data when frying food using the aforementioned oil contained in an oil tank; an index acquisition unit that acquires an index related to the deterioration of the aforementioned oil from the audio data acquired by the audio data acquisition unit; and a determination unit that determines the degree of deterioration of the aforementioned oil based on the aforementioned index acquired by the index acquisition unit.

[0008] The sound data acquisition unit of the deterioration prediction device acquires sound data of the oil during the preparation of fried foods such as tempura. The index acquisition unit extracts various sound components, such as frequency average and frequency standard deviation, from the sound data as indicators related to oil deterioration. Then, the determination unit determines the degree of oil deterioration, that is, whether it continues to deteriorate due to use, based on the indicators. In this way, this device can easily and accurately predict oil deterioration.

[0009] In the deterioration prediction device of the present invention, it is preferable to further include a notification unit, which notifies the degree of deterioration of the aforementioned grease or the timing of replacing the aforementioned grease; the notification unit makes the aforementioned notification when the determination unit determines, based on the degree of deterioration of the aforementioned grease, that it exceeds a predetermined replacement threshold.

[0010] According to this configuration, since the notification unit of the deterioration prediction device notifies the degree of deterioration of the grease, the user can keep track of the grease's usage status. Furthermore, the notification unit notifies the user of the appropriate time to replace the grease based on the determination result of the determination unit, starting from a pre-defined threshold. Therefore, the user can replace the grease at the appropriate time. Here, "time of replacement" can refer to the actual time to replace the grease, or it can refer to the remaining usable time estimated from the current degree of grease deterioration.

[0011] Furthermore, in the degradation prediction device of the present invention, the aforementioned indicators are preferably selected from one or more of the following: frequency average, frequency standard deviation, frequency median, frequency standard error, frequency maximum frequency, frequency first quartile, frequency third quartile, frequency quartile range, frequency centroid, frequency skewness, frequency kurtosis, spectral flatness, spectral entropy, spectral accuracy, sound complexity index, sound entropy, and dominant frequency.

[0012] Based on this configuration, the aforementioned indicators are selected from one or more that are highly correlated with oil deterioration. Therefore, this device can predict its deterioration with high accuracy.

[0013] The degradation prediction system of the present invention is composed of a detection device and a machine learning device, and predicts the degree of degradation of edible oils. The detection device includes: a sound data acquisition unit, which acquires sound data when frying food using the aforementioned oil contained in an oil tank; a memory unit, which stores a learning model generated by the aforementioned machine learning device that can determine the degradation of the aforementioned oil; and a determination unit, which uses the aforementioned learning model to determine the degree of degradation of the aforementioned oil from the aforementioned sound data. The machine learning device includes: a learning model generation unit, which extracts indicators related to the degradation of the aforementioned oil from the aforementioned sound data acquired by the sound data acquisition unit, and uses the aforementioned indicators to perform linear regression machine learning to generate the aforementioned learning model.

[0014] The degradation prediction system of the present invention is composed of a detection device and a machine learning device. The detection device consists of a sound data acquisition unit that acquires sound data of the oil when frying food, and a determination unit that uses a learning model to determine the degree of degradation of the oil.

[0015] Next, the learning model generation unit extracts indicators related to the deterioration of oils from the acquired sound data and performs linear regression machine learning. By updating the learning model, this system can easily and accurately predict the deterioration of oils.

[0016] In the degradation prediction system of the present invention, the aforementioned linear regression is preferably selected from one or more of simple regression, multiple regression, partial least squares (PLS) regression, or orthogonal projective partial least squares (OPLS) regression.

[0017] The learning model is generated using linear regression methods such as single regression, multiple regression, partial least squares (PLS) regression, and orthogonal projective partial least squares (OPLS) regression. Through this, the system can generate a learning model that can accurately determine the deterioration of oils.

[0018] Furthermore, in the degradation prediction system of the present invention, it is preferable that the aforementioned detection device and the aforementioned machine learning device are integrated.

[0019] For example, near an oil tank in a shop or factory, by installing a deterioration prediction system of one form of the present invention, users can obtain the prediction results of oil deterioration on the spot.

[0020] Furthermore, in the degradation prediction system of the present invention, it is preferable that the aforementioned detection device is located near the aforementioned oil tank in the shop or factory, and the aforementioned machine learning device is located at a distance from the aforementioned shop or factory.

[0021] The detection device, including the sound data acquisition unit, is installed near the oil tank in the shop or factory, but the machine learning device can be installed at a distance from the shop. Since the machine learning device is a different device from the detection device, the learning model generated by the machine learning device can be obtained through communication or the like.

[0022] Furthermore, in the degradation prediction system of the present invention, it is preferable that the aforementioned detection device includes: a first communication unit, which transmits the aforementioned sound data obtained by the aforementioned sound data acquisition unit to the aforementioned machine learning device; the aforementioned machine learning device includes a second communication unit that receives the aforementioned sound data from the aforementioned detection device.

[0023] The detection device, having a first communication unit, transmits sound data to the machine learning device. The machine learning device, having a second communication unit, receives the sound data and performs machine learning. When the detection device and the machine learning device are separate devices, this system can distribute the necessary tasks by having each communication unit transmit / receive data.

[0024] Furthermore, in the degradation prediction system of the present invention, the aforementioned first communication unit and the aforementioned second communication unit are preferably capable of wireless communication.

