Adaptive adjustment method for active matrix organic light-emitting display

Abstract

The invention relates to the technical field of organic light-emitting display data processing, in particular to an adaptive adjustment method for active matrix organic light-emitting display. In combination with current stress parameter fluctuation captured by the programmable current sampling module and heat distribution data of the temperature sensor network, the dynamic aging model micro-service cluster is input for parameter coupling calculation, and time-varying degradation characteristic parameters are output; the containerized simulation micro-service constructs a virtual aging scene based on the parameters and generates a compensation coefficient matrix; according to the method, the problem of compensation deviation caused by the fact that a traditional static model cannot adapt to current stress, temperature bias and a material relaxation coupling effect is solved, dynamic balance of display brightness, color gamut and power consumption is achieved, and the service life of a display panel is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of organic light-emitting display data processing, and particularly to an adaptive adjustment method for active matrix organic light-emitting displays. Background Art

[0002] The adaptive adjustment of active matrix organic light-emitting displays is based on an ambient light sensor and a real-time feedback system. By detecting the external light intensity and color temperature parameters, and combining with a temperature compensation circuit, the pixel drive current and the sub-pixel emission duration are dynamically adjusted to maintain the visual consistency of the screen brightness and color temperature. Under low ambient illumination, the system will reduce the drive voltage according to the gamma correction curve, and at the same time optimize the carrier injection efficiency of the organic light-emitting layer to balance power consumption and display accuracy; when high dynamic range content is detected, the pixel-level compensation algorithm will synchronously correct the gate voltage of the thin film transistor (TFT), compensate for the brightness attenuation caused by organic material aging, and reduce the afterimage effect in high-contrast scenes through dynamic refresh rate adjustment. This adjustment mechanism combines optoelectronic characteristic modeling and non-linear optimization algorithms, effectively extending the service life of the display while maintaining the color gamut coverage and ΔE color difference index.

[0003] In the prior art, the pixel-level compensation algorithm based on a preset aging model is difficult to accurately adapt to the time-varying characteristics of the degradation rate of organic light-emitting units. Its static parameter mapping relationship cannot reflect the non-linear aging trajectory under the combined action of drive current stress, temperature bias, and material intrinsic relaxation effects in real time, resulting in a cumulative deviation between the aging compensation coefficient and the actual attenuation amount of the pixel. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides an adaptive adjustment method for active matrix organic light-emitting displays, which solves the problem that the static parameter mapping relationship based on the preset aging model cannot adapt to the non-linear time-varying degradation characteristics of organic light-emitting units under the combined action of drive current stress, temperature bias, and material relaxation effects in real time, resulting in a cumulative deviation between the aging compensation coefficient and the actual attenuation amount.

[0005] To solve the above technical problems, the specific technical solution of the present invention is as follows: The adaptive adjustment method for active matrix organic light-emitting displays provided by the present invention includes: Collecting ambient light intensity and color temperature data by deploying distributed photosensitive sensor nodes, and transmitting the ambient light intensity and color temperature data to the dynamic aging model microservice cluster through the central data bus; Using a programmable current sampling module to capture the current stress parameter fluctuations in the drive circuit, and obtaining the thermal distribution data collected by the temperature sensor network, and inputting the current stress parameter fluctuations and the thermal distribution data into the dynamic aging model microservice cluster for parameter coupling calculation, and outputting time-varying degradation characteristic parameters; Based on the time-varying degradation characteristic parameters output by the dynamic aging model microservice cluster, a virtual aging scenario is constructed through containerized simulation microservices to generate a compensation coefficient matrix; The compensation coefficient matrix is input into a multi-objective loss function solver, and parameter optimization is performed in combination with a preset ΔE color difference threshold and power consumption constraint conditions to output optimized display driving parameters; The real-time adjustment of the optimized display driving parameters is executed through an edge-cloud collaborative architecture, and the compensation coefficient matrix is synchronized to the online learning microservice, and an incremental model update is triggered based on the optimized display driving parameters and the compensation coefficient matrix.

[0006] Furthermore, in the adaptive adjustment method for active matrix organic light-emitting display of the present invention, the data acquisition process of the distributed photosensitive sensor nodes includes: Each sensor node independently acquires ambient light parameters through a microservice architecture and uploads the ambient light parameters to the central data bus through a lightweight API protocol; After receiving multi-source data including ambient light parameters, current stress parameter fluctuations, and temperature distribution data, the central data bus calls the edge computing unit to perform abnormal pulse preprocessing on the current stress parameter fluctuations to generate filtered current stress time series data; The temperature sensor network, based on a topology-aware dynamic orchestration strategy, establishes a heat conduction path model according to the preprocessed current stress time series data to correct the spatial resolution deviation of the heat distribution data.

[0007] Furthermore, in the adaptive adjustment method for active matrix organic light-emitting display of the present invention, the construction of the dynamic aging model microservice cluster includes: The current stress cumulative calculation service, the temperature bias non-linear function solver, and the material relaxation coefficient matrix generator are split into independent microservices; Data pipeline coupling between the current stress cumulative calculation service and the material relaxation coefficient matrix generator is realized through an asynchronous message queue, where the filtered current stress time series data triggers real-time correction of the material relaxation coefficient matrix; The material relaxation coefficient matrix generator quantifies the mobility decay error by RPC calling the current density and mobility decay curves pre-stored in the experimental database and combining the corrected heat distribution data.

[0008] Furthermore, in the adaptive adjustment method for active matrix organic light-emitting display of the present invention, the operation process of the containerized simulation microservice includes: The LSTM network is deployed as a containerized inference microservice, and after receiving the filtered current stress time series data stream, it performs online weight fine-tuning to generate an adaptive time series correlation prediction model; The incremental learning algorithm is used to inject the organic material response delay data into the training set of the adaptive time series correlation prediction model, and update the weight parameters of the prediction model; Based on the adaptive time series correlation prediction model, the distributed simulation microservice parallel-computes the compensation strategy, and the consistent hashing algorithm is used to maintain the spatio-temporal correlation of the compensation parameters.

[0009] Furthermore, in the adaptive adjustment method for the active matrix organic light-emitting display of the present invention, the correction process of the material relaxation coefficient matrix includes: According to the corrected thermal distribution data output by the temperature sensor network, update the temperature bias parameter of the material intrinsic relaxation effect; Call the pre-stored current density and mobility decay curve in the experimental database, and combine the output result of the current stress cumulative calculation service to generate a mobility decay compensation gradient; Non-linearly couple the mobility decay compensation gradient and the current stress cumulative calculation result, and output the corrected relaxation coefficient matrix to the dynamic aging model microservice cluster.

[0010] Furthermore, in the adaptive adjustment method for the active matrix organic light-emitting display of the present invention, the construction process of the virtual aging scenario includes: Generate an initial pixel-level aging compensation strategy based on the time-varying degradation parameters output by the dynamic aging model microservice cluster; Split the pixel-level aging compensation initial strategy into three parallel microservice tasks: gamma correction weight calculation, sub-pixel emission duration optimization, and TFT gate voltage adjustment gradient; Through the service mesh, allocate the three parallel microservice tasks to the distributed computing nodes for collaborative execution, and use the transaction log to record the simulation status data of the virtual aging scenario.

