Intelligent street lamp constant illumination and time-sharing dimming cooperative control method and system and single lamp controller

Abstract

The application discloses a smart street lamp constant-illumination and time-sharing low-brightness cooperative control method and system and a single lamp controller, and particularly relates to the technical field of street lamp control. The single lamp controller stores a light efficiency reference quantity of a lamp, a known amplitude step is applied to a dimming setting value in operation, a steady-state ambient illumination total value before and after the step is collected and is subtracted to determine a current self-illumination contribution component. The component is deducted from the ambient illumination total value to obtain a real ambient light component, and the component is divided by the product of the current dimming setting value and the synchronously collected power to obtain a current actual light efficiency quantity. The light efficiency reference quantity is compared with the light efficiency quantity to determine a light decay factor. According to the light decay factor, a target illumination value and a current period operation constraint determined by a real-time clock and a time control strategy, a compensation release quantity is determined within a reserved dimming upper limit, and a dimming output is generated, which is used to eliminate the dimming oscillation problem caused by the self-light self-feedback.

Description

Technical Field

[0001] This invention relates to the field of street light control technology, and more specifically, to a method, system, and single-lamp controller for coordinated control of constant illuminance and time-sharing dimming of smart street lights. Background Technology

[0002] Existing smart streetlights typically handle constant illuminance control, time-based dimming control at night, and LED light decay compensation separately. Constant illuminance control relies on a photosensitive device on the pole to collect ambient illuminance and perform closed-loop dimming to maintain road surface illuminance. However, after the light is turned on, the total ambient illuminance value collected by the photosensitive device is actually the sum of the actual ambient light and the illuminance contribution from the light itself reflected back to the photosensitive device through the road surface and environment; these two are not separated. If the controller directly uses this total value as ambient light, the reading will increase when the light is brightened, misinterpreting it as an increase in ambient light and dimming it. After dimming, the reading will decrease again, leading to brightening again, thus generating self-oscillation and a persistently dim condition in steady state. This problem is particularly prominent in scenarios with low poles, high elevations, and highly reflective road surfaces where self-reflection from the light is significant.

[0003] Nighttime time-sharing dimming control typically reduces lights to a certain percentage based on pre-determined adjustment periods, without considering the actual ambient light and road illumination requirements. After dimming at night, some roads still fail to meet safety standards for illumination. LED light decay compensation often uses a server platform to pre-set a group light decay curve, compensating based on the cumulative runtime. This is open-loop, group-averaged, and non-in-situ; it cannot reflect individual lamp differences, nor can it determine whether the compensated road surface illumination meets standards. Under the same fixed power ratio, the actual illumination during the new lamp period and the aging period are different, with road surface illumination decreasing year by year. These three control methods operate independently, without forming a coordinated chain at the individual lamp controller level, making it difficult to simultaneously achieve illumination stability, energy saving, and aging compensation. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, this invention provides a method, system, and single-lamp controller for coordinated control of constant illuminance and time-sharing dimming in smart streetlights. In a single-lamp controller with only a single pole-mounted illuminance sensor, dimming execution, and electrical metering, a known step is applied to the dimming output, and the steady-state ambient illuminance before and after the step is differentially analyzed to separate the lamp's current self-illuminance contribution component. This same self-illuminance contribution component is then used for two independent control objectives: firstly, the component is subtracted from the total ambient illuminance to obtain the true ambient light component, used for constant illuminance dimming to eliminate self-light pollution; secondly, the component is compared with the synchronous power to obtain the current actual luminous efficacy, which is then compared with the initial luminous efficacy reference value to in-situ calculate the lamp's light decay factor for compensation, thus solving the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for coordinated control of constant illuminance and time-sharing dimming in smart streetlights, applied to a single-lamp controller having an ambient illuminance acquisition module, a real-time clock, an electrical metering module, a dimming execution module, and a remote communication module, comprising: Store the luminous efficacy reference value of this lamp. The luminous efficacy reference value characterizes the relationship between the lamp's own illuminance contribution in the initial state and the dimming output and power consumption. The dimming execution module applies a step of known amplitude to the dimming setpoint, and collects the steady-state total ambient illuminance before and after the step. The difference between the two steady-state total values ​​is divided by the difference in the dimming setpoint to obtain the unit dimming self-illuminance contribution response. The unit dimming self-illuminance contribution response is then multiplied by the current dimming setpoint to obtain the current self-illuminance contribution component of the lamp. The self-illuminance contribution component is then subtracted from the total ambient illuminance to obtain the true ambient light component. When the current period is in the self-sensing constant illuminance mode, the actual ambient light component is used as the input for determining whether to turn on the lights, whether to turn off the lights, and the calculation of constant illuminance dimming requirements. Divide its own illuminance contribution component by the product of the current dimming setting value and the power synchronously collected by the electrical metering module to obtain the current actual luminous efficacy. Then compare the current actual luminous efficacy with the luminous efficacy reference value to determine the current light decay factor of this lamp. Based on the light decay factor, target illuminance value, and the current time period operation constraints determined by the real-time clock and timing strategy, the compensation release amount is determined within the reserved dimming upper limit and a dimming output is generated, which is then executed by the dimming execution module.

[0006] Preferably, the light decay factor is obtained by: constructing a ratio sequence from the ratio of the current actual luminous efficacy to the luminous efficacy reference obtained multiple times; extracting the monotonically varying component in the long time window and the fluctuating component in the short time window of the ratio sequence; assigning the monotonically varying component to luminous efficacy decay and the fluctuating component to road surface reflectivity change; and updating the light decay factor only with the monotonically varying component; the long time window is an observation window for capturing the monotonically varying current actual luminous efficacy ratio sequence on a monthly or yearly scale, and the short time window is an observation window for capturing the daily cyclical fluctuation of the current actual luminous efficacy ratio sequence.

[0007] Preferably, the luminous efficacy attenuation trend index is determined by the product of the percentage of duration during which the ratio sequence has a negative downward slope and a sustained sign within a long time window and the slope amplitude; the reflection disturbance fluctuation index is determined by the magnitude of the detrended residual of the ratio sequence within a short time window and the absolute value of the normalized correlation coefficient between the detrended residual sequence and the rainfall indicator sequence, temperature and humidity state sequence, or daily periodic template; wherein, the daily periodic template is formed by the detrended residual of the same lamp position at the same historical moment, and the normalized correlation coefficient is calculated by the detrended residual sequence aligned according to the sampling time and the rainfall indicator sequence, temperature and humidity state sequence, or daily periodic template; when the ratio of the luminous efficacy attenuation trend index to the reflection disturbance fluctuation index is greater than 1, the luminous efficacy attenuation factor is updated; when the ratio is not greater than 1, the luminous efficacy attenuation factor update is frozen. To ensure the two indices can be compared on the same scale, the controller first normalizes each index to a dimensionless value by dividing it by its minimum resolvable value: the luminous efficacy decay trend index is normalized using the reference slope obtained by dividing the minimum resolvable change in the luminous decay factor by the duration of the long time window; the reflection disturbance fluctuation index is normalized using the minimum resolvable residual amplitude corresponding to the quantization resolution of the ambient illuminance acquisition module. After normalization, the two indices have the same dimension. When the ratio of the normalized luminous efficacy decay trend index to the reflection disturbance fluctuation index is greater than 1, it is determined to be actual luminous decay of the luminaire, and the luminous decay factor is updated with the monotonically slowly varying component within the long time window to determine the compensation release amount; when the ratio is not greater than 1, it is determined to be meteorological or dust disturbance of the road surface reflectivity, and the luminous decay factor update is frozen, maintaining the original compensation unchanged. Using "1" as a boundary indicates that the continuous monotonically downward intensity is comparable to the rapid fluctuation intensity that can be explained by meteorology and cycles, and at this time, the frozen side is processed from the perspective of suppressing false compensation.

