Fault prediction method, device and system for microwave oven based on data analysis

Abstract

The present invention relates to the field of data analysis technology, and discloses a method, device and system for predicting faults of a microwave oven based on data analysis. The method comprises: obtaining operation and maintenance data for preprocessing; performing block processing on the operation and maintenance data; classifying the operation and maintenance data into environmental data and equipment loss data to obtain target environmental data and target equipment loss data; performing environmental time series change analysis based on the target environmental data; performing operation loss prediction based on the environmental time series change data and the target equipment loss data by using the operation stage analysis; performing potential fault prediction based on the environmental time series change data and the operation loss prediction data by using the knowledge graph; performing correlation matching on the potential fault data and each fault management scheme, and generating a fault risk management plan based on the correlation matching result for management. The present invention can minimize the impact of faults and improve the reliability and safety of the microwave oven.

Description

Technical Field

[0001] The present invention mainly relates to the field of data analysis technology, and in particular to a method, device and system for predicting faults of a microwave oven based on data analysis. Background Art

[0002] Microwave ovens are common cooking appliances in modern life. In order to ensure that they do not affect the cooking needs of daily life, it is necessary to predict the faults of microwave ovens to foresee safety hazards in advance and make corresponding plans. For the fault prediction of microwave ovens, data analysis is usually performed directly based on relevant operation and maintenance data, without considering data classification of operation and maintenance data, resulting in too slow data analysis efficiency and making it difficult to reasonably use relevant operation and maintenance data. At the same time, in fault prediction, the analysis of environmental data and operating loss is a very important part. At present, relevant personnel usually conduct relevant analysis based on environmental data and loss data, but this method relies too much on the professional quality of relevant personnel and cannot guarantee the accuracy of environmental analysis and operation loss analysis, resulting in a large deviation between the obtained fault prediction results and the actual situation. At the same time, how to effectively improve the accuracy of potential fault prediction and the reliability of fault management plan matching is also an urgent problem to be solved at this stage. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides a microwave oven-steam-oven fault prediction method, device and system based on data analysis, which can minimize the impact of faults and improve the reliability and safety of the microwave oven-steam-oven.

[0004] In order to solve the above technical problems, the present invention provides a method for predicting faults of a microwave oven based on data analysis, which is applied to a cloud server, and the cloud server is deployed with a big data analysis engine; the method comprises:

[0005] The big data analysis engine of the cloud server obtains the operation and maintenance data of the microwave oven by using the data collection strategy, and pre-processes the operation and maintenance data to obtain the pre-processed operation and maintenance data;

[0006] The pre-processed operation and maintenance data is processed in blocks to obtain the block-processed operation and maintenance data;

[0007] Based on hypergraph clustering, the environmental data and equipment loss data of the block-processed operation and maintenance data are classified to obtain target environmental data and target equipment loss data;

[0008] Performing an environmental time series change analysis based on the target environmental data to obtain environmental time series change data;

[0009] Based on the environmental time series change data and the target equipment loss data, operation loss prediction is performed using operation stage analysis to obtain operation loss prediction data;

[0010] Constructing a knowledge graph, and using the knowledge graph to predict potential faults of the microwave oven based on the environmental time series change data and the operation loss prediction data, to obtain potential fault data;

[0011] A plurality of fault management schemes are obtained from a database, correlation matching is performed between the potential fault data and each fault management scheme to obtain a correlation matching result, and a fault risk management plan is generated based on the correlation matching result, and the microwave oven is managed based on the fault risk management plan.

[0012] Optionally, the using of a data collection strategy to acquire operation and maintenance data of the microwave oven and steamer, and preprocessing the operation and maintenance data to obtain the preprocessed operation and maintenance data includes:

[0013] Determine the data collection time and data collection scope based on the preset resource tag, and construct a data collection window based on the adaptation relationship between the data collection time and the data collection scope;

[0014] Determining a data collection strategy based on the data collection window, and acquiring operation and maintenance data of the microwave oven based on the data collection strategy;

[0015] Performing data cleaning on the operation and maintenance data to obtain the operation and maintenance data after the data cleaning;

[0016] Perform data integration processing on the operation and maintenance data after data cleaning to obtain pre-processed operation and maintenance data.

[0017] Optionally, the preprocessed operation and maintenance data is processed in blocks to obtain the block-processed operation and maintenance data, including:

[0018] A sliding window is set, and based on the sliding window, the pre-processed operation and maintenance data is processed into blocks by utilizing boundary connection to obtain the block-processed operation and maintenance data.

[0019] Optionally, the data classification of environment data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering to obtain target environment data and target equipment loss data includes:

[0020] Perform node coding and labeling on the block-processed operation and maintenance data to obtain several coding categories;

[0021] Obtaining a node coding matrix of each coding category, and constructing a hypergraph model based on the node coding matrix using data association support analysis;

[0022] Based on the hypergraph model, clustering is performed on the block-processed operation and maintenance data to obtain a clustering result;

[0023] The clustering result is used to classify the environmental data and the equipment loss data based on the tree classification and the network structure association classification method to obtain the target environmental data and the target equipment loss data.

[0024] Optionally, performing an environmental time series change analysis based on the target environmental data to obtain environmental time series change data includes:

[0025] Performing a fluctuation range analysis of the environmental data points based on the target environmental data to obtain the fluctuation range of the environmental data points;

[0026] Based on the target environmental data, a trend analysis is performed between environmental data points by using straight line fitting to obtain a change trend between environmental data points;

[0027] Based on the incremental probability transfer matrix, the fluctuation amplitude and the change trend are used to perform evolution analysis of the environmental time series to obtain evolution analysis data of the environmental time series;

[0028] Based on the fluctuation range and change trend, a hidden Markov model is used to perform environmental change prediction analysis to obtain environmental change prediction analysis data;

[0029] Based on the evolution analysis data and the environmental change prediction analysis data, an environmental time series change analysis is performed to obtain environmental time series change data.