[0011] Furthermore, in the adaptive adjustment method for the active matrix organic light-emitting display of the present invention, the optimization process of the multi-objective loss function solver includes: Receive the preset ΔE color difference threshold, afterimage suppression index, and power consumption constraint conditions, and construct a Pareto front analysis model; Through the two-way communication channel between the backpropagation microservice and the dynamic aging model microservice cluster, synchronously update the coupling weight parameters of the Pareto front analysis model and the dynamic aging model; Use differential privacy technology to desensitize the gradient sharing data output by the backpropagation microservice.

[0012] Furthermore, in the adaptive adjustment method for the active matrix organic light-emitting display of the present invention, the execution process of the edge-cloud collaborative architecture includes: Deploy the display driver chip register mapping microservice at the edge side to execute the optimized display driver parameter adjustment instruction output by the multi-objective loss function solver in real time; Synchronize the version snapshot of the compensation coefficient matrix to the cloud historical database to establish an aging parameter version control link based on timestamps; Obtain the brightness attenuation feedback data of the actual display panel through the subscription and publishing mode to trigger the incremental model update operation of the online learning microservice.

[0013] Furthermore, in the adaptive adjustment method of the active matrix organic light emitting display of the present invention, the incremental model update process includes: After receiving the brightness attenuation feedback data, the online learning microservice generates a historical database update instruction and attaches the version identifier of the compensation coefficient matrix; Route the update instruction to the dynamic aging model microservice cluster through the service mesh to trigger the parameter recalibration of the current stress accumulation calculation service and the material relaxation coefficient matrix generator; Adopt the federated learning framework to aggregate the brightness attenuation feedback data and historical aging parameters across devices to update the weight coefficients of the global dynamic aging model.

[0014] Furthermore, the adaptive adjustment method of the active matrix organic light emitting display of the present invention further includes: The distributed sensor microservice and the edge preprocessing unit reduce the computing load of the central node through asynchronous data processing to support the dynamic expansion of ultra-high resolution screens; The containerized simulation microservice dynamically allocates computing resources based on the elastic resource scheduling strategy to accelerate the verification efficiency of the compensation strategy for the virtual aging scenario; The incremental learning and federated learning microservices synchronize and optimize the time-varying adaptation parameters and cross-device generalization parameters of the dynamic aging model through a closed-loop data link; The service mesh and the version control middleware maintain the state consistency between microservices based on the simulated state data of the transaction log to prevent the display parameter jump caused by the update of local microservices.

[0015] Advantages of the present invention; Through the synergistic effect of the dynamic aging model microservice cluster and the containerized simulation microservice, the present invention significantly improves the adaptive adjustment accuracy and efficiency of the organic light-emitting display system. The distributed photosensitive sensors and multi-source data fusion technology are used to capture the ambient light, current stress, and thermal distribution parameters in real time. Combining with the asynchronous message queue, dynamic parameter coupling between microservices is realized, and the non-linear correlation between the material relaxation effect and the temperature bias is accurately quantified. Based on the adaptive time series prediction model and incremental learning mechanism of the LSTM network, through the spatio-temporal consistency generation strategy of the compensation coefficient matrix, the compensation deviation caused by the mismatch between the current stress accumulation and the material aging rate in the traditional static model is effectively eliminated. The federated learning framework under the edge-cloud collaborative architecture combines differential privacy technology to achieve privacy-protected aggregation of cross-device aging data, and synchronously optimizes the time-varying adaptability and generalization ability of the dynamic aging model. The service mesh and version control middleware maintain the microservice state consistency through transaction logs, and combine the gray release strategy to suppress the risk of parameter jumps, and finally achieve the dynamic balance of display brightness, color gamut coverage, and power consumption indicators, and extend the service life of the display panel. Description of the Drawings

[0016] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the drawings.

[0017] Figure 1 It is a flowchart of the adaptive adjustment method for active matrix organic light-emitting display provided by the embodiment of the present invention. Detailed Embodiments

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. The technical solutions provided by each embodiment of the present invention are described in detail below in conjunction with the drawings. To better understand the objectives of the present invention, the present invention is further described in detail below.

[0019] Please refer to Figure 1 , the adaptive adjustment method for active matrix organic light-emitting display provided by the present invention includes: Step S101, collecting the ambient light intensity and color temperature data through the deployment of distributed photosensitive sensor nodes, and transmitting the ambient light intensity and color temperature data to the dynamic aging model microservice cluster through the central data bus; Step S102: Use the programmable current sampling module to capture the current stress parameter fluctuations in the driving circuit, and obtain the thermal distribution data collected by the temperature sensor network. Input the current stress parameter fluctuations and the thermal distribution data into the dynamic aging model microservice cluster for parameter coupling calculation, and output the time-varying degradation characteristic parameters. Step S103: Based on the time-varying degradation characteristic parameters output by the dynamic aging model microservice cluster, construct a virtual aging scenario through the containerized simulation microservice to generate a compensation coefficient matrix. Step S104: Input the compensation coefficient matrix into the multi-objective loss function solver, and perform parameter optimization in combination with the preset ΔE color difference threshold and power consumption constraint conditions, and output the optimized display driving parameters. Step S105: Execute the real-time adjustment of the optimized display driving parameters through the edge-cloud collaborative architecture, and synchronize the compensation coefficient matrix to the online learning microservice, and trigger incremental model updates based on the optimized display driving parameters and the compensation coefficient matrix.

[0020] The adaptive adjustment method for active matrix organic light-emitting display provided by the present invention realizes the dynamic optimization of display parameters through multi-level data processing and microservice architecture. First, deploy distributed photosensitive sensor nodes. Each node independently collects ambient light intensity and color temperature data based on the microservice architecture, and transmits the original data to the central data bus through the lightweight API protocol. The central data bus integrates an edge computing unit, performs abnormal pulse filtering processing on the current stress parameter fluctuations of the driving circuit captured by the current sampling module, and at the same time receives the thermal distribution data generated by the temperature sensor network based on the topology-aware dynamic orchestration strategy to form a multi-source heterogeneous sensor data stream.

[0021] Input the processed current stress parameter fluctuations and thermal distribution data into the dynamic aging model microservice cluster. The microservice cluster is composed of a current stress cumulative calculation service, a temperature bias non-linear function solver, and a material relaxation coefficient matrix generator. Realize the data pipeline coupling between microservices through an asynchronous message queue. Among them, the filtered current stress time series data triggers the real-time correction of the material relaxation coefficient matrix, and combines the current density-mobility decay curve pre-stored in the experimental database to quantify the mobility decay error and output the time-varying degradation characteristic parameters. The time-varying degradation characteristic parameters characterize the non-linear aging trajectory of the organic light-emitting unit under the combined action of current stress, temperature bias, and material relaxation effects.

