Data analysis-based control method and system for a rechargeable induction cooker and induction cooker

Abstract

The application discloses a charging type electromagnetic oven control method and system based on data analysis and an electromagnetic oven, and particularly relates to the technical field of electromagnetic oven control; the original energy distribution data of the heating area of the electromagnetic oven is analyzed in time and frequency domains, the time-space change characteristics of the energy distribution are extracted, and the abnormal fluctuation index of the dynamic change of the heat field is identified. The significant feature mode of the position offset of the heating area is extracted through principal component analysis, and the identification model of the energy abnormal concentration area is established; the spatial position and boundary information of the heat concentration abnormal area in the real-time heating process are positioned through cluster analysis, the stability state of the current heating mode is judged, and the comprehensive evaluation result of the heating distribution is output. According to the comprehensive evaluation result, the excitation frequency and output power of each coil in the induction coil array of the electromagnetic oven are adjusted, the real-time balanced control of the heat distribution is realized, the problem that the existing electromagnetic oven is difficult to cope with the dynamic offset of the heating area under the fixed frequency control is effectively solved, and the heating uniformity is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of induction cooker control, and more particularly to a charging type induction cooker control method and system based on data analysis. BACKGROUND

[0002] An induction cooker generates high-frequency electric energy by applying an alternating magnetic field through an induction coil to stimulate heating, and the energy distribution in the actual heating area is prone to dynamic changes.

[0003] Existing induction cooker control systems generally rely on fixed frequency and power control strategies, and are difficult to identify position shifts and shape changes of the heating area under the action of high-frequency electromagnetism, resulting in abnormal heat concentration and making it difficult to effectively solve the problem of heat distribution unevenness causing heating performance degradation.

[0004] With the development and maturity of data analysis processing technology and intelligent control technology, the present application provides an intelligent control technology scheme based on data analysis to solve the above technical problems. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a charging type induction cooker control method and system based on data analysis to solve the problems raised in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0007] A charging type induction cooker control method based on data analysis, comprising the following steps:

[0008] S1: Collecting high-frequency current waveform data of the induction coil at different positions of the induction cooker, and obtaining original energy distribution data of the heating area based on a time sequence synchronization method;

[0009] S2: Based on the original energy distribution data, performing time-frequency domain feature analysis using a wavelet transform method, extracting the time-space variation characteristics of the energy distribution, and determining the abnormal fluctuation index of the thermal field dynamic change;

[0010] S3: Based on the abnormal fluctuation index, identifying the characteristic mode of the position shift of the heating area through a principal component analysis algorithm, and constructing an identification model of the energy abnormal concentration area;

[0011] S4: Using a clustering analysis algorithm, determining the spatial position and boundary information of the heat concentration abnormal area in the real-time heating process based on the identification model;

[0012] S5: Based on the spatial position and boundary information of the heat concentration abnormal area, judging the stability state of the current heating mode, and outputting a comprehensive evaluation result of whether the heating distribution is abnormal;

[0013] S6: dynamically adjusting the excitation frequency and output power of each coil in the induction coil array of the electromagnetic oven according to the comprehensive evaluation result.

[0014] In a preferred embodiment, S1 is specifically:

[0015] Collecting high-frequency current waveform data of each induction coil position in the induction coil array of the electromagnetic oven;

[0016] Based on the time sequence synchronization method, the high-frequency current waveform data of each induction coil position is time marked and aligned to generate high-frequency current waveform data;

[0017] According to the high-frequency current waveform data, the energy distribution value of each induction coil position is calculated to obtain the original energy distribution data of the heating area.

[0018] In a preferred embodiment, S2 is specifically:

[0019] Based on the original energy distribution data of the heating area, the original energy distribution data is analyzed in time and frequency domain by using wavelet transform method;

[0020] By performing multi-scale decomposition operation on the original energy distribution data, time-frequency domain feature components corresponding to different time and frequency are extracted;

[0021] According to the energy value change and spatial position change of the time-frequency domain feature components, the abnormal fluctuation index appearing in the dynamic change process of the thermal field is determined.

[0022] In a preferred embodiment, S3 is specifically:

[0023] Based on the abnormal fluctuation index appearing in the dynamic change process of the thermal field, the principal component analysis algorithm is used for feature dimension reduction processing to determine a number of principal components that mainly affect the dynamic change of the thermal field;

[0024] According to the determined number of principal components, the significant feature mode corresponding to the position shift of the heating area is extracted, and the significant feature mode includes energy concentration region position shift direction and shift amplitude characteristics;

[0025] According to the significant feature mode, an identification model of the energy abnormal concentration region is established.

[0026] In a preferred embodiment, S4 is specifically:

[0027] Based on the identification model of the energy abnormal concentration region, the clustering analysis algorithm is used to cluster the feature mode of the high-frequency current waveform data appearing in the real-time heating process;

[0028] determine the spatial position of the real-time energy concentration abnormal area by a clustering analysis algorithm, and calculate the spatial boundary position of the energy concentration abnormal area;

[0029] According to the spatial boundary position, generate the spatial position information and the spatial boundary information of the real-time heat concentration abnormal area.

[0030] In a preferred embodiment, S5 is specifically:

[0031] Based on the spatial position information and the spatial boundary information of the real-time heat concentration abnormal area, a stability determination model is used to determine the stability state of the real-time heating mode;

[0032] The stability determination model determines whether the real-time heating mode is in a stable state by comparing the difference between the spatial distribution characteristics of the real-time heat concentration abnormal area and the preset heat spatial distribution characteristics in a stable state;

[0033] According to the determination result of the stability determination model, a comprehensive evaluation result indicating whether the heating distribution in the real-time heating mode is abnormal is generated.

