A probe-based digital smart food thermometer with host receiver and wireless transmitter

Abstract

This invention belongs to the field of intelligent manufacturing equipment industry and discloses a probe-based digital intelligent food thermometer with host receiver and wireless transmitter. The thermometer's internal sensor collects the food insertion depth and receives and detects the food properties and ambient temperature via the host receiver. A food temperature deviation coefficient generation model is constructed using the food insertion depth, the detected food properties, the ambient temperature, and the corresponding food temperature deviation coefficient. The model generates multiple food temperature deviation coefficients to correct the temperature values ​​detected by the probe. Furthermore, a generative adversarial network model based on food insertion depth correction is used to expand the data volume of the food temperature deviation coefficient generation model to obtain a more accurate food temperature deviation coefficient. This invention utilizes multiple machine learning models to generate a large amount of data on the features affecting the deviation coefficient, and uses this data to generate a more accurate and efficient temperature correction coefficient, thereby improving the intelligence and accuracy of the food thermometer measurement.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent manufacturing equipment industry, and particularly relates to a probe digital intelligent food thermometer with host receiving and wireless transmission. BACKGROUND

[0002] With the development of science and technology, food thermometers are applied to people's lives and are used for temperature detection of food. At this time, the food thermometer is inserted into the food and used for temperature detection of the food so as to record the temperature of the food. The food thermometer can estimate the corresponding remaining cooking time and has an important influence on subsequent guidance of food storage and other key scene applications.

[0003] However, the existing food thermometer defines the overall temperature of the food through single or single position temperature detection, and fails to well consider the nature of the food and the influence of the environment on the temperature measured by the thermometer for intelligent correction of the temperature. This cannot guarantee the accuracy of temperature measurement, and further affects the temperature control of the food thermometer on the food and the intelligent cooking of the food, and has an adverse effect on the storage of the food.

[0004] Therefore, the existing technology has the following problems: the sensor in the thermometer collects which related optimal parameters, how to use the optimal parameters and the corresponding food temperature deviation coefficient to construct a food temperature deviation coefficient generation model, and how to use the food temperature deviation coefficient generated by the model to correct the temperature value detected by the probe. In the case of insufficient food temperature measurement correction data, how to use the scene adaptive data generation model to expand the data of the food temperature deviation coefficient generation model to obtain more accurate food temperature deviation coefficients, and use this type of generated data to generate more accurate and efficient temperature correction coefficients, how to optimize the generated data to improve the classification accuracy and robustness of the constructed food temperature deviation coefficient model, and improve the intelligence and accuracy of the food thermometer measurement. SUMMARY

[0005] To solve the above technical problems, the present application provides a probe digital intelligent food thermometer with host receiving and wireless transmission.

[0006] In a first aspect of the present application, a probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission is provided, comprising a sensor depth acquisition module, a host receiving module, a temperature difference coefficient correction feature generation module, a food temperature deviation coefficient generation model construction module and an intelligent food temperature correction module:

[0007] The sensor depth acquisition module is used for acquiring the insertion depth of the food;

[0008] The host receiving module is used to receive the food insertion depth, as well as to receive the detected food properties and ambient temperature, and to obtain the food temperature deviation coefficient set by the staff based on experience. It processes the food insertion depth, the detected food properties and the ambient temperature to obtain the first temperature difference coefficient correction feature.

[0009] Temperature difference coefficient correction feature generation model: The temperature difference coefficient correction feature is expanded by a generative adversarial network model based on the food insertion depth to obtain a second temperature difference coefficient correction feature; and the processed first temperature difference coefficient correction feature is used to obtain temperature difference coefficient correction feature training data;

[0010] Food temperature deviation coefficient generation model construction module: Receives and processes the temperature difference coefficient correction feature training data and the food temperature deviation coefficient to construct a food temperature deviation coefficient generation model;

[0011] Intelligent food temperature correction module: The food temperature deviation coefficient generation model is used to generate multiple food temperature deviation coefficients to correct the first food temperature detected by the probe and obtain the second food temperature.

[0012] Furthermore, the food insertion depth is obtained by measuring the distance sensor inside the thermometer;

[0013] The properties of the food being tested are calculated from the food's mass, volume, specific heat capacity, and moisture content.

