A fault diagnosis and optimization method for organic electroluminescent drive circuit

Abstract

The present invention relates to the field of organic electroluminescence technology. The present invention provides an organic electroluminescence drive circuit fault diagnosis optimization method, including obtaining basic circuit data and generating images, dividing circuit areas; establishing partition labels and performing random data screening, building a partition fault diagnosis model knowledge base; extracting historical fault information and performing cluster analysis, generating simulated fault tasks; matching real-time fault diagnosis models through simulated fault tests, obtaining fault diagnosis results; detecting diagnosis results and optimizing models. The present invention solves the problem that the prior art can only diagnose known faults, realizes real-time analysis and dynamic optimization of circuit faults, improves the accuracy and efficiency of fault diagnosis, reduces maintenance costs, and promotes the further development of organic electroluminescence technology.

Description

Technical Field

[0001] The invention relates to the technical field of organic electroluminescence, and in particular to a fault diagnosis and optimization method for an organic electroluminescence driving circuit. Background Art

[0002] Organic electroluminescent drive circuits are widely used in the fields of commercial display technology, computer display technology, industrial display technology, and traffic display technology. With the continuous development of organic electroluminescent technology, its wide application in small and medium-sized display screens and the higher requirements for brightness and life of medium and large-sized display screens have promoted the research and application of new organic electroluminescent structures (such as laminated organic electroluminescent structures) and driving technologies (such as AM organic electroluminescent technology). New organic electroluminescent technologies and structures will further improve the performance and stability of organic electroluminescent display devices and expand their application areas.

[0003] The organic electroluminescent drive circuit is one of the core parts of the organic electroluminescent technology. The performance of the organic electroluminescent drive circuit directly affects the display effect and application prospects of the organic electroluminescent. The structure of the organic electroluminescent drive circuit is relatively complex, but it has better resolution and color performance, and is suitable for large-size and large-information display. Using amorphous silicon TFT (a-SiTFT) or polycrystalline silicon (poly-SiTFT) switching components, the input signal is stored on the storage capacitor, so that the pixel remains connected during the frame period, so transient high brightness is not required, overcoming the shortcomings of PM organic electroluminescence and is not limited by duty cycle. Active drive organic electroluminescent drive circuits are mainly divided into two types according to the driving mode: p-SiTFT and a-Si:HTFT drive. Among them, the low-temperature polycrystalline silicon LTPSTFT in p-SiTFT has a higher electron mobility than a-Si:HTFT, and can provide a larger current for organic electroluminescence. The organic electroluminescent drive circuit can be used as a switch and drive element for organic electroluminescent pixels. This method is currently widely accepted.

[0004] During the actual use of the organic electroluminescent drive circuit, the failure of the organic electroluminescent drive circuit manifests itself in various forms, such as uneven brightness, color distortion, image flickering, etc. These failures are caused by various reasons, such as component aging, poor contact, unstable driving voltage, etc. Therefore, accurate judgment of the cause of the failure requires rich experience and professional testing equipment. The existing organic electroluminescent drive circuit fault diagnosis can only diagnose known faults, and it is impossible to analyze the organic electroluminescent drive circuit fault diagnosis model in real time. Therefore, the present invention provides an organic electroluminescent drive circuit fault diagnosis optimization method. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an organic electroluminescent drive circuit fault diagnosis optimization method to solve the problem that the existing organic electroluminescent drive circuit fault diagnosis can only diagnose known faults and cannot perform real-time organic electroluminescent drive circuit fault analysis and diagnosis models.

[0006] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0007] The present invention provides a method for fault diagnosis and optimization of an organic electroluminescent drive circuit, comprising:

[0008] Step S101, acquiring basic data of an organic electroluminescent drive circuit, collecting real-time operation data of the organic electroluminescent drive circuit, substituting the basic data of the organic electroluminescent drive circuit into a preset image generation model, obtaining a basic image of the organic electroluminescent drive circuit, configuring coordinates in the basic image of the organic electroluminescent drive circuit, dividing the mechanical energy area of ​​the organic electroluminescent drive circuit according to the configuration coordinates of the basic image of the organic electroluminescent drive circuit, and obtaining a partition of the organic electroluminescent drive circuit;

[0009] Step S102, establishing labels for the organic electroluminescent drive circuit partitions to obtain a partition label set, performing random data screening on the partition label set to obtain multiple groups of random partition label combinations, collecting historical data of each organic electroluminescent drive circuit partition, performing feature extraction on the historical data of each organic electroluminescent drive circuit partition to obtain partition data feature extraction results, matching the partition data feature extraction results in a preset fault analysis model knowledge base to obtain a fault diagnosis model set corresponding to the organic electroluminescent drive circuit partition, and establishing the fault diagnosis model set corresponding to the organic electroluminescent drive circuit partition into a partition fault diagnosis model knowledge base;

[0010] Step S103, extracting fault information from historical data of each organic electroluminescent drive circuit partition to obtain historical fault information of each organic electroluminescent drive circuit partition, performing cluster analysis on the historical fault information of each organic electroluminescent drive circuit partition to obtain a historical fault cluster analysis result of each organic electroluminescent drive circuit partition, matching the historical fault cluster analysis result of each organic electroluminescent drive circuit partition in a preset fault simulation library to obtain a simulated fault task set for each organic electroluminescent drive circuit partition;