[0022] Based on the time-varying degradation characteristic parameters, the containerized simulation microservices construct a virtual aging scenario. The simulation microservices deploy the LSTM network as a containerized inference unit, receive the time-series current data stream, perform online weight fine-tuning, and generate an adaptive time-series correlation prediction model. The gamma correction weight calculation, sub-pixel emission duration optimization, and TFT gate voltage adjustment gradient solving tasks are executed in parallel by distributed computing nodes. The consistent hashing algorithm is used to maintain the spatio-temporal correlation of the compensation parameters, and the transaction log is used to record the simulation state data of the virtual aging scenario, forming a compensation coefficient matrix.

[0023] The compensation coefficient matrix is input into the multi-objective loss function solver. Combining the preset ΔE color difference threshold and power consumption constraint conditions, a Pareto front analysis model is constructed. The coupled weight parameters are synchronously updated through the two-way communication channel between the backpropagation microservice and the dynamic aging model microservice cluster. The differential privacy technology is used to desensitize the gradient sharing data, and the optimized display driving parameters that meet the requirements of color gamut coverage, ghosting suppression, and power consumption balance are output.

[0024] The real-time adjustment of the display driving parameters is executed through the edge-cloud collaborative architecture. The display driver chip register mapping microservice is deployed at the edge to directly control the current and refresh rate, and at the same time, the version snapshot of the compensation coefficient matrix is synchronized to the cloud historical database. The online learning microservice obtains the brightness attenuation feedback data of the actual display panel based on the subscription-publish mode, and triggers incremental model updates after attaching the version identifier of the compensation coefficient matrix. The cross-device aging data is aggregated through the federated learning framework, and the global dynamic aging model parameters are updated in combination with the service mesh routing mechanism, forming a closed-loop for display parameter optimization. Each functional module maintains state consistency through the service mesh and version control middleware to prevent display anomalies caused by local updates, and realizes the adaptive adjustment ability of the display system throughout its life cycle.

[0025] Specifically, for the adaptive adjustment method of the active matrix organic light emitting display of the present invention, the data acquisition process of the distributed photosensitive sensor nodes includes: Each sensor node independently acquires the ambient light parameters through the microservice architecture and uploads the ambient light parameters to the central data bus through the lightweight API protocol; After receiving the multi-source data including the ambient light parameters, the current stress parameter fluctuations, and the temperature distribution data, the central data bus calls the edge computing unit to perform abnormal pulse preprocessing on the current stress parameter fluctuations and generates filtered current stress time-series data; Based on the topology-aware dynamic orchestration strategy, the temperature sensor network establishes a heat conduction path model according to the preprocessed current stress time-series data and corrects the spatial resolution deviation of the heat distribution data.

[0026] The data acquisition process of the distributed photosensitive sensor nodes realizes high-precision synchronous acquisition and preprocessing of environmental parameters through a multi-level collaborative architecture. Each sensor node is deployed with containerized microservices, and independently runs the environmental light parameter acquisition program. The light intensity signal is converted into a digital quantity through a preset photoelectric conversion calibration coefficient and encapsulated into a standardized data packet. The data packet is transmitted to the central data bus through a lightweight API protocol, and a message queue mechanism is used to achieve data flow control during multi-node concurrent uploads, avoiding transmission delays caused by bus congestion.

[0027] After receiving the multi-source heterogeneous data stream containing environmental light parameters, current stress parameter fluctuations, and temperature distribution data, the central data bus triggers the preprocessing process of the edge computing unit. The edge computing unit deploys a digital filtering algorithm to detect abnormal pulses in the current stress parameter fluctuations, and uses a sliding window mean filter to eliminate high-frequency noise interference, generating filtered current stress time-series data. This time-series data and the original thermal distribution data uploaded by the temperature sensor network are jointly input into the data fusion module, and the acquisition time series of multi-source data is matched through a timestamp alignment algorithm.

[0028] The temperature sensor network reconstructs the thermal distribution data processing link based on a topology-aware dynamic orchestration strategy, and establishes a heat conduction path model according to the filtered current stress time-series data. The model analyzes the spatial correlation between the current stress parameters and the temperature distribution, identifies the heat conduction path offset of the organic light-emitting display panel, and uses an interpolation algorithm to correct the spatial resolution deviation caused by the sparse layout of sensor nodes. The corrected thermal distribution data and the filtered current stress data are jointly used as input parameters for the dynamic aging model microservice cluster, forming a closed-loop logic link for sensor data acquisition, preprocessing, and model input.

[0029] Specifically, for the adaptive adjustment method of the active matrix organic light-emitting display of the present invention, the construction of the dynamic aging model microservice cluster includes: Splitting the current stress cumulative calculation service, temperature bias non-linear function solver, and material relaxation coefficient matrix generator into independent microservices; Realizing data pipeline coupling between the current stress cumulative calculation service and the material relaxation coefficient matrix generator through an asynchronous message queue, where the filtered current stress time-series data triggers real-time correction of the material relaxation coefficient matrix; The material relaxation coefficient matrix generator quantifies the mobility decay error by RPC calling the pre-stored current density and mobility decay curves in the experimental database and combining the corrected thermal distribution data.

[0030] The construction of the dynamic aging model microservice cluster realizes parameter coupling optimization by decoupling the core computing module from the distributed service collaboration. The current stress accumulation calculation service is deployed as an independent microservice, and the cumulative amount of the filtered current stress time-series data is calculated based on the time integration algorithm, and the current stress cumulative distribution map is output. The temperature bias non-linear function solver microservice receives the corrected heat distribution data, constructs the mapping relationship between the temperature gradient field and the current stress distribution by using the polynomial fitting algorithm, and generates the temperature compensation coefficient matrix.

[0031] A data pipeline between the current stress accumulation calculation service and the material relaxation coefficient matrix generator is established through an asynchronous message queue. When the filtered current stress time-series data reaches the preset time window threshold, the message queue is triggered to push a data update event to the material relaxation coefficient matrix generator. The material relaxation coefficient matrix generator starts the remote procedure call protocol, accesses the dataset of the current density and mobility decay curve pre-stored in the experimental database, and calculates the time-varying decay rate of the carrier mobility of the organic material in combination with the gradient parameters of the current temperature compensation coefficient matrix.

[0032] The non-linear coupling operation is performed on the mobility decay rate and the current stress cumulative distribution map, and the tensor decomposition algorithm is used to solve the spatial distribution characteristics of the material relaxation effect, and a dynamically updated relaxation coefficient matrix is generated. This matrix is synchronized to the temperature bias non-linear function solver microservice through the service mesh, forming a cross-feedback mechanism of current-temperature-material characteristics, and realizing the closed-loop iterative optimization of the aging model parameters. The data dependency between microservices is maintained by the version control middleware to ensure the temporal consistency during the parameter update process.

