A fault identification method for an all-electric kitchen appliance

Abstract

The application discloses a fault identification method of full-electric kitchen utensils, and relates to the technical field of fault identification. The method comprises the following steps. By introducing a non-uniform sub-grid division mechanism, the monitoring accuracy of the heating area can be focused on the key position, and the waste of resources in the fault-free area can be avoided, so that the detection efficiency is improved. At the same time, by using the power spectrum and harmonic distortion characteristics of electromagnetic radiation, early identification of abnormal signals is realized, and multi-dimensional cross verification is combined with temperature data, so that the accuracy and reliability of discrimination are further improved. By dynamically adjusting the sub-grid density, the application ensures that sufficient resolution is still available for accurate positioning in the case of initial fault or weak anomaly. By applying an excitation signal of a combination frequency and extracting the response change, the application can effectively identify the fault position, realize the whole-process closed-loop diagnosis from macroscopic screening to microscopic identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault identification, and in particular to a fault identification method for all-electric kitchenware. BACKGROUND

[0002] With the wide popularity of intelligent kitchen electrical equipment and smart home systems, all-electric kitchens, as an important embodiment of the integration of green energy-saving concepts and modern lifestyles, are gradually replacing traditional gas kitchens. Among them, all-electric kitchenware uses electromagnetic heating, resistance heating and other methods for cooking operations, and embeds various sensors and control modules to achieve fine control and multi-scenario intelligent linkage. However, with the increasing integration of functions and use frequency, the fault problems in the operation of all-electric kitchenware are increasingly prominent, especially the nonlinear harmonic disturbance caused by local anomalies of the electromagnetic system, loose electrical connections or material aging. If not identified in time, it not only affects the cooking efficiency, but also may cause safety hazards.

[0003] For example, CN118797299A discloses a sunlight kitchen non-technical reason fault identification method, device, equipment and medium, which obtains fault node information, pre-processes it and inputs it into a trained identification model, and then outputs a fault identification result. This method has a certain level of automation, reduces the uncertainty of manual evaluation, and improves the identification efficiency. However, the non-technical reason faults handled by this method are mostly problems caused by operation or human error, and its data source relies on the reporting information of terminal nodes or model training corpus. It does not consider the direct detection of technical level abnormal signals such as electromagnetic field disturbance and harmonic distortion, and is difficult to apply to the fault scene of all-electric kitchenware with complex power flow and highly correlated local excitation response. Especially in electromagnetic heating equipment, the coupling characteristics of harmonic distortion and spatial temperature gradient are important clues for early fault judgment, and this kind of method lacks the ability to perceive and model such characteristics.

[0004] For another example, CN112153373A proposes a fault identification method, device and storage medium for bright kitchen and bright stove equipment, which analyzes image frame information, image similarity and shooting time parameters in video stream to determine the state of the camera equipment. Although this method has practical value in video capture device state judgment and can solve the problem of camera false online, its core detection mechanism relies on image recognition algorithm and the processing object is mainly video device signal, without involving electromagnetic energy transmission and temperature gradient change and other physical quantity parameters. When facing all-electric kitchenware and other equipment with strong electrical characteristics, it cannot provide effective internal fault positioning means. In addition, this scheme does not introduce electromagnetic excitation response mechanism and does not use regional refinement analysis method, so it does not have the ability to distinguish small physical abnormalities in the heating area.

[0005] In summary, the existing kitchen equipment fault identification method mostly focuses on the operation level or image signal level, and cannot realize the collaborative perception and dynamic refined diagnosis of the internal electromagnetic and thermal abnormalities of the full-electric kitchen appliance, especially in the identification accuracy and response analysis capability of the fault area sub-grid level. SUMMARY

[0006] In view of the problems existing in the existing kitchen equipment fault identification, the present application is proposed.

[0007] Therefore, the problem to be solved by the present application is how to encode and refine the harmonic feature and temperature feature based on the sub-grid granularity, and realize accurate positioning of the abnormal area through electromagnetic excitation.

