Device and method for monitoring service life of organic electroluminescent material

Abstract

The invention provides a service life monitoring device and method of an organic electroluminescent material, and relates to the technical field of service life monitoring of luminescent materials. The system comprises a multispectral excitation module, a dynamic degradation module, a feature fusion network and a life prediction module. The multispectral excitation module is used for acquiring a multiband light-emitting image sequence of the organic electroluminescent material; the dynamic degradation module establishes a dynamic model of the material luminescence characteristic changing along with time, and quantifies the degradation rate; the feature fusion network extracts cross-band time sequence features; and the life prediction module outputs final life evaluation. By introducing multispectral excitation, efficient dynamic degradation analysis, feature fusion and accurate life prediction, the efficiency of material life monitoring is greatly improved. The real-time monitoring and advanced data analysis capabilities are realized, and the change of the material performance can be captured in time and early warning can be given out, so that the loss caused by material degradation is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring the lifespan of luminescent materials, and particularly to a device for monitoring the lifespan of organic electroluminescent materials and a method for monitoring the lifespan of organic electroluminescent materials. Background Art

[0002] Organic electroluminescent materials have received extensive attention in recent years due to their wide applications in displays and lighting devices. Such materials can emit light under the excitation of an electric current, and thus show great application potential in fields such as consumer electronics, automotive displays, and smart wearable devices. Compared with traditional liquid crystal display technologies, organic electroluminescent materials have higher contrast, wider viewing angles, and are thinner and lighter, which enables their rapid development in the market. The core of organic electroluminescent materials lies in their excellent optoelectronic properties. They are usually composed of organic molecules or polymers, and these materials can emit light efficiently when excited by an electric current. Organic electroluminescent materials can achieve a relatively high luminous efficiency and have lower energy consumption compared with traditional inorganic luminescent materials. Due to the characteristics of organic materials, organic electroluminescent materials can be fabricated into flexible screens and are suitable for devices of various shapes. In addition to displays, organic electroluminescent materials also show great application potential in high-end lighting, transparent displays, and other emerging markets.

[0003] Despite the significant advantages of organic electroluminescent materials, in the application process, the lifespan and stability of the materials have always been major challenges in the industry. Since organic electroluminescent materials are vulnerable to environmental factors, their luminous efficiency gradually decays over time. Existing lifespan monitoring methods mostly rely on off-line tests and cannot achieve real-time tracking of the material state. This results in a lack of timely warning when the material performance deteriorates, leading to potential losses. Moreover, it is impossible to deeply analyze the degradation mechanism of the material, and there is also a lack of an effective model to describe the change of luminous characteristics over time, making the prediction of the service life of the material inaccurate. Summary of the Invention

[0004] The present invention provides a device and a method for monitoring the lifespan of organic electroluminescent materials to solve the defects existing in the prior art.

[0005] On the one hand, the present invention provides a device for monitoring the lifespan of organic electroluminescent materials, including: A multi-spectral excitation module, configured to excite the organic electroluminescent materials through a tunable excitation light source and obtain a multi-band luminous image sequence of the organic electroluminescent materials.

[0006] A dynamic degradation module, configured to establish a dynamic model of the change of the material's luminous characteristics over time according to the multi-band luminous image sequence and quantify the degradation rate.

[0007] A feature fusion network is used to fuse multi-band luminescence image sequences and degradation rates, and extract cross-band temporal features.

[0008] A lifetime prediction module is used to generate a theoretical lifetime prediction value based on the degradation rate, obtain the degradation law of cross-band temporal features through a long short-term memory network, and combine the theoretical lifetime prediction value with the degradation law to correct the prediction deviation and output the final lifetime assessment.

[0009] According to a lifetime monitoring device for an organic electroluminescent material provided by the present invention, the multi-spectral excitation module includes a spectral switching unit, a time synchronization controller, and an ambient light suppression unit. The spectral switching unit is used to adjust the wavelength range of the excitation light source and capture the luminescence images of the organic electroluminescent material in different bands. The time synchronization controller is used to coordinate the timing synchronization of the excitation light source and the image acquisition device. The ambient light suppression unit is used to dynamically filter background stray light to obtain a multi-band luminescence image sequence.

[0010] According to a lifetime monitoring device for an organic electroluminescent material provided by the present invention, the process of dynamically filtering background stray light includes: A light sensor is set during the process of the tunable excitation light source exciting the organic electroluminescent material to monitor ambient light data in real time. The ambient light data includes intensity and wavelength characteristics.