[0025] The detection device is designed so that sound data can be transmitted wirelessly via the first communication unit for machine learning devices, thus miniaturizing the function of the detection device.

[0026] The degradation prediction method of the present invention is to predict the degree of degradation of edible oils. The degradation prediction method is characterized by having: a sound data acquisition step, which acquires sound data when the aforementioned oil is used to cook fried food;

[0026] The indicator extraction step involves extracting indicators related to the deterioration of the aforementioned oils from the sound data obtained in the aforementioned sound data acquisition step; and

[0026] The determination step is based on the aforementioned indicators obtained in the aforementioned indicator extraction step to determine the degree of deterioration of the aforementioned oil.

[0027] The degradation prediction method of the present invention acquires sound data of oil used in cooking fried foods such as tempura during the sound data acquisition step. Next, in the index extraction step, various sound components, such as frequency average and frequency standard deviation, are extracted from the sound data as indicators related to oil degradation. Furthermore, in the judgment step, the degree of oil degradation is determined based on these indicators, that is, whether it continues to degrade due to use. Therefore, this method can easily and accurately predict oil degradation.

[0028] The grease replacement system of the present invention is characterized by performing one or more of the following a) to e) based on the notification information related to the degree of deterioration of the aforementioned grease output by the aforementioned deterioration prediction device: a) notifying grease sellers to order new grease; b) notifying grease manufacturers to establish grease manufacturing or sales plans; c) notifying the overall management headquarters of shops or factories, or grease manufacturers, and providing suggestions or guidance on the use of grease to the shops or factories under their management; d) notifying waste oil recyclers or grease manufacturers to arrange waste oil recycling; e) notifying cleaning operators to arrange oil tank cleaning.

[0029] In the grease replacement system of the present invention, new grease is ordered from grease retailers based on notification information related to the degree of grease deterioration, for example, when the notification information has been sent a predetermined number of times. Furthermore, based on the notification information, grease manufacturers are notified to establish grease manufacturing or sales plans. Thus, this system can establish manufacturing and sales plans corresponding to the grease replacement schedule.

[0030] Furthermore, the grease replacement system, based on this notification information, notifies the coordinating headquarters of shops or factories, or grease manufacturers, and provides suggestions or guidance on the use of grease to the shops or factories under its coordination. For example, the coordinating headquarters guides each shop to use grease in a way that avoids waste and replaces it appropriately. Moreover, based on this notification information, it notifies waste oil recycling companies to arrange waste oil collection and notifies cleaning companies to arrange oil tank cleaning. Therefore, this system can quickly operate from the supply of grease to the disposal of waste oil.

[0031] The fryer system of the present invention includes a valve control unit, which controls the valve provided in the oil tank based on notification information related to the degree of deterioration of the aforementioned grease output from the aforementioned deterioration prediction device; and the aforementioned valve control unit automatically discards the aforementioned grease contained in the aforementioned oil tank.

[0032] The fryer system of the present invention controls the valve of the oil tank based on notification information related to the degree of deterioration of the grease. In this way, the system can automatically dispose of grease in use.

[0033] In the fryer system of the present invention, the aforementioned valve control unit preferably automatically supplies new oil to the aforementioned oil tank.

[0034] According to this configuration, the valve control unit controls the valve to automatically supply new oil to the oil tank. In this way, the user can reduce the series of work burdens from confirming the degree of deterioration of the grease and its disposal to the supply of new oil.

[0035] According to the present invention, the deterioration of oils can be predicted easily and with high accuracy. [Simplified Explanation of the Diagram]

[0036] Figure 1 is a schematic diagram illustrating the degradation prediction device and fryer of the first embodiment.

[0036] Figure 2 is a functional block diagram of the degradation prediction device in the first embodiment.

[0036] Figure 3 is a flowchart of the process for determining the deterioration of frying oil using a deterioration prediction device.

[0036] Figure 4 is a functional block diagram of the degradation prediction device (degradation prediction system) of the second embodiment.

[0036] Figure 5A shows the relationship between the calibration curve (also known as the standard curve or calibration curve) obtained by machine learning (single regression) and the heating time of the test data.

[0036] Figure 5B is a table showing the average and standard deviation of the predicted values ​​in Figure 5A.

[0036] Figure 6A shows the relationship between the calibration curve obtained by machine learning (regression) and the acid value of the test data.

[0036] Figure 6B is a table showing the average of the measured values, predicted values, and standard deviations of Figure 6A.

[0036] Figure 7A shows the relationship between the calibration curve obtained by machine learning (OPLS) and the heating time of the test data.

[0036] Figure 7B is a table showing the average and standard deviation of the predicted values ​​in Figure 7A.

[0036] Figure 8A shows the relationship between the calibration curve obtained by machine learning (OPLS) and the acid value of the test data.

[0036] Figure 8B is a table showing the average of the measured values, predicted values, and standard deviations of Figure 8A.

[0036] Figure 9A is a graph showing the relationship between the color of the calibration curve and the test data obtained by machine learning (PLS).

[0036] Figure 9B is a table showing the average of the measured values, predicted values, and standard deviations of Figure 9A.

[0036] Figure 10A shows the relationship between the calibration curve obtained by machine learning (PLS) and the viscosity increase rate of the test data.

[0036] Figure 10B is a table showing the average of the measured values, predicted values, and standard deviations of Figure 10A.