[0008] To solve the above technical problems, the present application provides the following technical solutions:

[0009] In a first aspect, the present application provides a fault identification method for a full-electric kitchen appliance, which comprises: dividing a heating area into non-uniform sub-grids, real-time collecting electromagnetic radiation power spectral density and characteristic harmonic distortion rate of each sub-grid, and encoding into a harmonic feature matrix; comparing the harmonic feature matrix with a pre-set mode library, extracting the sub-grid coordinates of distortion abnormalities, and marking as a suspected fault area B; collecting real-time temperature distribution data of the suspected fault area B, forming a temperature feature matrix; dynamically adjusting the sub-grid density in the B area according to the temperature gradient condition, and refining the harmonic feature matrix and the temperature feature matrix; at the coordinates of the refined suspected fault area B, applying an electromagnetic excitation signal of combined frequency, collecting excitation response and screening abnormal coordinates, and completing the identification of local fault coordinates.

[0010] As a preferred scheme of the fault identification method for the full-electric kitchen appliance, wherein: the division of the non-uniform sub-grid comprises: extracting the peak line of the temperature gradient change through gradient operation of the surface temperature data of the heating area; according to the peak line, marking the part with sharp temperature change in the area as a gradient sensitive zone, and marking the remaining part as a gradient gentle zone; using different grid scales to refine and divide the gradient sensitive zone and the gradient gentle zone, to generate a non-uniform sub-grid with adaptive scale characteristics.

[0011] As a preferred scheme of the fault identification method for the full-electric kitchen appliance, wherein: the generation of the harmonic feature matrix comprises: in each non-uniform sub-grid, extracting the power spectral density of electromagnetic radiation of each sub-grid, forming a data group P, and the data group P maintains sub-grid sequence consistency; using the power spectral density data in the data group P to calculate the harmonic distortion rate in each sub-grid by frequency band, to form a one-dimensional harmonic feature sequence corresponding to the sub-grid; according to the spatial arrangement rule of the sub-grid, encoding the one-dimensional harmonic feature sequence into a two-dimensional matrix according to the spatial position correspondence.

[0012] As a preferred scheme of the fault identification method of the all-electric kitchen utensil, wherein: the distortion anomaly is: the harmonic feature matrix adopts the normalization method based on the maximum-minimum value range, the feature vectors of each sub-grid are standardized to generate a standardized matrix; each sub-grid feature row vector in the standardized matrix is calculated with the weighted cosine similarity with the plurality of distortion mode vectors in the preset mode library to obtain a similarity matrix; from the similarity matrix, the sub-grid indexes with a similarity value lower than a set threshold are extracted to form a coordinate set.

[0013] As a preferred scheme of the fault identification method of the all-electric kitchen utensil, wherein: the dynamic adjustment of the sub-grid density in the suspected fault area B includes: synchronously collecting the current temperature values of all sub-grid units in the suspected fault area B and organizing them into a temperature feature matrix according to the sub-grid coordinates, maintaining a one-to-one correspondence with the sub-grid structure in the suspected fault area B; the local temperature gradient value of each sub-grid is calculated for the temperature feature matrix; for the temperature gradient amplitude of each sub-grid, a density adjustment factor F is set according to the amplitude size, different density adjustment factors F are given to different gradient units, and the sub-grid density is adjusted.

[0014] As a preferred scheme of the fault identification method of the all-electric kitchen utensil, wherein: the supplementary refinement of the harmonic and temperature feature matrix includes: the corresponding electromagnetic radiation power spectral density data and temperature values of the refined suspected fault area B sub-grid are collected one by one and supplemented to the refined harmonic feature matrix and the refined temperature feature matrix respectively, and the refined matrix maintains consistency with the refined sub-grid coordinates.

[0015] As a preferred scheme of the fault identification method of the all-electric kitchen utensil, wherein: the collection of excitation responses includes: a combined frequency excitation signal containing low-order frequencies and high-order frequencies is applied to the adjusted sub-grid coordinates in turn; the electromagnetic response signal of each sub-grid is collected after the suspected fault area B is excited, the response intensity is recorded in coordinates to obtain a list of enhanced feature values compared with before excitation, and the list is arranged according to the refined sub-grid index.