[0033] Specifically, for the adaptive adjustment method of the active matrix organic light-emitting display described in the present invention, the operation process of the containerized simulation microservice includes: The LSTM network is deployed as a containerized inference microservice, and after receiving the filtered current stress time-series data stream, it performs online weight fine-tuning to generate an adaptive time-series correlation prediction model; The organic material response delay data is injected into the training set of the adaptive time-series correlation prediction model by using the incremental learning algorithm to update the weight parameters of the prediction model; The distributed simulation microservice parallelly calculates the compensation strategy based on the adaptive time-series correlation prediction model, and uses the consistent hashing algorithm to maintain the spatio-temporal correlation of the compensation parameters.

[0034] The operation process of the containerized simulation microservice realizes the dynamic generation of compensation strategies through time series modeling and distributed computing technologies. The LSTM network is encapsulated as a containerized inference microservice, and elastic scaling is achieved based on the Kubernetes orchestration platform. After receiving the filtered current stress time series data stream, the online inference service is started. The LSTM network adopts a sliding time window division strategy to segment the input data, and fine-tunes the network weights through an adaptive learning rate adjustment algorithm to generate an adaptive time series correlation prediction model that can reflect the time-varying degradation characteristics of organic materials.

[0035] During the operation of the adaptive time series correlation prediction model, the incremental learning algorithm captures the organic material response delay data in real time, and performs time series alignment and normalization processing through a data buffer queue. The preprocessed delay data is injected into the model training set, and the backpropagation algorithm is used to update the weight parameters of the hidden layer of the LSTM network, so that the prediction model can dynamically adapt to the change trend of the material aging rate. The updated weight parameters generate a differential snapshot through the version control middleware, triggering the parameter synchronization event of the distributed simulation microservice.

[0036] Based on the updated adaptive time series correlation prediction model, the distributed simulation microservice splits the compensation strategy calculation task into three parallel subtasks: gamma correction weight, sub-pixel emission duration, and TFT gate voltage adjustment. The consistent hashing algorithm is used to construct a virtual node ring, and the subtasks are mapped to specific computing nodes according to the spatio-temporal tags of the compensation parameters. The spatio-temporal correlation of the compensation parameters is maintained through the redundant copy mechanism of the hash ring. The local compensation parameters generated by each node are aggregated and reconstructed through the service mesh, and a global compensation coefficient matrix matching the physical layout of the display panel is output.

[0037] Specifically, for the adaptive adjustment method of the active matrix organic light emitting display of the present invention, the correction process of the material relaxation coefficient matrix includes: Updating the temperature bias parameter of the material intrinsic relaxation effect according to the corrected heat distribution data output by the temperature sensor network; Calling the pre-stored current density and mobility decay curves in the experimental database, and combining the output results of the current stress cumulative calculation service to generate a mobility decay compensation gradient; Non-linearly coupling the mobility decay compensation gradient with the current stress cumulative calculation result, and outputting the corrected relaxation coefficient matrix to the dynamic aging model microservice cluster.

[0038] The correction process of the material relaxation coefficient matrix realizes the dynamic calibration of material properties through multi-source data fusion and nonlinear modeling. The corrected thermal distribution data output by the temperature sensor network is input into the material intrinsic relaxation effect analysis module. The polynomial interpolation algorithm is used to reconstruct the global temperature field of the display panel, and the temperature bias parameter is updated based on the thermodynamic relaxation equation. The temperature bias parameter is matched with the activation energy threshold in the organic material energy band structure database to generate the temperature-dependent carrier mobility correction coefficient.

[0039] When calling the current density and mobility decay curves stored in the experimental database, a remote data channel with the material property analysis microservice is established through the RPC protocol. The piecewise linear interpolation algorithm is used to map the current stress distribution map output by the current stress accumulation calculation service to the mobility decay curve, and a three-dimensional gradient vector field is constructed in combination with the temperature bias parameter to generate the mobility decay compensation gradient. The compensation gradient characterizes the non-linear decay rate of the organic material mobility under the coupled action of current stress and temperature.

[0040] The mobility decay compensation gradient and the current stress accumulation calculation result are subjected to tensor decomposition operation, and the coupling interference between different physical dimensions is eliminated through the orthogonal projection algorithm. The non-linear least squares method is used to optimize the weights of the decomposed eigenvectors, and the corrected relaxation coefficient matrix is output. This matrix is synchronized to the asynchronous message queue of the dynamic aging model microservice cluster through the service mesh, triggering the parameter recalibration event of the current stress accumulation calculation service, and forming a cross-iterative optimization loop of temperature-current stress-material properties. The version control middleware records the updated version identifier of the relaxation coefficient matrix, providing a data backtracking basis for incremental model updates.

[0041] Specifically, for the adaptive adjustment method of the active matrix organic light-emitting display of the present invention, the construction process of the virtual aging scenario includes: Generating an initial pixel-level aging compensation strategy based on the time-varying degradation parameters output by the dynamic aging model microservice cluster; Splitting the initial pixel-level aging compensation strategy into three parallel microservice tasks: gamma correction weight calculation, sub-pixel emission duration optimization, and TFT gate voltage adjustment gradient; Assigning the three parallel microservice tasks to distributed computing nodes for collaborative execution through the service mesh, and using the transaction log to record the simulation state data of the virtual aging scenario.

[0042] The construction process of the virtual aging scenario realizes the efficient verification of the compensation strategy through policy decomposition and distributed collaborative computing. The time-varying degradation parameters output by the dynamic aging model microservice cluster are input into the compensation strategy generation module. Based on the pixel-level current stress distribution map and the mobility decay gradient parameters, a weighted fusion algorithm is used to generate the initial aging compensation strategy. The initial strategy includes a set of compensation parameters in three dimensions: gamma voltage correction amount, sub-pixel emission time baseline, and TFT gate voltage offset amount.

[0043] When splitting the initial aging compensation strategy into parallel microservice tasks, task division is performed according to the data type of the compensation parameters and the characteristics of the computing load. The gamma correction weight calculation task calls the gamma lookup table data of the display panel, and combines the mobility decay gradient in the time-varying degradation parameters to generate the voltage compensation coefficient using an interpolation algorithm. The sub-pixel emission duration optimization task calculates the adjustment amount of the emission time baseline based on the response delay data of the organic material and the cumulative distribution of the current stress through a pulse width modulation model. The TFT gate voltage adjustment gradient solving task receives the temperature bias parameter and the mobility decay error data, and iteratively solves the gate voltage compensation amount using the gradient descent algorithm.

[0044] Through the load balancing strategy of the service mesh, the three parallel microservice tasks are assigned to distributed computing nodes, and the consistent hashing algorithm is used to maintain the mapping relationship between task assignment and the physical display area. When each node performs the calculation of the compensation parameters, the intermediate calculation results are synchronized through the message middleware, and the version identifier, timestamp, and node status data of the compensation coefficient are recorded using the transaction log. The transaction log data is input into the state tracing module of the virtual aging scenario, providing a historical data benchmark for the iterative optimization of the compensation strategy, and forming a closed-loop logical link for the generation, verification, and optimization of the compensation parameters.

