Artificial Intelligence-Based Automatic Equipment Control Methods and Systems

Abstract

This invention discloses an artificial intelligence-based automatic control method and system for equipment, belonging to the field of intelligent control technology. The method includes: acquiring display task data of an LED screen and adaptively associating it with the environment to generate multiple environmental display tasks; establishing a display control decision space based on these tasks; constructing a display control evaluation channel; performing optimization analysis on the decision space to generate a display control optimization space; performing mutation optimization on the optimization space through a display control fitness function to generate a display control optimization strategy; and adaptively controlling the LED screen. This invention solves the technical problem that existing LED screen display control methods are unable to adapt to diverse display task requirements under different environmental conditions and cannot balance display effect and energy consumption. It achieves the technical effect of dynamically optimizing the display quality and energy consumption of the LED screen through AI-based adaptive control, realizing the best balance between display effect and energy efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and more specifically to an automatic control method and system for equipment based on artificial intelligence. Background Technology

[0002] With the widespread application of LED display technology, especially in advertising, information dissemination, and traffic guidance, optimizing the display effect of LED screens under different environmental conditions has become a crucial technical issue. Traditional LED screen control methods rely primarily on fixed display parameters, such as brightness and contrast, and cannot adaptively adjust to real-time environmental changes. This results in display effects that cannot adapt to different lighting, temperature, and humidity conditions. Furthermore, existing control methods often lack effective energy efficiency management, frequently leading to energy waste while improving display quality. Summary of the Invention

[0003] This application provides an artificial intelligence-based automatic control method and system for devices, which addresses the technical problem that existing LED screen display control methods are unable to adapt to the diverse display task requirements under different environmental conditions and cannot balance display effect and energy consumption control.

[0004] The first aspect of this application provides an artificial intelligence-based automatic device control method, the method comprising: obtaining display task data of an LED screen, and adaptively associating the display task data with the environment to generate M environmental display tasks, where M is a positive integer greater than 1; performing control analysis on the LED screen according to the M environmental display tasks to establish M display control decision spaces; constructing a display control evaluation channel according to a display control evaluation factor; performing optimization analysis on the M display control decision spaces according to the display control evaluation channel to generate M display control optimization spaces; performing mutation optimization on the M display control optimization spaces according to a display control fitness function to generate M display control optimization strategies; and performing adaptive control on the LED screen based on the M environmental display tasks and the M display control optimization strategies.

[0005] A second aspect of this application provides an artificial intelligence-based automatic control system for devices, the system comprising: an environment display task generation module, which obtains display task data of an LED screen and performs adaptive environment association on the display task data to generate M environment display tasks, where M is a positive integer greater than 1; a display control parsing module, which performs control parsing on the LED screen according to the M environment display tasks to establish M display control decision spaces; an evaluation channel construction module, which constructs a display control evaluation channel according to display control evaluation factors; a display control optimization module, which performs optimization analysis on the M display control decision spaces according to the display control evaluation channels to generate M display control optimization spaces; a mutation optimization module, which performs mutation optimization on the M display control optimization spaces according to a display control fitness function to generate M display control optimization strategies; and an adaptive control module, which performs adaptive control on the LED screen based on the M environment display tasks and the M display control optimization strategies.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The AI-based automatic control method and system for devices provided in this application pertain to the field of intelligent control technology. By acquiring LED screen display task data and adaptively associating it with the environment, multiple environmental display tasks are generated. Based on these tasks, a display control decision space is established, and optimization analysis is performed using display control evaluation factors. Adaptive display control of the LED screen is then implemented based on the environmental display tasks and the optimal control strategy. This solves the technical problem that existing LED screen display control methods struggle to adapt to diverse display task requirements under different environmental conditions and cannot balance display quality and energy consumption. It achieves the technical effect of dynamically optimizing the display quality and energy consumption of the LED screen through AI-based adaptive control, thus realizing the optimal balance between display quality and energy efficiency. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic flowchart of an artificial intelligence-based automatic device control method provided in an embodiment of this application.

[0010] Figure 2 This is a schematic diagram of the structure of an artificial intelligence-based automatic control system for devices provided in an embodiment of this application.

[0011] Figure labeling: Environment display task generation module 11, display control parsing module 12, evaluation channel construction module 13, display control optimization module 14, mutation optimization module 15, adaptive control module 16. Detailed Implementation

[0012] This application provides an artificial intelligence-based automatic control method and system for devices, which addresses the technical problem that existing LED screen display control methods are unable to adapt to the diverse display task requirements under different environmental conditions and cannot balance display effect and energy consumption control.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, this application provides an artificial intelligence-based automatic control method for equipment, the method comprising:

[0016] P10: Obtain the display task data of the LED screen, and perform adaptive environment association on the display task data to generate M environment display tasks, where M is a positive integer greater than 1.