[0036] Figure 11 is a diagram illustrating the grease replacement system of the third embodiment.

[0036] Figure 12 is a diagram illustrating the fryer system of the fourth embodiment.

Implementation Method

[0037] Hereinafter, with reference to the drawings, one embodiment of the degradation prediction device of the present invention will be described.

[0038] [First Implementation Type]

[0038] First, referring to FIG1, the general outline of the degradation prediction device 1 and the fryer 20 of the first embodiment of the present invention will be described. As shown, the degradation prediction device 1 is mainly composed of an audio data acquisition unit 2 (the "audio data acquisition unit" of the present invention) and a processing unit 3. The audio data acquisition unit 2 is, for example, a highly directional microphone, and acquires the sound (such as the sound of bubbles bursting) when fried food is cooked by frying food in frying oil (the "oil" of the present invention) contained in the fryer 20.

[0039] The acquired sound (hereinafter, also referred to as sound data) is transmitted to the processing unit 3. Then, the processing unit 3 extracts characteristic quantities and analyzes the deterioration of the frying oil from these characteristic quantities. Details will be described later, but the processing unit 3 includes a display unit 5, a control unit 10, etc.

[0040] The deep fryer 20 has a box-shaped cabinet 21, and inside it is an oil tank 22 for holding frying oil. The temperature of the frying oil held in the oil tank 22 can be adjusted by a heater 23. For example, when making fried meat patties, the frying oil is adjusted to 180°C.

[0041] Furthermore, an oil drain pipe 25 is connected to the bottom surface of the oil tank 22 via a valve 24. The bottom surface of the oil tank 22 is shaped like a downward-sloping funnel to facilitate oil drainage. Deteriorated frying oil is discharged as waste oil by opening the valve 24. The waste oil tank 26 is located below the oil drain pipe 25 to collect waste oil.

[0042] The oil tank 22 is assumed to be used in large deep fryers such as those used in restaurants and izakayas, but it is not limited thereto. That is, the oil tank 22 can be used in smaller deep fryers, or it can be used as a deep frying appliance for home use.

[0043] In this embodiment, the sound data acquisition unit 2 is installed at a height of approximately 1 m above the fryer 20 (slightly above the oil tank 22). Normally, because cooking produces fumes, an exhaust fan (not shown) is installed above the fryer 20 to expel the fumes outdoors. The sound data acquisition unit 2 can be installed on the side of the exhaust fan, etc. Alternatively, the sound data acquisition unit 2 can simply be installed on the side of the cabinet 21, or near the oil tank 22 on a wall, ceiling, etc.

[0044] Figure 2 is a functional block diagram of the degradation prediction device 1 in the first embodiment.

[0045] The degradation prediction device 1 is composed of a sound data acquisition unit 2 and a processing unit 3. The processing unit 3 includes an input unit 4, a display unit 5, a memory unit 6, a notification unit 7, and a control unit 10. First, the sound data acquisition unit 2 acquires the sound of cooking meat patties, tempura, etc.

[0046] The sound data acquisition unit 2 can use a microphone, or record sound using the recording function of a camcorder or smartphone. For example, the sound data acquisition unit 2 sets the sound sampling rate to 48kHz and acquires sound data of the cooking time of the fried ingredients. Furthermore, in the sound data, the operating sound caused by the addition or removal of the fried ingredients is considered noise, so the sound of the first 10 seconds after the start of recording and the last 10 seconds before the end of recording are removed.

[0047] If the frying oil deteriorates, the fatty acids contained in the frying oil will be decomposed, and the sound during cooking will change slowly. The index acquisition unit 11 of the control unit 10 extracts the index (hereinafter, also referred to as index data) related to the deterioration of the frying oil from the acquired sound data, and the result receiving unit 12 receives the index data.

[0048] Most of the indicator data are the frequency characteristics of the sound during cooking, so the frequency mean (f_mean), frequency standard deviation (f_sd), frequency median (f_median), frequency standard error (f_sem), and frequency mode (f_mode) are used.

[0049] Other index data can be listed as follows: the first quartile (f_Q25) of the frequency at the 25th percentile from the minimum frequency, the third quartile (f_Q75) of the frequency at the 75th percentile from the minimum frequency, the frequency quartile range (f_IQR), the center of gravity (f_cent), the skewness (f_skewness), the kurtosis (f_kurtosis), the spectral flatness (f_sfm), the spectral entropy (f_sh), the spectral precision (prec), the sound complexity index (d.ACI), the sound entropy (dH), and the dominant frequency (dfnum). Furthermore, in sound analysis, Seewave (Sound Analysis and Synthesis) and ROPLLS ​​(PCA, PLS(-DA) and OPLS(-DA) for multivariate analysis) are used.

[0050] Specifically, the control unit 10 is a processor that controls and manages the entire degradation prediction device 1, and is composed of a CPU (Central Processing Unit) that executes a program with a prescribed control sequence. Such a program is, for example, stored in the memory unit 6 or other external memory media device.

[0051] The control unit 10 performs various processes of the degradation prediction device 1 by controlling the entire processing unit 3. For example, the control unit 10 activates the degradation prediction device 1 based on the predetermined input operation performed by the user (employee). The predetermined input operation is, for example, the operation of turning on the power of the degradation prediction device 1, the operation of setting the cooking time or the temperature of the frying oil.