[0014] Furthermore, the ambient temperature is calculated from the on-site measured temperature and the average ambient temperature.

[0015] Furthermore, the first temperature difference coefficient correction feature obtained by processing the food insertion depth, detecting the food properties, and the ambient temperature is achieved by horizontally concatenating feature vectors.

[0016] Furthermore, the temperature deviation coefficient of the food is between 0.91 and 0.96.

[0017] Furthermore, the generative adversarial network model employs an improved generative adversarial network model based on the food insertion depth.

[0018] Furthermore, the food temperature deviation coefficient generation model adopts a linear discriminant model.

[0019] Furthermore, the linear discriminant model employs an improved multi-class linear discriminant model.

[0020] Furthermore, the first food temperature detected by the probe is corrected to obtain the second food temperature, and the multiple food temperature deviation coefficients generated by the food temperature deviation coefficient generation model are used.

[0021] A probe-based digital intelligent food thermometer with host receiver and wireless transmitter is also provided, including a probe-based digital intelligent food temperature correction measurement system with host receiver and wireless transmitter to achieve intelligent food temperature correction measurement.

[0022] This invention employs an internal sensor within a thermometer to collect the food insertion depth. A host computer receives and detects the food properties and ambient temperature. Using the food insertion depth, the detected food properties, the ambient temperature, and the corresponding food temperature deviation coefficient, a food temperature deviation coefficient generation model is constructed. This model generates multiple temperature deviation coefficients to correct the temperature values ​​detected by the probe. Furthermore, a generative adversarial network model based on food insertion depth correction is used to expand the data volume of the food temperature deviation coefficient generation model, resulting in a more accurate food temperature deviation coefficient. This invention utilizes multiple types of machine learning models to generate a large amount of data on the features affecting the deviation coefficient, and uses this data to generate a more accurate and efficient temperature correction coefficient, thereby improving the intelligence and accuracy of food thermometer measurements. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the implementation of a probe-based digital intelligent food temperature correction and measurement system with host receiver and wireless transmitter according to the present invention.

[0024] Figure 2 This is a schematic diagram of the structure of a probe-based digital intelligent food temperature correction measurement system with host receiver and wireless transmitter according to the present invention.

[0025] Figure 3 This is a schematic diagram of the probe-based digital intelligent food thermometer with host receiver and wireless transmitter in this invention.

[0026] Figure 4 This is a schematic diagram of the improved generative adversarial network model in this invention;

[0027] Figure 5 This is a schematic diagram of the receiving host structure of the present invention; Detailed Implementation

[0028] The invention will now be further described in conjunction with the accompanying drawings and specific embodiments.

[0029] This invention discloses an intelligent measuring instrument for measuring the temperature of food, which belongs to the manufacturing of experimental and analytical instruments such as intelligent measuring instruments and meters, and is classified under the intelligent manufacturing equipment industry.

[0030] In a first aspect of the present invention, in order to solve the above-mentioned technical problems, a probe-based digital intelligent food thermometer with host receiver and wireless transmitter is provided.

[0031] In the first aspect of the invention, as shown in the appendixFigure 1 The diagram illustrates the implementation flow of a probe-based digital intelligent food temperature correction measurement system with host reception and wireless transmission. The system includes a sensor depth acquisition module, a host reception module, a temperature difference coefficient correction feature generation module, a food temperature deviation coefficient generation model construction module, and an intelligent food temperature correction module.

[0032] Sensor depth acquisition module: used to acquire the insertion depth of food.

[0033] The host receiving module is used to receive the food insertion depth, as well as to receive the detected food properties and ambient temperature, and to obtain the food temperature deviation coefficient set by the staff based on experience. It processes the food insertion depth, the detected food properties and the ambient temperature to obtain the first temperature difference coefficient correction feature.

[0034] Temperature difference coefficient correction feature generation model: The temperature difference coefficient correction feature is expanded by a generative adversarial network model based on the food insertion depth to obtain a second temperature difference coefficient correction feature; and the first temperature difference coefficient correction feature is processed together to obtain temperature difference coefficient correction feature training data;

[0035] Food temperature deviation coefficient generation model construction module: Receives and processes the temperature difference coefficient correction feature training data and the food temperature deviation coefficient to construct a food temperature deviation coefficient generation model;

[0036] Intelligent food temperature correction module: The food temperature deviation coefficient generation model is used to generate multiple food temperature deviation coefficients to correct the first food temperature detected by the probe and obtain the second food temperature.