[0008] Preferably, the method for obtaining the luminous efficacy reference quantity is as follows: during the installation or initialization phase, the total steady-state ambient illuminance and synchronous power are collected at at least two different dimming settings. The difference in the total illuminance between adjacent dimming settings is divided by the difference between adjacent dimming settings to obtain the unit dimming self-illuminance contribution response. Then, the unit dimming self-illuminance contribution response is normalized to 1 by the average value or steady-state representative value of adjacent sampled power to obtain the self-illuminance contribution under unit dimming and unit power, which is stored locally as the luminous efficacy reference quantity.

[0009] Preferably, the compensation release amount is obtained by: estimating the junction temperature or cumulative thermal stress of the light-emitting diode based on the power collected by the electrical metering module and the ambient temperature obtained from the meteorological data sent by the drive power supply casing temperature acquisition terminal or remote server, and using the estimated junction temperature or thermal stress contraction compensation release amount amplitude or step size; the release of the reserved dimming upper limit is a graded release, initially locked at the first upper limit, and gradually released to the second upper limit higher than the first upper limit as the light decay factor decreases; the first upper limit is the power upper limit corresponding to the initial dimming output locked when the target illuminance value is met in the initialization stage, and the second upper limit is the highest allowable power upper limit formed by the first upper limit plus the reserved dimming upper limit; the second upper limit does not exceed the rated output power of the lamp and the allowable power boundary of the junction temperature or cumulative thermal stress.

[0010] Preferably, the method for obtaining the self-illuminance contribution component is as follows: the total steady-state ambient illuminance value before and after the step is annotated with the acquisition time using a real-time clock, and the ambient light natural gradual change envelope is obtained by extrapolating the ambient light natural change rate within the adjacent short window outside the step window; when the ambient light gradual change between two acquisition times exceeds the ambient light natural gradual change envelope, the corresponding sample is discarded and then differential is performed; when the ambient illuminance acquisition module supports synchronous sampling with the dimming phase, the known amplitude step is formed by periodic micro-step disturbance, and the single lamp controller extracts the steady-state representative value of the synchronously acquired illuminance sequence according to the high phase segment and low phase segment of the periodic micro-step disturbance, respectively, and divides the difference between the steady-state representative value of the high phase segment and the steady-state representative value of the low phase segment by the difference between the dimming setting value of the high phase segment and the dimming setting value of the low phase segment to obtain the self-illuminance contribution response per unit dimming, and multiplies the self-illuminance contribution response per unit dimming by the current dimming setting value to obtain the self-illuminance contribution component.

[0011] In the periodic micro-step disturbance mode, the single-lamp controller synchronously collects illuminance sampling values ​​according to the phase marker of the micro-step disturbance. The illuminance sampling values ​​in the high-phase stable segment are generated into a high-phase steady-state representative value, and the illuminance sampling values ​​in the low-phase stable segment are generated into a low-phase steady-state representative value. The difference between the high-phase steady-state representative value and the low-phase steady-state representative value corresponds to the change in its own illuminance contribution caused by the micro-step disturbance. The single-lamp controller divides this difference by the difference between the high-phase dimming setting value and the low-phase dimming setting value to obtain the unit dimming self-illuminance contribution response.

[0012] Preferably, the known amplitude step is formed by an active test step or a brightness switch triggered by the current time-sharing dimming strategy; when the known amplitude step is formed by a brightness switch triggered by the current time-sharing dimming strategy, the single lamp controller collects the total steady-state ambient illuminance before and after the switch when the brightness switch occurs, in order to determine its own illuminance contribution component.

[0013] Preferably, the remote communication module reports a running data frame to the remote server. The running data frame includes the lamp identifier, cumulative running time, light decay factor, current actual luminous efficacy, and compensation release amount. The cumulative running time is used as a maintenance reference field, and the light decay factor and current actual luminous efficacy are used as the main fields for lamp-by-lamp aging sorting and abnormal warning. The module also receives the time control strategy frame and target illuminance value sent by the server.

[0014] To achieve the above objectives, the present invention provides the following technical solution: a single-lamp controller for executing the intelligent street light constant illuminance and time-sharing dimming coordinated control method, comprising: The system includes an ambient illuminance acquisition module, a real-time clock, an electrical metering module, a dimming execution module, a remote communication module, and a processor. The processor is configured as follows: Store the luminous efficacy reference value of this lamp. The luminous efficacy reference value characterizes the relationship between the lamp's own illuminance contribution in the initial state and the dimming output and power consumption. The dimming execution module applies a step of known amplitude to the dimming setpoint. The ambient illuminance acquisition module collects the steady-state total ambient illuminance before and after the step. The difference between the two steady-state total values ​​and the difference between the dimming setpoint are used to determine the current self-illuminance contribution component of the lamp. The self-illuminance contribution component is then subtracted from the total ambient illuminance to obtain the true ambient light component. When the strategy issued by the real-time clock and remote communication module determines that the current period is in the self-sensing constant illuminance mode, the real ambient light component is used as the input for the calculation of the light-on judgment, light-off judgment, and constant illuminance dimming requirements. The actual luminous efficacy is obtained by dividing the lamp's own illuminance contribution by the product of the current dimming setting and the power synchronously collected by the electrical metering module. The actual luminous efficacy is then compared with the luminous efficacy reference value to determine the lamp's current light decay factor. Based on the light decay factor, the target illuminance value, and the current time period operation constraints determined by the real-time clock and timing strategy, the compensation release amount is determined within the reserved dimming upper limit, and dimming output is generated and executed by the dimming execution module; among them, the self-illuminance contribution component is used to obtain the real ambient light component and determine the light decay factor.

[0015] To achieve the above objectives, the present invention provides the following technical solution: a smart street light constant illuminance and time-sharing brightness reduction coordinated control system, used to realize a smart street light constant illuminance and time-sharing brightness reduction coordinated control method, including the single lamp controller as described in claim 9 and a server connected to each single lamp controller via remote communication; each single lamp controller reports its light decay factor and current actual luminous efficacy to the server, and the server generates a lamp-by-lamp aging sort based on multiple light decay factors and sends a time control strategy and target illuminance value to each single lamp controller.

