A method for optimizing control parameters of a public space lighting system

Abstract

This invention relates to the field of intelligent lighting control in public spaces and optimization of control parameters, and discloses a method for optimizing control parameters of a public space lighting system. The scheme involves deploying multiple disturbance sensors in the public space, establishing a voltage-acceleration conversion through two-point calibration, and converting the voltage to calibration acceleration in real time during operation; weighting and fusing the equivalent disturbance from multiple data points within a time window, setting it to zero during startup; setting a reference brightness and maximum offset and limiting the amplitude; constructing a normalized response based on time averaging, setting it to zero in extreme cases; obtaining the variance and adaptive suppression ratio from the deviation, and then using residual weights and modulation corrections to output the final brightness; forming a disturbance trajectory and trend according to the refresh cycle, predicting the next cycle and limiting the amplitude, and generating timing commands.

Description

Technical Field

[0001] This invention relates to the field of intelligent lighting control in public spaces and optimization of control parameters, specifically a method for optimizing control parameters of a public space lighting system. Background Technology

[0002] Currently, public spaces such as underpasses, bridge arches, and parking areas under interchanges in cities generally employ fixed or adjustable lighting systems to improve safety and pedestrian comfort. Existing technologies commonly employ methods such as maintaining a constant illuminance for extended periods; switching brightness according to a schedule; or using simple light sensors or human body sensors for on / off control and coarse dimming. These systems typically only control slowly changing factors such as ambient brightness and pedestrian traffic, with control parameters largely based on experience, making fine-tuning difficult based on the dynamic characteristics of environmental disturbances.

[0003] In the space beneath urban overpasses, factors such as structural vibrations caused by vehicle loads, distant traffic shockwaves, and disturbances from surrounding construction often act on lighting equipment and the user's field of vision in a multi-source, time-delayed manner. For example, micro-vibrations of the road surface and bridge structure can cause complex relative displacements through the lamp holders, reflectors, and the position of the human eye, resulting in slight but frequent fluctuations in perceived illuminance over a short period. Existing lighting control technologies typically do not collect or analyze these disturbance signals, nor do they systematically map information such as the intensity, amplitude, and temporal distribution of disturbances into lighting control parameters. This can easily lead to perceptible flickering and brightness fluctuations when traffic flow changes or environmental disturbances suddenly increase, affecting the visual comfort and sense of safety for drivers and pedestrians. On the other hand, existing technologies often use fixed safety margins for setting lighting control parameters. To avoid insufficient illuminance in any situation, the upper limit of brightness is often set high, but there is a lack of a mechanism for fine-tuning based on the statistical characteristics of disturbances, making it difficult to converge to a brightness range that matches the actual disturbance level in a timely manner. This can easily lead to maintaining excessively high brightness even when disturbances are small or relatively stable, resulting in low energy efficiency. Furthermore, when disturbances are abrupt and have a delayed superposition effect, there is no way to perform feedforward time compensation for brightness changes, forcing reliance on simple thresholds or hysteresis adjustments, resulting in response lag and over-adjustment. In addition, existing public space lighting control schemes often focus on feedback control based on a single sensor, such as dimming based solely on a lux meter at a specific location or traffic flow detection results. They lack the ability to perform time-weighted fusion and statistical feature analysis of disturbance information from multiple measurement points, failing to extract information such as the average level, variance, and trend of the disturbance within a time window, let alone establish a clear mathematical mapping relationship between this information and control parameters. Therefore, existing technologies struggle to achieve refined, predictable, and parameterized optimization control of brightness output in real-world environments with multi-source delayed disturbances.

[0004] Therefore, this case aims to propose a method for optimizing control parameters of a public space lighting system. By deploying multiple disturbance sensors and linearly calibrating them, the original voltage signal is converted into an accurate disturbance acceleration value. Then, historical data is fused within a preset time window to obtain the real-time equivalent disturbance amount. Combining the reference brightness and offset range, a series of response coefficients, variance suppression ratios, and residual correction mechanisms are constructed to gradually obtain the theoretical brightness, the limited brightness, and the final brightness. Finally, the future brightness is predicted based on the disturbance trajectory at multiple discrete refresh times, forming a continuous lighting timing control command sequence. Summary of the Invention

[0005] This invention provides a method for optimizing control parameters of a public space lighting system, which helps to solve the problems mentioned in the background art.

[0006] This invention provides the following technical solution: a method for optimizing control parameters of a public space lighting system, comprising:

[0007] Disturbance sensors are deployed to collect vertical disturbance signals. A linear conversion between voltage and disturbance acceleration is established based on two sets of standard disturbances. During operation, the original voltage is converted into calibration disturbance acceleration.

[0008] Within the disturbance integration time window, time-weighted and fused data from each sensor are performed to calculate the current equivalent disturbance acceleration. The acceleration is set to zero when there is no complete window during the startup phase.

[0009] Set the reference illumination brightness and the maximum brightness offset, calculate the lower and upper limits of brightness, and impose upper and lower limit constraints on the output brightness;

[0010] The disturbance reference value is obtained by averaging over time, and the disturbance normalized response coefficient is constructed by combining the stability offset constant and the current equivalent disturbance acceleration. The coefficient is set to zero when both are zero.

[0011] The perturbation variance is constructed based on the deviation between the equivalent perturbation acceleration and the perturbation reference value. The adaptive brightness suppression ratio is obtained by combining the stability offset constant. The suppression ratio is set to zero when both the variance and the reference value are zero.

[0012] The theoretical luminance is calculated based on the reference illumination luminance, the perturbation normalized response coefficient, and the adaptive luminance suppression ratio, and upper and lower limit amplitude is applied;

[0013] Calculate the disturbance offset residual, construct the residual weighting factor and residual modulation coefficient according to the residual and the disturbance variance, correct the adaptive brightness suppression ratio, obtain the corrected brightness and limit it, and output the final brightness;

[0014] The disturbance response distribution trajectory is constructed according to the refresh cycle, the equivalent disturbance acceleration trajectory sequence is obtained, the disturbance trend information is generated based on the difference with the previous discrete refresh time, and the dimensionless disturbance trend coefficient is constructed by combining the disturbance reference value, disturbance variance and disturbance increment normalization scaling factor. The predicted brightness is calculated and limited, forming a lighting timing control command sequence that includes the final brightness output and the predicted limited brightness of the next cycle.

[0015] Optionally, the deployment of disturbance sensors to collect vertical disturbance signals, establishing a linear conversion between voltage and disturbance acceleration based on two sets of standard disturbances, and converting the original voltage into calibration disturbance acceleration during operation, specifically includes:

[0016] Multiple disturbance sensors are deployed at predetermined intervals and positions within the controlled lighting area under the overpass. Each disturbance sensor is assigned a unique number. Each disturbance sensor collects disturbance signals in the vertical direction and outputs a continuously changing raw voltage sequence.

[0017] Select the first and second sets of standard disturbance operation time points, and obtain the actual disturbance acceleration standard value through the standard accelerometer at each time point. At the same time, record the original voltage reading of each disturbance sensor at the corresponding time point to form two sets of calibration data corresponding to each sensor.