[0011] Step S104, sending the simulated fault tasks in the simulated fault task set of each organic electroluminescent drive circuit partition to the corresponding organic electroluminescent drive circuit partition in sequence, obtaining the simulated fault test results of the organic electroluminescent drive circuit partition, performing fault diagnosis model matching on the simulated fault test results of the organic electroluminescent drive circuit partition through the corresponding fault diagnosis model knowledge base, obtaining a real-time fault diagnosis model group for each partition, using each partition real-time fault diagnosis model group to perform data analysis on the simulated fault test results of the organic electroluminescent drive circuit partition, and obtaining a fault diagnosis result for each partition;

[0012] Step S105, detect the fault diagnosis result of each partition, obtain the fault diagnosis success information of each partition and the fault diagnosis unsuccessful information of each partition, retrieve the data information involved in the data processing process corresponding to the unsuccessful fault diagnosis information of each partition, train and optimize the fault diagnosis model according to the data information involved in the data processing process corresponding to the unsuccessful fault diagnosis information of each partition, and obtain the optimized fault diagnosis model, store the optimized fault diagnosis model in the partition fault diagnosis model knowledge base corresponding to the partition, and use the optimized fault diagnosis model stored in the partition fault diagnosis model knowledge base corresponding to the partition to perform fault diagnosis on the real-time operation data of the organic electroluminescent drive circuit of each partition.

[0013] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S101 comprises:

[0014] Acquire basic data of an organic electroluminescent drive circuit, the basic data of the organic electroluminescent drive circuit including circuit design data, component layout data, and circuit connection relationship data;

[0015] The basic data of the organic electroluminescent driving circuit is converted into a visual circuit image using a preset image generation model;

[0016] On the generated visualized circuit image, a coordinate system is established, where the coordinate system uses a three-dimensional coordinate system including an X coordinate, a Y coordinate, and a Z coordinate;

[0017] receiving a function of the organic electroluminescent driving circuit, and dividing the basic image into a plurality of regions according to the function of the organic electroluminescent driving circuit, each region corresponding to a combination of components in the circuit;

[0018] Assign a coordinate range to each divided area, which will be used for subsequent data analysis and fault diagnosis;

[0019] The results of the area division and coordinate configuration are recorded to form a complete record of the basic image configuration coordinates of the organic electroluminescent drive circuit.

[0020] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S101 comprises:

[0021] accurately aligning the base image of the organic electroluminescent driving circuit with a pre-established coordinate system to identify the structure in the organic electroluminescent driving circuit;

[0022] According to the functions of the organic electroluminescent driving circuit, the visualized circuit image is divided into different functional areas including input area, output area, control area and power supply area;

[0023] Determine the boundaries of each identified functional area and assign a coordinate range to each partition, the coordinate range is used to reflect the location and area of ​​the partition in the circuit;

[0024] Record the information of each partition including the name of each partition, the function of each partition and the boundary coordinates of each partition.

[0025] A partition and coordinate configuration table is formed.

[0026] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S102 includes:

[0027] Determine the screening parameters, including the proportion of random sampling, the number of label combinations in each group, and whether repeated sampling is allowed;

[0028] According to the random sampling algorithm, labels are extracted from the partition label set to form multiple groups of random partition label combinations;

[0029] Check the randomly drawn partition label combinations. If they do not meet the sampling quality requirements, including uneven sample distribution and many repeated sampling, adjust the screening conditions and parameter settings, and re-execute the random data screening steps.

[0030] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S102 includes:

[0031] Access the established fault analysis model knowledge base, which includes multiple fault diagnosis models and their corresponding feature descriptions or matching rules;

[0032] The data feature extraction results of each partition are matched one by one with the models in the fault analysis model knowledge base, and the matching degree between the data features and the models is quantified using the similarity calculation method;

[0033] According to the similarity calculation results, the fault diagnosis model with the highest matching degree is selected as the model corresponding to the partition;

[0034] The fault diagnosis models matched for each partition are summarized to form a complete set of fault diagnosis models.

[0035] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S103 includes:

[0036] Filter out fault information from the historical data of each organic electroluminescent drive circuit partition, the fault information including fault time, fault type and fault description;

[0037] Pre-process the screened fault information to identify the characteristics of cluster analysis, which include fault frequency, fault duration, and fault impact range;

[0038] According to the characteristics of fault information and clustering requirements, set the parameters of the clustering algorithm, which include the number of clusters, distance measurement, and number of iterations;

[0039] The preprocessed fault feature data is input into the selected clustering algorithm, clustering analysis is performed, and preliminary clustering results are obtained.

[0040] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S103 includes:

[0041] Match the clustering features of each partition with the simulation tasks in the fault simulation library one by one, and select the best matching simulation fault task for each cluster of each partition according to the matching degree;

[0042] Summarize the simulated fault tasks matched to each partition to form a simulated fault task set for the partition;

[0043] Classify or sort the simulated fault task set according to fault type, impact degree, occurrence frequency, etc.;

[0044] The simulated fault task set of each partition is output for fault simulation and fault diagnosis strategy formulation.

[0045] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, said step S104 comprises:

[0046] Collect the data generated by the simulated fault test from each organic electroluminescent drive circuit partition, pre-process the data,

[0047] Categorize data by partition and failure type;

[0048] According to the different partitions and fault types, a corresponding real-time fault diagnosis model is selected from a fault diagnosis model knowledge base, and the selected fault diagnosis model is loaded into a data analysis environment;

[0049] Input the preprocessed simulated fault test data into the corresponding fault diagnosis model, run the fault diagnosis model, and process and analyze the input data;

[0050] The fault diagnosis result of each partition is obtained from the fault diagnosis model, and the fault diagnosis result of each partition includes the fault type, location and severity.

