Integrated cooker temperature control method, system and equipment based on neural network
By using the predictive network model of convolutional neural network and gated cyclic unit in integrated foci temperature control, the problem of local small problems and slow learning speed in traditional BP neural network in integrated foci temperature control is solved, and the integrated foci is realized Accurate temperature control improves cooking convenience and safety.
Patent Information
- Application Number
- CN202411027838.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-12-13
AI Technical Summary
Traditional BP neural networks have small local problems and slow learning speed in integrated stove temperature control, resulting in the inability to accurately and efficiently control the temperature of the integrated stove, affecting the convenience, accuracy and safety of cooking.
The prediction network model of convolutional neural network unit and gated cycle unit is adopted to obtain and preprocess the operating data of the integrated stove, extract spatial and temporal characteristics, predict future temperature change trends, and adjust temperature control parameters according to the prediction results.
Accurate control of the temperature of the integrated stove is achieved, improving the convenience, accuracy and safety of cooking.
Smart Images
Figure CN119126868A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of integrated stove control technology, and more specifically to a neural network-based integrated stove temperature control method, system and equipment. Background Art
[0002] The temperature control method of integrated stoves based on neural networks is a new type of temperature control technology that has emerged in recent years with the development of intelligent and deep learning technologies. This method uses the powerful learning and prediction capabilities of neural networks to accurately control the temperature control process of integrated stoves, thereby improving the efficiency and quality of cooking. In temperature control systems, for most traditional temperature control systems, especially in application scenarios like integrated stoves, BP neural networks are widely popular because they are relatively simple, easy to implement, and have significant effects.
[0003] However, the basic BP neural network has the following two main defects: (1) The traditional BP network is a nonlinear optimization problem, and it is inevitable that there will be local minima. The weights and thresholds of the network are constantly modified in the direction of local improvement, trying to achieve a global solution that minimizes the error function, but in fact, what is often obtained is the local optimal point; (2) During the learning process, the error function decreases slowly, the learning speed is slow, and it is easy to have a long-term error flat area, that is, a platform. Therefore, the use of traditional BP neural networks cannot accurately and efficiently control the temperature of integrated stoves, which affects the convenience, accuracy and safety of cooking.
[0004] Therefore, how to provide a neural network-based integrated stove temperature control method, system and device that can solve the above problems is an issue that technical personnel in this field urgently need to solve. Summary of the invention
[0005] In view of this, the present invention provides an integrated stove temperature control method, system and device based on a neural network to solve the technical problems existing in the above-mentioned prior art.
[0006] In order to achieve the above object, the present invention provides a technical solution:
[0007] A temperature control method for an integrated stove based on a neural network comprises the following steps:
[0008] Acquire the operation data of the integrated stove and pre-process the operation data;
[0009] The preprocessed operation data is used as the input of the neural network, and the prediction network model is obtained by learning and training the neural network;
[0010] Predict the temperature change trend of the integrated stove in the future based on the prediction network model and output the prediction results;
[0011] Adjust the temperature control parameters of the integrated stove according to the prediction results.
[0012] Preferably, the integrated stove operation data includes: temperature data, time series data, and power data of the integrated stove.
[0013] Preferably, the preprocessing of the operating data includes:
[0014] De-noising and standardizing the temperature data and power data respectively;
[0015] Perform sliding window word extraction on the processed temperature data and power data to obtain the word vector matrix in the processed operation data;
[0016] Arrange the word vector matrix according to the time series data to obtain the word vector matrix that changes over time.
[0017] Preferably, the structure of the prediction network model includes a convolutional neural network unit and a gated recurrent unit, wherein the convolutional neural network unit extracts spatial features of the operating data, and the gated recurrent unit extracts time series data from the operating data to capture long-term dependencies in the operating data.