[0011] Adjust the wavelength of the excitation light source, and use the time synchronization controller to coordinate the timing of the excitation light source and the image acquisition device.

[0012] Compare the collected image with the ambient light data, and use the ambient light intensity for pixel value subtraction to dynamically adjust the image.

[0013] Use a dynamic filtering algorithm to update the filtering parameters in real time according to the change of the ambient light data to obtain a multi-band luminescence image sequence.

[0014] According to a lifetime monitoring device for an organic electroluminescent material provided by the present invention, the dynamic degradation module includes a time series decomposition unit, a dynamic model construction unit, and a parameter optimization unit. The time series decomposition unit is used to decompose the multi-band luminescence image into a basic brightness component and a degradation component. The dynamic model construction unit is used to construct a non-linear degradation model based on the degradation component to obtain the degradation rate of the organic electroluminescent material. The parameter optimization unit is used to fit the non-linear degradation model parameters through an optimization algorithm to quantify the degradation difference in the local area.

[0015] According to a lifetime monitoring device for an organic electroluminescent material provided by the present invention, the process of obtaining the degradation rate of the organic electroluminescent material includes: Select a power-law function to construct an initial model for the degradation of the organic electroluminescent material.

[0016] The polynomial regression algorithm is selected to analyze the characteristics of the degradation components and identify the key factors affecting the degradation of organic electroluminescent materials. The key factors include light intensity, temperature change, and usage time.

[0017] According to the key factors, the least squares method is used to fit the initial model to obtain a non-linear degradation model.

[0018] The degradation parameters in the non-linear degradation model are extracted to obtain the degradation rate of the organic electroluminescent materials. The degradation parameters include the material attenuation constant and the characteristic time constant.

[0019] According to a life monitoring device for organic electroluminescent materials provided by the present invention, the feature fusion network includes a feature extraction unit, a feature fusion unit, and a feature selection unit. The feature extraction unit is used to extract the temporal features of the multi-band luminescence image sequence through a convolutional neural network. The feature fusion unit is used to fuse the temporal features and the degradation rate using a fusion algorithm to generate a feature set. The feature selection unit is used to screen the key features in the feature set through the principal component analysis method.

[0020] According to a life monitoring device for organic electroluminescent materials provided by the present invention, the life prediction module includes a theoretical prediction unit, a long short-term memory network unit, and a prediction correction unit. The theoretical prediction unit is used to construct a theoretical life model and generate a theoretical life prediction value of the organic electroluminescent materials in combination with the degradation rate. The long short-term memory network unit is used to construct a degradation prediction model based on the long short-term memory network, input the key features, and output the degradation law. The prediction correction unit is used to combine the theoretical life prediction value and the degradation law, and correct the prediction deviation through a regression analysis algorithm to obtain the final life evaluation.

[0021] According to a life monitoring device for organic electroluminescent materials provided by the present invention, the process of generating the theoretical life prediction value of the organic electroluminescent materials includes: Collect the initial performance data and historical degradation rates of the organic electroluminescent materials. The initial performance data includes the initial brightness and usage conditions.

[0022] According to the initial performance data and historical degradation rates, a theoretical life model is constructed using the degradation exponential decay formula of the organic electroluminescent materials.

[0023] Set the critical brightness value of the organic electroluminescent materials, and calculate the time required to reach the critical brightness value according to the degradation rate to obtain the theoretical life prediction value.

[0024] According to a life monitoring device for organic electroluminescent materials provided by the present invention, the process of constructing a degradation prediction model based on the long short-term memory network includes: Collect historical degradation data of the organic electroluminescent material, where the historical degradation data includes a historical multi-band luminescence image sequence and the degradation law of the corresponding historical organic electroluminescent material.

[0025] Construct a historical dynamic model based on the historical multi-band luminescence image sequence and obtain the historical degradation rate.

[0026] Extract the historical time-series features of the historical multi-band luminescence image sequence, generate a historical feature set in combination with the historical degradation rate, and screen the historical key features in the historical feature set.

[0027] Construct a long short-term memory network basic model, and set the number of network layers, the number of neurons in each layer, and the activation function of the basic model.

[0028] Use the historical key features as the input and the corresponding historical degradation law as the output to train the basic model, retain the model parameters that meet the test accuracy, and obtain a degradation prediction model.

[0029] On the other hand, the present invention also provides a method for monitoring the lifespan of an organic electroluminescent material, including: Use a tunable excitation light source to excite the organic electroluminescent material, and obtain a luminescence image sequence of the organic electroluminescent material at different bands after excitation.