[0016] As a preferred scheme of the fault identification method of the all-electric kitchen utensil, wherein: the screening of abnormal coordinates includes: each value in the feature value list is subjected to a normalization operation according to the global maximum value and the minimum value to obtain a standardized feature value sequence ES; the gradient ΔE of each sub-grid feature value in ES and the feature values of the surrounding adjacent sub-grids is calculated to obtain the feature value fluctuation amplitude of each sub-grid; according to a set abnormal threshold, the gradients ΔE are screened one by one, and the sub-grid coordinates with a gradient value exceeding the abnormal threshold are extracted and marked as a local fault coordinate set.

[0017] In a second aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program instructs the processor to implement the steps of the fault identification method for the all-electric kitchen appliance according to the first aspect of the present application.

[0018] In a third aspect, the present application provides a computer-readable storage medium storing a computer program, wherein the computer program instructs the processor to implement the steps of the fault identification method for the all-electric kitchen appliance according to the first aspect of the present application.

[0019] The present application has the following beneficial effects: By introducing a non-uniform sub-grid division mechanism, the present application focuses the monitoring accuracy of the heating area on key positions, avoiding the waste of resources in fault-free areas, thereby improving the detection efficiency. At the same time, by utilizing the power spectrum and harmonic distortion characteristics of electromagnetic radiation, early identification of abnormal signals is realized, and multi-dimensional cross-validation is further combined with temperature data to further improve the accuracy and reliability of the discrimination. By dynamically adjusting the sub-grid density, the present application ensures that sufficient resolution is still available for accurate positioning in the case of initial fault or weak anomaly. Finally, by applying an excitation signal of a combination frequency and extracting the response change, the fault position can be effectively identified, realizing the whole-process closed-loop diagnosis from macroscopic screening to microscopic identification. In summary, the present application not only significantly improves the fault identification sensitivity and positioning accuracy of the all-electric kitchen appliance, but also provides an expandable health monitoring framework for intelligent kitchen appliances. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0021] Fig. 1 Flowchart of the fault identification method for the all-electric kitchen appliance;

[0022] Fig. 2 Flowchart of dynamically adjusting the sub-grid density in the suspected fault area B in the fault identification method for the all-electric kitchen appliance. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0024] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details set forth in this description. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the present application.

[0025] Secondly, the "one embodiment" or "an embodiment" referred to herein means a specific feature, structure, or characteristic under discussion. Thus, "one embodiment" does not mean a single embodiment nor is it to be taken individually or selectively from other embodiments.

[0026] Embodiment 1

[0027] Reference Figs. 1-2 For the first embodiment of the present application, the embodiment provides a fault identification method for an all-electric kitchen appliance, as shown in the figure, comprising the following operation steps:

[0028] S1: The heating area is divided into non-uniform subgrids, the electromagnetic radiation power spectral density and the characteristic harmonic distortion rate of each subgrid are collected in real time, and the harmonic feature matrix is coded.

[0029] S1.1: The division of the subgrid comprises the following steps:

[0030] The surface temperature data of the heating area is subjected to gradient operation to extract the peak line with significant temperature gradient change. It should be noted that the gradient operation is completed by calculating the spatial derivative of the surface temperature distribution, which can effectively detect the boundary line with abnormal temperature change rate, thereby clearly defining the spatial imbalance of the heat distribution inside the heating area. The extraction of the above peak line is dedicated to representing the high-variation boundary of the heat distribution in the heating area. Compared with the traditional uniform division method, the present application avoids the technical defect of ignoring the details of local thermal gradient change through this boundary extraction.

[0031] According to the peak line, the part with sharp temperature change in the area is marked as the gradient sensitive area, and the remaining part is marked as the gradient gentle area, thereby forming a preliminary non-uniform division logic. Among them, the marking logic of the gradient sensitive area is determined according to the gradient amplitude threshold value, which ensures the effective capture of the small but significant heat change area, and the division standard of the gradient gentle area ensures the efficient allocation of computing resources, avoiding excessive refinement of the low change area.