[0037] See attached Figure 2 The diagram shows the module structure of a probe-based digital intelligent food temperature correction measurement system with host receiver and wireless transmitter.

[0038] Appendix Figure 3 This refers to a probe-based digital food thermometer with a host receiver and wireless transmitter, which is a standard construction for food thermometers for those skilled in the art.

[0039] Furthermore, the food insertion depth is obtained by measuring the distance sensor inside the thermometer;

[0040] The properties of the food being tested are calculated from the food mass, food volume, specific heat capacity, and moisture content. The calculation formula is as follows:

[0041]

[0042] In the formula, To test the properties of food, For specific heat capacity, For food quality, For food volume, Moisture content;

[0043] The temperature change of food after processing is mainly determined by its thermodynamic properties and storage conditions. This invention specifically prioritizes food mass, volume, specific heat capacity, and moisture content for determining these parameters. This is because the invention has found that food mass, specific heat capacity, and moisture content significantly influence food temperature changes. To ensure accurate and efficient temperature measurement, this invention specifically prioritizes these parameters. For large-mass foods, internal temperature changes lag behind surface temperature. Food mass is measured using conventional techniques. However, as food volume increases, heat transfer from the surface to the center takes longer, leading to a larger internal and external temperature difference. Therefore, temperature changes take longer, reducing the time required for temperature probes (such as thermometers) to measure temperature. For foods with high specific heat capacity and high moisture content, instantaneous temperature measurement by the thermometer is necessary to minimize the impact of temperature changes.

[0044] Furthermore, the ambient temperature is calculated from the on-site measured temperature and the average ambient temperature, and the calculation formula is as follows:

[0045]

[0046] In the formula, The ambient temperature is... For on-site temperature measurement, The ambient average temperature.

[0047] This invention uses the temperature measured by the on-site thermometer and the predicted average ambient temperature to correct the measured ambient temperature, thus obtaining a more accurate ambient temperature. It eliminates the error between the measured temperature and the ambient temperature. Given that ambient temperature has a significant impact on food temperature measurement, the data feature processing is more precise to obtain a more accurate food temperature deviation coefficient for measuring the internal temperature of food.

[0048] Furthermore, the first temperature difference coefficient correction feature obtained by processing the food insertion depth, detecting the food properties, and the ambient temperature is achieved by horizontally concatenating feature vectors.

[0049] Furthermore, the food temperature deviation coefficient ranges from 0.91 to 0.96. This invention finds that food temperature deviation correction generally does not exhibit large fluctuations, which is determined by the requirements of intelligent food temperature measurement and practical considerations. Therefore, the food temperature deviation coefficient used for training the model in this invention ranges from 0.91 to 0.96. This is also consistent with the subsequent requirement of expanding the temperature difference coefficient correction feature to obtain a second temperature difference coefficient correction feature based on the generative adversarial network model using the food insertion depth. This allows for obtaining more high-quality training data, enabling the model parameters to be trained to better suit the probe-based digital intelligent food temperature measurement correction in this application scenario.

[0050] As attached Figure 4 As shown, Generative Adversarial Networks (GANs) are deep learning models that generate data through adversarial training. Their core principle is based on a two-player zero-sum game in game theory, where two neural networks—a generator and a discriminator—compete against each other and work together to optimize the data. The generator (G) receives a random noise vector z (usually sampled from a Gaussian distribution) and generates fake data through a neural network, aiming to make the generated data distribution approximate the real data distribution.

[0051] Discriminator (D): Receives real and generated data, and outputs a scalar probability value representing the likelihood that the input data comes from the true distribution. Its goal is to accurately distinguish between real and fake data.

[0052] Furthermore, the generative adversarial network model adopts an improved generative adversarial network model based on the food insertion depth, and its generation activation function is:

[0053]

[0054] In the formula, To generate activation function values, n is the number of the first temperature difference coefficient correction features used to train the food temperature deviation coefficient generation model. Let be the food insertion depth for the i-th training data.