[0052] The input unit 4 is a switch that accepts input operations from the user, such as operation buttons or operation keys. The input unit 4 is not limited to this and may also be constructed using a touch panel. Furthermore, the input unit 4 accepts predetermined input operations from the user before the execution of the processing performed by the degradation prediction device 1, and transmits the signal based on the user's input operation to the control unit 10.

[0053] The display unit 5 displays various items for the user to input. For example, when the user selects the type of food to be cooked, the display unit 5 displays the type of food based on the data related to the type of food stored in the memory unit 6. Also, when the notification unit 7 notifies the user of the degree of deterioration of the frying oil, the display unit 5 displays the point that it must be replaced as an auxiliary notification.

[0054] The memory unit 6 is constructed using semiconductor memory or magnetic memory, and stores various information and programs used to activate the degradation prediction device 1. In addition to storing acquired sound data and learning models, the memory unit 6 also stores data related to the food being prepared. For example, for each type of processed food, the memory unit 6 stores relevant data relating the displayed sound data to the degree of degradation of the frying oil. Furthermore, the memory unit 6 stores threshold information for notifications that varies depending on the type of food.

[0055] When the notification unit 7 determines that the deterioration level of the frying oil exceeds a predetermined threshold, it notifies the user. In this way, the notification unit 7 informs the user of the appropriate time to replace the frying oil. Here, "appropriate time to replace" refers to the actual time to replace the frying oil (displayed as "It is time to replace."). Furthermore, the notification unit 7 can also notify the user of the current deterioration level of the frying oil (displayed as "The current deterioration level is 50%."), and can also notify the user of the remaining usable time estimated from the deterioration level (displayed as "20 hours of usable time remaining.").

[0056] The notification unit 7 is equipped with a speaker and can provide notifications through auditory methods such as sound guidance and alarms. Furthermore, the notification unit 7 can also provide notifications through visual methods such as displaying images, text, colors, and emitting light. For example, the display unit 5 can be used to display images or text for notification, or light-emitting elements such as LEDs can be used for notification. Notifications provided by the notification unit 7 are not limited to visual or auditory methods; any combination of these methods or any method that allows the user to objectively identify when to change the frying oil, such as vibration, can be used.

[0057] The comparison and judgment unit 13 of the control unit 10 compares the acquired sound data with relevant data corresponding to the type of food being cooked using the frying oil to determine the degree of deterioration of the frying oil. The sound generated during cooking with the frying oil contained in the oil tank 22 depends on the type of food being cooked. The most appropriate time to replace the frying oil also varies depending on the type of food being cooked.

[0058] The relevant data is pre-memorized in the memory unit 6. When comparing, the comparison and determination unit 13 obtains the relevant data from the memory unit 6 and determines the degree of deterioration of the frying oil. Furthermore, the relevant data is generated by the machine learning unit 14, but it does not necessarily need to be generated inside the deterioration prediction device 1; relevant data provided from the outside can also be used.

[0059] When the control unit 10 determines that the degree of deterioration of the frying oil exceeds a predetermined threshold corresponding to the food type, it controls the notification unit 7 to issue a notification. This threshold is predetermined for each type of food. The threshold can also be appropriately changed according to the user. Furthermore, multiple thresholds can be set.

[0060] Next, referring to Figure 3, a flowchart of the degradation determination of frying oil using the degradation prediction device 1 will be explained. Also, Figure 3 is a flowchart when the threshold value of the standard for replacing frying oil is preset.

[0061] First, the user obtains information related to the food being cooked and makes the necessary settings (step 10). The sound during cooking varies depending on whether the fried food is a meat patty or tempura, so the degradation prediction device 1 is set to the settings corresponding to the fried food. Then, proceed to step 20.

[0062] In step 20, index data is generated from the sound during cooking. Specifically, the sound data acquisition unit 2 acquires the sound (sound data) during cooking and transmits it to the processing unit 3 to generate index data such as frequency average (f_mean). Then, proceed to step 30.

[0063] In step 30, relevant data is obtained from the memory unit. This relevant data is necessary for determining the degree of deterioration of the frying oil in subsequent steps. Then, proceed to step 40.

[0064] In step 40, the two data are compared to determine the degree of deterioration of the frying oil. Specifically, the comparison and determination unit 13 of the control unit 10 compares the sound data with the relevant data. Then, the process proceeds to step 50.

[0065] Next, it is determined whether the degree of deterioration of the frying oil exceeds a predetermined threshold (step 50). Although the threshold varies depending on the food material being fried, if it exceeds the threshold, the process proceeds to step 60, and if it does not exceed the threshold, the process returns to step 20.

[0066] When the deterioration of the frying oil exceeds a predetermined threshold (step 50: Yes), the user is notified (step 60). Specifically, the notification is sent via the notification unit 7 to urge the user to replace the frying oil. Afterward, the series of processes are terminated.

[0067] [Second Implementation Type]

[0067] Next, referring to FIG4, an outline of the degradation prediction system 100 of the second embodiment of the present invention will be described. The degradation prediction system 100 is mainly composed of a detection device 30 and a machine learning device 40. The detection device 30 and the machine learning device 40 are connected by a network NW and can transmit / receive various data to each other.