[0051] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S105 comprises:

[0052] Filter out unsuccessful information from the fault diagnosis results of each partition, and record detailed information of the unsuccessful information, where the detailed information of the unsuccessful information includes the fault type, the diagnosis error, and the timestamp;

[0053] Collect data information involved in the unsuccessful information data processing process, including input data, intermediate processing results and model output;

[0054] Identify the collected data information, identify the data anomalies that lead to unsuccessful diagnosis, and obtain the identification results;

[0055] Based on the recognition results, it is necessary to determine the optimization objectives of the fault diagnosis model, which include feature extraction, algorithm improvement, and parameter adjustment.

[0056] Furthermore, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S105 comprises:

[0057] Selecting an optimized fault diagnosis model from a partition fault diagnosis model knowledge base corresponding to the partition, and loading the selected optimized fault diagnosis model into the fault diagnosis system;

[0058] Input the preprocessed real-time operation data into the optimized fault diagnosis model, run the optimized fault diagnosis model, and process the input real-time data;

[0059] Fault diagnosis results are obtained from the optimized fault diagnosis model, and the fault diagnosis results include fault type, fault location, fault severity and recommended solutions.

[0060] Beneficial effects of the present invention:

[0061] The present invention collects the operation data of the organic electroluminescent drive circuit in real time and performs fault diagnosis. The present invention can timely discover and handle faults in the circuit, significantly improving the reliability and stability of the system. This real-time performance is crucial to improving the continuous normal operation of the display device.

[0062] The present invention utilizes partition label sets and random data screening technology to construct a partition fault diagnosis model knowledge base, and performs targeted fault diagnosis model matching for each partition, which can effectively improve the accuracy of fault diagnosis and reduce false positives and negative negatives.

[0063] In the process of fault diagnosis, the present invention conducts in-depth analysis of unsuccessful information, and trains and optimizes the fault diagnosis model accordingly. This dynamic optimization capability enables the model to adapt to the fault diagnosis needs in different situations, and continuously improves the accuracy and efficiency of diagnosis. By constructing a partitioned fault diagnosis model knowledge base, the present invention can perform targeted fault diagnosis on different areas of the organic electroluminescent drive circuit, thereby covering a wider range of fault types and scenarios. This comprehensiveness improves the applicability and practicality of fault diagnosis.

[0064] By discovering and diagnosing faults in advance, the present invention can reduce equipment downtime and maintenance costs caused by faults. At the same time, the optimized fault diagnosis model can reduce dependence on professionals and reduce labor costs.

[0065] The present invention provides an innovative method for optimizing fault diagnosis of an organic electroluminescent drive circuit, which provides strong support for the development of organic electroluminescent technology. By continuously improving the accuracy and efficiency of fault diagnosis, the present invention helps promote the application of organic electroluminescent technology in a wider range of fields.

[0066] In summary, the present invention significantly improves the real-time, accuracy, comprehensiveness and dynamic optimization capability of fault diagnosis of organic electroluminescent drive circuits through a series of innovative technical means, reduces maintenance costs, and promotes the further development of organic electroluminescent technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0068] Figure 1 A schematic diagram of a fault diagnosis and optimization method for an organic electroluminescent drive circuit provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. 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. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0070] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0071] The present invention provides a method for fault diagnosis and optimization of an organic electroluminescent drive circuit, comprising:

[0072] Step S101, acquiring basic data of an organic electroluminescent drive circuit, collecting real-time operation data of the organic electroluminescent drive circuit, substituting the basic data of the organic electroluminescent drive circuit into a preset image generation model, obtaining a basic image of the organic electroluminescent drive circuit, configuring coordinates in the basic image of the organic electroluminescent drive circuit, dividing the mechanical energy area of ​​the organic electroluminescent drive circuit according to the configuration coordinates of the basic image of the organic electroluminescent drive circuit, and obtaining a partition of the organic electroluminescent drive circuit;

[0073] In step S101, the present invention provides an initial step of a method for optimizing fault diagnosis of an organic electroluminescent drive circuit, which specifically involves the following sub-steps:

[0074] Obtain basic data: Collect basic data of the organic electroluminescent drive circuit. These basic data include but are not limited to circuit design data, component layout data, and circuit connection relationship data. These data are the basis for circuit analysis and fault diagnosis in subsequent steps.

[0075] Image generation: Substitute the collected basic data into the preset image generation model. This image generation model can convert complex circuit data into a visual circuit image, making the circuit structure and component layout more intuitive and easy to understand.

[0076] Configure coordinate system: Configure a coordinate system on the generated visual circuit image. This coordinate system usually uses a three-dimensional coordinate system, including X coordinates, Y coordinates, and Z coordinates, so that each component and area in the circuit can be accurately located.

[0077] Area division: According to the function and structural characteristics of the organic electroluminescent drive circuit, combined with the configured coordinate system, the circuit is divided into regions. The purpose of this step is to divide the circuit into multiple relatively independent functional regions, each region corresponding to a group of component combinations in the circuit.

[0078] Partition record: Assign a coordinate range to each divided area and record the name, function, boundary coordinates and other information of each area. This information will be used in the subsequent data analysis and fault diagnosis process to improve the ability to accurately locate and analyze faults in the circuit.

[0079] Through step S101, the present invention realizes the preliminary analysis and regional division of the organic electroluminescent driving circuit, and provides the necessary basic data and structural information for the construction of the fault diagnosis model and real-time fault diagnosis in the subsequent steps. This regional division method based on image generation and coordinate configuration makes the circuit analysis more intuitive and accurate, and helps to improve the accuracy and efficiency of fault diagnosis.

