Intelligent night induction dimming method and system for mobile solar light tower

Abstract

The application discloses a mobile solar lighthouse intelligent night induction dimming method and system, which comprises the following steps: obtaining initial environmental parameters, starting a red light main mode, locking a red light proportion according to a fog density threshold, and superimposing yellow light on the basis of the red light proportion to form a preliminary spectrum output. If the salinity changes, the yellow light proportion in the spectrum combination is optimized by using a linear interpolation algorithm according to the salinity change trend, the spectrum matching ratio is determined, and then the matching degree is verified by a spectrum analyzer. If the spectrum matching ratio does not reach the expectation, the red and yellow light matching ratio is iteratively adjusted by using a gradient descent algorithm, the spectrum matching sea area penetration requirement is ensured, and the dynamic spectrum combination and real-time wave band iteration based on environmental variables in the mobile solar lighthouse are realized, so that the light penetration and stability in the complex fog weather scene are ensured.

Description

Technical Field

[0001] This invention relates to intelligent dimming technology for solar-powered lighthouses, and more particularly to an intelligent nighttime sensing dimming method and system for mobile solar-powered lighthouses. Background Technology

[0002] Mobile solar-powered lighthouses are crucial for maritime navigation and safety, providing reliable visual guidance for ships at night and in adverse weather conditions. Their intelligent dimming technology is therefore essential. Traditional lighthouses rely heavily on fixed light sources or manual adjustments, making them ill-suited to the dynamic changes in the marine environment. Especially in low-visibility scenarios such as fog, existing methods fall short in handling complex environmental variables. They often fail to dynamically adjust spectral combinations based on real-time environmental data, resulting in insufficient light penetration or excessive energy consumption, thus failing to meet the needs of long-distance ship navigation.

[0003] The core challenge lies in achieving dynamic spectral combination and real-time adaptation to environmental variables. In foggy marine scenarios, light penetration is affected by factors such as fog density and sea salinity, requiring the lighthouse system to quickly sense environmental changes and adjust its spectral output. For example, red light has strong penetration in fog, but an excessively high proportion of red light may lead to an overly singular light source, making it difficult to meet visibility requirements under complex weather conditions such as sleet. Adding yellow light can improve light softness, but an imbalance in the proportion may weaken the overall penetration effect. This dynamic balance of spectral combination requires the system to iterate rapidly across different bands while maintaining output stability. However, existing technologies struggle to achieve continuous band scanning and real-time environmental data linkage when processing multi-band spectral superposition, resulting in light output that cannot accurately match actual needs. For instance, if the system cannot adjust the red and yellow light ratio in time when fog density suddenly increases in a certain sea area, the light may not be able to penetrate the dense fog, making it difficult for ships to identify the lighthouse signal.

[0004] Therefore, how to achieve dynamic spectral combination and real-time band iteration based on environmental variables in mobile solar light towers to ensure the penetration and stability of light in complex foggy scenarios has become the key issue of this study. Summary of the Invention

[0005] This invention provides an intelligent nighttime sensing dimming method and system for mobile solar light towers, which enables dynamic spectral combination and real-time band iteration based on environmental variables in mobile solar light towers to ensure the penetration and stability of light in complex foggy weather scenarios.

[0006] This invention provides an intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse, executed by a computer, comprising:

[0007] Collect fog density and salinity data of the marine environment to determine the initial environmental parameters;

[0008] Based on the initial environmental parameters, the red light-dominant mode of the RGB-LED array is activated, and the spectral band is set to start from red light to obtain the preliminary spectral output;

[0009] If the fog density data is higher than a preset density threshold, the proportion of red light in the initial spectral output is locked by the RGB-LED array, and a yellow light component is superimposed to obtain an adjusted spectral combination.

[0010] Based on the salinity change trend corresponding to real-time salinity data, a linear interpolation algorithm is used to optimize the proportion of yellow light in the adjusted spectral combination and determine the spectral ratio.

[0011] Based on the spectral ratio, the bands of the spectral ratio are scanned by a spectral analyzer to determine the spectral matching result, wherein the spectral matching result includes the degree of matching between the spectral output and the band required for sea area penetration;

[0012] If the spectral matching result does not reach the preset matching degree, the gradient descent algorithm is used to iteratively adjust the ratio of red and yellow light in the spectral composition to obtain the target spectral combination that meets the preset matching degree.

[0013] The target spectral combination is used for illumination execution and adjustment of light characteristics of the light source device, which includes the RGB-LED array.

[0014] The intelligent night-sensing dimming method for mobile solar-powered lighthouses provided by the present invention further includes:

[0015] Based on the target spectral combination, the output spectrum of the RGB-LED array is controlled, and a closed-loop feedback mechanism is used to continuously monitor environmental indicators.

[0016] If the environmental indicators change, fog density data and salinity data are reacquired as the environmental data set.

[0017] Based on the aforementioned environmental data set, a weighted average algorithm is used to calculate the comprehensive impact factors of fog density and salinity, thereby obtaining the environmental impact weights.

[0018] Based on the environmental impact weights and combined with the preset spectral ratio model, the spectral ratio data of the target spectral combination is updated to obtain the adjusted spectral ratio.

[0019] If the adjusted spectral ratio exceeds the preset threshold range, the adjusted spectral ratio is optimized using a linear regression algorithm to determine the optimized spectral ratio.

[0020] Based on the optimized spectral ratio, the light distribution parameters are calculated using spectral analysis tools to obtain a set of light distribution parameters.

[0021] The spectral ratio model is determined based on a preset spectral ratio training set.

[0022] The intelligent night-sensing dimming method for mobile solar-powered lighthouses provided by the present invention further includes:

[0023] Acquire GPS data from the mobile solar-powered lighthouse;

[0024] If the fog density data exceeds a preset density threshold, the fog density change trend is predicted using a linear regression algorithm to obtain a predicted fog density value.

[0025] Based on the predicted fog density and the salinity data, a comprehensive marine environmental index is determined.

[0026] If the comprehensive marine environment index is greater than the preset environmental state threshold, the lighthouse operating parameters are updated based on the GPS data to obtain the real-time operating status of the mobile solar lighthouse.

[0027] Based on the real-time operating status and the initial environmental parameters, the lighthouse operating modes are classified using the K-means clustering algorithm to determine the target operating mode;

[0028] The target operating mode is used to activate the red light-dominant mode of the RGB-LED array in combination with the initial environmental parameters to obtain a preliminary spectral output.

[0029] According to the intelligent night-time sensing dimming method for mobile solar-powered light towers provided by the present invention, if the fog density data is higher than a preset density threshold, the red light proportion in the initial spectral output is locked by the RGB-LED array, and a yellow light component is superimposed to obtain an adjusted spectral combination, including:

[0030] If the fog density data is higher than the preset density threshold, then based on the fog density data, the fog density data is processed by a spectral analysis algorithm to determine the red light ratio adjustment requirement and obtain the red light fixed ratio parameter;

[0031] Based on the red light fixed ratio parameter, the red light ratio of the RGB-LED array is fixed to generate an initial spectral combination;

[0032] Based on the initial spectral combination, the yellow light component is superimposed to obtain the intermediate spectral combination;

[0033] Based on the intermediate spectral combination, a spectral analysis algorithm is used to analyze the intermediate spectral combination to obtain its spectral characteristics;

[0034] Based on the spectral characteristics and the preset target attributes, the adjusted spectral combination is determined.