[0055] In this invention, n represents the number of feature vectors collected for training. During the expansion and generation of temperature difference coefficient correction features, it was found that the depth of the thermometer insertion into the food significantly affects the generated data. This allows the generated data to be more closely bound to a reasonable range, while also improving the food temperature deviation coefficient corresponding to the generated temperature difference coefficient correction features so that it is mostly within the set range of 0.91 to 0.96. Therefore, the food insertion depth of the actual collected data in the training dataset is used to improve the generative adversarial network model. Specifically, the most important generative activation function is improved, making the generative adversarial network model more generalizable to the generated data and improving the applicability of the model in the scenario of thermometer measurement correction.

[0056] Furthermore, the food temperature deviation coefficient generation model adopts a linear discriminant model, and its calculation formula is as follows:

[0057]

[0058] In the formula, This is the output food temperature deviation coefficient. It is a normal vector perpendicular to the plane. The input temperature difference coefficient is corrected for the feature. These are constant features of the linear discriminant model.

[0059] In this embodiment, the food temperature deviation coefficient generation model uses a conventional linear discriminant model, which has a certain classification error, but the obtained food temperature deviation coefficient has high accuracy.

[0060] Furthermore, the linear discriminant model adopts an improved multi-class linear discriminant model, and its calculation formula is as follows:

[0061]

[0062] In the formula, The output is the food temperature deviation coefficient, where n is the number of the first temperature difference coefficient correction features used to train the food temperature deviation coefficient generation model, and u is the number of the second temperature difference coefficient correction features used to train the food temperature deviation coefficient generation model. The sum of the eigenvectors for correcting the temperature difference coefficient obtained from the i-th acquisition is given. The sum of the eigenvectors for correcting the temperature difference coefficient generated for the kth time is given. To correct the number of features for the temperature difference coefficient obtained from centralized collection and processing of the training dataset, To generate and correct the number of features for the temperature difference coefficient obtained from the training dataset, It is a normal vector perpendicular to the plane. The input temperature difference coefficient is corrected for the feature. Constant features for improving multi-class linear discriminant models.

[0063] In order to ensure that the generation of food temperature deviation coefficients is more accurate and to minimize the error of the generated data when using the generated data for subsequent classification, this invention uses the relationship between the generated data and the collected data to correct the model. This can effectively reduce the model's misclassification rate, improve the model's robustness, and enable the model to be applied to personalized classification scenarios, especially in the acquisition of food temperature deviation coefficients.

[0064] In this embodiment, according to The value is used to output the food temperature deviation coefficient. This invention can be set such that when... When the value is less than -5, the food temperature deviation coefficient is 0.91. When the value is greater than or equal to -5 and less than 1, the food temperature deviation coefficient is 0.92. When the value is greater than or equal to 1, it is less than 0, and the food temperature deviation coefficient is 0.93. When the value is greater than 0 and less than or equal to 1, the food temperature deviation coefficient is 0.94. When the value is greater than 1 and less than 5, the food temperature deviation coefficient is 0.95. When the value is greater than or equal to 5, the food temperature deviation coefficient is 0.96.

[0065] Furthermore, the first food temperature detected by the probe is corrected to obtain the second food temperature. The multiple food temperature deviation coefficient generated using the food temperature deviation coefficient generation model is calculated using the following formula:

[0066]

[0067] In the formula, The second food temperature, The temperature of the first food sample is measured in the p-th temperature detection. Let Q be the food temperature deviation coefficient for the p-th temperature measurement, and let Q be the total number of food temperature measurements.

[0068] In this embodiment, food is typically tested multiple times. The food temperature deviation coefficient obtained through model processing is used to correct the food temperature from multiple tests to obtain the food temperature closest to the actual situation.

[0069] Another aspect of this embodiment provides a probe-based digital intelligent food thermometer with host reception and wireless transmission, including a probe-based digital intelligent food temperature correction measurement system with host reception and wireless transmission to achieve intelligent food temperature correction measurement.

