Cooking machine motor overheating protection method and system for high-temperature environment

Abstract

The invention relates to the technical field of chef machine motor protection, in particular to a chef machine motor overheating protection method and system oriented to a high-temperature environment, and the method comprises the steps: firstly collecting chef machine monitoring data, then obtaining first temperature prediction data according to a preset prediction model, and obtaining second temperature prediction data based on a preset thermodynamic model; and then combining the first temperature prediction data and the second temperature prediction data to adjust the initial overheating protection threshold value to obtain a target overheating protection threshold value, and finally performing predictive protection adjustment on the chef machine according to the target overheating protection threshold value. Compared with the prior art, the method has the advantages that the overheating protection threshold is dynamically adjusted according to the results of the preset prediction model and the preset thermodynamic model so as to reduce misjudgment, and meanwhile, the situation of shutdown waiting is avoided by adopting a predictive protection adjustment mode; the problems of low accuracy and low efficiency of a protection mode based on a fixed threshold in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of chef machine motor protection, and particularly to an overheat protection method and system for a chef machine motor facing a high-temperature environment. Background Art

[0002] Currently, most of the overheat protection methods adopted by chef machines in high-temperature environments on the market are based on thresholds for protection. That is to say, a specific temperature value is set in advance as the threshold. Once the temperature sensor inside the chef machine detects that the body temperature reaches or exceeds this threshold, the overheat protection device will be triggered, and then corresponding protection measures will be taken. However, this threshold-based overheat protection method has some obvious disadvantages.

[0003] On the one hand, relying solely on the temperature threshold to judge overheating is prone to errors. For example, when the chef machine is mixing overly hard dough, the motor torque current is large in a short period of time, and the threshold-based protection method cannot accurately distinguish this situation, and it is very likely to misidentify the normal temperature rise caused by overly hard dough as an overheat phenomenon, resulting in misjudgment. On the other hand, the current overheat protection method based on the judgment threshold is too rigid, resulting in the chef machine having to adopt the method of stopping and waiting for cooling for protection, increasing the working duration and affecting the working efficiency of the equipment. Especially in a high-temperature environment, the existing method extremely affects the user experience.

[0004] Therefore, people need an overheat protection method for a chef machine motor with high accuracy and high efficiency. Summary of the Invention

[0005] Therefore, the present invention provides an overheat protection method and system for a chef machine motor facing a high-temperature environment to solve the problems of low accuracy and low efficiency of the existing fixed-threshold-based protection method.

[0006] The present invention provides an overheat protection method for a chef machine motor facing a high-temperature environment, including: Collecting chef machine monitoring data; Based on the chef machine monitoring data, obtaining first temperature prediction data according to a preset prediction model; Based on the chef machine monitoring data, obtaining second temperature prediction data according to a preset thermodynamic model; Obtaining an initial overheat protection threshold, and adjusting the initial overheat protection threshold by combining the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold; Performing predictive protection adjustment on the chef machine according to the target overheat protection threshold.

[0007] Further, based on the chef machine monitoring data, obtaining first temperature prediction data according to a preset prediction model, including: Obtain the monitoring data of the cooking machine to get a monitoring data sequence; Set a sliding window in the monitoring data sequence, and obtain input data according to the content of the sliding window, where the input data includes a vector sequence, each vector in the vector sequence corresponds to a collection time, and each vector includes the motor current, motor speed, ambient temperature, and ambient humidity of the cooking machine; Input the input data into a preset prediction model to obtain the output data of the preset prediction model, where the preset prediction model is a prediction model based on deep learning, and the output data includes predicting the temperature change slope of the cooking machine within a preset time period; Obtain the first temperature prediction data according to the output data.

[0008] Furthermore, the preset prediction model includes an input layer, two LSTM hidden layers, a Dropout layer, a fully connected layer, and an output layer connected in sequence.