[0035] According to the intelligent night-time sensing dimming method for mobile solar-powered light towers provided by the present invention, the step of optimizing the proportion of yellow light in the adjusted spectral combination based on the salinity change trend corresponding to real-time salinity data and determining the spectral ratio using a linear interpolation algorithm includes:

[0036] The real-time salinity data is processed using a linear interpolation algorithm to generate a continuous salinity distribution characterizing salinity changes and to determine the salinity change trend.

[0037] Based on the salinity change trend, the yellow light ratio adjustment coefficient is calculated to obtain a preliminary ratio adjustment scheme;

[0038] If the preliminary ratio adjustment scheme meets the preset spectral ratio threshold, then the ratio of the preliminary ratio adjustment scheme is optimized through the spectral analysis model to determine the first ratio adjustment scheme;

[0039] If the initial ratio adjustment scheme does not meet the spectral ratio threshold, the mapping relationship between salinity change and yellow light ratio is recalculated using a regression analysis algorithm to obtain a second ratio adjustment scheme.

[0040] The spectral ratio is determined based on either the first or the second ratio adjustment scheme.

[0041] According to the intelligent night-time sensing dimming method for mobile solar-powered lighthouses provided by the present invention, the step of determining the spectral matching result by scanning the bands of the spectral ratio using a spectral analyzer based on the spectral ratio includes:

[0042] Based on the spectral ratio, an initial spectral distribution is generated by scanning the bands of the spectral ratio using a spectral analyzer.

[0043] If the band range of the initial spectral distribution does not match the preset sea area penetration requirement band, the scanning parameters of the spectral analyzer are adjusted to obtain the optimized spectral distribution corresponding to the spectral ratio.

[0044] The optimized spectral distribution is compared with the required waveband for sea area penetration to determine the matching degree score;

[0045] If the matching score is lower than a preset score threshold, the bands of the spectral ratio are reconfigured to generate updated spectral data.

[0046] Based on the updated spectral data, a support vector machine algorithm is used to classify it and determine the classification result, wherein the classification result includes the degree of matching between the updated spectral data and the required waveband for the sea area.

[0047] Based on the classification results, the band parameters are iteratively adjusted to obtain the spectral matching results.

[0048] According to the intelligent night-time sensing dimming method for a mobile solar-powered lighthouse provided by the present invention, if the spectral matching result does not reach a preset matching degree, a gradient descent algorithm is used to iteratively adjust the ratio of red and yellow light in the spectral composition to obtain a target spectral combination that satisfies the preset matching degree, including:

[0049] If the spectral matching result does not reach the preset matching degree, the matching error value corresponding to the spectral matching result is calculated by the spectral analysis model to obtain the initial red light ratio and yellow light ratio;

[0050] Based on the matching error value, the gradient direction of the initial red light ratio and yellow light ratio is calculated using the gradient descent algorithm to determine the adjustment step size parameter;

[0051] Based on the adjustment step size parameter, the initial red light ratio and yellow light ratio are updated to obtain a new spectral combination;

[0052] Based on the new spectral combination, the new matching error value is recalculated using the spectral analysis model;

[0053] If the new matching error value does not meet the convergence judgment condition, the gradient descent algorithm is used to iteratively optimize the new matching error value to determine the target spectral combination.

[0054] This invention also provides an intelligent night-time sensing dimming system for a mobile solar-powered lighthouse, comprising:

[0055] The data acquisition module is used to collect fog density and salinity data of the marine environment to determine the initial environmental parameters;

[0056] The activation module is used to activate the red light-dominant mode of the RGB-LED array based on the initial environmental parameters, set the spectral band to start from red light, and obtain the initial spectral output;

[0057] A fixing module is used to lock the proportion of red light in the initial spectral output through the RGB-LED array and superimpose the yellow light component if the fog density data is higher than a preset density threshold, so as to obtain an adjusted spectral combination.

[0058] The optimization module is used to optimize the proportion of yellow light in the adjusted spectral combination based on the salinity change trend corresponding to real-time salinity data, and to determine the spectral ratio.

[0059] The scanning module is used to scan the bands of the spectral ratio using a spectral analyzer based on the spectral ratio to determine the spectral matching result, wherein the spectral matching result includes the degree of matching between the spectral output and the band required for sea area penetration;

[0060] The adjustment module is used to iteratively adjust the ratio of red and yellow light in the spectral composition using a gradient descent algorithm if the spectral matching result does not reach the preset matching degree, so as to obtain the target spectral combination that meets the preset matching degree.

[0061] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the program, it implements an intelligent night-sensing dimming method for a mobile solar lighthouse as described in any of the preceding claims.

[0062] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the intelligent night-sensing dimming method for a mobile solar lighthouse as described in any of the preceding claims.

[0063] The present invention provides an intelligent nighttime sensing and dimming method and system for mobile solar-powered lighthouses. By acquiring initial environmental parameters, a red-light-dominant mode is activated, and the red light ratio is locked based on a fog density threshold. Yellow light is then superimposed on this red light ratio to form a preliminary spectral output. If salinity changes, a linear interpolation algorithm is used to optimize the yellow light ratio in the spectral combination based on the salinity trend, determining the spectral ratio. The matching degree is then verified using a spectral analyzer. If the spectral ratio does not meet expectations, a gradient descent algorithm is used to iteratively adjust the red and yellow light ratio to ensure the spectrum adapts to the penetration requirements of the sea area. This achieves dynamic spectral combination and real-time band iteration based on environmental variables in mobile solar-powered lighthouses, ensuring the penetration and stability of light in complex foggy conditions. Attached Figure Description

[0064] Figure 1 This is one of the flowcharts illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention;

[0065] Figure 2 This is the second flowchart illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention.

[0066] Figure 3 This is the third flowchart illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention.

[0067] Figure 4 This is the fourth flowchart illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention.

[0068] Figure 5 This is the fifth flowchart illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention.

[0069] Figure 6 This is the sixth flowchart illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention.

[0070] Figure 7 This is the seventh flowchart illustrating the intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse provided in this embodiment of the invention.

[0071] Figure 8 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0072] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0073] Reference Figure 1 This invention provides an intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse, comprising the following steps:

[0074] Step 100: Collect fog density and salinity data of the marine environment to determine the initial environmental parameters;

[0075] Initial environmental parameters are generated by collecting fog density and salinity data of the marine environment. These initial environmental parameters can be a set of fog density and salinity data, a weighted sum of fog density and salinity data, or a result of analysis and processing. A sensor network can be deployed within a pre-defined target marine area. Key locations of the sensor network can be near the coast, in waterways, and in ecologically sensitive areas of the target marine area. The sensor network can include a laser scattering fog density meter and a conductivity-salinity sensor. The fog density meter is used to collect fog density data of the marine environment, and the conductivity-salinity sensor is used to collect salinity data of the marine environment. The fog density data is collected by the fog density meter by emitting a specific wavelength laser and analyzing the scattering intensity of atmospheric particulate matter, automatically recording visibility values ​​hourly. The salinity data is obtained by the salinity sensor by real-time measurement and conversion of seawater conductivity. The sampling depth for salinity data can be set to 0.5 meters below the sea surface to avoid surface interference.