[0070] Appendix Figure 5 The receiving host structure of the present invention is a host receiving box that houses a thermometer and is used to process relevant data. It has no special structure, except that the food temperature deviation coefficient generation model described in this application is installed inside to process the relevant data.

[0071] This invention employs an internal sensor within a thermometer to collect the food insertion depth. A host computer receives and detects the food properties and ambient temperature. Using the food insertion depth, the detected food properties, the ambient temperature, and the corresponding food temperature deviation coefficient, a food temperature deviation coefficient generation model is constructed. This model generates multiple temperature deviation coefficients to correct the temperature values ​​detected by the probe. Furthermore, a generative adversarial network model based on food insertion depth correction is used to expand the data volume of the food temperature deviation coefficient generation model, resulting in a more accurate food temperature deviation coefficient. This invention utilizes multiple types of machine learning models to generate a large amount of data on the features affecting the deviation coefficient, and uses this data to generate a more accurate and efficient temperature correction coefficient, thereby improving the intelligence and accuracy of food thermometer measurements.

[0072] The combination of multiple embodiments of the present invention can achieve all the above effects, but it is not required that each embodiment of the present invention achieve all the above advantages and effects, because each embodiment of the present invention can constitute a separate technical solution and make one or more contributions to the prior art.

[0073] For any module structures not specifically defined in this invention, the existing technical specifications shall prevail. The existing technical specifications mentioned in the foregoing background and specific embodiments sections are considered part of this invention and are used to understand the meaning of certain technical features or parameters. The scope of protection of this invention is determined by the actual contents of the claims.

Claims

1. A probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission, comprising a sensor depth acquisition module, a host receiving module, a temperature difference coefficient correction feature generation module, a food temperature deviation coefficient generation model construction module, and an intelligent food temperature correction module, characterized in that: the sensor depth acquisition module is used to acquire the food insertion depth; the host receiving module is used to receive the food insertion depth, and is also used to receive the detected food properties and the ambient temperature, and obtain the food temperature deviation coefficient set by the staff according to experience, and process the food insertion depth, the detected food properties, and the ambient temperature to obtain the first temperature difference coefficient correction feature; the temperature difference coefficient correction feature generation model is used to expand the temperature difference coefficient correction feature based on the generated adversarial network model based on the food insertion depth to obtain the second temperature difference coefficient correction feature; and the first temperature difference coefficient correction feature obtained by processing is used to obtain the temperature difference coefficient correction feature training data; the food temperature deviation coefficient generation model construction module receives and processes the temperature difference coefficient correction feature training data and the food temperature deviation coefficient to construct the food temperature deviation coefficient generation model; the intelligent food temperature correction module uses the food temperature deviation coefficient generation model to generate the food temperature deviation coefficient of multiple detections to correct the first food temperature detected by the probe to obtain the second food temperature; the food insertion depth, the detected food properties, and the ambient temperature are processed to obtain the first temperature difference coefficient correction feature by using the feature vector transverse splicing method; and the food temperature deviation coefficient generation model uses a linear discriminant model or an improved multi-class linear discriminant model. 2.The probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission according to claim 1, characterized in that: the food insertion depth is obtained by measuring the distance from the sensor inside the thermometer; and the detected food properties are obtained by calculating the food quality, the food volume, the specific heat capacity, and the water content. 3.The probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission according to claim 1, characterized in that: the ambient temperature is obtained by calculating the on-site temperature and the average ambient temperature. 4.The probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission according to claim 1 or 2 or 3, characterized in that: the food temperature deviation coefficient is in the range of 0.91 to 0.

96. 5.The probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission according to claim 4, characterized in that: the generated adversarial network model is an improved generated adversarial network model based on the food insertion depth. 6.The probe digital intelligent food temperature correction measurement system with host receiving and wireless transmission according to claim 1, characterized in that: the food temperature deviation coefficient of multiple detections generated by the food temperature deviation coefficient generation model is used to correct the first food temperature detected by the probe to obtain the second food temperature. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ 7. A digital smart food thermometer with host-receiving and wireless-transmitting probe, characterized in that, A digital smart food temperature correction measurement system with host receiver and wireless transmission including a probe as claimed in any one of claims 1 to 6 to enable smart food temperature correction measurement.

Images