[0068] The detection device 30 includes a sound data acquisition unit 2, an input unit 4, a display unit 5, a memory unit 6, a notification unit 7, a communication unit 8, and a control unit 10. Furthermore, the control unit 10 includes a comparison and determination unit 13. Also, except for the communication unit 8, all its components are the same as those of the processing unit 3 in the first embodiment, so descriptions are omitted.

[0069] In the detection device 30, when the sound data acquisition unit 2 acquires the sound of cooking meat patties, tempura, etc., the comparison and determination unit 13 compares the acquired sound data with relevant data corresponding to the type of food being cooked, and determines the degree of deterioration of the frying oil.

[0070] Furthermore, the communication unit 8 (the "first communication unit" of this invention) automatically transmits voice data to the machine learning device 40 via a network NW. This communication can be wired or wireless, such as Wi-Fi (registered trademark) or Bluetooth (registered trademark). Since the degradation prediction system 100 only requires a detection device 30 in the shop or factory (near the oil tank 22), the device can be miniaturized.

[0071] The machine learning device 40 includes a communication unit 48 (the "second communication unit" of this invention) and a learning model generation unit 50. Audio data is automatically received via the communication unit 48 of the machine learning device 40. The machine learning device 40 can be positioned away from the fryer 20. Alternatively, the detection device 30 and the machine learning device 40 can be integrated into a single system.

[0072] The learning model generation unit 50 includes an index acquisition unit 51, a memory unit 52, and a calibration curve generation unit 53. The index acquisition unit 51 acquires index data related to the deterioration of frying oil from the received sound data and stores the index data in the memory unit 52. The calibration curve generation unit 53 performs so-called teacher-guided learning, generating a calibration curve (model) from the stored index data (explanatory variables) through linear regression analysis.

[0073] The types of linear regression (analysis) include simple regression, multiple regression, partial least squares (PLS) regression, orthogonal partial least squares (OPLS) regression, etc., but one or more of these can be used.

[0074] A single regression system is a method that predicts a single objective variable using a single explanatory variable, while a multiple regression system is a method that predicts a single objective variable using multiple explanatory variables. Furthermore, the (orthogonal projective) partial least squares regression system is a method that extracts principal components by maximizing the covariance between the principal components (which can be obtained solely through principal component analysis of the explanatory variables) and the objective variable, which are among the few characteristic quantities. Moreover, the (orthogonal projective) partial least squares regression system is suitable when the number of explanatory variables exceeds the number of samples, and when the correlation between the explanatory variables is high.

[0075] Figures 5A and 5B show the relationship between the calibration curve obtained by machine learning and the heating time (predicted value and measured value) of the test data.

[0076] The straight line M1 in Figure 5A is the test line (model formula) obtained by single regression analysis using the frequency mean (f_mean). The horizontal axis of the graph represents the predicted value of heating time [h], and the vertical axis represents the measured value of heating time [h]. The "○" symbol in the figure is a plot of the predicted value obtained from the frequency mean (f_mean).

[0077] Figure 5B shows the measured heating time (frying time), the average of the 5 predicted values, and the standard deviation. For example, the average predicted value relative to the measured heating time of 8 [h] is 8.9 [h], and the standard deviation is 1.4. The predicted values ​​are roughly around the straight line M1 (see Figure 5A), and the degree of unevenness is also small. Therefore, the calibration curve obtained by single regression analysis confirms that it has a certain degree of accuracy.

[0078] Figures 6A and 6B show the relationship between the calibration curve obtained by machine learning and the acid value (predicted value and measured value) of the test data.

[0079] The straight line M2 in Figure 6A is the check line (model formula) obtained by repeated regression analysis using the frequency mean (f_mean) and the flatness (f_sfm) obtained from the spectrum. The horizontal axis of the graph represents the predicted value of acid value, and the vertical axis represents the measured value of acid value. The "○" symbol in the graph is a plot of the predicted values ​​of acid value obtained from the frequency mean (f_mean) and the flatness (f_sfm).

[0080] Figure 6B shows the measured values ​​of heating time, acid value, average of 5 predicted values, and standard deviation. For example, the measured acid value relative to the measured heating time of 8 [h] is 0.16, the average predicted value is 0.11, and the standard deviation is 0.10. Some predicted acid values ​​also exist on the straight line M2 (see Figure 6A). Due to the small degree of heterogeneity, it is confirmed that the accuracy of the calibration curve obtained by repeated regression analysis is high.

[0081] Figures 7A and 7B show the relationship between the calibration curve obtained by machine learning and the heating time (predicted value and measured value) of the test data.

[0082] The straight line M3 in Figure 7A is the test line (model formula) obtained by orthogonal projective partial least square regression (OPLS) analysis. The horizontal axis of the graph represents the predicted value of heating time [h], and the vertical axis represents the measured value of heating time [h]. The "○" symbol in the figure is a plot of the predicted value of heating time obtained from the frequency average (f_mean).

[0083] Figure 7B shows a view of the measured heating time (frying time), the average of the 5 predicted values, and the standard deviation. For example, the average predicted value relative to the measured heating time of 8 [h] is 9.0 [h], and the standard deviation is 1.8. Some predicted values ​​also exist on the straight line M3 (see Figure 7A), and the degree of non-uniformity is relatively small. Therefore, the accuracy of the calibration curve obtained by the orthogonal projective partial least square regression analysis is also confirmed.