[0016] The technical effects and advantages of this invention are as follows: (1) The constant illuminance closed loop is driven by step difference to remove the component of the lamp itself, so that the constant illuminance closed loop will not oscillate due to its own light self-feedback and the steady-state dimming will disappear. The lamp becomes usable in strong reflection scenarios; the light decay is back-calculated by the ratio of its own illuminance contribution component to the synchronous power relative to the initial reference, so that the light decay compensation is free from the platform preset curve, realizing lamp-by-lamp in-situ self-calibration and with lamp-by-lamp aging sorting and abnormal warning.

[0017] (2) The difference between the time constants of the two types of disturbances is used to separate the change in road surface reflectivity from the change in lamp light decay. Rain and dust accumulation are not misjudged as light decay, and the compensation is not erroneous. The release amount of compensation is constrained by junction temperature or accumulated thermal stress to achieve balance in restoring road surface illuminance and inhibiting accelerated aging. The above effects are shared by the same step differential measurement. That is, a single measurement simultaneously closes the two control targets of pollution-free constant illuminance and self-calibrated light decay, which will be a synergistic effect that exceeds the simple assembly of the three types of control. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall process of the collaborative control method of the present invention.

[0019] Figure 2 This is a schematic diagram of the module connection of the single-lamp controller of the present invention.

[0020] Figure 3 This is a schematic diagram illustrating the principle of the present invention: step differential separation of its own illuminance contribution component and restoration of the true ambient light component.

[0021] Figure 4 This is a schematic diagram of the process of separating and compensating for time constants based on the current actual luminous efficacy sequence of the present invention. Detailed Implementation

[0022] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0023] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0024] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this application and its application or use.

[0025] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0026] See Figures 1 to 4 This invention provides a method for coordinated control of constant illuminance and time-sharing dimming of smart streetlights, as shown in the figure, comprising: Step 1: Store the luminous efficacy reference value of this lamp. The luminous efficacy reference value represents the relationship between the lamp's own illuminance contribution in the initial state and the dimming output and power consumption. Step 2: The dimming execution module applies a known amplitude step to the dimming setpoint, and collects the steady-state total ambient illuminance before and after the step. The difference between the two steady-state total values ​​is divided by the difference in the dimming setpoint to obtain the unit dimming self-illuminance contribution response. The unit dimming self-illuminance contribution response is then multiplied by the current dimming setpoint to obtain the current self-illuminance contribution component of the lamp. The self-illuminance contribution component is subtracted from the total ambient illuminance to obtain the true ambient light component. Step 3: Determine the operating constraints for the current time period using the real-time clock and time control strategy; when the current time period is in the self-sensing constant illuminance mode, use the actual ambient light component as the input for calculating the light-on judgment, light-off judgment, and constant illuminance dimming requirements. Step 4: Divide the lamp's own illuminance contribution component by the product of the current dimming setting and the power synchronously collected by the electrical metering module to obtain the current actual luminous efficacy. Then compare the current actual luminous efficacy with the luminous efficacy reference value to determine the current light decay factor of the lamp. Based on the light decay factor, the target illuminance value, and the current time period operation constraints determined by the real-time clock and time control strategy, determine the compensation release amount within the reserved dimming upper limit and generate dimming output, which is then executed by the dimming execution module. The self-illuminance contribution component is used in step two to obtain the real ambient light component and in step four to determine the light decay factor.

[0027] Furthermore, step one is used to establish the luminous efficacy reference quantity of the luminaire in its initial state: Background: This step occurs during the initialization phase after the single-lamp controller is installed on the light pole and powered on for the first time, or after the lamp is replaced. At this time, the lamp has not yet experienced light decay, and the road surface and environment are in a known clean state. This is the only opportunity to establish a reference benchmark upon which all subsequent judgments depend. If a benchmark that can be compared over a long period is not established at this stage, any luminous efficacy observations obtained during subsequent operation will lack a comparison object, and back-calculation of light decay will be impossible. Therefore, the actual scenario for this step is: after the construction personnel complete the power-on and alignment of the lamp, the single-lamp controller automatically completes a two-point sampling and establishes the benchmark without the intervention of a manual illuminance meter, relying solely on its existing environmental illuminance acquisition module, dimming execution module, and electrical metering module.

[0028] The core objective of this step is to determine and store the luminous efficacy reference value R0 in the initial clean state of the luminaire. This value characterizes the illuminance contribution of the luminaire itself, reflected from the road surface and environment back to the photosensitive device, relative to the dimming setting and power consumption. The physical basis is as follows: the total ambient illuminance E_total received by the photosensitive device on the pole is a linear superposition of the actual ambient light component E_amb and the luminaire's own illuminance contribution component E_self; that is, the total ambient illuminance equals the sum of the actual ambient light component and the luminaire's own illuminance contribution component. The luminaire's own illuminance contribution component E_self originates from the luminous flux emitted by the LED, reflected from the road surface and environment, and re-enters the photosensitive device. Its magnitude is proportional to the product of the road surface reflectivity, the installation geometry factor, the current LED luminous efficacy, and the dimming setting. When the installation geometry factor remains fixed at the same luminaire position, the quotient obtained by dividing the luminaire's own illuminance contribution component by the dimming setting and then by the power only varies with the road surface reflectivity and the LED luminous efficacy, thus serving as an observational measure characterizing luminous efficacy for long-term tracking.

[0029] In this step, the outputs of the ambient illuminance acquisition module, the dimming execution module, and the electrical metering module are all closed-loop. Specifically: the dimming execution module generates two known, different dimming settings; the ambient illuminance acquisition module reads the total ambient illuminance value after each dimming setting has stabilized; and the electrical metering module reads the corresponding power. The outputs of these three modules are then differentially analyzed and normalized within the controller to form a separate luminous efficacy reference quantity. The key points are synchronization and windowing; the sampling between the two modules must fall within the same time window where the natural ambient light remains essentially constant, ensuring that the actual ambient light component is included in both samplings.

[0030] Furthermore, the specific implementation of step one includes: Steady-state sampling of two dimming setpoints. During the initialization phase, the single-lamp controller first sends a first dimming setpoint to the LED driver power supply via the dimming execution module, which is recorded as the first setpoint. After the lamp's light output stabilizes, the ambient illuminance acquisition module reads the total ambient illuminance value at this time, which is recorded as the first total value. At the same time, the electrical metering module reads the corresponding power, which is recorded as the first power. Subsequently, the controller sends a second dimming setpoint via the dimming execution module, which is recorded as the second setpoint. The second setpoint and the first setpoint maintain a difference that can be distinguished by the ambient illuminance acquisition module. After the lamp's light output stabilizes again, the second total value and the second power are read.

[0031] In a preferred embodiment, the first setpoint is set to 50% and the second setpoint to 80%. This range ensures that the difference between the two self-illuminance contribution components is significantly higher than the reading noise of the photosensitive device, while avoiding the locking zone of the reserved dimming upper limit due to approaching full output, thus guaranteeing stable and resolvable differential results. This provides subsequent differential analysis with two clean and repeatable sampling points, avoiding the inherent defect of single-point sampling in separating ambient light, and also avoiding insufficient signal-to-noise ratio caused by excessively small setpoint differences.

