A temperature control method for a microwave oven based on a temperature difference discrimination kernel

Abstract

The application discloses a microwave oven temperature control method based on temperature difference distinguishing kernel and relates to the technical field of microwave cooking.The microwave oven temperature control method based on temperature difference distinguishing kernel provided by the application firstly obtains the temperatures at two time points (time point t1 and time point t2) by a non-contact FOV sensor in an induction stage, distinguishes single quantity and multiple quantities (double quantity and triple quantity) by calculating the temperature change rates at the two time points, then distinguishes the multiple quantities of food into double quantity and triple quantity based on a calculated similarity distinguishing kernel, and finally calculates the temperature change values between different adjacent time points to obtain the remaining time.The application can identify the types and sizes of food materials by the method of calculating the temperature difference distinguishing kernel based on the temperature data obtained by the non-contact sensor, further determine the heating time, realize the optimization of cooking effect, and improve the user experience.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of microwave cooking, and relates to a novel microwave oven temperature control method based on temperature difference resolution kernels, in particular to a temperature control method based on resolution kernels, multiple portion differentiation and adaptive dishes. BACKGROUND

[0002] Compared with traditional cooking methods, microwave ovens have the advantages of fast heating speed, simple operation, low energy consumption, etc., and can effectively shorten food cooking time and retain nutritional ingredients. Especially under the fast pace of life, microwave ovens have become a standard configuration of household kitchen equipment. However, despite the significant advantages of microwave heating, existing microwave ovens still have many technical bottlenecks in terms of heating uniformity and intelligent temperature control effect, and need to be improved. At present, microwave ovens are usually equipped with non-contact temperature sensors for real-time monitoring of the temperature in the oven cavity. However, such sensors have the following disadvantages during microwave heating: the water vapor released by the food during the heating process changes the radiation characteristics of the surface of the target food, causing measurement deviation of the non-contact temperature sensor. Non-contact sensors usually measure the temperature at a specific fixed position and cannot obtain the temperature distribution of the entire food, which can easily cause local overheating or undercooking. The sensor temperature measurement result is significantly affected by the material, shape and surface characteristics of the container, and the measurement accuracy is therefore limited. Due to the above problems, the existing temperature control algorithm based on non-contact sensor data has difficulties in identifying the food type and portion, and the heating effect is poor, especially when dealing with complex food categories and irregularly shaped objects, the effect of intelligent temperature control is limited.

[0003] To address these challenges, in recent years, the research of control theory and intelligent algorithms has provided new solutions for microwave oven temperature control technology. In particular, in the aspect of discrimination enhancement, through the combination of multi-sensor fusion and intelligent algorithms, it is expected to break through the current technical bottlenecks: by combining infrared thermal imaging technology and array temperature sensors, high-precision detection of food surface and internal temperature distribution can be realized, thereby providing more comprehensive data support for temperature control algorithms. Through machine learning and data fusion technology, multi-source temperature data is converted into a high-resolution three-dimensional temperature distribution map, providing support for real-time dynamic adjustment during the subsequent heating process. Research based on physical property correction algorithms compensates for temperature measurement deviations caused by container material and surface characteristics, further improving temperature control accuracy. In addition, machine learning as an advanced intelligent control method, machine learning algorithms can automatically extract rules from a large amount of data without relying on artificial definition of rules and fuzzy sets like fuzzy control. For complex food temperature changes and nonlinear heating characteristics, machine learning can build accurate mathematical models through data-driven methods, significantly improving prediction accuracy and control effect. In summary, the combination of resolution enhancement technology and machine learning intelligent algorithms will be the key research direction of future microwave oven intelligent temperature control. This technology can improve food temperature measurement accuracy and achieve intelligent food recognition and temperature control optimization, which is expected to significantly improve the user experience and market competitiveness of microwave ovens.