[0084] Figures 8A and 8B show the relationship between the calibration curve obtained by machine learning and the acid value (predicted and measured values) of the test data. Here, "acid value" refers to the value determined by the standard oil and fat analysis method 2.3.1-2013.

[0085] The straight line M4 in Figure 8A is the test line (model form) obtained by orthogonal projective partial least square regression (OPLS) analysis. The horizontal axis of the graph represents the predicted value of acid price, and the vertical axis represents the measured value of acid price. The "○" symbol in the graph is a plot of the predicted value of acid price obtained from index data such as frequency average (f_mean).

[0086] Figure 8B shows the measured values ​​of heating time, acid value, average of 5 predicted values, and standard deviation. For example, the measured acid value relative to the measured heating time of 8 [h] is 0.16 [h], the average predicted value is 0.13, and the standard deviation is 0.12. The unevenness of the predicted acid value is small, thus confirming the high accuracy of the calibration curve obtained by orthogonal projective partial least squares regression analysis.

[0087] Figures 9A and 9B show the relationship between the calibration curve obtained by machine learning and the color (predicted value and measured value) of the test data. The "color" referred to here is the hue of the frying oil, which is "Y+10R" determined by the standard oil analysis method 2.2.1.1-1996.

[0088] The straight line M5 in Figure 9A is the test line (model) obtained by partial least squares regression (PLS) analysis. The horizontal axis of the graph represents the predicted value of color, and the vertical axis represents the measured value of color. The "○" symbol in the graph is a plot of the predicted value of color obtained from index data such as frequency average (f_mean).

[0089] Figure 9B shows the measured values ​​of heating time and color, the average of the five predicted values, and the standard deviation. For example, the measured value of color is 6.5 relative to the measured value of heating time 8 [h], the average predicted value is 6.9, and the standard deviation is 1.6. The unevenness of the predicted color values ​​is relatively small, thus confirming that the calibration curve obtained by partial least squares regression analysis has a certain degree of accuracy.

[0090] Figures 10A and 10B show the relationship between the calibration curve obtained through machine learning and the viscosity increase rate (predicted value and measured value) of the test data. Here, "viscosity" refers to the value of the degree of stickiness (viscosity) of frying oil measured by a commercially available viscometer, such as the E-type viscometer (TVE-25H: manufactured by Toki Sangyo Co., Ltd.). In this study, the viscosity increase rate [%] relative to heating time was investigated.

[0091] If the viscosity of the frying oil at the time of initial use (viscosity at the beginning of use) is taken as Vs, the viscosity of the frying oil will increase as it continues to deteriorate due to repeated frying of fried foods. If the viscosity after the start of use is taken as Vt, the "viscosity increase rate" is defined as the ratio of the increase in viscosity relative to Vs (=Vt-Vs).

[0092] The straight line M6 in Figure 10A is the calibration curve (model formula) obtained by partial least squares regression (PLS) analysis. Furthermore, the horizontal axis represents the predicted value of viscosity increase rate [%], and the vertical axis represents the measured value of viscosity increase rate [%]. The "○" symbol in the figure is a plot of the predicted value of viscosity increase rate obtained from index data such as frequency average (f_mean).

[0093] Figure 10B shows the measured values ​​of heating time and viscosity increase rate, the average of the five predicted values, and the standard deviation. For example, the measured viscosity increase rate relative to the measured heating time of 8 [h] is 3.52, the average predicted value is 3.87, and the standard deviation is 0.57. The predicted viscosity increase rate obtained by plotting the figure shows a small degree of non-uniformity, thus confirming that the measurement system obtained by partial least squares regression analysis has high accuracy.

[0094] Thus, the calibration curve generation unit 53 generates the calibration curve from the index data through linear regression analysis. However, the linear regression can utilize any of the following: single regression, multiple regression, partial least squares (PLS) regression, or orthogonal projective partial least squares (OPLS) regression. The actual generated calibration curve is based on the evaluation results of the acid value, color, viscosity increase rate, etc. of the frying oil as it changes with heating time. The accuracy of the degree of deterioration is high, and the deterioration of the frying oil can be accurately predicted and determined from the relevant sound index data.

[0095] In the degradation prediction system 100 in Figure 4, the machine learning device 40 may be set in a remote location far away from the store, and the detection device 30 is a detection server and the machine learning device 40 is a machine learning server.

[0096] At this time, the detection server on the store side shall have at least: a sound data acquisition unit that acquires sound data when frying food; a communication unit that transmits / receives various data (sound data, judgment results, etc.) with the machine learning server; and a notification unit that notifies the degree of deterioration of the frying oil and the timing of replacement based on the judgment results.

[0097] Furthermore, the remote machine learning server system shall at least include: a communication unit for transmitting / receiving various data with the detection server; a learning model generation unit for extracting indicators related to the deterioration of frying oil from the received sound data, performing linear regression machine learning using the indicators, and generating a learning model that can determine the deterioration of frying oil; a memory unit for storing the generated learning model; and a determination unit for determining the degree of deterioration of frying oil using the learning model.

[0098] According to this configuration, on the machine learning server side, the learning model generation unit performs machine learning using received sound data to generate a learning model. Next, the determination unit uses the learning model to determine the degree of deterioration of the frying oil and transmits the determination result to the detection server side. On the detection server side, the notification unit notifies the frying oil of the replacement time based on the received determination result. In this way, the tasks can be divided in such a manner: receiving sound data on the machine learning server side, performing the determination until the result is returned to the detection server.