[0017] Furthermore, step P10 in this embodiment of the application also includes:

[0018] P11: Divide the display control time zone to obtain multiple predetermined display time zones; P12: Perform environmental prediction on the LED screen based on the multiple predetermined display time zones to obtain a multi-time zone display environment prediction set; P13: Perform clustering and fusion on the multi-time zone display environment prediction set to generate M display environment feature vectors; P14: Perform association mapping on the display task data according to the M display environment feature vectors to generate the M environment display tasks.

[0019] It should be understood that obtaining display task data for the LED screen may include various information such as the content to be displayed (e.g., text, images, videos), display time, and display priority. This display task data is then adaptively correlated with the environment to generate corresponding display tasks for different environments. For example, in a bright light environment, a high-brightness display task may be generated to ensure the displayed content is clearly visible; in a dark environment, a low-brightness display task with high contrast may be generated to avoid glare and ensure display quality.

[0020] Specifically, the LED screen's display control time zone is first divided to obtain multiple predetermined display time zones. This division is defined based on different environmental conditions, such as day-night cycles, seasonal changes, and the impact of special events, ensuring that the LED screen's display tasks can flexibly respond to the needs of different time periods. Each time zone corresponds to different display characteristics, allowing for dynamic adjustments based on real-time environmental data. Specifically, the division of display control time zones can be based on a preset timetable (e.g., by hour or by schedule) or determined by real-time changes in environmental sensing data, thus ensuring that the LED screen's display content remains consistent with changes in the external environment.

[0021] Next, environmental prediction is performed on the LED screen based on the aforementioned multiple predetermined display time zones to obtain a multi-time zone display environment prediction set. Environmental prediction utilizes existing environmental data, historical records, and relevant algorithm models to estimate the environmental conditions of the LED screen within each predetermined display time zone. For example, by analyzing historical meteorological data and ambient light sensor data, it is predicted that the ambient light intensity in different predetermined display time zones may be strong, weak, or dark, and that the colors of surrounding objects may be predominantly warm or cool tones, forming a set containing environmental prediction information for multiple time zones, providing a data foundation for subsequent feature vector generation.

[0022] After obtaining environmental prediction sets from multiple time zones, the process enters the clustering and fusion stage. These environmental prediction data undergo cluster analysis to generate M display environment feature vectors. This process uses clustering algorithms to categorize similar environmental features, with each category corresponding to a feature vector. These feature vectors concisely express the display characteristics under each environment. For example, time zone environmental predictions with similar light intensities and surrounding object colors are grouped together to form a single display environment feature vector. Each feature vector contains key features of that time zone environment, such as average light intensity and dominant color distribution. These feature vectors can more concisely and efficiently represent different environmental scenarios, providing clear environmental feature guidance for subsequent display task data association.

[0023] Finally, based on the generated M display environment feature vectors, the display task data is correlated and mapped to generate M environment display tasks. Correlation mapping involves matching the display task data with the corresponding environment feature vectors, binding each display task to specific environmental features. For example, if a display task is to play a video, based on its associated display environment feature vector, it is determined that in a bright light environment, the video's brightness and contrast should be increased, while in a dark light environment, brightness should be reduced and color saturation optimized. This generates environment display tasks tailored to different environmental characteristics, achieving adaptive optimization of the LED screen's display effect to meet the optimal display requirements in different environments. This method significantly improves the LED screen's adaptability, enabling it to flexibly respond to various environmental changes.

[0024] P20: Based on the M environmental display tasks, control and analyze the LED screen to establish M display control decision spaces.

[0025] Furthermore, step P20 in this embodiment of the application also includes:

[0026] P21: Extract the m-th environmental display task based on the M environmental display tasks, where m is a positive integer greater than 1, and 1 ≤ m ≤ M; P22: Perform control retrieval on the LED screen based on the m-th environmental display task to obtain the m-th display control sample set; P23: Perform trigger feature parsing on the m-th display control sample set to generate the m-th display control trigger domain; P24: Based on the m-th environmental display task, perform control parameter parsing on the LED screen based on the m-th display control trigger domain to generate the m-th display control decision space.

[0027] Optionally, based on the generated M environmental display tasks, the LED screen is controlled and analyzed. Based on the analysis of each environmental display task, the corresponding control parameter range, control strategy selection, etc. are determined, and M display control decision spaces are established.

[0028] First, based on the aforementioned M environment display tasks, the m-th environment display task is extracted one by one (where m is a positive integer greater than 1, 1 ≤ m ≤ M). Each environment display task represents a specific display requirement. These tasks, through adaptive analysis and association with environmental conditions, define different display targets. The key to the extraction process is to extract the multiple environment display tasks generated through the previous environment association one by one, so as to perform targeted control and analysis for each specific environment display task. For example, if M is 5, representing 5 display tasks under different environmental characteristics, then the 2nd, 3rd, 4th, 5th, and so on environment display tasks are extracted sequentially to provide specific objects for subsequent control retrieval and analysis.