[0032] Differential illuminance elimination eliminates the real ambient light component. The controller subtracts the first total value from the second total value to obtain the difference between the two total ambient illuminance values. Since the two samples fall within the same time window and the real ambient light component remains consistent, the real ambient light component in this difference is completely canceled out, leaving only the difference in the self-illuminance contribution component between the two dimming settings. The controller then divides this difference by the difference between the second and first settings to obtain the unit dimming self-illuminance contribution response. The differential illuminance used here is based on the physical fact that the real ambient light component is a common mode within the same window, while the self-illuminance contribution component is a differential mode, eliminating the common mode and retaining the differential mode. The effect of this technique is that, without relying on an external illuminance meter or turning off the lamps, it extracts the self-illuminance contribution component, which was previously impossible to observe separately, from the total illuminance for the first time, clearing the way for the establishment of a luminous efficacy reference quantity.

[0033] Power normalization and baseline solidification. The controller takes the power of the operating segment during the two samplings, divides the above-mentioned unit dimming self-illuminance contribution response by the power, and obtains the self-illuminance contribution under unit dimming and unit power, i.e., the luminous efficacy reference quantity R0. Then, R0, along with the calibration time and installation geometry factor identifier, is written into local non-volatile storage as the anchor point for lifelong comparison of this lamp position. The physical meaning of the luminous efficacy reference quantity R0 is the normalized representation of the product of road surface reflectivity, installation geometry factor, and initial LED luminous efficacy. The effect of this technology is that it transforms the one-time, condition-sensitive raw reading into a stable reference with clear dimensions that can be directly compared by any operation observation many years later, making it possible to back-calculate the light decay of each lamp in place without relying on the group light decay curve preset on the server platform.

[0034] In an alternative approach, the two-point sampling can be expanded to three or more points. A linear fit is then performed on different dimming settings and the corresponding total ambient illuminance, using the fit slope to replace the difference between the two points and further suppress single-point reading noise. This is equivalent to two-point differential sampling and can be performed independently. Another alternative implementation is possible when the luminaire has been removed but the pole and road surface remain unchanged. In this case, the luminous efficacy of the LED can be measured without altering the previously installed geometric factor markings, saving recalibration time. The above method for obtaining the luminous efficacy reference quantity is completely separated from the operational logic of subsequent steps and can be applied independently or used in a decoupled combination.

[0035] Furthermore, step two is used to perform step differential separation of its own light and restore the real ambient light to drive constant illuminance: Background: This step occurs when the luminaire enters normal nighttime operation and is configured for self-sensing constant illuminance mode by the time control strategy. The actual scenario is as follows: the light is on, and the total ambient illuminance value read by the photosensitive device on the pole includes both the actual ambient light from the sky and surrounding area, as well as the light reflected back from the road surface after the light itself illuminates. If this total value is directly used as ambient light to drive the constant illuminance closed loop, it will result in self-oscillations: the light turns on, the reading increases, the controller misinterprets this as brightening and dims, the reading decreases after dimming, and then the light is brightened again, leading to a consistently dimmed reading in steady state. The premise of this step is to separate the light from the total illuminance without turning off the light or adding a second sensor, restoring the true ambient light, and feeding it into the constant illuminance closed loop. The self-illuminance contribution component separated in this step will be reused in step four.

[0036] The core objective of this step is to separate the lamp's current self-illuminance contribution component E_self in real time during operation, and subtract this component from the total ambient illuminance value E_total to obtain the true ambient light component E_amb after removing the lamp's self-illuminance component. This true ambient light component is then used as the sole input for determining whether the lamp is turned on, off, or for calculating constant illuminance dimming requirements. Its physical basis remains the linear superposition of illuminance: assuming the natural ambient light remains essentially constant within a short window, actively and knowingly changing the dimming setpoint will cause the change in the total ambient illuminance value to originate entirely from the change in the self-illuminance contribution component, while the true ambient light component is canceled out in the change. In other words, the controller knowingly changes the lamp's brightness and observes the change in reading; the changed portion must be caused by the lamp itself, thus separating the lamp's component from the ambient component. This is similar to the two-point calibration in Step 1, the difference being that Step 1 establishes a baseline in a clean initial state, while this step extracts the current value in real time in any operating state.

[0037] The synergistic relationship between this step and steps three and four lies in the fact that the product of the single step differential in this step has two paths: one is the real ambient light component E_amb, which is used to supply the decontaminated ambient light to the downstream constant illuminance closed loop; the other is the self-illuminance contribution component E_self, which is taken by step four to calculate light decay. In other words, with only one active step and one photosensitive device, this step simultaneously provides input for two independent control objectives. This arrangement of measuring once and closing two loops simultaneously allows the solution to break away from the conventional approach of assembling constant illuminance control, time-sharing brightness reduction, and light decay compensation separately, resulting in a synergistic effect that exceeds conventional R&D expectations after combining various technical features.

[0038] Furthermore, the specific implementation of step two includes: A known step is applied, and the steady-state total value before and after is collected. When the current time period is in self-sensing constant illuminance mode and the luminaire is at a certain steady-state dimming setting value (denoted as the setting value before the step), the controller, through the dimming execution module, increases the dimming setting value by a known amplitude to the setting value after the step. After the luminaire's light output stabilizes again, the ambient illuminance acquisition module reads the steady-state ambient illuminance total value before and after the step, denoted as the total value before the step and the total value after the step. This known amplitude is the difference between the two dimming setting values ​​issued by the dimming execution module itself, a quantity that the controller can set and read, rather than an externally unknown quantity. In a preferred embodiment, this known amplitude is subject to two constraints: first, the amplitude does not exceed the upper limit of brightness jumps that are imperceptible to the human eye, preventing pedestrians or drivers from perceiving streetlight flickering; second, the interval between two adjacent steps is not less than the light output stabilization time required for the dimming execution module to complete the step response and enter a steady state, ensuring that the step frequency is below the range of frequent brightness changes perceptible to the human eye, thus avoiding visual disturbance. The effect of this technique is that it provides a single, imperceptible brightness change in exchange for a single observation pair that can be used for differential analysis, without sacrificing visual comfort and safety for road traffic.

[0039] The differential sensor extracts the lamp's own illuminance contribution component and reconstructs the true ambient light component. The controller subtracts the total value after the step change from the total value before the step change to obtain the change in the total ambient illuminance value. This change is divided by the difference between the setpoint after the step change and the setpoint before the step change to obtain the lamp's own illuminance contribution response under the current operating conditions. This response is then multiplied by the current actual dimming setpoint to obtain the lamp's current own illuminance contribution component, E_self. The controller then subtracts the own illuminance contribution component from the current total ambient illuminance value to obtain the true ambient light component, E_amb. To prevent non-negligible gradual changes in ambient light within the step window from contaminating the differential sensor, the controller uses a real-time clock to annotate the acquisition time for the two readings before and after the step change, and extrapolates the natural rate of change of ambient illuminance within the adjacent short window outside the step window to obtain the ambient light natural gradual change envelope. When the ambient light gradual change between the two acquisition times exceeds the ambient light natural gradual change envelope, the step window is deemed unclean, the sample is discarded, and a re-acquisition is scheduled. The effect of this technology is to switch the input of the constant illuminance closed loop from the total illuminance containing the lamp component to the real ambient light excluding the lamp component, thereby fundamentally eliminating the oscillation and steady-state dimness caused by self-light feedback.