[0076] During a continuous 72-hour data acquisition period, fog density and salinity data, along with corresponding auxiliary parameters such as tidal cycle, wind speed and direction, and solar radiation intensity, are recorded simultaneously. Furthermore, the fog density and salinity data can be validated through a triple verification process, including: automatically removing outliers from the fog density or salinity data based on the aforementioned auxiliary parameters (e.g., removing salinity data with sudden drops caused by ship wakes); calibrating the probe using a standard sodium chloride solution to check for equipment drift errors; or verifying the spatial consistency of the fog density or salinity data using data from nearby marine observation stations. After verification, the spatiotemporal distribution characteristics of fog density are calculated based on the fog density data. These characteristics can include the diurnal and nighttime fog concentration averages, and a salinity gradient heatmap is plotted to identify areas affected by freshwater input. The final output initial environmental parameters can include the baseline values ​​and fluctuation ranges of core indicators. Fog density data is expressed using percentiles, and salinity data is annotated with tidal influence deviations.

[0077] Step 200: Based on the initial environmental parameters, activate the red light-dominant mode of the RGB-LED array, set the spectral band to start from red light, and obtain the preliminary spectral output;

[0078] Initial environmental parameters are acquired by collecting light intensity and color temperature data through sensors. Based on these parameters, a set of environmental parameters is determined, including fog density, salinity, light intensity, and color temperature data. Lighting conditions are assessed using preset thresholds. If the light intensity is below a preset threshold, the color temperature is below a preset threshold, or the fog density is above a preset threshold, the RGB-LED array is activated, generating an array activation signal. The RGB-LED array is then configured using this activation signal, setting a red light dominance mode and adjusting the red light channel duty cycle to generate a red light dominance configuration. Based on this configuration, the spectral output is controlled, and the output spectrum is detected using a spectrophotometer to obtain preliminary spectral data. For this preliminary spectral data, the K-means algorithm is applied to perform spectral clustering to determine the spectral band distribution characteristics. Based on these characteristics, the current distribution of the RGB-LED array is adjusted to optimize red light intensity control, generating an optimized spectral output. By optimizing the spectral output, the stability of the red light-dominant mode was verified using a spectral analyzer, and preliminary spectral output was obtained.

[0079] Specifically, activating the red-dominant mode requires adjusting the RGB channel ratios. The red channel PWM duty cycle is set to 90% (corresponding to a value of 230), while the green and blue channels have duty cycles of 10% (value 25) to ensure red dominance and a target color temperature of 2700K. Spectral band configuration starts with red light, with a wavelength range of 620-630nm and a peak at 625nm. A spectrophotometer is used to simulate the spectral distribution, setting the initial band to 620nm, with 5nm increments covering the 620-700nm range, generating a spectral data array. The analysis process uses Fourier transform to convert the spectral data to the frequency domain, checks if the peak wavelength is stable at 625nm, calculates the spectral variance, and obtains a standard deviation of 2.3nm, confirming that the spectral output meets expectations.

[0080] Step 300: If the fog density data is higher than a preset density threshold, the proportion of red light in the initial spectral output is locked by the RGB-LED array, and the yellow light component is superimposed to obtain the adjusted spectral combination;

[0081] When the monitoring system detects that the real-time fog density data is higher than the preset density threshold, that is, the fog density data exceeds the preset safety threshold, for example, the fog density data exceeds 0.5 g / m³ corresponding to visibility of less than 500 meters. 3 The system will automatically initiate spectral adjustment. Based on the high scattering characteristics of fog particles on short-wavelength light, it analyzes the current spectral output effect and determines the initial spectral output red light band (620-750nm) as the core lighting anchor point due to its optimal atmospheric penetration at longer wavelengths. By accessing the independent drive module of the RGB-LED array, the current output of the red light chip group is locked to a preset peak intensity, while the activation ratio of the blue light chip group (450-495nm) is dynamically suppressed to below 10% to avoid visual blurring caused by the diffusion of short-wavelength light in the fog.

[0082] Based on enhanced red-light-dominated illumination, the system simultaneously injects yellow light components (570-590nm) to obtain an adjusted spectral combination. This wavelength selection is based on two considerations: firstly, yellow light wavelengths fall between the red and green spectra, retaining strong penetrability (three times higher fog penetration coefficient than blue-green light) while supplementing the mid-range visually sensitive area lacking red light; secondly, the superposition of yellow and red light generates fog light color coordinates that meet maritime standards, avoiding color fatigue caused by pure red light. The spectral synthesis process can be achieved using PWM pulse width modulation technology. Preferably, the duty cycle of the red and yellow light chipsets is controlled at a golden ratio of 3:1, ultimately outputting a low-scattering spectral combination with amber as the main tone (main peak 620nm, secondary peak 585nm) and a color temperature of 2200K.

[0083] Step 400: Based on the salinity change trend corresponding to the real-time salinity data, a linear interpolation algorithm is used to optimize the proportion of yellow light in the adjusted spectral combination to determine the spectral ratio.

[0084] When a persistent change in sea surface salinity is detected, for example, due to tidal fluctuations or freshwater inflow causing salinity fluctuations exceeding ±2 PSU / hour, a dynamic spectral optimization program will be initiated. The current and voltage signals collected in real-time by the salinity sensor are corrected by the temperature compensation module and converted into standard salinity values ​​(accuracy ±0.1 PSU). The control unit simultaneously records the timestamp and constructs a salinity change curve to determine the salinity trend. Specifically, the slope of the salinity change curve can be used to determine whether salinity has changed. For example, if the absolute value of the curve's slope is consistently greater than a preset slope threshold, it is considered a significant salinity trend change. When a persistent change in salinity data is detected based on the salinity trend, a linear interpolation algorithm is used to optimize the proportion of yellow light in the adjusted spectral combination to obtain the spectral ratio.

[0085] Furthermore, based on the current direction (increasing / decreasing) and rate of salinity change, a linear interpolation algorithm is used to dynamically adjust the proportion of yellow light in the spectrum: In scenarios of increasing salinity (e.g., 28 PSU → 32 PSU): High-salinity water bodies enhance the absorption of short-wavelength light, requiring a gradual reduction in the proportion of yellow light. Starting with an initial salinity of 28 PSU corresponding to a yellow light proportion of 30%, and ending with a target salinity of 35 PSU corresponding to a yellow light proportion of 15%, the yellow light weight is calculated using linear interpolation based on real-time salinity values. For example, when salinity rises to 30 PSU, the yellow light proportion is 25.7%. In scenarios of decreasing salinity (e.g., 32 PSU → 28 PSU): Increased freshwater input increases water body transmittance, requiring an increase in yellow light to compensate for visual clarity. Using a baseline of 15% yellow light proportion at 35 PSU and 30% yellow light proportion at 28 PSU, when salinity drops to 29 PSU, the yellow light proportion is 27.9%.