[0009] Furthermore, the loss function used during the training of the preset prediction model is: ; where, represents the loss, represents the prediction error, represents the thermodynamic consistency error, and are respectively preset loss weights, where the prediction error is obtained based on the difference between the output result of the preset prediction model corresponding to the input sample data and the true label corresponding to the input sample data, and the thermodynamic consistency error is obtained based on the difference between the output result of the preset thermodynamic model corresponding to the input sample data and the true label corresponding to the input sample data.

[0010] Furthermore, obtain an initial overheat protection threshold, and combine the first temperature prediction data and the second temperature prediction data to adjust the initial overheat protection threshold to obtain a target overheat protection threshold, including: Obtain the relative humidity according to the monitoring data of the cooking machine; Perform a weighted sum on the first temperature prediction data and the second temperature prediction data to obtain a temperature change preference value; Obtain the target overheat protection threshold according to the initial overheat protection threshold, the temperature change preference value, and the relative humidity.

[0011] Furthermore, obtaining the target overheat protection threshold according to the initial overheat protection threshold, the temperature change preference value, and the relative humidity includes: Obtain the target overheat protection threshold through the following formula: ; where, is the target overheat protection threshold, is the initial overheat protection threshold, is the temperature change preference value, is the preset adjustment rate, is the relative humidity, is the preset humidity compensation coefficient.

[0012] Furthermore, perform weighted summation on the first temperature prediction data and the second temperature prediction data to obtain the temperature change preference value, including: Obtain the first weight corresponding to the first temperature prediction data and the second weight corresponding to the second temperature prediction data; Obtain the actual temperature data and the historical temperature prediction data output by the preset prediction model; Calculate the difference between the actual temperature data and the historical temperature prediction data, and compare the difference with the preset threshold; If the difference is less than the preset threshold, increase the first weight and decrease the second weight; If the difference is not less than the preset threshold, increase the second weight and decrease the first weight.

[0013] Furthermore, perform predictive protection adjustment on the cooking machine according to the target overheat protection threshold, including: Establish a set of policy description vectors based on the monitoring data of the cooking machine; Based on the preset thermodynamics model, with the optimization goal of the temperature not exceeding the target overheat protection threshold and the shortest processing time, optimize the set of policy description vectors through an optimization algorithm to obtain the optimal policy description vector; Adjust the cooking machine according to the optimal policy description vector.

[0014] Furthermore, the preset thermodynamics model includes: The differential equation of temperature change: ; Wherein, is the equivalent heat capacity of the cooking machine, is the real-time heating power of the cooking machine, is the comprehensive heat dissipation power of the cooking machine, is the differential of the second temperature prediction data, is the differential of time; The real-time heating power is calculated by the following formula: ; Wherein, is the motor current, is the resistance of each phase in the motor stator winding; The comprehensive heat dissipation power of the cooking machine is calculated by the following formula: ; Among them, is the natural convection heat dissipation power, is the heat conduction heat dissipation power of the shell, is the air-cooling heat dissipation power; The natural convection heat dissipation power is calculated by the following formula: ; Among them, is the preset convective heat transfer coefficient, is the surface area of the natural convection heat dissipation surface of the cooking machine, is the temperature difference between the natural convection heat dissipation surface of the cooking machine and the environment; The heat conduction heat dissipation power of the shell is calculated by the following formula: ; Among them, is the thermal conductivity of the shell material, is the area of the shell, is the thickness of the shell, is the temperature difference between the inside and outside of the shell; The air-cooling heat dissipation power is calculated by the following formula: ; Among them, is the air density, is the specific heat capacity at constant pressure of the air, is the air flow rate of the cooling fan, and is the air temperature at the air inlet of the cooling fan.

[0015] The present invention also provides a motor overheat protection system for a cooking machine facing a high-temperature environment, including: An acquisition and monitoring module for acquiring the monitoring data of the cooking machine; A temperature prediction module for obtaining first temperature prediction data based on the monitoring data of the cooking machine according to a preset prediction model; A theoretical calculation module for obtaining second temperature prediction data based on the monitoring data of the cooking machine according to a preset thermodynamics model; A dynamic adjustment module for obtaining an initial overheat protection threshold, and adjusting the initial overheat protection threshold in combination with the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold; A prediction protection module for performing predictive protection adjustment on the cooking machine according to the target overheat protection threshold.