[0045] Specifically, for the adaptive adjustment method of the active matrix organic light emitting display of the present invention, the optimization process of the multi-objective loss function solver includes: Receiving the preset ΔE color difference threshold, afterimage suppression index, and power consumption constraint conditions, and constructing a Pareto front analysis model; Synchronously updating the coupling weight parameters of the Pareto front analysis model and the dynamic aging model through the two-way communication channel between the backpropagation microservice and the dynamic aging model microservice cluster; Using differential privacy technology to desensitize the gradient sharing data output by the backpropagation microservice.

[0046] The optimization process of the multi-objective loss function solver realizes the search for the global optimal solution of the display parameters through multi-constraint fusion and privacy protection mechanisms. After receiving the preset ΔE color difference threshold, afterimage suppression index, and power consumption constraint conditions, a multi-objective normalization algorithm is used to map the constraint parameters of different dimensions to a unified evaluation space, and a Pareto front analysis model is constructed. The model generates a candidate solution set through the non-dominated sorting algorithm, and calculates the gamut coverage rate, afterimage index, and power consumption level evaluation values corresponding to each solution based on the time-varying degradation parameters output by the dynamic aging model microservice cluster.

[0047] After establishing a two-way communication channel between the backpropagation microservice and the dynamic aging model microservice cluster, the weight update gradient of the Pareto front analysis model is transmitted to the dynamic aging model through the gradient sharing protocol. The temperature bias non-linear function solver in the dynamic aging model microservice cluster receives the gradient data, and uses the adaptive momentum optimization algorithm to adjust the coupling weight parameters of the temperature compensation coefficient matrix. The synchronized updated weight parameters are fed back to the Pareto front analysis model through the version control middleware, forming a cross-model parameter collaborative update mechanism.

[0048] When performing privacy protection processing on the gradient sharing data output by the backpropagation microservice, differential privacy technology is used to inject Gaussian noise perturbations. The noise injection intensity is controlled through the privacy budget allocation algorithm, and local perturbations are performed on the feature dimensions of the gradient tensor to generate the desensitized gradient update amount. The desensitized gradient is transmitted to the dynamic aging model microservice cluster through the secure channel of the service mesh, blocking the sensitive data leakage path while maintaining the model optimization accuracy. The noise parameters and perturbation strategies during the privacy processing are dynamically configured through independent microservices, forming a traceable privacy protection operation log.

[0049] Specifically, for the adaptive adjustment method of the active matrix organic light emitting display described in the present invention, the execution process of the edge-cloud collaborative architecture includes: Deploy the display driver chip register mapping microservice at the edge side, and execute the optimized display driver parameter adjustment instruction output by the multi-objective loss function solver in real time; Synchronize the version snapshot of the compensation coefficient matrix to the cloud historical database, and establish a version control link of the aging parameters based on the time stamp; Obtain the brightness attenuation feedback data of the actual display panel through the subscription and publication mode, and trigger the incremental model update operation of the online learning microservice.

[0050] The execution process of the edge-cloud collaborative architecture realizes the dynamic optimization of the display system through hierarchical computing and data closed-loop. A display driver chip register mapping microservice is deployed at the edge side, directly docking with the optimized display driver parameters output by the multi-objective loss function solver. The parameters include gamma correction voltage values, sub-pixel light emission duration benchmarks, and TFT gate voltage offsets. The microservice converts the parameters into a binary instruction set recognizable by the driver chip register through the hardware abstraction layer interface, and updates the current control module and refresh rate control unit of the display panel in real time.

[0051] When synchronizing the version snapshot of the compensation coefficient matrix to the cloud historical database, an event-triggered timestamp generation mechanism is adopted, and the version identifier and generation time node of the compensation matrix are appended after each compensation strategy update. The historical database constructs a version control link, records the parameter changes of adjacent version compensation matrices through a differential comparison algorithm, and forms a traceable aging parameter evolution map. The data structure of the version control link adopts a key-value storage mode, and a mapping relationship between compensation parameters and virtual aging scenario simulation states is established with timestamps as indexes.

[0052] When obtaining the brightness attenuation feedback data of the actual display panel through the publish-subscribe mode, an optoelectronic sensor array is embedded in the display driver circuit, and the sub-pixel brightness attenuation is periodically collected and encapsulated into a standardized message. The message is broadcast to the subscription queue of the online learning microservice through the message middleware. Before triggering the incremental model update operation, data integrity verification and time window alignment operations are first performed. After the update operation is started, the service mesh routes to the corresponding dynamic aging model microservice node according to the compensation coefficient matrix version identifier, and completes the closed-loop iterative optimization of aging parameters and compensation strategies.

[0053] Specifically, for the adaptive adjustment method of the active matrix organic light-emitting display of the present invention, the incremental model update process includes: After receiving the brightness attenuation feedback data, the online learning microservice generates a historical database update instruction and appends the version identifier of the compensation coefficient matrix; The update instruction is routed to the dynamic aging model microservice cluster through the service mesh, triggering the parameter recalibration of the current stress accumulation calculation service and the material relaxation coefficient matrix generator; The federated learning framework is used to aggregate the brightness attenuation feedback data and historical aging parameters across devices, and update the weight coefficients of the global dynamic aging model.

[0054] The incremental model update process realizes the continuous optimization of the aging model through data closed-loop and distributed learning. After the online learning microservice receives the brightness attenuation feedback data collected by the optoelectronic sensor array of the display panel, it executes a data preprocessing process, including outlier filtering and time window alignment operations, to generate a standardized attenuation feature vector. The feature vector is appended with the version identifier of the compensation coefficient matrix and pushed to the historical database update instruction generation module through a message queue. The version identifier includes the compensation strategy generation timestamp and the hash check code.

[0055] When routing the update instruction to the dynamic aging model microservice cluster through the service mesh, a content-addressable routing strategy based on the version identifier is adopted to accurately match the computing node corresponding to the currently effective version of the compensation coefficient matrix. When triggering the parameter recalibration process of the current stress accumulation calculation service, the sliding window algorithm is used to adjust the time integration weight coefficient of the current stress accumulation calculation in combination with the spatial distribution information in the brightness attenuation feature vector. The recalibration operation of the material relaxation coefficient matrix generator is synchronously started, and the interpolation reference point of the mobility attenuation curve is corrected based on the difference between the attenuation feature vector and the historical compensation parameters.

[0056] When using the federated learning framework for cross-device data aggregation, a local differential privacy module is deployed at the edge to perform noise injection and feature desensitization processing on the brightness attenuation feedback data and historical aging parameters. The processed data is uploaded to the federated learning server through a secure multi-party computing protocol, and a weighted average algorithm is used to fuse the feature parameter distributions of multiple devices to generate the weight update gradient of the global dynamic aging model. The gradient is distributed to the dynamic aging model microservice clusters of each device through the service mesh, and combined with the parameter evolution graph recorded by the version control middleware, the incremental iterative update of the aging model parameters is completed, forming a dynamic adaptation closed-loop of the display panel degradation characteristics and compensation strategy.