[0040] The constant illuminance judgment is driven by the actual ambient light component. The controller uses the actual ambient light component E_amb as the sole input to participate in the calculation of light-on judgment, light-off judgment, and constant illuminance dimming requirements: when the actual ambient light component is lower than the illuminance level corresponding to the light-on judgment, the lights are turned on; when it is higher than the illuminance level corresponding to the light-off judgment, the lights are turned off; when it is in between, the required dimming output is calculated based on the difference between the target illuminance value and the actual ambient light component. Since the actual ambient light component with the lamp component removed is used here, the controller's dimming action will no longer lower its own input reading, and the closed loop is no longer self-excited. The effect of this technology is that in scenarios with significant self-reflection of the lamp, such as highly reflective road surfaces and low poles, the light-sensing constant illuminance control becomes stable and usable, dimming no longer jitters, and the steady state at night is no longer too dim.

[0041] In an optional equivalent implementation, when the ambient illuminance acquisition module supports synchronous sampling with the dimming phase, the single step in technique A can be replaced by applying a periodic perturbation to the dimming output and performing correlation demodulation on the synchronously acquired illuminance to extract the self-illuminance contribution component with the same frequency as the perturbation from the frequency domain. This is suitable for the evening and early morning periods when natural ambient light fluctuates rapidly and a single step window is difficult to maintain cleanliness. Specifically, the controller superimposes a periodic perturbation of known frequency onto the dimming output, multiplies the illuminance sequence synchronously acquired by the ambient illuminance acquisition module with the perturbation reference signal, and then performs low-pass integration. The resulting DC component is proportional to the self-illuminance contribution component, and the component is extracted in the frequency domain accordingly.

[0042] In another optional embodiment, the controller reuses the brightness switching triggered by the current time-sharing dimming strategy itself as the step source—since the time-sharing strategy is supposed to switch brightness at a set time, the controller collects the steady-state total value before and after the switch to perform differential analysis when the switch occurs, thus eliminating the need for additional active interference and achieving self-light separation with zero additional interference. When encountering abnormal operating conditions such as unavailable dimming steps or communication interruption, the single-lamp controller enters a degradation backup path: it uses the self-illuminance contribution response obtained from the most recent effective separation as a temporary value to maintain operation, and updates it again after the operating conditions recover. The above-mentioned self-light separation feature and the downstream constant illuminance driving feature are logically independent and can be implemented independently.

[0043] Furthermore, step three is used for determining the current operating mode and converting the time-sharing brightness reduction power boundary value: Background: This step follows the output of step two and occurs at the point where the operating mode for the current time period is determined and switched. In practice, the single-lamp controller needs to first determine whether to base its output on ambient light or a timetable, and then decide whether to send the real ambient light component reconstructed in step two for constant illuminance dimming, or to use the time-based dimming power boundary value given by the time-controlled strategy to constrain the output. This step itself is a standard process branch and state determination, but it is a necessary link between steps two and four, therefore it is disclosed in a simplified but complete manner.

[0044] The core objective of this step is to determine the current operating mode for the current time period by combining the current time of the real-time clock and the time control strategy issued by the remote communication module, and to select the control criteria based on this operating mode. The operating modes include at least three states: self-sensing constant illuminance mode, fixed brightness mode, and time-division dimming mode. The real-time clock provides the current time, day of the week, and date, while the time control strategy provides the mode and brightness ratio to be used for each time period and date. The intersection of these two parameters uniquely determines the current state to be entered.

[0045] Furthermore, the specific implementation of step three includes: Operating mode determination. The controller reads the current time from the real-time clock and retrieves the mode field corresponding to the time period from the locally cached time control strategy. If the self-sensing constant illuminance mode is matched, the entry condition is met, and the actual ambient light component restored in step two is called to calculate the constant illuminance. If the fixed brightness mode or time-sharing dimming mode is matched, the entry condition is met, and the brightness ratio of the time period is converted into the time-sharing dimming power boundary value. The state determination only determines the flow direction, and the entry conditions and subsequent call objects of each state are clearly defined to avoid dangling branches.

[0046] The conversion of time-sharing dimming power boundary values. When the current time period is in time-sharing dimming mode or fixed brightness mode, the controller converts the brightness ratio configured for that time period into time-sharing dimming power boundary values, which serve as the upper limit constraint for the current dimming output, rather than directly as the final output. The significance of this processing is that the time-sharing strategy is no longer a hard execution of reducing the brightness to a fixed value at a given time, but is transformed into a boundary condition that works in conjunction with constant illuminance requirements and light decay compensation in step four—this is the key connection in rewriting the time-sharing strategy from an execution item to a constraint item.

[0047] Furthermore, step four is used for in-situ optical attenuation reverse calculation, time constant separation, and dual-exponential gating and bounded compensation collaborative output: Background: As lighting fixtures are used for months and years, the luminous efficacy of LEDs gradually decreases. The actual road illuminance corresponding to the same dimming setting is different in the new lamp period and the aging period. If the initial fixed ratio of time-based dimming is still applied, the road illuminance will decrease year by year. At the same time, the road surface becomes wet in rainy weather and accumulates dust over time, which also changes the reflectivity. Both of these factors will reduce the contribution of the illuminance read by the photosensitive device. A single decrease cannot distinguish whether it is due to lamp aging or road surface dirt or wetness. This step aims to solve the problem of measuring the true light decay of a lamp in situ, lamp by lamp, under the condition of only a single photosensitive device and an electrical metering module, and to compensate for it without being misled by road disturbances or accelerating aging due to the compensation itself.

[0048] The core objectives of this step are twofold: First, to compare the self-illuminance contribution component E_self separated in step two with the power P synchronously collected by the electrical metering module to obtain the current actual luminous efficacy R, and then compare it with the luminous efficacy benchmark R0 fixed in step one to infer the current light decay factor of the lamp in situ; Second, based on the light decay factor, the target illuminance value, and the time-division power reduction boundary value given in step three, to determine the compensation release amount within the reserved dimming upper limit and generate the final dimming output. The physical link is a clearly deducible chain: the self-illuminance contribution component is proportional to the product of road surface reflectivity, installation geometry factor, LED luminous efficacy, and dimming setting value. Therefore, dividing the self-illuminance contribution component by the dimming setting value and then by the power yields the current actual luminous efficacy, which is proportional to the product of road surface reflectivity, installation geometry factor, and LED luminous efficacy. Under the premise that the installation geometry factor remains unchanged at the same lamp position, the ratio of the current actual luminous efficacy to the luminous efficacy benchmark is equal to the product of the ratio of road surface reflectivity and the ratio of LED luminous efficacy. The LED luminous efficacy ratio is precisely the light decay factor sought.

