An organic electroluminescent device operation detection method and system

Abstract

The present invention relates to the field of organic electroluminescent technology. The present invention provides an operation detection method and system for an organic electroluminescent device, which aims to solve the problem that the prior art cannot accurately detect the operating performance of the device under different states. By acquiring device configuration information and detection tasks, the device operating status is detected in real time, historical data is collected and a deep learning model is trained, and an operation model group corresponding to the device status is established. A time period detection strategy is adopted, the best model is matched to process data, the detection task is output, and the scheduling instructions are optimized. If the error of the detection result exceeds the range, the scheduling model is optimized and the fault is warned. The present invention improves the detection accuracy, enhances the fault warning capability, prolongs the service life of the device, and improves the user experience, providing effective support for the operation and maintenance of organic electroluminescent devices.

Description

Technical Field

[0001] The present invention relates to the technical field of organic electroluminescence, and in particular to an operation detection method and system of an organic electroluminescence device. Background Art

[0002] The operation detection method of organic electroluminescent devices is a technology based on the electroluminescence phenomenon, that is, when the organic electroluminescent device is stimulated by electric current, the organic electroluminescent device will produce electronic excitation and then emit photons. This process can be detected and analyzed by specific testing equipment. In practical applications, it usually involves using a DC power supply to apply current to the organic electroluminescent device, and using a light intensity measurement device to capture and measure the light energy released. The key performance parameters of the organic electroluminescent device, such as luminous intensity, uniformity, response time, and color accuracy, can be evaluated, so as to fully understand its operating status and performance. This method has the advantages of non-destructive, high sensitivity and high resolution, and is an indispensable and important means in the product development, quality control and performance evaluation of organic electroluminescent devices.

[0003] In practical applications, the operation detection method of organic electroluminescent devices is widely used in many fields. First, in the field of electronic displays, organic electroluminescent devices are a new generation of display technology, and their performance is directly related to the image quality of the display device. Therefore, through electroluminescence detection, organic electroluminescent device displays can provide high-quality image display effects in various environments.

[0004] The existing operation detection method of organic electroluminescent devices has the following technical pain points: the realization of multi-color display requires precise control of the coating and luminescent properties of different luminescent materials, which puts higher requirements on the detection technology. The existing operation detection method of organic electroluminescent devices cannot accurately detect the operating performance of organic electroluminescent devices under different states, resulting in the inability to accurately detect faults in organic electroluminescent devices. For this reason, the present invention proposes an operation detection method and system for organic electroluminescent devices. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an operation detection method and system for an organic electroluminescent device, which solves the problem that the existing operation detection method for an organic electroluminescent device cannot accurately detect the operating performance of the organic electroluminescent device under different states, resulting in an inability to accurately detect faults in the organic electroluminescent device.

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

[0007] In a first aspect, the present invention provides an operation detection method of an organic electroluminescent device, comprising:

[0008] Step S101, obtaining organic electroluminescent device configuration information and device detection tasks, wherein the organic electroluminescent device configuration information includes basic physical parameters, structural parameters, material properties and working environment parameters, and the device detection tasks include detection targets, detection parameter indicators, detection conditions and expected detection results;

[0009] Step S102, detecting the operating state of the organic electroluminescent device to obtain real-time operating state data of the organic electroluminescent device, wherein the real-time operating state data of the organic electroluminescent device includes luminous brightness data, luminous efficiency data, current density data, voltage state data, and luminous chromaticity data;

[0010] Step S103, collecting operation detection history data of the organic electroluminescent device, performing cluster analysis on the operation detection history data of the organic electroluminescent device to obtain the historical data cluster analysis result, using the operation detection history data of the organic electroluminescent device and the historical data cluster analysis result to train the deep learning model to obtain the operation model group of the organic electroluminescent device;

[0011] Step S104, establishing a detection period for the device detection task, the detection period includes a first detection period, a second detection period, a third detection period and a fourth detection period, performing state detection on the organic electroluminescent device in the first detection period, the second detection period, the third detection period and the fourth detection period respectively, and obtaining the device operation state in the detection period, the device operation state in the detection period includes the device operation state in the first detection period, the device operation state in the second detection period, the device operation state in the third detection period and the device operation state in the fourth detection period;

[0012] Step S105, matching each detection period in the device operation state of the detection period with the detection task generation model in the operation model group of the organic electroluminescent device, obtaining the detection task generation model corresponding to each detection period, using the detection task generation model corresponding to each detection period to process the data of each detection period, obtaining the organic electroluminescent device detection task of each period, sending the organic electroluminescent device detection task of each period to the corresponding detection equipment, receiving the data after the detection equipment executes the organic electroluminescent device in each period, substituting the data after the detection equipment executes the organic electroluminescent device in each period into the organic electroluminescent device operation evaluation model, and obtaining the organic electroluminescent device evaluation result.

[0013] Furthermore, the operation detection method of the organic electroluminescent device provided by the present invention, the step S103 comprises:

[0014] The operation detection history data of the organic electroluminescent device includes the operation detection data of the organic electroluminescent device and the organic electroluminescent device state corresponding to the operation detection data of the organic electroluminescent device. The operation detection data of the organic electroluminescent device includes the operation detection task of the organic electroluminescent device, the operation detection process data of the organic electroluminescent device and the operation detection result data of the organic electroluminescent device.

[0015] Preprocessing the collected operation detection history data of the organic electroluminescent device, extracting features from the preprocessed data, wherein the extracted features include the operation status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data;

[0016] Use clustering algorithms to perform cluster analysis on the preprocessed historical data of operation detection, and divide the historical data into different categories based on the similarities and differences of the data;

[0017] For each clustering result, a corresponding deep learning model is constructed, and the deep learning model is used to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the clustering;

[0018] The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimizing algorithms, the model can accurately predict the operating status of organic electroluminescent devices.

[0019] According to the clustering analysis results and the prediction results of the deep learning model, a correspondence between the operating model in the operating model group of the organic electroluminescent device and the state of the organic electroluminescent device is established, and the deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster.

[0020] Furthermore, the operation detection method of the organic electroluminescent device provided by the present invention, the step S104 comprises:

[0021] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods, namely, a first detection period, a second detection period, a third detection period and a fourth detection period;

[0022] Determine the start time and end time of each detection period, the detection period is divided to cover the entire operation cycle of the device, and the operation data of the organic electroluminescent device can be obtained in each period;

[0023] According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. The detection equipment measures the voltage, current, brightness and color temperature of the organic electroluminescent device during operation;

[0024] The detection device performs status detection in the first detection period, starts the detection device and starts to collect the operation data of the organic electroluminescent device, and after the data collection is completed, processes and analyzes the data of the first detection period to obtain the operation status of the device in the first detection period;

[0025] Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively, and obtain corresponding operation status data;

[0026] The device operating status data obtained in the four detection time periods are sorted to form an operating status data set, and the device operating status is extracted from the operating status data set, and the results of the device operating status of the first detection time period, the device operating status of the second detection time period, the device operating status of the third detection time period, and the device operating status of the fourth detection time period are output respectively.

[0027] Furthermore, the operation detection method of the organic electroluminescent device provided by the present invention, the step S105 comprises:

[0028] According to the device operation state data of the first detection period, the second detection period, the third detection period and the fourth detection period, performing detection task generation model matching in the operation model group of the organic electroluminescent device;

[0029] The basis for matching is the similarity between the running status data and the training data in the model. The distance measurement method is used for matching. The matching result is that each detection period corresponds to a detection task generation model;

[0030] Preprocess the data of each detection period, and input the preprocessed data into the detection task generation model corresponding to each detection period;

[0031] The detection task generation model corresponding to each detection period is calculated based on the input data, and outputs the detection task of the organic electroluminescent device within the period. The detection task includes device performance evaluation, fault prediction and parameter adjustment.