[0080] Step S102, establishing labels for the organic electroluminescent drive circuit partitions to obtain a partition label set, performing random data screening on the partition label set to obtain multiple groups of random partition label combinations, collecting historical data of each organic electroluminescent drive circuit partition, performing feature extraction on the historical data of each organic electroluminescent drive circuit partition to obtain partition data feature extraction results, matching the partition data feature extraction results in a preset fault analysis model knowledge base to obtain a fault diagnosis model set corresponding to the organic electroluminescent drive circuit partition, and establishing the fault diagnosis model set corresponding to the organic electroluminescent drive circuit partition into a partition fault diagnosis model knowledge base;

[0081] In step S102, the present invention further deepens the process of the organic electroluminescent drive circuit fault diagnosis optimization method. Through a series of fine operations, a corresponding fault diagnosis model knowledge base is established for each circuit partition. The following is a detailed analysis of this step:

[0082] Labels are established for the organic electroluminescent drive circuit partitions obtained in step S101. Each partition is assigned a unique label to identify and distinguish different circuit areas. These labels form a partition label set, which facilitates subsequent data processing and model matching.

[0083] Perform random data screening on the partition label set. By setting certain screening parameters (such as the proportion of random sampling, the number of label combinations in each group, whether repeated sampling is allowed, etc.), multiple groups of random partition label combinations are extracted from the label set. This step is designed to simulate different fault scenarios and combinations to improve the generalization ability of the fault diagnosis model.

[0084] For each organic electroluminescent drive circuit partition, its historical operation data is collected. These data include the performance of the circuit under different working conditions and the fault conditions that occur, which is an important basis for subsequent data feature extraction and model matching.

[0085] Based on the collected historical data, feature extraction is performed on the historical data of each partition. This step aims to mine useful information for fault diagnosis from the original data, such as characteristic parameters such as voltage fluctuation, current change, and temperature anomaly. Through feature extraction, complex original data can be converted into feature vectors that are easy for the model to process.

[0086] The extracted partition data features are matched in the preset fault analysis model knowledge base. The fault analysis model knowledge base is a database containing multiple fault diagnosis models and their corresponding feature descriptions or matching rules. Through similarity calculation and other methods, the most suitable fault diagnosis model is matched for each partition. Finally, the fault diagnosis model set corresponding to each partition is sorted out to establish a partition fault diagnosis model knowledge base.

[0087] Through step S102, the present invention establishes a targeted fault diagnosis model knowledge base for each partition of the organic electroluminescent drive circuit. These models can accurately judge and predict the fault conditions in the circuit operation based on the historical data and characteristic parameters of the partition. The implementation of this step provides a solid foundation for real-time fault diagnosis and model optimization in subsequent steps.

[0088] Step S103, extracting fault information from historical data of each organic electroluminescent drive circuit partition to obtain historical fault information of each organic electroluminescent drive circuit partition, performing cluster analysis on the historical fault information of each organic electroluminescent drive circuit partition to obtain a historical fault cluster analysis result of each organic electroluminescent drive circuit partition, matching the historical fault cluster analysis result of each organic electroluminescent drive circuit partition in a preset fault simulation library to obtain a simulated fault task set for each organic electroluminescent drive circuit partition;

[0089] In step S103, the present invention constructs a more comprehensive and accurate fault diagnosis system by deeply analyzing the historical fault data of the organic electroluminescent drive circuit partition. The following is a detailed analysis of this step:

[0090] Fault information is extracted from the historical data of each organic electroluminescent drive circuit partition. This information includes but is not limited to the time when the fault occurred, the fault type (such as uneven brightness, color distortion, image flickering, etc.) and the fault description. The extraction of fault information is the basis for subsequent clustering analysis and model matching.

[0091] Next, cluster analysis is performed on the historical fault information of each partition. Cluster analysis is a process of dividing a data set into multiple classes or clusters, so that the similarity between data points in the same cluster is high, while the similarity between data points in different clusters is low. In this step, by setting the parameters of the clustering algorithm (such as the number of clusters, distance measurement, number of iterations, etc.), the historical fault information is clustered according to its characteristics (such as fault frequency, fault duration, fault impact range, etc.), and the historical fault clustering analysis results of each partition are obtained.

[0092] The results of cluster analysis are matched with the preset fault simulation library. The fault simulation library is a database containing a variety of simulated fault tasks and fault scenarios, which are designed to simulate various fault conditions that occur in actual circuit operation. Through matching, the set of simulated fault tasks that best matches the historical fault cluster analysis results is selected for each partition.

[0093] According to the matching results, a simulated fault task set is constructed for each organic electroluminescent drive circuit partition. The task set contains simulation tasks for various fault conditions that may occur in the partition, which is used for subsequent fault simulation and fault diagnosis strategy formulation.

[0094] Through step S103, the present invention realizes in-depth mining and utilization of historical fault data. Through cluster analysis and fault simulation library matching, not only the inherent laws and characteristics of circuit partition faults are revealed, but also an important reference basis is provided for subsequent fault simulation and real-time fault diagnosis. The implementation of this step makes the fault diagnosis method more scientific, comprehensive and efficient.