[0099] The learning model is generated by a machine learning server, for example, and is updated each time new sound data is acquired. This eliminates the need for a learning model that transmits / receives large amounts of data, allowing a shop with a detection server to determine when to change the frying oil.

[0100] [Third Implementation Type]

[0100] Next, referring to FIG11, an outline of the grease replacement system 200 of the third embodiment of the present invention will be described.

[0101] Figure 11 is a schematic diagram of the grease replacement system 200. As shown, the grease replacement system 200 consists of shops A to C equipped with a deterioration prediction device 1 and fryers 20', a general management headquarters H that coordinates shops A to C, a manufacturer (grease manufacturer) X that manufactures the frying oil used in shops A to C, a retailer (wholesaler or retailer) Y, and a waste oil recycling company Z. Sometimes, the grease manufacturer sells directly to customers; therefore, the retailer Y is a concept that includes the grease manufacturer.

[0102] In the first embodiment, the notification unit 7 of the degradation prediction device 1 notifies the user of the main points via a speaker, display unit 5, etc., when the degradation level of the frying oil is determined to exceed a predetermined threshold. However, in this embodiment, in addition to such notification, notification information regarding the degradation level of the frying oil is also output. The notification information may be the content indicating that the degradation level of the frying oil exceeds the threshold, or it may be a warning that the degradation level is about to exceed the threshold.

[0103] As shown in the figure, when the notification information is sent from store B (Izakaya) to the coordination headquarters H, the coordination headquarters H analyzes the number or frequency of the notification information received. Not only store B, but also store A (tempura store) and store C (pork cutlet store) can be advised or guided on whether the frying oil is used properly, whether it should be replaced appropriately, or whether it will cause waste, as needed.

[0104] The General Coordination Headquarters H System can manage not only multiple stores, but also multiple factories equipped with fryers. Furthermore, the General Coordination Headquarters H System can exist within a store or factory and manage multiple fryers within the facility.

[0105] This notification information was also sent to the frying oil manufacturer X and the retailer Y. Manufacturer X received the notification information and established a frying oil manufacturing or sales plan. Retailer Y received the notification information and ordered new frying oil, purchasing frying oil P from manufacturer X. Retailer Y then allocated the new frying oil P to shop B (and, as needed, shops A and C).

[0106] Furthermore, the notification information is sent to the oil recycling operator Z (or manufacturer X). The recycling operator Z receives the notification information and arranges for the collection of waste oil Q. For example, when the recycling operator Z receives the notification information a predetermined number of times, it visits shop B and collects waste oil Q from the oil tank 22 of the fryer 20'.

[0107] Furthermore, this notification information can be sent to the cleaning service provider (illustration omitted). Upon receiving the notification information, the cleaning service provider visits shop B to clean the inside or vicinity of the oil tank 22 of the fryer 20'. In this way, the grease replacement system 200 can quickly supply frying oil to shops A through C, handle waste disposal, and clean the area.

[0108] Furthermore, according to the notification, automating the replacement of frying oil in the store would further reduce the burden on users (store staff). If a notification message is output indicating that the deterioration level of the frying oil exceeds a threshold, the replacement of the frying oil will begin automatically.

[0109] [Fourth Implementation Type]

[0109] Finally, referring to FIG12, an outline of the fryer system 300 of the fourth embodiment of the present invention will be described.

[0110] Figure 12 shows the deterioration prediction device 1 and the fryer 20' constituting the fryer system 300 of this embodiment. Furthermore, for the fryer 20', the same reference numerals are given to components that are the same as those of the fryer 20 of the first embodiment, and the description is omitted.

[0111] As shown in the figure, a valve control device 61 (the "valve control unit" of the present invention) and a new oil tank 62 are provided near the fryer 20'. The new oil tank 62 contains unused frying oil, and frying oil is supplied to the oil tank 22 via the oil supply pipe 63.

[0112] If the valve control device 61 receives notification information from the deterioration prediction device 1 indicating that the frying oil needs to be replaced (if the threshold is exceeded), it first sends a control signal to the valve 24' to open the valve 24'. Thereby, the waste oil is automatically discharged into the waste oil tank 26 via the drain pipe 25.

[0113] After a sufficient period of time, the valve control device 61 sends a control signal to valve 24' again to close valve 24'. Then, the valve control device 61 sends a control signal to valve 64, which is located midway in the oil supply pipe 63, to open valve 64. In this way, new oil is automatically supplied to the oil tank 22. Furthermore, the amount of new oil supplied can be detected by the water level sensor in the new oil tank 62, or valve 64 can be opened only for a predetermined time.

[0114] According to this embodiment, the valve control device 61 controls valves 24' and 64 based on notification information transmitted from the deterioration prediction device 1, thereby automatically discharging the frying oil in use. Furthermore, the valve control device 61 controls the automatic supply of new oil from the new oil tank 62, thereby allowing the user to determine the degree of deterioration of the frying oil and set it as waste oil until new oil is supplied, thus reducing the workload.