[0030] Construct a dynamic model of the luminescence characteristics changing with time based on the multi-band luminescence image sequence, and quantify the degradation rate of the material through the dynamic model.

[0031] Fuse the multi-band luminescence image sequence and the degradation rate, and extract cross-band time-series features.

[0032] Generate a theoretical lifespan prediction value based on the degradation rate, analyze the extracted time-series features through a long short-term memory network to obtain the degradation law of the organic electroluminescent material, and combine the theoretical lifespan prediction value with the degradation law to obtain a final lifespan assessment.

[0033] Set a lifespan warning threshold for the organic electroluminescent material, divide the warning levels according to the final lifespan assessment, and trigger corresponding warning signals.

[0034] The life monitoring device and method for an organic electroluminescent material provided by the present invention can effectively excite the organic electroluminescent material through a tunable excitation light source, capture a sequence of luminescence images at different wavelengths, comprehensively obtain the luminescence characteristics of the material, provide rich spectral information, and provide high-quality data support for subsequent analysis. Through ambient light suppression and timing synchronization, it is possible to ensure that the influence of background noise is minimized during the imaging process, improving the usability and accuracy of the images. By deeply analyzing the sequence of multi-wavelength luminescence images, a dynamic model of the material is established in real time, thereby quantitatively showing the change of the material's luminescence characteristics over time. It can not only quantify the degradation rate, but also systematically analyze the stability of the material under different usage conditions and environments, providing early warnings for potential failure modes. This makes the evaluation of the material's service life more scientific and reasonable, and can effectively reduce the failure risk. By fusing multi-wavelength luminescence image features and degradation rates, cross-wavelength timing features are extracted, enhancing the depth and breadth of feature analysis. Combining convolutional neural networks and principal component analysis ensures the retention of key information and reduces unnecessary redundant data, providing a more comprehensive perspective for the performance evaluation of the material, and helping to further understand the comprehensive influencing factors of material degradation. By fusing theoretical prediction values with the degradation law obtained through a long short-term memory network, in-depth analysis and accurate prediction of the material's performance are achieved. By using regression analysis methods to correct prediction biases, the dynamic degradation characteristics of the material can be captured more accurately, enabling the final life evaluation to be obtained based on conditions close to actual use. At the same time, warning levels can be divided according to the evaluation results and corresponding warning signals can be triggered, helping relevant personnel to take timely measures to reduce R & D and production losses caused by material failures, and improving the safety and reliability of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0036] Figure 1 It is a schematic structural diagram of a life monitoring device for an organic electroluminescent material provided by an embodiment of the present invention; Figure 2 It is a schematic flowchart of generating a theoretical life prediction value of an organic electroluminescent material in an embodiment of the present invention; Figure 3 It is a schematic flowchart of a life monitoring method for an organic electroluminescent material provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts fall within the protection scope of the present invention.

[0038] The following will describe Figures 1-3 a life monitoring device and method for an organic electroluminescent material of the present invention.

[0039] Figure 1 is a schematic structural diagram of a life monitoring device for an organic electroluminescent material provided by an embodiment of the present invention.

[0040] As Figure 1 shown, for a life monitoring device and method for an organic electroluminescent material provided by an embodiment of the present invention, the execution subject can be a life monitoring device for an organic electroluminescent material. The system includes a multi-spectral excitation module, a dynamic degradation module, a feature fusion network, and a life prediction module.

[0041] The multi-spectral excitation module is used to excite the organic electroluminescent material through a tunable excitation light source to obtain a multi-band luminescence image sequence of the organic electroluminescent material.

[0042] The multi-spectral excitation module includes a spectral switching unit, a time synchronization controller, and an ambient light suppression unit. The spectral switching unit is used to adjust the wavelength range of the excitation light source to capture the luminescence images of the organic electroluminescent material in different bands. The time synchronization controller is used to coordinate the timing synchronization between the excitation light source and the image acquisition device. The ambient light suppression unit is used to dynamically filter background stray light to obtain a multi-band luminescence image sequence.

[0043] The function of the spectral switching unit is to adjust the wavelength range of the excitation light source so as to be able to capture the luminescence characteristics of the organic electroluminescent material in multiple bands. The specific implementation process is as follows: Select a tunable excitation light source, such as lasers or LED light sources with different wavelengths. The required excitation wavelength can be accurately set through a control system. According to the luminescence characteristics of the material and the required band to be measured, a series of excitation wavelengths are preset. When the wavelength of the excitation light source gradually changes, a relevant image acquisition device, such as a high-sensitivity camera, simultaneously captures images to record the spectral response of the material at each excitation wavelength.