[0095] Step S104, sending the simulated fault tasks in the simulated fault task set of each organic electroluminescent drive circuit partition to the corresponding organic electroluminescent drive circuit partition in sequence, obtaining the simulated fault test results of the organic electroluminescent drive circuit partition, performing fault diagnosis model matching on the simulated fault test results of the organic electroluminescent drive circuit partition through the corresponding fault diagnosis model knowledge base, obtaining a real-time fault diagnosis model group for each partition, using each partition real-time fault diagnosis model group to perform data analysis on the simulated fault test results of the organic electroluminescent drive circuit partition, and obtaining a fault diagnosis result for each partition;

[0096] In step S104, the present invention further deepens the application of the organic electroluminescent drive circuit fault diagnosis optimization method, and realizes accurate diagnosis of the fault condition of each circuit partition through the application of simulated fault testing and real-time fault diagnosis model. The following is a detailed analysis of this step:

[0097] The simulated fault tasks in the simulated fault task set constructed for each organic electroluminescent driving circuit partition in step S103 are sent to the corresponding circuit partition in sequence. This step is intended to test the response and performance of the circuit partition by simulating the fault conditions that occur in actual operation.

[0098] After executing the simulated fault task, collect and obtain the simulated fault test results of each circuit partition. These results reflect the actual performance of the circuit partition under the simulated fault condition, including key information such as fault phenomenon and impact degree.

[0099] The obtained simulated fault test results are matched with the corresponding fault diagnosis model knowledge base. In step S102, a partition fault diagnosis model knowledge base has been established for each circuit partition, which contains multiple fault diagnosis models for the partition. Through matching, the most appropriate real-time fault diagnosis model group is selected for each circuit partition.

[0100] The simulated fault test results are analyzed using selected real-time fault diagnosis model groups. Based on advanced algorithms and technologies, these model groups can deeply interpret and process the test results to accurately diagnose the fault type, location, and severity of the circuit partition.

[0101] Finally, the fault diagnosis results of each circuit partition are output. These results not only contain the specific information of the fault, but also include the recommended solutions for the fault, which provides an important reference for subsequent fault repair and maintenance.

[0102] Step S105, detect the fault diagnosis result of each partition, obtain the fault diagnosis success information of each partition and the fault diagnosis unsuccessful information of each partition, retrieve the data information involved in the data processing process corresponding to the unsuccessful fault diagnosis information of each partition, train and optimize the fault diagnosis model according to the data information involved in the data processing process corresponding to the unsuccessful fault diagnosis information of each partition, and obtain the optimized fault diagnosis model, store the optimized fault diagnosis model in the partition fault diagnosis model knowledge base corresponding to the partition, and use the optimized fault diagnosis model stored in the partition fault diagnosis model knowledge base corresponding to the partition to perform fault diagnosis on the real-time operation data of the organic electroluminescent drive circuit of each partition.

[0103] In step S105, the present invention further improves the efficiency of the organic electroluminescent drive circuit fault diagnosis optimization method, and realizes continuous optimization of the fault diagnosis model through strict detection of the fault diagnosis results and in-depth analysis of unsuccessful diagnosis information. The following is a detailed analysis of this step:

[0104] The fault diagnosis results of each circuit partition are tested. This step aims to identify which diagnoses are successful (i.e., the faults are accurately identified and located) and which are unsuccessful (i.e., the faults are not correctly identified or located).

[0105] According to the test results, the successful and unsuccessful fault diagnosis information of each partition is recorded. The successful information includes the type and location of the fault that is correctly diagnosed, while the unsuccessful information includes the details of the fault that is not correctly diagnosed, as well as key information such as the diagnostic error.

[0106] For each partition's unsuccessful fault diagnosis information, retrieve all data information involved in the corresponding data processing process. This information includes but is not limited to input data, intermediate processing results, model output, etc. Through in-depth analysis of this information, identify the reasons for the unsuccessful diagnosis, such as data anomalies, inaccurate model matching, etc.

[0107] Based on the results of unsuccessful information data analysis, the fault diagnosis model is trained and optimized. The optimization goals include feature extraction improvement, algorithm optimization, parameter adjustment, etc. Through continuous training and optimization, the model's fault diagnosis capability and accuracy are improved.

[0108] The optimized fault diagnosis model is stored in the partition fault diagnosis model knowledge base corresponding to the partition. In this way, each circuit partition has a fault diagnosis model library that has been continuously optimized and improved. In the subsequent fault diagnosis process, these optimized models can be directly used to diagnose the real-time operation data of the organic electroluminescent drive circuit, thereby improving the accuracy and efficiency of the diagnosis.

[0109] Finally, the optimized fault diagnosis model is used to perform fault diagnosis on the real-time operation data of the organic electroluminescent drive circuit of each partition. This step improves the ability of any circuit fault to be discovered and diagnosed in a timely manner in practical applications, providing a strong guarantee for the stable operation of the circuit.

[0110] By implementing step S105, the present invention realizes closed-loop optimization of the organic electroluminescent drive circuit fault diagnosis method. From the detection of fault diagnosis results to the analysis of unsuccessful information, and then to the training and optimization of the model, a continuous improvement cycle is formed. This process not only improves the accuracy of fault diagnosis, but also improves the timeliness and practicality of the diagnosis method, providing strong support for the further development of organic electroluminescent technology.

[0111] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S101 includes:

[0112] Acquire basic data of an organic electroluminescent drive circuit, the basic data of the organic electroluminescent drive circuit including circuit design data, component layout data, and circuit connection relationship data;

[0113] The basic data of the organic electroluminescent driving circuit is converted into a visual circuit image using a preset image generation model;

[0114] On the generated visualized circuit image, a coordinate system is established, where the coordinate system uses a three-dimensional coordinate system including an X coordinate, a Y coordinate, and a Z coordinate;

[0115] receiving a function of the organic electroluminescent driving circuit, and dividing the basic image into a plurality of regions according to the function of the organic electroluminescent driving circuit, each region corresponding to a combination of components in the circuit;

[0116] Assign a coordinate range to each divided area, which will be used for subsequent data analysis and fault diagnosis;

[0117] The results of the area division and coordinate configuration are recorded to form a complete record of the basic image configuration coordinates of the organic electroluminescent drive circuit.