[0004] Patent CN 106482180B proposes a semiconductor microwave oven and its heating control method and device, aiming to achieve uniform heating of food by adjusting the working parameters of the microwave source. The method first acquires the initial phase and initial frequency of the microwave source, and uses them as the starting parameters to start the heating process. Subsequently, by randomly changing the characteristics of the microwave field, it effectively reduces the local overheating or undercooked phenomenon caused by standing wave effects in traditional microwave ovens. This random phase and frequency adjustment method is suitable for a variety of food types, significantly improving the adaptability to complex heating requirements. However, although this method performs well in improving heating uniformity, it may still be difficult to flexibly adjust the overall heating power in microwave ovens with fixed power. In addition, the results of the random algorithm have certain uncertainty, which may affect the control accuracy.

[0005] Patent CN 117135787A proposes a precise control heating algorithm for frozen food based on a commercial microwave oven. Through precise segmented heating control and dynamic adjustment of key parameters, efficient and uniform heating of frozen food is achieved. The method first scans the heating code of the food, obtains the precise heating algorithm data matched with the specific food, and provides support for subsequent operations. Then, the entire heating process is divided into multiple time periods, and the heating task of each stage is executed step by step to achieve dynamic adjustment and fine control. Through the combination of food heating code and algorithm data, this method provides a customized heating scheme for different frozen foods, effectively improving the heating effect. At the same time, by dynamically adjusting the microwave power and stirring speed, the problem of uneven heating is further reduced, especially suitable for cases where there is a significant temperature gradient inside the frozen food. However, for foods without matching heating codes, the heating accuracy may be limited. In addition, the design of multi-variable joint regulation and segmented heating puts high requirements on hardware performance and algorithm efficiency.

[0006] Therefore, it is of great practical significance to develop a temperature control method that accurately distinguishes multiple portions of food and accurately identifies the quantity of food. SUMMARY

[0007] The purpose of the present application is to solve the problem of how to intelligently distinguish multiple portions during the heating process of a microwave oven, and to overcome the difficulty of distinguishing multiple portions based on non-contact infrared temperature sensor data. The present application uses machine learning to obtain the characteristics of the target portion curve (such as double or triple portions), and then calculates the distance between the features of the portion curve to be identified (i.e. the unknown portion placed in the microwave oven) and the target portion features in the feature space to distinguish the portion.

[0008] To achieve the above purpose, the present application adopts the following technical solutions:

[0009] A microwave oven temperature control method based on temperature difference resolution kernel, including the following contents:

[0010] Collect the surface temperature of the food to be identified by a non-contact temperature sensor;

[0011] Calculate the temperature change rate of the food between the initial time and a certain time during the heating process, and use the change rate to distinguish single and multiple portions;

[0012] Use the equal interval difference method to extract the feature vector of the temperature curve, and determine the temperature difference resolution parameter through machine learning method;

[0013] Calculate the feature vector of the target portion curve by averaging a large amount of experimental data, which is called the resolution kernel;

[0014] Calculate the Euclidean distance between the to-be-resolved portion feature vector and the resolution kernel in the feature space, and distinguish the food portion by the size of the Euclidean distance.

[0015] Preferably, the formula for calculating the temperature change rate is:

[0016]

[0017] Wherein, α represents the temperature change rate; Respectively represent the temperature at t1, t2 moment;

[0018] The standard change rate is K, when α≥K, then judge the food portion is single; Otherwise, it is judged that the food portion is multiple.

[0019] Preferably, the function of the resolution kernel is:

[0020]

[0021] Wherein, Φ represents the resolution kernel; δ represents the feature vector of the temperature curve, δ i The difference between the i-th time interval; the superscript T Indicates the transpose operation.

[0022] Preferably, the formula for calculating the Euclidean distance is:

[0023]

[0024] Wherein, δ u Indicates the to-be-resolved portion feature vector; d indicates the Euclidean distance between the to-be-resolved portion feature vector δ u And the resolution kernel Φ.