[0032] Further, the gradient sensitive area and the gradient gentle area are further divided by using different grid scales to generate non-uniform subgrids with adaptive scale characteristics, and the specific operation is as follows:

[0033] a. For the gradient sensitive area and the gradient gentle area, two scale factors Fs and Fp are set in advance, where Fs is smaller than Fp, for controlling the size of the refined grid. Wherein, the scale factors Fs and Fp are matched according to the target thermal resolution requirement, wherein Fs is used to meet the fine thermal change detection requirement, and Fp is used to ensure the low calculation load of the gentle area, thereby giving the sub-grid size controllability. By adopting the double scale factor mechanism, the invention effectively avoids the limitation problem of over-coarse in the sensitive area and over-fine in the gentle area faced by the conventional single scale division, and realizes the regional adaptive adjustment of the grid division scale.

[0034] b. In the marked gradient sensitive area, the scale factor Fs is used to refine the grid in the area, and the grid is preferentially divided along the temperature gradient direction in the refinement process, thereby forming a fine grid group Ns, wherein Ns has an arrangement feature matched with the gradient direction, and the adaptability of the sub-grid to thermal change is improved through directional refinement.

[0035] Specifically, in the refinement process, the logic of preferentially dividing along the temperature gradient direction is specially set, so that the refined grid can obtain higher resolution in the direction of sharp gradient change. The fine grid group Ns formed thereby is not only fine in scale, but also has an arrangement feature matched with the gradient direction, and the adaptability and response sensitivity of the sub-grid to thermal change are improved through directional refinement, so as to ensure that the thermal dynamic change in the gradient sensitive area can be captured with high precision.

[0036] c. For the gradient gentle area, the scale factor Fp is used for uniform division of a larger scale to generate a coarse grid group Np, wherein Np maintains spatial layout uniformity and forms a splicable structure with the fine grid group Ns at the boundary, and through the coarse and fine grid splicing logic, the problem of grid fault or incoherent transition is avoided. It should be noted that the boundary smoothing mechanism is designed through the coarse and fine grid splicing logic in this step, thereby effectively avoiding the problem of grid fault or incoherent transition, and ensuring that the grid division of the entire heating area reaches the optimal state in terms of spatial continuity and data splicing.

[0037] d. The obtained Ns and Np are spliced in a spatially adjacent relationship to form an overall non-uniform sub-grid structure G, which contains fine-scale sub-grids in the gradient sensitive area and coarse-scale sub-grids in the gradient gentle area, and the boundary splicing is continuous. This design point realizes the scale adaptive grid design driven by thermal gradient.

[0038] S1.2: The generation of the harmonic feature matrix includes the following operation steps:

[0039] In each non-uniform sub-grid, the power spectral density of the electromagnetic radiation of each sub-grid is extracted to form a data set P, which maintains the consistency of the sub-grid sequence, i.e., each sub-grid number is one-to-one mapped to the corresponding power spectral density data, ensuring that the data structure is completely aligned with the grid division in the spatial dimension;

[0040] The power spectral density data in the data set P is used to calculate the harmonic distortion rate in each sub-grid in different frequency bands to form a one-dimensional harmonic feature sequence corresponding to the sub-grid. In this step, the calculation of the harmonic distortion rate uses a multi-band ratio algorithm, which can accurately describe the harmonic components in the electromagnetic signal;

[0041] According to the spatial arrangement rule of the sub-grid, the one-dimensional harmonic feature sequence is encoded into a two-dimensional matrix according to the spatial position correspondence. This two-dimensional matrix accurately reflects the harmonic features and maintains consistency with the non-uniform sub-grid structure G in the matrix structure, thereby realizing the fusion coding of the thermal field spatial features and the electromagnetic harmonic dynamic features. The spatial coding logic of the matrix arranges the rows and columns according to the position coordinates of the sub-grid in the X-Y plane, so that the row and column structure of the matrix directly maps the actual grid distribution of the heating area, realizing high consistency between the data structure and the physical structure

[0042] S2: Compare the harmonic feature matrix with the pre-set pattern library, extract the coordinates of the distortion abnormal sub-grid, and mark it as a suspected fault area B.