[0049] The key challenge here lies in removing the road surface reflectance ratio from the aforementioned ratios. Otherwise, the darkening of the road surface during rainy weather would be misinterpreted as lamp light decay, triggering unnecessary compensation, while dust accumulation would slowly lower the readings and be mistakenly interpreted as aging. This step utilizes the fundamental difference in the time scale between the two types of disturbances to achieve this separation: LED light decay is a monotonous, irreversible, slowly changing process on a monthly or yearly scale, while road surface reflectance changes are a rapidly changing, reversible process on a daily scale, fluctuating with rainfall and dust accumulation. The time constants of the two differ by at least two orders of magnitude. Accordingly, this step extracts the monotonically slowly changing component within a long time window and the fluctuating component within a short time window from the historical sequence of the current actual luminous efficacy. The monotonically slowly changing component is attributed to lamp light decay and used to update the light decay factor, while the fluctuating component is attributed to changes in road surface reflectance and does not participate in the light decay factor update. This time constant separation criterion is the fundamental difference between this step and all other methods that rely on cumulative duration to look up preset light decay curves: it does not look up tables or rely on group averages, but directly solves the true light decay in situ from the lamp's own observation sequence based on the frequency band differences between the two types of physical processes.

[0050] Furthermore, to avoid the lack of objective criteria for determining whether short-term fluctuations should be accepted, this step constructs two complementary indices to perform confidence gating on each decline in luminous efficacy. The first is the luminous efficacy decay trend index: first, the downward slope of the current actual luminous efficacy sequence segment by segment within a long time window is taken; then, the slope amplitude is multiplied by the proportion of durations where the slope is negative and the sign is persistent, resulting in a trend index. This index characterizes the intensity of a continuous monotonic decline, corresponding to the monotonic and irreversible physical characteristic of LED luminous efficacy decay. The second is the reflection disturbance fluctuation index: first, the residual amplitude of the current actual luminous efficacy sequence after removing the trend within a short time window is taken; then, the degree of synchronization between this residual and rainfall, temperature and humidity, or daily cycles is normalized, resulting in a fluctuation index. This index characterizes the intensity of rapidly changing reversible disturbances, corresponding to the daily cyclical nature of road surface reflectivity with weather and dust accumulation. The degree of synchronization is characterized by the correlation coefficient between the detrended residual sequence and the rainfall indicator sequence or daily periodic template: the higher the correlation coefficient, the more attributable the fluctuation to the meteorological and dust accumulation variations in road surface reflectivity, and the larger the reflection disturbance fluctuation index. The two indices are not simply weighted sums, but are gated based on their relative strength: only when the light efficiency decay trend index is dominant relative to the reflection disturbance fluctuation index is the current downtrend considered a true light decay, and the light decay factor is updated accordingly; when the reflection disturbance fluctuation index is not lower than the light efficiency decay trend index, it is considered a road surface reflection disturbance, and the light decay factor update is frozen. Dominance here means the ratio of the light efficiency decay trend index to the reflection disturbance fluctuation index is greater than one; when this ratio is greater than one, it is considered a true light decay, and the light decay factor is updated; when this ratio is not greater than one, the update is frozen. This complementary gating is based on the inverse relationship between the monotonic persistence and rapid fluctuation of the two physical processes, rather than any artificially set amplitude threshold.

[0051] The reflection disturbance fluctuation index is obtained by multiplying two factors: the first factor is the residual amplitude, and the second factor is the degree of synchronization. The residual amplitude is calculated as follows: a sample of the current actual light efficiency sequence within a short time window is taken. First, the trend term within that short time window is fitted and subtracted to obtain a detrended residual sequence. Then, a robust representative value of the amplitude of this detrended residual sequence (such as the median or root mean square of the absolute residual values) is taken as the residual amplitude, which is a non-negative real number representing the strength of rapid fluctuations within the short time window. The degree of synchronization is calculated as follows: the above detrended residual sequence is aligned with the rainfall indicator sequence, temperature and humidity state sequence, and daily cycle template according to the sampling time, and the normalized correlation coefficient is calculated. The largest absolute value of each normalized correlation coefficient is taken as the degree of synchronization, which is a dimensionless proportion between 0 and 1. The closer the value is to 1, the more the residual fluctuation can be explained by rainfall, temperature and humidity, or daily cycles, and the more it should be attributed to meteorological and dust accumulation changes in road surface reflectivity, rather than LED light decay. The normalized correlation coefficient is the sum of the products of two sequences aligned by sampling time, each minus its own mean, divided by the product of the standard deviation and the number of samples for each sequence, with a value between -1 and +1. The daily cycle template is a daily cycle reference sequence formed by aggregating the detrended residuals of the same lamp position at the same historical moment. The significance of the reflection disturbance fluctuation index is that it only increases significantly when there is a significant fluctuation within a short time window, and this fluctuation is highly synchronized with rainfall or the daily cycle. If the fluctuation is large but unrelated to any meteorological or periodic factors, the degree of synchronization is close to zero, and the reflection disturbance fluctuation index is very small, preventing the actual light decay from being mistakenly treated as road disturbance and causing a freeze update.

[0052] Furthermore, the specific implementation of step four includes: The current actual luminous efficacy and light decay factor are calculated in situ. The controller takes the self-illuminance contribution component E_self output from step two, divides it by the current dimming setting, and then divides it by the power P synchronously collected by the electrical metering module at the same operating point to obtain the current actual luminous efficacy R. The current actual luminous efficacy R is compared with the locally stored luminous efficacy reference value R0 to obtain the ratio between the two. After time constant separation and confidence gating in the next technical point, the component attributable to the change in the luminous efficacy of the light-emitting diode is the light decay factor. The physical meaning of the light decay factor is the ratio of the current luminous efficacy of the light-emitting diode to the initial luminous efficacy, and its value gradually decreases from 1 as the lamp ages. The effect of this technical point is: no external illuminance meter is needed, no server is needed to obtain the light decay curve, and an in-situ observation reflecting the true aging degree of the lamp can be obtained with only one sampling of the lamp's operation, and each lamp is independent and undependent.

[0053] Furthermore, the current actual luminous efficacy is obtained by dividing the lamp's current self-illuminance contribution component by the current dimming setting value, and then by the power synchronously collected by the electrical metering module at the same operating point. The self-illuminance contribution component is a positive real number in lux, derived from the step difference extraction in step two, used to provide the absolute amount of the lamp's current self-illuminance; the dimming setting value is a dimensionless ratio between zero and one, issued by the dimming execution module, used to normalize different brightness operating points; the power is a positive real number in watts, used to normalize different power operating points; the resulting current actual luminous efficacy is a positive real number, representing the illuminance contribution of the lamp's own illumination reflected back to the photosensitive device under unit dimming and unit power, serving as an operational observation that can be directly compared with the luminous efficacy benchmark.