[0057] Specifically, the adaptive adjustment method for the active matrix organic light-emitting display described in the present invention further includes: The distributed sensor microservice and the edge preprocessing unit reduce the computing load of the central node through asynchronous data processing, supporting the dynamic expansion of ultra-high-resolution screens; The containerized simulation microservice dynamically allocates computing resources based on an elastic resource scheduling strategy to accelerate the verification efficiency of the compensation strategy for the virtual aging scenario; The incremental learning and federated learning microservices synchronously optimize the time-varying adaptation parameters and cross-device generalization parameters of the dynamic aging model through a closed-loop data link; The service mesh and the version control middleware maintain the state consistency between microservices based on the simulated state data of the transaction log, preventing display parameter jumps caused by local microservice updates.

[0058] The distributed sensor microservices and the edge preprocessing unit adopt a message queue asynchronous processing mechanism to achieve dynamic allocation of computing loads. Each sensor node slices the collected raw data and pushes it to the local message queue of the edge computing node. The edge preprocessing unit deploys a streaming processing engine to perform parallel filtering and feature extraction operations on the data slices, generating standardized preprocessed data packets. The preprocessing results are temporarily stored through a distributed caching mechanism, and only the key feature data is uploaded to the central node, reducing the transmission overhead of processing massive data on high-resolution screens and supporting linear expansion of computing resources when the screen resolution is dynamically expanded.

[0059] The containerized simulation microservices dynamically manage computing resources based on the Kubernetes elastic scheduler, and evaluate the computing task loads according to the complexity of the compensation strategy for virtual aging scenarios. The scheduler automatically triggers horizontal scaling operations of container instances by monitoring container resource utilization metrics, and increases the number of simulation computing nodes during peak periods. A priority queue is used to manage the verification tasks of different compensation strategies, and a time-limit aware algorithm is combined to allocate computing resources, shortening the verification cycle of high-priority strategies and improving the simulation efficiency of complex aging scenarios.

[0060] The incremental learning microservices and the federated learning microservices build a dual-channel data closed-loop link. The incremental learning microservices generate local model update gradients through the real-time feedback data of the display panel, and the federated learning microservices aggregate the encrypted gradient data of multiple devices. The dual-channel adopts a time window synchronization mechanism to coordinate the update rhythm, triggers the federated aggregation operation after the local gradient reaches the convergence threshold, and fuses the cross-device feature distributions through a weighted average algorithm to synchronously update the time-varying adaptation parameters and device generalization parameters of the dynamic aging model, achieving a balance between model personalization and generality.

[0061] The service mesh and the version control middleware build a state consistency maintenance system based on transaction logs. The service mesh attaches a transaction identifier and an operation sequence number during communication between microservices, and the version control middleware records the state change events of each microservice in the distributed transaction log. When a local microservice version update is detected, the state consistency is verified through a log replay mechanism, and a two-phase commit protocol is used to coordinate cross-service parameter updates. For the detected risk of display parameter jumps, a version rollback mechanism is triggered to restore to the nearest stable state, and the update is gradually promoted in combination with a gray release strategy to maintain a smooth transition during the display parameter adjustment process.

[0062] In the specific implementation manner of the present invention, the dynamic adaptive adjustment of the organic light-emitting display system is realized through a multi-level technical architecture. Distributed photosensitive sensor nodes are deployed around the display panel. Each node operates independently based on the microservice architecture, collects ambient light intensity and color temperature data through the optoelectronic conversion module, and transmits the standardized data packets to the central data bus using the lightweight API protocol. The programmable current sampling module captures the current stress parameter fluctuations in the driving circuit in an event-driven mode, and forms a multi-source heterogeneous data stream including ambient light, current stress, and temperature by combining the thermal distribution data generated by the temperature sensor network based on the topology-aware dynamic orchestration strategy. The edge computing unit performs sliding window mean filtering and abnormal pulse detection on the current stress data to generate the filtered current stress time series data; the temperature sensor network establishes a heat conduction path model based on the current stress time series distribution, and corrects the spatial resolution deviation of the thermal distribution caused by the sparse layout of the sensor nodes through the interpolation algorithm.

[0063] The dynamic aging model microservice cluster consists of a current stress accumulation calculation service, a temperature bias non-linear function solver, and a material relaxation coefficient matrix generator. Each microservice realizes data pipeline coupling through an asynchronous message queue. The filtered current stress time series data triggers the material relaxation coefficient matrix generator to start a remote procedure call, access the pre-stored current density-mobility decay curve in the experimental database, and calculate the mobility decay error by combining the corrected thermal distribution data. The error and the current stress cumulative distribution map are non-linearly coupled through the tensor decomposition algorithm to output time-varying degradation characteristic parameters, which reflect the synergistic effects of driving current stress, temperature gradient, and material relaxation effect in real time. The containerized simulation microservice constructs a virtual aging scenario based on this parameter, deploys the LSTM network as a containerized inference unit, performs online weight fine-tuning after receiving the current stress time series data stream, and generates an adaptive time series correlation prediction model. The incremental learning algorithm injects the organic material response delay data in real time, updates the weights of the model hidden layer to adapt to the change of the aging rate, and parallelly executes the tasks of gamma correction weight, sub-pixel emission duration, and TFT gate voltage adjustment through the distributed computing nodes. The consistent hashing algorithm is used to maintain the spatio-temporal correlation of the compensation parameters, and a global compensation coefficient matrix is generated.

[0064] The multi-objective loss function solver receives the ΔE color difference threshold, the afterimage suppression index, and the power consumption constraint condition, constructs a Pareto front analysis model, and synchronously updates the coupled weight parameters through the two-way communication channel between the backpropagation microservice and the dynamic aging model. The differential privacy technology injects Gaussian noise perturbation into the gradient sharing data to generate the desensitized gradient update amount and transmits it to the dynamic aging model microservice cluster. The edge side deploys the display driver chip register mapping microservice, converts the optimized driver parameters into binary instruction sets to adjust the pixel drive current and refresh rate in real time; the version snapshot of the compensation coefficient matrix is attached with a timestamp identifier through the event trigger mechanism and synchronized to the cloud historical database to construct an aging parameter version control link. The online learning microservice obtains the brightness attenuation feedback data collected by the display panel photoelectric sensor based on the subscription-publish mode, triggers the federated learning framework to aggregate cross-device feature parameters, and updates the global dynamic aging model weights in combination with the service mesh routing strategy to form a dynamic adaptation closed-loop of the compensation strategy and the actual attenuation amount. The service mesh and version control middleware maintain the microservice state consistency through transaction logs, adopt the gray release strategy to gradually promote parameter updates, suppress the display parameter jumps caused by local microservice version iterations, and achieve continuous optimization and balance of display brightness, color gamut coverage, and power consumption indicators.