[0032] Furthermore, the operation detection method of the organic electroluminescent device provided by the present invention, the step S105 comprises:

[0033] Collecting the test result data after executing the test task of the organic electroluminescent device in each time period, obtaining the test result data of each time period, comparing the test result data of each time period with the basic physical parameters and structural parameters in the configuration information of the organic electroluminescent device, and obtaining the test result error comparison result; if the data error in the test result error comparison result exceeds the preset range, optimizing the preset organic electroluminescent device scheduling model with the test result error comparison result and the test result data of each time period through a genetic algorithm, and obtaining the optimized organic electroluminescent device scheduling model;

[0034] Receive user demand instructions, use the organic electroluminescent device scheduling model to process the user demand instructions, output the optimized real-time organic electroluminescent device operation scheduling instructions, and detect the operation of the organic electroluminescent device after executing the real-time organic electroluminescent device operation scheduling instructions. If the detection result meets the preset error value, stop optimizing the organic electroluminescent device scheduling model. If the detection result does not meet the preset error value, use the preset organic electroluminescent device fault detection model to generate a fault detection command to perform fault detection on the organic electroluminescent device. If there is a faulty organic electroluminescent device in the fault detection result, generate fault warning information.

[0035] In a second aspect, the present invention provides an organic electroluminescent device operation detection system, which uses the organic electroluminescent device operation detection method as described above, comprising:

[0036] A data acquisition unit, which acquires the configuration information of the organic electroluminescent device and the device detection task. The configuration information of the organic electroluminescent device includes basic physical parameters, structural parameters, material characteristics and working environment parameters. The device detection task includes detection targets, detection parameter indicators, detection conditions and expected detection results.

[0037] A detection unit detects the operating state of the organic electroluminescent device and obtains real-time operating state data of the organic electroluminescent device, wherein the real-time operating state data of the organic electroluminescent device includes luminous brightness data, luminous efficiency data, current density data, voltage state data and luminous chromaticity data;

[0038] A data reorganization unit collects historical operation detection data of the organic electroluminescent device, performs cluster analysis on the historical operation detection data of the organic electroluminescent device, obtains cluster analysis results of the historical data, and uses the historical operation detection data of the organic electroluminescent device and the cluster analysis results of the historical data to train a deep learning model to obtain an operation model group of the organic electroluminescent device;

[0039] A data matching unit is used to establish a detection period for a device detection task, wherein the detection period includes a first detection period, a second detection period, a third detection period, and a fourth detection period, and to perform state detection on the organic electroluminescent device in the first detection period, the second detection period, the third detection period, and the fourth detection period, respectively, to obtain a device operation state in the detection period, wherein the device operation state in the detection period includes a device operation state in the first detection period, a device operation state in the second detection period, a device operation state in the third detection period, and a device operation state in the fourth detection period;

[0040] The data processing unit matches each detection period in the device operation state of the detection period with the detection task generation model in the operation model group of the organic electroluminescent device to obtain the detection task generation model corresponding to each detection period, uses the detection task generation model corresponding to each detection period to process the data of each detection period to obtain the organic electroluminescent device detection task of each period, sends the organic electroluminescent device detection task of each period to the corresponding detection equipment, receives the data after the detection equipment executes the organic electroluminescent device in each period, substitutes the data after the detection equipment executes the organic electroluminescent device in each period into the organic electroluminescent device operation evaluation model to obtain the organic electroluminescent device evaluation result.

[0041] Furthermore, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data reorganization unit comprises:

[0042] The operation detection history data of the organic electroluminescent device includes the operation detection data of the organic electroluminescent device and the organic electroluminescent device state corresponding to the operation detection data of the organic electroluminescent device. The operation detection data of the organic electroluminescent device includes the operation detection task of the organic electroluminescent device, the operation detection process data of the organic electroluminescent device and the operation detection result data of the organic electroluminescent device.

[0043] Preprocessing the collected operation detection history data of the organic electroluminescent device, extracting features from the preprocessed data, wherein the extracted features include the operation status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data;

[0044] Use clustering algorithms to perform cluster analysis on the preprocessed historical data of operation detection, and divide the historical data into different categories based on the similarities and differences of the data;

[0045] For each clustering result, a corresponding deep learning model is constructed, and the deep learning model is used to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the clustering;

[0046] The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimizing algorithms, the model can accurately predict the operating status of organic electroluminescent devices.

[0047] According to the clustering analysis results and the prediction results of the deep learning model, a correspondence between the operating model in the operating model group of the organic electroluminescent device and the state of the organic electroluminescent device is established, and the deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster.

[0048] Furthermore, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data matching unit is further used for:

[0049] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods, namely, a first detection period, a second detection period, a third detection period and a fourth detection period;

[0050] Determine the start time and end time of each detection period, the detection period is divided to cover the entire operation cycle of the device, and the operation data of the organic electroluminescent device can be obtained in each period;

[0051] According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. The detection equipment measures the voltage, current, brightness and color temperature of the organic electroluminescent device during operation;

[0052] The detection device performs status detection in the first detection period, starts the detection device and starts to collect the operation data of the organic electroluminescent device, and after the data collection is completed, processes and analyzes the data of the first detection period to obtain the operation status of the device in the first detection period;

[0053] Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively, and obtain corresponding operation status data;

[0054] The device operating status data obtained in the four detection time periods are sorted to form an operating status data set, and the device operating status is extracted from the operating status data set, and the results of the device operating status of the first detection time period, the device operating status of the second detection time period, the device operating status of the third detection time period, and the device operating status of the fourth detection time period are output respectively.

[0055] Furthermore, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data processing unit is further used for:

[0056] According to the device operation state data of the first detection period, the second detection period, the third detection period and the fourth detection period, performing detection task generation model matching in the operation model group of the organic electroluminescent device;

[0057] The basis for matching is the similarity between the running status data and the training data in the model. The distance measurement method is used for matching. The matching result is that each detection period corresponds to a detection task generation model;

[0058] Preprocess the data of each detection period, and input the preprocessed data into the detection task generation model corresponding to each detection period;

[0059] The detection task generation model corresponding to each detection period is calculated based on the input data, and outputs the detection task of the organic electroluminescent device within the period. The detection task includes device performance evaluation, fault prediction and parameter adjustment.

[0060] Furthermore, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data processing unit is further used for:

[0061] Collecting the test result data after executing the test task of the organic electroluminescent device in each time period, obtaining the test result data of each time period, comparing the test result data of each time period with the basic physical parameters and structural parameters in the configuration information of the organic electroluminescent device, and obtaining the test result error comparison result; if the data error in the test result error comparison result exceeds the preset range, optimizing the preset organic electroluminescent device scheduling model with the test result error comparison result and the test result data of each time period through a genetic algorithm, and obtaining the optimized organic electroluminescent device scheduling model;

[0062] Receive user demand instructions, use the organic electroluminescent device scheduling model to process the user demand instructions, output the optimized real-time organic electroluminescent device operation scheduling instructions, and detect the operation of the organic electroluminescent device after executing the real-time organic electroluminescent device operation scheduling instructions. If the detection result meets the preset error value, stop optimizing the organic electroluminescent device scheduling model. If the detection result does not meet the preset error value, use the preset organic electroluminescent device fault detection model to generate a fault detection command to perform fault detection on the organic electroluminescent device. If there is a faulty organic electroluminescent device in the fault detection result, generate fault warning information.

[0063] Beneficial effects of the present invention:

[0064] The present invention collects the operation detection history data of the organic electroluminescent device and uses the deep learning model for training. The present invention can construct an operation model group corresponding to the different states of the organic electroluminescent device, so that in real-time detection, the model that matches the current state can be matched more accurately, thereby improving the accuracy of the detection result. The time-division detection strategy further refines the detection process, and can capture the changes in the operation performance of the organic electroluminescent device in different time periods, avoiding the errors and one-sidedness caused by single time point detection.

[0065] Based on the detection results, the present invention adds error comparison and model optimization steps. When the error of the detection result exceeds the preset range, the scheduling model will be optimized to improve the accuracy of the output instruction, which is helpful to discover the potential faults of the organic electroluminescent device in advance and avoid the expansion of the fault. Through the preset organic electroluminescent device fault detection model, the present invention can immediately generate early warning information when a fault is detected, provide users with timely maintenance prompts, and prevent the fault from causing further damage to the organic electroluminescent device.

[0066] The generation and execution of real-time scheduling instructions enables organic electroluminescent devices to be dynamically adjusted according to user needs or current operating status, thereby providing a more personalized display effect and improving user experience. Accurate detection and optimized scheduling help to extend the service life of organic electroluminescent devices and reduce replacement and repair costs due to failures, which is of great significance to the popularization and application of organic electroluminescent devices. The operation detection method and system provided by the present invention provide new ideas and means for performance evaluation and fault diagnosis of organic electroluminescent devices, which will help promote the further development of organic electroluminescent technology and improve its application level and competitiveness in the display field.