[0016] The beneficial effects of adopting the above embodiments are: The present invention provides a method and system for overheat protection of a mixer motor facing a high-temperature environment. First, it collects mixer monitoring data. Then, based on the mixer monitoring data and according to a preset prediction model, it obtains first temperature prediction data, and based on the mixer monitoring data and according to a preset thermodynamics model, it obtains second temperature prediction data. After that, it acquires an initial overheat protection threshold, and combines the first temperature prediction data and the second temperature prediction data to adjust the initial overheat protection threshold to obtain a target overheat protection threshold. Finally, according to the target overheat protection threshold, it performs predictive protection adjustment on the mixer. Compared with the prior art, the present invention uses a preset prediction model to predict the temperature, uses a preset thermodynamics model for theoretical calculation to eliminate prediction errors, and dynamically adjusts the overheat protection threshold based on the results of both to reduce misjudgment situations. At the same time, it uses the method of predictive protection adjustment to avoid the occurrence of downtime waiting, solving the problems of low accuracy and low efficiency of the protection method based on a fixed threshold in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flowchart of the method for overheat protection of a mixer motor facing a high-temperature environment provided by the present invention; Figure 2 is Figure 1 a specific step diagram of step S102 in Figure 3 is Figure 1 a specific step diagram of step S104 in Figure 4 is a system structure diagram of the overheat protection system for a mixer motor facing a high-temperature environment provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] In conjunction with Figure 1 shown, a specific embodiment of the present invention discloses a method for overheat protection of a mixer motor facing a high-temperature environment, including: S101. Collect mixer monitoring data; S102. Based on the mixer monitoring data, obtain first temperature prediction data according to a preset prediction model; S103. Based on the mixer monitoring data, obtain second temperature prediction data according to a preset thermodynamics model; S104. Obtain the initial overheat protection threshold, and combine the first temperature prediction data and the second temperature prediction data to adjust the initial overheat protection threshold to obtain the target overheat protection threshold; S105. Perform predictive protection adjustment on the cooking machine according to the target overheat protection threshold.

[0020] In the above content, the monitoring data of the cooking machine refers to various data that can be collected by the monitoring system during the operation of the cooking machine, including but not limited to current, voltage, torque, temperature, humidity, etc. The overheat protection threshold is a critical value set for the overheat protection of the cooking machine motor. When the temperature inside the cooking machine reaches or exceeds this threshold, the overheat protection mechanism will be triggered to take corresponding protection measures.

[0021] The preset prediction model is a model used to predict future data. Its main function is to predict the temperature change trend of the cooking machine during subsequent operation based on the collected monitoring data of the cooking machine, and dynamically adjust the overheat protection threshold based on this. It can be imagined that in actual applications, the environment and purpose of users using the cooking machine are highly unpredictable, and there are still certain errors in the prediction results of the preset prediction model.

[0022] To eliminate this error, the present invention also takes into account the working condition rules of the cooking machine in most cases (that is, in most cases, when the cooking machine is in use, most of its parameters such as current remain unchanged). Assuming that other parameters such as current remain unchanged, the preset thermodynamic model is used to constrain the prediction results of the preset prediction model.

[0023] The preset thermodynamic model is a theoretical calculation model constructed based on the basic principles of thermodynamics. It mainly considers the physical processes such as heat generation, transfer, and dissipation during the operation of the cooking machine. Through the analysis of the monitoring data and the combination of thermodynamic formulas and parameters, this model calculates the temperature distribution and changes of each component inside the cooking machine. Compared with the preset prediction model, the preset thermodynamic model pays more attention to starting from the physical essence and deeply analyzing the source and destination of heat, and can provide more accurate theoretical support for temperature prediction. By comprehensively adjusting the overheat protection threshold based on the theoretical calculation results of the preset thermodynamic model and the prediction results of the preset prediction model, the prediction error caused by different environments and usage purposes can be eliminated, and the rationality of the dynamic adjustment of the overheat protection threshold can be improved.