[0030] Optionally, performing operation loss prediction based on the environment time series change data and the target equipment loss data by using operation stage analysis to obtain operation loss prediction data includes:

[0031] Generate a loss rate curve based on the target equipment loss data, and obtain a parameter range of an operation stage based on the loss rate curve;

[0032] Performing a conversion analysis of the operation phase based on the parameter interval to obtain operation phase conversion data;

[0033] Based on the operation phase conversion data, the environmental time series change data is used to perform loss evolution to obtain loss evolution data;

[0034] Derivation of uncertainty is performed based on the environmental time series variation data and the target equipment loss data to obtain uncertainty derivation parameters;

[0035] Based on the loss evolution data and uncertainty derivation parameters, the operation loss prediction is performed using the loss simulation process to obtain the operation loss prediction data.

[0036] Optionally, the constructing of the knowledge graph and using the knowledge graph to predict potential failures of the microwave oven based on the environmental time series change data and the operation loss prediction data to obtain potential failure data include:

[0037] Acquire historical fault information and a topological structure network of the microwave oven, wherein the historical fault information includes historical environmental information and historical operation loss information, and perform hierarchical classification of environmental content information and hierarchical classification of loss content information based on the historical fault information to obtain hierarchical classification data of environmental content information and hierarchical classification data of loss content information;

[0038] Constructing a fault information matrix based on the environmental content information hierarchical classification data, the loss content information hierarchical classification data and the topological structure network;

[0039] Perform fault coupling impact degree analysis based on the historical fault information to obtain fault coupling impact degree data;

[0040] Constructing a fault tree relationship based on the fault information matrix and the fault coupling influence degree data to obtain a fault tree structure diagram, and constructing a knowledge graph based on the fault tree structure diagram;

[0041] A potential fault prediction model is constructed based on the knowledge graph in combination with a Bayesian network prediction model, and potential faults of the microwave oven are predicted based on the potential fault prediction model using the environmental time series change data and the operating loss prediction data to obtain potential fault data.

[0042] Optionally, performing correlation matching on the potential fault data and each fault management solution to obtain a correlation matching result includes:

[0043] Acquire a first label text string of the potential fault data and a second label text string of each fault management solution;

[0044] Performing similarity calculation on the first label text string and each second label text string based on the directed graph to obtain corresponding similarity calculation results;

[0045] Obtain a difference string between the first label text string and each second label text string;

[0046] Based on the difference character string, performing difference calculation on the first label text character string and each second label text character string by using decision diagram conversion to obtain corresponding difference calculation results;

[0047] Based on the similarity calculation result and the difference calculation result, correlation matching is performed on the potential fault data and each fault management solution to obtain a correlation matching result.

[0048] In addition, the present invention also provides a microwave oven fault prediction device based on data analysis, which is applied to a cloud server, and the cloud server is deployed with a big data analysis engine; the device comprises:

[0049] A data preprocessing module, in which the big data analysis engine of the cloud server uses a data collection strategy to obtain the operation and maintenance data of the microwave oven and steamer, and preprocesses the operation and maintenance data to obtain the preprocessed operation and maintenance data;

[0050] The data block module is used to block the preprocessed operation and maintenance data to obtain the block-processed operation and maintenance data;

[0051] A data classification module is used to classify the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering to obtain target environmental data and target equipment loss data;

[0052] An environment time series change analysis module, used to perform environment time series change analysis based on the target environment data to obtain environment time series change data;

[0053] An operation loss prediction module, used to perform operation loss prediction based on the environment time series change data and the target equipment loss data by using operation stage analysis to obtain operation loss prediction data;

[0054] A potential fault prediction module is used to construct a knowledge graph, and to predict potential faults of the microwave oven based on the environmental time series change data and the operation loss prediction data using the knowledge graph to obtain potential fault data;

[0055] The fault risk management module is used to obtain several fault management schemes in the database, perform correlation matching on the potential fault data and each fault management scheme, obtain the correlation matching result, generate a fault risk management plan based on the correlation matching result, and manage the microwave oven based on the fault risk management plan.

[0056] In addition, the present invention also provides a microwave oven fault prediction system based on data analysis, the system includes a cloud server, the cloud server is deployed with a big data analysis engine, and the system is configured to execute the above-mentioned microwave oven fault prediction method based on data analysis.

[0057] In an embodiment of the present invention, by classifying the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering, more accurate target environmental data and target equipment loss data can be obtained, the efficiency of subsequent data analysis can be improved, and the relevant operation and maintenance data can be used more reasonably. The environmental time series change analysis based on the evolution analysis data and environmental change prediction analysis data generated by the target environmental data can effectively improve the accuracy of the environmental time series analysis. Based on the environmental time series change data and the target equipment loss data, the operation loss prediction is performed using the operation stage analysis. Taking into account the environmental time series changes and combining the operation stage analysis, the accumulated loss of the microwave oven in each operation stage can be more accurately analyzed. At the same time, uncertainty derivation is added to make the obtained operation loss prediction data more reliable. Based on the environmental time series change data and the operation loss prediction data, the knowledge graph is used to predict the potential faults of the microwave oven, avoid misjudgment and omission of potential faults, and achieve more accurate potential fault prediction. A fault risk management plan is generated based on the correlation matching result obtained by matching the potential fault data with each fault management plan. The microwave oven / steamer is managed based on the fault risk management plan to ensure that when a device fails, the impact of the failure can be minimized and the reliability and safety of the microwave oven / steamer is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 is a schematic flow chart of a method for predicting a fault of a microwave oven-steamer based on data analysis in an embodiment of the present invention;

[0060] Figure 2 Schematic diagram of the structure of a microwave oven-steamer-oven fault prediction system based on data analysis in an embodiment of the present invention;