[0118] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S101 includes:

[0119] accurately aligning the base image of the organic electroluminescent driving circuit with a pre-established coordinate system to identify the structure in the organic electroluminescent driving circuit;

[0120] According to the functions of the organic electroluminescent driving circuit, the visualized circuit image is divided into different functional areas including input area, output area, control area and power supply area;

[0121] Determine the boundaries of each identified functional area and assign a coordinate range to each partition, the coordinate range is used to reflect the location and area of ​​the partition in the circuit;

[0122] Record the information of each partition including the name of each partition, the function of each partition and the boundary coordinates of each partition.

[0123] A partition and coordinate configuration table is formed.

[0124] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S102 includes:

[0125] Determine the screening parameters, including the proportion of random sampling, the number of label combinations in each group, and whether repeated sampling is allowed;

[0126] According to the random sampling algorithm, labels are extracted from the partition label set to form multiple groups of random partition label combinations;

[0127] Check the randomly drawn partition label combinations. If they do not meet the sampling quality requirements, including uneven sample distribution and many repeated sampling, adjust the screening conditions and parameter settings, and re-execute the random data screening steps.

[0128] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S102 includes:

[0129] Access the established fault analysis model knowledge base, which includes multiple fault diagnosis models and their corresponding feature descriptions or matching rules;

[0130] The data feature extraction results of each partition are matched one by one with the models in the fault analysis model knowledge base, and the matching degree between the data features and the models is quantified using the similarity calculation method;

[0131] According to the similarity calculation results, the fault diagnosis model with the highest matching degree is selected as the model corresponding to the partition;

[0132] The fault diagnosis models matched for each partition are summarized to form a complete set of fault diagnosis models.

[0133] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S103 includes:

[0134] Filter out fault information from the historical data of each organic electroluminescent drive circuit partition, the fault information including fault time, fault type and fault description;

[0135] Pre-process the screened fault information to identify the characteristics of cluster analysis, which include fault frequency, fault duration, and fault impact range;

[0136] According to the characteristics of fault information and clustering requirements, set the parameters of the clustering algorithm, which include the number of clusters, distance measurement, and number of iterations;

[0137] The preprocessed fault feature data is input into the selected clustering algorithm, clustering analysis is performed, and preliminary clustering results are obtained.

[0138] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S103 includes:

[0139] Match the clustering features of each partition with the simulation tasks in the fault simulation library one by one, and select the best matching simulation fault task for each cluster of each partition according to the matching degree;

[0140] Summarize the simulated fault tasks matched to each partition to form a simulated fault task set for the partition;

[0141] Classify or sort the simulated fault task set according to fault type, impact degree, occurrence frequency, etc.;

[0142] The simulated fault task set of each partition is output for fault simulation and fault diagnosis strategy formulation.

[0143] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S104 includes:

[0144] Collect the data generated by the simulated fault test from each organic electroluminescent drive circuit partition, pre-process the data,

[0145] Categorize data by partition and failure type;

[0146] According to the different partitions and fault types, a corresponding real-time fault diagnosis model is selected from a fault diagnosis model knowledge base, and the selected fault diagnosis model is loaded into a data analysis environment;

[0147] Input the preprocessed simulated fault test data into the corresponding fault diagnosis model, run the fault diagnosis model, and process and analyze the input data;

[0148] The fault diagnosis result of each partition is obtained from the fault diagnosis model, and the fault diagnosis result of each partition includes the fault type, location and severity.

[0149] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S105 includes:

[0150] Filter out unsuccessful information from the fault diagnosis results of each partition, and record detailed information of the unsuccessful information, where the detailed information of the unsuccessful information includes the fault type, the diagnosis error, and the timestamp;

[0151] Collect data information involved in the unsuccessful information data processing process, including input data, intermediate processing results and model output;

[0152] Identify the collected data information, identify the data anomalies that lead to unsuccessful diagnosis, and obtain the identification results;

[0153] Based on the recognition results, it is necessary to determine the optimization objectives of the fault diagnosis model, which include feature extraction, algorithm improvement, and parameter adjustment.

[0154] Specifically, the organic electroluminescent drive circuit fault diagnosis optimization method provided by the present invention, step S105 includes:

[0155] Selecting an optimized fault diagnosis model from a partition fault diagnosis model knowledge base corresponding to the partition, and loading the selected optimized fault diagnosis model into the fault diagnosis system;

[0156] Input the preprocessed real-time operation data into the optimized fault diagnosis model, run the optimized fault diagnosis model, and process the input real-time data;

[0157] Fault diagnosis results are obtained from the optimized fault diagnosis model, and the fault diagnosis results include fault type, fault location, fault severity and recommended solutions.

[0158] The present invention provides an organic electroluminescent drive circuit fault diagnosis optimization method. Through a series of innovative steps and technical means, the present invention effectively solves the problem that the prior art can only diagnose known faults and cannot analyze the organic electroluminescent drive circuit fault diagnosis model in real time. Specifically, the present invention adopts the following measures:

[0159] Obtain the basic data (such as design data, component layout data, circuit connection relationship data) and real-time operation data of the organic electroluminescent drive circuit. Use the preset image generation model to convert the basic data into a visual circuit image, configure a coordinate system on the image, divide the circuit into regions, and obtain the organic electroluminescent drive circuit partition.