[0067] In summary, the present invention provides strong support for the operation, maintenance and management of organic electroluminescent devices by improving detection accuracy, enhancing fault warning capabilities, improving user experience and device life, and promoting the development of organic electroluminescent technology, and has significant beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] 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.

[0069] Figure 1 A schematic diagram of an operation detection method for an organic electroluminescent device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0070] 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.

[0071] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0072] In a first aspect, the present invention provides an operation detection method of an organic electroluminescent device, comprising:

[0073] Step S101, obtaining organic electroluminescent device configuration information and device detection tasks, wherein the organic electroluminescent device configuration information includes basic physical parameters, structural parameters, material properties and working environment parameters, and the device detection tasks include detection targets, detection parameter indicators, detection conditions and expected detection results;

[0074] In step S101, the method of the present invention first focuses on obtaining detailed configuration information of the organic electroluminescent device and a specific device detection task. This step is the basis for subsequent detection and analysis, and improves the pertinence and effectiveness of the detection process.

[0075] The organic electroluminescent device configuration information includes the following:

[0076] Basic physical parameters: Describe the basic physical characteristics of the device, such as size, weight, electrode structure, etc. Physical parameters are crucial to understanding the basic performance and limitations of the device.

[0077] Structural parameters: Structural parameters involve the layer structure of the device, including the thickness of the organic layer and electrode layer, material selection, and interlayer structure.

[0078] Material properties: Material properties refer to the physical and chemical properties of the organic materials and electrode materials that constitute the device, such as band structure, mobility, stability, etc. These properties directly affect the luminous efficiency and life of the device.

[0079] Working environment parameters: Working environment parameters describe the environmental conditions in which the device is working, such as temperature, humidity, light intensity, etc.

[0080] Test objective: Clarify the purpose of the test, such as evaluating the overall performance of the device, detecting performance changes under specific conditions, or fault diagnosis.

[0081] Detection parameter indicators: According to the detection target, determine the specific parameters that need to be monitored and measured, such as luminous brightness, luminous efficiency, current density, voltage stability, luminous chromaticity, etc.

[0082] Test conditions: Set the environmental conditions during testing, such as temperature, humidity, applied voltage or current, etc., to simulate the actual working conditions of the device.

[0083] Expected test results: Based on the understanding of the device and the test objectives, a reasonable test result range or expected performance indicators are preset for subsequent comparative analysis with the actual test results.

[0084] By fully acquiring these configuration information and detection tasks, step S101 provides a solid foundation for subsequent detection steps, improves the pertinence and accuracy of the detection process, and helps to more effectively evaluate the performance and stability of the device.

[0085] Step S102, detecting the operating state of the organic electroluminescent device to obtain real-time operating state data of the organic electroluminescent device, wherein the real-time operating state data of the organic electroluminescent device includes luminous brightness data, luminous efficiency data, current density data, voltage state data, and luminous chromaticity data;

[0086] In step S102, the organic electroluminescent device operation detection method of the present invention focuses on real-time detection of the device's operating state to obtain a series of key operating state data. The following is a detailed analysis of this step:

[0087] Step S102: Detect the operating state of the organic electroluminescent device. The goal of this step is to collect key performance parameters of the organic electroluminescent device in real-time operation. These data reflect the current operating state of the device. Specifically, the data to be obtained include:

[0088] Luminous brightness is an important indicator to measure the luminous intensity of organic electroluminescent devices, usually measured in nits. By measuring luminous brightness, we can evaluate the brightness performance of the device under different conditions and the stability of the brightness over time.

[0089] Luminous efficiency reflects the ability of organic electroluminescent devices to convert electrical energy into light energy, usually expressed in lumens / watt (lm / W) or candela / ampere (cd / A). High luminous efficiency means that the device can produce more light while consuming the same amount of electrical energy, which is of great significance for reducing energy consumption and extending battery life.

[0090] Current density refers to the current passing through a device per unit area, usually in milliamperes per square centimeter (mA / cm²). Current density is closely related to the brightness and efficiency of the device. By monitoring current density, we can understand the device's workload and potential overheating risks.

[0091] The voltage status data reflects the voltage changes of the device during operation. A stable voltage status is essential to ensure the normal operation of the device and extend its service life. By monitoring the voltage status, voltage fluctuations or anomalies can be discovered in time to avoid damage to the device due to unstable voltage.

[0092] Luminescence chromaticity refers to the color and purity of the light emitted by an organic electroluminescent device, usually expressed in color temperature (K) and color coordinates (x, y). By measuring luminescence chromaticity, the color rendering ability of the device, as well as the stability and consistency of the color over time, can be evaluated.

[0093] In order to obtain these real-time data, it is necessary to use special test equipment and sensors to monitor the organic electroluminescent device in real time, and transmit the data to the data processing unit for analysis and processing. Through the real-time detection of step S102, the operating status of the organic electroluminescent device can be fully understood, providing an important basis for subsequent performance evaluation, fault diagnosis and optimization adjustment.

[0094] Step S103, collecting operation detection history data of the organic electroluminescent device, performing cluster analysis on the operation detection history data of the organic electroluminescent device to obtain the historical data cluster analysis result, using the operation detection history data of the organic electroluminescent device and the historical data cluster analysis result to train the deep learning model to obtain the operation model group of the organic electroluminescent device;

[0095] In step S103, the present invention proposes an innovative method to improve the accuracy of the operation detection of the organic electroluminescent device, specifically by collecting historical detection data, performing cluster analysis, and using these data to train a deep learning model, and finally obtaining an operation model group corresponding to the state of the organic electroluminescent device. The following is a detailed analysis of this step:

[0096] The operation detection history data of the organic electroluminescent device is collected, and the data content includes the operation detection data of the organic electroluminescent device and its corresponding status information.

[0097] Operation test data: covers the device's operation test tasks, real-time data during the test process (such as voltage, current, brightness, etc.), and the final test result data.

[0098] Device status information: refers to the specific status of the device during detection, such as normal operation, performance degradation, failure, etc.

[0099] The purpose of clustering historical data is to group similar data points into one category in order to better understand and analyze the internal structure and rules of the data. Clustering algorithms (such as K-means, hierarchical clustering, etc.) are used to perform cluster analysis on preprocessed historical data. Preprocessing includes steps such as data cleaning, missing value processing, and normalization to improve the accuracy and consistency of the data. Based on the similarities and differences of the data, the historical data is divided into different categories. Each category represents one or more similar organic electroluminescent device states.

[0100] A corresponding deep learning model is constructed for each clustering result. These models are designed to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the cluster. The deep learning model is trained using the clustered historical data. By continuously adjusting the model parameters and optimizing the algorithm, the model can accurately predict the operating state of the organic electroluminescent device.

[0101] The corresponding relationship between the running model group and the device state is established. According to the cluster analysis results and the prediction results of the deep learning model, the corresponding relationship between the models in the running model group of the organic electroluminescent device and the device state is established. Each deep learning model is marked as corresponding to the organic electroluminescent device state under a specific cluster. A complete running model group is formed by multiple deep learning models corresponding to different device states.

[0102] Through step S103, the present invention uses historical detection data to construct an operation model group closely related to the state of the organic electroluminescent device, which can more accurately reflect the operating performance of the device in different states, and provide strong support for subsequent state detection and fault diagnosis. At the same time, because the model is trained based on a large amount of historical data, it has better generalization ability and robustness, and can adapt to detection needs under different environments and conditions.

[0103] Step S104, establishing a detection period for the device detection task, the detection period includes a first detection period, a second detection period, a third detection period and a fourth detection period, performing state detection on the organic electroluminescent device in the first detection period, the second detection period, the third detection period and the fourth detection period respectively, and obtaining the device operation state in the detection period, the device operation state in the detection period includes the device operation state in the first detection period, the device operation state in the second detection period, the device operation state in the third detection period and the device operation state in the fourth detection period;

[0104] In step S104, the present invention proposes a method for performing time-division state detection on an organic electroluminescent device, aiming to more comprehensively and accurately capture the operating performance of the device in different time periods. The following is a detailed analysis of this step:

[0105] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into multiple detection periods to more carefully observe the changes in the device state. In this embodiment, the entire monitoring cycle is divided into four detection periods: a first detection period, a second detection period, a third detection period, and a fourth detection period.