[0061] Figure 3 It is a schematic diagram of the structural composition of a microwave oven fault prediction device based on data analysis in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0063] See also Figure 1 , Figure 1 1 is a flow chart of a method for predicting a fault of a microwave oven based on data analysis in an embodiment of the present invention. The method is applied to a cloud server, and a big data analysis engine is deployed on the cloud server. The method includes:

[0064] S11: The big data analysis engine of the cloud server obtains the operation and maintenance data of the microwave oven by using the data collection strategy, and pre-processes the operation and maintenance data to obtain the pre-processed operation and maintenance data;

[0065] In the specific implementation process of the present invention, the operation and maintenance data of the microwave oven and steamer are obtained by using a data collection strategy, and the operation and maintenance data are preprocessed to obtain the preprocessed operation and maintenance data, including: determining the data collection time and the data collection range based on a preset resource tag, and constructing a data collection window based on the adaptation relationship between the data collection time and the data collection range; determining the data collection strategy based on the data collection window, and obtaining the operation and maintenance data of the microwave oven and steamer based on the data collection strategy; performing data cleaning processing on the operation and maintenance data to obtain the operation and maintenance data after data cleaning processing; performing data integration processing on the operation and maintenance data after data cleaning processing to obtain the preprocessed operation and maintenance data.

[0066] Specifically, the data collection time and data collection scope are determined based on preset resource tags. The preset resource tags are call resource tags determined according to the topological network of the cloud server and the amount of storage data of the microwave oven. The data collection window is constructed based on the adaptation relationship between the data collection time and the data collection scope. The adaptation relationship is obtained by matching the tags of the data collection time and the data collection scope. The data collection strategy is determined based on the data collection window. The big data analysis engine of the cloud server obtains the operation and maintenance data of the microwave oven based on the data collection strategy. The operation and maintenance data includes operation and maintenance environment data, operation and maintenance equipment loss data, operation and maintenance type data, and operation and maintenance time data. The operation and maintenance data is cleaned, and abnormal characters and garbled characters are cleaned to obtain the operation and maintenance data after data cleaning. Performing data integration processing on the operation and maintenance data after data cleaning processing, selecting each data element to be processed that needs to be integrated from the operation and maintenance data after data cleaning processing, performing pairing processing on each data element to be processed, obtaining a matching relationship between each data element to be processed, generating a data relationship graph based on the matching relationship using a preset topology graph, and performing data integration processing on the operation and maintenance data after data cleaning processing according to the data relationship graph can effectively improve the efficiency of data integration. After completing the data integration processing, the preprocessing of the operation and maintenance data is completed, and the preprocessed operation and maintenance data is obtained.

[0067] S12: performing block processing on the pre-processed operation and maintenance data to obtain block-processed operation and maintenance data;

[0068] In the specific implementation process of the present invention, the pre-processed operation and maintenance data is processed in blocks to obtain the block-processed operation and maintenance data, including: setting a sliding window, and based on the sliding window, the pre-processed operation and maintenance data is processed in blocks using boundary connection to obtain the block-processed operation and maintenance data.

[0069] Specifically, a sliding window is set, and the pre-processed operation and maintenance data is initially divided into blocks according to the minimum offset value and the offset grouping interval to obtain a number of initial block points. The initial block points are subjected to sliding block judgment according to the sliding window. In the process of sliding block judgment of each initial block point by the sliding window, the hash value in the sliding window is calculated. If the hash value meets the preset requirements of the block, the initial block point is retained. If the hash value does not meet the preset requirements of the block, the initial block point is removed, and the retained initial block points are used as target block points. The block point performs block processing on the pre-processed operation and maintenance data to obtain several operation and maintenance data blocks. During the block processing, the last data of each data block except the last operation and maintenance data block is processed with the first data of the next operation and maintenance data block. The boundary connection processing is used to determine whether there is redundancy in the initial block point. If there is redundancy, the redundant initial block point is removed to obtain the final block-processed operation and maintenance data, which can effectively improve the accuracy of data segmentation, so as to improve the efficiency of subsequent data analysis and avoid occupying too many computing resources.

[0070] S13: classifying the environmental data and the equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering to obtain target environmental data and target equipment loss data;

[0071] In the specific implementation process of the present invention, the data classification of environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering to obtain target environmental data and target equipment loss data includes: node coding and labeling of the block-processed operation and maintenance data to obtain several coding categories; obtaining the node coding matrix of each coding category, and constructing a hypergraph model based on the node coding matrix using data association support analysis; clustering the block-processed operation and maintenance data based on the hypergraph model to obtain clustering results; and using the clustering results based on the tree classification and mesh structure association classification method to classify environmental data and equipment loss data to obtain target environmental data and target equipment loss data.