[0029] Next, a control retrieval is performed on the LED screen based on the m-th environment display task to obtain the m-th display control sample set. The control retrieval is based on the requirements of the m-th environment display task, searching for relevant control samples in existing control data or historical records. These control samples may include brightness adjustment values, color calibration parameters, and display mode selections previously used by the LED screen under similar environments. For example, if the m-th environment display task is a display task under specific light intensity and color backgrounds, the retrieved sample set may include samples of control parameters such as 50% brightness and 6500K color temperature previously used by the LED screen under similar lighting and color backgrounds, providing a data foundation for subsequent trigger feature analysis.

[0030] Then, trigger feature analysis is performed on the m-th display control sample set to generate the m-th display control trigger domain. Trigger feature analysis analyzes which factors or parameters in the control sample set are key factors that trigger specific display controls on the LED screen. For example, analysis of the sample set reveals that under specific light intensities, when the ambient light intensity changes beyond a certain threshold, the LED screen needs to adjust its brightness; or when the contrast between the displayed content's color and the background color is below a certain value, color saturation needs to be adjusted, etc. The set of these key factors constitutes the m-th display control trigger domain, clarifying under what conditions control and adjustment of the LED screen are required.

[0031] Finally, based on the m-th environmental display task and the corresponding display control trigger domain, the control parameters of the LED screen are analyzed to generate the m-th display control decision space. Control parameter analysis involves determining the range and combination of control parameters that the LED screen might use under that task, based on key factors in the trigger domain and the specific requirements of the m-th environmental display task. For example, if the trigger domain specifies that brightness adjustment is needed under specific lighting changes, then during control parameter analysis, a brightness adjustment range, such as from 30% to 70%, and possible adjustment steps will be determined. Simultaneously, other relevant parameters will be considered, such as whether color calibration parameters need to be adjusted synchronously. This ultimately forms a decision space containing multiple possible combinations of control parameters, providing diverse options for subsequent control optimization to achieve precise control of the LED screen under different environmental display tasks.

[0032] These steps enable the generation of an independent control decision space for each environmental display task, thereby achieving refined display control.

[0033] P30: Establish a display control evaluation channel based on display control evaluation factors. These evaluation factors include visual comfort, response latency, and image quality.

[0034] Furthermore, step P30 in this embodiment of the application also includes:

[0035] P31: Based on the display control evaluation factors, retrieve evaluation records to obtain historical sets of visual comfort evaluation, response delay evaluation, and image quality evaluation; P32: Train a visual comfort evaluation model based on the historical set of visual comfort evaluation; P33: Train a response delay evaluation model based on the historical set of response delay evaluation; P34: Train an image quality evaluation model based on the historical set of image quality evaluation; P35: Perform distillation learning based on the visual comfort evaluation model, the response delay evaluation model, and the image quality evaluation model to generate the display control evaluation channel.

[0036] Specifically, a display control evaluation channel is established based on display control evaluation factors. These factors can be diverse, such as display clarity, color accuracy, power consumption, and display stability. These factors allow for the evaluation of different control schemes, determining whether they meet the expected display effects and performance requirements. Establishing an evaluation channel is equivalent to creating an evaluation system that enables subsequent quantitative evaluation of various schemes within the control decision space.

[0037] For example, the evaluation records are first retrieved based on display control evaluation factors to obtain historical sets of visual comfort evaluations, response latency evaluations, and image quality evaluations. For instance, by retrieving historical data, evaluation records related to visual comfort, response latency, and image quality are collected. These records can come from user feedback, system logs, test reports, etc., providing a data foundation for subsequent model training. For example, the historical set of visual comfort evaluations may contain user ratings of comfort under different display parameter settings; the historical set of response latency evaluations may record response times under different control commands; and the historical set of image quality evaluations may contain ratings for image sharpness, color accuracy, etc.

[0038] Next, a visual comfort evaluation model is trained based on the historical set of visual comfort evaluations. Visual comfort is a user's subjective feeling when viewing displayed content, and is usually related to factors such as brightness, contrast, and flicker frequency. Using machine learning algorithms, such as linear regression, decision trees, or neural networks, a model capable of predicting visual comfort under different display parameter settings is trained using data from the historical visual comfort evaluation set. For example, the model can learn that users have higher comfort scores at specific brightness and contrast settings, thus providing a basis for subsequent evaluations.

[0039] Simultaneously, a response latency evaluation model is trained based on the historical response latency evaluation dataset. Response latency refers to the time delay between user input or environmental changes and the change displayed on the LED screen. This factor is crucial to user experience, especially in dynamic or interactive displays. By analyzing data from the historical response latency evaluation dataset and utilizing appropriate machine learning algorithms, a model capable of evaluating response latency under different control schemes is trained. For example, the model can learn to achieve lower response latency under specific hardware configurations and software optimizations, thus providing a reference for optimizing control strategies.

[0040] An image quality assessment model is trained based on a historical dataset of image quality evaluations. Image quality primarily includes visual quality parameters such as sharpness, color reproduction, and contrast. By analyzing historical evaluation data, image quality can be quantified, and a corresponding evaluation model can be constructed. This model can predict the image quality of LED screens under various environmental conditions and provide a scientific basis for optimizing display effects.