[0043] First, a normalization method based on the maximum-minimum value range is used on the harmonic feature matrix to standardize the feature vectors of each sub-grid and generate a standardized matrix. It should be noted that through this standardization process, on the one hand, the comparison deviation caused by amplitude differences between different sub-grid features can be eliminated, and on the other hand, the feature components can be ensured to be in the uniform 0 to 1 interval during the subsequent similarity calculation process, improving the stability of the calculation and the fairness of the comparison. In addition, the standardized matrix in this step maintains a one-to-one correspondence with the original harmonic feature matrix in the spatial position, ensuring accurate traceability during subsequent coordinate extraction.

[0044] Second, the feature row vector of each sub-grid in the standardized matrix is calculated with the weighted cosine similarity with the multiple distortion mode vectors in the pre-set pattern library to obtain a similarity matrix.

[0045] Specifically, the preset mode library contains a plurality of representative distortion modes, each mode is saved in the form of a feature vector, for representing the known electromagnetic radiation harmonic abnormal characteristics. In order to realize accurate comparison, a weighted cosine similarity calculation method is used for vector comparison in this step. The weighted cosine similarity refers to introducing a weight factor on the basis of the traditional cosine similarity formula, for adjusting the influence degree of each feature component on the similarity calculation result. Through the weight, the application can specifically improve the sensitivity to the distortion characteristics of certain frequency bands, thereby enhancing the accuracy of fault detection.

[0046] After the above calculation, a similarity matrix is obtained, each element value in the matrix represents the similarity score of the corresponding sub-grid feature and the distortion mode in the mode library, the value range is 0 to 1, wherein 1 represents complete matching and 0 represents complete mismatch.

[0047] Further, from the similarity matrix, the sub-grid indexes with a similarity value lower than a set threshold are extracted to form a coordinate set; the screened coordinate set is grouped according to the spatial proximity rule, the sub-grids with mutual connectivity are aggregated into a continuous block, and the aggregated block is marked as a suspected fault area B. Through the connectivity grouping method, the spatial consistency of the abnormal recognition result is improved, and the misjudgment of isolated abnormal points is avoided.

[0048] Specifically, the application uses a connectivity detection-based method to analyze the coordinates in the coordinate set, and aggregates the sub-grid coordinates with mutual connectivity into a continuous block. The connectivity refers to the coordinates being adjacent to each other in a two-dimensional space, such as four-adjacent or eight-adjacent. Through this aggregation operation, the sub-grid group forming a continuous abnormal distribution can be effectively identified, and the spatial isolated and scattered abnormal points are excluded to avoid misjudgment caused by measurement fluctuations or local noise.

[0049] It is worth noting that through the connectivity grouping method in this step, the application significantly improves the spatial coherence and reliability of the abnormal detection result, compared with the traditional point-by-point detection method, avoiding the false alarm problem caused by isolated abnormal points, thereby enhancing the reliability in actual application scenarios.

[0050] S3: Collecting real-time temperature distribution data of the suspected fault area B to form a temperature feature matrix; dynamically adjusting the sub-grid density in the B area according to the temperature gradient, and refining the harmonic feature matrix and the temperature feature matrix.

[0051] In the embodiment of the application, as shown in Fig. 2 The dynamic adjustment of the sub-grid density in the suspected fault area B includes the following operation steps:

[0052] S3.1: Synchronize the current temperature values of all sub-grid units in the suspected fault area B, and organize them into a temperature feature matrix according to the sub-grid coordinates, maintaining a one-to-one correspondence with the sub-grid structure in the suspected fault area B.

[0053] S3.2: Calculate the local temperature gradient value of each sub-grid for the temperature feature matrix.

[0054] In particular, the method based on central difference or adjacent average difference is used (the present application is not limited to this), and the temperature value of each sub-grid unit is subtracted from the temperature values of its four adjacent or eight adjacent units. Through the above gradient calculation, the local area with rapid temperature change or transition within the suspected fault area B can be effectively identified, thereby providing data basis for subsequent sub-grid density adjustment.

