Adaptive control method for fire emergency lighting fixtures

Abstract

The present invention discloses an adaptive control method for fire emergency lighting fixtures, which relates to the technical field of lighting control. The method includes: deploying a group of fire emergency lighting fixtures with ceramic phosphor, using blue laser excitation to generate visible light of a predetermined color, and at the same time achieving ns-level fiber break control through the chromatographic reflection of visible light, and combining a generative adversarial network to predict and identify the attenuation of the light conversion efficiency of the lighting fixtures; generating a three-dimensional model of a building through simulation, mapping the lighting fixture identification to the model, performing fixed-point optimization control analysis of the lighting fixtures, dynamically adjusting the laser parameters, and realizing the adaptive regulation of emergency lighting in the target building. The present invention solves the technical problems that the existing fire emergency lighting system has potential electrical fire hazards, and insufficient anti-interference ability and emergency response ability, and achieves the technical effect of improving the intrinsic safety, anti-interference ability and emergency response ability of the emergency lighting system through the dynamic optimization of optoelectronic separation technology and generative adversarial network.

Description

Technical Field

[0001] The present invention relates to the technical field of lighting control, and specifically relates to an adaptive control method for fire emergency lighting fixtures. Background Art

[0002] Existing fire emergency lighting systems usually provide power and control signals to lighting fixtures through traditional wire connection methods. This design has the following technical defects and safety hazards: on the one hand, since wires are vulnerable to electromagnetic interference, signal transmission may be unstable, affecting the normal startup and lighting effect of the lighting fixtures; on the other hand, the wire power supply method has the risk of short circuit or electric spark, and in flammable and explosive places (such as chemical plants, gas stations) or ancient buildings, it is extremely easy to cause fires or explosions. Moreover, traditional emergency lighting systems still have deficiencies in light source performance prediction and dynamic control, and cannot be intelligently adjusted according to the actual environment and the attenuation of lighting fixture performance, thus affecting the emergency response ability of the overall system. Summary of the Invention

[0003] This application provides an adaptive control method for fire emergency lighting fixtures, which is used to solve the technical problems that the existing fire emergency lighting system has potential electrical fire hazards, and insufficient anti-interference ability and emergency response ability.

[0004] This application provides an adaptive control method for fire emergency lighting fixtures, and the method includes: deploying predetermined fire emergency lighting equipment inside a target building, where the predetermined fire emergency lighting equipment includes a fire emergency lighting fixture group, the fire emergency lighting fixture group includes a number of fire emergency lighting fixtures, and each fire emergency lighting fixture is installed with ceramic phosphor, and the ceramic phosphor can emit visible light of a predetermined color greater than the blue light wavelength under the irradiation of blue laser; combining a generative adversarial network, and respectively predicting the attenuation of the light conversion efficiency of the number of fire emergency lighting fixtures according to the number of working characteristics, the number of working environmental conditions, and the number of element ratios of the number of fire emergency lighting fixtures, and obtaining a number of predicted light conversion efficiencies to identify the number of fire emergency lighting fixtures; combining the layout structure diagram of the target building to simulate and generate a three-dimensional building model, and mapping and rendering the number of identified fire emergency lighting fixtures to the three-dimensional building model to generate a three-dimensional building emergency lighting model; using the three-dimensional building emergency lighting model to perform fixed-point optimization control analysis on the predetermined fire emergency lighting equipment, and outputting a dynamic optimization control plan for adaptive regulation of the emergency lighting of the target building.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] The adaptive control method for fire emergency lighting fixtures provided by this application relates to the field of lighting control technology. By deploying a group of fire emergency lighting fixtures with ceramic phosphor, visible light of a predetermined color is generated by blue light excitation. At the same time, ns-level fiber break control is achieved through the chromatographic reflection of visible light, and a generative adversarial network is combined for predicting light efficiency attenuation. Through building three-dimensional modeling and dynamic optimization control, an adaptive regulation scheme is output to achieve efficient, safe, and intelligent management of emergency lighting, solving the technical problems that the existing fire emergency lighting system has potential electrical fire hazards, and insufficient anti-interference ability and emergency response ability, and achieving the technical effect of improving the inherent safety, anti-interference ability, and emergency response ability of the emergency lighting system through optoelectronic separation technology and dynamic optimization of the generative adversarial network. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0008] Figure 1 It is a schematic flow chart of the adaptive control method for fire emergency lighting fixtures provided by the embodiment of this application;

[0009] Figure 2 It is a schematic flow chart of the adaptive regulation of emergency lighting for the target building in the adaptive control method for fire emergency lighting fixtures provided by the embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0010] This application provides an adaptive control method for fire emergency lighting fixtures to solve the technical problems that the existing fire emergency lighting system has potential electrical fire hazards, and insufficient anti-interference ability and emergency response ability.

[0011] The following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the drawings in the embodiments of this application. Obviously, the described embodiments are only some embodiments of this application, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the description of this application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, 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 comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices.