[0018] Preferably, the convolutional neural network unit includes: an input layer, a first convolutional layer, a second convolutional layer, a third convolutional layer, a pooling layer, a fully connected layer and a first output layer, and a residual block is set between the first convolutional layer, the second convolutional layer and the third convolutional layer;
[0019] The word vector matrix that changes with time is received through the input layer, and the word vector matrix that changes with time is sequentially convolved through the first convolution layer, the second convolution layer, and the third convolution layer to extract spatial features; maximum pooling is performed through the pooling layer; after classification through the fully connected layer, it is output through the first output layer.
[0020] Preferably, the computational expression of the convolutional neural network is:
[0021]
[0022] Where i=1 is the starting window; f is the activation function; d is the dimension of the word vector, h is the height of the convolution kernel; b is the bias top; W[j,k] is the element in the jth row and kth column of the convolution kernel; X[i+j-1,k] is the kth dimension from the i+j-1th word in the word vector matrix. Since i=1, and the convolution kernel looks forward h words from the current position.
[0023] Preferably, the gated recurrent unit comprises: a GRU layer, a normalized LN layer, an attention mechanism layer and a second output layer;
[0024] The output of the convolutional neural network unit is input into the GRU layer after being normalized by the LN layer. The attention weight of each time step is calculated by the attention mechanism layer. The feature vector output by the GRU layer is weighted and summed to obtain the final result and output it through the second output layer.
[0025] Preferably, the attention mechanism layer calculates the attention weight of each time step, including:
[0026]
[0027] s m =tanh(w T g m );
[0028]
[0029] In the formula, a m is the weight coefficient, g m is the feature vector output by the GRU layer, c m is the spatial eigenvector, s m c m With g m The degree of correlation, w T is the model parameter, s z It means there are z correlation levels in total.
[0030] An integrated stove temperature control system based on a neural network, comprising:
[0031] Acquisition module: acquires the operation data of the integrated stove and pre-processes the operation data;
[0032] Building module: taking the preprocessed running data as the input of the neural network, and obtaining the prediction network model by learning and training the neural network;
[0033] Prediction module: predicts the temperature change trend of the integrated stove in the future based on the prediction network model and outputs the prediction results;
[0034] Control module: adjust the temperature control parameters of the integrated stove according to the prediction results.
[0035] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements a temperature control method for an integrated stove based on a neural network when executing the computer program.
[0036] It can be seen from the above technical solution that compared with the prior art, the present invention discloses a temperature control method, system and device for an integrated stove based on a neural network, which predicts the future temperature change trend of the integrated stove through a predictive network model including a convolutional neural network unit and a gated loop unit, and automatically adjusts the control parameters according to the prediction results to achieve precise temperature control, thereby improving the convenience, accuracy and safety of cooking. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0038] Figure 1 It is a schematic diagram of the method flow of the present invention;
[0039] Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION
[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0041] The purpose of the present invention is to provide an integrated stove temperature control method, system and device based on a neural network, including: obtaining the operation data of the integrated stove and preprocessing the operation data; using the preprocessed operation data as the input of the neural network, and obtaining a prediction network model by learning and training the neural network; predicting the temperature change trend of the integrated stove in the future period of time according to the prediction network model, and outputting the prediction result; adjusting the temperature control parameters of the integrated stove according to the prediction result. A solution is provided for the existing technology that cannot accurately and efficiently control the temperature of the integrated stove.
[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0043] See also Figure 1 The embodiment of the present invention discloses a temperature control method of an integrated stove based on a neural network, comprising the following steps:
[0044] Acquire the operation data of the integrated stove and pre-process the operation data;
[0045] The preprocessed operation data is used as the input of the neural network, and the prediction network model is obtained by learning and training the neural network;
[0046] Predict the temperature change trend of the integrated stove in the future based on the prediction network model and output the prediction results;
[0047] Adjust the temperature control parameters of the integrated stove according to the prediction results.
[0048] In a specific embodiment, the integrated stove operation data includes: temperature data, time series data, and power data of the integrated stove.
[0049] In a specific embodiment, preprocessing the operation data includes:
[0050] De-noising and standardizing the temperature data and power data respectively;
[0051] Perform sliding window word extraction on the processed temperature data and power data to obtain the word vector matrix in the processed operation data;
[0052] Arrange the word vector matrix according to the time series data to obtain the word vector matrix that changes over time.