[0034] In a preferred embodiment, S6 is specifically:

[0035] According to the comprehensive evaluation result indicating whether the heating distribution in the real-time heating mode is abnormal, determine the induction coil that needs to adjust the excitation frequency and the output power;

[0036] For the induction coil that needs to be adjusted, the excitation frequency adjustment amount and the output power adjustment amount of the induction coil are calculated according to the spatial distribution characteristics and the spatial boundary information of the real-time heat concentration abnormal area;

[0037] The calculated excitation frequency adjustment amount and output power adjustment amount of the induction coil are applied to the electromagnetic induction coil array to adjust the excitation frequency and output power of each induction coil in the electromagnetic induction coil array in real time.

[0038] On the other hand, the application provides a charging type electromagnetic induction cooker control system based on data analysis, comprising:

[0039] Data acquisition module: acquire high-frequency current waveform data of different positions of the electromagnetic induction coil, and obtain original energy distribution data of the heating area based on a time sequence synchronization method;

[0040] Feature extraction module: based on the original energy distribution data, perform time-frequency domain feature analysis by wavelet transform method, extract the time-space variation characteristics of the energy distribution, and determine the abnormal fluctuation index of the heat field dynamic change;

[0041] Pattern recognition module: based on abnormal fluctuation indicators, identify the characteristic pattern of heating area position deviation through principal component analysis algorithm, and build an identification model of energy abnormal concentration area;

[0042] Area positioning module: using clustering analysis algorithm, determine the spatial position and boundary information of the heat concentration abnormal area in real-time heating process based on the identification model;

[0043] Stability evaluation module: based on the spatial position and boundary information of the heat concentration abnormal area, judge the stability state of the current heating mode, and output the comprehensive evaluation result of whether the heating distribution is abnormal;

[0044] Parameter control module: dynamically adjust the excitation frequency and output power of each coil in the induction coil array of the induction cooker according to the comprehensive evaluation result.

[0045] On the other hand, the application provides a rechargeable induction cooker, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to realize the above-mentioned data analysis-based rechargeable induction cooker control method.

[0046] Beneficial effects: compared with the prior art, the application collects high-frequency current waveform data at different positions of the induction coil of the induction cooker, obtains original energy distribution data of the heating area based on a time sequence synchronization method, realizes the improvement of spatial resolution of the heating state, extracts time-frequency domain features through wavelet transform, can capture dynamic changes of energy distribution on multiple scales, and identify heat field instability characteristics; combined with the principal component analysis algorithm, identify the heat abnormal concentration trend, and build an identification model, which has the ability to identify the position deviation of the heating area; using clustering analysis algorithm to determine the spatial position and boundary information of the heat concentration abnormal area, enhances the positioning accuracy of the abnormality; through the stability state determination mechanism, output the comprehensive evaluation result of the heating distribution abnormality, realize the discrimination of the heat distribution state; through dynamic adjustment of the excitation frequency and output power, fine control of the output response of each induction coil, make the magnetic field action and heat distribution in the heating area realize real-time coupling matching, overall improve the heating uniformity, adaptability and intelligent control level of the induction cooker. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 A data analysis-based rechargeable induction cooker control method of the application is shown in the figure;

[0048] Figure 2 A structure diagram of a data analysis-based rechargeable induction cooker control system of the application is shown in the figure. DETAILED DESCRIPTION

[0049] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those ordinarily skilled in the art without creative work fall within the scope of the present application.

[0050] Embodiment 1

[0051] Figure 1 A charging type electromagnetic oven control method based on data analysis is given, which comprises the following steps:

[0052] S1: collecting high-frequency current waveform data of the electromagnetic oven induction coil at different positions, and obtaining original energy distribution data of the heating area based on a time sequence synchronization method;

[0053] S2: based on the original energy distribution data, performing time-frequency domain feature analysis by using a wavelet transform method, extracting time-space change features of the energy distribution, and determining an abnormal fluctuation index of the thermal field dynamic change;

[0054] S3: based on the abnormal fluctuation index, identifying a characteristic mode of the heating area position offset by using a principal component analysis algorithm, and constructing an identification model of the energy abnormal concentration area;

[0055] S4: determining spatial position and boundary information of the heat concentration abnormal area in a real-time heating process based on the identification model by using a clustering analysis algorithm;

[0056] S5: judging the stability state of the current heating mode based on the spatial position and boundary information of the heat concentration abnormal area, and outputting a comprehensive evaluation result of whether the heating distribution is abnormal;

[0057] S6: dynamically adjusting the excitation frequency and output power of each coil in the electromagnetic oven induction coil array according to the comprehensive evaluation result.

[0058] S1: collecting high-frequency current waveform data of the electromagnetic oven induction coil at different positions, and obtaining original energy distribution data of the heating area based on a time sequence synchronization method, comprising:

[0059] collecting high-frequency current waveform data of each induction coil position in the electromagnetic oven induction coil array;

[0060] Specifically, the induction coil array of the induction cooker includes multiple independently controlled induction coils, each equipped with a sensor for measuring high-frequency current waveforms. The high-frequency current waveform data includes the amplitude, phase, and frequency of the high-frequency current. The amplitude reflects the energy output of the coil, the phase reflects the timing relationship between the induction coils, and the frequency reflects the energy transfer characteristics. The measurement method for the amplitude, phase, and frequency of the high-frequency current is as follows: a high-speed current sensor and a synchronous clock module are installed at each induction coil location. The high-speed current sensor measures the amplitude of the high-frequency current in each coil in real time, while the phase detection module records the current phase of each coil in real time. A frequency meter is used to measure the frequency of the current waveform of each coil in real time to obtain the high-frequency current waveform information.