[0013] Embodiment 1, as Figure 1 shown, the present application provides an adaptive control method for fire emergency lighting fixtures, and the method includes:

[0014] P10: Deploy predetermined fire emergency lighting equipment inside the target building, where the predetermined fire emergency lighting equipment includes a fire emergency lighting fixture group, the fire emergency lighting fixture group includes several fire emergency lighting fixtures, and each fire emergency lighting fixture is installed with ceramic phosphor, and the ceramic phosphor can emit visible light of a predetermined color greater than the blue light wavelength under the irradiation of blue laser.

[0015] Among them, the predetermined fire emergency lighting equipment further includes a photoelectric lighting centralized power supply and an optical fiber module. Among them, the photoelectric lighting centralized power supply is embedded with a laser generator, and the laser generator is used to emit blue laser; the optical fiber module is connected to the fire emergency lighting fixture group, and the blue laser emitted by the laser generator is projected onto the several fire emergency lighting fixtures to emit visible light of a predetermined color greater than the blue light wavelength, where the predetermined color is green, red or white.

[0016] It should be understood that deploying the predetermined fire emergency lighting equipment inside the target building, which includes a fire emergency lighting fixture group, a photoelectric lighting centralized power supply, and an optical fiber module, can achieve efficient and safe emergency lighting functions. The fire emergency lighting fixture group consists of multiple independent fire emergency lighting fixtures, and each fixture is installed with ceramic phosphor, which is a special material that can be excited when irradiated by blue laser and emit visible light of a predetermined color greater than the blue light wavelength. These visible lights include green, red or white, which can respectively meet the different needs of evacuation indication, emergency warning, and ambient lighting.

[0017] Among them, the photoelectric lighting centralized power supply is the core component of the entire system. It is embedded with a laser generator, and the main function of the laser generator is to generate blue laser with a wavelength in the range of 450 - 490 nm. This blue laser has the characteristics of high energy and high directivity, and is very suitable for efficient long-distance transmission through optical fibers. In order to ensure the transmission effect of the blue light, the system designs an optical fiber module. The optical fiber module is a device specifically used for optical signal transmission, which has the characteristics of low loss and high reliability, and can stably transmit the blue laser emitted by the laser generator to each fire emergency lighting fixture.

[0018] When the blue laser is projected onto the fire emergency lighting fixture through the optical fiber module, the ceramic phosphor in the fixture starts to work under the excitation of the blue light and emits visible light of a predetermined color through an energy conversion process. For example, green light can be used to mark evacuation routes and emergency exits, red light is used for danger warning or emergency area guidance, and white light is used to provide high-brightness ambient lighting to ensure that people can clearly identify the surrounding environment and complete the evacuation smoothly.

[0019] The above design makes full use of the technical principle of optoelectronic separation, and at the same time realizes the nS-level fiber break control technology through the reflection of the visible light chromatogram, so that the entire system completely avoids the participation of current in the transmission path from the photoelectric lighting centralized power supply to the fire emergency lighting fixture, effectively reducing the hidden danger of electrical fires. In addition, through the combination of blue laser and ceramic phosphor, the lighting fixture can achieve efficient and stable lighting, meeting the emergency needs in different scenarios. In short, by integrating optical fiber transmission, laser technology and fluorescence excitation principle, a safe, reliable and versatile fire emergency lighting solution can be created.

[0020] P20: Combining with a generative adversarial network, according to the several working characteristics, several working environmental conditions and several element ratios of the several fire emergency lighting fixtures, respectively predict the light conversion efficiency attenuation of the several fire emergency lighting fixtures, and obtain several predicted light conversion efficiencies to label the several fire emergency lighting fixtures.

[0021] Further, step P20 of the embodiment of the present application further includes:

[0022] P21: Obtain the working characteristics, working environmental conditions of the several fire emergency luminaires, and the element ratios of the several ceramic phosphors. Among them, the working characteristic is the cumulative on-time of the luminaire, and the working environmental conditions at least include the average temperature, average temperature difference, average humidity, and average humidity difference in a predetermined historical time zone; P22: Randomly select the first working characteristic, the first working environmental condition, and the first element ratio of the first fire emergency luminaire; P23: Combine with the pre-constructed efficiency decay prediction plug-in of the generative adversarial network to perform decay prediction on the first working characteristic, the first working environmental condition, and the first element ratio, and output the first decay ratio; P24: Subtract the first decay ratio from 1 to obtain the first predicted light conversion efficiency, and add it to the several predicted light conversion efficiencies.

[0023] It should be understood that in the fire emergency lighting system, the generative adversarial network (GAN) can be combined to predict the decay of the light conversion efficiency of several fire emergency luminaires, providing an accurate performance status identifier for system optimization.

[0024] First, obtain the working characteristics, working environmental conditions of the several fire emergency luminaires, and the element ratios of the ceramic phosphors. The working characteristic mainly refers to the cumulative on-time of the luminaire, that is, the cumulative working time of the luminaire from the first start to the current moment, which is a key influencing factor for the change of the light conversion efficiency. The working environmental conditions include the average temperature, average temperature difference, average humidity, and average humidity difference in a predetermined historical time zone (such as the past 24 hours), etc. These parameters can reflect the stability and response ability of the luminaire operating environment. The element ratio refers to the proportion of different components in the ceramic phosphor in the luminaire, which directly affects the light efficiency and stability of the visible light after blue light excitation.