[0160] The organic electroluminescent drive circuit is partitioned and labeled, and multiple groups of random partition label combinations are obtained through random data screening. The historical data of each partition is collected, and data features are extracted and matched in the preset fault analysis model knowledge base to obtain the fault diagnosis model set corresponding to each partition and establish the partition fault diagnosis model knowledge base.

[0161] Fault information is extracted from the historical data of each partition, and cluster analysis is performed to obtain the historical fault cluster analysis results. These results are matched in the preset fault simulation library to generate a set of simulated fault tasks for each partition.

[0162] The simulated fault tasks are sent to the corresponding circuit partitions in sequence to perform simulated fault tests and collect the test results. Model matching is performed through the corresponding fault diagnosis model knowledge base, and the test results are analyzed using the real-time fault diagnosis model group to obtain the fault diagnosis results of each partition.

[0163] The fault diagnosis results of each partition are tested to distinguish between successful and unsuccessful information. For unsuccessful information, the data information involved in the relevant data processing process is retrieved to train and optimize the fault diagnosis model. The optimized fault diagnosis model is stored back in the fault diagnosis model knowledge base corresponding to the partition, and these optimized models are used to perform fault diagnosis on real-time operation data.

[0164] The present invention can timely discover and handle faults by real-time collection of operating data and fault diagnosis, thereby improving the reliability and stability of the system. The fault diagnosis model is continuously optimized according to the fault diagnosis results, so that it can adapt to the fault diagnosis needs in different situations and improve the accuracy and efficiency of diagnosis. By constructing a partitioned fault diagnosis model knowledge base, targeted fault diagnosis can be performed on different areas of the organic electroluminescent drive circuit, covering a wider range of fault types and scenarios.

[0165] In summary, the present invention effectively solves the limitations of organic electroluminescent drive circuit fault diagnosis in the prior art through a series of innovative technical means, and improves the real-time, accuracy and comprehensiveness of fault diagnosis.

[0166] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if the modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include the modifications and variations, and the above-described embodiments of the present invention do not constitute a limitation on the scope of protection of the present invention.

Claims

1. A method for fault diagnosis and optimization of an organic electroluminescent drive circuit, characterized in that: include: Step S101, acquiring basic data of an organic electroluminescent drive circuit, collecting real-time operation data of the organic electroluminescent drive circuit, substituting the basic data of the organic electroluminescent drive circuit into a preset image generation model, obtaining a basic image of the organic electroluminescent drive circuit, configuring coordinates in the basic image of the organic electroluminescent drive circuit, dividing the mechanical energy area of ​​the organic electroluminescent drive circuit according to the configuration coordinates of the basic image of the organic electroluminescent drive circuit, and obtaining a partition of the organic electroluminescent drive circuit; Step S102, establishing labels for the organic electroluminescent drive circuit partitions to obtain a partition label set, performing random data screening on the partition label set to obtain multiple groups of random partition label combinations, collecting historical data of each organic electroluminescent drive circuit partition, performing feature extraction on the historical data of each organic electroluminescent drive circuit partition to obtain partition data feature extraction results, matching the partition data feature extraction results in a preset fault analysis model knowledge base to obtain a fault diagnosis model set corresponding to the organic electroluminescent drive circuit partition, and establishing the fault diagnosis model set corresponding to the organic electroluminescent drive circuit partition into a partition fault diagnosis model knowledge base; Step S103, extracting fault information from historical data of each organic electroluminescent drive circuit partition to obtain historical fault information of each organic electroluminescent drive circuit partition, performing cluster analysis on the historical fault information of each organic electroluminescent drive circuit partition to obtain a historical fault cluster analysis result of each organic electroluminescent drive circuit partition, matching the historical fault cluster analysis result of each organic electroluminescent drive circuit partition in a preset fault simulation library to obtain a simulated fault task set for each organic electroluminescent drive circuit partition; Step S104, sending the simulated fault tasks in the simulated fault task set of each organic electroluminescent drive circuit partition to the corresponding organic electroluminescent drive circuit partition in sequence, obtaining the simulated fault test results of the organic electroluminescent drive circuit partition, performing fault diagnosis model matching on the simulated fault test results of the organic electroluminescent drive circuit partition through the corresponding fault diagnosis model knowledge base, obtaining a real-time fault diagnosis model group for each partition, using each partition real-time fault diagnosis model group to perform data analysis on the simulated fault test results of the organic electroluminescent drive circuit partition, and obtaining a fault diagnosis result for each partition; Step S105, detect the fault diagnosis result of each partition, obtain the fault diagnosis success information of each partition and the fault diagnosis unsuccessful information of each partition, retrieve the data information involved in the data processing process corresponding to the unsuccessful fault diagnosis information of each partition, train and optimize the fault diagnosis model according to the data information involved in the data processing process corresponding to the unsuccessful fault diagnosis information of each partition, and obtain the optimized fault diagnosis model, store the optimized fault diagnosis model in the partition fault diagnosis model knowledge base corresponding to the partition, and use the optimized fault diagnosis model stored in the partition fault diagnosis model knowledge base corresponding to the partition to perform fault diagnosis on the real-time operation data of the organic electroluminescent drive circuit of each partition.