[0061] Based on the timing synchronization method, the high-frequency current waveform data at each induction coil position is time-marked and aligned to generate high-frequency current waveform data;

[0062] Specifically, the timing synchronization method is implemented as follows: First, a unified reference clock source is set up in the induction coil array of the induction cooker. The reference clock source is connected to the synchronization clock module at each induction coil position to achieve synchronized distribution of the unified clock signal. The high-frequency current waveform data collected at each induction coil position is marked with timestamp information provided by the reference clock source, thus enabling all collected high-frequency current waveform data to be aligned and synchronized using a unified time scale, ensuring accurate time correspondence between data at different positions. The collected high-frequency current waveform data with timestamp information is uploaded to a cloud server via a wireless network, and the cloud server performs timing synchronization of the high-frequency current waveform data based on the timestamp information.

[0063] After achieving data timing synchronization, the high-frequency current waveform data at each induction coil position is combined to form high-frequency current waveform data. Specifically, the high-frequency current waveform data records the current amplitude, current phase, and current frequency at each induction coil position at every moment, forming a continuous set of information that can reflect the real-time energy state of the entire heating area.

[0064] Based on the high-frequency current waveform data, the energy distribution value at each induction coil position is calculated to obtain the original energy distribution data of the heating area.

[0065] Specifically, the method for calculating the energy distribution value at each inductive coil position is as follows: first, the instantaneous power value at the position corresponding to each inductive coil is calculated using the synchronized high-frequency current waveform data. The instantaneous power value is calculated by multiplying the measured high-frequency current amplitude at each inductive coil with the instantaneous voltage amplitude at the corresponding position of the coil, and correcting through the power factor. The power factor is determined by the phase difference of the high-frequency current, and is represented as the cosine value of the phase difference between the high-frequency current and the coil voltage. When calculating the instantaneous power value, the correction factor is included in the product, i.e. the instantaneous power value is equal to the current amplitude multiplied by the voltage amplitude multiplied by the cosine value of the phase difference between the current amplitude and the voltage amplitude.

[0066] The instantaneous power value at the position corresponding to each inductive coil is integrated under a unified time reference to obtain the energy cumulative value at each inductive coil position within a certain time period. By summarizing and arranging the energy cumulative values at each inductive coil position, the original energy distribution data of the electromagnetic stove heating area can be obtained. The original energy distribution data reflects the real-time distribution characteristics of energy transmission at each inductive coil position, and can show the energy concentration at each position in the entire electromagnetic stove heating area. The original energy distribution data is a set of energy cumulative values obtained by integrating a plurality of power change sequences generated based on each inductive coil position within a certain time period.

[0067] S2: Based on the original energy distribution data, a wavelet transform method is used for time-frequency domain feature analysis to extract the spatio-temporal variation characteristics of the energy distribution and determine the abnormal fluctuation index of the thermal field dynamic change, including:

[0068] Based on the original energy distribution data of the heating area, the wavelet transform method is used to jointly analyze the time domain and frequency domain of the original energy distribution data;

[0069] Specifically, after obtaining the original energy distribution data of the electromagnetic stove heating area, in order to analyze the dynamic characteristic change of the energy distribution of the electromagnetic stove in the real-time heating process, the wavelet transform method is selected to jointly analyze the time domain and frequency domain of the original energy distribution data. The wavelet transform can reflect the variation characteristics of the signal in the time dimension and the frequency dimension, can identify the local abnormal information under different time points and different frequency components, and can identify the energy abnormal change of the inductive coil array heating area of the electromagnetic stove.

[0070] By performing a multi-scale decomposition operation on the original energy distribution data, time-frequency domain feature components corresponding to different time instants and different frequencies are extracted;

[0071] Specifically, data preprocessing is performed on the original energy distribution data to ensure the effectiveness of the wavelet analysis. The preprocessing method includes removing abnormal data points caused by measurement noise or data acquisition errors, and correcting data missing by interpolation method to ensure the continuity and smoothness of the data.

[0072] After the data preprocessing is completed, the original energy distribution data is continuously subjected to multiple wavelet transform decomposition operations to obtain low-frequency and high-frequency components of different scales: the original energy distribution data is subjected to one-time wavelet decomposition to obtain an approximate low-frequency component and several detail high-frequency components; the obtained approximate low-frequency component is further subjected to two-time, three-time or even higher-time multi-scale wavelet decomposition to gradually obtain characteristic components at different scales.

[0073] From the low-frequency and high-frequency components of different scales obtained through multi-scale decomposition, the energy values and spatial positions corresponding to each specific moment are extracted to form a complete set of time-frequency domain characteristic components. For example, at a specific moment, the signal intensity on different scale components is calculated, the energy sizes of different positions are analyzed, and the corresponding positions and scales are recorded to form a set of time-frequency domain characteristic components. The specific calculation method is as follows: for each scale characteristic component, the energy value at the corresponding moment is obtained by squaring and summing the energy distribution values at the scale.

[0074] According to the energy value changes and spatial position changes of the time-frequency domain characteristic components, an abnormal fluctuation index appearing in the dynamic change process of the thermal field is determined.