[0025] Subsequently, randomly select the first working characteristic, the first working environmental condition, and the first element ratio of a certain fire emergency luminaire (such as the first luminaire) as the input data for decay prediction. The random selection mechanism here can effectively avoid data bias and ensure the universality and accuracy of the prediction results.

[0026] Then, through the efficiency decay prediction plug-in pre-constructed by the generative adversarial network (GAN), calculate and analyze the above input data of the luminaire. The generative adversarial network is a deep learning model. Through the adversarial training of the generator and the discriminator, it can capture hidden non-linear relationships in complex and multi-dimensional data, thereby accurately predicting the decay trend of the light conversion efficiency. Specifically, GAN generates a decay model based on the input data, combines with the operating performance of the luminaire in the actual environment, and outputs the first decay ratio of the first luminaire. The decay ratio reflects the degree to which the light conversion efficiency of the luminaire decreases due to the usage time, environmental impact, and material aging.

[0027] Finally, according to the predicted first attenuation ratio, calculate the first predicted light conversion efficiency of the first lamp. The calculation formula can be: light conversion efficiency = 1 - attenuation ratio. The obtained predicted light conversion efficiency value directly reflects the light efficiency state of the lamp, and at the same time, add this value to the set of predicted light conversion efficiencies of all lamps in the system. Through this set, a comprehensive identification of the light efficiency states of all fire emergency lamps in the target building can be carried out.

[0028] By introducing GAN technology, this method effectively combines multi-dimensional working characteristics and environmental parameters to achieve accurate prediction of lamp performance, which not only provides a scientific basis for the dynamic regulation of the system but also ensures the reliability and intelligence of the fire emergency lighting system.

[0029] Furthermore, to pre-build an efficiency attenuation prediction plug-in, step P23 of the embodiment of the present application further includes:

[0030] P23-1: Collect a sample working characteristic set, a sample working environmental condition set, and a sample element ratio set, and label the lamp light conversion efficiency attenuation ratios under different sample working characteristics, sample working environmental conditions, and sample element ratios to obtain a sample attenuation ratio set; P23-2: Use the sample working characteristic set, the sample working environmental condition set, the sample element ratio set, and the sample attenuation ratio set as a training set, and equally divide it into P parts of supervised training data, where P is an integer greater than 10; P23-3: Combine a generative adversarial network to construct P efficiency attenuation prediction nodes, where each efficiency attenuation prediction node includes a generator and a discriminator; P23-4: Use the P parts of supervised training data to train the P efficiency attenuation prediction nodes respectively until the mean square error loss function converges, and obtain P trained efficiency attenuation prediction nodes, which constitute the efficiency attenuation prediction plug-in.

[0031] Optionally, in order to construct a high-precision efficiency attenuation prediction plug-in, through the processes of sample collection, data annotation, training data division, and training of the generative adversarial network (GAN), multiple efficient prediction nodes can be finally formed.

[0032] First, collect a sample working characteristic set, a sample working environmental condition set, and a sample element ratio set, and label the lamp light conversion efficiency attenuation ratios under different sample conditions to generate a sample attenuation ratio set. Here, the sample working characteristics include parameters such as the cumulative on-time of the lamp, the sample working environmental conditions cover historical temperature, humidity, and their fluctuation ranges, and the sample element ratio is the specific information of the different material ratios in the ceramic phosphor. The annotation of these data is a key step, quantitatively evaluating the light efficiency change of the lamp through real experiments or historical data to form a system's attenuation ratio benchmark.

[0033] Next, the above four types of sample data (working feature set, environmental condition set, element ratio set, attenuation ratio set) are used as the training set and equally divided into P portions of supervised training data to ensure that each data set has a balanced feature distribution and representativeness. Here, P is an integer greater than 10 to ensure sufficient sample distribution to support the efficient training of the model.

[0034] Then, P efficiency attenuation prediction nodes are constructed using the Generative Adversarial Network (GAN) technology. Each node includes two key modules: a generator and a discriminator. The generator is responsible for generating efficiency attenuation prediction values based on the input samples, while the discriminator evaluates and provides feedback by comparing the generated values with the true values, gradually optimizing the prediction ability of the generator. This adversarial training mechanism of GAN can capture complex non-linear relationships, enabling the model to exhibit excellent prediction accuracy under multi-dimensional input data.

[0035] Subsequently, the P portions of supervised training data are respectively used to train these P prediction nodes. During the training process, the Mean Squared Error (MSE) is used as the loss function, and the deviation between the prediction result and the true value is measured by minimizing the MSE value. The training is iterated until the loss functions of all nodes converge, that is, the prediction error of the model drops to an acceptable range.

[0036] Finally, all the P trained efficiency attenuation prediction nodes are integrated to form a complete efficiency attenuation prediction plugin. This plugin can accurately predict the light conversion efficiency attenuation of the lamp in real time according to the input working features, environmental conditions, and element ratios, providing key data support for subsequent dynamic optimization control. Through this process, the efficiency attenuation prediction plugin combines GAN technology with multi-dimensional training samples to achieve the performance prediction goals of high accuracy and high robustness, ensuring that the fire emergency lighting system can intelligently adapt to complex environmental changes and provide stable and reliable lighting services in practical applications.