[0024] Predictive protection adjustment refers to taking a series of actions in advance to adjust the operating state of the cooking machine before it reaches the overheat state according to the target overheat protection threshold, so as to avoid excessive temperature. Different from the traditional protection method of shutting down and waiting for cooling, predictive protection adjustment pays more attention to early intervention and optimization. For example, when it is predicted that the temperature of the cooking machine is about to approach the target overheat protection threshold, the system can automatically adjust the speed of the motor to reduce its load; or start a more efficient heat dissipation device to increase air circulation and accelerate heat dissipation; it can also optimize the working process of the cooking machine, reasonably arrange tasks, and reduce unnecessary energy consumption and heat generation.

[0025] In summary, through these measures taken in advance, the present invention realizes the paradigm innovation of "prediction - adjustment" replacing "detection - power off", and can effectively reduce the temperature, extend its service life, and improve work efficiency without affecting the normal operation of the cooking machine.

[0026] It can be imagined that the preset prediction model can be implemented by any existing technology, and the present invention provides a preferred method. Specifically, in combination with Figure 2 As shown, in another embodiment, the above step S102. Based on the monitoring data of the cooking machine, according to the preset prediction model, obtain the first temperature prediction data, which specifically includes: S201. Obtain the monitoring data of the cooking machine to obtain a monitoring data sequence; S202. Set a sliding window in the monitoring data sequence, and obtain input data according to the content of the sliding window. Among them, the input data includes a vector sequence, each vector in the vector sequence corresponds to a collection time, and each vector includes the motor current, motor speed, ambient temperature, and ambient humidity of the cooking machine; S203. Input the input data into the preset prediction model to obtain the output data of the preset prediction model. Among them, the preset prediction model is a prediction model based on deep learning, and the output data includes predicting the temperature change slope of the cooking machine within a preset time period; S204. Obtain the first temperature prediction data according to the output data.

[0027] In the above content, the deep learning model can automatically learn complex non-linear relationships and patterns from a large amount of monitoring data without manual feature extraction, effectively mine the deep information in the data, and improve the prediction accuracy. The monitoring data of the cooking machine includes various dimensions of data such as current, voltage, torque, temperature, and humidity. The deep learning model can effectively process these high-dimensional data, comprehensively consider the influence of various factors on temperature, and provide a more comprehensive and accurate prediction. Through the sliding window technology, the deep learning model can update the input data in real time, quickly adapt to the dynamic changes of the operating state of the cooking machine, timely capture the trend of temperature changes, and enhance the real-time performance and adaptability of the model. Compared with the traditional linear prediction model, the deep learning model can better fit the complex data distribution, more accurately predict the slope of temperature changes, thus providing a more reliable basis for overheat protection and effectively preventing the occurrence of overheat phenomena.

[0028] In a more specific embodiment, the above preset prediction model is an LSTM model, which includes an input layer, two LSTM hidden layers, a Dropout layer, a fully connected layer, and an output layer connected in sequence. Among them, the data in the input layer is processed by sliding window Z-score normalization to avoid the influence of long-term drift (such as the temperature change throughout the day in the kitchen). The first LSTM hidden layer and the second LSTM hidden layer both include 32 LSTM units to achieve lightweight. The double-layer LSTM design can extract features layer by layer. The first layer captures short-term fluctuations (such as current mutations), and the second layer learns long-term dependencies (such as continuous high-load trends). The Dropout layer can prevent overfitting (especially when the data noise in the kitchen scenario is large) and improve the robustness of the model to occasional sensor failures. The number of units in the fully connected layer is 16, which can compress the high-dimensional state (32 dimensions) output by the LSTM into a low-dimensional space and extract the most sensitive feature combination for temperature rise prediction.