[0054] To ensure clarity, the core physical logic link of this technical point is presented below, expressed by two formulas, while the remaining processing steps are described in words.

[0055] R = E_self / (D × P), In the formula, the current actual luminous efficacy R is a positive real number, representing the illuminance contribution of the lamp itself after reflection back to the photosensitive device under unit dimming and unit power. Its function is to serve as an operational observation that can be directly compared with the luminous efficacy benchmark. The self-illuminance contribution component E_self, in lux, is a positive real number, derived from the step difference extraction in step two, and its function is to provide the absolute amount of the lamp's current self-light. The dimming setpoint D is a dimensionless ratio ranging from zero to one, issued by the dimming execution module, and its function is to normalize different brightness operating points. The power P, in watts, is a positive real number, synchronously collected by the electrical metering module at the same operating point, and its function is to normalize different power operating points.

[0056] The value of the light decay factor is determined by the ratio of the current actual luminous efficacy to the luminous efficacy reference value. This ratio is physically equal to the product of the ratio of the current road surface reflectivity to the initial road surface reflectivity and the ratio of the current luminous efficacy of the LED to the initial luminous efficacy. The controller uses time constant separation and double exponential gating to identify and remove the term that changes with the road surface reflectivity. The remaining term, which only reflects the ratio of the luminous efficacy of the LED to the initial luminous efficacy, is used as the light decay factor. The light decay factor, a dimensionless ratio ranging from 0 to 1, characterizes the ratio of the current luminous efficacy of the LED to its initial luminous efficacy. Its function is to drive the determination of subsequent compensation release. Time constant separation and dual-exponential confidence gating are also used. The controller processes the historical sequence of the ratio of the current actual luminous efficacy R to the luminous efficacy reference value R0 within long and short time windows: the long time window captures monotonically slow variations on a monthly or yearly scale, while the short time window captures cyclical fluctuations on a daily scale. Based on this, the controller calculates the luminous efficacy decay trend index and the reflection disturbance fluctuation index: the former is obtained by multiplying the slope amplitude by the proportion of the duration of the downward slope with a negative and persistent sign within the long time window; the latter is obtained by normalizing the detrended residual amplitude within the short time window with the synchronization degree of rainfall, temperature and humidity, or daily cycles. The controller then compares the relative strengths of the two factors: when the luminous efficacy decay trend index is superior to the reflection disturbance fluctuation index, the current downlink is considered a true luminous decay, and the luminous decay factor is updated with a monotonically slowly varying component over a long time window; when the reflection disturbance fluctuation index is not lower than the luminous efficacy decay trend index, it is judged as a road surface reflection disturbance, and the luminous decay factor update is frozen. The effect of this technology is that under conditions of significant fluctuations in road surface reflectivity, such as rain or dust accumulation, the luminous decay factor is not mistakenly updated, and the compensation is not erroneously adjusted. Under the gradual trend of true aging of the lamps, the luminous decay factor is updated stably and monotonically.

[0057] Bounded compensation release and coordinated output. The controller will determine the current light decay state and whether the target illuminance value can be reached under the current time-sharing boundary based on the updated light decay factor, target illuminance value, and the time-sharing dimming power boundary value given in step three. If the requirements can be met, the time-sharing dimming power boundary value will be executed first to maintain energy saving. If the requirements cannot be met but there is still a margin in the reserved dimming upper limit, the compensation release amount will be determined within the reserved dimming upper limit. The dimming setting value that needs to be increased will be calculated back according to the illuminance gap corresponding to the light decay factor, and the release compensation will be performed based on this. Moreover, the compensation release is not unconstrained full release: First, the power of the electrical metering module is combined with the ambient temperature to estimate the junction temperature or cumulative thermal stress of the light-emitting diode - the size of the "junction temperature" is equal to the product of the ambient temperature heating resistance and power consumption - then the amplitude or step size of this compensation release amount is reduced based on the estimated junction temperature or thermal stress to suppress the positive feedback effect of increased drive current caused by release compensation, increased current causing increased junction temperature, increased junction temperature causing further reduction in luminous efficacy, and accelerated aging. The final controller coordinates the constant illuminance dimming requirement, the time-sharing dimming power boundary value, and the compensation release amount within the boundary to generate a dimming duty cycle or power upper limit correction value, which is then executed by the dimming conversion circuit or relay of the dimming execution module. The effect of this technology is to restore the road illuminance lamp by lamp according to the actual light decay of the lamp, while using junction temperature constraints to balance the restoration of illuminance and delay aging, avoiding the problem of the lamp overheating and decaying with increasing heat.

[0058] In one optional embodiment, the release of the reserved dimming upper limit can be in one stage, starting from the first upper limit and gradually releasing to another upper limit higher than the first upper limit as the light decay factor decreases. This conforms to the characteristics of gradual aging within the life cycle of the lamp. In this equivalent embodiment, junction temperature estimation can be directly measured by an independent temperature acquisition terminal instead of estimation based on power and ambient temperature. They are logically equivalent and can be implemented independently. When communication is abnormal, the controller can be degraded to a backup operation: continue the above steps of light decay back-calculation and compensation release according to the timing control strategy of the local real-time clock and cache, and re-report after communication is restored; alternatively, the light-off period or the lowest output period can be selected to have a known reference calibrated photosensitive device respond, distinguishing the dust accumulation or aging of the photosensitive device itself from the light decay of the lamp, preventing the sensor from being misjudged as lamp light decay due to dirt; the steps applied by multiple adjacent single lamp controllers can be staggered in time or use mutually orthogonal perturbation sequences to suppress the crosstalk component introduced by the adjacent lamp steps into the total ambient illuminance of the lamp, avoiding the crosstalk component being included in the illuminance contribution component of the lamp itself. These features are independent of each other and can operate independently or be decoupled and combined.

[0059] Furthermore, the light decay factor refers to the ratio of the current luminous efficacy of the LED to its initial luminous efficacy, a dimensionless ratio between 0 and 1. Its input signals include the self-illuminance contribution component output in step two, the luminous efficacy reference value solidified in step one, and the power synchronously acquired by the electrical metering module, obtained by the ambient illuminance acquisition module and the electrical metering module at the same operating point. When generating the light decay factor, the current actual luminous efficacy is first obtained by dividing the self-illuminance contribution component by the current dimming setting and the power. This is then compared to the luminous efficacy reference value to obtain the ratio. Subsequently, the historical sequence of this ratio is separated into long and short time windows and subjected to dual exponential gating. The factor is updated with a monotonically slowly varying component only when it is determined to be true light decay, and the determination result is written to local storage. When the light decay represented by the light decay factor makes the target illuminance value unattainable, the controller releases compensation within the reserved dimming upper limit; when the light decay factor update is frozen by gating, the controller maintains the original compensation unchanged.