[0072] Specifically, node coding is performed on the operation and maintenance data after block processing, and node coding is performed through a preset coding table. The preset coding table is constructed according to the records of historical data classification. Nodes are created through the preset coding table, each node represents a coding category, and each coding category corresponds to a corresponding data category dimension, and several coding categories are obtained. The node coding matrix of each coding category is obtained, and the coding frequency distribution is calculated according to several coding categories. The node coding matrix of each coding category is generated through matrix coding query in the software. A hypergraph model is constructed based on the node coding matrix using data association support analysis. An initial hypergraph model is constructed based on the node coding matrix using a preset hypergraph template. In the initial hypergraph model, hyperedges are used to represent the association of the data categories represented by the vertices connected to it. The association rule support is used as the weight of each hyperedge. The node coding matrix is ​​clustered by coding category to obtain the corresponding hypergraph data. The architecture of the coding category can be displayed through the hypergraph data. The initial hypergraph model is adjusted according to the corresponding hypergraph data, and the relationship between data categories can be quickly obtained to obtain the final hypergraph model. Based on the hypergraph model, clustering processing is performed on the block-processed operation and maintenance data, and classes are initialized for the block-processed operation and maintenance data. All classes in the hypergraph model are traversed, and all classes that can be merged are obtained according to the merging index analysis. A new hypergraph model is constructed for all classes that can be merged, and the weighted mean square deviation of all edges in each cluster segmentation in the new hypergraph model is calculated to determine whether the weighted mean square deviation is greater than or equal to a preset threshold. If the weighted mean square deviation is greater than or equal to the preset threshold, class initialization, traversal of the hypergraph model, acquisition of merged classes, construction of the new hypergraph model, and calculation of the weighted mean square deviation are repeated until a preset number of iterations is reached. If the weighted mean square deviation is less than the preset threshold, the clustering processing is terminated to obtain the clustering result. A more accurate clustering result can be obtained, and data can be quickly and effectively processed, so as to mine data with more demand value. The association classification method based on tree classification and mesh structure utilizes the clustering results to classify the environmental data and equipment loss data, and uses the clustering results as data classification features. The data classification of environmental data and equipment loss data is performed by constructing a tree structure using the data classification features. The data in the environmental data and equipment loss data are linked and classified through the association classification of the mesh structure. At the same time, the tree classification and the association classification of the mesh structure use a logistic regression classifier to obtain the target environmental data and target equipment loss data. The target environmental data includes environmental temperature, humidity, etc. The target equipment loss data includes loss and maintenance data at each time point, etc., which can more accurately search and classify the required target environmental data and target loss data in the block-processed operation and maintenance data, providing sufficient data support for subsequent environmental time series change analysis and operation loss prediction.

[0073] S14: performing an environmental time series change analysis based on the target environmental data to obtain environmental time series change data;

[0074] In the specific implementation process of the present invention, the environment time series change analysis is performed based on the target environment data to obtain the environment time series change data, including: performing a fluctuation amplitude analysis of the environment data points based on the target environment data to obtain the fluctuation amplitude of the environment data points; performing a trend analysis between the environment data points based on the target environment data using straight line fitting to obtain the change trend between the environment data points; performing an evolution analysis of the environment time series based on the incremental probability transfer matrix using the fluctuation amplitude and the change trend to obtain the evolution analysis data of the environment time series; performing an environmental change prediction analysis based on the fluctuation amplitude and the change trend using a hidden Markov model to obtain the environmental change prediction analysis data; performing an environmental time series change analysis based on the evolution analysis data and the environmental change prediction analysis data to obtain the environmental time series change data.

[0075] Specifically, the fluctuation amplitude of the environmental data points is analyzed based on the target environmental data, the difference between every two environmental data points is calculated in the target environmental data, and the fluctuation amplitude is calculated according to the difference between every two environmental data points to obtain the fluctuation amplitude of the environmental data points. Based on the target environmental data, the trend analysis between environmental data points is performed using straight line fitting, reference data points are selected in the target environmental data, and the data points in the target environmental data and the reference data points are linearly fitted using the least squares method to obtain the target trend straight line, and the slope value of the target trend straight line is calculated, that is, the change trend between the environmental data points is obtained. Based on the incremental probability transfer matrix, the fluctuation amplitude and change trend are used to perform evolution analysis of the environmental time series, historical environmental evolution data are obtained, the change amplitude of each environment in the historical environmental evolution data is calculated, the occurrence probability of each environmental change amplitude is calculated, the incremental probability is calculated according to the occurrence probability, and the incremental probability transfer matrix is ​​constructed according to the incremental probability. The fluctuation amplitude and change trend are used to analyze the data evolution of the environmental data in each time series period through the incremental probability transfer matrix, and the evolution analysis data of the environmental time series is obtained, that is, the state evolution of the environmental data in the time series period, so as to achieve high-precision environmental time series evolution analysis. Based on the fluctuation amplitude and change trend, a hidden Markov model is used to predict and analyze environmental changes, obtain different modal sequences of historical environmental evolution data, obtain state probability matrices based on different modal sequences, and construct a hidden Markov model based on the state probability matrix and the incremental probability transfer matrix. According to the hidden Markov model, the fluctuation amplitude and change trend are used to predict the change state of environmental data, that is, predict the probability of occurrence of the next environmental data point, and obtain environmental change prediction and analysis data. Based on the evolution analysis data and environmental change prediction and analysis data, an environmental time series change analysis is performed, and the current environmental evolution probability is obtained through the evolution analysis data. The current environmental change probability in each time series period is obtained according to the environmental change prediction and analysis data. According to the environmental evolution probability and the environmental change probability in each time series period, the time series change of the environment in which the microwave oven is located is known, and the environmental time series change data is obtained. The environmental time series change data can analyze the impact of the environment on the operation of the device, such as excessive changes in the ambient temperature will cause the device to overheat or insufficient heat dissipation.

[0076] S15: performing operation loss prediction based on the environment time series change data and the target equipment loss data by using operation stage analysis to obtain operation loss prediction data;

[0077] In the specific implementation process of the present invention, the operation loss prediction is performed based on the environment time series change data and the target equipment loss data using the operation stage analysis to obtain the operation loss prediction data, including: generating a loss rate curve based on the target equipment loss data, and obtaining the parameter range of the operation stage based on the loss rate curve; performing operation stage conversion analysis based on the parameter range to obtain operation stage conversion data; performing loss evolution based on the operation stage conversion data using the environment time series change data to obtain loss evolution data; performing uncertainty derivation based on the environment time series change data and the target equipment loss data to obtain uncertainty derivation parameters; performing operation loss prediction based on the loss evolution data and the uncertainty derivation parameters using the loss simulation process to obtain operation loss prediction data.