[0041] Finally, using a distillation learning method, the visual comfort evaluation model, response latency evaluation model, and image quality evaluation model are distilled to generate a display control evaluation channel. Distillation learning is a knowledge transfer technique that improves the generalization ability and efficiency of a model by fusing knowledge from multiple models into a comprehensive model. In this step, the knowledge from the three evaluation models—visual comfort, response latency, and image quality—is fused to generate a comprehensive display control evaluation channel. This evaluation channel can comprehensively consider multiple evaluation factors, fully assess different display control schemes, and provide accurate evaluation criteria for subsequent control strategy optimization.

[0042] These steps enable precise control of display tasks based on evaluation factors, as well as dynamic adjustment of display settings to improve visual comfort, reduce response latency, and optimize image quality, thereby enhancing user experience and achieving intelligent display control.

[0043] P40: Based on the display control evaluation channel, perform optimization analysis on the M display control decision spaces to generate M display control optimization spaces.

[0044] Furthermore, step P40 in this embodiment of the application also includes:

[0045] P41: Simulate control of the LED screen according to each display control decision in the m-th display control decision space to obtain multiple display control simulation sets; P42: Input the multiple display control simulation sets into the display control evaluation channel to obtain multiple display control evaluation sets; P43: Set display control evaluation constraints based on the display control evaluation factors; P44: Based on the multiple display control evaluation sets, perform evaluation constraint optimization on the m-th display control decision space according to the display control evaluation constraints to generate the m-th display control optimization space.

[0046] It should be understood that by utilizing the established display control evaluation channel, optimization analysis is performed on M display control decision spaces to generate M display control optimization spaces. This process involves screening and optimizing various control schemes within each decision space through the evaluation channel to identify the superior control scheme and form an optimization space. For example, within a decision space, there may be multiple brightness adjustment schemes. The evaluation channel assesses these schemes, identifying the brightness adjustment scheme that achieves a good balance between power consumption and display clarity, and incorporating it into the optimization space.

[0047] Specifically, for each display control decision within the m-th display control decision space, simulated control operations on the LED screen are executed, resulting in multiple display control simulation sets. This simulation control process is based on various possible combinations of control parameters encompassed within the decision space. By simulating the application effects of these parameters in a real display environment, corresponding display data is collected. This display data constitutes the display control simulation sets, providing fundamental material for subsequent evaluation and analysis. For example, under a specific display control decision, simulation control might involve setting specific parameters such as brightness, contrast, and color saturation, and recording the display effect data of the LED screen under these parameter settings.

[0048] Subsequently, these display control simulation sets are input into the previously established display control evaluation channel. As a comprehensive evaluation system, the display control evaluation channel can objectively and comprehensively evaluate the input simulation sets based on pre-defined evaluation factors, such as visual comfort, response latency, and image quality. Through this evaluation process, multiple display control evaluation sets are obtained, each containing the scores or evaluation results of the corresponding simulation set on various evaluation factors. These evaluation sets provide quantitative basis for subsequent optimization analysis, enabling the system to clearly define the performance of each control decision across different evaluation dimensions.

[0049] Next, display control evaluation constraints are set based on the display control evaluation factors. This step ensures that the optimization process focuses on control decisions that meet specific performance requirements. Evaluation constraints are set according to actual application needs and display effect standards. For example, visual comfort scores can be set to not fall below a certain threshold, response latency cannot exceed a certain time, and image quality scores must reach a certain level. These constraints provide clear optimization directions and boundaries for subsequent optimization analysis, making the generated optimization space more practical and targeted.

[0050] Finally, based on multiple display control evaluation sets, and according to the display control evaluation constraints, the m-th display control decision space is optimized by evaluating the constraints. This process involves selecting the decisions that perform better on various evaluation factors from numerous control decisions while satisfying the evaluation constraints, thus generating the m-th display control optimization space. The control decisions within the optimization space are more aligned with the optimization objective than the original decision space, resulting in better display performance for the LED screen. This process can use optimization algorithms, such as genetic algorithms or other optimization methods, to gradually narrow down the decision space by optimizing the evaluation values ​​under constraints, ultimately generating a display control optimization space that meets the constraints. This optimization space contains the combination of control strategies that provides the best display performance under given environmental conditions.

[0051] These steps enable optimized analysis of the display control decision space, allowing for full evaluation and optimization of display control decisions in each environment. This improves the display performance of the LED screen and ensures optimal visual experience, response speed, and image quality under different environmental conditions.

[0052] P50: Based on the display control fitness function, perform mutation optimization on the M display control optimization spaces to generate M display control optimization strategies.

[0053] Furthermore, step P50 in this embodiment of the application also includes:

[0054] P50a: Assign weights to the display control evaluation factors to generate the display control fitness function. P51: Perform fitness analysis on the m-th display control optimization space based on the display control fitness function to obtain the display control fitness sequence. P52: Perform mutation and expansion optimization on the m-th display control optimization space based on the display control fitness sequence to generate the m-th control expanded optimization space. P53: Perform energy consumption minimization optimization on the m-th control expanded optimization space to generate the m-th display control optimization strategy.