[0086] Step 500: Based on the spectral ratio, the bands of the spectral ratio are scanned by a spectral analyzer to determine the spectral matching result, wherein the spectral matching result includes the degree of matching between the spectral output and the band required for sea area penetration;

[0087] After dynamic adjustment of the spectral ratio, the system uses an integrated fiber optic spectrometer to perform high-precision verification of the actual output spectrum of the RGB-LED array. This device performs a full-band scan in the 400-800nm ​​visible light range at a sampling frequency of 5 times per second, focusing on capturing the intensity distribution of the target bands (620-750nm in the red region and 570-590nm in the yellow region) to determine the spectral ratio bands. Then, the spectral ratio bands are compared with the required sea penetration bands to obtain the spectral matching results. The required sea penetration bands include the red light penetration wavelength range and the yellow light absorption wavelength range, the band area ratio of the red and yellow light bands being within the deviation threshold of the target ratio, and the range of seawater transmittance data. The spectral matching results can also be determined by the spectral matching index between the spectral output and the required sea penetration bands.

[0088] Furthermore, the matching degree analysis of the spectral band ratio with the required band for sea area penetration includes three core dimensions: **Main peak position consistency:** Checking whether the red light main peak is stable at 635±2nm (optimal penetration wavelength) and whether the yellow light secondary peak remains at 585±3nm (avoiding moisture absorption peaks). **Energy ratio accuracy:** Calculating the integral area ratio of the red and yellow light bands, and strictly controlling the deviation threshold of the target ratio (e.g., 60% red light: 25% yellow light) within ±3%. **Sea area penetration adaptability:** Importing the measured spectrum into a pre-stored seawater transmittance database (containing absorption curves under different salinity and turbidity) to simulate and calculate the spectral penetration efficiency in a real-time salinity environment (e.g., 30 PSU). Key verifications include whether the 570-590nm yellow light band avoids the strong absorption band of high-salinity seawater (520-540nm), and whether the scattering attenuation rate of the red light band in a fog-salt coupling environment is less than 15%. The analysis results generate a spectral matching index, which can be calculated by combining band overlap (weight 40%), transmission efficiency (weight 50%), and color coordinate deviation (weight 10%). When the spectral matching index is greater than or equal to 0.9, the optimization is considered effective; if 0.8 ≤ spectral matching index < 0.9, the system fine-tunes the yellow light drive current to compensate for intensity drift; if the spectral matching index < 0.8, an alarm is triggered, the system reverts to the basic red light mode, and manual diagnostics are initiated.

[0089] Step 600: If the spectral matching result does not reach the preset matching degree, the gradient descent algorithm is used to iteratively adjust the ratio of red and yellow light in the spectral composition to obtain the target spectral combination that meets the preset matching degree.

[0090] The target spectral combination is used for illumination execution and adjustment of light characteristics of the light source device, which includes the RGB-LED array.

[0091] When the spectral matching result does not reach the preset matching degree, the system will initiate a gradient descent algorithm to iteratively optimize the ratio of red to yellow light. This process first uses the current spectral ratio (e.g., 60% red light / 25% yellow light) as the initial point, calculates the negative gradient direction of the spectral matching index, i.e., by slightly perturbing the proportion of light (±0.5%) and measuring the rate of change of the spectral matching result, to determine the fastest optimization path to improve the matching degree. If the deviation in the energy ratio of red to yellow light is detected as the main mismatch factor (contribution > 70%), the algorithm will gradually adjust the weights of red and yellow light along the gradient direction: each iteration synchronously modifies the PWM duty cycle of the red and yellow light driving signals, i.e., modifies the ratio of red to yellow light. The initial step size can be set to 2%, and it will dynamically shrink as the convergence speed increases.

[0092] The iteration process can incorporate three intelligent constraints: physical boundary protection: the proportion of red light is always anchored in the basic penetration range of 55-65%, and the proportion of yellow light is strictly limited to an adjustable window of 15-30%, avoiding the sacrifice of core optical performance in pursuit of matching degree; environmental parameter coupling: real-time salinity data is substituted into the transmittance prediction model, and the attenuation curve of the output spectrum in the real sea area is simulated after each ratio adjustment to ensure that the optimization direction conforms to the current optical characteristics of the water body; local optimum avoidance: when the matching degree is less than 0.5% after 3 consecutive iterations, random perturbation (yellow light ±3% mutation) is actively injected to escape potential convergence traps.

[0093] The intelligent nighttime sensing and dimming method for mobile solar-powered lighthouses provided by this invention acquires initial environmental parameters, activates a red light-dominant mode, and locks the red light ratio based on a fog density threshold. Yellow light is then superimposed on this red light ratio to form a preliminary spectral output. If salinity changes, a linear interpolation algorithm is used to optimize the yellow light ratio in the spectral combination based on the salinity change trend, determining the spectral ratio. The matching degree is then verified using a spectral analyzer. If the spectral ratio does not meet expectations, a gradient descent algorithm is used to iteratively adjust the red and yellow light ratio to ensure the spectrum adapts to the penetration requirements of the sea area. This achieves dynamic spectral combination and real-time band iteration based on environmental variables in mobile solar-powered lighthouses, ensuring the penetration and stability of light in complex foggy conditions.

[0094] In one embodiment, please refer to Figure 2 It also includes:

[0095] Step 700: Based on the target spectral combination, control the output spectrum of the RGB-LED array and continuously monitor environmental indicators using a closed-loop feedback mechanism;

[0096] Step 800: If the environmental indicators change, then reacquire fog density data and salinity data as an environmental data set.

[0097] Step 900: Based on the environmental data set, a weighted average algorithm is used to calculate the comprehensive impact factors of fog density and salinity to obtain the environmental impact weights;

[0098] Step 1000: Based on the environmental impact weights and combined with the preset spectral ratio model, update the spectral ratio data of the target spectral combination to obtain the adjusted spectral ratio.

[0099] Step 1100: If the adjusted spectral ratio exceeds a preset threshold range, the adjusted spectral ratio is optimized using a linear regression algorithm to determine the optimized spectral ratio.

[0100] Step 1200: Based on the optimized spectral ratio, use a spectral analysis tool to calculate the light distribution parameters and obtain a set of light distribution parameters;

[0101] The light distribution parameter set is used to perform illumination and adjust the characteristics of light in the light source device. The light source device includes the RGB-LED array, and the spectral ratio model is determined based on a preset spectral ratio training set.

[0102] After obtaining the target spectral combination, the RGB-LED array is controlled to output the spectrum of that target spectral combination, and a closed-loop feedback mechanism is used to continuously monitor the environmental indicators of the mobile solar-powered lighthouse when outputting the target spectral combination. If a change in environmental indicators is detected, fog density and salinity data are collected through sensors to obtain an environmental data set. Based on the environmental data set, a preset weighted average algorithm is used to calculate the comprehensive influence factor of fog density and salinity, determining the environmental impact weight. Then, using the environmental impact weight and a preset spectral ratio model, the spectral ratio data of the target spectral combination is updated to obtain the adjusted spectral ratio. If the adjusted spectral ratio exceeds a preset threshold range, the ratio parameters of the adjusted spectral ratio are optimized using a linear regression algorithm to determine the optimized spectral ratio. Based on the optimized spectral ratio, spectral analysis tools are used to calculate the light distribution parameters, obtaining a light distribution parameter set. This light distribution parameter set is used to determine the characteristics of the illumination execution and adjustment light of the light source device, which includes an RGB-LED array. Therefore, by adjusting the output configuration of the light source device through the set of light distribution parameters, the final light source output scheme is determined, so that the light source device generates control commands according to the final light source output scheme and sends them to the light source device, so that the light source device can complete the real-time adjustment of the spectral ratio according to the control commands.