[0115] The above-described deterioration prediction device, deterioration prediction system, and grease replacement system are merely examples of embodiments of the present invention and can be appropriately modified according to their uses, purposes, etc. This example illustrates a regression analysis performed by extracting the frequency mean (f_mean) and the flatness (f_sfm) obtained from the spectrum of sound data; however, frequency standard deviation (f_sd), dominant frequency (dfnum), etc., can also be applied to the deterioration prediction of frying oil.

[0116] Furthermore, in the degradation prediction system, the tasks performed by each component device can be changed. In the degradation prediction system 100 shown in Figure 4, the detection device 30 and the machine learning device 40 are separate devices. However, the amount of data in the sound data or learning model is large, and communication is time-consuming and expensive. Therefore, a new control device can be installed to remotely instruct the detection device containing the machine learning unit and to receive notification information such as the degree of degradation of the frying oil from the detection device.

Claims

1. A degradation prediction device for predicting the degree of degradation of edible oils, the degradation prediction device comprising: The sound data acquisition unit acquires sound data when cooking fried food using the aforementioned oil contained in the oil tank; The indicator extraction unit extracts indicators related to the deterioration of the aforementioned grease from the sound data obtained by the aforementioned sound data acquisition unit; and The determination unit determines the degree of deterioration of the aforementioned oil based on the aforementioned indicators captured by the aforementioned indicator acquisition unit.

2. The degradation prediction device as described in claim 1 further includes a notification unit that notifies the degree of degradation of the aforementioned grease or the timing of replacing the aforementioned grease. The aforementioned notification is issued when the aforementioned determination department determines, based on the degree of deterioration of the aforementioned grease, that it exceeds the pre-defined replacement threshold.

3. The degradation prediction device as described in claim 1 or 2, wherein, The aforementioned indicators are selected from one or more of the following: frequency average, frequency standard deviation, frequency median, frequency standard error, frequency maximum frequency, frequency first quartile, frequency third quartile, frequency quartile range, frequency centroid, frequency skewness, frequency kurtosis, spectral flatness, spectral entropy, spectral accuracy, sound complexity index, sound entropy, and dominant frequency.

4. A degradation prediction system, comprising a detection device and a machine learning device, for predicting the degree of degradation of edible oils, wherein... The aforementioned detection device is equipped with: The sound data acquisition unit acquires sound data when cooking fried food using the aforementioned oil contained in the oil tank; The memory unit stores the learning model generated by the aforementioned machine learning device, which can determine the deterioration of the aforementioned grease; and The determination unit uses the aforementioned learning model to determine the degree of deterioration of the aforementioned grease from the aforementioned sound data; The aforementioned machine learning device has the following features: The learning model generation unit extracts indicators related to the deterioration of the aforementioned oil from the aforementioned sound data obtained by the aforementioned sound data acquisition unit, and uses the aforementioned indicators to perform linear regression machine learning to generate the aforementioned learning model.

5. The degradation prediction system as described in claim 4, wherein, The aforementioned linear regression system is selected from one or more of the following: simple regression, multiple regression, partial least squares (PLS) regression, or orthogonal projective partial least squares (OPLS) regression.

6. The degradation prediction system as described in claim 4 or 5, wherein, The aforementioned detection device and the aforementioned machine learning device are integrated into one unit.

7. The degradation prediction system as described in claim 4 or 5, wherein, The aforementioned detection device is located near the aforementioned oil tank in the shop or factory, and the aforementioned machine learning device is located at a distance from the aforementioned shop or factory.

8. The degradation prediction system as described in claim 4 or 5, wherein, The aforementioned detection device is equipped with a first communication unit, which transmits the aforementioned sound data obtained by the aforementioned sound data acquisition unit to the aforementioned machine learning device; The aforementioned machine learning device is equipped with a second communication unit that receives the aforementioned sound data from the aforementioned detection device.

9. The degradation prediction system as described in claim 8, wherein, The aforementioned first communication unit and the aforementioned second communication unit can perform wireless communication.

10. A degradation prediction method for predicting the degree of degradation of edible oils, the degradation prediction method comprising: The sound data acquisition step involves acquiring sound data when using the aforementioned oil to cook fried food; The indicator extraction step involves extracting indicators related to the deterioration of the aforementioned oils from the sound data obtained in the aforementioned sound data acquisition step; and The determination step is based on the aforementioned indicators obtained in the aforementioned indicator acquisition step to determine the degree of deterioration of the aforementioned oil.

11. A grease replacement system, which performs one or more of the following a) to e) based on notification information related to the degree of deterioration of the aforementioned grease output from the deterioration prediction device described in claim 1 or 2; a) Notify oil and grease vendors to order new oil and grease; b) Notify oil and fat manufacturers to formulate oil and fat manufacturing or sales plans; c) Notify the coordinating headquarters of the store or factory, or the oil manufacturer, and provide suggestions or guidance on the use of oil to the stores or factories under its coordination; d) Notify waste oil recycling operators or oil manufacturers to arrange for the recycling of waste oil; e) Notify the cleaning service provider to arrange for the cleaning of the oil tank.

12. A deep fryer system comprising a valve control unit that controls a valve located in an oil tank based on notification information related to the degree of deterioration of the aforementioned oil, output from the deterioration prediction device described in claim 1 or 2. The aforementioned valve control unit automatically disposes of the aforementioned grease contained in the aforementioned oil tank.

13. The fryer system as described in claim 12, wherein, The aforementioned valve control unit automatically supplies new oil to the aforementioned oil tank.

Images