[0037] Further, for predicting the first attenuation ratio, step P23 of the embodiment of the present application further includes:

[0038] P23-5: Obtain the first working characteristic and the first working environment condition, wherein the first working characteristic is the first cumulative opening time, and the first working environment condition includes the first temperature mean, the first temperature difference mean, the first humidity mean and the first humidity difference mean; P23-6: Perform lamp attenuation scale evaluation based on the first cumulative opening time, the first temperature mean, the first temperature difference mean, the first humidity mean and the first humidity difference mean, and output the first attenuation scale coefficient, wherein the attenuation scale coefficient is positively correlated with the cumulative opening time, the temperature mean, the temperature difference mean, the humidity mean and the humidity difference mean; P23-7: Calculate the coefficient ratio of the first attenuation scale coefficient to the maximum attenuation scale coefficient in the predetermined historical window, and multiply it by P and round it to get the first prediction node number; P23-8: Randomly select the efficiency attenuation prediction nodes of the first prediction node number from the P efficiency attenuation prediction nodes, perform attenuation prediction on the first working characteristic, the first working environment condition and the first element ratio, and output the first attenuation ratio after the mean calculation.

[0039] In a possible embodiment of the present application, in the process of predicting the first attenuation ratio, the embodiment of the present application realizes accurate and efficient light conversion efficiency prediction through in-depth analysis of the first working characteristics and the first working environment conditions, combined with the branch optimization selection of the lamp attenuation scale assessment and efficiency attenuation prediction node.

[0040] First, the first working characteristic and the first working environment condition are obtained, wherein the first working characteristic is the cumulative opening time of the lamp, which is one of the core factors of the lamp light efficiency decay and reflects the total use time of the lamp. The first working environment condition includes the first temperature mean, the first temperature difference mean, the first humidity mean and the first humidity difference mean. These environmental parameters can fully characterize the environmental changes and stability of the lamp operation, and have an important impact on the light efficiency decay.

[0041] Next, based on the above parameters, the lamp attenuation scale is evaluated and the first attenuation scale coefficient is output. The attenuation scale coefficient is an indicator that quantifies the trend of the lamp's light efficiency decline, and is positively correlated with the cumulative turn-on time, the average temperature, the average temperature difference, the average humidity, and the average humidity difference. For example, the longer the cumulative turn-on time and the more severe the environmental fluctuations (such as temperature difference and humidity difference), the greater the attenuation scale coefficient, reflecting that the degree to which the lamp's light efficiency is affected is higher.

[0042] Then, calculate the ratio of the first attenuation scale coefficient to the maximum attenuation scale coefficient within a predetermined historical window to obtain the ratio of the current luminaire to the historical most severe attenuation condition. Multiply this ratio by P (the total number of efficiency attenuation prediction nodes) and round up to obtain the number of the first prediction nodes. This step realizes the optimization of prediction adaptability by dynamically allocating the number of prediction nodes: the larger the attenuation scale, the more prediction nodes are allocated to ensure the prediction accuracy; when the attenuation scale is small, the number of allocated nodes is correspondingly reduced to improve the system calculation efficiency.

[0043] Subsequently, randomly select the nodes with the number of the first prediction nodes among the P efficiency attenuation prediction nodes to perform attenuation prediction on the first operating characteristics, the first operating environment conditions, and the first element ratio. The calculation result of each prediction node reflects the independent evaluation ability of the node for the current input parameters. Finally, calculate the mean value of the output results of all selected nodes to obtain the first attenuation ratio.

[0044] This integrated learning method greatly improves the analysis efficiency while ensuring the prediction accuracy by dynamically selecting the number of prediction branches. By only using an appropriate number of prediction nodes to participate in the calculation, it avoids resource waste and at the same time ensures the sufficient learning ability of the model for complex input parameters. The finally output first attenuation ratio provides reliable data support for the subsequent calculation of the light conversion efficiency and the optimization control, further enhancing the intelligence and robustness of the fire emergency lighting system.

[0045] P30: Combine the layout structure diagram of the target building to simulate and generate a three-dimensional building model, map and render several identified fire emergency luminaires to the three-dimensional building model to generate a three-dimensional building emergency lighting model.

[0046] Specifically, in the process of combining the layout structure diagram of the target building to simulate and generate a three-dimensional building model, the embodiment of the present application maps the data of the identified fire emergency luminaires into the building model through digital modeling and rendering technology to construct a complete three-dimensional building emergency lighting model.

[0047] First, obtain the layout structure diagram from the target building. This structure diagram is important basic data reflecting the building space distribution and functional areas, including room positions, passage distributions, exit positions, and height information, etc. The layout structure Figure 1 is generally stored in CAD or BIM formats. These formats can accurately describe the geometric features and spatial relationships of the building and are the key inputs for subsequent three-dimensional modeling.

[0048] Next, using professional simulation modeling tools (such as Revit or 3D Max), the layout structure diagram is converted into a three-dimensional building model. During this process, the simulation tool utilizes the geometric information in the layout structure diagram (such as floor plans, sectional views, and elevation views) to construct a three-dimensional digital model that reflects the actual structure of the building. This model not only includes the spatial layout of the building but also can reflect details such as wall materials and door and window distributions based on the input data, making the model highly realistic.