[0055] Optionally, based on the display control fitness function, mutation optimization is performed on the M display control optimization spaces. The fitness function is a function that measures how well a control scheme adapts to the environment, and it can be used to further optimize the control scheme within the optimization space. Mutation optimization is an optimization algorithm that, by introducing mutation operations, can explore new control schemes, avoid getting trapped in local optima, and thus find a control strategy that is more adaptable to environmental changes.

[0056] Specifically, the display control evaluation factors are first weighted to generate the display control fitness function. This process assigns different weights to evaluation factors such as visual comfort, response latency, and image quality based on actual application needs and the importance of display effects. For example, if visual comfort is more critical in a specific application scenario, it is given a higher weight; while if response latency has a significant impact on the application, its weight is increased accordingly. In this way, the generated display control fitness function can comprehensively reflect the importance of different evaluation factors to display control decisions, providing a quantitative basis for subsequent fitness analysis and mutation optimization. For example, assume that the display control evaluation factors include visual comfort V, response latency R, and image quality I, and their weights are w respectively. V ,w R ,w I Then the fitness function F can be expressed as: F = w V ·V+w R ·R+w I ·I; where the weight w Vw R w I Satisfy w V +w R +w I =1, and w V ,w R ,w I ≥0. This function comprehensively considers three evaluation factors: visual comfort, response latency, and image quality, reflecting their importance in fitness evaluation through weight allocation. In practical applications, the weight allocation can be adjusted according to specific needs and scenarios to adapt to different optimization objectives.

[0057] Next, fitness analysis is performed on the m-th display control optimization space according to the display control fitness function to obtain the display control fitness sequence. Fitness analysis is performed by substituting each control decision in the optimization space into the fitness function and calculating its fitness value, thus forming a fitness sequence. This sequence reflects the degree of fitness of each control decision after comprehensively considering all evaluation factors, that is, its applicability and performance in the current environment. For example, a control decision may score highly in visual comfort but score poorly in response delay; after calculation using the fitness function, its fitness value can comprehensively reflect this characteristic.

[0058] Subsequently, based on the display control fitness sequence, the m-th display control optimization space is expanded through mutation to generate the m-th expanded control optimization space. Mutation expansion optimization is an optimization algorithm that adjusts and expands the control decisions within the optimization space by introducing mutation operations. Mutation operations can randomly change certain control parameters or introduce new parameter combinations to explore new control decisions. This process aims to avoid the optimization process getting trapped in local optima. By generating new control decisions through mutation operations, the optimization space is further expanded, providing more possibilities for finding better control strategies. For example, through mutation operations, a new control decision that achieves a better balance between energy consumption and display performance may be discovered.

[0059] Finally, based on the expanded optimization space of the m-th control, energy consumption minimization optimization is performed to generate the m-th display control optimization strategy. Energy consumption minimization optimization involves finding the control decisions with the lowest energy consumption while meeting display effect requirements within the expanded optimization space. This process evaluates the energy consumption of each control decision and combines it with the display control fitness function to select the control decision that meets both display effect requirements and has the lowest energy consumption, thereby generating the final display control optimization strategy. For example, through energy consumption minimization optimization, a control decision with the lowest energy consumption and a high fitness value under specific brightness and color settings might be selected as the final optimization strategy.

[0060] Through these steps, the system can perform intelligent control optimization based on multiple evaluation factors. By using mutation expansion and energy consumption minimization optimization, the generated display control optimization strategy can minimize energy consumption while maintaining high display quality, achieving more efficient and intelligent LED screen display control. This greatly improves the system's adaptability and optimization capabilities, meeting the dual requirements of display effect and energy consumption in complex environments.

[0061] Furthermore, step P52 in this embodiment of the application also includes:

[0062] P52-1: Based on the display control fitness sequence, perform mutation weight allocation on the m-th display control optimization space to obtain the mutation weight allocation result; P52-2: Based on the mutation weight allocation result, mutate the m-th display control optimization space to generate the m-th display control mutation space; P52-3: Based on the display control fitness function, perform optimization analysis on the m-th display control mutation space according to the predetermined display control fitness to obtain the m-th control extended optimization space.

[0063] It should be understood that a more optimized extended optimization space for the m-th control can be generated by further performing more refined mutation operations and optimization analysis on the m-th display control optimization space.

[0064] First, mutation weights are assigned to the m-th display control optimization space based on the display control fitness sequence to obtain the mutation weight assignment results. This process involves assigning different mutation weights to each control decision in the fitness sequence based on its fitness value. Control decisions with lower fitness values, i.e., those that perform poorly under the current evaluation system, are given higher mutation weights to introduce greater changes through mutation operations and explore new control strategies; while control decisions with higher fitness values ​​are given lower mutation weights to maintain their relatively stable and superior characteristics. For example, a control decision may score low in the fitness sequence because it performs poorly in terms of response latency; therefore, it will be given a higher weight in the mutation weight assignment so that response latency-related parameters can be adjusted in subsequent mutation operations.