[0044] The role of the time synchronization controller is to ensure the timing synchronization between the excitation light source and the image acquisition device to ensure the real-time response of the material captured at each wavelength. The implementation process is as follows: Implement the timing control between the excitation light source and the image acquisition device through hardware or software. For example, use a microcontroller or a field-programmable gate array to control the on and off times of the excitation light source.

[0045] When the light source switches to a specified wavelength, the time synchronization controller sends a trigger signal to the camera to ensure that the image is captured at an appropriate time point.

[0046] With each switch of the excitation wavelength, the time synchronization controller will quickly coordinate the camera to capture the luminescent image of the corresponding band, ensuring that each picture corresponds to accurate excitation conditions.

[0047] The process of dynamically filtering background stray light includes: Set up a light sensor during the process of the tunable excitation light source exciting the organic electroluminescent material to monitor the ambient light data in real time. The ambient light data includes intensity and wavelength characteristics.

[0048] Adjust the wavelength of the excitation light source and use the time synchronization controller to coordinate the timing of the excitation light source and the image acquisition device.

[0049] Compare the captured image with the ambient light data, perform pixel value subtraction using the ambient light intensity, and dynamically adjust the image.

[0050] Use a dynamic filtering algorithm to update the filtering parameters in real time according to the changes in the ambient light data to obtain a multi-band luminescent image sequence.

[0051] The dynamic degradation module is used to establish a dynamic model of the material's luminescence characteristics changing with time based on the multi-band luminescent image sequence and quantify the degradation rate.

[0052] The dynamic degradation module includes a timing decomposition unit, a dynamic model construction unit, and a parameter optimization unit. The timing decomposition unit is used to decompose the multi-band luminescent image into a basic brightness component and a degradation component. The dynamic model construction unit is used to construct a non-linear degradation model based on the degradation component to obtain the degradation rate of the organic electroluminescent material. The parameter optimization unit is used to fit the non-linear degradation model parameters through an optimization algorithm to quantify the degradation differences in local regions.

[0053] The timing decomposition unit is responsible for decomposing the multi-band luminescent image sequence into a basic brightness component and a degradation component for subsequent analysis. The specific implementation process is as follows: Preprocess the multi-band luminescent image sequence to ensure image quality, including denoising and normalization processing.

[0054] Adopt time-domain analysis or frequency-domain analysis techniques, such as wavelet transform, singular value decomposition, etc., to decompose each image into two parts: a basic brightness component and a degradation component. The basic brightness component represents the normal luminescence characteristics of the material under stable conditions, and the degradation component represents additional attenuation or loss characteristics.

[0055] By comparing the luminance data within a time series, analyze the changes in luminescence at different time points, and extract the base luminance and degradation components at each time point for constructing a dynamic model.

[0056] The process of obtaining the degradation rate of the organic electroluminescent material includes: Select a power-law function to construct an initial model for the degradation of the organic electroluminescent material.

[0057] Select a polynomial regression algorithm to analyze the characteristics of the degradation components and identify the key factors affecting the degradation of the organic electroluminescent material. The key factors include light intensity, temperature change, and usage time.

[0058] According to the key factors, use the least squares method to fit the initial model to obtain a non-linear degradation model.

[0059] Extract the degradation parameters in the non-linear degradation model to obtain the degradation rate of the organic electroluminescent material. The degradation parameters include the material attenuation constant and the characteristic time constant.

[0060] The parameter optimization unit is responsible for fitting the parameters of the non-linear degradation model through an optimization algorithm to quantify the degradation differences in local regions. The specific implementation process is as follows: Determine the optimization objective function, such as minimizing the mean square error between the predicted value and the actual observed value, to measure the goodness of fit of the model.

[0061] Adopt common optimization algorithms, such as the gradient descent method, genetic algorithm, or other techniques suitable for non-linear optimization, to find the optimal model parameters.

[0062] Input the observed degradation component data into the optimization algorithm, and iteratively adjust the model parameters until the error reaches an acceptable range. This process can also adopt a cross-validation method to ensure the generalization ability of the model.

[0063] Through the optimized model and parameters, calculate the degradation rates of different local regions to obtain more specific material degradation information.