[0025] Preferably, the food portion is distinguished by the size of the Euclidean distance, which specifically includes the following contents:

[0026]

[0027] Wherein, x represents the food portion; d0 represents the distance threshold;

[0028] For food categories with large initial temperature difference, add a regularization term related to the initial temperature:

[0029]

[0030] Wherein, η represents the weight factor, 0<η<1; Indicates the average initial temperature of the target portion; t0 represents the initial temperature of the to-be-resolved portion feature vector; and then:

[0031]

[0032] D0, wherein D represents the Euclidean distance between the to-be-resolved portion feature vector of the food with a large initial temperature difference and the resolving kernel, and D0 represents a threshold value of D.

[0033] Preferably, the method takes the temperature change rate as input, realizes temperature compensation by calculating the relationship between the actual temperature and the target temperature, and specifically includes the following contents:

[0034] (1) Calculate the temperature change value between adjacent time points:

[0035] ΔT1 = T' t+1 -T' t

[0036] ΔT2 = T' t+2 -T' t+1

[0037] wherein ΔT1 is the temperature change between the i-th time point and the i+1-th time point; and ΔT2 is the temperature change between the i+1-th time point and the i+2-th time point.

[0038] (2) Calculate the temperature change ratio:

[0039]

[0040] wherein β is the temperature change ratio.

[0041] (3) According to the temperature change value calculated in step (1) and the temperature change ratio calculated in step (2), and in combination with a pre-set remaining time calculation model, predict the remaining time:

[0042] When α ≥ a, the remaining time prediction formula is:

[0043]

[0044] wherein t estimate_left_time1 represents the predicted remaining time when α ≥ a; α represents the temperature change rate; and a represents a pre-set temperature change rate threshold value.

[0045] When α < a and d ≤ γ, the remaining time prediction formula is:

[0046]

[0047] wherein γ represents a pre-set Euclidean distance threshold value; and t estimate_left_time2 represents the predicted remaining time when α < a. represents the real-time temperature collected by the microwave oven-mounted temperature sensor at t4 when α < a and d ≤ γ.

[0048] When α γ, the remaining time prediction formula is:

[0049]

[0050] wherein, t estimate_left_time3 represents the predicted remaining time when alpha is less than a; represents the real-time temperature collected by the microwave oven-mounted temperature sensor at the t5 moment when alpha is less than a and d is greater than gamma;

[0051] (4) The intermediate calibration temperature corresponding to different temperature change rates alpha is compensated, and the compensation is as follows:

[0052] When alpha is greater than or equal to a, the intermediate calibration temperature The calculation formula of the compensation time of the intermediate calibration temperature

[0053]

[0054] wherein, t award_time1 represents the predicted remaining time when alpha is greater than or equal to a;

[0055] When alpha is less than a and d is less than or equal to gamma, the intermediate calibration temperature The calculation formula of the compensation time of the intermediate calibration temperature

[0056]

[0057] wherein, t award_time2 represents the predicted remaining time when alpha is greater than or equal to a;

[0058] When alpha is less than a and d is greater than gamma, the intermediate calibration temperature The calculation formula of the compensation time of the intermediate calibration temperature

[0059]

[0060] wherein, t award_time3 represents the predicted remaining time when alpha is greater than or equal to a;

[0061] (5) The final heating time is obtained by comprehensively considering the contents of steps (1)-(4):

[0062] t left_time =t estimate_left_time +t award_time

[0063] wherein, t estimate_left_time is the final predicted remaining time, and t award_time is the final compensation time.

[0064] Beneficial effects:

[0065] (1) The microwave oven temperature control method based on the temperature difference resolution core of the present application can distinguish single serving and multiple servings by comparing the temperature changes at different moments.

[0066] (2) The microwave oven temperature control method based on the temperature difference resolution kernel of the application, by calculating the temperature difference resolution kernel, by introducing the distance calculation rule, the double and triple portions of the food are effectively classified, and the subsequent heating time is determined.