[0018] Under the condition that the two sets of voltage readings of each disturbance sensor are not equal, based on the linear relationship between the standard value of the actual disturbance acceleration and the two sets of voltage readings of the corresponding disturbance sensor, the proportional coefficient and offset coefficient of the corresponding disturbance sensor are calculated respectively, and a linear conversion relationship between the original voltage of the corresponding disturbance sensor and the disturbance acceleration is established.

[0019] When a disturbance sensor obtains the same original voltage reading in two sets of standard disturbance operations, it is determined that the disturbance sensor with the corresponding number cannot obtain an effective linear fitting result under the current calibration conditions. The disturbance sensor with the corresponding number is replaced or the layout is adjusted, and the standard disturbance operation is re-executed until all disturbance sensors involved in the control obtain two sets of unequal voltage readings and complete the linear calibration.

[0020] At any point during system operation, the raw voltage output of each disturbance sensor is read, and a linear conversion is performed according to the proportional coefficient and offset coefficient of the corresponding disturbance sensor to obtain the calibrated disturbance acceleration output of the corresponding disturbance sensor at the current moment, thus forming a multi-sensor disturbance acceleration data set.

[0021] Optionally, the step of performing time-weighted fusion of data from each sensor within the disturbance integration time window to calculate the current equivalent disturbance acceleration, and setting it to zero when there is no complete window during the startup period, specifically includes:

[0022] Set the disturbance integration time window length to a preset second-level time length;

[0023] Construct a time-weighted perturbation kernel function, and normalize the weighting coefficients within the entire perturbation integration time window so that the sum of the weighting coefficients at all historical moments within the perturbation integration time window equals one.

[0024] When the system running time is greater than or equal to the length of the disturbance integration time window, for any current moment, according to the preset disturbance time weighting kernel function, the historical disturbance acceleration of each disturbance sensor within the disturbance integration time window range is traced back from the current moment and weighted and integrated to obtain the weighted disturbance acceleration of each disturbance sensor at the current moment. Then, the weighted disturbance acceleration of all disturbance sensors is averaged by quantity to obtain the equivalent disturbance acceleration at the current moment.

[0025] In the initial stage of system startup, when the running time is less than the length of the disturbance integration time window, if historical disturbance data within the complete disturbance integration time window cannot be obtained, the equivalent disturbance acceleration is uniformly set to zero as a boundary condition for the startup stage.

[0026] Optionally, the setting of the reference illumination brightness and the maximum brightness offset, the calculation of the lower and upper limits of brightness, and the imposition of upper and lower limit constraints on the output brightness specifically include:

[0027] In the controlled area under the viaduct, the reference lighting brightness is set according to the road function, safety standards and environmental requirements. The reference lighting brightness is the target brightness of the lighting system when the disturbance is zero or small.

[0028] Sets the maximum allowable brightness offset relative to the reference illumination brightness, which limits the width of the brightness adjustment range;

[0029] The lower limit of lighting brightness is calculated by subtracting the maximum brightness offset from the reference lighting brightness, and the lower limit of lighting brightness is used as the minimum brightness that the system cannot fall below under any operating state.

[0030] The upper limit of lighting brightness is calculated by adding the maximum brightness offset to the reference lighting brightness, and the upper limit of lighting brightness is used as the highest brightness that the system cannot exceed in any operating state.

[0031] In the subsequent brightness calculation and adjustment process, upper and lower limits are imposed on all theoretical brightness and corrected brightness to ensure that the brightness output is always between the lower limit and the upper limit of the lighting brightness.

[0032] Optionally, the step of obtaining the disturbance reference value by time averaging, constructing the disturbance normalized response coefficient by combining the stability offset constant and the current equivalent disturbance acceleration, and setting the coefficient to zero when both are zero, specifically includes:

[0033] When the system running time is greater than or equal to the disturbance integral time window length, at each current moment, the equivalent disturbance acceleration before the current moment is averaged over time according to the disturbance integral time window length to obtain the disturbance reference value at the current moment.

[0034] During the startup phase, when the system runtime is less than the length of the disturbance integration time window, the disturbance reference value is uniformly processed to zero.

[0035] A stability offset constant is set, which is a positive number, and it participates in the calculation together with the equivalent disturbance acceleration and the disturbance reference value during the construction of the disturbance normalized response function;

[0036] At any given moment, when the absolute value of the equivalent disturbance acceleration and the absolute value of the disturbance reference value are both zero, the disturbance normalized response coefficient is set to zero.

[0037] At any given moment, when the absolute value of the equivalent disturbance acceleration and the absolute value of the disturbance reference value are not both zero, the disturbance normalized response coefficient is calculated based on the proportional relationship between the current value of the equivalent disturbance acceleration, the absolute value of the equivalent disturbance acceleration, the absolute value of the disturbance reference value, and the stability offset constant.

[0038] Optionally, the step of constructing the perturbation variance based on the deviation between the equivalent perturbation acceleration and the perturbation reference value, and obtaining the adaptive brightness suppression ratio by combining it with the stability offset constant, and setting the suppression ratio to zero when both the variance and the reference value are zero, specifically includes:

[0039] When the system running time is greater than or equal to the disturbance integration time window length, at each current moment, the deviation between the equivalent disturbance acceleration before the current moment and the disturbance reference value is averaged according to the disturbance integration time window length to construct the disturbance variance at the current moment.

[0040] During the period when the system runtime is less than the length of the disturbance integration time window, the disturbance variance is set to zero as a boundary treatment for the startup phase.

[0041] At any given moment, when both the disturbance variance and the disturbance reference value are zero, the adaptive brightness suppression ratio is set to zero.

[0042] At any given moment, when the disturbance variance and the disturbance reference value are not both zero, the adaptive brightness suppression ratio is calculated based on the proportional relationship between the combined results of the disturbance variance, the absolute value of the disturbance reference value, and the stable offset constant.

[0043] Optionally, the step of calculating the theoretical brightness based on the reference illumination brightness, the perturbation normalized response coefficient, and the adaptive brightness suppression ratio, and performing upper and lower limit limiting, specifically includes:

[0044] At any given moment, read the reference illumination brightness, the perturbation normalized response coefficient, and the adaptive brightness suppression ratio, and calculate the theoretical brightness output according to the product and difference relationship between the three.

[0045] At any given moment, the calculated theoretical brightness output is compared with the lower limit of lighting brightness. When the theoretical brightness output is lower than the lower limit of lighting brightness, the theoretical brightness output is corrected to the lower limit of lighting brightness.

[0046] At any given moment, compare the theoretical brightness output with the upper limit of the lighting brightness. If the theoretical brightness output is higher than the upper limit of the lighting brightness, correct the theoretical brightness output to the upper limit of the lighting brightness.

[0047] After completing the dual limiting correction of the lower and upper limits, the corrected brightness is used as the brightness output after limiting adjustment, and provided to the subsequent residual correction mechanism and lighting drive module.

[0048] Optionally, the calculation of the disturbance offset residual, constructing a residual weighting factor and a residual modulation coefficient based on the residual and the disturbance variance, correcting the adaptive brightness suppression ratio, obtaining the corrected brightness and limiting it, and outputting the final brightness specifically includes:

[0049] At any given moment, the difference between the current equivalent disturbance acceleration and the current disturbance reference value is defined as the disturbance offset residual.