[0065] The explanations of the technical feature terms of the present invention are as follows: Distributed photosensitive sensor nodes: refer to multiple independent sensing units deployed around the display panel. Each node collects ambient light intensity and color temperature data in real time through a photoelectric conversion module and operates independently based on the microservice architecture. The nodes communicate with the central data bus through a lightweight API protocol to achieve concurrent acquisition and low-latency transmission of multi-source data, avoid single-point failures, and adapt to the expansion requirements of different screen sizes.

[0066] Dynamic aging model microservice cluster: a distributed computing module composed of a current stress accumulation calculation service, a temperature bias nonlinear function solver, and a material relaxation coefficient matrix generator. Each microservice realizes data interaction through an asynchronous message queue. Among them, the current stress accumulation service calculates the time integral effect of the drive current, the temperature bias solver establishes a nonlinear mapping between the thermal distribution and the current stress, and the material relaxation generator quantifies the material aging rate in combination with the experimental database to jointly construct time-varying degradation characteristic parameters to replace the traditional static model.

[0067] Containerized simulation microservice: encapsulates the LSTM network inference module using containerization technology (such as Docker / Kubernetes) to achieve elastic scaling of computing resources. The microservice receives the filtered current stress time series data stream, divides it through a sliding time window and adjusts the adaptive learning rate, and online fine-tunes the network weights to generate an adaptive time series prediction model to support the dynamic simulation of virtual aging scenarios.

[0068] Multi-objective Loss Function Solver: An integrated multi-dimensional optimization module that incorporates the ΔE color difference threshold, afterimage suppression metrics, and power consumption constraints. It evaluates the non-dominated sorting results of the candidate solution set through the Pareto front analysis model. Combining the two-way communication of the backpropagation microservice and the dynamic aging model, it synchronously updates the model coupling weights to balance the conflicting objectives of display quality and energy consumption.

[0069] Edge-Cloud Collaborative Architecture: The edge side deploys a microservice for register mapping of the display driver chip to directly execute the optimized drive parameter adjustment instructions; the cloud stores the historical versions of the compensation coefficient matrix and the evolution data of the aging parameters. Through the subscription-publishing mode, it synchronizes the brightness attenuation feedback data to form a hierarchical optimization system for local real-time control and global model update.

[0070] Incremental Learning and Federated Learning Microservices: The incremental learning microservice updates the weights of the LSTM network through real-time brightness attenuation data to adapt to the change in the material aging rate; the federated learning framework aggregates the desensitized data of multiple devices and uses the weighted average algorithm to fuse the cross-device feature distributions to improve the generalization ability of the model. The two coordinate the update rhythm through the time window synchronization mechanism to resolve the contradiction between data privacy and model generalization.

[0071] Service Mesh and Version Control Middleware: The service mesh manages the communication routing and load balancing between microservices, and attaches a transaction identifier to ensure the temporal consistency of data transfer; the version control middleware records parameter change events to the distributed transaction log, supports state rollback and gray release, and prevents display parameter jumps caused by partial updates.

[0072] Asynchronous Message Queue: It uses message middleware (such as RabbitMQ / Kafka) to achieve decoupled communication between microservices. When the filtered current stress data reaches the time window threshold, it triggers a queue push event to the material relaxation coefficient generator to ensure the real-time and reliability of the data pipeline and avoid service blocking.

[0073] Consistent Hashing Algorithm: A virtual node ring is constructed in the distributed simulation microservice, and the computing tasks are mapped to specific nodes according to the spatio-temporal tags (such as pixel coordinates, timestamps) of the compensation parameters. Through the redundant replica mechanism, it maintains the spatio-temporal correlation of the parameters and prevents data loss caused by node failures.

[0074] Differential Privacy Technology: During the multi-objective optimization process, Gaussian noise perturbation is injected into the gradient data output by the backpropagation microservice, and the noise intensity is controlled by the privacy budget. Local desensitization is implemented on the feature dimensions of the gradient tensor to block the leakage path of sensitive information (such as the aging trajectory of specific pixels) and meet the privacy compliance requirements.

[0075] Description of the Synergistic Relationship of the Technical Features of the Present Invention: Data flow closed-loop: Distributed sensors → Edge preprocessing → Dynamic aging model → Simulation microservices → Multi-objective optimization → Driving parameter adjustment → Online learning feedback, forming a complete closed-loop from data collection to model iteration.

[0076] Dynamic adaptation mechanism: Asynchronous communication of microservice clusters and containerized elastic scheduling support real-time update of aging model parameters; Dual-channel optimization of federated learning and incremental learning balances the requirements of model personalization and generality.

[0077] Risk-resistant design: The state consistency maintenance and version rollback mechanism of the service mesh ensure the stability of the system during frequent updates; The combination of differential privacy and edge computing takes into account both data utility and privacy security.

[0078] Through the combination of the above technical features, the present invention realizes the real-time tracking and compensation of the non-linear aging trajectory of organic light-emitting units under the combined action of current stress, temperature bias and material relaxation effect, solves the cumulative deviation problem caused by traditional static models, and significantly improves the life and visual consistency of display panels.

[0079] The present invention solves the problem that traditional static aging models cannot adapt to non-linear time-varying degradation characteristics through dynamic data collection, real-time model iteration and closed-loop feedback mechanism. The specific technical solutions are as follows: The present invention uses distributed photosensitive sensors, programmable current sampling modules and temperature sensor networks to collect ambient light, current stress and thermal distribution data in real time, and the central data bus calls the edge computing unit for multi-source data preprocessing. The filtered current stress time series data and the corrected thermal distribution data are input into the dynamic aging model microservice cluster, and the cumulative calculation of current stress, the solution of non-linear temperature bias function and the generation of material relaxation coefficient are split into independent microservices. The data pipeline coupling between services is realized through an asynchronous message queue, and combined with the current density-mobility decay curve in the experimental database, the mobility decay error is quantified, and the time-varying degradation characteristic parameters are constructed. This parameter reflects the combined action of current stress, temperature bias and material relaxation effect in real time, replacing the traditional static parameter mapping relationship.

[0080] Based on the time-varying degradation parameters, the containerized simulation microservice constructs a virtual aging scenario, and uses the LSTM network to fine-tune online to generate an adaptive time series correlation prediction model. The incremental learning algorithm injects organic material response delay data in real time to update the model weights to adapt to the change of aging rate. The distributed simulation microservice splits the compensation strategy into parallel tasks of gamma correction weight, sub-pixel light emission duration and TFT gate voltage adjustment, and uses the consistent hashing algorithm to maintain the spatio-temporal correlation of compensation parameters. The tasks are assigned to distributed nodes through the service mesh to generate a global compensation coefficient matrix, solving the spatio-temporal deviation accumulation problem of the static compensation strategy.