[0029] Furthermore, the pure machine learning model may output unreasonable results due to noise data or overfitting (such as predicting an instant temperature drop of 100°C). To further improve the accuracy of the present invention, when training the preset prediction model of the present invention, the output of the preset thermodynamic model is further used as the prior knowledge of machine learning to construct a hybrid loss function. That is, the present invention also provides an embodiment, in which the loss function used when training the preset prediction model is: ; Among them, represents the loss, represents the prediction error, represents the thermodynamic consistency error, and They are preset loss weights respectively, where the prediction error is obtained based on the difference between the output result of the preset prediction model corresponding to the input sample data and the true label corresponding to the input sample data, and the thermodynamic consistency error is obtained based on the difference between the output result of the preset thermodynamic model corresponding to the input sample data and the true label corresponding to the input sample data.

[0030] In the above process, the input sample data refers to the data prepared in advance for training the preset prediction model, and the true label refers to the true output result corresponding to the input sample data. In this embodiment, by introducing the output of the thermodynamic model as prior knowledge when training the preset prediction model, constructing a hybrid loss function, and combining physical laws to constrain the prediction result, the occurrence of overfitting caused by pure data-driven is avoided, forcing the machine learning prediction to be consistent with the thermodynamic model in the trend of change. The advantages of the two different methods are complementary, enabling the model to automatically learn and adapt to physical laws during the training process, effectively improving the rationality and accuracy of the preset prediction model, and providing a more reliable prediction basis for the overheat protection of the cooking machine.

[0031] Similarly, it can be understood that the preset thermodynamic model can also be implemented using any model according to the actual specific situation, and the present invention provides a preferred method. The preset thermodynamic model in step S103 above specifically includes: Differential equation of temperature change: ; Where is the equivalent heat capacity of the cooking machine, is the real-time heating power of the cooking machine, is the comprehensive heat dissipation power of the cooking machine, is the differential of the second temperature prediction data, is the differential of time. It can be understood that according to the calculation results of temperature and time in the above formula, the temperature data of the cooking machine at any moment can be calculated.

[0032] The real-time heating power is calculated by the following formula: ; Where is the motor current, is the resistance of each phase in the motor stator winding (at the actual temperature); The comprehensive heat dissipation power of the cooking machine is calculated by the following formula: ; Where is the natural convection heat dissipation power, is the heat conduction heat dissipation power of the housing, is the air-cooled heat dissipation power; The natural convection heat dissipation power is calculated by the following formula: ; Wherein, is the preset convective heat transfer coefficient, is the surface area of the natural convection heat dissipation surface of the cooking machine, is the temperature difference between the natural convection heat dissipation surface of the cooking machine and the environment; The heat dissipation power of the shell conduction is calculated by the following formula: ; Wherein, is the thermal conductivity of the shell material, is the area of the shell, is the thickness of the shell, is the temperature difference between the inside and outside of the shell; The air-cooled heat dissipation power is calculated by the following formula: ; Wherein, is the air density, is the specific heat capacity at constant pressure of the air, is the air flow rate of the cooling fan, and is the air temperature at the air inlet of the cooling fan.

[0033] Further, as shown in Figure 3 In a feasible solution, the above step S104. Obtain the initial overheat protection threshold, and combine the first temperature prediction data and the second temperature prediction data to adjust the initial overheat protection threshold to obtain the target overheat protection threshold, specifically including: S301. Obtain the relative humidity according to the monitoring data of the cooking machine; S302. Perform a weighted sum of the first temperature prediction data and the second temperature prediction data to obtain a temperature change preference value; S303. Obtain the target overheat protection threshold according to the initial overheat protection threshold, the temperature change preference value and the relative humidity.

[0034] In this embodiment, the consideration of relative humidity is added. Relative humidity is one of the important factors affecting the temperature change of the cooking machine (mainly affecting heat dissipation). By incorporating relative humidity into the calculation, the influence of the environment on the temperature of the cooking machine can be considered more comprehensively, so that the adjustment of the overheat protection threshold can better adapt to different working environments and improve the generality and adaptability of the model.

[0035] Specifically, in the process of the above step S303. Obtain the target overheat protection threshold according to the initial overheat protection threshold, the temperature change preference value and the relative humidity, a specific and feasible formula for calculating the target overheat protection threshold is: ; Wherein, is the target overheat protection threshold, is the initial overheat protection threshold, is the temperature change preference value, is the preset adjustment rate, is the relative humidity, is the preset humidity compensation coefficient.