[0053] Specifically, denoising and cleaning data includes: removing outliers, missing values, etc.
[0054] Standardization processing includes: data formatting: converting time data into a format suitable for model processing (such as time series data). And data normalization / standardization: converting data of different dimensions to the same scale to improve model training efficiency and accuracy.
[0055] In a specific embodiment, the structure of the prediction network model includes a convolutional neural network unit and a gated recurrent unit, wherein the convolutional neural network unit extracts spatial features of the operating data, and the gated recurrent unit extracts time series data in the operating data to capture long-term dependencies in the operating data.
[0056] Specifically, the overall structure of the prediction network model connects convolutional neural network units and gated recurrent units in series. Specifically, the spatial features of the running data are first extracted through the convolutional neural network unit, and then the extracted feature vector (or sequence data) is input into the gated recurrent unit to capture the time series features. Finally, according to the specific task (such as classification, regression, etc.), a fully connected layer and other structures can be added after the gated recurrent unit for output, so that the prediction network model can capture the spatial and temporal features of the data at the same time, thereby understanding the data more comprehensively.
[0057] In a specific embodiment, the convolutional neural network unit includes: an input layer, a first convolutional layer, a second convolutional layer, a third convolutional layer, a pooling layer, a fully connected layer, and a first output layer, and a residual block is set between the first convolutional layer, the second convolutional layer, and the third convolutional layer;
[0058] The word vector matrix that changes with time is received through the input layer, and the word vector matrix that changes with time is sequentially convolved through the first convolutional layer, the second convolutional layer, and the third convolutional layer to extract spatial features; the maximum pooling is performed through the pooling layer; after classification through the fully connected layer, it is output through the first output layer.
[0059] Specifically, the convolutional neural network unit obtains effective information through the above-mentioned input layer, first convolutional layer, second convolutional layer, third convolutional layer, pooling layer, fully connected layer and first output layer structure, automatically generates feature vectors of the data, reduces the difficulty of feature extraction and data reconstruction, and improves the quality of data features.
[0060] In a specific embodiment, the computational expression of the convolutional neural network, for each possible window in X (from the 1st word to the n-h+1th word), is expressed as:
[0061]
[0062] Where i=1, which is the starting window (i is the starting index of the current window, i.e., starting from 1); f is the activation function; d is the dimension of the word vector, h is the height of the convolution kernel; b is the bias top; W[j,k] is the element in the jth row and kth column of the convolution kernel; X[i+j-1,k] is the kth dimension from the i+j-1th word in the word vector matrix, since i=1, and the convolution kernel looks forward h words from the current position.
[0063] In a specific embodiment, the gated recurrent unit includes: a GRU layer, a normalized LN layer, an attention mechanism layer, and a second output layer;
[0064] The output of the convolutional neural network unit is input into the GRU layer after the normalization LN layer. The attention weight of each time step is calculated through the attention mechanism layer. The feature vector output by the GRU layer is weighted and summed to obtain the final result and output through the second output layer, retaining important features and reducing the interference caused by unnecessary features.
[0065] In a specific embodiment, the attention mechanism layer calculates the attention weight for each time step, including:
[0066]
[0067] s m =tanh(w T g m );
[0068]
[0069] In the formula, a m is the weight coefficient, g m is the feature vector output by the GRU layer, c m is the spatial eigenvector, sm c m With g m The degree of correlation, w T are model parameters, sz It means there are z correlation levels in total.
[0070] See also Figure 2 , an integrated stove temperature control system based on neural network, comprising:
[0071] Acquisition module: acquires the operation data of the integrated stove and pre-processes the operation data;
[0072] Building module: taking the preprocessed running data as the input of the neural network, and obtaining the prediction network model by learning and training the neural network;
[0073] Prediction module: predicts the temperature change trend of the integrated stove in the future based on the prediction network model and outputs the prediction results;
[0074] Control module: adjust the temperature control parameters of the integrated stove according to the prediction results.