[0049] Subsequently, the already identified fire emergency lighting fixtures are mapped into the three-dimensional building model. Each fire emergency lighting fixture identification data includes the position coordinates of the fixture, light conversion efficiency, lighting range, and a predetermined color (e.g., green for evacuation indication, red for warning, and white for ambient lighting). During the mapping process, by matching the fixture identification data with specific coordinate points in the three-dimensional building model, the fixtures are accurately placed at their actual installation positions in the model.

[0050] To further enhance the visualization effect of the model, rendering technology is used to dynamically represent the fire emergency lighting fixtures. For example, the functional types of different fixtures can be visually marked by colors, the lighting range of the fixtures can be visualized in the form of a light cone, or the lighting effects of the fixtures in different environments (such as darkness, smoke-filled, etc.) can be dynamically simulated. This visualization process not only improves the intuitiveness of the three-dimensional model but also provides a reliable basis for the subsequent optimization of the lighting system.

[0051] Finally, a complete three-dimensional model of building emergency lighting is generated. This model integrates the building structure, fixture distribution, and their performance status, and can dynamically reflect the working conditions of the emergency lighting system in the building. Through the above steps, the construction of the three-dimensional model of building emergency lighting not only realizes digital management but also provides a precise analysis tool for the optimization and control of the fire emergency system, helping to improve the efficiency and reliability of emergency lighting.

[0052] P40: Using the three-dimensional model of building emergency lighting, perform the fixed-point optimization control analysis of the predetermined fire emergency lighting equipment, and output a dynamic optimization control plan for the adaptive regulation of emergency lighting in the target building.

[0053] Furthermore, as Figure 2 shown, step P40 of the embodiment of the present application further includes:

[0054] P41: Receive fire sensing data at a fixed point, and within the building three-dimensional model, perform fire spread simulation within a predetermined time zone based on the fire sensing data, and output the predicted fire spread area; P42: Analyze the illumination intensity of several fire emergency lighting fixtures based on the predicted fire spread area, and output several expected illumination intensities; P43: Use the building emergency lighting three-dimensional model, combine several predicted light conversion efficiencies to compensate the several expected illumination intensities, and determine several target illumination intensities; P44: Based on the several target illumination intensities, perform optimization analysis of the emission control parameters of the laser generator within the predetermined time zone, output an optimized control scheme, and perform laser emission control within the predetermined time zone according to the optimized control scheme; P45: Continue to receive fire sensing data to perform fixed-point optimization control analysis, output a dynamic optimization control scheme for adaptive control of the emergency lighting of the target building.

[0055] It should be understood that when performing fixed-point optimization control analysis of the predetermined fire emergency lighting equipment using the building emergency lighting three-dimensional model, the embodiments of the present application form a real-time adaptive control scheme through the dynamic linkage between the fire sensing data and the lighting system.

[0056] First, receive fire sensing data at a fixed point in the building three-dimensional model. The fire sensing data includes information such as temperature, smoke concentration, and flame position, which are provided in real time by sensors distributed in the building. Based on this data, simulate the fire spread within a predetermined time zone, and use simulation algorithms to predict the fire spread path and area in the building. The fire spread simulation provides a key environmental basis for subsequent emergency lighting optimization.

[0057] Next, based on the predicted fire spread area, the system analyzes the illumination intensity of relevant fire emergency lighting fixtures. By calculating the lighting requirements in the fire-affected area and combining the evacuation path and visibility requirements of personnel, output several expected illumination intensities. For example, the fixtures in high-fire-risk areas need to provide stronger lighting to ensure clear visibility of the evacuation route.

[0058] Then, use the building emergency lighting three-dimensional model, combine several previously calculated predicted light conversion efficiencies, and compensate these expected illumination intensities. The light conversion efficiency compensation is an adjustment based on the current performance state of the fixture (such as the attenuation ratio) to ensure that the actual lighting effect matches the expected demand. Through calculation, determine the target illumination intensity of each fixture to meet the emergency lighting requirements.

[0059] Furthermore, based on the target illumination intensity, the system performs an optimization analysis on the emission control parameters of the laser generator. The control parameters include laser power, emission direction, distribution intensity, etc. Through the optimization analysis, an optimized control scheme is generated, and based on this, the operating state of the laser generator within a predetermined time period is adjusted. For example, the laser power is increased in areas where high brightness is required, while the power is reduced in low-demand areas, optimizing energy consumption while ensuring the lighting quality in key areas.

[0060] Finally, continue to receive fire sensing data and adjust the optimization scheme in real time. This dynamic optimization continuously updates the control strategy by continuously monitoring the spread of the fire and combining the deviation analysis between the current illumination intensity and the target intensity, ensuring that the intensity deviation of the overall lighting system is minimized and achieving the best adaptive control effect.