[0064] The feature fusion network is used to fuse multi-band luminescence image sequences and degradation rates to extract cross-band time series features.

[0065] The feature fusion network includes a feature extraction unit, a feature fusion unit, and a feature selection unit. The feature extraction unit is used to extract the time series features of multi-band luminescence image sequences through a convolutional neural network. The feature fusion unit is used to fuse the time series features and degradation rates using a fusion algorithm to generate a feature set. The feature selection unit is used to screen the key features in the feature set through the principal component analysis method.

[0066] The main function of the feature extraction unit is to extract the temporal features of the multi-band luminescence image sequence through a convolutional neural network. The specific implementation process is as follows: Before starting feature extraction, first standardize the multi-band image sequence to ensure that the input data is within the same range and reduce the impact of differences during model training.

[0067] Design a suitable convolutional neural network architecture, including multiple convolutional layers, pooling layers, and fully connected layers, to extract hierarchical temporal features. The convolutional layers are used to capture local features, and the pooling layers are used to reduce the feature dimension and prevent overfitting.

[0068] Take the multi-band image sequence as input and perform forward propagation through the convolutional neural network. The extracted features will represent the luminescence features of the material at different times, including information such as shape, texture, and light intensity changes.

[0069] At the end of the convolutional neural network, combine the fully connected layer to output a feature vector to form a high-dimensional representation.

[0070] The feature fusion unit is used to fuse the temporal features and the degradation rate using a fusion algorithm to generate a combined feature set. The implementation process includes: Input the temporal feature vector from the feature extraction unit and the degradation rate data from the dynamic degradation module. Ensure that these two features are consistent in time for effective fusion.

[0071] Select a suitable fusion algorithm according to the characteristics and requirements of the features, including weighted summation, concatenation, averaging, or attention mechanism-based fusion methods.

[0072] Combine the temporal features and the degradation rate through the selected fusion algorithm to generate a comprehensive feature set. For example, assign weights to different features in weighted summation to highlight more important temporal features or degradation information.

[0073] The fused feature set will serve as the basis for subsequent feature selection, integrating degradation and temporal information.

[0074] The feature selection unit is used to filter out key features from the feature set through the principal component analysis method, reduce the data dimension, and optimize the model performance. The specific implementation process is as follows: Standardize the fused feature set to ensure that the mean of each feature is 0 and the variance is 1. Then calculate the covariance matrix of this feature set to analyze the correlation between different features.

[0075] Perform eigenvalue decomposition on the covariance matrix to extract the eigenvectors and corresponding eigenvalues. The eigenvalues describe the variance of the corresponding features in the data, and the eigenvectors describe the directions in the original data space.

[0076] Select the top k eigenvectors as the principal components according to the magnitudes of the eigenvalues. These principal components can retain as much variability of the original data as possible to obtain the key features.

[0077] The lifespan prediction module is used to generate a theoretical lifespan prediction value based on the degradation rate, obtain the degradation law of cross-band time series features through a long short-term memory network, combine the theoretical lifespan prediction value and the degradation law, correct the prediction deviation, and output the final lifespan assessment.

[0078] The lifespan prediction module includes a theoretical prediction unit, a long short-term memory network unit, and a prediction correction unit. The theoretical prediction unit is used to construct a theoretical lifespan model and generate a theoretical lifespan prediction value of the organic electroluminescent material in combination with the degradation rate. The long short-term memory network unit is used to construct a degradation prediction model based on the long short-term memory network, input the key features, and output the degradation law. The prediction correction unit is used to combine the theoretical lifespan prediction value and the degradation law, and correct the prediction deviation through a regression analysis algorithm to obtain the final lifespan assessment.

[0079] Figure 2 It is a schematic flow diagram for generating the theoretical lifespan prediction value of the organic electroluminescent material in an embodiment of the present invention.

[0080] As Figure 2 shown, the process of generating the theoretical lifespan prediction value of the organic electroluminescent material includes: Collect the initial performance data and historical degradation rates of the organic electroluminescent material. The initial performance data includes the initial brightness and usage conditions.

[0081] According to the initial performance data and historical degradation rates, construct a theoretical lifespan model using the degradation exponential decay formula of the organic electroluminescent material.

[0082] Set the critical brightness value of the organic electroluminescent material, and calculate the time required to reach the critical brightness value according to the degradation rate to obtain the theoretical lifespan prediction value.