[0075] Specifically, the determination method of the abnormal fluctuation index is as follows: the energy value change trend of the time-frequency domain characteristic components at different time points and different scales is analyzed, the abnormal fluctuation signals obviously deviating from the normal preset range are identified, and the position, intensity and frequency scale of the abnormal fluctuation signals are recorded. The abnormal fluctuation index is defined as follows: at each specific moment, the energy values of the characteristic components extracted at each frequency scale are analyzed, the difference between the energy values at adjacent moments at a specific frequency scale is taken as an absolute value to obtain the fluctuation intensity of the energy at the frequency scale; the difference between the energy values of adjacent spatial positions at the same moment is also taken as an absolute value to determine the energy difference amplitude of the spatial position; the fluctuation intensity at the frequency scale and the energy difference amplitude of the spatial position are multiplied to obtain the abnormal fluctuation value under the joint action of the frequency scale and the spatial position; the abnormal fluctuation values obtained under the joint action of all frequency scales and spatial positions are accumulated to obtain the abnormal fluctuation index of the entire dynamic change process of the thermal field.

[0076] S3: based on the abnormal fluctuation index, a feature mode of the position deviation of the heating area is identified through principal component analysis algorithm to construct an identification model of the energy abnormal concentration area, including:

[0077] Based on the abnormal fluctuation index appearing in the dynamic change process of the thermal field, the principal component analysis algorithm is used for feature dimension reduction processing to determine several principal components mainly affecting the dynamic change of the thermal field.

[0078] Specifically, the basic principle of the principal component analysis algorithm is to project high-dimensional feature data into a new lower-dimensional space to remove redundant information in the data and only keep key features that can fully describe the changes of the original data. The abnormal fluctuation indicators are standardized. The standardization method is to subtract the average value of the overall abnormal fluctuation indicators from each abnormal fluctuation indicator and then divide by the standard deviation of the overall abnormal fluctuation indicators.

[0079] The standardized abnormal fluctuation indicators are input into the principal component analysis algorithm, and the eigenvalues corresponding to the eigenvectors are determined by calculating the covariance matrix of the abnormal fluctuation indicators and the eigenvalues and eigenvectors of the covariance matrix. The eigenvalue represents the degree of contribution of the principal component represented by the corresponding eigenvector to the overall abnormal fluctuation indicator change. The principal component analysis algorithm sorts the eigenvalues according to their sizes, and selects the eigenvectors with larger eigenvalues and cumulative contribution rates above a predetermined threshold to combine a few principal components. The selection criteria for eigenvalues are: sort the eigenvalues from large to small, accumulate the contribution rate of each eigenvalue, i.e. the proportion of each eigenvalue in the sum of all eigenvalues, and stop selecting when the cumulative contribution rate reaches a predetermined threshold (e.g. more than ninety percent), thereby ensuring that the principal components can fully represent the main feature information of the original data.

[0080] According to the determined principal components, the significant feature patterns corresponding to the position shift of the energy concentration region are extracted, including the energy concentration region position shift direction and shift amplitude features;

[0081] The significant feature pattern includes the energy concentration region position shift direction and shift amplitude. The method for extracting the significant feature pattern is: first, according to the obtained principal components, taking the energy concentration region distribution features of each position under the normal state of the heating area as the reference benchmark, comparing the real-time principal components, calculating the difference size and spatial difference features between the real-time state and the normal state, and identifying whether the energy concentration region appears obvious spatial shift phenomenon; calculating the position change of the shifted region, i.e. the shift direction (such as along the long axis or short axis direction of the inductor coil array) and the shift amplitude (specifically expressed as the distance difference between the center position of the shifted region and the normal position).

[0082] The method for calculating the difference size and spatial difference features between the real-time state and the normal state is:

[0083] The normal state principal component feature data, i.e. the principal component data set obtained by pre-recording or training under the normal working state of the induction cooker, are determined as the standard reference benchmark data.

[0084] The abnormal fluctuation indicator data of the real-time state is projected into the principal component space through the same principal component analysis method to obtain the principal component feature data set of the real-time state.

[0085] Comparing the difference between the real-time state principal component feature data and the normal state principal component feature data in the principal component space position by position:

[0086] Corresponding to each inductive coil position, the difference between the real-time state principal component feature data and the normal state principal component feature data is calculated respectively;

[0087] The absolute values of the above differences are added to quantify the difference size.

[0088] The method for calculating the spatial difference feature is:

[0089] Analyzing the positions where the real-time state differs most from the normal state in the principal component space;

[0090] Calculating the amplitude and spatial distribution range of the position principal component data deviation to determine the position and area range where the obvious spatial abnormal fluctuation occurs.

[0091] Determination of the offset direction: determining the spatial coordinates of each inductive coil array in the heating area (such as explicitly dividing the inductive coil array space into a long axis and a short axis two spatial directions);

[0092] Taking the position center of the energy concentration area in the normal state as the original reference, the coordinates of the position center of the energy concentration area in the real-time state are determined;

[0093] By the difference in coordinate position, the change direction of the real-time state energy concentration area center relative to the normal state center is determined, and the offset direction is judged to be along the long axis direction, the short axis direction, or a certain specific direction between the two axes;

[0094] For example, if the real-time state energy concentration area center coordinates move more relative to the normal state center coordinates in the long axis direction, the offset direction is defined as along the long axis direction.

[0095] The calculation method of the offset amplitude is: first, determine the center position of the energy concentration area in the normal working state of the electromagnetic oven heating area, and the center position is obtained by weighted average of the energy distribution values at each inductive coil position in the normal working state. Then determine the center position of the energy concentration area in the real-time state, that is, when the energy abnormal concentration or offset occurs, the current actual center position is calculated by the same weighted average method through the real-time obtained energy distribution value corresponding to each inductive coil position. The actual center position calculated in real time and the normal state center position are compared respectively in the electromagnetic oven inductive coil array space coordinates, and the specific calculation method is to calculate the Euclidean distance of the two center positions in the spatial coordinates to determine the distance size of the position offset. The distance size of the position offset is defined as the offset amplitude, which reflects the actual spatial offset degree of the real-time energy concentration area relative to the normal state center position.