[0061] Furthermore, step P44 of the embodiment of the present application further includes:

[0062] P44-1: Obtain the emission control parameter space of the laser generator and randomly select multiple initial emission control parameters. Among them, the emission control parameters at least include laser power, laser beam divergence angle, and laser frequency; P44-2: Perform lighting simulation according to the multiple initial emission control parameters and output multiple sets of simulated illumination intensities. Among them, each set of simulated illumination intensities includes several simulated illumination intensities; P44-3: Based on the several target illumination intensities, calculate the mapping deviation for each of the multiple sets of simulated illumination intensities, output multiple sets of illumination intensity deviations, and calculate multiple mean intensity deviations; P44-4: Evaluate the lighting fitness of the multiple initial emission control parameters according to the multiple mean intensity deviations and output multiple lighting fitness values. Among them, the lighting fitness is negatively correlated with the mean intensity deviation; P44-5: With the emission control parameter space as a constraint, based on the multiple lighting fitness values and multiple initial emission control parameters, perform an optimization analysis of the emission control parameters until convergence, output the optimal emission control parameters, and set them as the optimized control scheme.

[0063] Optionally, during the process of optimizing the emission control parameters of the laser generator, the embodiment of the present application finally determines the optimal emission control parameters through the exploration of the emission parameter space, lighting simulation and deviation evaluation, and fitness-based optimization analysis.

[0064] First, obtain the emission control parameter space of the laser generator. The emission control parameter space is a set of all feasible parameters, at least including three key dimensions: laser power, laser beam divergence angle, and laser frequency. Among them, the laser power determines the intensity of the light, the laser beam divergence angle affects the lighting coverage range, and the laser frequency plays a key role in the energy distribution and stability of the laser. Randomly select multiple initial emission control parameters from this parameter space as the starting point of the optimization process.

[0065] Next, lighting simulation is performed based on the selected multiple initial emission control parameters. The lighting simulation reproduces the actual lighting effects of each set of emission control parameters in the three-dimensional model of building emergency lighting through simulation technology, and outputs multiple sets of simulated lighting intensities. Each set of simulated lighting intensities includes several simulated lighting intensities, which respectively correspond to the lighting effects in different areas of the building.

[0066] Then, based on the predetermined target lighting intensity of the system, mapping deviation calculation is performed on each set of simulated lighting intensities. The mapping deviation calculation is used to evaluate the difference between the simulated lighting intensity and the target lighting intensity, outputs multiple sets of lighting intensity deviations, and further calculates the mean intensity deviation of each deviation set. The mean deviation reflects the overall lighting deviation degree of the current emission parameter combination and is a key indicator for measuring parameter adaptability.

[0067] Subsequently, lighting fitness evaluation is performed according to the calculated multiple mean intensity deviations. Lighting fitness is an index for evaluating the effect of the emission control parameter combination, and has a negative correlation with the mean intensity deviation, that is, the smaller the mean deviation, the higher the fitness. In this way, the fitness is used to rank the advantages and disadvantages of the initial emission control parameters, providing a basis for subsequent optimization.

[0068] Finally, with the emission control parameter space as the constraint condition, based on multiple lighting fitnesses and the initial emission control parameters, emission control parameter optimization analysis is performed. The optimization analysis uses an iterative algorithm (such as a genetic algorithm or a particle swarm optimization algorithm) to explore within the emission control parameter space, gradually adjusting the parameter combination to improve the fitness. Through multiple iterative calculations until the fitness converges, the optimal emission control parameters are finally output. This parameter combination is the optimized control scheme and can be directly used for the emission control of the laser generator.

[0069] Furthermore, step P44-5 of the embodiment of the present application further includes:

[0070] P44-51: Arrange the multiple initial emission control parameters in descending order of lighting fitness to generate an initial solution sequence, and select the first predetermined proportion of solutions in the initial solution sequence as preferred individuals to obtain a preferred individual population; P44-52: According to a predetermined scheme, perform individual crossover and individual mutation on the preferred individual population to obtain an updated individual population; P44-53: Repeat the selection, crossover, and mutation steps until a preset number of iterations is reached, and output the individual with the maximum fitness in the current updated individual population as the optimal emission control parameter.

[0071] Specifically, in the process of determining the optimal emission control parameters, further through the crossover and mutation optimization mechanism based on the preferred individual population, intelligent search for the emission control parameters is realized, and finally the optimal solution with the highest fitness is output.

[0072] First, according to the previously calculated lighting fitness, arrange multiple initial emission control parameters in descending order to generate an initial solution sequence. Each solution in the solution sequence represents a set of emission control parameter combinations, including laser power, laser beam divergence angle, and laser frequency. The fitness of these parameter combinations reflects the degree of their matching with the target illumination intensity. Subsequently, select a number of solutions with the highest fitness from the initial solution sequence according to a predetermined ratio (such as the top 30% or 50%) to form a preferred individual population. The preferred individual population represents the emission control parameter combinations with better performance in the current solution space and provides a high-quality basis for subsequent optimization.

[0073] Next, for the preferred individual population, perform individual crossover and individual mutation operations according to a predetermined scheme. The crossover operation refers to selecting some parameters from two preferred individuals for exchange. For example, combine the laser power of one individual with the laser beam divergence angle of another individual to generate a new solution. This way can explore new regions of the solution space and avoid falling into local optima. The mutation operation is to make small random adjustments to the individual parameters, such as slightly increasing the laser frequency or decreasing the laser power, thereby introducing diversity into the solution space. Through the crossover and mutation operations, generate a new updated individual population and recalculate and evaluate its fitness.