[0083] The process of constructing a degradation prediction model based on the long short-term memory network includes: Collect the historical degradation data of the organic electroluminescent material. The historical degradation data includes the historical multi-band luminescence image sequence and the corresponding degradation law of the historical organic electroluminescent material.

[0084] Construct a historical dynamic model based on the historical multi-band luminescence image sequence and obtain the historical degradation rate.

[0085] Extract the historical time series features of the historical multi-band luminescence image sequence, generate a historical feature set in combination with the historical degradation rate, and screen the historical key features in the historical feature set.

[0086] Build a basic long short-term memory network model, and set the number of network layers, the number of neurons in each layer, and the activation function of the basic model.

[0087] Use historical key features as input and the corresponding historical degradation laws as output to train the basic model. Retain the model parameters that meet the test accuracy to obtain a degradation prediction model.

[0088] The prediction correction unit is used to combine the theoretical life prediction value and the degradation law, and correct the prediction deviation through a regression analysis algorithm. The process includes: Receive relevant data from the theoretical prediction unit and the long short-term memory network unit, including the theoretical life prediction value and the degradation law. The theoretical life prediction value comes from the life estimate generated by the theoretical prediction unit based on the degradation rate. The degradation law comes from the dynamic law of material degradation extracted and output by the long short-term memory network unit according to the key features.

[0089] The prediction correction unit performs prediction correction by constructing a regression model. The process includes: Select a suitable regression algorithm according to the data characteristics and requirements, such as linear regression, ridge regression, or Lasso regression.

[0090] Formulate an objective function to be optimized by regression, that is, the error between the theoretical prediction value and the prediction value output by the degradation prediction model based on the long short-term memory network. The objective function includes the mean square error or the absolute error.

[0091] Use the collected training data to train the regression model so that it can incorporate the interaction between the theoretical life prediction and the degradation law. The specific steps are as follows: Prepare a training sample set according to the theoretical life prediction value and the degradation law. Each sample contains the theoretical prediction value and the output value of the long short-term memory network degradation prediction model.

[0092] Use the prepared sample data to train the regression model. During the training process, continuously adjust the model parameters to minimize the value of the objective function and improve the generalization ability of the model on the test set.

[0093] After the training is completed, evaluate the performance of the regression model, and use methods such as cross-validation to test the stability and accuracy of the model to ensure that it can better capture the relationship between the theoretical prediction value and the degradation law.

[0094] Apply the trained regression model to correct the life prediction results. The specific steps are as follows: Use the regression model to predict the input combination of the theoretical life prediction and the degradation law to obtain the corrected life prediction value.

[0095] Based on the corrected results output by the model, a final life assessment that comprehensively considers theoretical prediction and degradation factors is obtained. This life assessment will reflect the more real material performance status and can effectively guide the decisions on the use and maintenance of materials.

[0096] It also includes a feedback control module, which is used to divide the warning levels according to the final life assessment and trigger corresponding warning signals.

[0097] The feedback control module includes a warning level division unit, a warning information generation unit, and a system feedback unit. The warning level division unit classifies the life assessment results into different warning levels by setting thresholds. The warning level is output according to the final life assessment.

[0098] Once the assessment result reaches a certain warning threshold, the warning signal generation unit triggers the generation of corresponding alarm signals to generate an alarm, reminding relevant personnel in a visual or auditory manner.

[0099] The system feedback unit continuously monitors and feedbacks the operating state of the control system, conducts real-time monitoring and data collection on the system operating conditions to optimize the subsequent work process.

[0100] In summary, this embodiment provides a life monitoring device for organic electroluminescent materials. By using a tunable excitation light source to effectively excite the organic electroluminescent materials and capturing the luminescence image sequences in different bands, it can comprehensively obtain the luminescence characteristics of the materials, provide rich spectral information, and provide high-quality data support for subsequent analysis. Through ambient light suppression and timing synchronization, it can ensure that the influence of background noise is minimized during the imaging process, improving the usability and accuracy of the images. By deeply analyzing the multi-band luminescence image sequences, a dynamic model of the materials is established in real time, thereby quantitatively showing the change of the material luminescence characteristics over time. It can not only quantify the degradation rate but also systematically analyze the stability of the materials under different usage conditions and environments, providing early warnings for potential failure modes. It makes the life assessment of the materials more scientific and reasonable, and can effectively reduce the failure risk. By fusing the multi-band luminescence image features and the degradation rate, the cross-band timing features are extracted, enhancing the depth and breadth of feature analysis. Combining convolutional neural network and principal component analysis ensures the retention of key information and reduces unnecessary redundant data, providing a more comprehensive perspective for the performance assessment of the materials, which helps to further understand the comprehensive influencing factors of material degradation. By fusing the theoretical prediction values with the degradation laws obtained through long short-term memory network, in-depth analysis and accurate prediction of the material performance are realized. By using the regression analysis method to correct the prediction deviation, it can more accurately capture the dynamic degradation characteristics of the materials, enabling the final life assessment to be obtained on the basis of approaching the real usage conditions.