[0078] Specifically, a loss rate curve is generated based on the target device loss data, the loss rate of the microwave oven at each moment is calculated based on the target device loss data, and the loss rate curve is plotted, and the parameter interval of the operation stage is obtained based on the loss rate curve, that is, the loss rate interval of each operation stage is obtained. Based on the parameter interval, a conversion analysis of the operation stage is performed, and a corresponding change curve is generated according to the parameter interval of each operation stage, and the similarity between the corresponding change curves is calculated. The conversion of the operation stage is analyzed according to the similarity between the change curves, that is, if the similarity is greater than or equal to the preset similarity threshold, the operation stage is converted to obtain the operation stage conversion data. Based on the operation phase conversion data, the environmental time series change data is used to perform loss evolution. According to the operation phase conversion data and the environmental time series change data, a fusion algorithm composed of a successive optimization algorithm and an extended Kalman filter algorithm is used to construct a deduction model. According to the operation phase conversion data and the environmental time series change data, a loss-time fitting curve is generated. According to the deduction model and the loss-time fitting curve, preliminary loss evolution is performed to obtain preliminary loss evolution data. The preliminary loss evolution data is corrected according to the self-learning correction coefficient, which is a correction coefficient obtained through a large number of experiments, to obtain the final loss evolution data. Based on the environmental time series change data and the target equipment loss data, uncertainty derivation is performed. Bayesian joint probability calculation is performed using multivariate Gaussian distribution through the environmental time series change data and the target equipment loss data to obtain Bayesian joint probability. According to the Bayesian joint probability, a posteriori parameters are calculated using likelihood function and Taylor formula. According to the posteriori parameters, uncertainty derivation is performed using preset rate deviation combined with Heaviside function to obtain uncertainty derivation parameters. Since there is uncertainty in the occurrence of potential faults, uncertainty derivation is performed. Based on the loss evolution data and uncertainty derivation parameters, the loss simulation process is used to predict the operation loss. The loss evolution data and uncertainty derivation parameters are input into the loss simulation process to predict the operation loss, and the operation loss prediction data is obtained. The operation loss prediction data can be used to understand the loss caused by the continued operation of the microwave oven under the environmental time series changes and the current loss conditions.

[0079] S16: constructing a knowledge graph, and using the knowledge graph to predict potential faults of the microwave oven based on the environmental time series change data and the operation loss prediction data, to obtain potential fault data;

[0080] In the specific implementation process of the present invention, the knowledge graph is constructed, and the potential fault prediction of the microwave oven is performed using the knowledge graph based on the environmental time series change data and the operation loss prediction data to obtain the potential fault data, including: obtaining the historical fault information and topological structure network of the microwave oven, the historical fault information includes historical environmental information and historical operation loss information, and hierarchical classification of environmental content information and hierarchical classification of loss content information are performed based on the historical fault information to obtain the hierarchical classification data of environmental content information and the hierarchical classification data of loss content information; based on the hierarchical classification data of environmental content information and the loss content information, the hierarchical classification data of environmental content information and the loss content information are obtained. A fault information matrix is ​​constructed based on the hierarchical classification data of capacity information and the topological structure network; a fault coupling impact degree analysis is performed based on the historical fault information to obtain fault coupling impact degree data; a fault tree relationship is constructed based on the fault information matrix and the fault coupling impact degree data to obtain a fault tree structure diagram, and a knowledge graph is constructed based on the fault tree structure diagram; a potential fault prediction model is constructed based on the knowledge graph combined with a Bayesian network prediction model, and based on the potential fault prediction model, potential faults of the microwave oven / steamer are predicted using the environmental time series change data and the operating loss prediction data to obtain potential fault data.

[0081] Specifically, the historical fault information and topological structure network of the microwave oven are obtained, the historical fault information includes historical environmental information and historical operating loss information, and the environmental content information and loss content information are hierarchically classified based on the historical fault information, the semantic keywords in the historical environmental information and the historical operating loss information are obtained, the levels of the environmental content information and the loss content information are determined according to the semantic keywords, and hierarchical classification is performed according to the levels of the environmental content information and the loss content information to obtain environmental content information hierarchical classification data and loss content information hierarchical classification data. A fault information matrix is ​​constructed based on the environmental content information hierarchical classification data, the loss content information hierarchical classification data, and the topological structure network, and the environmental content information hierarchical classification data and the loss content information hierarchical classification data are used as matrix input parameters, and the fault information matrix is ​​constructed in combination with the topological structure network. Based on the historical fault information, the fault coupling influence degree analysis is performed, the coupling relationship and coupling strength data between each fault category information in the historical fault information are obtained, the node metamodel and coupling relationship metamodel are constructed based on the coupling relationship and coupling strength data, the coupling relationship information between each node of the fault function diagram model is initialized based on the node metamodel and the coupling relationship metamodel, the initial model of the fault function diagram is obtained, the initial model is optimized and updated using the hierarchical information, the fault function model is obtained, and the fault coupling influence degree data is obtained according to the fault function model. Based on the fault information matrix and the fault coupling influence degree data, the fault tree relationship is constructed, the corresponding positions of the root node and the remaining nodes in the tree structure are determined according to the fault information matrix and the fault coupling influence degree data, and then the fault tree structure diagram is obtained, and the knowledge graph is constructed based on the fault tree structure diagram. A potential fault prediction model is constructed based on the knowledge graph in combination with a Bayesian network prediction model, the parameters of the Bayesian network prediction model are adjusted by changing the correction coefficient, and the potential faults of the microwave oven are predicted based on the potential fault prediction model using the environmental time series change data and the operating loss prediction data. The environmental time series change data and the operating loss prediction data are input into the potential fault prediction model to obtain the potential fault data, so that the potential faults of the microwave oven can be known in time, the prediction accuracy and efficiency of the potential faults can be improved, and the misjudgment and omission of potential faults can be avoided.

[0082] S17: Obtain several fault management plans from the database, perform correlation matching on the potential fault data and each fault management plan, obtain a correlation matching result, generate a fault risk management plan based on the correlation matching result, and manage the microwave oven based on the fault risk management plan.