[0103] Furthermore, the transmission effect of the light distribution parameters can be verified cyclically using a spectral output device. Real-time data processing is employed to maintain the stability of the spectral combination, resulting in the final spectral distribution result. Initial light distribution parameters are acquired through the spectral output device, and the initial spectral distribution of the spectral output device is determined using a spectral analysis algorithm, yielding initial light distribution data. Based on the initial light distribution data, a cyclic verification mechanism is used to detect the transmission effect. If the detection result deviates from a preset transmission threshold, the parameters of the spectral output device are adjusted, resulting in adjusted light distribution data. The adjusted light distribution data is analyzed through a real-time data processing module, and the frequency characteristics of the spectral combination are extracted using a fast Fourier transform algorithm, yielding a stability index for the spectral combination. Based on the stability index of the spectral combination, it is determined whether the stability meets a preset stability threshold. If not, the light distribution parameters of the spectral output device are adjusted using a parameter optimization algorithm, resulting in optimized light distribution data. The optimized light distribution data is then cyclically verified again using the spectral output device to obtain real-time feedback data on the transmission effect, resulting in verified light distribution data. Based on the verified light distribution data, data fusion technology is used to evaluate the stability of the spectral combination, yielding the final light distribution result. The final photometric results are formatted by the result acquisition module to generate standardized spectral photometric data, resulting in an outputtable final result.

[0104] This embodiment integrates multiple environmental factors through weighted averaging, then uses a preset model for rapid response, and finally employs linear regression and spectral analysis for bottom-line optimization, forming a three-layer response mechanism. This overcomes the limitations of traditional single-factor regulation and achieves a dynamic weighted mechanism for both fog density and salinity, solving the problem of the synergistic influence of multiple attenuation effects in complex marine environments. By using a linear regression constraint, the neural network output is forcibly mapped to a physically realizable parameter space, eliminating the risk of spectral runaway due to model overfitting. From environmental perception, weight calculation, model prediction, boundary optimization to parameter verification, an autonomous decision-making closed loop is formed, resulting in a significantly faster response speed than traditional methods and meeting the needs of scenarios with changing environments.

[0105] In one embodiment, please refer to Figure 3 It also includes:

[0106] Step 1300: Obtain GPS data for the mobile solar-powered lighthouse;

[0107] Step 1400: If the fog density data exceeds a preset density threshold, the fog density change trend is predicted by a linear regression algorithm to obtain a predicted fog density value.

[0108] Step 1500: Based on the predicted fog density and the salinity data, determine the comprehensive marine environmental index;

[0109] Step 1600: If the comprehensive marine environment index is greater than the preset environmental state threshold, then based on the GPS data, update the lighthouse operating parameters to obtain the real-time operating status of the mobile solar lighthouse.

[0110] Step 1700: Based on the real-time operating status and the initial environmental parameters, classify the lighthouse operating modes using the K-means clustering algorithm to determine the target operating mode;

[0111] The target operating mode is used to activate the red light-dominant mode of the RGB-LED array in combination with the initial environmental parameters to obtain a preliminary spectral output.

[0112] GPS data provides the real-time latitude and longitude coordinates of the lighthouse, which can be obtained through a GPS module to acquire the geographical location information of the mobile solar-powered lighthouse. Sensors are used to collect fog density and salinity data in the sea area, generating environmental monitoring data. If the collected fog density data exceeds a preset density threshold, a linear regression algorithm is used to predict the fog density change trend, obtaining a predicted fog density value. This predicted fog density value characterizes the fog density change trend over a period of time. After obtaining the predicted fog density value, the marine environmental comprehensive index I is calculated based on the predicted fog density value and salinity data. The formula for calculating the marine environmental comprehensive index is as follows:

[0113] I = w1 * Mhw2 * S

[0114] Where M is the predicted fog density, S is the salinity data, and w1 and w2 are preset weights to determine the overall environmental status. The overall environmental index is compared with a preset environmental status threshold. If the overall environmental index is greater than the preset environmental status threshold, the solar power supply of the lighthouse is adjusted, generating a power adjustment command. Based on the power adjustment command and geographical location information, the lighthouse operating parameters are updated to obtain the real-time operating status. Using the real-time operating status and environmental monitoring data, the lighthouse operating modes are classified using a K-means clustering algorithm to determine the optimized target operating mode. The target operating mode of the solar-powered lighthouse is obtained, and combined with the initial environmental parameters, the red light-dominant mode of the RGB-LED array is activated to obtain the preliminary spectral output.

[0115] Furthermore, a high-precision GPS chip can be used to acquire location data, collecting latitude and longitude data once per second with an accuracy of 2.5 meters.

[0116] In this embodiment, environmental prediction, location services, and operational mode decision-making are linked in three dimensions to achieve a dual-drive mechanism of "GPS positioning + fog prediction," breaking through the spatiotemporal limitations of traditional environmental monitoring. Based on electronic chart data binding location and optical strategies, the same lighthouse automatically outputs differentiated spectra in waterways, fishing areas, and protected areas, improving compliance while reducing the rate of ecological disturbance complaints. K-means clustering dynamically updates the center point coordinates every 24 hours, continuously learning marine environmental patterns, resulting in a lower misjudgment rate compared to fixed threshold classification methods. Through linear regression prediction of fog density, the identification of high-risk conditions is advanced from "after the event occurs" to "15-30 minutes before the event occurs," providing a critical time window for ship avoidance.

[0117] In one embodiment, please refer to Figure 4 If the fog density data is higher than a preset density threshold, the proportion of red light in the initial spectral output is locked by the RGB-LED array, and a yellow light component is superimposed to obtain an adjusted spectral combination, including:

[0118] Step 301: If the fog density data is higher than a preset density threshold, then based on the fog density data, the fog density data is processed by a spectral analysis algorithm to determine the red light ratio adjustment requirement and obtain the red light fixed ratio parameter;

[0119] Step 302: Based on the fixed red light ratio parameter, control the RGB-LED array to fix the red light ratio and generate an initial spectral combination;

[0120] Step 303: Based on the initial spectral combination, the yellow light component is superimposed to obtain the intermediate spectral combination;

[0121] Step 304: Based on the intermediate spectral combination, a spectral analysis algorithm is used to analyze the intermediate spectral combination to obtain the spectral characteristics of the intermediate spectral combination;

[0122] Step 305: Based on the spectral features and the preset target attributes, determine the adjusted spectral combination.