[0065] Next, the m-th display control optimization space is mutated based on the mutation weight allocation result to generate the m-th display control mutation space. The mutation operation randomly adjusts or changes the parameters of the control decisions within the optimization space according to the assigned mutation weights. Specifically, for control decisions assigned higher mutation weights, a larger range of parameter adjustments or the introduction of new parameter combinations are performed; while for control decisions with lower mutation weights, smaller fine-tuning is done. In this way, the generated m-th display control mutation space contains more diverse control decisions, providing richer choices for subsequent optimization analysis. For example, for a control decision that performs poorly in terms of energy consumption but has high visual comfort, the mutation operation may adjust its energy-related parameters, such as brightness adjustment strategies, while keeping visual comfort parameters unchanged, thereby generating a new control decision.

[0066] Finally, based on the display control fitness function, the system performs optimization analysis on the m-th display control mutation space according to the predetermined display control fitness, obtaining the m-th extended optimization space. Guided by the fitness function, all control decisions within the mutation space are evaluated, and those control decisions with fitness values ​​meeting predetermined criteria are selected. The predetermined display control fitness is a threshold set according to actual application requirements, used to ensure that the selected control decisions, after comprehensively considering all evaluation factors, can meet certain performance requirements. Through this process, the final m-th extended optimization space not only includes high-quality decisions from the original optimization space, but also introduces new and better control decisions through mutation operations, further improving the quality and diversity of the optimization space. For example, after optimization analysis, some new control decisions may be found to achieve a better balance between energy consumption and display effect, and thus be included in the final extended optimization space.

[0067] Through these steps, precise mutation and expansion of the display control optimization space are achieved, generating a more diverse and efficient display control mutation space. Under the dual effects of mutation weight allocation and optimization analysis, the system can dynamically adjust the display control strategy according to different environments and task requirements, thereby achieving the optimal balance between display performance and energy efficiency.

[0068] P60: Based on the M environmental display tasks, adaptive control is performed on the LED screen according to the M display control optimization strategies.

[0069] Specifically, the display control optimization strategy obtained through a series of complex optimization processes is applied to the actual LED screen control to achieve adaptive control of the LED screen and ensure that its display effect reaches the best state in different environments.

[0070] In this stage, the system selects the corresponding display control optimization strategy based on the needs of each environmental display task and applies these strategies to the LED screen control process in real time. Specifically, each environmental display task is matched with its corresponding display control optimization strategy, and then the display parameters of the LED screen are dynamically adjusted according to the specific parameter settings of these strategies. For example, for an environmental display task that needs to be displayed in a bright light environment, the corresponding display control optimization strategy may include parameters such as increasing brightness and adjusting color saturation to ensure that the displayed content is clearly visible and the colors are accurate under bright light; while for a task that needs to be displayed in a dark light environment, the strategy may include reducing brightness and optimizing contrast to avoid the screen being too bright and dazzling, while ensuring the readability of the displayed content.

[0071] In practice, the system selects the most suitable strategy from M display control optimization strategies based on the characteristics of the display task, such as ambient light intensity and the type of displayed content. Then, according to the specific parameters in the strategy, such as brightness adjustment values, color calibration parameters, and display mode selection, the corresponding hardware parameters of the LED screen are adjusted through control circuits or software algorithms. This process is performed in real time, enabling rapid responses to changes in the environment and different display tasks, ensuring that the LED screen is always in an optimal display state.

[0072] The key to adaptive control lies in the system's ability to perceive environmental changes in real time and quickly adjust its display strategy. This process not only considers optimizing the display effect but also includes controlling energy efficiency, thereby avoiding unnecessary energy consumption and improving the system's intelligence.

[0073] Through these steps, the system can dynamically adjust the display settings of the LED screen according to different environmental changes and display task requirements, so as to achieve the optimal display effect, response speed and energy efficiency, and ensure that the LED screen can maintain the best working state in various complex environments.

[0074] In summary, the embodiments of this application have at least the following technical effects:

[0075] This application achieves dynamic adjustment and optimization of LED screen display parameters through the construction of an adaptive environment association and control decision space. By employing multi-dimensional display control evaluation factors and optimization analysis, the display control strategy is optimized, improving display performance and response speed. Simultaneously, through mutation optimization and fitness functions, energy consumption is reduced, and energy efficiency is optimized. Finally, through intelligent adjustment based on environmental data and task requirements, the display performance and energy efficiency of the LED screen are improved in different environments.

[0076] It achieves the technical effect of dynamically optimizing the display quality and energy consumption of LED screens through AI-based adaptive control, thus realizing the best balance between display effect and energy efficiency.