[0083] In the specific implementation process of the present invention, the correlation matching of the potential fault data and each fault management scheme to obtain the correlation matching result includes: obtaining a first label text string of the potential fault data and a second label text string of each fault management scheme; performing similarity calculation on the first label text string and each second label text string based on a directed graph to obtain a corresponding similarity calculation result; obtaining a difference string between the first label text string and each second label text string; performing difference calculation on the first label text string and each second label text string using a decision graph conversion based on the difference string to obtain a corresponding difference calculation result; performing correlation matching on the potential fault data and each fault management scheme based on the similarity calculation result and the difference calculation result to obtain a correlation matching result.

[0084] Specifically, several fault management schemes are obtained from the database, and the fault management scheme is a scheme for managing the risk of potential faults of the microwave oven. The first label text string of the potential fault data and the second label text string of each fault management scheme are obtained. The similarity of the first label text string and each second label text string is calculated based on a directed graph, and the first label text string is used as the first node and the second label text string is used as the second node. The joint probability distribution of the first label text string and each second label text string is calculated, and a directed graph is constructed using the first node and the second node according to the joint probability distribution. The similarity between each node is calculated using the first label text string and each second label text string according to the directed graph, that is, the similarity between the first label text string and each second label text string is calculated to obtain the corresponding similarity calculation result. Based on the difference string, the difference calculation of the first label text string and each second label text string is performed using a decision diagram conversion, the difference string of the first label text string and each second label text string is calculated based on a hash function, a first difference calculation result is generated according to the difference string, the first label text string and each second label text string are converted into a binary decision diagram to obtain the corresponding first binary decision diagram and each second binary decision diagram, the difference of the first binary decision diagram and each second binary decision diagram is calculated to obtain the second difference calculation result, and the final difference is calculated according to the first difference calculation result and the second difference calculation result, that is, the corresponding difference calculation result is obtained. Based on the similarity calculation result and the difference calculation result, the potential fault data and each fault management scheme are matched with the correlation, that is, the correlation between the potential fault data and each fault management scheme is calculated according to the similarity calculation result and the difference calculation result, and the correlation matching result is obtained, so as to avoid the instability and incompleteness of the correlation matching caused by a single calculation result. The fault management scheme with the greatest correlation is used as the fault risk management plan, and the microwave oven is managed based on the fault risk management plan. This can ensure that when a device fails, the impact of the failure can be minimized and the reliability and safety of the microwave oven is improved.

[0085] In an embodiment of the present invention, by classifying the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering, more accurate target environmental data and target equipment loss data can be obtained, the efficiency of subsequent data analysis can be improved, and the relevant operation and maintenance data can be used more reasonably. The environmental time series change analysis based on the evolution analysis data and environmental change prediction analysis data generated by the target environmental data can effectively improve the accuracy of the environmental time series analysis. Based on the environmental time series change data and the target equipment loss data, the operation loss prediction is performed using the operation stage analysis. Taking into account the environmental time series changes and combining the operation stage analysis, the accumulated loss of the microwave oven in each operation stage can be more accurately analyzed. At the same time, uncertainty derivation is added to make the obtained operation loss prediction data more reliable. Based on the environmental time series change data and the operation loss prediction data, the knowledge graph is used to predict the potential faults of the microwave oven, avoid misjudgment and omission of potential faults, and achieve more accurate potential fault prediction. A fault risk management plan is generated based on the correlation matching result obtained by matching the potential fault data with each fault management plan. The microwave oven / steamer is managed based on the fault risk management plan to ensure that when a device fails, the impact of the failure can be minimized and the reliability and safety of the microwave oven / steamer is improved. Example

[0086] See also Figure 2 , Figure 2 It is a schematic diagram of the structural composition of a microwave oven-steamer-oven fault prediction system based on data analysis in an embodiment of the present invention, wherein the system includes a cloud server, and the cloud server is deployed with a big data analysis engine; the system is configured to execute the microwave oven-steamer-oven fault prediction method based on data analysis described in the embodiment.

[0087] During the specific implementation of the present invention, the cloud server is used to predict potential failures of the microwave oven and steamer based on the acquired operation and maintenance data. A big data analysis engine is deployed in the cloud server. The big data analysis engine can provide more efficient data analysis. The big data analysis engine can obtain more accurate analysis results through automated data collation and analysis processes, that is, it can obtain more accurate potential failure prediction data and match more reliable risk management plans, so as to better pre-manage the risks of the microwave oven and steamer. Figure 2 The system shown does not constitute a limitation of all components, and may include more or less components than shown, or combine certain components.

[0088] In the specific implementation process of the present invention, the specific implementation methods of the system items can be referred to the above embodiments, which will not be repeated here.

[0089] In an embodiment of the present invention, by classifying the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering, more accurate target environmental data and target equipment loss data can be obtained, the efficiency of subsequent data analysis can be improved, and the relevant operation and maintenance data can be used more reasonably. The environmental time series change analysis based on the evolution analysis data and environmental change prediction analysis data generated by the target environmental data can effectively improve the accuracy of the environmental time series analysis. Based on the environmental time series change data and the target equipment loss data, the operation loss prediction is performed using the operation stage analysis. Taking into account the environmental time series changes and combining the operation stage analysis, the accumulated loss of the microwave oven in each operation stage can be more accurately analyzed. At the same time, uncertainty derivation is added to make the obtained operation loss prediction data more reliable. Based on the environmental time series change data and the operation loss prediction data, the knowledge graph is used to predict the potential faults of the microwave oven, avoid misjudgment and omission of potential faults, and achieve more accurate potential fault prediction. A fault risk management plan is generated based on the correlation matching result obtained by matching the potential fault data with each fault management plan. The microwave oven / steamer is managed based on the fault risk management plan to ensure that when a device fails, the impact of the failure can be minimized and the reliability and safety of the microwave oven / steamer is improved. Example

[0090] See also Figure 3 , Figure 3 : is a schematic diagram of a structure of a microwave oven fault prediction device based on data analysis in an embodiment of the present invention, the device is applied to a cloud server, and the cloud server is deployed with a big data analysis engine; the device includes:

[0091] Data preprocessing module 31: a big data analysis engine for a cloud server acquires operation and maintenance data of the microwave oven using a data collection strategy, and preprocesses the operation and maintenance data to obtain preprocessed operation and maintenance data;

[0092] Data block module 32: used to perform block processing on the pre-processed operation and maintenance data to obtain the block-processed operation and maintenance data;

[0093] Data classification module 33: used for classifying the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering to obtain target environmental data and target equipment loss data;

[0094] The environment time series change analysis module 34 is used to perform environment time series change analysis based on the target environment data to obtain environment time series change data;

[0095] Operation loss prediction module 35: used to perform operation loss prediction based on the environment time series change data and the target equipment loss data by using operation stage analysis to obtain operation loss prediction data;

[0096] Potential fault prediction module 36: used for constructing a knowledge graph, and using the knowledge graph to predict potential faults of the microwave oven based on the environmental time series change data and the operation loss prediction data, to obtain potential fault data;

[0097] Fault risk management module 37: used to obtain several fault management plans in the database, perform correlation matching on the potential fault data and each fault management plan, obtain correlation matching results, and generate a fault risk management plan based on the correlation matching results, and manage the microwave oven based on the fault risk management plan.

[0098] In the specific implementation process of the present invention, the specific implementation methods of the device items can be referred to the above embodiments, which will not be repeated here.

[0099] In an embodiment of the present invention, by classifying the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering, more accurate target environmental data and target equipment loss data can be obtained, the efficiency of subsequent data analysis can be improved, and the relevant operation and maintenance data can be used more reasonably. The environmental time series change analysis based on the evolution analysis data and environmental change prediction analysis data generated by the target environmental data can effectively improve the accuracy of the environmental time series analysis. Based on the environmental time series change data and the target equipment loss data, the operation loss prediction is performed using the operation stage analysis. Taking into account the environmental time series changes and combining the operation stage analysis, the accumulated loss of the microwave oven in each operation stage can be more accurately analyzed. At the same time, uncertainty derivation is added to make the obtained operation loss prediction data more reliable. Based on the environmental time series change data and the operation loss prediction data, the knowledge graph is used to predict the potential faults of the microwave oven, avoid misjudgment and omission of potential faults, and achieve more accurate potential fault prediction. A fault risk management plan is generated based on the correlation matching result obtained by matching the potential fault data with each fault management plan. The microwave oven / steamer is managed based on the fault risk management plan to ensure that when a device fails, the impact of the failure can be minimized and the reliability and safety of the microwave oven / steamer is improved.

[0100] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0101] In addition, the above is a detailed introduction to a microwave oven fault prediction method based on data analysis and related devices provided in an embodiment of the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for predicting faults of a microwave oven based on data analysis, characterized in that: Applied to a cloud server, the cloud server is deployed with a big data analysis engine; the method comprises: The big data analysis engine of the cloud server obtains the operation and maintenance data of the microwave oven by using the data collection strategy, and pre-processes the operation and maintenance data to obtain the pre-processed operation and maintenance data; The pre-processed operation and maintenance data is processed in blocks to obtain the block-processed operation and maintenance data; Based on hypergraph clustering, the environmental data and equipment loss data of the block-processed operation and maintenance data are classified to obtain target environmental data and target equipment loss data; Performing an environmental time series change analysis based on the target environmental data to obtain environmental time series change data; Based on the environmental time series change data and the target equipment loss data, operation loss prediction is performed using operation stage analysis to obtain operation loss prediction data; Constructing a knowledge graph, and using the knowledge graph to predict potential faults of the microwave oven and steamer based on the environmental time series change data and the operation loss prediction data to obtain potential fault data, specifically including: obtaining historical fault information and a topological structure network of the microwave oven and steamer, wherein the historical fault information includes historical environmental information and historical operation loss information, and hierarchical classification of environmental content information and hierarchical classification of loss content information are performed based on the historical fault information to obtain environmental content information hierarchical classification data and loss content information hierarchical classification data; constructing a fault information matrix based on the environmental content information hierarchical classification data, the loss content information hierarchical classification data and the topological structure network; analyzing the degree of fault coupling influence based on the historical fault information to obtain fault coupling influence data; constructing a fault tree relationship based on the fault information matrix and the fault coupling influence data to obtain a fault tree structure diagram, and constructing a knowledge graph based on the fault tree structure diagram; constructing a potential fault prediction model based on the knowledge graph combined with a Bayesian network prediction model, and predicting potential faults of the microwave oven and steamer based on the potential fault prediction model using the environmental time series change data and the operation loss prediction data to obtain potential fault data; A plurality of fault management schemes are obtained from a database, correlation matching is performed between the potential fault data and each fault management scheme to obtain a correlation matching result, and a fault risk management plan is generated based on the correlation matching result, and the microwave oven is managed based on the fault risk management plan.

2. The method for predicting faults of a microwave oven based on data analysis according to claim 1, characterized in that: The method of using the data collection strategy to obtain the operation and maintenance data of the microwave oven and steamer, and preprocessing the operation and maintenance data to obtain the preprocessed operation and maintenance data includes: Determine the data collection time and data collection scope based on the preset resource tag, and construct a data collection window based on the adaptation relationship between the data collection time and the data collection scope; Determining a data collection strategy based on the data collection window, and acquiring operation and maintenance data of the microwave oven based on the data collection strategy; Performing data cleaning on the operation and maintenance data to obtain the operation and maintenance data after the data cleaning; Perform data integration processing on the operation and maintenance data after data cleaning to obtain pre-processed operation and maintenance data.

3. The method for predicting faults of a microwave oven based on data analysis according to claim 1, characterized in that: The block processing of the pre-processed operation and maintenance data to obtain the block-processed operation and maintenance data includes: A sliding window is set, and based on the sliding window, the pre-processed operation and maintenance data is processed into blocks by utilizing boundary connection to obtain the block-processed operation and maintenance data.