[0123] If the fog density exceeds a preset threshold, environmental fog density data is collected by sensors, and a signal processing algorithm is used to determine the fog density value. The fog density value is then processed using a spectral analysis algorithm to determine the red light ratio adjustment requirement, resulting in a fixed red light ratio parameter. Based on this parameter, the RGB-LED array is controlled to maintain a fixed red light output ratio, generating an initial spectral combination. This initial spectral combination is then superimposed with a yellow light component, and the yellow light intensity is adjusted via the light source control module to obtain an intermediate spectral combination. A spectral analysis algorithm is used to detect the intermediate spectral combination, determining if it meets preset spectral requirements and obtaining its spectral characteristics. Based on the spectral characteristics of the intermediate spectral combination and preset target attributes, the output of the RGB-LED array is adjusted to generate the final spectral combination, resulting in the adjusted spectral combination. For example, based on target attributes where the penetrability target exceeds a preset band threshold, the spectral characteristics of the intermediate spectral combination, including bimodal position, energy ratio, or background noise, can be iteratively optimized until the penetrability target exceeds the preset band threshold, resulting in the adjusted spectral combination. Furthermore, the output effect of the final spectral combination can be verified using an environmental sensing module to determine its stability.

[0124] This embodiment proposes a dynamic spectral optimization process to address the penetration problem in foggy weather in a phased manner. First, the basic proportion of red light is locked to ensure penetration, then yellow light is superimposed to improve visibility, and finally, fine-tuning is achieved through spectral analysis. This design is more intelligent than the traditional single spectral output, solves the risk of insufficient penetration caused by the red light fixing mechanism, uses yellow light superposition to solve visual fatigue caused by pure red light, and ensures spectral effectiveness through closed-loop verification.

[0125] In one embodiment, please refer to Figure 5 The step of optimizing the proportion of yellow light in the adjusted spectral combination based on the salinity change trend corresponding to real-time salinity data and determining the spectral ratio using a linear interpolation algorithm includes:

[0126] Step 401: The real-time salinity data is processed using a linear interpolation algorithm to generate a continuous salinity distribution characterizing salinity changes and to determine the salinity change trend.

[0127] Step 402: Based on the salinity change trend, calculate the yellow light ratio adjustment coefficient to obtain a preliminary ratio adjustment scheme;

[0128] Step 403: If the preliminary ratio adjustment scheme meets the preset spectral ratio threshold, then the ratio of the preliminary ratio adjustment scheme is optimized by the spectral analysis model to determine the first ratio adjustment scheme.

[0129] Step 404: If the preliminary ratio adjustment scheme does not meet the spectral ratio threshold, then the mapping relationship between salinity change and yellow light ratio is recalculated through regression analysis algorithm to obtain a second ratio adjustment scheme.

[0130] Step 405: Determine the spectral ratio based on the first ratio adjustment scheme or the second ratio adjustment scheme.

[0131] Real-time salinity data of the marine environment is acquired by sensors. A linear interpolation algorithm is used to process the real-time salinity data, generating a continuous salinity distribution to determine the salinity change trend. Based on the salinity change trend, the proportion adjustment coefficient of the yellow light component is calculated to obtain a preliminary proportion adjustment scheme. If the preliminary proportion adjustment scheme meets a preset spectral ratio threshold, the proportion of the preliminary proportion adjustment scheme is optimized using a spectral analysis model to determine the optimized yellow light component proportion, resulting in a first proportion adjustment scheme. If the preliminary proportion adjustment scheme does not meet the preset spectral ratio threshold, a regression analysis algorithm is used to recalculate the mapping relationship between salinity change and the yellow light component proportion, resulting in a second proportion adjustment scheme. Then, a spectral ratio is generated based on either the first or second proportion adjustment scheme. The output of the yellow light component of the light source device is adjusted according to the ratio of red and yellow light in the spectral ratio to determine the operating parameters of the light source device.

[0132] This embodiment shortens the system's response delay time when a salinity front passes by by combining linear interpolation with rate of change calculation, which is 6 times faster than the traditional moving average method. Furthermore, through a dual-path decision mechanism, fine-tuning the spectral model when the threshold is met is an energy-saving strategy, while initiating regression analysis when the threshold is not met is a safety net. This ensures efficiency under normal conditions and solves the reliability problem when extreme salinity changes occur.

[0133] In one embodiment, please refer to Figure 6 The step of determining the spectral matching result by scanning the spectral bands of the spectral ratio using a spectral analyzer based on the spectral ratio includes:

[0134] Step 501: Based on the spectral ratio, scan the bands of the spectral ratio using a spectral analyzer to generate an initial spectral distribution;

[0135] Step 502: If the band range of the initial spectral distribution does not match the preset sea area penetration requirement band, then adjust the scanning parameters of the spectral analyzer to obtain the optimized spectral distribution corresponding to the spectral ratio.

[0136] Step 503: Compare the optimized spectral distribution with the required waveband for sea area penetration to determine the matching degree score;

[0137] Step 504: If the matching degree score is lower than the preset score threshold, the bands of the spectral ratio are reconfigured to generate updated spectral data.

[0138] Step 505: Based on the updated spectral data, classify it using a support vector machine algorithm to determine the classification result, wherein the classification result includes the degree of matching between the updated spectral data and the required waveband for sea area penetration;

[0139] Step 506: Based on the classification results, iteratively adjust the band parameters to obtain the spectral matching results.

[0140] The raw spectral data of the spectral bands are acquired using a spectrometer to generate an initial spectral distribution. If the band range of the initial spectral distribution does not match the preset sea penetration requirement bands, the scanning parameters of the spectrometer are adjusted to obtain an optimized spectral distribution. The optimized spectral distribution is compared with the preset sea penetration requirement model, and the matching score is calculated to determine the consistency of the spectral output. Through consistency analysis, if the matching score is lower than a preset threshold, the spectral bands are reconfigured to generate updated spectral data. Based on the updated spectral data, a support vector machine algorithm is used to classify the updated spectral data, obtaining the classification result. The classification result is used to determine whether the updated spectral data meets the sea penetration requirement. The classification result includes the matching degree between the updated spectral data and the sea penetration requirement bands. It can be seen that if the matching score in the classification result is greater than the preset threshold, the updated spectral data meets the sea penetration requirement and no optimization is needed; if the matching score in the classification result is less than or equal to the preset threshold, the classification result indicates that the requirement is not met, and the band parameters are iteratively adjusted to obtain new spectral band data. The new spectral ratio data was used to conduct a final verification with the marine penetration demand model, and spectral matching results were generated.

[0141] Specifically, the spectral composition is scanned using a spectrometer, initially set to a scanning range of 400-1000 nanometers, covering the visible and near-infrared regions to meet the penetration requirements of the sea area. A high-resolution spectrometer is used to acquire spectral intensity data of the spectral composition in 0.5-nanometer increments. Assuming the target composition is 40% blue light (450 nm), 35% green light (550 nm), and 25% red light (650 nm), actual spectral data is acquired using the instrument to generate a spectral curve. To verify whether the spectral output matches the sea area penetration requirements, the root mean square error (RMSE) algorithm is used to calculate the deviation between the actual spectrum and the target composition. The formula for calculating RMSE is as follows:

[0142]

[0143] Among them, I 实际For the measured strength, I 目标 Let n be the target intensity and n be the number of band points. Assuming the measured intensity is 0.42 at 450 nm, 0.33 at 550 nm, and 0.26 at 650 nm, the calculated RMSE is 0.0173. Analyzing the RMSE value, if it is less than 0.02, it indicates a good match between the spectral output and the ocean penetration requirements, suitable for deep-sea optical communication or underwater detection. Otherwise, adjust the light source drive current, for example, increasing the current in the 450 nm band from 100 mA to 120 mA, re-acquire data, and repeat the calculation until the RMSE meets the threshold. Finally, generate a spectral matching result report, including band intensity, deviation value, and matching conclusion, stored in JSON format for easy system integration and subsequent analysis.