[0055] S3.3: For the gradient amplitude of each sub-grid, set a density adjustment factor F according to the amplitude size, assign different density adjustment factors F to different gradient units, and adjust the sub-grid density. Through gradient-driven density control, the limitations of the conventional uniform refinement method are overcome.

[0056] Specifically, according to the gradient amplitude of each sub-grid, it is divided into multiple gradient levels, and different levels are assigned corresponding density adjustment factors F. The F value defines the number of sub-grids that need to be subdivided in the refinement process. For example, for the area with rapid change with gradient amplitude higher than threshold T1, set F = 4, i.e. the original unit is subdivided into 4 small grids; for the area with gradient amplitude in the middle range T2, set F = 2; and for the stable area with gradient lower than T3, set F = 1, i.e. maintain the original density without refinement.

[0057] Through this gradient-driven density control strategy, the present application breaks through the conventional uniform refinement method for the whole area, and can realize high-density refinement for key high-gradient areas, while avoiding redundant refinement for low-gradient areas, thereby effectively balancing detection resolution and computational resource consumption, improving detection efficiency and accuracy.

[0058] After completing the above-mentioned dynamic adjustment of sub-grid density, the present step also performs a supplementary refinement data collection operation on the refined suspected fault area B, and the supplementary refinement harmonic and temperature feature matrix includes the following operation steps:

[0059] The electromagnetic radiation power spectral density data and temperature values of the completed sub-grid of the suspected fault area B are collected one by one, and are supplemented to the refined harmonic feature matrix and refined temperature feature matrix, respectively. The refined matrix maintains consistency with the refined sub-grid coordinates.

[0060] To ensure that the refined harmonic feature matrix and the refined temperature feature matrix are compatible with the original matrix structure, the application preferably adopts a matrix refinement mechanism based on coordinate mapping, rather than a simple data appending method. Specifically, after completing the dynamic adjustment of the sub-grid density in the suspected fault area B, the refined matrix is reconstructed based on the coordinate system of the adjusted sub-grid, ensuring that the number of rows and columns of the matrix corresponds one-to-one to the number of refined sub-grids, thereby avoiding the problem of dimension expansion.

[0061] It should be particularly pointed out that through the dynamic refinement strategy in this step, local high-precision detection and global resource optimization are effectively combined in the suspected fault area. Compared with the uniform sub-grid encryption method commonly used in the prior art, the gradient amplitude driven density adjustment method adopted by the application can automatically perceive abnormal gradient areas in the temperature field and specifically improve the detection resolution of these key areas, avoiding the waste of computing resources on irrelevant smooth areas. At the same time, the refined harmonic and temperature feature matrices have higher spatial resolution and feature richness, providing a more solid data foundation for subsequent fault type determination and accurate positioning, further improving the fault detection capability and adaptability in complex electromagnetic environments.

[0062] S4: Apply an electromagnetic excitation signal of a combined frequency at the coordinates of the refined suspected fault area B, collect the excitation response and select abnormal coordinates to complete the discrimination of local fault coordinates.

[0063] In the embodiment of the application, collecting the excitation response includes:

[0064] S4.1: Apply a combined frequency excitation signal containing low frequency and high frequency to the adjusted sub-grid coordinates in turn.

[0065] Specifically, the combined frequency signal is composed of a low frequency component and a high frequency component, wherein the low frequency part is used to excite the macroscopic response of the material structure in the area B, and the high frequency part is used to specifically activate the resonance characteristics of micro-scale defects and abnormal units.

[0066] When applying the excitation, a coordinate positioning excitation method is adopted, that is, electromagnetic signals are applied to each sub-grid unit in turn according to the index order of the refined sub-grid coordinates, to ensure that each sub-grid response signal corresponds to its spatial coordinates one-to-one, avoiding data confusion.

[0067] S4.2: Collect the electromagnetic response signals of each sub-grid in the suspected fault area B after applying the excitation, coordinate the response intensity, and obtain a list of characteristic values that are enhanced after comparison with the pre-excitation state. To facilitate feature extraction, further compare the response intensity of this excitation with the baseline response intensity before excitation (i.e., the static state), calculate the response enhancement characteristic values, and use them to highlight the local abnormal units under excitation.