[0050] Construct residual weighting factors based on the perturbation offset residual and the current perturbation variance;

[0051] The residual modulation coefficients are constructed based on the residual weighting factor, and the residual modulation coefficients are compressed using an exponential form.

[0052] The residual modulation coefficients are combined with the adaptive luminance suppression ratio to form the corrected luminance suppression ratio;

[0053] The corrected luminance output is recalculated based on the corrected luminance suppression ratio, reference illumination luminance, and disturbance normalized response coefficient.

[0054] The corrected brightness output is compared with the lower limit and upper limit of the lighting brightness. The corrected brightness output is then subjected to a limiting process to obtain the final brightness output, which is then provided to the lighting driver module and the next step of timing control.

[0055] Optionally, the step of constructing a disturbance response distribution trajectory according to the refresh cycle, obtaining an equivalent disturbance acceleration trajectory sequence, generating disturbance trend information based on the difference with the previous discrete refresh time, and constructing a dimensionless disturbance trend coefficient by combining the disturbance reference value, disturbance variance, and disturbance increment normalization scaling factor, calculating and limiting the predicted brightness, and forming a lighting timing control command sequence that includes the final brightness output and the predicted limited brightness for the next cycle, specifically including:

[0056] Set the lighting control refresh cycle to generate a discrete refresh time sequence starting from the system startup time, with each discrete refresh time corresponding to a continuous time point;

[0057] Set the disturbance history tracking duration, and determine the maximum number of historical steps that can be traced at each discrete refresh time based on the refresh cycle and the disturbance history tracking duration;

[0058] At any discrete refresh time, backtrack several refresh cycles according to the refresh cycle, collect the equivalent disturbance acceleration history values ​​within the disturbance history tracking time range, and form the equivalent disturbance acceleration trajectory sequence corresponding to the current discrete refresh time.

[0059] At any discrete refresh time, the equivalent disturbance acceleration at the previous discrete refresh time is subtracted from the current equivalent disturbance acceleration to construct the equivalent disturbance increment for the current refresh cycle. The equivalent disturbance increment is set to zero in the first refresh cycle after the system starts.

[0060] At any discrete refresh time, based on the current disturbance reference value, the current disturbance variance, and the stability offset constant, a normalized scaling factor for the disturbance increment is constructed, and the equivalent disturbance increment is compared with the normalized scaling factor to obtain the dimensionless disturbance trend coefficient.

[0061] At any discrete refresh time, set the dimming response coefficient, and calculate the predicted brightness for the next refresh cycle based on the dimensionless disturbance trend coefficient, dimming response coefficient, and maximum brightness offset of the current refresh cycle.

[0062] At any discrete refresh time, the predicted brightness for the next refresh cycle is subjected to upper and lower limit limiting processing. When the predicted brightness is lower than the lower limit of the lighting brightness, the predicted brightness is set to the lower limit of the lighting brightness. When the predicted brightness is higher than the upper limit of the lighting brightness, the predicted brightness is set to the upper limit of the lighting brightness, thus obtaining the predicted limited brightness.

[0063] At any discrete refresh time, the final brightness output is combined with the predicted limited brightness of the next refresh cycle to form a lighting timing control command sequence. The lighting drive module uses the final brightness output in the current refresh cycle and the corresponding predicted limited brightness in the next refresh cycle, forming a continuous lighting timing control process.

[0064] The present invention has the following beneficial effects:

[0065] 1. This scheme first deploys multiple disturbance sensors in the controlled area under the viaduct. Two sets of standard disturbance operations are used to obtain data comparing the actual acceleration with the original sensor voltage, establishing a linear conversion relationship for each sensor. A dual-time-point calibration strategy designed for the complex vibration environment ensures accurate calculation of the proportional and offset coefficients even when the sensor output changes only slightly, guaranteeing accurate and reliable converted acceleration data. This improves the accuracy and consistency of sensor data, providing a reliable foundation for subsequent fusion and response, avoiding misjudgments and fluctuations caused by individual sensor differences, and making the input data for real-time control more stable.

[0066] 2. This solution employs a time-weighted kernel function to perform weighted convolution on historical data from multiple sensors, using a weighting curve designed to be weak first, then strong, then weak again. This preserves real-time trends while suppressing sudden noise. During startup, the equivalent disturbance acceleration is automatically set to zero at the boundary, avoiding control jitter caused by insufficient historical data. This fusion method considers both short-term disturbance effects and the average effect over longer periods, resulting in a more objective and stable equivalent disturbance. This improves the accuracy of disturbance extraction, ensuring the lighting system maintains a stable response during system startup and periods of severe disturbance, overcoming the shortcomings of existing technologies that rely excessively on historical data or suffer from severe real-time jitter.

[0067] 3. The solution first sets a reference brightness based on road function and environmental safety requirements, then calculates the upper and lower limits based on the maximum allowable brightness deviation, and applies consistent limiting to all subsequent brightness values. This decouples the reference brightness from disturbance effects, ensuring minimum lighting requirements while avoiding excessive brightness and waste. Furthermore, the same upper and lower limit constraints are applied in every dynamic calculation stage to ensure consistency. This achieves both safe lighting and further energy savings under different disturbance scenarios, overcoming the problems of inconsistent brightness or excessive brightness caused by traditional static upper and lower limits or a lack of unified constraints.

[0068] 4. This scheme constructs a normalized response coefficient between the equivalent disturbance and its historical benchmark, and forces it to zero when both the benchmark and the current value are zero, effectively avoiding jitter caused by division by zero or extremely small denominators. Furthermore, it calculates the disturbance variance based on the deviation between the equivalent disturbance and the benchmark, and then combines this with a stable offset constant to obtain the adaptive brightness suppression ratio. By incorporating both variance and mean into the brightness suppression ratio model, it can distinguish between persistent disturbances and instantaneous deviations, and through smooth ratio adjustment, it balances response speed and stability. It increases brightness suppression when disturbances increase and automatically relaxes suppression when disturbances decrease, making brightness adjustment more intelligent and gentle, overcoming the predicament of existing technologies that cannot balance response and stability by relying solely on instantaneous or mean suppression.

[0069] 5. To achieve direct control over lighting brightness, the scheme calculates the theoretical brightness by multiplying the reference brightness by the difference between two suppression coefficients, and then adjusts it within the same upper and lower limits. By applying both suppression coefficients simultaneously to the reference brightness, the adjustment result is influenced by both the normalized response and the suppression ratio, thus reflecting the combined effect of the average disturbance characteristics and the current offset characteristics. The theoretical brightness can respond quickly to environmental changes while remaining within a safe and controllable range, ensuring both lighting comfort and energy efficiency.

[0070] 6. This solution further applies residual weights and residual modulation functions to dynamically adjust the adaptive suppression ratio for current disturbance offset residuals: the residual weight factor measures the importance of the offset in the overall fluctuation, and the residual modulation coefficient is exponentially compressed to highlight significant offsets. The combination of the two can enhance the suppression of sudden large-amplitude disturbances. A piecewise correction mechanism is introduced to keep the control gentle during stable disturbances and rapidly strengthen the suppression during sudden disturbances. This improves the system's robustness and safety margin to critical disturbances, avoids severe lighting fluctuations or delayed response caused by sudden disturbances, and improves control accuracy and user experience compared to traditional static correction or no correction mechanism.