[0144] In this embodiment, the dynamic adjustment mechanism of spectrometer parameters maintains 99% data validity even in environments with strong electromagnetic interference (such as ship radar). Compared with traditional fixed parameter scanning, the data availability rate is increased by 45%, which is especially suitable for complex maritime scenarios. The innovative integration of dual evaluation of band coverage and intensity matching solves the shortcomings of traditional methods that focus on position rather than energy. The parameters are adjusted in a targeted manner based on the classification results to avoid blindly traversing and optimizing.

[0145] In one embodiment, please refer to Figure 7 If the spectral matching result does not reach the preset matching degree, a gradient descent algorithm is used to iteratively adjust the ratio of red and yellow light in the spectral composition to obtain a target spectral combination that satisfies the preset matching degree, including:

[0146] Step 601: If the spectral matching result does not reach the preset matching degree, the matching error value corresponding to the spectral matching result is calculated by the spectral analysis model to obtain the initial red light ratio and yellow light ratio.

[0147] Step 602: Based on the matching error value, the gradient direction of the initial red light ratio and yellow light ratio is calculated using the gradient descent algorithm to determine the adjustment step size parameter;

[0148] Step 603: Based on the adjustment step size parameter, update the initial red light ratio and yellow light ratio to obtain a new spectral combination;

[0149] Step 604: Based on the new spectral combination, recalculate the new matching error value using the spectral analysis model;

[0150] Step 605: If the new matching error value does not meet the convergence judgment condition, the gradient descent algorithm is used to iteratively optimize the new matching error value to determine the target spectral combination.

[0151] If the spectral matching result does not reach the preset matching degree, i.e., the spectral matching result does not meet expectations, the matching error value corresponding to the spectral matching result is calculated using the spectral analysis model to obtain the initial red light ratio and yellow light ratio. Based on the matching error value, the gradient descent algorithm is used to calculate the gradient direction of the initial red light ratio and yellow light ratio to determine the adjustment step size parameter. The red light ratio and yellow light ratio are updated by adjusting the step size parameter to obtain a new spectral combination. For the new spectral combination, the spectral analysis model is used to recalculate the new matching error value to determine whether the convergence judgment condition is met. If the matching error value does not meet the convergence judgment condition, the gradient descent algorithm is returned to continue iterative optimization to obtain the updated red light ratio and yellow light ratio. Through the iterative optimization process, the red light ratio and yellow light ratio that meet the convergence judgment condition are obtained, and the target spectral combination is determined. Furthermore, based on the target spectral combination, the spectral analysis model can also be used to verify the matching error value to obtain a further optimized spectral combination.

[0152] For example, during spectral matching, if the initial spectral matching result does not meet expectations—for instance, if the mean square error (MSE) between the red (wavelength 630nm) and yellow (wavelength 580nm) ratios of the target spectrum and the standard spectrum is greater than 0.05—then a gradient descent algorithm is needed to iteratively adjust the red and yellow ratios to optimize the spectral combination. First, define the initial ratios, assuming a red ratio of 0.6 and a yellow ratio of 0.4. The red and yellow intensities of the target spectrum are 0.55 and 0.45, respectively. Calculate the mean square error (MSE) between the current spectrum and the target spectrum. The formula for calculating the MSE is as follows:

[0153]

[0154] Among them, I 真实 The spectral intensity of the current spectrum. This represents the average spectral intensity of the target spectrum.

[0155] The calculated MSE is 0.005. The learning rate is set to 0.01, and the iteration count is capped at 1000 or stops when the MSE is less than 0.001. The gradient descent algorithm updates the proportions by calculating the partial derivatives of the MSE with respect to the red and yellow light proportions. The MSE is recorded for each iteration. If, after the 50th iteration, the MSE drops to 0.0009, which is less than the threshold of 0.001, the iteration stops, resulting in an optimized spectral combination of 0.5502 for red and 0.4498 for yellow. If the target is not met after 1000 iterations, the combination with the lowest MSE is selected. The final spectral combination is verified using a spectral analyzer to confirm that the difference in chromaticity coordinates between the combination and the target spectrum is less than 0.01, ensuring matching accuracy.

[0156] In this embodiment, the spectral matching problem is transformed into an error function optimization problem. By using a red / yellow light error separation calculation mechanism, the band crosstalk problem caused by traditional mixed gradients is solved, thereby improving the accuracy of spectral combination calculation.

[0157] The intelligent night-sensing dimming device for a mobile solar-powered lighthouse provided by the present invention will be described below. The intelligent night-sensing dimming device for a mobile solar-powered lighthouse described below can be referred to in correspondence with the intelligent night-sensing dimming method for a mobile solar-powered lighthouse described above.

[0158] This invention also provides an intelligent night-time sensing dimming system for a mobile solar-powered lighthouse, comprising:

[0159] The data acquisition module is used to collect fog density and salinity data of the marine environment to determine the initial environmental parameters;

[0160] The activation module is used to activate the red light-dominant mode of the RGB-LED array based on the initial environmental parameters, set the spectral band to start from red light, and obtain the initial spectral output;

[0161] A fixing module is used to lock the proportion of red light in the initial spectral output through the RGB-LED array and superimpose the yellow light component if the fog density data is higher than a preset density threshold, so as to obtain an adjusted spectral combination.

[0162] The optimization module is used to optimize the proportion of yellow light in the adjusted spectral combination based on the salinity change trend corresponding to real-time salinity data, and to determine the spectral ratio.

[0163] The scanning module is used to scan the bands of the spectral ratio using a spectral analyzer based on the spectral ratio to determine the spectral matching result, wherein the spectral matching result includes the degree of matching between the spectral output and the band required for sea area penetration;

[0164] The adjustment module is used to iteratively adjust the ratio of red and yellow light in the spectral composition using a gradient descent algorithm if the spectral matching result does not reach the preset matching degree, so as to obtain the target spectral combination that meets the preset matching degree.

[0165] Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840. The processor 810, communications interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions from the memory 830 to execute the intelligent night-sensing dimming method of the mobile solar-powered lighthouse.

[0166] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0167] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent night-sensing dimming method for a mobile solar lighthouse provided by the methods described above.