[0036] Furthermore, still considering the different usage purposes and habits of different users, when performing weighted summation on the first temperature prediction data and the second temperature prediction data, the weights can be feedback-adjusted according to the previous prediction results to further improve the accuracy of the present invention. Specifically, in a new embodiment, the above step S302 of performing weighted summation on the first temperature prediction data and the second temperature prediction data to obtain the temperature change preference value specifically includes: Obtain the first weight corresponding to the first temperature prediction data and the second weight corresponding to the second temperature prediction data; Obtain the actual temperature data and the historical temperature prediction data output by the preset prediction model; Calculate the difference between the actual temperature data and the historical temperature prediction data, and compare the difference with the preset threshold; If the difference is less than the preset threshold, increase the first weight and decrease the second weight; If the difference is not less than the preset threshold, increase the second weight and decrease the first weight.

[0037] The historical temperature prediction data refers to the data obtained when the preset prediction model was last called on the cooking machine on the premise that other working parameters of the cooking machine changed little, and the historical temperature prediction data and the actual temperature data correspond to the same moment.

[0038] By obtaining the actual temperature data and the historical temperature prediction data output by the preset prediction model and calculating their difference, the prediction accuracy of the preset prediction model can be evaluated. According to the comparison result of the difference with the preset threshold, the weights are adjusted, so that in the subsequent weighted summation process, more accurate prediction data will be given higher weights, thereby improving the accuracy of the temperature change preference value.

[0039] For example, when the results obtained from multiple predictions are inaccurate, this method will gradually increase the proportion of the second weight. At this time, when dynamically adjusting the overheat protection threshold, more emphasis will be placed on considering the calculation results of the preset thermodynamic model to ensure accuracy and improve the adaptability of the present invention.

[0040] Further, predictive protection adjustment refers to a strategy adjustment method that does not involve downtime waiting and performs early intervention on the cooking machine. For example, assume that when the cooking machine processes high-hardness dough, the current suddenly increases, causing the temperature to rise. After calculation and detection, it is possible to reach the overheat protection threshold within the next 20 seconds. At this time, the predictive protection adjustment method can be: reducing the motor speed by 15% 10 seconds in advance, so that the heating power is reduced to 42W, and avoiding triggering the power-off protection.

[0041] Similarly, it can be understood that in practice, the specific predictive protection adjustment measures adopted can also be flexibly designed according to the specific situation, and the present invention does not limit it.

[0042] For example, in an ideal situation, assume that the computing power of the cooking machine is sufficient, or the optimal strategy can be found through remote computing. In step S105 above. According to the target overheat protection threshold, when performing predictive protection adjustment on the cooking machine, the following method can be specifically used to determine the predictive protection adjustment measures: Based on the monitoring data of the cooking machine, establish a set of strategy description vectors; Based on a preset thermodynamic model, with the temperature not exceeding the target overheat protection threshold and the shortest processing time as the optimization goal, optimize the set of strategy description vectors through an optimization algorithm to obtain the optimal strategy description vector; Adjust the cooking machine according to the optimal strategy description vector.

[0043] The above process uses an optimization algorithm to obtain the best predictive protection adjustment strategy. Among them, the set of strategy description vectors refers to a set composed of vectors representing different adjustment strategies of the cooking machine. The specific encoding method of the vector can be flexibly determined according to the actual situation. For example: a numerical sequence can be used as the strategy description vector, and each numerical value in the vector corresponds to a time slice, and the content of the numerical value itself represents the power of the motor. When generating the set of strategy description vectors, an initial strategy description vector without zero elements can be generated first, and then the positions in the initial strategy description vector can be randomly set to 0 or adjusted to obtain a large amount of vector data as the set of strategy description vectors. The set of strategy description vectors mainly serves as the data basis for subsequent optimization algorithm optimization. After obtaining the optimal strategy description vector, the average value can be taken segment by segment according to the content in the optimal strategy description vector to obtain the adjustment strategy of the motor power.