4. The method for predicting faults of a microwave oven based on data analysis according to claim 1, characterized in that: The method of classifying the environmental data and the equipment loss data of the block-processed operation and maintenance data based on the hypergraph clustering to obtain the target environmental data and the target equipment loss data includes: Perform node coding and labeling on the block-processed operation and maintenance data to obtain several coding categories; Obtaining a node coding matrix of each coding category, and constructing a hypergraph model based on the node coding matrix using data association support analysis; Based on the hypergraph model, clustering is performed on the block-processed operation and maintenance data to obtain a clustering result; The clustering result is used to classify the environmental data and the equipment loss data based on the tree classification and the network structure association classification method to obtain the target environmental data and the target equipment loss data.

5. The method for predicting faults of a microwave oven based on data analysis according to claim 1, characterized in that: The step of performing an environmental time series change analysis based on the target environmental data to obtain environmental time series change data includes: Performing a fluctuation range analysis of the environmental data points based on the target environmental data to obtain the fluctuation range of the environmental data points; Based on the target environmental data, a trend analysis is performed between environmental data points by using straight line fitting to obtain a change trend between environmental data points; Based on the incremental probability transfer matrix, the fluctuation amplitude and the change trend are used to perform evolution analysis of the environmental time series to obtain evolution analysis data of the environmental time series; Based on the fluctuation range and change trend, a hidden Markov model is used to perform environmental change prediction analysis to obtain environmental change prediction analysis data; Based on the evolution analysis data and the environmental change prediction analysis data, an environmental time series change analysis is performed to obtain environmental time series change data.

6. The method for predicting faults of a microwave oven based on data analysis according to claim 1, characterized in that: The step of performing operation loss prediction based on the environment time series change data and the target equipment loss data by using the operation stage analysis to obtain the operation loss prediction data includes: Generate a loss rate curve based on the target equipment loss data, and obtain a parameter range of an operation stage based on the loss rate curve; Performing a conversion analysis of the operation phase based on the parameter interval to obtain operation phase conversion data; Based on the operation phase conversion data, the environmental time series change data is used to perform loss evolution to obtain loss evolution data; Derivation of uncertainty is performed based on the environmental time series variation data and the target equipment loss data to obtain uncertainty derivation parameters; Based on the loss evolution data and uncertainty derivation parameters, the operation loss prediction is performed using the loss simulation process to obtain the operation loss prediction data.

7. The method for predicting faults of a microwave oven based on data analysis according to claim 1, characterized in that: The performing correlation matching on the potential fault data and each fault management scheme to obtain a correlation matching result includes: Acquire a first label text string of the potential fault data and a second label text string of each fault management solution; Performing similarity calculation on the first label text string and each second label text string based on the directed graph to obtain corresponding similarity calculation results; Obtain a difference string between the first label text string and each second label text string; Based on the difference character string, performing difference calculation on the first label text character string and each second label text character string by using decision diagram conversion to obtain corresponding difference calculation results; Based on the similarity calculation result and the difference calculation result, correlation matching is performed on the potential fault data and each fault management solution to obtain a correlation matching result.

8. A device for predicting faults of a microwave oven based on data analysis, characterized in that: Applied to a cloud server, the cloud server is deployed with a big data analysis engine; the device comprises: A data preprocessing module, in which the big data analysis engine of the cloud server uses a data collection strategy to obtain the operation and maintenance data of the microwave oven and steamer, and preprocesses the operation and maintenance data to obtain the preprocessed operation and maintenance data; The data block module is used to block the preprocessed operation and maintenance data to obtain the block-processed operation and maintenance data; A data classification module is used to classify the environmental data and equipment loss data of the block-processed operation and maintenance data based on hypergraph clustering to obtain target environmental data and target equipment loss data; An environment time series change analysis module, used to perform environment time series change analysis based on the target environment data to obtain environment time series change data; An operation loss prediction module, used to perform operation loss prediction based on the environment time series change data and the target equipment loss data by using operation stage analysis to obtain operation loss prediction data; A potential fault prediction module is used to construct a knowledge graph, and based on the environmental time series change data and the operation loss prediction data, the knowledge graph is used to predict the potential faults of the microwave oven and steamer, and obtain the potential fault data, specifically including: obtaining the historical fault information and topological structure network of the microwave oven and steamer, wherein the historical fault information includes the historical environmental information and the historical operation loss information, and hierarchical classification of the environmental content information and the hierarchical classification of the loss content information are performed based on the historical fault information to obtain the hierarchical classification data of the environmental content information and the hierarchical classification data of the loss content information; based on the hierarchical classification data of the environmental content information and the loss content information, the hierarchical classification data of the loss content information are obtained ... A fault information matrix is ​​constructed based on information hierarchical classification data and a topological structure network; a fault coupling influence degree analysis is performed based on the historical fault information to obtain fault coupling influence degree data; a fault tree relationship is constructed based on the fault information matrix and the fault coupling influence degree data to obtain a fault tree structure diagram, and a knowledge graph is constructed based on the fault tree structure diagram; a potential fault prediction model is constructed based on the knowledge graph combined with a Bayesian network prediction model, and based on the potential fault prediction model, the environmental time series change data and the operating loss prediction data are used to predict the potential faults of the microwave oven and steamer to obtain potential fault data; The fault risk management module is used to obtain several fault management schemes in the database, perform correlation matching on the potential fault data and each fault management scheme, obtain the correlation matching result, generate a fault risk management plan based on the correlation matching result, and manage the microwave oven based on the fault risk management plan.

9. A microwave oven fault prediction system based on data analysis, characterized in that: The system includes a cloud server, on which a big data analysis engine is deployed. The system is configured to execute the microwave oven fault prediction method based on data analysis as described in any one of claims 1 to 7.

Images