[0106] Determine the start and end time coverage period of the detection period, and improve the detection period divided to cover the entire operation cycle of the organic electroluminescent device, so as to obtain comprehensive status data. The length of each detection period can be adjusted according to actual needs, but it should be improved to obtain sufficient operation data in each period for status analysis.

[0107] According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. These devices are used to measure the key indicators of organic electroluminescent devices during operation, such as voltage, current, brightness, color temperature, etc. Before the detection, all detection equipment is calibrated to improve the accuracy and reliability of the measurement results.

[0108] During each test period, the test equipment is started and the operating data of the organic electroluminescent device is collected. These data will be used for subsequent status analysis and evaluation. After the data collection is completed, the raw data is preprocessed and analyzed to extract the key indicators and features reflecting the operating status of the device.

[0109] Based on the processed data, the operating status of the organic electroluminescent device in each detection period is extracted, and this status information will be used for subsequent model matching and performance evaluation. The device operating status data obtained in the four detection periods are sorted to form a complete operating status data set. This data set contains the device state change information during the entire monitoring period.

[0110] Through step S104, the present invention realizes the state detection of the organic electroluminescent device in different time periods. This method can more carefully capture the operating performance changes of the device in different time periods, and provide richer data support for subsequent state evaluation and fault diagnosis. At the same time, by reasonably configuring the detection equipment and parameters, the accuracy and reliability of the detection results are improved.

[0111] Step S105, matching each detection period in the device operation state of the detection period with the detection task generation model in the operation model group of the organic electroluminescent device, obtaining the detection task generation model corresponding to each detection period, using the detection task generation model corresponding to each detection period to process the data of each detection period, obtaining the organic electroluminescent device detection task of each period, sending the organic electroluminescent device detection task of each period to the corresponding detection equipment, receiving the data after the detection equipment executes the organic electroluminescent device in each period, substituting the data after the detection equipment executes the organic electroluminescent device in each period into the organic electroluminescent device operation evaluation model, and obtaining the organic electroluminescent device evaluation result.

[0112] In step S105, the present invention proposes a data processing method based on time-segment state matching and deep learning model, which is used to comprehensively and accurately evaluate the operating state of the organic electroluminescent device. The following is a detailed analysis of this step:

[0113] According to the organic electroluminescent device operating state data of the four detection periods (the first, second, third, and fourth detection periods) obtained in step S104, matching is performed in the pre-constructed organic electroluminescent device operating model groups. These model groups are trained based on historical detection data and deep learning algorithms and can reflect the operating behaviors and characteristics of organic electroluminescent devices under different states. The basis for matching is the similarity between the operating state data and the training data in the model. By using distance measurement, similarity calculation and other methods, an optimal matching model is selected for each detection period.

[0114] After selecting the matching model for each time period, the data of the time period is preprocessed to meet the requirements of the model input, which includes data cleaning, normalization, feature extraction and other operations. The preprocessed data is input into the detection task generation model of the corresponding time period. The model predicts and calculates based on the input data and outputs the detection tasks of the organic electroluminescent device in the time period. These tasks include device performance evaluation, fault prediction and parameter adjustment suggestions.

[0115] The organic electroluminescent device detection tasks for each period are sent to the corresponding detection equipment, which performs corresponding detection operations according to the task requirements, such as measuring key indicators such as voltage, current, brightness, color temperature, etc. After the detection equipment completes the detection task, it sends the collected data back to the data processing unit. These data reflect the actual operating status and performance of the organic electroluminescent device during the period.

[0116] Substitute the received test equipment execution data into the organic electroluminescent device operation evaluation model, which is another model built based on deep learning or traditional algorithms, used to comprehensively analyze the data and give the final device status evaluation results. After the evaluation model processes the data, it outputs the evaluation results of the organic electroluminescent device, which includes key information such as the device's performance score, failure probability, and remaining service life, providing decision support for subsequent device maintenance, optimization, or replacement.

[0117] Through step S105, the present invention realizes an efficient and accurate method for evaluating the state of an organic electroluminescent device. The method can comprehensively capture the operating performance changes of the device in different time periods through steps such as time period detection, model matching and data processing, and give scientific evaluation results based on a deep learning algorithm, which is of great significance for improving the reliability of organic electroluminescent devices, extending their service life and reducing maintenance costs.

[0118] Specifically, the operation detection method of the organic electroluminescent device provided by the present invention, the step S103 comprises:

[0119] The operation detection history data of the organic electroluminescent device includes the operation detection data of the organic electroluminescent device and the organic electroluminescent device state corresponding to the operation detection data of the organic electroluminescent device. The operation detection data of the organic electroluminescent device includes the operation detection task of the organic electroluminescent device, the operation detection process data of the organic electroluminescent device and the operation detection result data of the organic electroluminescent device.

[0120] Preprocessing the collected operation detection history data of the organic electroluminescent device, extracting features from the preprocessed data, wherein the extracted features include the operation status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data;

[0121] Use clustering algorithms to perform cluster analysis on the preprocessed historical data of operation detection, and divide the historical data into different categories based on the similarities and differences of the data;

[0122] For each clustering result, a corresponding deep learning model is constructed, and the deep learning model is used to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the clustering;

[0123] The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimizing algorithms, the model can accurately predict the operating status of organic electroluminescent devices.

[0124] According to the clustering analysis results and the prediction results of the deep learning model, a correspondence between the operating model in the operating model group of the organic electroluminescent device and the state of the organic electroluminescent device is established, and the deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster.

[0125] In step S103, the operation detection method of the organic electroluminescent device provided by the present invention specifically includes the following detailed steps, aiming to construct a deep learning model group that can accurately predict the operation status of the organic electroluminescent device by collecting and analyzing historical detection data:

[0126] Collect the operation and detection history data of organic electroluminescent devices, which includes the operation and detection tasks of the devices, the real-time data during the detection process (such as voltage, current, brightness, etc.), and the final detection result data. Preprocess the collected historical data to remove noise, fill in missing values, and perform data normalization to improve the accuracy and consistency of the data.

[0127] Operating status: Status indicators such as normal operation, performance degradation, and failure.

[0128] Run the detection task parameters: such as detection target, detection conditions, etc.

[0129] Operation process data change trend: such as the change trend of parameters such as voltage, current, brightness, etc. over time.

[0130] Operation result indicators: such as luminous efficiency, color accuracy and other performance indicators.

[0131] Use clustering algorithms (such as K-means, hierarchical clustering, etc.) to perform cluster analysis on the preprocessed historical data. The purpose of clustering is to group similar data points into one category in order to better understand and analyze the inherent structure and laws of the data.

[0132] According to the similarity and difference of the data, the historical data are divided into different categories, each of which represents one or more similar organic electroluminescent device states.

[0133] For each clustering result, a corresponding deep learning model is constructed. These models are designed to learn and simulate the operating behavior and state changes of the organic electroluminescent devices under the clustering.

[0134] The selection and design of the model should be based on the characteristics of the data and the analysis requirements, including convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), etc.

[0135] The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimization algorithms (such as gradient descent, Adam optimizer, etc.), the model can accurately predict the operating status of organic electroluminescent devices.

[0136] During the training process, the performance of the model can be evaluated through methods such as cross-validation, and the model can be adjusted and optimized based on the evaluation results.

[0137] According to the cluster analysis results and the prediction results of the deep learning model, the correspondence between the models in the organic electroluminescent device operation model group and the device status is established.

[0138] The deep learning model corresponding to each cluster is marked to correspond to the state of the organic electroluminescent device under the cluster, so that in practical applications, a suitable model can be selected for prediction and analysis according to the current state of the device.

[0139] Through the above steps, the present invention successfully constructs an operation model group closely related to the state of the organic electroluminescent device in step S103. These models can learn and simulate the operation behavior and state changes of the device in different states based on historical data, providing strong support for subsequent state detection and fault diagnosis.