[0101] Based on the same general inventive concept, the present invention also protects a method for monitoring the lifespan of an organic electroluminescent material. The following describes the method for fault detection of an organic electroluminescent device based on image processing provided by the present invention. The method for fault detection of an organic electroluminescent device based on image processing described below can be correspondingly referred to in relation to the device for monitoring the lifespan of an organic electroluminescent material described above.

[0102] Figure 3 It is a schematic flow diagram of a method for monitoring the lifespan of an organic electroluminescent material provided by an embodiment of the present invention.

[0103] As Figure 3 shown, a method for monitoring the lifespan of an organic electroluminescent material includes: Using a tunable excitation light source to excite the organic electroluminescent material, and after excitation, obtaining a sequence of luminescence images of the organic electroluminescent material at different wavelength bands.

[0104] Constructing a dynamic model of the luminescence characteristics changing with time based on the multi-band luminescence image sequence, and quantifying the degradation rate of the material through the dynamic model.

[0105] Fusing the multi-band luminescence image sequence and the degradation rate, and extracting cross-band temporal features.

[0106] Generating a theoretical lifespan prediction value based on the degradation rate, and analyzing the temporal features extracted through a long short-term memory network to obtain the degradation law of the organic electroluminescent material. Combining the theoretical lifespan prediction value with the degradation law to obtain the final lifespan assessment.

[0107] Setting a lifespan warning threshold for the organic electroluminescent material, dividing the warning levels according to the final lifespan assessment, and triggering corresponding warning signals.

[0108] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0109] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A device for monitoring the life of an organic electroluminescent material, characterized in that: include: A multi-spectral excitation module is used to excite the organic electroluminescent material through a tunable excitation light source to obtain a multi-band luminescent image sequence of the organic electroluminescent material; A dynamic degradation module, used to establish a dynamic model of the material luminescence characteristics changing over time according to the multi-band luminescence image sequence, and quantify the degradation rate; A feature fusion network, used for fusing the multi-band luminous image sequence and the degradation rate to extract cross-band time series features; The life prediction module is used to generate a theoretical life prediction value based on the degradation rate, obtain the degradation law of the cross-band time series characteristics through a long short-term memory network, combine the theoretical life prediction value with the degradation law, correct the prediction deviation and output the final life evaluation.

2. The device for monitoring the lifetime of an organic electroluminescent material according to claim 1, characterized in that: The multi-spectral excitation module includes a spectrum switching unit, a time synchronization controller and an ambient light suppression unit; the spectrum switching unit is used to adjust the wavelength range of the excitation light source to capture the luminous images of organic electroluminescent materials in different bands; the time synchronization controller is used to coordinate the timing synchronization between the excitation light source and the image acquisition device; the ambient light suppression unit is used to dynamically filter background stray light to obtain a multi-band luminous image sequence.

3. The device for monitoring the life of an organic electroluminescent material according to claim 2, characterized in that: The process of dynamically filtering background stray light includes: During the process of the tunable excitation light source exciting the organic electroluminescent material, a light sensor is provided to monitor the ambient light data in real time, wherein the ambient light data includes intensity and wavelength characteristics; Adjust the wavelength of the excitation light source and coordinate the timing of the excitation light source and the image acquisition device using a time synchronization controller; Comparing the captured image with the ambient light data, performing pixel value subtraction using the ambient light intensity, and dynamically adjusting the image; A dynamic filtering algorithm is used to update filtering parameters in real time according to changes in the ambient light data to obtain a multi-band luminous image sequence.

4. The device for monitoring the lifetime of an organic electroluminescent material according to claim 1, characterized in that: The dynamic degradation module includes a time series decomposition unit, a dynamic model construction unit and a parameter optimization unit; the time series decomposition unit is used to decompose the multi-band luminous image into a basic brightness component and a degradation component; The dynamic model building unit is used to build a nonlinear degradation model according to the degradation components to obtain the degradation rate of the organic electroluminescent material; the parameter optimization unit is used to fit the nonlinear degradation model parameters through an optimization algorithm to quantify the degradation difference in the local area.