[0071] 7. Finally, the scheme constructs a disturbance trajectory sequence within the lighting control refresh cycle. Based on the acceleration increment and normalized scaling factor compared to the previous refresh time, a dimensionless trend coefficient is calculated to predict and limit the brightness of the next cycle, forming a continuous control command with the current final brightness. By employing a dimensionless coefficient combining short-term historical trajectories and incremental trends, short-term brightness prediction is achieved without complex models. This allows for advance adjustment of lighting to cope with upcoming disturbance changes, reducing delays and overshoot. Compared to existing methods that rely solely on real-time feedback, this prediction mechanism significantly improves the foresight and stability of the lighting system. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0073] 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.

[0074] Example, refer to Figure 1 A method for optimizing control parameters of a public space lighting system, comprising:

[0075] Disturbance sensors are deployed to collect vertical disturbance signals. A linear conversion between voltage and disturbance acceleration is established based on two sets of standard disturbances. During operation, the original voltage is converted into calibration disturbance acceleration.

[0076] Within the disturbance integration time window, time-weighted and fused data from each sensor are performed to calculate the current equivalent disturbance acceleration. The acceleration is set to zero when there is no complete window during the startup phase.

[0077] Set the reference illumination brightness and the maximum brightness offset, calculate the lower and upper limits of brightness, and impose upper and lower limit constraints on the output brightness;

[0078] The disturbance reference value is obtained by averaging over time, and the disturbance normalized response coefficient is constructed by combining the stability offset constant and the current equivalent disturbance acceleration. The coefficient is set to zero when both are zero.

[0079] The perturbation variance is constructed based on the deviation between the equivalent perturbation acceleration and the perturbation reference value. The adaptive brightness suppression ratio is obtained by combining the stability offset constant. The suppression ratio is set to zero when both the variance and the reference value are zero.

[0080] The theoretical luminance is calculated based on the reference illumination luminance, the perturbation normalized response coefficient, and the adaptive luminance suppression ratio, and upper and lower limit amplitude is applied;

[0081] Calculate the disturbance offset residual, construct the residual weighting factor and residual modulation coefficient according to the residual and the disturbance variance, correct the adaptive brightness suppression ratio, obtain the corrected brightness and limit it, and output the final brightness;

[0082] The disturbance response distribution trajectory is constructed according to the refresh cycle, the equivalent disturbance acceleration trajectory sequence is obtained, the disturbance trend information is generated based on the difference with the previous discrete refresh time, and the dimensionless disturbance trend coefficient is constructed by combining the disturbance reference value, disturbance variance and disturbance increment normalization scaling factor. The predicted brightness is calculated and limited, forming a lighting timing control command sequence that includes the final brightness output and the predicted limited brightness of the next cycle.

[0083] This scheme employs a step-by-step control parameter optimization method. First, multiple disturbance sensors are deployed within the controlled public space. Based on two sets of standard disturbances, a precise linear conversion relationship between voltage and disturbance acceleration is established, converting the original sensor voltages into calibrated disturbance acceleration data in real time. This step resolves the issues of inconsistent sensor responses and difficulty in effectively calibrating reading errors, ensuring the accuracy and consistency of input data. Subsequently, time-weighted fusion of multi-sensor data is applied within the disturbance integration time window to extract the equivalent disturbance acceleration at the current moment. Boundary processing is introduced during system startup to zero out incomplete data, eliminating interference from insufficient historical data or window edge effects on the fusion results and ensuring a smooth system startup during initialization. Next, a reference lighting brightness and maximum brightness offset are set. By calculating the upper and lower limits of brightness and constraining the output brightness, the safety and energy consumption risks caused by excessively low lower or high upper limits of brightness in traditional lighting systems under different environmental disturbances are addressed. In constructing the normalized response coefficient and adaptive suppression ratio for disturbances, this method dynamically generates the response coefficient by combining the deviation between the current disturbance and the historical benchmark, and sets it to zero under special boundary conditions. This avoids division by zero or jitter caused by extreme or zero disturbances, thus improving numerical stability. Furthermore, by calculating the disturbance offset residual and introducing residual weights and modulation coefficients, this scheme achieves secondary correction based on the original suppression ratio, effectively distinguishing between sudden disturbances and stable fluctuations, and improving the response accuracy and robustness to critical disturbances. Finally, this scheme constructs the disturbance response trajectory according to the brightness control refresh cycle, and predicts the brightness of the next cycle by combining the data difference at the previous refresh time and the normalized scaling factor, achieving proactive predictive control, making brightness adjustments in advance, and reducing response delay and overshoot risk. Compared with existing methods that rely solely on real-time feedback or static models, this scheme improves the stability, comfort, and energy efficiency of public space lighting systems in complex environments through a multi-dimensional, phased, dynamic correction, and prediction approach.

[0084] The deployed disturbance sensors collect vertical disturbance signals, and establish a linear conversion between voltage and disturbance acceleration based on two sets of standard disturbances. During operation, the original voltage is converted into a calibration disturbance acceleration, specifically including:

[0085] Multiple disturbance sensors are deployed at predetermined intervals and positions within the controlled lighting area under the overpass. Each disturbance sensor is assigned a unique number. Each disturbance sensor collects disturbance signals in the vertical direction and outputs a continuously changing raw voltage sequence.

[0086] Select the first and second sets of standard disturbance operation time points, and obtain the actual disturbance acceleration standard value through the standard accelerometer at each time point. At the same time, record the original voltage reading of each disturbance sensor at the corresponding time point to form two sets of calibration data corresponding to each sensor.

[0087] Under the condition that the two sets of voltage readings of each disturbance sensor are not equal, based on the linear relationship between the standard value of the actual disturbance acceleration and the two sets of voltage readings of the corresponding disturbance sensor, the proportional coefficient and offset coefficient of the corresponding disturbance sensor are calculated respectively, and a linear conversion relationship between the original voltage of the corresponding disturbance sensor and the disturbance acceleration is established.

[0088] When a disturbance sensor obtains the same original voltage reading in two sets of standard disturbance operations, it is determined that the disturbance sensor with the corresponding number cannot obtain an effective linear fitting result under the current calibration conditions. The disturbance sensor with the corresponding number is replaced or the layout is adjusted, and the standard disturbance operation is re-executed until all disturbance sensors involved in the control obtain two sets of unequal voltage readings and complete the linear calibration.

[0089] At any point during system operation, the raw voltage output of each disturbance sensor is read, and a linear conversion is performed according to the proportional coefficient and offset coefficient of the corresponding disturbance sensor to obtain the calibrated disturbance acceleration output of the corresponding disturbance sensor at the current moment, thus forming a multi-sensor disturbance acceleration data set.