[0140] Specifically, the operation detection method of the organic electroluminescent device provided by the present invention, the step S104 comprises:

[0141] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods, namely, a first detection period, a second detection period, a third detection period and a fourth detection period;

[0142] Determine the start time and end time of each detection period, the detection period is divided to cover the entire operation cycle of the device, and the operation data of the organic electroluminescent device can be obtained in each period;

[0143] According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. The detection equipment measures the voltage, current, brightness and color temperature of the organic electroluminescent device during operation;

[0144] The detection device performs status detection in the first detection period, starts the detection device and starts to collect the operation data of the organic electroluminescent device, and after the data collection is completed, processes and analyzes the data of the first detection period to obtain the operation status of the device in the first detection period;

[0145] Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively, and obtain corresponding operation status data;

[0146] The device operating status data obtained in the four detection time periods are sorted to form an operating status data set, and the device operating status is extracted from the operating status data set, and the results of the device operating status of the first detection time period, the device operating status of the second detection time period, the device operating status of the third detection time period, and the device operating status of the fourth detection time period are output respectively.

[0147] In step S104, the present invention proposes a method for performing time-divided state detection on an organic electroluminescent device. By subdividing the monitoring cycle, the operating performance of the organic electroluminescent device in different time periods is more comprehensively captured, thereby improving the accuracy and reliability of the detection. The following is a detailed analysis of this step:

[0148] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is reasonably divided into four detection periods: the first detection period, the second detection period, the third detection period and the fourth detection period. This division helps to more carefully observe the changes in the device status over time.

[0149] Determine the start and end time of each detection period, and improve the detection period after division to cover the entire operation cycle of the organic electroluminescent device. At the same time, sufficient operation data should be obtained in each period to support subsequent status analysis and evaluation.

[0150] According to the specific requirements of the inspection task, the corresponding inspection equipment and parameters are configured. These equipment are used to measure the key indicators of organic electroluminescent devices during operation, such as voltage, current, brightness and color temperature. All inspection equipment is calibrated before inspection to improve the accuracy and reliability of the measurement results.

[0151] In each test period, start the configured test equipment and start collecting the operation data of the organic electroluminescent device. During the data collection process, the continuity and integrity of the data should be improved to avoid data loss or abnormality.

[0152] After the data collection is completed, the collected data is processed and analyzed. By extracting key indicators and features, the operating status of the organic electroluminescent device during the period is evaluated.

[0153] Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively. After each period ends, corresponding operation status data is obtained.

[0154] The device operation status data obtained in the four detection periods are sorted to form a complete operation status data set, which contains the state change information of the organic electroluminescent device in the entire monitoring period.

[0155] The device operating status is extracted from the sorted operating status data set. By analyzing and comparing the data of different time periods, the results of the device operating status of the first detection period, the second detection period, the third detection period, and the fourth detection period are output respectively. These status results will provide important basis for subsequent model matching, performance evaluation, and fault diagnosis.

[0156] Through the above steps, the present invention realizes the time-division state detection of the organic electroluminescent device in step S104. This method not only improves the comprehensiveness and accuracy of the detection, but also helps to more deeply understand the change law of device performance over time, and provides strong support for the research and development, production and use of organic electroluminescent devices.

[0157] Specifically, the method for detecting the operation of an organic electroluminescent device provided by the present invention, step S105, comprises:

[0158] According to the device operation state data of the first detection period, the second detection period, the third detection period and the fourth detection period, performing detection task generation model matching in the operation model group of the organic electroluminescent device;

[0159] The basis for matching is the similarity between the running status data and the training data in the model. The distance measurement method is used for matching. The matching result is that each detection period corresponds to a detection task generation model;

[0160] Preprocess the data of each detection period, and input the preprocessed data into the detection task generation model corresponding to each detection period;

[0161] The detection task generation model corresponding to each detection period is calculated based on the input data, and outputs the detection task of the organic electroluminescent device within the period. The detection task includes device performance evaluation, fault prediction and parameter adjustment.

[0162] Each detection period in the device operation state of the detection period is matched with the detection task generation model in the operation model group of the organic electroluminescent device to obtain the detection task generation model corresponding to each detection period. The detection task generation model corresponding to each detection period is used to process the data of each detection period to obtain the organic electroluminescent device detection task of each period, send the task to the corresponding detection device, receive and process the data returned by the detection device, and finally obtain the evaluation result of the organic electroluminescent device.

[0163] The matching detection task generation model is matched in the operation model group of the organic electroluminescent device according to the device operation status data of the first detection period, the second detection period, the third detection period and the fourth detection period, to find the detection task generation model that is most similar to the data of each detection period.

[0164] Data preprocessing: preprocess the data of each detection period to meet the input requirements of the detection task generation model. Preprocessing may include data cleaning, normalization, feature extraction and other operations.

[0165] Generate a test task, input the preprocessed data into the matched test task generation model, and the model calculates and outputs the test task of the organic electroluminescent device within the period based on the input data. The test task may include device performance evaluation, fault prediction, and parameter adjustment.

[0166] Execute the detection task and send the organic electroluminescent device detection task of each time period to the corresponding detection equipment. The detection equipment performs corresponding detection operations according to the task requirements, such as measuring key indicators such as voltage, current, brightness, and color temperature.

[0167] Receive and process data, receive data returned by the detection equipment after performing the detection task, and substitute these data into the organic electroluminescent device operation evaluation model. The evaluation model calculates and outputs the evaluation results of the organic electroluminescent device based on the input data.

[0168] If the detection result has a large error with the expectation, the preset organic electroluminescent device scheduling model is optimized to improve the accuracy of the output instruction. If there is still a situation that does not meet the preset error value after optimization, the preset organic electroluminescent device fault detection model is used to perform fault detection, and a warning message is generated when a fault is detected.

[0169] Receive user demand instructions, use the optimized organic electroluminescent device scheduling model to process the user demand instructions, and output the optimized real-time organic electroluminescent device operation scheduling instructions. After executing the scheduling instructions, the operating status of the organic electroluminescent device is detected to verify the scheduling effect. Through time-division detection, model matching, data processing, optimization and fault warning steps, the operating status of the organic electroluminescent device is continuously monitored and evaluated to ensure the stable operation and performance optimization of the device.

[0170] The time-segment detection strategy improves the comprehensiveness and accuracy of detection and can capture the changes in the device's operating performance in different time periods. The introduction of deep learning models improves the accuracy and reliability of detection results, providing strong support for device fault diagnosis and performance optimization. The generation and execution of real-time scheduling instructions enables devices to dynamically adjust according to user needs or current status, improving user experience and device life.

[0171] The specific steps to build a detection task generation model are as follows:

[0172] The collected operation detection history data of the organic electroluminescent device includes the operation detection task of the device, the operation detection process data and the operation detection result data.

[0173] The collected historical data is preprocessed, including data cleaning, missing value processing, normalization and other steps to improve the accuracy and consistency of the data. Features are extracted from the preprocessed data, including the operating status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data.

[0174] Use clustering algorithms (such as K-means, hierarchical clustering, etc.) to perform cluster analysis on the preprocessed historical data, and divide the historical data into different categories based on the similarities and differences of the data. For each clustering result, build a corresponding deep learning model. These models are designed to learn and simulate the operating behavior and state changes of organic electroluminescent devices under the cluster.

[0175] The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimization algorithms (such as gradient descent, Adam optimizer, etc.), the model can accurately predict the operating status of organic electroluminescent devices.

[0176] According to the cluster analysis results and the prediction results of the deep learning model, the corresponding relationship between the model and the device state in the operation model group of the organic electroluminescent device is established. The deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster. The trained model is verified to ensure its accuracy and robustness in practical applications. According to the verification results, the model is adjusted and optimized as necessary. A detection task generation model that can accurately predict the operation state of the organic electroluminescent device is constructed.