[0075] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a temperature control method for an integrated stove based on a neural network is implemented.
[0076] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0077] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A temperature control method for an integrated stove based on a neural network, characterized in that: The following steps are involved: Acquire the operation data of the integrated stove and pre-process the operation data; The preprocessed running data is used as an input of a neural network, and a prediction network model is obtained by learning and training the neural network, wherein the structure of the prediction network model includes a convolutional neural network unit and a gated recurrent unit; Predict the temperature change trend of the integrated stove in the future based on the prediction network model and output the prediction results; Adjust the temperature control parameters of the integrated stove according to the prediction results.
2. The temperature control method of an integrated stove based on a neural network according to claim 1, characterized in that: The integrated stove operation data includes: temperature data, time series data, and power data of the integrated stove.
3. The temperature control method of an integrated stove based on a neural network according to claim 2, characterized in that: The preprocessing of the operation data includes: De-noising and standardizing the temperature data and power data respectively; Perform sliding window word extraction on the processed temperature data and power data to obtain the word vector matrix in the processed operation data; Arrange the word vector matrix according to the time series data to obtain the word vector matrix that changes over time.
4. The temperature control method of an integrated stove based on a neural network according to claim 3, characterized in that: The convolutional neural network unit extracts the spatial features of the operating data, and the gated recurrent unit extracts the time series data in the operating data to capture the long-term dependencies in the operating data.
5. The temperature control method of an integrated stove based on a neural network according to claim 4, characterized in that: The convolutional neural network unit includes: an input layer, a first convolutional layer, a second convolutional layer, a third convolutional layer, a pooling layer, a fully connected layer and a first output layer, and a residual block is set between the first convolutional layer, the second convolutional layer and the third convolutional layer; The word vector matrix that changes with time is received through the input layer, and the word vector matrix that changes with time is sequentially convolved through the first convolution layer, the second convolution layer, and the third convolution layer to extract spatial features; maximum pooling is performed through the pooling layer; after classification through the fully connected layer, it is output through the first output layer.
6. The temperature control method of an integrated stove based on a neural network according to claim 4, characterized in that: The computational expression of the convolutional neural network is: Where i=1 is the starting window; f is the activation function; d is the dimension of the word vector, h is the height of the convolution kernel; b is the bias top; W[j,k] is the element in the jth row and kth column of the convolution kernel; X[i+j-1,k] is the kth dimension from the i+j-1th word in the word vector matrix.
7. The temperature control method of an integrated stove based on a neural network according to claim 4, characterized in that: The gated recurrent unit includes: a GRU layer, a normalized LN layer, an attention mechanism layer and a second output layer; The output of the convolutional neural network unit is input into the GRU layer after being normalized by the LN layer. The attention weight of each time step is calculated by the attention mechanism layer. The feature vector output by the GRU layer is weighted and summed to obtain the final result and output it through the second output layer.
8. The temperature control method of an integrated stove based on a neural network according to claim 7, characterized in that: The attention mechanism layer calculates the attention weight for each time step, including: S m =tanh(w T g m ); In the formula, a m is the weight coefficient, g m is the feature vector output by the GRU layer, c m is the spatial eigenvector, sm c m With g m The degree of correlation, w T are model parameters, sz It means there are z correlation levels in total.
9. A neural network-based integrated stove temperature control system using the neural network-based integrated stove temperature control method according to any one of claims 1 to 8, characterized in that: include: Acquisition module: acquires the operation data of the integrated stove and pre-processes the operation data; Building module: taking the preprocessed running data as the input of the neural network, and obtaining a prediction network model by learning and training the neural network, wherein the structure of the prediction network model includes a convolutional neural network unit and a gated recurrent unit; Prediction module: predicts the temperature change trend of the integrated stove in the future based on the prediction network model and outputs the prediction results; Control module: adjust the temperature control parameters of the integrated stove according to the prediction results.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the neural network-based integrated stove temperature control method as described in any one of claims 1 to 8 is implemented.