[0068] Further, in the embodiments of the present application, the screening of abnormal coordinates comprises:

[0069] S4.3: Perform normalization operation on each value in the list of characteristic values according to the global maximum and minimum values to obtain a standardized characteristic value sequence ES; the purpose of normalization processing is to eliminate the magnitude difference between the response intensities of different sub-grids, ensure that the subsequent fluctuation detection operation is performed on a unified numerical scale, and improve the accuracy of abnormality discrimination.

[0070] S4.4: For each sub-grid characteristic value in ES, calculate the gradient ΔE of the characteristic value with the surrounding neighboring sub-grid characteristic values to obtain the fluctuation amplitude of the characteristic value of each sub-grid.

[0071] In specific calculation, for each sub-grid unit, select the standardized characteristic values of its four adjacent or eight adjacent units, perform mean difference or maximum difference operation to obtain the response gradient ΔE of the sub-grid.

[0072] Through this local gradient calculation method, the region with dramatic changes in local response signals can be effectively detected, and these changes are usually directly related to internal micro-defects, cracks or abnormal structures, thus having higher fault indication significance.

[0073] S4.5: According to the set abnormal threshold, screen the gradient ΔE one by one, extract the sub-grid coordinates whose gradient values exceed the abnormal threshold, and mark them as a local fault coordinate set. This coordinate set is the specific local fault point inside the region B identified in this step, which is used for subsequent repair, compensation or detailed diagnosis.

[0074] It should be noted that the combined frequency excitation + response gradient screening mechanism used in this step can activate the electromagnetic response modes of both macroscopic structures and microscopic defects through combined frequency excitation, thereby enhancing the response contrast of fault points. By calculating the ΔE gradient and screening the abnormal threshold, the present application breaks through the traditional method of relying only on absolute response value screening, effectively eliminates the false alarm risk caused by background noise and response fluctuation, and ensures the high accuracy and robustness of fault coordinate discrimination.

[0075] Further, the harmonic distortion rate, temperature gradient and other information corresponding to the local fault coordinates are fused to generate fault feature data, which is submitted to the upper computer for classification management.

[0076] The embodiment also provides a computer device suitable for the fault identification method of the all-electric kitchen appliance, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the fault identification method of the all-electric kitchen appliance proposed in the above embodiment.

[0077] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0078] The embodiment also provides a storage medium having a computer program stored thereon, and the program is executed by a processor to realize the fault identification method of the all-electric kitchen appliance proposed in the above embodiment.

[0079] To sum up, by introducing the non-uniform sub-grid division mechanism, the monitoring accuracy of the heating area is focused on the key position, and the waste of resources in the fault-free area is avoided, so that the detection efficiency is improved. At the same time, by using the power spectrum and harmonic distortion characteristics of electromagnetic radiation, early identification of abnormal signals is realized, and multi-dimensional cross verification is further combined with temperature data to further improve the accuracy and reliability of discrimination. By dynamically adjusting the sub-grid density, the present application ensures that sufficient resolution is still available for accurate positioning in the case of initial failure or weak anomaly. Finally, by applying an excitation signal of a combination frequency and extracting the response change, the fault position can be effectively identified, and the whole process of closed-loop diagnosis from macroscopic screening to microscopic identification is realized. To sum up, the present application not only significantly improves the fault identification sensitivity and positioning accuracy of the all-electric kitchen appliance, but also provides an expandable health monitoring framework for intelligent kitchen appliances.