[0096] According to the significant feature mode, an identification model of the energy abnormal concentration area is established;

[0097] Specifically, according to the obtained significant feature mode including the position offset direction and the offset amplitude, an identification model of the energy abnormal concentration area is established. The establishment method of the identification model is as follows: taking the obtained significant feature mode data as training data, selecting a supervised machine learning method (such as a support vector machine or a decision tree algorithm), taking the significant feature mode data as input, and taking the spatial position label corresponding to the energy abnormal concentration area as output, a model capable of automatically identifying the abnormal energy concentration area is trained. A plurality of sets of actual collected energy concentration abnormal area significant feature mode data are used for model training and cross-validation, and through multiple verifications and optimization of parameters such as the selection of the kernel function in the support vector machine algorithm or the node division threshold in the decision tree algorithm, the optimal model parameter configuration is determined to ensure that the model has high recognition accuracy and stability.

[0098] S4: using a clustering analysis algorithm, determining the spatial position and boundary information of the heat concentration abnormal area in the real-time heating process based on the identification model, including:

[0099] Based on the identification model of the energy abnormal concentration area, a clustering analysis algorithm is used to cluster the feature mode of the high-frequency current waveform data appearing in the real-time heating process;

[0100] The spatial position of the energy concentration abnormal area appearing in real time is determined by the clustering analysis algorithm, and the spatial boundary position of the energy concentration abnormal area is calculated;

[0101] According to the spatial boundary position, the spatial position information and the spatial boundary information of the real-time heat concentration abnormal area are generated;

[0102] Specifically, in order to accurately locate the energy abnormal concentration area appearing in the actual operation of the induction cooker, clustering analysis needs to be performed on the real-time collected high-frequency current waveform data to accurately locate the position and boundary information of the energy abnormal concentration area in real-time state.

[0103] In the real-time operation process, according to the identification model of the energy abnormal concentration area, the feature mode of the real-time high-frequency current waveform data collected from each position of the induction cooker induction coil array is subjected to clustering analysis. The clustering analysis algorithm used is density clustering analysis. The energy abnormal concentration area appearing in the induction coil array region of the induction cooker has obvious spatial concentration characteristics, and the density clustering analysis algorithm can effectively identify the spatial dense area in the data distribution.

[0104] First, the real-time high-frequency current waveform data is preprocessed and features are extracted. The feature extraction is to extract the feature mode significantly related to the energy concentration area from the high-frequency current waveform data obtained in real time at each inductive coil position, including the amplitude feature of the high-frequency current at each coil position, the phase difference feature of the high-frequency current, and the frequency change feature of the high-frequency current. After feature extraction, all feature data is standardized to ensure uniform data scale between different features, facilitating clustering analysis.

[0105] The standardized feature data is used as input, and the density clustering analysis algorithm is used to cluster the data. The process of implementing the density clustering analysis algorithm is as follows: first, determine the data point density determination parameters in the data space, including the data density threshold parameter and the minimum data sample number parameter. The determination method is: through repeated clustering experiments on historical data, adjust the parameters until the clustering results can stably and effectively identify the abnormal energy concentration areas confirmed in history. After determining the data density threshold parameter and the minimum data sample number parameter, the density clustering analysis algorithm is executed according to the following method: determine whether the number of points within a specified spatial radius around each data point in the data space reaches the set minimum sample number one by one, if so, define the region as a density core area; then connect all density core areas to form larger spatial aggregation areas until all energy abnormal concentration areas are identified.

[0106] Through the above clustering process, the specific spatial position of the real-time energy concentration abnormal area is determined. The determination of the spatial position is expressed as the center position of each cluster, that is, by averaging the spatial coordinate positions of all data points in each cluster, the specific spatial center position of the abnormal concentration area is obtained, which is located in each specific coil position area of the induction coil array of the electromagnetic oven.

[0107] In order to determine the specific spatial boundary position of the energy concentration abnormal area, the boundary position of each cluster is calculated. The calculation method of the boundary position is: for each determined energy abnormal concentration area, starting from the spatial center position of the cluster, the spatial distance of the data points from the center position in each direction is calculated one by one, when the energy data point density of the distance from the center position is significantly lower than the density threshold parameter set by the density clustering algorithm, the distance position is defined as the spatial boundary position of the region; through the spatial distance calculation in different directions, the complete cluster spatial boundary position data is obtained.

[0108] According to the calculated spatial boundary position, a set of spatial position information and spatial boundary information of the real-time heat concentration abnormal area is generated: taking the center spatial position of each cluster as a reference, combining the calculated spatial boundary position in each direction, the spatial range and contour boundary of the energy abnormal concentration area in the electromagnetic oven heating area under the real-time state are marked, and a data set of the specific real-time heat concentration abnormal area spatial position information and spatial boundary information is formed.