[0177] Specifically, the method for detecting the operation of an organic electroluminescent device provided by the present invention, step S105, comprises:

[0178] Collecting the test result data after executing the test task of the organic electroluminescent device in each time period, obtaining the test result data of each time period, comparing the test result data of each time period with the basic physical parameters and structural parameters in the configuration information of the organic electroluminescent device, and obtaining the test result error comparison result; if the data error in the test result error comparison result exceeds the preset range, optimizing the preset organic electroluminescent device scheduling model with the test result error comparison result and the test result data of each time period through a genetic algorithm, and obtaining the optimized organic electroluminescent device scheduling model;

[0179] Receive user demand instructions, use the organic electroluminescent device scheduling model to process the user demand instructions, output the optimized real-time organic electroluminescent device operation scheduling instructions, and detect the operation of the organic electroluminescent device after executing the real-time organic electroluminescent device operation scheduling instructions. If the detection result meets the preset error value, stop optimizing the organic electroluminescent device scheduling model. If the detection result does not meet the preset error value, use the preset organic electroluminescent device fault detection model to generate a fault detection command to perform fault detection on the organic electroluminescent device. If there is a faulty organic electroluminescent device in the fault detection result, generate fault warning information.

[0180] In step S105, the organic electroluminescent device operation detection method of the present invention not only involves model matching and data processing, but also further includes the optimization of the detection results and the fault warning mechanism. This step is intended to improve the accuracy of the detection results and respond to user needs in a timely manner. At the same time, it can issue a warning when a potential fault is detected to ensure the stable operation of the organic electroluminescent device. The following is a detailed analysis of this step:

[0181] The test result data after executing the organic electroluminescent device test task in each time period is collected to obtain the test result data of each time period. The test result data of each time period is compared with the basic physical parameters (such as size, weight, electrode structure, etc.) and structural parameters (such as organic layer thickness, electrode layer material, etc.) in the organic electroluminescent device configuration information. Through the comparison, the test result error comparison result is obtained. This step helps to evaluate the accuracy of the test results and identify the source of the error.

[0182] If the data error in the detection result error comparison result exceeds the preset range, it is considered that the current detection result has a large uncertainty or deviation. In this case, the detection result error comparison result and the detection result data of each time period are input into the preset organic electroluminescent device scheduling model, and the model is optimized by optimization algorithms such as genetic algorithms. Genetic algorithm is an optimization algorithm that simulates natural selection and genetic mechanisms, and gradually approaches the optimal solution through an iterative search process. In this step, it is used to adjust the parameters of the scheduling model to reduce the error of the detection result and improve the prediction accuracy of the model.

[0183] Receive user demand instructions, such as adjusting the working mode, brightness, color temperature, etc. of the organic electroluminescent device.

[0184] Use the optimized organic electroluminescent device scheduling model to process user demand instructions and output optimized real-time organic electroluminescent device operation scheduling instructions. Execute the real-time scheduling instructions, adjust the operating status of the organic electroluminescent device, and monitor it in real time. Detect the operation of the organic electroluminescent device after executing the real-time scheduling instructions to obtain new operating status data. Compare the new operating status data with the preset error value to evaluate the scheduling effect.

[0185] If the detection result meets the preset error value, the scheduling is considered successful and further optimization of the organic electroluminescent device scheduling model is stopped.

[0186] If the test result still does not meet the preset error value, the above optimization process is repeated until the error requirement is met or the preset optimization number limit is reached.

[0187] During the detection process, if a fault detection command is generated using a preset organic electroluminescent device fault detection model to perform fault detection on the organic electroluminescent device, and there is a faulty organic electroluminescent device in the fault detection result, a fault warning message is immediately generated. The fault warning message contains information such as the fault type, fault location, and fault severity, which helps users or maintenance personnel to take timely countermeasures to prevent the fault from further deteriorating or causing more serious consequences.

[0188] Through the implementation of step S105, the present invention not only improves the accuracy and reliability of the operation detection of the organic electroluminescent device, but also improves the stable operation of the organic electroluminescent device and the timely response to user needs through the real-time scheduling and fault warning mechanism, which is of great significance for increasing the service life of the organic electroluminescent device, reducing maintenance costs and improving user experience.

[0189] In a second aspect, the present invention provides an organic electroluminescent device operation detection system, which uses the organic electroluminescent device operation detection method as described above, comprising:

[0190] A data acquisition unit, which acquires the configuration information of the organic electroluminescent device and the device detection task. The configuration information of the organic electroluminescent device includes basic physical parameters, structural parameters, material characteristics and working environment parameters. The device detection task includes detection targets, detection parameter indicators, detection conditions and expected detection results.

[0191] A detection unit detects the operating state of the organic electroluminescent device and obtains real-time operating state data of the organic electroluminescent device, wherein the real-time operating state data of the organic electroluminescent device includes luminous brightness data, luminous efficiency data, current density data, voltage state data and luminous chromaticity data;

[0192] A data reorganization unit collects historical operation detection data of the organic electroluminescent device, performs cluster analysis on the historical operation detection data of the organic electroluminescent device, obtains cluster analysis results of the historical data, and uses the historical operation detection data of the organic electroluminescent device and the cluster analysis results of the historical data to train a deep learning model to obtain an operation model group of the organic electroluminescent device;

[0193] A data matching unit is used to establish a detection period for a device detection task, wherein the detection period includes a first detection period, a second detection period, a third detection period, and a fourth detection period, and to perform state detection on the organic electroluminescent device in the first detection period, the second detection period, the third detection period, and the fourth detection period, respectively, to obtain a device operation state in the detection period, wherein the device operation state in the detection period includes a device operation state in the first detection period, a device operation state in the second detection period, a device operation state in the third detection period, and a device operation state in the fourth detection period;

[0194] The data processing unit matches each detection period in the device operation state of the detection period with the detection task generation model in the operation model group of the organic electroluminescent device to obtain the detection task generation model corresponding to each detection period, uses the detection task generation model corresponding to each detection period to process the data of each detection period to obtain the organic electroluminescent device detection task of each period, sends the organic electroluminescent device detection task of each period to the corresponding detection equipment, receives the data after the detection equipment executes the organic electroluminescent device in each period, substitutes the data after the detection equipment executes the organic electroluminescent device in each period into the organic electroluminescent device operation evaluation model to obtain the organic electroluminescent device evaluation result.

[0195] Specifically, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data reorganization unit comprises:

[0196] The operation detection history data of the organic electroluminescent device includes the operation detection data of the organic electroluminescent device and the organic electroluminescent device state corresponding to the operation detection data of the organic electroluminescent device. The operation detection data of the organic electroluminescent device includes the operation detection task of the organic electroluminescent device, the operation detection process data of the organic electroluminescent device and the operation detection result data of the organic electroluminescent device.

[0197] Preprocessing the collected operation detection history data of the organic electroluminescent device, extracting features from the preprocessed data, wherein the extracted features include the operation status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data;

[0198] Use clustering algorithms to perform cluster analysis on the preprocessed historical data of operation detection, and divide the historical data into different categories based on the similarities and differences of the data;

[0199] For each clustering result, a corresponding deep learning model is constructed, and the deep learning model is used to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the clustering;

[0200] The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimizing algorithms, the model can accurately predict the operating status of organic electroluminescent devices.

[0201] According to the clustering analysis results and the prediction results of the deep learning model, a correspondence between the operating model in the operating model group of the organic electroluminescent device and the state of the organic electroluminescent device is established, and the deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster.

[0202] Specifically, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data matching unit is further used for:

[0203] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods, namely, a first detection period, a second detection period, a third detection period and a fourth detection period;

[0204] Determine the start time and end time of each detection period, the detection period is divided to cover the entire operation cycle of the device, and the operation data of the organic electroluminescent device can be obtained in each period;

[0205] According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. The detection equipment measures the voltage, current, brightness and color temperature of the organic electroluminescent device during operation;

[0206] The detection device performs status detection in the first detection period, starts the detection device and starts to collect the operation data of the organic electroluminescent device, and after the data collection is completed, processes and analyzes the data of the first detection period to obtain the operation status of the device in the first detection period;

[0207] Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively, and obtain corresponding operation status data;

[0208] The device operating status data obtained in the four detection time periods are sorted to form an operating status data set, and the device operating status is extracted from the operating status data set, and the results of the device operating status of the first detection time period, the device operating status of the second detection time period, the device operating status of the third detection time period, and the device operating status of the fourth detection time period are output respectively.

[0209] Specifically, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data processing unit is further used for:

[0210] According to the device operation state data of the first detection period, the second detection period, the third detection period and the fourth detection period, performing detection task generation model matching in the operation model group of the organic electroluminescent device;

[0211] The basis for matching is the similarity between the running status data and the training data in the model. The distance measurement method is used for matching. The matching result is that each detection period corresponds to a detection task generation model;

[0212] Preprocess the data of each detection period, and input the preprocessed data into the detection task generation model corresponding to each detection period;

[0213] The detection task generation model corresponding to each detection period is calculated based on the input data, and outputs the detection task of the organic electroluminescent device within the period. The detection task includes device performance evaluation, fault prediction and parameter adjustment.