[0077] Example 2, based on the same inventive concept as the artificial intelligence-based automatic device control method in the foregoing examples, such as... Figure 2 As shown, this application provides an artificial intelligence-based automatic control system for equipment. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0078] An environment display task generation module 11 is used to obtain display task data of the LED screen and perform adaptive environment association on the display task data to generate M environment display tasks, where M is a positive integer greater than 1.

[0079] The display control analysis module 12 is used to control and analyze the LED screen according to the M environmental display tasks, and establish M display control decision spaces.

[0080] The evaluation channel construction module 13 is used to construct a display control evaluation channel based on display control evaluation factors. These display control evaluation factors include visual comfort, response latency, and image quality.

[0081] Display control optimization module 14 is used to perform optimization analysis on the M display control decision spaces according to the display control evaluation channel, and generate M display control optimization spaces.

[0082] The mutation optimization module 15 is used to perform mutation optimization on the M display control optimization spaces according to the display control fitness function, and generate M display control optimization strategies.

[0083] An adaptive control module 16 is used to adaptively control the LED screen based on the M environmental display tasks and the M display control optimization strategies.

[0084] Furthermore, the environment display task generation module 11 is also used to perform the following steps:

[0085] The display control time zone is divided to obtain multiple predetermined display time zones; the LED screen is subjected to environmental prediction based on the multiple predetermined display time zones to obtain a multi-time zone display environment prediction set; the multi-time zone display environment prediction set is clustered and fused to generate M display environment feature vectors; the display task data is associated and mapped according to the M display environment feature vectors to generate the M environment display tasks.

[0086] Furthermore, the display control parsing module 12 is also used to perform the following steps:

[0087] Extract the m-th environmental display task from the M environmental display tasks, where m is a positive integer greater than 1, and 1 ≤ m ≤ M; perform control retrieval on the LED screen based on the m-th environmental display task to obtain the m-th display control sample set; perform trigger feature parsing on the m-th display control sample set to generate the m-th display control trigger domain; and perform control parameter parsing on the LED screen based on the m-th environmental display task and the m-th display control trigger domain to generate the m-th display control decision space.

[0088] Furthermore, the evaluation channel construction module 13 is also used to perform the following steps:

[0089] Based on the display control evaluation factors, evaluation records are retrieved to obtain historical sets of visual comfort evaluation, response delay evaluation, and image quality evaluation. A visual comfort evaluation model is trained based on the visual comfort evaluation historical set; a response delay evaluation model is trained based on the response delay evaluation historical set; and an image quality evaluation model is trained based on the image quality evaluation historical set. Distillation learning is then performed on the visual comfort evaluation model, the response delay evaluation model, and the image quality evaluation model to generate the display control evaluation channel.

[0090] Furthermore, the display control optimization module 14 is also used to perform the following steps:

[0091] The LED screen is simulated and controlled according to each display control decision in the m-th display control decision space to obtain multiple display control simulation sets; the multiple display control simulation sets are input into the display control evaluation channel to obtain multiple display control evaluation sets; display control evaluation constraints are set based on the display control evaluation factors; based on the multiple display control evaluation sets, the m-th display control decision space is optimized according to the display control evaluation constraints to generate the m-th display control optimization space.

[0092] Furthermore, the mutation optimization module 15 is also used to perform the following steps:

[0093] The display control evaluation factors are weighted to generate the display control fitness function. The fitness function is used to analyze the fitness of the m-th display control optimization space to obtain a display control fitness sequence. Based on the fitness sequence, the m-th display control optimization space is expanded and modified to generate an expanded m-th control optimization space. Energy consumption minimization is performed on the expanded m-th control optimization space to generate the m-th display control optimization strategy.

[0094] Furthermore, the mutation optimization module 15 is also used to perform the following steps:

[0095] Based on the display control fitness sequence, the m-th display control optimization space is subjected to mutation weight allocation to obtain the mutation weight allocation result; based on the mutation weight allocation result, the m-th display control optimization space is mutated to generate the m-th display control mutation space; based on the display control fitness function, the m-th display control mutation space is optimized and analyzed according to the predetermined display control fitness to obtain the m-th control extended optimization space.