[0109] S5: Based on the spatial position and boundary information of the heat concentration abnormal area, the stability state of the current heating mode is judged, and a comprehensive evaluation result of whether the heating distribution is abnormal is output, including:

[0110] Based on the spatial position information and spatial boundary information of the real-time heat concentration abnormal area, a stability determination model is used to judge the stability state of the real-time heating mode;

[0111] The stability determination model compares the difference between the spatial distribution characteristics of the real-time heat concentration abnormal area and the preset heat spatial distribution characteristics under the stable state, and determines whether the real-time heating mode is in a stable state;

[0112] According to the determination result of the stability determination model, a comprehensive evaluation result indicating whether the heating distribution in the real-time heating mode is abnormal is generated;

[0113] Specifically, the stability determination model requires defining the heat spatial distribution characteristics of the electromagnetic oven heating mode in the stable state as the stable reference characteristic data. The determination method of the stable reference characteristic data is: under the designed best stable working condition of the electromagnetic oven induction coil array, the energy distribution data at each induction coil position is collected for a long time, and these data are integrated and averaged in time to determine the spatial position information and boundary characteristics of the energy distribution under the stable state, which ensures the representativeness and accuracy of the stable reference characteristic data.

[0114] In real-time operation, the stability determination model compares the spatial position information and the spatial boundary information of the real-time heat concentration abnormal area with the pre-determined stable benchmark feature data based on the spatial distribution. The spatial distribution comparison analysis is specifically as follows: taking the spatial position determined by the stable benchmark feature data as a reference standard, the spatial overlap degree and the spatial difference degree of the real-time heat concentration abnormal area and the stable state are calculated in real time, including spatial coincidence rate calculation and spatial offset calculation. The spatial coincidence rate calculation method is as follows: the real-time heat concentration abnormal area and the stable benchmark area are superimposed in space, and the ratio of the spatial intersection area to the stable benchmark area is calculated to determine the spatial overlap degree of the real-time state and the stable state. The spatial offset calculation method is as follows: the spatial distance between the center positions of the real-time heat concentration area and the stable benchmark area is calculated to represent the spatial position deviation degree of the real-time heat concentration area relative to the stable state.

[0115] After obtaining the spatial coincidence rate and the spatial offset, the stability determination model determines the stability state according to the pre-defined threshold rule: through statistical analysis of a large amount of historical stable and non-stable state data in the actual operation state of the induction cooker, a minimum threshold of the spatial coincidence rate and a maximum threshold of the spatial offset are set. When the spatial coincidence rate calculated in real time is lower than the minimum threshold of the spatial coincidence rate or the spatial offset exceeds the maximum threshold, it is determined that the current real-time heating mode is in a non-stable state; otherwise, it is considered that the real-time heating mode is in a stable state.

[0116] In order to ensure the accuracy and reliability of the stability determination model, the specific values of the spatial coincidence rate and the spatial offset threshold parameters are optimized through multiple training and verification of historical data during the model establishment process. The optimization method is as follows: the actual abnormal state samples and the normal stable state samples in the historical data are repeatedly calculated, and the spatial coincidence rate and the spatial offset threshold are continuously adjusted to accurately distinguish the historical state until the highest determination accuracy of the stability determination model is achieved, so as to determine the best determination threshold parameters.

[0117] The stability determination model comprehensively analyzes the spatial difference calculation results of the real-time state and the stable state, and outputs the comprehensive evaluation results of whether the real-time heating mode is in a stable state according to the calculated spatial coincidence rate and spatial offset. The generation method of the comprehensive evaluation results is as follows: when the spatial coincidence rate calculated in real time is lower than the minimum threshold of the spatial coincidence rate or the spatial offset exceeds the maximum threshold, the comprehensive evaluation result is heating distribution abnormality, indicating that the real-time heating mode is in a non-stable state and needs heating adjustment control measures; otherwise, the comprehensive evaluation result is heating distribution normality, indicating that the real-time heating mode is in a stable state.

[0118] The above stability determination model can accurately capture the slight deviation changes in the spatial position of the real-time energy concentration area, timely discover potential heating imbalance or abnormal conditions, and ensure that the electromagnetic oven is always in a safe and stable working state, thereby effectively improving the heating precision, efficiency and service life of the electromagnetic oven.

[0119] S6: dynamically adjusting the excitation frequency and output power of each coil in the array of induction coils of the electromagnetic oven according to the comprehensive evaluation result, including:

[0120] According to the comprehensive evaluation result indicating whether the heating distribution in the real-time heating mode is abnormal, determine the induction coil that needs to adjust the excitation frequency and output power;

[0121] Specifically, according to the comprehensive evaluation result, determine the induction coil that needs to be dynamically adjusted: when the comprehensive evaluation result is that the heating distribution is abnormal, i.e., according to the spatial position information and spatial boundary information of the real-time heat concentration abnormal area, identify and determine the induction coil in the abnormal area or near the abnormal area, i.e., the specific coil that needs to be dynamically adjusted. Compare and match the spatial position of the real-time spatial abnormal area with the layout position of the array of induction coils of the electromagnetic oven, and determine the induction coil in the spatial abnormal area as the target adjustment object.

[0122] For the induction coil that needs to be adjusted, according to the spatial distribution characteristics and spatial boundary information of the real-time heat concentration abnormal area, calculate the excitation frequency adjustment amount and output power adjustment amount of the induction coil;

[0123] Specifically, according to the spatial distribution characteristics and spatial boundary information of the real-time heat concentration abnormal area, calculate the target energy output state that each induction coil needs to reach, i.e., set the target output energy distribution value of each coil position. The determination method of the target output energy distribution value is: compare the energy concentration state of the current abnormal area with the ideal energy distribution in the stable state, calculate the energy difference between the two, and determine the energy output target required for each induction coil according to the size and spatial position of the difference value.