[0067] (3) The fuzzy rule microwave oven temperature control method based on the analog heat capacity calculation of the application introduces the temperature difference resolution kernel as a variable, converts the real-time temperature data obtained by the non-contact FOV sensor into heating time, and finally realizes the control of the microwave oven on the heating of the food. BRIEF DESCRIPTION OF DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, a brief introduction will now be given to the drawings involved in the embodiments. Obviously, the drawings in the following description are only schematic illustrations of some embodiments of the application, and based on these drawings, other forms of drawings can also be constructed by those skilled in the art without creative efforts.

[0069] Figure 1 The flow chart of the heating control method of the microwave oven in embodiment 1 of the application;

[0070] Figure 2 The basic block diagram of the fuzzy control theory in embodiment 1 of the application;

[0071] Figure 3 The flow chart of the microwave oven one portion distinguishing algorithm in embodiment 1 of the application;

[0072] Figure 4 The flow chart of the two and three portion distinguishing algorithm based on the resolution kernel in embodiment 1 of the application;

[0073] Figure 5 The flow chart of the multi-portion temperature compensation mechanism in embodiment 1 of the application. DETAILED DESCRIPTION

[0074] The application will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the application and not to limit the scope of the application. In addition, it should be understood that after reading the content taught by the application, those skilled in the art can make various modifications or modifications to the application, and these equivalent forms also fall within the scope defined by the claims attached to the present application. At the same time, it should be pointed out that if there is no conflict, the embodiments of the present application and the various features in the embodiments can be combined with each other, and all fall within the protection scope of the present application. In addition, although the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in an order different from that here.

[0075] The present application relates to a kind of microwave oven temperature control method based on temperature difference resolution kernel, traditional microwave oven carries the temperature control algorithm based on non-contact temperature sensor, rely on the radiation characteristic of target food surface, but due to heating process, food surface can change its radiation characteristic due to water vapor or oil evaporation, cause temperature measurement deviation.Temperature control algorithm usually relies on single-point temperature measurement, it is difficult to dynamically adjust heating strategy in real time to cope with rapidly changing temperature distribution, especially when heating complex structure or non-uniform object is weak.In order to solve the problem of single-point temperature measurement and realize the intelligent distinction of multiple food portions, the present application proposes a new type of temperature control algorithm based on the idea of temperature difference resolution kernel, a portion is distinguished using a period of time before starting heating, then the characteristics of target portion curve (such as two portions or three portions) are obtained by statistical or machine learning method, then the difference between the characteristics of to-be-identified portion curve (unknown portion placed in microwave oven) and target portion is calculated to distinguish the portion.Based on the above content, the microwave oven temperature control method based on temperature difference resolution kernel proposed by the present application will be described below in combination with related drawings and specific examples, and the specific content is as follows.

[0076] Example 1:

[0077] Please refer to Figures 1-5 The present application proposes a microwave oven temperature control method based on temperature difference resolution kernel, and the specific steps are as follows:

[0078] (1) At the beginning of heating, use non-contact FOV temperature sensor to record initial temperature Temperature is expressed in degrees Celsius (℃).

[0079] (2) Emit microwaves to microwave oven cavity at 600W power, record t2 temperature when heated to t2 Temperature is expressed in degrees Celsius (℃).

[0080] (3) The elimination theory of installation error is as follows, the temperature data collected by the standardized equipment without installation error can be recorded as T, the temperature measurement error generated by installation can be recorded as μ, temperature is expressed in degrees Celsius (℃); When calculating the specific heat capacity change rate α:

[0081]

[0082] Based on the above formula, calculate the temperature change rate α, and calibrate the standard change rate K, when α≥X, then judge the food portion as single portion, otherwise, judge the food portion as multiple portions, at this time, temperature difference resolution kernel can be used to distinguish;

[0083] (4) The formula for calculating portion feature vector δ, temperature difference resolution kernel Φ and distance d in feature space is as follows:

[0084]

[0085] wherein d is the unknown serving feature vector δ u The distance between the resolution kernel Φ and the unknown serving feature vector δ in the feature space, the serving feature vector δ is obtained by using the equal interval difference method on the temperature curve, and each dimension component δ k denotes the temperature difference value of the kth time interval.