[0168] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0169] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent nighttime sensing and dimming of a mobile solar-powered lighthouse, characterized in that, Executed by a computer, including: Collect fog density and salinity data of the marine environment to determine the initial environmental parameters; Based on the initial environmental parameters, the red light-dominant mode of the RGB-LED array is activated, and the spectral band is set to start from red light to obtain the preliminary spectral output; If the fog density data is higher than a preset density threshold, the proportion of red light in the initial spectral output is locked by the RGB-LED array, and a yellow light component is superimposed to obtain an adjusted spectral combination. Based on the salinity change trend corresponding to real-time salinity data, a linear interpolation algorithm is used to optimize the proportion of yellow light in the adjusted spectral combination to determine the spectral ratio. Specifically, this includes: processing the real-time salinity data using a linear interpolation algorithm to generate a continuous salinity distribution characterizing salinity changes and determining the salinity change trend; calculating the yellow light proportion adjustment coefficient based on the salinity change trend to obtain a preliminary proportion adjustment scheme; if the preliminary proportion adjustment scheme meets a preset spectral ratio threshold, optimizing the proportion of the preliminary proportion adjustment scheme using a spectral analysis model to determine a first proportion adjustment scheme; if the preliminary proportion adjustment scheme does not meet the spectral ratio threshold, recalculating the mapping relationship between salinity change and the yellow light proportion using a regression analysis algorithm to obtain a second proportion adjustment scheme; and determining the spectral ratio based on either the first or second proportion adjustment scheme. Based on the spectral ratio, the bands of the spectral ratio are scanned using a spectral analyzer to determine the spectral matching result. The spectral matching result includes the degree of matching between the spectral output and the required sea area penetration bands. Specifically, this includes: based on the spectral ratio, scanning the bands of the spectral ratio using a spectral analyzer to generate an initial spectral distribution; if the band range of the initial spectral distribution does not match the preset sea area penetration requirement bands, adjusting the scanning parameters of the spectral analyzer to obtain an optimized spectral distribution corresponding to the spectral ratio; comparing the optimized spectral distribution with the required sea area penetration bands to determine a matching degree score; if the matching degree score is lower than a preset score threshold, reconfiguring the bands of the spectral ratio to generate updated spectral data; based on the updated spectral data, classifying it using a support vector machine algorithm to determine the classification result, wherein the classification result includes the degree of matching between the updated spectral data and the required sea area penetration bands; and based on the classification result, iteratively adjusting the band parameters to obtain the spectral matching result. If the spectral matching result does not reach the preset matching degree, the gradient descent algorithm is used to iteratively adjust the red and yellow light ratios of the spectral composition to obtain a target spectral combination that meets the preset matching degree. Specifically, this includes: if the spectral matching result does not reach the preset matching degree, calculating the matching error value corresponding to the spectral matching result using a spectral analysis model to obtain the initial red and yellow light ratios; based on the matching error value, using the gradient descent algorithm to calculate the gradient direction of the initial red and yellow light ratios to determine the adjustment step size parameter; based on the adjustment step size parameter, updating the initial red and yellow light ratios to obtain a new spectral combination; based on the new spectral combination, recalculating the new matching error value using the spectral analysis model; if the new matching error value does not meet the convergence judgment condition, continuing to use the gradient descent algorithm to iteratively optimize the new matching error value to determine the target spectral combination. The target spectral combination is used for illumination execution and adjustment of light characteristics of the light source device, which includes the RGB-LED array.

2. The intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse according to claim 1, characterized in that, Also includes: Based on the target spectral combination, the output spectrum of the RGB-LED array is controlled, and a closed-loop feedback mechanism is used to continuously monitor environmental indicators. If the environmental indicators change, fog density data and salinity data are reacquired as the environmental data set. Based on the aforementioned environmental data set, a weighted average algorithm is used to calculate the comprehensive impact factors of fog density and salinity, thereby obtaining the environmental impact weights. Based on the environmental impact weights and combined with the preset spectral ratio model, the spectral ratio data of the target spectral combination is updated to obtain the adjusted spectral ratio. If the adjusted spectral ratio exceeds the preset threshold range, the adjusted spectral ratio is optimized using a linear regression algorithm to determine the optimized spectral ratio. Based on the optimized spectral ratio, the light distribution parameters are calculated using spectral analysis tools to obtain a set of light distribution parameters. The spectral ratio model is determined based on a preset spectral ratio training set.

3. The intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse according to claim 1, characterized in that, Also includes: Acquire GPS data from the mobile solar-powered lighthouse; If the fog density data exceeds a preset density threshold, the fog density change trend is predicted using a linear regression algorithm to obtain a predicted fog density value. Based on the predicted fog density and the salinity data, a comprehensive marine environmental index is determined. If the comprehensive marine environment index is greater than the preset environmental state threshold, the lighthouse operating parameters are updated based on the GPS data to obtain the real-time operating status of the mobile solar lighthouse. Based on the real-time operating status and the initial environmental parameters, the lighthouse operating modes are classified using the K-means clustering algorithm to determine the target operating mode; The target operating mode is used to activate the red light-dominant mode of the RGB-LED array in combination with the initial environmental parameters to obtain a preliminary spectral output.

4. The intelligent nighttime sensing dimming method for a mobile solar-powered lighthouse according to claim 1, characterized in that, If the fog density data is higher than a preset density threshold, the proportion of red light in the initial spectral output is locked by the RGB-LED array, and a yellow light component is superimposed to obtain an adjusted spectral combination, including: If the fog density data is higher than the preset density threshold, then based on the fog density data, the fog density data is processed by a spectral analysis algorithm to determine the red light ratio adjustment requirement and obtain the red light fixed ratio parameter; Based on the red light fixed ratio parameter, the red light ratio of the RGB-LED array is fixed to generate an initial spectral combination; Based on the initial spectral combination, the yellow light component is superimposed to obtain the intermediate spectral combination; Based on the intermediate spectral combination, a spectral analysis algorithm is used to analyze the intermediate spectral combination to obtain its spectral characteristics; Based on the spectral characteristics and the preset target attributes, the adjusted spectral combination is determined.

5. An intelligent night-time sensing dimming system for a mobile solar-powered lighthouse, characterized in that, The intelligent nighttime sensing dimming method for mobile solar-powered lighthouses as described in any one of claims 1-4 includes: The data acquisition module is used to collect fog density and salinity data of the marine environment to determine the initial environmental parameters; The activation module is used to activate the red light-dominant mode of the RGB-LED array based on the initial environmental parameters, set the spectral band to start from red light, and obtain the initial spectral output; A fixing module is used to lock the proportion of red light in the initial spectral output through the RGB-LED array and superimpose the yellow light component if the fog density data is higher than a preset density threshold, so as to obtain an adjusted spectral combination. The optimization module is used to optimize the proportion of yellow light in the adjusted spectral combination based on the salinity change trend corresponding to real-time salinity data, and to determine the spectral ratio. The scanning module is used to scan the bands of the spectral ratio using a spectral analyzer based on the spectral ratio to determine the spectral matching result, wherein the spectral matching result includes the degree of matching between the spectral output and the band required for sea area penetration; An adjustment module is used to iteratively adjust the ratio of red and yellow light in the spectral composition using a gradient descent algorithm if the spectral matching result does not reach a preset matching degree, so as to obtain a target spectral combination that meets the preset matching degree. The target spectral combination is used for illumination execution and adjustment of light characteristics of the light source device, which includes the RGB-LED array.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the intelligent night-sensing dimming method for the mobile solar-powered lighthouse as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the intelligent night-sensing dimming method for the mobile solar lighthouse as described in any one of claims 1 to 4.

Images