[0124] Calculate the excitation frequency adjustment amount and output power adjustment amount of each specific induction coil respectively:

[0125] The excitation frequency adjustment amount calculation method is: analyze the influence relationship of the current induction coil real-time frequency on the energy transmission efficiency and spatial distribution characteristics, and determine the quantitative relationship between the high-frequency current frequency change and the induction heating depth and regional distribution change; according to the quantitative relationship, calculate the frequency difference value between the current energy output state and the target output state of each induction coil, and determine the frequency value required for each specific coil to be adjusted through linear or nonlinear proportional relationship: the frequency adjustment amount is equal to the difference value between the current frequency and the target frequency obtained according to the target energy output state.

[0126] The calculation method of the output power adjustment amount is as follows: first, the difference between the current real-time output power of each induction coil and the target power is determined, the real-time coil current amplitude and voltage amplitude are calculated, and power correction is performed according to the target power requirement. The power correction method is as follows: the target output power is subtracted from the current output power, and then corrected according to the corresponding power factor, so as to obtain the output power value that each induction coil needs to adjust.

[0127] The calculation parameters (such as the coefficient of the relationship between frequency and energy distribution, the correction coefficient of power adjustment) of the above frequency power dynamic adjustment algorithm are verified and optimized through a large amount of historical operation data.

[0128] The calculated excitation frequency adjustment amount and output power adjustment amount of the induction coil are applied to the induction coil array of the induction cooker respectively, and the excitation frequency and output power of each induction coil in the induction coil array of the induction cooker are adjusted in real time.

[0129] Specifically, the cloud server automatically issues an instruction based on the analysis result, and the induction cooker sends the calculated excitation frequency adjustment amount and output power adjustment amount of the induction coil to the frequency control module and the power control module of the induction cooker respectively according to the instruction, and realizes real-time frequency and power adjustment of each induction coil through a high-precision frequency regulator and a power regulator. After real-time adjustment, the actual output state of each coil in the induction coil array of the induction cooker will be consistent with the target state, the abnormal energy concentration area is corrected in time, so as to ensure that the overall thermal field distribution quickly recovers to a stable state.

[0130] Embodiment 2

[0131] The difference between the embodiment 2 and the embodiment 1 of the present application is that the embodiment 2 is to introduce a charging type induction cooker control system based on data analysis.

[0132] Figure 2 The structure diagram of the charging type induction cooker control system based on data analysis is given, and the charging type induction cooker control system based on data analysis comprises:

[0133] The data acquisition module acquires high-frequency current waveform data of the induction coil at different positions of the induction cooker, and obtains original energy distribution data of the heating area based on a time sequence synchronization method.

[0134] The feature extraction module performs time-frequency domain feature analysis on the original energy distribution data by using a wavelet transform method, extracts the time-space change feature of the energy distribution, and determines the abnormal fluctuation index of the thermal field dynamic change.

[0135] Pattern recognition module: based on the abnormal fluctuation index, the characteristic pattern of the heating area position deviation is recognized through the principal component analysis algorithm, and the identification model of the energy abnormal concentration area is constructed;

[0136] Region positioning module: the clustering analysis algorithm is adopted, and the spatial position and boundary information of the heat concentration abnormal area in the real-time heating process are determined based on the identification model;

[0137] Stability evaluation module: based on the spatial position and boundary information of the heat concentration abnormal area, the stability state of the current heating mode is judged, and the comprehensive evaluation result of whether the heating distribution is abnormal is output;

[0138] Parameter control module: according to the comprehensive evaluation result, the excitation frequency and output power of each coil in the induction coil array of the electromagnetic induction cooker are dynamically adjusted.

[0139] Embodiment 3

[0140] The application provides a kind of rechargeable electromagnetic induction cooker, including memory, processor and computer program stored in memory and can be run on processor, when processor executes computer program, realize the control method of rechargeable electromagnetic induction cooker based on data analysis described above.

[0141] The rechargeable electromagnetic induction cooker provided by the application not only has charging and discharging functions, overcomes the limitation of traditional electric electromagnetic induction cooker power cord bundle, can meet different use scenarios, also has a rechargeable electromagnetic induction cooker control system based on data analysis, which enables the rechargeable electromagnetic induction cooker to have data analysis function, can make the electromagnetic induction cooker heat more evenly, has higher adaptability and intelligent control level, compared with the traditional electromagnetic induction cooker on the market, has better application prospect and commercial value.

[0142] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center through a wired (for example, infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0143] Those of ordinary skill in the art can realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0144] Those of ordinary skill in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device, and module can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0145] In several embodiments provided in the present application, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed ones can be indirect coupling or communication connection through some interfaces, devices, or modules, which can be electrical, mechanical, or other forms.

[0146] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0147] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0148] If the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products, and the computer software products are stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and various program code storage media.