[0086] The distance d between the resolution kernel and the unknown serving feature vector in the feature space is calculated based on the above formula, when d≤d0, the serving is determined as two servings, otherwise it is determined as three servings;

[0087] (5) Calculate the temperature change value between adjacent time points:

[0088] ΔT1=T' t+1 -T' t

[0089] ΔT2=T' t+2 -T' t+1

[0090] wherein ΔT1 is the temperature change between the ith time point and the ith+1 time point; ΔT2 is the temperature change between the ith+1 time point and the ith+2 time point;

[0091] (6) Further calculate the temperature change value:

[0092]

[0093] wherein β is the temperature change ratio.

[0094] (7) According to the calculated temperature change rate and ratio, combined with the pre-set remaining time calculation model, predict the remaining time:

[0095] When α≥a, use the formula to predict the remaining time, wherein t estimate_left_time1 denotes the predicted remaining time when α≥a;

[0096] When α to predict the remaining time, wherein t estimate_left_time2 denotes the predicted remaining time when α denotes the real-time temperature collected by the microwave oven-mounted temperature sensor at the t4 moment when α

[0097] When α to predict the remaining time, wherein t estimate_left_time3 denotes the predicted remaining time when α denotes the real-time temperature collected by the microwave oven-mounted temperature sensor at the t5 moment when α

[0098] (8) By compensating the intermediate calibration temperature corresponding to different temperature change rates a, specifically as follows:

[0099] When a >= a, the compensation time of the intermediate calibration temperature at this time is Wherein, t award_time1 represents the predicted remaining time when a >= a; When a < a and d <= g, the compensation time of the intermediate calibration temperature at this time is

[0100] Wherein, t award_time2 represents the predicted remaining time when a >= a; When a g, the compensation time of the intermediate calibration temperature at this time is Wherein, t award_time3 represents the predicted remaining time when a >= a;

[0101] (9) The final heating time is t left_time = t estimate_left_time + t award_time , wherein t estimate_left_time is the final predicted remaining time, and t award_time is the final compensation time.

[0102] The microwave oven temperature control method based on the temperature difference resolution kernel of the application can effectively classify the heating food material quantity under the premise of not increasing the number of sensors, and realize the elimination of installation errors, which not only helps to improve the heating effect and energy efficiency, but also meets the diversified heating needs, promotes the development of the home appliance industry to the intelligent and fine direction, brings more convenient and high-quality use experience to users, and has significant environmental protection and economic benefits.

[0103] In the remaining technical features in the embodiment, those skilled in the art can flexibly select them according to actual conditions to meet different specific actual needs. However, it is obvious to those skilled in the art that the specific details do not have to be used to implement the application. In other examples, in order to avoid confusion of the application, the composition, structure or component of the formula is not specifically described, and is within the technical solution limited by the claims of the application.

[0104] In the remaining technical features in the embodiment, those skilled in the art can flexibly select them according to actual conditions to meet different specific actual needs. However, it is obvious to those skilled in the art that the specific details do not have to be used to implement the application. In other examples, in order to avoid confusion of the application, the composition, structure or component of the formula is not specifically described, and is within the technical solution limited by the claims of the application.

[0105] ​​Modifications and changes can be made to the present application without departing from the spirit and scope of the application. In the above description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the present application can be practiced without these specific details. In other instances, well-known techniques are not described in detail in order to avoid obscuring the present application. The preceding description, as such, is not to be understood in a restrictive sense, and not to be dedicated to the generic field only, but is to be accorded the broadest scope indicative of the principles and technical scope of the present application.

[0106] The principles and implementations of the present application have been described above with the specific examples. The above description is only used to help understand the method and core idea of the present application. For those skilled in the art, according to the idea of the present application, the specific implementation and application range can be changed. In summary, the content of the present description should not be understood as a limitation of the present application.