[0090] Further specific implementation steps include:

[0091] Lighting to be deployed in the controlled lighting area under the overpass One disturbance sensor, numbered as Each disturbance sensor acquires the vertical disturbance signal and outputs the raw voltage. ;in, This represents the total number of disturbance sensors deployed in the controlled area under the viaduct; Index the sensor numbers; For the first A disturbance sensor at continuous time The original voltage output; It is a continuous-time variable;

[0092] Two sets of standard disturbance operation time points were selected. and At each time point, the corresponding standard value of the actual disturbance acceleration is obtained using a standard accelerometer, and is denoted as follows: Simultaneously record the first The original voltage outputs of the disturbance sensors at these two time points are denoted as follows: and ;in, , These are the consecutive time points corresponding to the first and second groups of calibration operations, respectively. , At the calibration time , The standard value of the actual disturbance acceleration measured by a standard accelerometer; , The first Each disturbance sensor at calibration time , The original voltage reading;

[0093] If and only if satisfying When, calculate the first The linear calibration coefficients for each disturbance sensor are as follows: , ;in, For the first The linear slope coefficient of voltage and acceleration of each disturbance sensor; For the first Linear calibration intercept of each disturbance sensor;

[0094] when If the sensor fails to achieve a valid linear fit under the current calibration conditions, it is determined that the sensor cannot complete an effective linear fit. The sensor is then replaced until a valid linear fit is obtained. Valid calibration data;

[0095] At any given runtime, the first The calibrated disturbance acceleration output of each disturbance sensor is: ;in, For the first A disturbance sensor at continuous time The actual disturbance acceleration value.

[0096] The step of performing time-weighted fusion of sensor data within the disturbance integration time window to calculate the current equivalent disturbance acceleration, and setting it to zero when there is no complete window during the startup period, specifically includes:

[0097] Set the disturbance integration time window length to a preset second-level time length;

[0098] Construct a time-weighted perturbation kernel function, and normalize the weighting coefficients within the entire perturbation integration time window so that the sum of the weighting coefficients at all historical moments within the perturbation integration time window equals one.

[0099] When the system running time is greater than or equal to the length of the disturbance integration time window, for any current moment, according to the preset disturbance time weighting kernel function, the historical disturbance acceleration of each disturbance sensor within the disturbance integration time window range is traced back from the current moment and weighted and integrated to obtain the weighted disturbance acceleration of each disturbance sensor at the current moment. Then, the weighted disturbance acceleration of all disturbance sensors is averaged by quantity to obtain the equivalent disturbance acceleration at the current moment.

[0100] In the initial stage of system startup, when the running time is less than the length of the disturbance integration time window, if historical disturbance data within the complete disturbance integration time window cannot be obtained, the equivalent disturbance acceleration is uniformly set to zero as a boundary condition for the startup stage.

[0101] Further specific implementation steps include:

[0102] Set the disturbance integration time window length to Second;

[0103] The perturbation-time-weighted kernel function is constructed as follows: , and satisfy ;in, This is the historical time offset variable used in integration operations; The time-weighted kernel function for perturbation represents the time offset. The weighting coefficients for historical disturbance accelerations;

[0104] During system uptime Then, the perturbation accelerations of each sensor are convolved and fused using a time-weighted kernel to obtain the equivalent perturbation acceleration, specifically: ;in, For continuous time intervals The equivalent perturbation acceleration; For the first A disturbance sensor at a historical moment The value of the disturbance acceleration;

[0105] When the system just starts up and During this stage, .

[0106] The process of setting the reference illumination brightness and the maximum brightness offset, calculating the lower and upper limits of brightness, and imposing upper and lower limit constraints on the output brightness specifically includes:

[0107] In the controlled area under the viaduct, the reference lighting brightness is set according to the road function, safety standards and environmental requirements. The reference lighting brightness is the target brightness of the lighting system when the disturbance is zero or small.

[0108] Sets the maximum allowable brightness offset relative to the reference illumination brightness, which limits the width of the brightness adjustment range;

[0109] The lower limit of lighting brightness is calculated by subtracting the maximum brightness offset from the reference lighting brightness, and the lower limit of lighting brightness is used as the minimum brightness that the system cannot fall below under any operating state.

[0110] The upper limit of lighting brightness is calculated by adding the maximum brightness offset to the reference lighting brightness, and the upper limit of lighting brightness is used as the highest brightness that the system cannot exceed in any operating state.

[0111] In the subsequent brightness calculation and adjustment process, upper and lower limits are imposed on all theoretical brightness and corrected brightness to ensure that the brightness output is always between the lower limit and the upper limit of the lighting brightness.

[0112] Further specific implementation steps include:

[0113] Set the reference lighting brightness for the area to be controlled under the viaduct. lumen;

[0114] Set the maximum allowable brightness offset relative to the reference brightness as follows: lumen;

[0115] Based on the reference luminance and the maximum luminance offset, the lower and upper limits of the illumination luminance are calculated respectively, as follows: , ;in, , These represent the minimum and maximum allowable values ​​for lighting brightness, respectively.

[0116] The process of obtaining the time-averaged disturbance reference value, constructing the disturbance normalized response coefficient by combining the stability offset constant and the current equivalent disturbance acceleration, and setting the coefficient to zero when both are zero, specifically includes:

[0117] When the system running time is greater than or equal to the disturbance integral time window length, at each current moment, the equivalent disturbance acceleration before the current moment is averaged over time according to the disturbance integral time window length to obtain the disturbance reference value at the current moment.

[0118] During the startup phase, when the system runtime is less than the length of the disturbance integration time window, the disturbance reference value is uniformly processed to zero.

[0119] A stability offset constant is set, which is a positive number, and it participates in the calculation together with the equivalent disturbance acceleration and the disturbance reference value during the construction of the disturbance normalized response function;

[0120] At any given moment, when the absolute value of the equivalent disturbance acceleration and the absolute value of the disturbance reference value are both zero, the disturbance normalized response coefficient is set to zero.

[0121] At any given moment, when the absolute value of the equivalent disturbance acceleration and the absolute value of the disturbance reference value are not both zero, the disturbance normalized response coefficient is calculated based on the proportional relationship between the current value of the equivalent disturbance acceleration, the absolute value of the equivalent disturbance acceleration, the absolute value of the disturbance reference value, and the stability offset constant.

[0122] Further specific implementation steps include:

[0123] During system uptime At that time, the disturbance reference value is calculated as follows: ;in, For continuous time intervals The perturbation reference acceleration;

[0124] During system uptime season ;

[0125] Set stable offset constant ;

[0126] Construct the normalized response function for the disturbance, specifically as follows:

[0127] S401, when and season ;in, For continuous time intervals The perturbation normalized response coefficient;

[0128] S402, when and When they are not both 0, let: .

[0129] The perturbation variance is constructed based on the deviation between the equivalent perturbation acceleration and the perturbation reference value. This variance, combined with a stable offset constant, yields the adaptive brightness suppression ratio. The suppression ratio is set to zero when both the variance and the reference value are zero. Specifically, this includes:

[0130] When the system running time is greater than or equal to the disturbance integration time window length, at each current moment, the deviation between the equivalent disturbance acceleration before the current moment and the disturbance reference value is averaged according to the disturbance integration time window length to construct the disturbance variance at the current moment.

[0131] During the period when the system runtime is less than the length of the disturbance integration time window, the disturbance variance is set to zero as a boundary treatment for the startup phase.

[0132] At any given moment, when both the disturbance variance and the disturbance reference value are zero, the adaptive brightness suppression ratio is set to zero.