[0149] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0150] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A control method for a rechargeable induction cooker based on data analysis, characterized in that, Includes the following steps: S1: Collect high-frequency current waveform data at different locations of the induction coil of the induction cooker, and obtain the original energy distribution data of the heating area based on the time-series synchronization method, specifically: Collect high-frequency current waveform data at each position of the induction coil in the induction coil array of the induction cooker; Based on the timing synchronization method, the high-frequency current waveform data at each induction coil position is time-marked and aligned to generate high-frequency current waveform data; Based on the high-frequency current waveform data, the energy distribution value at each induction coil position is calculated to obtain the original energy distribution data of the heating area. S2: Based on the original energy distribution data, wavelet transform is used to perform time-frequency domain feature analysis, extract the spatiotemporal variation characteristics of energy distribution, and determine the abnormal fluctuation index of dynamic changes in the thermal field. S3: Based on the abnormal fluctuation index, the characteristic patterns of heating area positional shifts are identified using principal component analysis algorithm, and an identification model for areas of abnormal energy concentration is constructed, specifically: Based on the abnormal fluctuation indicators that appear during the dynamic changes of the thermal field, the principal component analysis algorithm is used to perform feature dimensionality reduction to determine several principal components that play a major role in the dynamic changes of the thermal field. Based on the identified principal components, significant feature patterns are extracted when the position of the corresponding heating region shifts. The significant feature patterns include the shift direction and magnitude of the energy concentration region position. Based on salient feature patterns, establish an identification model for regions with concentrated energy anomalies; S4: Using a clustering analysis algorithm, the spatial location and boundary information of abnormal heat concentration areas during real-time heating are determined based on the identification model; S5: Based on the spatial location and boundary information of the abnormal heat concentration area, determine the stability state of the current heating mode and output a comprehensive evaluation result on whether the heating distribution is abnormal; S6: Dynamically adjust the excitation frequency and output power of each coil in the induction coil array of the induction cooker based on the comprehensive evaluation results.

2. The data analysis-based control method for a rechargeable induction cooker according to claim 1, characterized in that, S2, specifically: Based on the original energy distribution data of the heating region, wavelet transform method is used to perform joint time-domain and frequency-domain analysis on the original energy distribution data; By performing multi-scale decomposition on the original energy distribution data, time-frequency domain feature components corresponding to different times and frequencies are extracted; Based on the changes in energy values ​​and spatial locations of the characteristic components in the time and frequency domains, the abnormal fluctuation indicators that occur during the dynamic changes of the thermal field are determined.

3. The data analysis-based control method for a rechargeable induction cooker according to claim 2, characterized in that, S4, specifically: Based on the identification model of energy anomaly concentration areas, a clustering analysis algorithm is used to cluster the characteristic patterns of high-frequency current waveform data that appear during real-time heating. The spatial location of real-time energy concentration anomaly regions is determined by cluster analysis algorithm, and the spatial boundary location of the energy concentration anomaly regions is calculated. Based on the spatial boundary location, generate real-time spatial location information and spatial boundary information of the heat concentration anomaly area.

4. The data analysis-based control method for a rechargeable induction cooker according to claim 3, characterized in that, S5, specifically: Based on the spatial location and boundary information of the real-time heat concentration anomaly area, a stability judgment model is used to judge the stability state of the real-time heating mode. The stability assessment model determines whether the real-time heating mode is in a stable state by comparing the spatial distribution characteristics of the real-time heat concentration anomaly area with the heat spatial distribution characteristics under the preset stable state. Based on the determination results of the stability judgment model, a comprehensive evaluation result is generated indicating whether the heating distribution in the real-time heating mode is abnormal.

5. The data analysis-based control method for a rechargeable induction cooker according to claim 4, characterized in that, S6, specifically: Based on the comprehensive evaluation results indicating whether the heating distribution is abnormal in the real-time heating mode, the induction coils whose excitation frequency and output power need to be adjusted are determined. For induction coils that need adjustment, the excitation frequency adjustment and output power adjustment are calculated based on the spatial distribution characteristics and spatial boundary information of the real-time heat concentration anomaly area. The calculated excitation frequency adjustment and output power adjustment of the induction coil are applied to the induction coil array of the induction cooker to adjust the excitation frequency and output power of each induction coil in the induction coil array in real time.

6. A data analysis-based control system for a rechargeable induction cooker, used to implement the data analysis-based control method for a rechargeable induction cooker as described in any one of claims 1-5, characterized in that, include: Data acquisition module: Collects high-frequency current waveform data at different locations of the induction coil of the induction cooker, and obtains the original energy distribution data of the heating area based on a time-series synchronization method, specifically: Collect high-frequency current waveform data at each position of the induction coil in the induction coil array of the induction cooker; Based on the timing synchronization method, the high-frequency current waveform data at each induction coil position is time-marked and aligned to generate high-frequency current waveform data; Based on the high-frequency current waveform data, the energy distribution value at each induction coil position is calculated to obtain the original energy distribution data of the heating area. Feature extraction module: Based on the original energy distribution data, wavelet transform is used to perform time-frequency domain feature analysis, extract the spatiotemporal variation features of energy distribution, and determine the abnormal fluctuation index of dynamic changes in the thermal field; Pattern recognition module: Based on abnormal fluctuation indicators, it identifies characteristic patterns of heating area positional shifts using principal component analysis algorithm, and constructs an identification model for areas of abnormal energy concentration. Specifically: Based on the abnormal fluctuation indicators that appear during the dynamic changes of the thermal field, the principal component analysis algorithm is used to perform feature dimensionality reduction to determine several principal components that play a major role in the dynamic changes of the thermal field. Based on the identified principal components, significant feature patterns are extracted when the position of the corresponding heating region shifts. The significant feature patterns include the shift direction and magnitude of the energy concentration region position. Based on salient feature patterns, establish an identification model for regions with concentrated energy anomalies; Regional positioning module: Employs clustering analysis algorithms to determine the spatial location and boundary information of areas with abnormal heat concentration during real-time heating based on an identification model; Stability assessment module: Based on the spatial location and boundary information of the heat concentration anomaly area, it determines the stability status of the current heating mode and outputs a comprehensive evaluation result on whether the heating distribution is abnormal; Parameter control module: Dynamically adjusts the excitation frequency and output power of each coil in the induction coil array of the induction cooker based on the comprehensive evaluation results.

7. A rechargeable induction cooker, 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, it implements the data analysis-based control method for a rechargeable induction cooker according to any one of claims 1-5.

Images