Claims

1. A microwave oven temperature control method based on a temperature difference resolution kernel, characterized in that, Includes the following: The surface temperature of the food to be identified is collected using a non-contact temperature sensor. Calculate the rate of temperature change of food between the initial moment and a certain moment during the heating process, and use the rate of change to distinguish between single servings and multiple servings; The feature vector of the temperature curve is extracted using the equal interval difference method, and the temperature difference resolution parameter is determined by machine learning. The characteristic vector of the target fraction curve is calculated by averaging a large amount of experimental data, and it is called the resolution kernel. Calculate the Euclidean distance in the feature space between the feature vector of the portion size to be distinguished and the discrimination kernel, and distinguish the portion size of food by the magnitude of the Euclidean distance; wherein, the formula for calculating the Euclidean distance is: in, This represents the feature vector of the component to be distinguished; d Represents the feature vector of the component to be distinguished With distinguishing nuclei The Euclidean distance between them; Distinguishing food portion sizes based on the Euclidean distance specifically includes the following: in, x Indicates food portion size; d 0 indicates the distance threshold; For food types with large initial temperature differences, add a regularization term related to the initial temperature: in, Indicates the weighting factor. ; The average initial temperature representing the target portion; t 0 represents the initial temperature of the feature vector of the component to be distinguished; thus, we obtain: in, D This represents the Euclidean distance between the feature vector of the portion size of food to be distinguished and the resolution kernel, indicating a large initial temperature difference. D 0 represents D The threshold; The method takes the temperature change rate as input and achieves temperature compensation by calculating the relationship between the actual temperature and the target temperature. Specifically, it includes the following: (1) Calculate the temperature change between adjacent time points: in, For the first i Time point and the i Temperature changes between +1 time points; For the first i +1 time point and the i Temperature changes between +2 time points; (2) Calculate the temperature change ratio: in, β This represents the ratio of temperature changes. (3) Based on the temperature change value calculated in step (1) and the temperature change ratio calculated in step (2), and combined with the pre-set remaining time calculation model, predict the remaining time: when α ≥ a When the remaining time is predicted, the formula is: in, Indicates when α ≥ a Predicted remaining time; α Indicates the rate of temperature change; a This indicates the preset temperature change rate threshold. when and When the remaining time is predicted, the formula is: in, γ This represents the preset Euclidean distance threshold; Indicates when Predicted remaining time; Indicates when and Microwave ovens equipped with temperature sensors t Real-time temperature collected at 4 o'clock; when and When the remaining time is predicted, the formula is: in, Indicates when Predicted remaining time; Indicates when and The microwave oven is equipped with a temperature sensor in the first t Real-time temperature collected at 5 o'clock; (4) By analyzing different rates of temperature change The corresponding intermediate calibration temperature is compensated, as follows: when At that time, intermediate calibration temperature The formula for calculating the compensation time is: in, Indicates when Predicted remaining time; when and At that time, intermediate calibration temperature The formula for calculating the compensation time is: in, Indicates when Predicted remaining time; when and At that time, intermediate calibration temperature The formula for calculating the compensation time is: in, Indicates when Predicted remaining time; (5) Combining the contents described in steps (1) to (4), the final heating time is: in, To ultimately predict the remaining time, This is the final compensation period.

2. The microwave oven temperature control method based on a temperature difference resolution kernel according to claim 1, characterized in that, The formula for calculating the rate of temperature change is: in, α Indicates the rate of temperature change; They represent t 1. t Temperature at time 2; The standard change rate is K, when α If the value is greater than or equal to K, then the ingredient quantity is determined to be a single serving; otherwise, the ingredient quantity is determined to be multiple servings.

3. The microwave oven temperature control method based on a temperature difference resolution kernel according to claim 2, characterized in that, The function of the resolution kernel is expressed as: in, Indicates the resolving nucleus; The characteristic vector representing the temperature curve, Indicates the first i Temperature difference over a period of time; superscript T This indicates the transpose operation.

Images