[0133] At any given moment, when the disturbance variance and the disturbance reference value are not both zero, the adaptive brightness suppression ratio is calculated based on the proportional relationship between the combined results of the disturbance variance, the absolute value of the disturbance reference value, and the stable offset constant.

[0134] Further specific implementation steps include:

[0135] During system uptime At that time, the perturbation variance function is constructed as follows: ;in, For continuous time intervals The variance of the perturbation acceleration;

[0136] During system uptime season ;

[0137] At any given time, an adaptive suppression ratio function is constructed based on the disturbance variance and the baseline value, specifically as follows:

[0138] like and Then let Otherwise, let: ;in, For continuous time intervals Adaptive brightness suppression ratio.

[0139] The calculation of theoretical brightness based on reference illumination luminance, perturbation normalized response coefficient, and adaptive luminance suppression ratio, followed by upper and lower limit limiting, specifically includes:

[0140] At any given moment, read the reference illumination brightness, the perturbation normalized response coefficient, and the adaptive brightness suppression ratio, and calculate the theoretical brightness output according to the product and difference relationship between the three.

[0141] At any given moment, the calculated theoretical brightness output is compared with the lower limit of lighting brightness. When the theoretical brightness output is lower than the lower limit of lighting brightness, the theoretical brightness output is corrected to the lower limit of lighting brightness.

[0142] At any given moment, compare the theoretical brightness output with the upper limit of the lighting brightness. If the theoretical brightness output is higher than the upper limit of the lighting brightness, correct the theoretical brightness output to the upper limit of the lighting brightness.

[0143] After completing the dual limiting correction of the lower and upper limits, the corrected brightness is used as the brightness output after limiting adjustment, and provided to the subsequent residual correction mechanism and lighting drive module.

[0144] Further specific implementation steps include:

[0145] At any given time, the real-time brightness output is calculated based on the reference brightness, the perturbation normalized response function, and the suppression ratio function, specifically as follows: ;in, For public lighting systems at continuous times Theoretical brightness output;

[0146] The adjusted brightness obtained after amplitude limiting is: ;in, For continuous time intervals Brightness output after limiting adjustment.

[0147] The calculation of the disturbance offset residual, the construction of residual weighting factors and residual modulation coefficients based on the residual and disturbance variance, the correction of the adaptive brightness suppression ratio, the obtaining of the corrected brightness and limiting, and the output of the final brightness specifically include:

[0148] At any given moment, the difference between the current equivalent disturbance acceleration and the current disturbance reference value is defined as the disturbance offset residual.

[0149] Construct residual weighting factors based on the perturbation offset residual and the current perturbation variance;

[0150] The residual modulation coefficients are constructed based on the residual weighting factor, and the residual modulation coefficients are compressed using an exponential form.

[0151] The residual modulation coefficients are combined with the adaptive luminance suppression ratio to form the corrected luminance suppression ratio;

[0152] The corrected luminance output is recalculated based on the corrected luminance suppression ratio, reference illumination luminance, and disturbance normalized response coefficient.

[0153] The corrected brightness output is compared with the lower limit and upper limit of the lighting brightness. The corrected brightness output is then subjected to a limiting process to obtain the final brightness output, which is then provided to the lighting driver module and the next step of timing control.

[0154] Further specific implementation steps include:

[0155] At any given time, construct the perturbation offset residual function. ;in, For continuous time intervals The disturbance offset residual;

[0156] Constructing a residual weighting factor function based on residuals and variance: ;in, For continuous time intervals The residual weighting factor;

[0157] Constructing a residual modulation function based on residual weighting factors ;in, For continuous time intervals The residual modulation coefficient;

[0158] The suppression ratio function is corrected using the residual modulation function to obtain the corrected suppression ratio function, as follows: ;in, For continuous time intervals Corrected suppression ratio;

[0159] Based on the corrected suppression ratio function, the corrected brightness output is calculated as follows: ;in, For continuous time intervals Corrected brightness output;

[0160] Correcting brightness output After amplitude limiting, the final brightness output is obtained as follows: ;in, For continuous time intervals The final brightness output.

[0161] The process involves constructing a disturbance response distribution trajectory according to the refresh cycle, obtaining an equivalent disturbance acceleration trajectory sequence, generating disturbance trend information based on the difference with the previous discrete refresh time, and constructing a dimensionless disturbance trend coefficient by combining the disturbance reference value, disturbance variance, and disturbance increment normalization scaling factor. This process calculates and limits the predicted brightness, forming a lighting timing control command sequence that includes the final brightness output and the predicted limited brightness for the next cycle. Specifically, this includes:

[0162] Set the lighting control refresh cycle to generate a discrete refresh time sequence starting from the system startup time, with each discrete refresh time corresponding to a continuous time point;

[0163] Set the disturbance history tracking duration, and determine the maximum number of historical steps that can be traced at each discrete refresh time based on the refresh cycle and the disturbance history tracking duration;

[0164] At any discrete refresh time, backtrack several refresh cycles according to the refresh cycle, collect the equivalent disturbance acceleration history values ​​within the disturbance history tracking time range, and form the equivalent disturbance acceleration trajectory sequence corresponding to the current discrete refresh time.

[0165] At any discrete refresh time, the equivalent disturbance acceleration at the previous discrete refresh time is subtracted from the current equivalent disturbance acceleration to construct the equivalent disturbance increment for the current refresh cycle. The equivalent disturbance increment is set to zero in the first refresh cycle after the system starts.

[0166] At any discrete refresh time, based on the current disturbance reference value, the current disturbance variance, and the stability offset constant, a normalized scaling factor for the disturbance increment is constructed, and the equivalent disturbance increment is compared with the normalized scaling factor to obtain the dimensionless disturbance trend coefficient.

[0167] At any discrete refresh time, set the dimming response coefficient, and calculate the predicted brightness for the next refresh cycle based on the dimensionless disturbance trend coefficient, dimming response coefficient, and maximum brightness offset of the current refresh cycle.

[0168] At any discrete refresh time, the predicted brightness for the next refresh cycle is subjected to upper and lower limit limiting processing. When the predicted brightness is lower than the lower limit of the lighting brightness, the predicted brightness is set to the lower limit of the lighting brightness. When the predicted brightness is higher than the upper limit of the lighting brightness, the predicted brightness is set to the upper limit of the lighting brightness, thus obtaining the predicted limited brightness.

[0169] At any discrete refresh time, the final brightness output is combined with the predicted limited brightness of the next refresh cycle to form a lighting timing control command sequence. The lighting drive module uses the final brightness output in the current refresh cycle and the corresponding predicted limited brightness in the next refresh cycle, forming a continuous lighting timing control process.