[0044] In addition, it can be understood that the above content can adopt a variety of optimization algorithms, such as genetic algorithms, particle swarm optimization algorithms, etc., which can be flexibly selected and adjusted according to the specific situation to achieve the best optimization effect. The specific process of the above-mentioned optimization algorithms and the specific design methods for achieving the optimization goals mentioned in this embodiment are all prior arts that can be understood by those skilled in the art, so the present invention will not be described in detail.

[0045] Combined withFigure 4 As shown, the present invention also provides a food processor motor overheating protection system for high temperature environments, comprising: The collection and monitoring module 410 is used to collect the monitoring data of the chef machine; The temperature prediction module 420 is used to obtain first temperature prediction data based on the food processor monitoring data and a preset prediction model; Theoretical calculation module 430, used to obtain second temperature prediction data based on the food processor monitoring data and a preset thermodynamic model; The dynamic adjustment module 440 is used to obtain an initial overheat protection threshold, and adjust the initial overheat protection threshold in combination with the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold; The predictive protection module 450 is used to perform predictive protection adjustment on the chef machine according to the target overheat protection threshold.

[0046] It should be noted here that the corresponding system provided in the above embodiments can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can be found in the corresponding contents in the above method embodiments, which will not be repeated here.

[0047] The present invention provides a method and system for overheat protection of a chef machine motor in a high-temperature environment, which first collects monitoring data of the chef machine, then obtains first temperature prediction data based on the monitoring data of the chef machine according to a preset prediction model, and obtains second temperature prediction data based on the monitoring data of the chef machine according to a preset thermodynamic model, then obtains an initial overheat protection threshold, and adjusts the initial overheat protection threshold in combination with the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold, and finally performs predictive protection adjustment on the chef machine according to the target overheat protection threshold. Compared with the prior art, the present invention uses a preset prediction model to predict the temperature, uses a preset thermodynamic model to perform theoretical calculations to eliminate the prediction error, and dynamically adjusts the overheat protection threshold based on the results of the two to reduce the situation of misjudgment, and at the same time uses a predictive protection adjustment method to avoid the situation of downtime waiting, which solves the problem of low accuracy and low efficiency of the protection method based on fixed thresholds in the prior art.

[0048] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0049] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for overheating protection of a chef machine motor in a high temperature environment, characterized in that: include: Collect chef machine monitoring data; Based on the monitoring data of the chef machine and according to a preset prediction model, first temperature prediction data is obtained; Based on the monitoring data of the chef machine and the preset thermodynamic model, second temperature prediction data is obtained; Acquire an initial overheat protection threshold, and adjust the initial overheat protection threshold in combination with the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold; Make predictive protection adjustments to the kitchen machine based on the target overheat protection threshold.

2. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 1, characterized in that: Based on the monitoring data of the chef machine and according to the preset prediction model, first temperature prediction data is obtained, including: Get the monitoring data of the chef machine and obtain the monitoring data sequence; A sliding window is set in the monitoring data sequence, and input data is obtained according to the content of the sliding window, wherein the input data includes a vector sequence, each vector in the vector sequence corresponds to a collection time, and each vector includes a motor current, a motor speed, an ambient temperature, and an ambient humidity of the food processor; Inputting the input data into a preset prediction model to obtain output data of the preset prediction model, wherein the preset prediction model is a prediction model based on deep learning, and the output data includes predicting the temperature change slope of the food processor within a preset time period; According to the output data, first temperature prediction data is obtained.

3. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 2, characterized in that: The preset prediction model includes an input layer, two LSTM hidden layers, a Dropout layer, a fully connected layer, and an output layer connected in sequence.

4. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 3, characterized in that: The loss function used in the preset prediction model training is: ； in, Indicates loss, represents the prediction error, represents the thermodynamic consistency error, and are respectively the preset loss weights, wherein the prediction error is obtained by the difference between the output result of the preset prediction model corresponding to the input sample data and the true label corresponding to the input sample data, and the thermodynamic consistency error is obtained by the difference between the output result of the preset thermodynamic model corresponding to the input sample data and the true label corresponding to the input sample data.

5. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 1, characterized in that: Acquiring an initial overheat protection threshold, and adjusting the initial overheat protection threshold in combination with the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold, including: According to the monitoring data of the chef machine, the relative humidity is obtained; Performing weighted summation on the first temperature prediction data and the second temperature prediction data to obtain a temperature change preference value; The target overheat protection threshold is obtained according to the initial overheat protection threshold, the temperature change preference value and the relative humidity.

6. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 5, characterized in that: According to the initial overheat protection threshold, the temperature change preference value and the relative humidity, the target overheat protection threshold is obtained, including: The target overheat protection threshold is obtained by the following formula: ； in, is the target overheat protection threshold, is the initial overheat protection threshold, is the temperature change preference value, is the preset adjustment rate, is the relative humidity, It is the preset humidity compensation coefficient.

7. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 5, characterized in that: The first temperature prediction data and the second temperature prediction data are weighted and summed to obtain a temperature change preference value, including: Obtain a first weight corresponding to the first temperature prediction data and a second weight corresponding to the second temperature prediction data; Obtain actual temperature data and historical temperature prediction data output by a preset prediction model; Calculate the difference between the actual temperature data and the historical temperature prediction data, and compare the difference with the preset threshold; If the difference is less than the preset threshold, the first weight is increased and the second weight is decreased; If the difference is not less than the preset threshold, the second weight is increased and the first weight is decreased.

8. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 1, characterized in that: According to the target overheat protection threshold, the kitchen machine is adjusted predictively for protection, including: According to the monitoring data of the chef machine, a set of strategy description vectors is established; Based on the preset thermodynamic model, the optimization goal is to optimize the strategy description vector set by using the optimization algorithm, with the temperature not exceeding the target overheat protection threshold and the processing time being the shortest. The optimal strategy description vector is obtained. Adjust the chef machine according to the optimal policy description vector.

9. The method for overheating protection of a chef machine motor in a high temperature environment according to claim 1, characterized in that: Preset thermodynamic models, including: Differential equation for temperature change: ； in, is the equivalent heat capacity of the food processor, The real-time heating power of the food processor. The comprehensive heat dissipation power of the chef machine. is the differential of the second temperature prediction data, is the differential of time; The real-time heating power is calculated by the following formula: ； in, is the motor current, is the resistance of each phase in the motor stator winding; The comprehensive heat dissipation power of the chef machine is calculated by the following formula: ； in, is the natural convection heat dissipation power, The heat dissipation power of the shell is The heat dissipation power is air cooling; The natural convection heat dissipation power is calculated by the following formula: ； in, is the preset convective heat transfer coefficient, The surface area of ​​the natural convection heat dissipation surface of the chef machine. The temperature difference between the natural convection heat dissipation surface of the kitchen machine and the environment; The heat dissipation power of the shell is calculated by the following formula: ； in, is the thermal conductivity of the shell material, is the area of ​​the shell, is the shell thickness, is the temperature difference between the inside and outside of the shell; The air cooling power is calculated by the following formula: ； in, is the air density, is the constant pressure specific heat capacity of air, is the air flow of the cooling fan, and is the air temperature at the cooling fan air inlet.

10. A food processor motor overheat protection system for high temperature environments, characterized in that: include: Collection monitoring module, used to collect chef machine monitoring data; A temperature prediction module, for obtaining first temperature prediction data based on the food processor monitoring data and a preset prediction model; Theoretical calculation module, used for obtaining second temperature prediction data based on the monitoring data of the chef machine and a preset thermodynamic model; A dynamic adjustment module is used to obtain an initial overheat protection threshold, and adjust the initial overheat protection threshold in combination with the first temperature prediction data and the second temperature prediction data to obtain a target overheat protection threshold; The predictive protection module is used to make predictive protection adjustments to the chef machine according to a target overheat protection threshold.

Images