[0214] Specifically, the present invention provides an operation detection system for an organic electroluminescent device, wherein the data processing unit is further used for:

[0215] Collecting the test result data after executing the test task of the organic electroluminescent device in each time period, obtaining the test result data of each time period, comparing the test result data of each time period with the basic physical parameters and structural parameters in the configuration information of the organic electroluminescent device, and obtaining the test result error comparison result; if the data error in the test result error comparison result exceeds the preset range, optimizing the preset organic electroluminescent device scheduling model with the test result error comparison result and the test result data of each time period through a genetic algorithm, and obtaining the optimized organic electroluminescent device scheduling model;

[0216] Receive user demand instructions, use the organic electroluminescent device scheduling model to process the user demand instructions, output the optimized real-time organic electroluminescent device operation scheduling instructions, and detect the operation of the organic electroluminescent device after executing the real-time organic electroluminescent device operation scheduling instructions. If the detection result meets the preset error value, stop optimizing the organic electroluminescent device scheduling model. If the detection result does not meet the preset error value, use the preset organic electroluminescent device fault detection model to generate a fault detection command to perform fault detection on the organic electroluminescent device. If there is a faulty organic electroluminescent device in the fault detection result, generate fault warning information.

[0217] The present invention provides an organic electroluminescent device operation detection method and system, aiming to solve the problem that the existing organic electroluminescent device operation detection method cannot accurately detect the operation performance of the device in different states, and thus cannot accurately detect the faults of the device. The specific technical solution is as follows:

[0218] Obtain the configuration information of the organic electroluminescent device, including basic physical parameters, structural parameters, material properties, and working environment parameters. At the same time, clarify the device detection tasks, including detection targets, detection parameter indicators, detection conditions, and expected detection results. Detect the real-time operating status of the organic electroluminescent device and obtain luminous brightness data, luminous efficiency data, current density data, voltage state data, and luminous chromaticity data. Collect the operation detection history data of the organic electroluminescent device, including operation detection data and corresponding status. Preprocess and feature extract the historical data, and then use the clustering algorithm to perform cluster analysis to obtain different data categories. For each clustering result, build a corresponding deep learning model, and use the clustered historical data to train the model so that the model can accurately predict the operating status of the organic electroluminescent device. Finally, establish an operation model group of the organic electroluminescent device, each model corresponding to a specific state.

[0219] According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods (the first detection period, the second detection period, the third detection period, and the fourth detection period). In each detection period, the detection equipment is started to collect data, and the device operation status in the period is obtained through processing and analysis.

[0220] According to the device operation status data obtained in each detection period, match it in the operation model group of the organic electroluminescent device to find the most suitable detection task generation model. Use the matched detection task generation model to process the data of each period, output the detection task (such as performance evaluation, fault prediction, parameter adjustment), and send the task to the corresponding detection equipment for execution. Receive and execute the data returned by the device, substitute it into the operation evaluation model, and obtain the final evaluation result. There is also an optimization of the detection result and a fault warning mechanism. If there is a large error between the detection result and the expectation, the scheduling model is optimized. If there is still a situation that does not meet the preset error value after optimization, fault detection is performed, and a warning message is generated when a fault is detected.

[0221] Through the above technical scheme, the present invention can more accurately detect the operating performance of organic electroluminescent devices under different states, timely discover and warn of potential faults, improve the accuracy and reliability of detection, and provide strong support for product development, quality control and performance evaluation of organic electroluminescent devices.

[0222] Obviously, those skilled in the art can 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 equivalent technologies, the present invention is also intended to include the modifications and variations. 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 detecting the operation of an organic electroluminescent device, characterized in that: include: Step S101, obtaining organic electroluminescent device configuration information and device detection tasks, wherein the organic electroluminescent device configuration information includes basic physical parameters, structural parameters, material properties and working environment parameters, and the device detection tasks include detection targets, detection parameter indicators, detection conditions and expected detection results; Step S102, detecting the operating state of the organic electroluminescent device to obtain real-time operating state data of the organic electroluminescent device, wherein the real-time operating state data of the organic electroluminescent device includes luminous brightness data, luminous efficiency data, current density data, voltage state data, and luminous chromaticity data; Step S103, collecting operation detection history data of the organic electroluminescent device, performing cluster analysis on the operation detection history data of the organic electroluminescent device to obtain the historical data cluster analysis result, using the operation detection history data of the organic electroluminescent device and the historical data cluster analysis result to train the deep learning model to obtain the operation model group of the organic electroluminescent device; Step S104, establishing a detection period for the device detection task, the detection period includes a first detection period, a second detection period, a third detection period and a fourth detection period, performing state detection on the organic electroluminescent device in the first detection period, the second detection period, the third detection period and the fourth detection period respectively, and obtaining the device operation state in the detection period, the device operation state in the detection period includes the device operation state in the first detection period, the device operation state in the second detection period, the device operation state in the third detection period and the device operation state in the fourth detection period; Step S105, matching each detection period in the device operation state of the detection period with the detection task generation model in the operation model group of the organic electroluminescent device, obtaining the detection task generation model corresponding to each detection period, using the detection task generation model corresponding to each detection period to process the data of each detection period, obtaining the organic electroluminescent device detection task of each period, sending the organic electroluminescent device detection task of each period to the corresponding detection equipment, receiving the data after the detection equipment executes the organic electroluminescent device in each period, substituting the data after the detection equipment executes the organic electroluminescent device in each period into the organic electroluminescent device operation evaluation model, and obtaining the organic electroluminescent device evaluation result.

2. The method for detecting the operation of an organic electroluminescent device according to claim 1, characterized in that: The step S103 includes: The operation detection history data of the organic electroluminescent device includes the operation detection data of the organic electroluminescent device and the organic electroluminescent device state corresponding to the operation detection data of the organic electroluminescent device. The operation detection data of the organic electroluminescent device includes the operation detection task of the organic electroluminescent device, the operation detection process data of the organic electroluminescent device and the operation detection result data of the organic electroluminescent device. Preprocessing the collected operation detection history data of the organic electroluminescent device, extracting features from the preprocessed data, wherein the extracted features include the operation status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data; Use clustering algorithms to perform cluster analysis on the preprocessed historical data of operation detection, and divide the historical data into different categories based on the similarities and differences of the data; For each clustering result, a corresponding deep learning model is constructed, and the deep learning model is used to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the clustering; The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimizing algorithms, the model can accurately predict the operating status of organic electroluminescent devices. According to the clustering analysis results and the prediction results of the deep learning model, a correspondence between the operating model in the operating model group of the organic electroluminescent device and the state of the organic electroluminescent device is established, and the deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster.

3. The method for detecting the operation of an organic electroluminescent device according to claim 1, characterized in that: The step S104 includes: According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods, namely, a first detection period, a second detection period, a third detection period and a fourth detection period; Determine the start time and end time of each detection period, the detection period is divided to cover the entire operation cycle of the device, and the operation data of the organic electroluminescent device can be obtained in each period; According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. The detection equipment measures the voltage, current, brightness and color temperature of the organic electroluminescent device during operation; The detection device performs status detection in the first detection period, starts the detection device and starts to collect the operation data of the organic electroluminescent device, and after the data collection is completed, processes and analyzes the data of the first detection period to obtain the operation status of the device in the first detection period; Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively, and obtain corresponding operation status data; The device operating status data obtained in the four detection time periods are sorted to form an operating status data set, and the device operating status is extracted from the operating status data set, and the results of the device operating status of the first detection time period, the device operating status of the second detection time period, the device operating status of the third detection time period, and the device operating status of the fourth detection time period are output respectively.