[0170] Further specific implementation steps include:

[0171] Set the lighting control refresh cycle to seconds, discrete refresh time is ;in, The continuous time intervals from system startup; For discrete-time indexing; For the first The continuous time corresponding to each discrete refresh moment;

[0172] Set the duration for tracking disturbance history. Second;

[0173] At any discrete time Above, a perturbation trajectory sequence is formed: ;in, This serves as an index for historical points in the perturbation trajectory sequence; In the time span The maximum number of historical steps that can be included within the content must meet the following conditions. ; For relative to discrete time The perturbation equivalent acceleration trajectory sequence;

[0174] The perturbation increment is constructed as follows:

[0175] when season ;in, In the first The equivalent disturbance increment of each refresh time relative to the previous refresh time;

[0176] when season ;

[0177] The normalized scaling factor is constructed as follows: ;in, In the first The scale factor used to normalize the perturbation increment at each refresh time;

[0178] Constructing a dimensionless perturbation trend coefficient: ;in, For the first The relative trend of the equivalent intensity of the disturbance within each refresh cycle;

[0179] Set the dimming response coefficient to ;

[0180] Based on the disturbance trend coefficient Predicted brightness value for the next refresh cycle: ;in, In the first The brightness value of the next refresh cycle predicted by the perturbation trend of the previous refresh cycle;

[0181] right Perform limiting processing again to obtain the predicted limiting brightness: ;

[0182] in, This is the result after limiting the predicted brightness for the next refresh cycle;

[0183] At each discrete time This constitutes a short-term lighting timing control command sequence: ;in, For discrete time The corresponding set of lighting timing control instructions includes the final brightness used in the current cycle and the predicted limited brightness used in the next cycle;

[0184] The lighting driver module uses the command at the current moment. It will arrive in the next refresh cycle. When adopted .

[0185] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0186] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing control parameters of a public space lighting system, characterized in that, include: Disturbance sensors are deployed to collect vertical disturbance signals. A linear conversion between voltage and disturbance acceleration is established based on two sets of standard disturbances. During operation, the original voltage is converted into calibration disturbance acceleration. Within the disturbance integration time window, time-weighted and fused data from each sensor are performed to calculate the current equivalent disturbance acceleration. The acceleration is set to zero when there is no complete window during the startup phase. Set the reference illumination brightness and the maximum brightness offset, calculate the lower and upper limits of brightness, and impose upper and lower limit constraints on the output brightness; The disturbance reference value is obtained by averaging over time, and the disturbance normalized response coefficient is constructed by combining the stability offset constant and the current equivalent disturbance acceleration. The coefficient is set to zero when both are zero. The perturbation variance is constructed based on the deviation between the equivalent perturbation acceleration and the perturbation reference value. The adaptive brightness suppression ratio is obtained by combining the stability offset constant. The suppression ratio is set to zero when both the variance and the reference value are zero. The theoretical luminance is calculated based on the reference illumination luminance, the perturbation normalized response coefficient, and the adaptive luminance suppression ratio, and upper and lower limit amplitude is applied; Calculate the disturbance offset residual, construct the residual weighting factor and residual modulation coefficient according to the residual and the disturbance variance, correct the adaptive brightness suppression ratio, obtain the corrected brightness and limit it, and output the final brightness; The disturbance response distribution trajectory is constructed according to the refresh cycle, the equivalent disturbance acceleration trajectory sequence is obtained, the disturbance trend information is generated based on the difference with the previous discrete refresh time, and the dimensionless disturbance trend coefficient is constructed by combining the disturbance reference value, disturbance variance and disturbance increment normalization scaling factor. The predicted brightness is calculated and limited, forming a lighting timing control command sequence that includes the final brightness output and the predicted limited brightness of the next cycle. The deployed disturbance sensors collect vertical disturbance signals, and establish a linear conversion between voltage and disturbance acceleration based on two sets of standard disturbances. During operation, the original voltage is converted into a calibration disturbance acceleration, specifically including: Multiple disturbance sensors are deployed at predetermined intervals and positions within the controlled lighting area under the overpass. Each disturbance sensor is assigned a unique number. Each disturbance sensor collects disturbance signals in the vertical direction and outputs a continuously changing raw voltage sequence. Select the first and second sets of standard disturbance operation time points, and obtain the actual disturbance acceleration standard value through the standard accelerometer at each time point. At the same time, record the original voltage reading of each disturbance sensor at the corresponding time point to form two sets of calibration data corresponding to each sensor. Under the condition that the two sets of voltage readings of each disturbance sensor are not equal, based on the linear relationship between the standard value of the actual disturbance acceleration and the two sets of voltage readings of the corresponding disturbance sensor, the proportional coefficient and offset coefficient of the corresponding disturbance sensor are calculated respectively, and a linear conversion relationship between the original voltage of the corresponding disturbance sensor and the disturbance acceleration is established. When a disturbance sensor obtains the same original voltage reading in two sets of standard disturbance operations, it is determined that the disturbance sensor with the corresponding number cannot obtain an effective linear fitting result under the current calibration conditions. The disturbance sensor with the corresponding number is replaced or the layout is adjusted, and the standard disturbance operation is re-executed until all disturbance sensors involved in the control obtain two sets of unequal voltage readings and complete the linear calibration. At any point during system operation, the raw voltage output of each disturbance sensor is read, and a linear conversion is performed according to the proportional coefficient and offset coefficient of the corresponding disturbance sensor to obtain the calibrated disturbance acceleration output of the corresponding disturbance sensor at the current moment, thus forming a multi-sensor disturbance acceleration data set.

2. The method for optimizing control parameters of a public space lighting system according to claim 1, characterized in that, The step of performing time-weighted fusion of sensor data within the disturbance integration time window to calculate the current equivalent disturbance acceleration, and setting it to zero when there is no complete window during the startup period, specifically includes: Set the disturbance integration time window length to a preset second-level time length; Construct a time-weighted perturbation kernel function, and normalize the weighting coefficients within the entire perturbation integration time window so that the sum of the weighting coefficients for all historical moments within the perturbation integration time window equals one. When the system running time is greater than or equal to the length of the disturbance integration time window, for any current moment, according to the preset disturbance time weighting kernel function, the historical disturbance acceleration of each disturbance sensor within the disturbance integration time window range is traced back from the current moment and weighted and integrated to obtain the weighted disturbance acceleration of each disturbance sensor at the current moment. Then, the weighted disturbance acceleration of all disturbance sensors is averaged by quantity to obtain the equivalent disturbance acceleration at the current moment. In the initial stage of system startup, when the running time is less than the length of the disturbance integration time window, if historical disturbance data within the complete disturbance integration time window cannot be obtained, the equivalent disturbance acceleration is uniformly set to zero as a boundary condition for the startup stage.

3. The method for optimizing control parameters of a public space lighting system according to claim 2, characterized in that, The process of setting the reference illumination brightness and the maximum brightness offset, calculating the lower and upper limits of brightness, and imposing upper and lower limit constraints on the output brightness specifically includes: In the controlled area under the viaduct, the reference lighting brightness is set according to the road function, safety standards and environmental requirements. The reference lighting brightness is the target brightness of the lighting system when the disturbance is zero or small. Sets the maximum allowable brightness offset relative to the reference illumination brightness, which limits the width of the brightness adjustment range; The lower limit of lighting brightness is calculated by subtracting the maximum brightness offset from the reference lighting brightness, and the lower limit of lighting brightness is used as the minimum brightness that the system cannot fall below under any operating state. The upper limit of lighting brightness is calculated by adding the maximum brightness offset to the reference lighting brightness, and the upper limit of lighting brightness is used as the highest brightness that the system cannot exceed in any operating state. In the subsequent brightness calculation and adjustment process, upper and lower limits are imposed on all theoretical and corrected brightness to ensure that the brightness output is always between the lower limit and the upper limit of the lighting brightness.