5. The device for monitoring the lifetime of an organic electroluminescent material according to claim 4, characterized in that: The process of obtaining the degradation rate of the organic electroluminescent material includes: A power law function is selected to construct an initial model of organic electroluminescent material degradation; A polynomial regression algorithm is used to analyze the characteristics of the degradation components and identify key factors affecting the degradation of the organic electroluminescent material, wherein the key factors include light intensity, temperature change and usage time; According to the key factors, the initial model is fitted by the least square method to obtain a nonlinear degradation model; The degradation parameters in the nonlinear degradation model are extracted to obtain the degradation rate of the organic electroluminescent material, wherein the degradation parameters include a material attenuation constant and a characteristic time constant.

6. The device for monitoring the lifetime of an organic electroluminescent material according to claim 1, characterized in that: The feature fusion network includes a feature extraction unit, a feature fusion unit and a feature selection unit; The feature extraction unit is used to extract the time series features of the multi-band luminescence image sequence through a convolutional neural network; the feature fusion unit is used to fuse the time series features and the degradation rate using a fusion algorithm to generate a feature set; the feature selection unit is used to screen the key features in the feature set through a principal component analysis method.

7. The device for monitoring the lifetime of an organic electroluminescent material according to claim 6, characterized in that: The lifetime prediction module includes a theoretical prediction unit, a long short-term memory network unit and a prediction correction unit; the theoretical prediction unit is used to construct a theoretical lifetime model, and generate a theoretical lifetime prediction value of the organic electroluminescent material in combination with the degradation rate; the long short-term memory network unit is used to construct a degradation prediction model based on the long short-term memory network, input the key features, and output the degradation law; the prediction correction unit is used to combine the theoretical lifetime prediction value and the degradation law, and correct the prediction deviation through a regression analysis algorithm to obtain a final lifetime assessment.

8. The device for monitoring the lifetime of an organic electroluminescent material according to claim 1, characterized in that: The process of generating a theoretical lifetime prediction for an organic electroluminescent material involves: Collecting initial performance data and historical degradation rates of organic electroluminescent materials, wherein the initial performance data includes initial brightness and usage conditions; Based on the initial performance data and the historical degradation rate, a theoretical life model is constructed using an exponential decay formula for degradation of organic electroluminescent materials; The critical brightness value of the organic electroluminescent material is set, and the time required to reach the critical brightness value is calculated according to the degradation rate to obtain a theoretical life prediction value.

9. The device for monitoring the lifetime of an organic electroluminescent material according to claim 1, characterized in that: The process of building a degradation prediction model based on long short-term memory network includes: Collecting historical degradation data of organic electroluminescent materials, wherein the historical degradation data includes historical multi-band luminescent image sequences and corresponding historical degradation laws of organic electroluminescent materials; Building a historical dynamic model according to the historical multi-band luminescence image sequence, and obtaining a historical degradation rate; Extracting historical time series features of the historical multi-band luminescence image sequence, generating a historical feature set in combination with the historical degradation rate, and screening historical key features in the historical feature set; Constructing a long short-term memory network basic model, setting the number of network layers, the number of neurons in each layer and the activation function of the basic model; The historical key features are used as input and the corresponding historical degradation laws are used as output, the basic model is trained, and the model parameters that meet the test accuracy are retained to obtain a degradation prediction model.

10. A method for monitoring the life of an organic electroluminescent material, using the device for monitoring the life of an organic electroluminescent material as claimed in any one of claims 1 to 9, characterized in that: The organic electroluminescent device fault detection method based on image processing comprises: Using a tunable excitation light source to excite the organic electroluminescent material, after the excitation, a sequence of luminescent images of the organic electroluminescent material in different wavelength bands is obtained; A dynamic model of the time-varying luminescence characteristics is constructed based on a multi-band luminescence image sequence, and the degradation rate of the material is quantified through the dynamic model; The multi-band luminescence image sequence and degradation rate are fused to extract the cross-band temporal features; Generate theoretical lifetime prediction value based on degradation rate, and obtain degradation law of organic electroluminescent material by extracting time series characteristics through long short-term memory network analysis. Combine theoretical lifetime prediction value with degradation law to obtain final lifetime assessment. Set the lifetime warning threshold of the organic electroluminescent material, divide the warning level according to the final lifetime evaluation, and trigger the corresponding alarm signal.

Images