[0081] The multi-objective loss function solver combines ΔE color difference, afterimage suppression, and power consumption constraints to construct a Pareto front analysis model, and synchronously updates the coupling weights through bidirectional communication between the backpropagation microservice and the dynamic aging model. The edge side executes optimization-driven parameter adjustment, and the cloud historical database records the snapshot of the compensation matrix version, forming an aging parameter evolution link. The online learning microservice obtains brightness attenuation feedback data based on the subscription-publish mode, triggers the federated learning framework to aggregate cross-device features, and updates the parameters of the global dynamic aging model. The service mesh and version control middleware maintain state consistency, prevent parameter jumps through the gray release and rollback mechanisms, and achieve continuous dynamic adaptation of the compensation coefficient and the actual attenuation amount.

Claims

1. An adaptive adjustment method for an active matrix organic light emitting display, characterized in that: include: Collect ambient light intensity and color temperature data by deploying distributed photosensitive sensor nodes, and transmit the ambient light intensity and color temperature data to the dynamic aging model microservice cluster through a central data bus; A programmable current sampling module is used to capture the current stress parameter fluctuation in the driving circuit, and the thermal distribution data collected by the temperature sensor network is obtained. The current stress parameter fluctuation and thermal distribution data are input into the dynamic aging model microservice cluster for parameter coupling calculation, and the time-varying degradation characteristic parameters are output; Based on the time-varying degradation characteristic parameters output by the dynamic aging model microservice cluster, a virtual aging scenario is constructed through containerized simulation microservices to generate a compensation coefficient matrix; Input the compensation coefficient matrix into a multi-objective loss function solver, perform parameter optimization in combination with a preset ΔE color difference threshold and power consumption constraint, and output optimized display driving parameters; The optimized display driver parameters are adjusted in real time through the edge and cloud collaborative architecture, and the compensation coefficient matrix is ​​synchronized to the online learning microservice, and an incremental model update is triggered based on the optimized display driver parameters and the compensation coefficient matrix.

2. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: The data collection process of the distributed photosensitive sensor nodes includes: Each sensor node independently collects ambient light parameters through a microservice architecture, and uploads the ambient light parameters to a central data bus through a lightweight API protocol; After receiving multi-source data including ambient light parameters, current stress parameter fluctuations and temperature distribution data, the central data bus calls the edge computing unit to perform abnormal pulse preprocessing on the current stress parameter fluctuations to generate filtered current stress time series data; The temperature sensor network is based on a topology-aware dynamic orchestration strategy, and a heat conduction path model is established according to the preprocessed current stress time series data to correct the spatial resolution deviation of the thermal distribution data.

3. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: The construction of the dynamic aging model microservice cluster includes: Split the current stress accumulation calculation service, temperature bias nonlinear function solver, and material relaxation coefficient matrix generator into independent microservices; A data pipeline coupling is implemented between the current stress accumulation calculation service and the material relaxation coefficient matrix generator through an asynchronous message queue, wherein the filtered current stress time series data triggers the real-time correction of the material relaxation coefficient matrix; The material relaxation coefficient matrix generator calls the current density and mobility attenuation curve pre-stored in the experimental database through RPC, and quantifies the mobility attenuation error in combination with the corrected thermal distribution data.

4. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: The operation process of the containerized simulation microservice includes: Deploy the LSTM network as a containerized inference microservice, perform online weight fine-tuning after receiving the filtered current stress time series data stream, and generate an adaptive time series association prediction model; Injecting the organic material response delay data into the training set of the adaptive timing association prediction model using an incremental learning algorithm to update the weight parameters of the prediction model; The compensation strategy is calculated in parallel based on the adaptive time series correlation prediction model through distributed simulation microservices, and the consistent hashing algorithm is used to maintain the temporal and spatial correlation of the compensation parameters.

5. The adaptive adjustment method of active matrix organic light emitting display according to claim 3, characterized in that: The correction process of the material relaxation coefficient matrix includes: updating the temperature bias parameter of the intrinsic relaxation effect of the material according to the corrected thermal distribution data output by the temperature sensor network; Invoke the current density and mobility decay curve pre-stored in the experimental database, and generate a mobility decay compensation gradient in combination with the output result of the current stress accumulation calculation service; The mobility attenuation compensation gradient is nonlinearly coupled with the current stress accumulation calculation result, and a corrected relaxation coefficient matrix is ​​output to the dynamic aging model microservice cluster.

6. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: The construction process of the virtual aging scene includes: Generate an initial strategy for pixel-level aging compensation based on the time-varying degradation parameters output by the dynamic aging model microservice cluster; The initial strategy of pixel-level aging compensation is split into three parallel microservice tasks: gamma correction weight calculation, sub-pixel light emission duration optimization, and TFT gate voltage gradient adjustment; The three parallel microservice tasks are distributed to distributed computing nodes for collaborative execution through a service grid, and transaction logs are used to record simulation status data of the virtual aging scenario.

7. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: The optimization process of the multi-objective loss function solver includes: Receive the preset ΔE color difference threshold, afterimage suppression index and power consumption constraint conditions, and build a Pareto frontier analysis model; Synchronously updating coupling weight parameters of the Pareto frontier analysis model and the dynamic aging model through a bidirectional communication channel between the backpropagation microservice and the dynamic aging model microservice cluster; Differential privacy technology is used to desensitize the gradient shared data output by the back-propagation microservice.

8. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: The execution process of the edge and cloud collaborative architecture includes: Deploy a display driver chip register mapping microservice at the edge to execute the optimized display driver parameter adjustment instructions output by the multi-objective loss function solver in real time; Synchronize the version snapshot of the compensation coefficient matrix to the cloud historical database to establish an aging parameter version control link based on timestamp; The brightness attenuation feedback data of the actual display panel is obtained through the subscription and publishing mode to trigger the incremental model update operation of the online learning microservice.

9. The adaptive adjustment method of active matrix organic light emitting display according to claim 8, characterized in that: The incremental model updating process includes: After receiving the brightness attenuation feedback data, the online learning microservice generates a history database update instruction and attaches a version identifier of the compensation coefficient matrix; Routing the update instruction to the dynamic aging model microservice cluster through the service grid to trigger parameter recalibration of the current stress accumulation calculation service and the material relaxation coefficient matrix generator; A federated learning framework is used to aggregate the brightness attenuation feedback data and historical aging parameters across devices to update the weight coefficients of the global dynamic aging model.

10. The adaptive adjustment method of active matrix organic light emitting display according to claim 1, characterized in that: Also includes: The distributed sensor microservice and edge preprocessing unit reduce the computing load of the central node through asynchronous data processing to support the dynamic expansion of ultra-high-resolution screens; The containerized simulation microservice dynamically allocates computing resources based on an elastic resource scheduling strategy to accelerate the efficiency of verifying the compensation strategy for the virtual aging scenario; The incremental learning and federated learning microservices synchronously optimize the time-varying adaptability parameters and cross-device generalization parameters of the dynamic aging model through a closed-loop data link; The service grid and version control middleware are based on simulated state data of transaction logs.

Images