[0074] Subsequently, repeat the optimization steps of selection, crossover, and mutation. In each iteration, the quality of the preferred individual population is continuously improved, and the new updated population gradually approaches the solution with higher fitness. The system records the individual with the highest fitness in the current population in each iteration until the preset number of iterations is reached, ensuring that the optimization process fully explores the solution space while controlling the computational cost.

[0075] Finally, after reaching the preset number of iterations, set the individual with the highest fitness in the current updated individual population as the optimal emission control parameter. This parameter combination has been optimized through multiple rounds and can achieve the best balance in dimensions such as laser power, divergence angle, and frequency, ensuring that the system achieves the optimal configuration of energy efficiency and performance while meeting the target illumination intensity. Through this optimization strategy, efficiently search in the complex multi-dimensional parameter space, and the finally output optimal emission control parameter can significantly improve the performance and reliability of the fire emergency lighting system, while ensuring flexibility and stability in practical applications.

[0076] In summary, the embodiments of the present application have at least the following technical effects:

[0077] Through the optoelectronic separation technology, this application completely eliminates the risks of electrical fires such as short circuits and arcing that may be caused by the traditional wire power supply method, and significantly improves the safety of the system in flammable, explosive places and special environments. By using generative adversarial networks and 3D building models, dynamic optimization control of the emergency lighting system is achieved, and the laser parameters are adjusted in real time according to environmental changes to ensure precise matching of lighting coverage and intensity, thereby improving the emergency response ability.

[0078] It achieves the technical effect of improving the inherent safety, anti-interference ability and emergency response ability of the emergency lighting system through the optoelectronic separation technology and the dynamic optimization of generative adversarial networks.

[0079] It should be noted that the above sequence of embodiments of this application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the above specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0080] The above are only the preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this application shall be included within the protection scope of this application.

[0081] This specification and the drawings are only exemplary descriptions of this application and are considered to cover any and all modifications, changes, combinations or equivalents within the scope of this application. Obviously, those skilled in the art can make various changes and modifications to this application without departing from the scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application is intended to include these changes and modifications.

Claims

1. The adaptive control method of fire emergency lighting is characterized in that: Methods include: Deploy predetermined fire emergency lighting equipment in the target building, wherein the predetermined fire emergency lighting equipment includes a fire emergency lamp group, the fire emergency lamp group includes a plurality of fire emergency lamps, each of which is equipped with ceramic phosphors, and the ceramic phosphors can excite a predetermined color visible light greater than the wavelength of blue light under the irradiation of a blue laser; Combined with a generative adversarial network, according to a number of working characteristics, a number of working environment conditions and a number of element ratios of the said number of fire emergency lamps, respectively predict the light conversion efficiency attenuation of the said number of fire emergency lamps, obtain a number of predicted light conversion efficiencies to mark the said number of fire emergency lamps, the element ratio refers to the ratio of different components in the ceramic phosphor in the lamp; Combined with the layout structure diagram of the target building, a three-dimensional building model is generated by simulation, and a plurality of identification fire emergency lamps are mapped and rendered to the three-dimensional building model to generate a three-dimensional building emergency lighting model; Using the three-dimensional model of building emergency lighting, performing fixed-point optimization control analysis of the predetermined fire emergency lighting equipment, and outputting a dynamic optimization control scheme to perform adaptive regulation of the emergency lighting of the target building; Combined with the generative adversarial network, according to the working characteristics, working environment conditions and element ratios of the fire emergency lamps, the light conversion efficiency attenuation of the fire emergency lamps is predicted respectively to obtain a number of predicted light conversion efficiencies, including: Obtaining several working characteristics and several working environment conditions of the several fire emergency lamps, and several element ratios of several ceramic phosphors, wherein the working characteristic is the cumulative opening time of the lamps, and the working environment conditions at least include the temperature mean, temperature difference mean, humidity mean, and humidity difference mean in a predetermined historical time zone; Randomly select a first working characteristic, a first working environment condition and a first element ratio of a first fire emergency lighting fixture; In combination with the pre-built efficiency attenuation prediction plug-in of the generative adversarial network, attenuation prediction is performed on the first working characteristic, the first working environment condition and the first element ratio, and a first attenuation ratio is output; Subtracting the first attenuation ratio from 1 to obtain a first predicted light conversion efficiency, and adding the first predicted light conversion efficiency to the plurality of predicted light conversion efficiencies; Using the three-dimensional model of building emergency lighting, executing the fixed-point optimization control analysis of the predetermined fire emergency lighting equipment, and outputting a dynamic optimization control scheme to perform adaptive regulation of the emergency lighting of the target building, including: receiving fire sensor data at a fixed point, performing a fire spread simulation within a predetermined time zone in the three-dimensional building model according to the fire sensor data, and outputting a predicted fire spread area; Performing lighting intensity analysis of a plurality of fire emergency lamps according to the predicted fire spread area, and outputting a plurality of expected lighting intensities; Using the three-dimensional model of building emergency lighting, combined with a plurality of predicted light conversion efficiencies, the plurality of expected lighting intensities are compensated to determine a plurality of target lighting intensities; Based on the illumination intensities of the several targets, performing an optimization analysis of emission control parameters of the laser generator in the predetermined time zone, outputting an optimization control scheme, and performing laser emission control in the predetermined time zone according to the optimization control scheme; Continue to receive fire sensor data to perform fixed-point optimization control analysis, and output dynamic optimization control solutions to perform adaptive regulation of emergency lighting in target buildings.