[0080] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A fault identification method for all-electric kitchen appliances, characterized in that: include: The heating area is divided into non-uniform sub-grids, and the electromagnetic radiation power spectral density and characteristic harmonic distortion rate of each sub-grid are collected in real time and encoded into a harmonic characteristic matrix. The harmonic feature matrix is ​​compared with a pre-set mode library to extract the sub-grid coordinates of the distortion anomaly and mark them as suspected fault area B. Real-time temperature distribution data is collected for suspected fault area B to form a temperature feature matrix; the sub-grid density in area B is dynamically adjusted according to the temperature gradient, and the harmonic feature matrix and temperature feature matrix are refined. At the coordinates of the refined suspected fault area B, an electromagnetic excitation signal of a combined frequency is applied, the excitation response is collected, and abnormal coordinates are screened to complete the identification of local fault coordinates. The generation of the harmonic feature matrix includes: extracting the power spectral density of electromagnetic radiation in each non-uniform subgrid to form a data group P, wherein the data group P maintains the consistency of the subgrid sequence. Using the power spectral density data in data set P, the harmonic distortion rate in each subgrid is calculated by frequency band, forming a one-dimensional harmonic characteristic sequence corresponding to the subgrid; according to the spatial arrangement rules of the subgrid, the one-dimensional harmonic characteristic sequence is encoded into a two-dimensional matrix according to the spatial position correspondence.

2. The fault identification method for all-electric kitchen appliances as described in claim 1, characterized in that: The division of the non-uniform subgrid includes: The surface temperature data of the heated area is used to extract the peak lines with significant temperature gradient changes through gradient calculation; Based on the peak line, the areas with drastic temperature changes within the region are marked as gradient-sensitive areas, and the remaining areas are marked as gradient-flat areas. Different grid scales are used to refine the gradient-sensitive region and the gradient-flat region, generating non-uniform sub-grids with adaptive scale characteristics.

3. The fault identification method for all-electric kitchen appliances as described in claim 1, characterized in that: The distortion anomaly is: A normalization method based on the maximum-minimum range is used to normalize the feature vectors of each subgrid to generate a normalized matrix. The feature row vector of each subgrid in the normalized matrix is ​​compared with multiple distortion mode vectors in the preset mode library to calculate the weighted cosine similarity and obtain the similarity matrix. From the similarity matrix, the indices of all subgrids with similarity values ​​lower than a set threshold are extracted to form a coordinate set.

4. The fault identification method for all-electric kitchen appliances as described in claim 1, characterized in that: Dynamically adjusting the sub-mesh density within the suspected fault region B includes: The current temperature value is collected synchronously for all sub-grid cells in the suspected fault area B, and organized into a temperature feature matrix according to the sub-grid coordinates, maintaining a one-to-one correspondence with the sub-grid structure in the suspected fault area B. For the temperature feature matrix, calculate the local temperature gradient value of each subgrid; For each subgrid temperature gradient magnitude, a density adjustment factor F is set according to the magnitude of the magnitude. Different density adjustment factors F are assigned to different gradient units to adjust the subgrid density.

5. The fault identification method for all-electric kitchen appliances as described in claim 4, characterized in that: The refined harmonic characteristic matrix and the refined temperature characteristic matrix include: For the refined suspected fault area B subgrid, the corresponding electromagnetic radiation power spectral density data and temperature values ​​are collected one by one, and then added to the refined harmonic feature matrix and refined temperature feature matrix respectively. The refined matrix maintains the consistency with the coordinates of the refined subgrid.

6. The fault identification method for all-electric kitchen appliances as described in claim 1, characterized in that: The acquisition stimulus response includes: The combined frequency excitation signal containing low and high frequencies is sequentially applied to the adjusted subgrid coordinates; for the suspected fault region B after excitation, the electromagnetic response signal of each subgrid is collected, the response intensity is recorded in coordinate form, and a list of enhanced feature values ​​after comparison with before excitation is obtained, which are uniformly arranged according to the refined subgrid index.

7. The fault identification method for all-electric kitchen appliances as described in claim 6, characterized in that: The filtered abnormal coordinates include: Normalize each value in the eigenvalue list according to the global maximum and minimum values ​​to obtain the standardized eigenvalue sequence ES; For each subgrid eigenvalue in ES, calculate the gradient ΔE between it and the eigenvalues ​​of its neighboring subgrids to obtain the eigenvalue fluctuation amplitude of each subgrid; Based on the set anomaly threshold, the gradient ΔE is screened one by one, and the coordinates of the subgrids whose gradient values ​​exceed the anomaly threshold are extracted and marked as local fault coordinate sets.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the fault identification method for all-electric kitchen appliances according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the fault identification method for any one of the all-electric kitchen appliances according to claims 1 to 7.

Images