4. The method for optimizing control parameters of a public space lighting system according to claim 3, characterized in that, The process of obtaining the time-averaged disturbance reference value, constructing the disturbance normalized response coefficient by combining the stability offset constant and the current equivalent disturbance acceleration, and setting the coefficient to zero when both are zero, specifically includes: When the system running time is greater than or equal to the disturbance integral time window length, at each current moment, the equivalent disturbance acceleration before the current moment is averaged over time according to the disturbance integral time window length to obtain the disturbance reference value at the current moment. During the startup phase, when the system runtime is less than the length of the disturbance integration time window, the disturbance reference value is uniformly processed to zero. A stability offset constant is set, which is a positive number, and it participates in the calculation together with the equivalent disturbance acceleration and the disturbance reference value during the construction of the disturbance normalized response function; At any given moment, when the absolute value of the equivalent disturbance acceleration and the absolute value of the disturbance reference value are both zero, the disturbance normalized response coefficient is set to zero. At any given moment, when the absolute value of the equivalent disturbance acceleration and the absolute value of the disturbance reference value are not both zero, the disturbance normalized response coefficient is calculated based on the proportional relationship between the current value of the equivalent disturbance acceleration, the absolute value of the equivalent disturbance acceleration, the absolute value of the disturbance reference value, and the stability offset constant.

5. The method for optimizing control parameters of a public space lighting system according to claim 4, characterized in that, The perturbation variance is constructed based on the deviation between the equivalent perturbation acceleration and the perturbation reference value. This variance, combined with a stable offset constant, yields the adaptive brightness suppression ratio. The suppression ratio is set to zero when both the variance and the reference value are zero. Specifically, this includes: When the system running time is greater than or equal to the disturbance integration time window length, at each current moment, the deviation between the equivalent disturbance acceleration before the current moment and the disturbance reference value is averaged according to the disturbance integration time window length to construct the disturbance variance at the current moment. During the period when the system runtime is less than the length of the disturbance integration time window, the disturbance variance is set to zero as a boundary treatment for the startup phase. At any given moment, when both the disturbance variance and the disturbance baseline value are zero, the adaptive brightness suppression ratio is set to zero. At any given moment, when the disturbance variance and the disturbance reference value are not both zero, the adaptive brightness suppression ratio is calculated based on the proportional relationship between the combined results of the disturbance variance, the absolute value of the disturbance reference value, and the stable offset constant.

6. The method for optimizing control parameters of a public space lighting system according to claim 5, characterized in that, The calculation of theoretical brightness based on reference illumination luminance, perturbation normalized response coefficient, and adaptive luminance suppression ratio, followed by upper and lower limit limiting, specifically includes: At any given moment, read the reference illumination brightness, the perturbation normalized response coefficient, and the adaptive brightness suppression ratio, and calculate the theoretical brightness output according to the product and difference relationship between the three. At any given moment, the calculated theoretical brightness output is compared with the lower limit of lighting brightness. When the theoretical brightness output is lower than the lower limit of lighting brightness, the theoretical brightness output is corrected to the lower limit of lighting brightness. At any given moment, compare the theoretical brightness output with the upper limit of the lighting brightness. If the theoretical brightness output is higher than the upper limit of the lighting brightness, correct the theoretical brightness output to the upper limit of the lighting brightness. After completing the dual limiting correction of the lower and upper limits, the corrected brightness is used as the brightness output after limiting adjustment, and provided to the subsequent residual correction mechanism and lighting drive module.

7. The method for optimizing control parameters of a public space lighting system according to claim 6, characterized in that, The calculation of the disturbance offset residual, the construction of residual weighting factors and residual modulation coefficients based on the residual and disturbance variance, the correction of the adaptive brightness suppression ratio, the obtaining of the corrected brightness and limiting, and the output of the final brightness specifically include: At any given moment, the difference between the current equivalent disturbance acceleration and the current disturbance reference value is defined as the disturbance offset residual. Construct residual weighting factors based on the perturbation offset residual and the current perturbation variance; The residual modulation coefficients are constructed based on the residual weighting factor, and the residual modulation coefficients are compressed using an exponential form. The residual modulation coefficients are combined with the adaptive luminance suppression ratio to form the corrected luminance suppression ratio; The corrected luminance output is recalculated based on the corrected luminance suppression ratio, reference illumination luminance, and disturbance normalized response coefficient. The corrected brightness output is compared with the lower limit and upper limit of the lighting brightness. The corrected brightness output is then subjected to a limiting process to obtain the final brightness output, which is then provided to the lighting driver module and the next step of timing control.

8. The method for optimizing control parameters of a public space lighting system according to claim 7, characterized in that, The process involves constructing a disturbance response distribution trajectory according to the refresh cycle, obtaining an equivalent disturbance acceleration trajectory sequence, generating disturbance trend information based on the difference with the previous discrete refresh time, and constructing a dimensionless disturbance trend coefficient by combining the disturbance reference value, disturbance variance, and disturbance increment normalization scaling factor. This process calculates and limits the predicted brightness, forming a lighting timing control command sequence that includes the final brightness output and the predicted limited brightness for the next cycle. Specifically, this includes: Set the lighting control refresh cycle to generate a discrete refresh time sequence starting from the system startup time, with each discrete refresh time corresponding to a continuous time point; Set the disturbance history tracking duration, and determine the maximum number of historical steps that can be traced at each discrete refresh time based on the refresh cycle and the disturbance history tracking duration; At any discrete refresh time, backtrack several refresh cycles according to the refresh cycle, collect the equivalent disturbance acceleration history values ​​within the disturbance history tracking time range, and form the equivalent disturbance acceleration trajectory sequence corresponding to the current discrete refresh time. At any discrete refresh time, the equivalent disturbance acceleration at the previous discrete refresh time is subtracted from the current equivalent disturbance acceleration to construct the equivalent disturbance increment for the current refresh cycle. The equivalent disturbance increment is set to zero in the first refresh cycle after the system starts. At any discrete refresh time, based on the current disturbance reference value, the current disturbance variance, and the stability offset constant, a normalized scaling factor for the disturbance increment is constructed, and the equivalent disturbance increment is compared with the normalized scaling factor to obtain the dimensionless disturbance trend coefficient. At any discrete refresh time, set the dimming response coefficient, and calculate the predicted brightness for the next refresh cycle based on the dimensionless disturbance trend coefficient, dimming response coefficient, and maximum brightness offset of the current refresh cycle. At any discrete refresh time, the predicted brightness for the next refresh cycle is subjected to upper and lower limit limiting processing. When the predicted brightness is lower than the lower limit of the lighting brightness, the predicted brightness is set to the lower limit of the lighting brightness. When the predicted brightness is higher than the upper limit of the lighting brightness, the predicted brightness is set to the upper limit of the lighting brightness, thus obtaining the predicted limited brightness. At any discrete refresh time, the final brightness output is combined with the predicted limited brightness of the next refresh cycle to form a lighting timing control command sequence. The lighting drive module uses the final brightness output in the current refresh cycle and the corresponding predicted limited brightness in the next refresh cycle, forming a continuous lighting timing control process.

Images