4. The method for detecting the operation of an organic electroluminescent device according to claim 1, characterized in that: The step S105 includes: According to the device operation state data of the first detection period, the second detection period, the third detection period and the fourth detection period, performing detection task generation model matching in the operation model group of the organic electroluminescent device; The basis for matching is the similarity between the running status data and the training data in the model. The distance measurement method is used for matching. The matching result is that each detection period corresponds to a detection task generation model; Preprocess the data of each detection period, and input the preprocessed data into the detection task generation model corresponding to each detection period; The detection task generation model corresponding to each detection period is calculated based on the input data, and outputs the detection task of the organic electroluminescent device within the period. The detection task includes device performance evaluation, fault prediction and parameter adjustment.

5. The method for detecting the operation of an organic electroluminescent device according to claim 4, characterized in that: The step S105 includes: Collecting the test result data after executing the test task of the organic electroluminescent device in each time period, obtaining the test result data of each time period, comparing the test result data of each time period with the basic physical parameters and structural parameters in the configuration information of the organic electroluminescent device, and obtaining the test result error comparison result; if the data error in the test result error comparison result exceeds the preset range, optimizing the preset organic electroluminescent device scheduling model with the test result error comparison result and the test result data of each time period through a genetic algorithm, and obtaining the optimized organic electroluminescent device scheduling model; Receive user demand instructions, use the organic electroluminescent device scheduling model to process the user demand instructions, output the optimized real-time organic electroluminescent device operation scheduling instructions, and detect the operation of the organic electroluminescent device after executing the real-time organic electroluminescent device operation scheduling instructions. If the detection result meets the preset error value, stop optimizing the organic electroluminescent device scheduling model. If the detection result does not meet the preset error value, use the preset organic electroluminescent device fault detection model to generate a fault detection command to perform fault detection on the organic electroluminescent device. If there is a faulty organic electroluminescent device in the fault detection result, generate fault warning information.

6. An organic electroluminescent device operation detection system, using the organic electroluminescent device operation detection method according to any one of claims 1 to 5, characterized in that: include: A data acquisition unit, used to acquire organic electroluminescent device configuration information and device detection tasks, wherein the organic electroluminescent device configuration information includes basic physical parameters, structural parameters, material characteristics and working environment parameters, and the device detection tasks include detection targets, detection parameter indicators, detection conditions and expected detection results; A detection unit, used to detect the operating state of the organic electroluminescent device, and obtain real-time operating state data of the organic electroluminescent device, wherein the real-time operating state data of the organic electroluminescent device includes luminous brightness data, luminous efficiency data, current density data, voltage state data, and luminous chromaticity data; A data reorganization unit is used to collect operation detection historical data of the organic electroluminescent device, perform cluster analysis on the operation detection historical data of the organic electroluminescent device, obtain the historical data cluster analysis results, and use the operation detection historical data of the organic electroluminescent device and the historical data cluster analysis results to train a deep learning model to obtain an operation model group of the organic electroluminescent device; A data matching unit is used to establish a detection period for a device detection task, the detection period includes a first detection period, a second detection period, a third detection period and a fourth detection period, and to perform state detection on the organic electroluminescent device in the first detection period, the second detection period, the third detection period and the fourth detection period, respectively, to obtain a device operation state in the detection period, the device operation state in the detection period includes a device operation state in the first detection period, a device operation state in the second detection period, a device operation state in the third detection period and a device operation state in the fourth detection period; A data processing unit is used to match each detection period in the device operation state of the detection period with a detection task generation model in the operation model group of the organic electroluminescent device, obtain the detection task generation model corresponding to each detection period, use the detection task generation model corresponding to each detection period to perform data processing on the data of each detection period, obtain the organic electroluminescent device detection task of each period, send the organic electroluminescent device detection task of each period to the corresponding detection equipment, receive the data after the detection equipment executes the organic electroluminescent device in each period, substitute the data after the detection equipment executes the organic electroluminescent device in each period into the organic electroluminescent device operation evaluation model, and obtain the organic electroluminescent device evaluation result.

7. The operation detection system of the organic electroluminescent device according to claim 6, characterized in that: The data reorganization unit comprises: The operation detection history data of the organic electroluminescent device includes the operation detection data of the organic electroluminescent device and the organic electroluminescent device state corresponding to the operation detection data of the organic electroluminescent device. The operation detection data of the organic electroluminescent device includes the operation detection task of the organic electroluminescent device, the operation detection process data of the organic electroluminescent device and the operation detection result data of the organic electroluminescent device. Preprocessing the collected operation detection history data of the organic electroluminescent device, extracting features from the preprocessed data, wherein the extracted features include the operation status of the organic electroluminescent device, parameters in the operation detection task, change trend data in the operation process data, and indicators in the operation result data; Use clustering algorithms to perform cluster analysis on the preprocessed historical data of operation detection, and divide the historical data into different categories based on the similarities and differences of the data; For each clustering result, a corresponding deep learning model is constructed, and the deep learning model is used to learn and simulate the operating behavior and state changes of the organic electroluminescent device under the clustering; The deep learning model is trained using clustered historical data. By continuously adjusting model parameters and optimizing algorithms, the model can accurately predict the operating status of organic electroluminescent devices. According to the clustering analysis results and the prediction results of the deep learning model, a correspondence between the operating model in the operating model group of the organic electroluminescent device and the state of the organic electroluminescent device is established, and the deep learning model corresponding to each cluster is marked as corresponding to the state of the organic electroluminescent device under the cluster.

8. The operation detection system of the organic electroluminescent device according to claim 6, characterized in that: The data matching unit is further used for: According to the actual operation and monitoring requirements of the organic electroluminescent device, the entire monitoring cycle is divided into four detection periods, namely, a first detection period, a second detection period, a third detection period and a fourth detection period; Determine the start time and end time of each detection period, the detection period is divided to cover the entire operation cycle of the device, and the operation data of the organic electroluminescent device can be obtained in each period; According to the requirements of the detection task, the corresponding detection equipment and parameters are configured. The detection equipment measures the voltage, current, brightness and color temperature of the organic electroluminescent device during operation; The detection device performs status detection in the first detection period, starts the detection device and starts to collect the operation data of the organic electroluminescent device, and after the data collection is completed, processes and analyzes the data of the first detection period to obtain the operation status of the device in the first detection period; Repeat the above steps to perform status detection on the organic electroluminescent device in the second detection period, the third detection period and the fourth detection period respectively, and obtain corresponding operation status data; The device operating status data obtained in the four detection time periods are sorted to form an operating status data set, and the device operating status is extracted from the operating status data set, and the results of the device operating status of the first detection time period, the device operating status of the second detection time period, the device operating status of the third detection time period, and the device operating status of the fourth detection time period are output respectively.

9. The operation detection system of the organic electroluminescent device according to claim 6, characterized in that: The data processing unit is further used for: According to the device operation state data of the first detection period, the second detection period, the third detection period and the fourth detection period, performing detection task generation model matching in the operation model group of the organic electroluminescent device; The basis for matching is the similarity between the running status data and the training data in the model. The distance measurement method is used for matching. The matching result is that each detection period corresponds to a detection task generation model; Preprocess the data of each detection period, and input the preprocessed data into the detection task generation model corresponding to each detection period; The detection task generation model corresponding to each detection period is calculated based on the input data, and outputs the detection task of the organic electroluminescent device within the period. The detection task includes device performance evaluation, fault prediction and parameter adjustment.

10. The operation detection system of the organic electroluminescent device according to claim 9, characterized in that: The data processing unit is further used for: Collecting the test result data after executing the test task of the organic electroluminescent device in each time period, obtaining the test result data of each time period, comparing the test result data of each time period with the basic physical parameters and structural parameters in the configuration information of the organic electroluminescent device, and obtaining the test result error comparison result; if the data error in the test result error comparison result exceeds the preset range, optimizing the preset organic electroluminescent device scheduling model with the test result error comparison result and the test result data of each time period through a genetic algorithm, and obtaining the optimized organic electroluminescent device scheduling model; Receive user demand instructions, use the organic electroluminescent device scheduling model to process the user demand instructions, output the optimized real-time organic electroluminescent device operation scheduling instructions, and detect the operation of the organic electroluminescent device after executing the real-time organic electroluminescent device operation scheduling instructions. If the detection result meets the preset error value, stop optimizing the organic electroluminescent device scheduling model. If the detection result does not meet the preset error value, use the preset organic electroluminescent device fault detection model to generate a fault detection command to perform fault detection on the organic electroluminescent device. If there is a faulty organic electroluminescent device in the fault detection result, generate fault warning information.

Images