2. The organic electroluminescent drive circuit fault diagnosis optimization method according to claim 1, characterized in that: The step S101 includes: Acquire basic data of an organic electroluminescent drive circuit, the basic data of the organic electroluminescent drive circuit including circuit design data, component layout data, and circuit connection relationship data; The basic data of the organic electroluminescent driving circuit is converted into a visual circuit image using a preset image generation model; On the generated visualized circuit image, a coordinate system is established, and the coordinate system adopts a three-dimensional coordinate system including an X coordinate, a Y coordinate, and a Z coordinate; receiving a function of the organic electroluminescent driving circuit, and dividing the basic image into a plurality of regions according to the function of the organic electroluminescent driving circuit, each region corresponding to a combination of components in the circuit; Assign a coordinate range to each divided area, which will be used for subsequent data analysis and fault diagnosis; The results of the area division and coordinate configuration are recorded to form a complete record of the basic image configuration coordinates of the organic electroluminescent drive circuit.

3. The organic electroluminescent drive circuit fault diagnosis optimization method according to claim 2, characterized in that: The step S101 includes: accurately aligning the base image of the organic electroluminescent driving circuit with a pre-established coordinate system to identify the structure in the organic electroluminescent driving circuit; According to the functions of the organic electroluminescent driving circuit, the visualized circuit image is divided into different functional areas including input area, output area, control area and power supply area; Determine the boundaries of each identified functional area and assign a coordinate range to each partition, the coordinate range is used to reflect the location and area of ​​the partition in the circuit; The information of each partition is recorded, including the name of each partition, the function of each partition, and the boundary coordinates of each partition. A partition and coordinate configuration table is formed.

4. The organic electroluminescent drive circuit fault diagnosis optimization method according to claim 1, characterized in that: The step S102 includes: Determine the screening parameters, including the proportion of random sampling, the number of label combinations in each group, and whether repeated sampling is allowed; According to the random sampling algorithm, labels are extracted from the partition label set to form multiple groups of random partition label combinations; Check the randomly drawn partition label combinations. If they do not meet the sampling quality requirements, including uneven sample distribution and many repeated sampling, adjust the screening conditions and parameter settings, and re-execute the random data screening steps.

5. The organic electroluminescent drive circuit fault diagnosis optimization method according to claim 4, characterized in that: The step S102 includes: Access the established fault analysis model knowledge base, which includes multiple fault diagnosis models and their corresponding feature descriptions or matching rules; The data feature extraction results of each partition are matched one by one with the models in the fault analysis model knowledge base, and the matching degree between the data features and the models is quantified using the similarity calculation method; According to the similarity calculation results, the fault diagnosis model with the highest matching degree is selected as the model corresponding to the partition; The fault diagnosis models matched for each partition are summarized to form a complete set of fault diagnosis models.

6. The organic electroluminescent drive circuit fault diagnosis and optimization method according to claim 1, characterized in that: The step S103 includes: Filter out fault information from the historical data of each organic electroluminescent drive circuit partition, the fault information including fault time, fault type and fault description; Pre-process the screened fault information to identify the characteristics of cluster analysis, which include fault frequency, fault duration, and fault impact range; According to the characteristics of fault information and clustering requirements, set the parameters of the clustering algorithm, including the number of clusters, distance measurement, and number of iterations; The preprocessed fault feature data is input into the selected clustering algorithm, clustering analysis is performed, and preliminary clustering results are obtained.

7. The organic electroluminescent drive circuit fault diagnosis and optimization method according to claim 6, characterized in that: The step S103 includes: Match the clustering features of each partition with the simulation tasks in the fault simulation library one by one, and select the best matching simulation fault task for each cluster of each partition according to the matching degree; Summarize the simulated fault tasks matched to each partition to form a simulated fault task set for the partition; Classify or sort the simulated fault task set according to fault type, impact degree, occurrence frequency, etc.; The simulated fault task set of each partition is output for fault simulation and fault diagnosis strategy formulation.

8. The organic electroluminescent drive circuit fault diagnosis and optimization method according to claim 1, characterized in that: The step S104 includes: Collect the data generated by the simulated fault test from each organic electroluminescent drive circuit partition, pre-process the data, Categorize data by partition and failure type; According to the different partitions and fault types, a corresponding real-time fault diagnosis model is selected from a fault diagnosis model knowledge base, and the selected fault diagnosis model is loaded into a data analysis environment; Input the preprocessed simulated fault test data into the corresponding fault diagnosis model, run the fault diagnosis model, and process and analyze the input data; The fault diagnosis result of each partition is obtained from the fault diagnosis model, and the fault diagnosis result of each partition includes the fault type, location and severity.

9. The organic electroluminescent drive circuit fault diagnosis and optimization method according to claim 1, characterized in that: The step S105 includes: Filter out unsuccessful information from the fault diagnosis results of each partition, and record detailed information of the unsuccessful information, where the detailed information of the unsuccessful information includes the fault type, the diagnosis error, and the timestamp; Collect data information involved in the unsuccessful information data processing process, including input data, intermediate processing results and model output; Identify the collected data information, identify the data anomalies that lead to unsuccessful diagnosis, and obtain the identification results; Based on the recognition results, it is necessary to determine the optimization objectives of the fault diagnosis model, which include feature extraction, algorithm improvement, and parameter adjustment.

10. The organic electroluminescent drive circuit fault diagnosis and optimization method according to claim 9, characterized in that: The step S105 includes: Selecting an optimized fault diagnosis model from a partition fault diagnosis model knowledge base corresponding to the partition, and loading the selected optimized fault diagnosis model into the fault diagnosis system; Input the preprocessed real-time operation data into the optimized fault diagnosis model, run the optimized fault diagnosis model, and process the input real-time data; Fault diagnosis results are obtained from the optimized fault diagnosis model, and the fault diagnosis results include fault type, fault location, fault severity and recommended solutions.

Images