2. The adaptive control method of fire emergency lighting according to claim 1, characterized in that: The predetermined fire emergency lighting equipment also includes a photoelectric lighting centralized power supply and an optical fiber module, wherein the photoelectric lighting centralized power supply is embedded with a laser generator, and the laser generator is used to emit a blue laser; The optical fiber module is connected to the fire emergency lamp group, and emits a predetermined color visible light greater than the wavelength of the blue light by projecting the blue laser emitted by the laser generator to the plurality of fire emergency lamps, wherein the predetermined color is green, red or white.

3. The adaptive control method of fire emergency lighting according to claim 1, characterized in that: Pre-built efficiency decay prediction plugins, including: Collecting a sample working characteristic set, a sample working environment condition set and a sample element ratio set, and marking the light conversion efficiency attenuation ratio of lamps under different sample working characteristics, sample working environment conditions and sample element ratios to obtain a sample attenuation ratio set; The sample working feature set, the sample working environment condition set, the sample element ratio set and the sample attenuation ratio set are used as training sets, and are equally divided into P supervised training data, where P is an integer greater than 10; Combined with the generative adversarial network, P efficiency decay prediction nodes are constructed, where each efficiency decay prediction node includes a generator and a discriminator; The P pieces of supervised training data are used to train the P efficiency decay prediction nodes respectively until the mean square error loss function converges, and the trained P efficiency decay prediction nodes are obtained to form the efficiency decay prediction plug-in.

4. The adaptive control method of fire emergency lighting according to claim 3, characterized in that: Performing attenuation prediction on the first working characteristic, the first working environment condition and the first element ratio, and outputting a first attenuation ratio, includes: Acquire a first working characteristic and a first working environment condition, wherein the first working characteristic is a first accumulated opening time, and the first working environment condition includes a first temperature mean, a first temperature difference mean, a first humidity mean, and a first humidity difference mean; Perform lamp attenuation scale assessment according to the first cumulative opening time, the first temperature average, the first temperature difference average, the first humidity average and the first humidity difference average, and output a first attenuation scale coefficient, wherein the attenuation scale coefficient is positively correlated with the cumulative opening time, the temperature average, the temperature difference average, the humidity average and the humidity difference average; Calculate the coefficient ratio of the first attenuation scale coefficient to the maximum attenuation scale coefficient in a predetermined historical window, multiply the coefficient by P and round it to obtain the first prediction node quantity; Efficiency attenuation prediction nodes of the first number of prediction nodes are randomly selected from the P efficiency attenuation prediction nodes, attenuation prediction is performed on the first working characteristic, the first working environment condition and the first element ratio, and a first attenuation ratio is output after mean calculation.

5. The adaptive control method of fire emergency lighting according to claim 1, characterized in that: Based on the illumination intensities of the several targets, an optimization analysis of emission control parameters of the laser generator in the predetermined time zone is performed, and an optimization control scheme is output, including: Acquire the emission control parameter space of the laser generator, and randomly select a plurality of initial emission control parameters, wherein the emission control parameters at least include laser power, laser beam divergence angle and laser frequency; Performing lighting simulation according to the multiple initial emission control parameters, and outputting multiple simulated lighting intensity sets, wherein each simulated lighting intensity set includes a plurality of simulated lighting intensities; Taking the several target lighting intensities as references, respectively performing mapping deviation calculations on the multiple simulated lighting intensity sets, outputting multiple lighting intensity deviation sets, and calculating multiple intensity deviation mean values; Performing lighting adaptability evaluation on the multiple initial emission control parameters according to the multiple intensity deviation mean values, and outputting multiple lighting adaptability values, wherein the lighting adaptability value is negatively correlated with the intensity deviation mean value; Taking the emission control parameter space as a constraint, based on the multiple lighting adaptability and multiple initial emission control parameters, an emission control parameter optimization analysis is performed until convergence, and the optimal emission control parameters are output and set as the optimization control scheme.

6. The adaptive control method of fire emergency lighting according to claim 5, characterized in that: Based on the multiple lighting adaptability and the multiple initial emission control parameters, an emission control parameter optimization analysis is performed until convergence, and an optimal emission control parameter is output, including: Arrange multiple initial emission control parameters from large to small according to the lighting adaptability to generate an initial solution sequence, and select the first predetermined proportion of solutions in the initial solution sequence as preferred individuals to obtain a preferred individual population; According to a predetermined plan, performing individual crossover and individual mutation on the preferred individual population to obtain an updated individual population; The steps of selection, crossover, and mutation are repeatedly performed until a preset number of iterations is reached, and the individual with the largest fitness in the current updated individual population is output as the optimal emission control parameter.

Images