[0096] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0097] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0098] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for automatic control of a device based on artificial intelligence, characterized by, The method comprises: obtaining display task data of an LED screen and adaptively associating the display task data with an environment to generate M environment display tasks, M being a positive integer greater than 1; controlling and analyzing the LED screen according to the M environment display tasks to establish M display control decision spaces; building a display control evaluation channel according to a display control evaluation factor; performing optimization analysis on the M display control decision spaces according to the display control evaluation channel to generate M display control optimization spaces; performing mutation optimization on the M display control optimization spaces according to a display control fitness function to generate M display control optimization strategies; adaptively controlling the LED screen based on the M environment display tasks and according to the M display control optimization strategies; wherein the adaptive association of the display task data with the environment to generate M environment display tasks comprises: dividing a display control time zone to obtain a plurality of predetermined display time zones; performing environment prediction on the LED screen based on the plurality of predetermined display time zones to obtain a multi-time-zone display environment prediction set; performing cluster fusion on the multi-time-zone display environment prediction set to generate M display environment feature vectors; performing association mapping on the display task data according to the M display environment feature vectors to generate the M environment display tasks; wherein the mutation optimization on the M display control optimization spaces according to the display control fitness function to generate M display control optimization strategies comprises: performing fitness analysis on the mth display control optimization space according to the display control fitness function to obtain a display control fitness sequence; performing mutation expansion optimization on the mth display control optimization space based on the display control fitness sequence to generate an mth control expansion optimization space; performing energy minimization optimization on the mth control expansion optimization space to generate an mth display control optimization strategy; wherein the mutation expansion optimization on the mth display control optimization space based on the display control fitness sequence to generate an mth control expansion optimization space comprises: performing mutation weight distribution on the mth display control optimization space based on the display control fitness sequence to obtain a mutation weight distribution result; performing mutation on the mth display control optimization space based on the mutation weight distribution result to generate an mth display control mutation space; performing optimization analysis on the mth display control mutation space according to a predetermined display control fitness based on the display control fitness function to obtain the mth control expansion optimization space.

2. The method of claim 1, wherein, The controlling and analyzing of the LED screen according to the M environment display tasks to establish M display control decision spaces comprises: extracting an mth environment display task according to the M environment display tasks, m being a positive integer greater than 1 and 1≤m≤M; performing control retrieval on the LED screen according to the mth environment display task to obtain an mth display control sample set; performing trigger feature analysis on the mth display control sample set to generate an mth display control trigger domain; Based on the mth environment display task, a control parameter of the LED screen is parsed according to the mth display control trigger domain, and an mth display control decision space is generated.

3. The method of claim 1, wherein, According to the display control evaluation factor, a display control evaluation channel is built, including: Based on the display control evaluation factor, an evaluation record retrieval is performed to obtain a visual comfort evaluation history set, a response delay evaluation history set, and an image quality evaluation history set; According to the visual comfort evaluation history set, a visual comfort evaluation model is trained; According to the response delay evaluation history set, a response delay evaluation model is trained; According to the image quality evaluation history set, an image quality evaluation model is trained; According to the visual comfort evaluation model, the response delay evaluation model, and the image quality evaluation model, distillation learning is performed to generate the display control evaluation channel.

4. The method of claim 1, wherein, According to the display control evaluation channel, the M display control decision spaces are optimized and analyzed to generate M display control optimization spaces, including: According to each display control decision in the mth display control decision space, the LED screen is simulated and controlled to obtain a plurality of display control simulation sets; The plurality of display control simulation sets are input into the display control evaluation channel to obtain a plurality of display control evaluation sets; Based on the display control evaluation factor, a display control evaluation constraint is set; Based on the plurality of display control evaluation sets, the mth display control decision space is evaluated and optimized according to the display control evaluation constraint to generate an mth display control optimization space.

5. The method of claim 1, wherein, The display control evaluation factor includes visual comfort, response delay, and image quality.

6. The method of claim 1, wherein, The display control evaluation factor is assigned a weight to generate a display control fitness function.

7. An artificial intelligence-based device automatic control system, characterized by, The system includes: An environment display task generation module, which is used to obtain display task data of an LED screen and perform adaptive environment association on the display task data to generate M environment display tasks, M being a positive integer greater than 1; A display control parsing module, which is used to parse the LED screen according to the M environment display tasks to establish M display control decision spaces; An evaluation channel building module, which is used to build a display control evaluation channel according to a display control evaluation factor; A display control optimization module, which is used to optimize and analyze the M display control decision spaces according to the display control evaluation channel to generate M display control optimization spaces; A mutation optimization module, which is used to mutate and optimize the M display control optimization spaces according to a display control fitness function to generate M display control optimization strategies; An adaptive control module, which is used to perform adaptive control on the LED screen according to the M display control optimization strategies based on the M environment display tasks. The environment display task generation module is further configured to divide a display control time zone, obtain a plurality of predetermined display time zones, perform environment prediction on the LED screen based on the plurality of predetermined display time zones, obtain a multi-time-zone display environment prediction set, perform clustering fusion on the multi-time-zone display environment prediction set, generate M display environment feature vectors, perform associated mapping on the display task data according to the M display environment feature vectors, and generate the M environment display tasks. The mutation optimization module is further configured to perform fitness analysis on the mth display control optimization space according to the display control fitness function, obtain a display control fitness sequence, perform mutation expansion optimization on the mth display control optimization space based on the display control fitness sequence, generate an mth control expansion optimization space, perform energy consumption minimization optimization according to the mth control expansion optimization space, and generate an mth display control optimization strategy. The mutation optimization module is further configured to perform mutation weight distribution on the mth display control optimization space based on the display control fitness sequence, obtain a mutation weight distribution result, perform mutation on the mth display control optimization space based on the mutation weight distribution result, generate an mth display control mutation space, perform optimization analysis on the mth display control mutation space according to a predetermined display control fitness based on the display control fitness function, and obtain the mth control expansion optimization space.

Images