[0060] Furthermore, under the premise that the geometric factor of the same lamp position remains unchanged, the ratio of the current actual luminous efficacy to the luminous efficacy reference is equal to the product of the ratio of the current road surface reflectivity to the initial road surface reflectivity and the ratio of the current luminous efficacy of the light-emitting diode to the initial luminous efficacy; where the ratio of the luminous efficacy of the light-emitting diode is the light decay factor to be sought, and the ratio of the road surface reflectivity is the interference term that needs to be removed.

[0061] Furthermore, the luminous efficacy decay trend index is a measure of the sustained monotonically declining intensity of the current actual luminous efficacy sequence within a long time window. Its input is the historical sequence of the current actual luminous efficacy, generated by the controller segmentally calculating the downward slope of the sequence and multiplying the percentage of durations with negative slopes and sustained signs by the slope amplitude. The reflection disturbance fluctuation index is a measure of the intensity of rapid and reversible fluctuations in the sequence within a short time window. It is generated by the controller normalizing the residual amplitude after removing the trend from the sequence, and the correlation coefficient between the detrended residual sequence and the rainfall indicator sequence or daily cycle template. The daily cycle template is a daily cycle reference sequence formed from the detrended residuals of the same lamp position at the same historical moment. The two are not weighted together but are gated by relative strength: when the luminous efficacy decay trend index is superior to the reflection disturbance fluctuation index, the controller allows updating the luminous efficacy decay factor with a monotonically slowly varying component and releasing compensation accordingly; when the reflection disturbance fluctuation index is not lower than the luminous efficacy decay trend index, the controller prohibits updating the luminous efficacy decay factor and freezes the compensation release action. The reason why this gating method is fundamentally physically incompatible with the traditional method of looking up the light decay curve based on cumulative duration is that the light decay of LEDs is a monotonous and irreversible process on a monthly or yearly scale, while the change in road surface reflectivity is a reciprocating and reversible process on a daily scale. The time constants of the two differ by at least two orders of magnitude. Therefore, they can only be distinguished based on the physical criterion of the relative strength of monotonous continuity and rapid fluctuation. On the other hand, the lookup table method discards time scale information and cannot make a judgment on whether to accept or ignore the same downward illuminance. The two methods cannot replace each other.

[0062] In an optional equivalent implementation, the relative strength of the two indices can be determined by comparing the energy of the two channels after separating the current actual luminous efficacy sequence into low-pass and high-pass filters. This is logically equivalent to the segmented slope method and can be implemented independently. When encountering an abnormal operating condition where there are insufficient samples of the current actual luminous efficacy sequence, the single-lamp controller enters a degradation protection path: pausing the update of the luminous decay factor, using the most recent effective luminous decay factor to maintain compensation, and resuming updates after sufficient sample accumulation.

[0063] The explanation explains that for a known amplitude step, the upper bound of the amplitude is taken as the dimming setting value increment that is higher than the noise envelope of the ambient illuminance acquisition module but lower than the upper bound of the brightness jump perceptible to the human eye. During the installation phase of this lamp, the relative change in road illuminance before and after the step and the light output stabilization time are recorded and solidified into a brightness jump variable calibration table. The interval between two adjacent steps is not less than the light output stabilization time required for the dimming execution module to complete the step response and enter a steady state, avoiding transient readings. There is a hysteresis dead zone in the real ambient light component driving the lamp-on and lamp-off judgments. That is, when the light level corresponding to the lamp-on judgment is lower than the light level corresponding to the lamp-off judgment, there is a hysteresis band between the two. When the real ambient light component fluctuates within this hysteresis band, there is no start / stop or switching function, avoiding frequent lamp-on and lamp-off during dusk or when clouds pass by. The update dead zone of the light decay factor is determined by the quantization resolution of the ambient illuminance acquisition module, the median absolute deviation of the sampling noise, and the power resolution of the electrical metering. If the change in the ratio of the current actual luminous efficacy to the luminous efficacy reference value relative to the last update value does not exceed the minimum resolvable change, the light decay factor is not updated to avoid compensation jitter caused by reading noise.

[0064] In the hardware and software acquisition process, the total ambient illuminance value is obtained by converting the data from a phototransistor or photosensor circuit to an analog-to-digital converter. The continuously sampled values ​​covering a stable light output range are sorted by magnitude and the result is the median or the median value within the stable range. The dimming setting value is sent from the controller to the LED driver via a pulse-width modulation interface or a digital dimming interface, and can be read back by the controller. The power is read by the metering chip via a serial interface. The self-illuminance contribution component, the current actual luminous efficacy, the light decay factor, and the compensation release are all intermediate values ​​generated internally by the controller and written to local non-volatile memory. The final dimming output is converted by the dimming conversion circuit into a dimming signal that the driver can receive, or by a relay switching on / off mechanism as the execution endpoint. None of these methods rely on external preset curves or manual illuminance meters; for newly installed lamps, this manual is sufficient to independently complete the acquisition, calculation, and execution.

[0065] In one possible embodiment, when abnormal conditions such as the inability to apply a dimming step, a low number of current actual luminous efficacy sequence samples, or communication interruption occur, the single-lamp controller enters a degraded backup mode: it stops updating its own light decay factor, maintains operation using the most recent effective light decay factor and the most recent effective self-illuminance contribution response, and continues constant illuminance dimming and compensation release according to the timing control strategy of the local real-time clock and cache. After the operating conditions are restored, it recalibrates, resamples, and starts updating again. The self-light separation feature, light decay inverse calculation feature, and junction temperature constraint compensation feature are logically independent and can be used individually or decoupled into a non-bound combination. The other two usage modes of the single-lamp controller (i.e., only implementing self-light separation and constant illuminance decontamination without implementing light decay inverse calculation) and only implementing light decay inverse calculation and bounded compensation, while the ambient light obtained by other methods are all solutions selected by the present invention.

[0066] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for coordinated control of constant illuminance and time-sharing dimming in smart streetlights, applied to a single-lamp controller with an ambient illuminance acquisition module, a real-time clock, an electrical metering module, a dimming execution module, and a remote communication module, characterized in that: include: Store the luminous efficacy reference value of this lamp. The luminous efficacy reference value characterizes the relationship between the lamp's own illuminance contribution in the initial state and the dimming output and power consumption. The dimming execution module applies a known amplitude step to the dimming setpoint, and collects the steady-state total ambient illuminance before and after the step. The difference between the two steady-state total values ​​is divided by the difference in the dimming setpoint to obtain the unit dimming self-illuminance contribution response. The unit dimming self-illuminance contribution response is then multiplied by the current dimming setpoint to obtain the current self-illuminance contribution component of the lamp. The self-illuminance contribution component is subtracted from the total ambient illuminance to obtain the true ambient light component. The known amplitude step is formed by an active test step or a brightness switch triggered by the current time-sharing dimming strategy. When the known amplitude step is formed by a brightness switch triggered by the current time-sharing dimming strategy, the single lamp controller collects the steady-state total ambient illuminance before and after the brightness switch occurs to determine the self-illuminance contribution component. The operating constraints for the current time period are determined by the real-time clock and the time control strategy. When the current time period is in the self-sensing constant illuminance mode, the actual ambient light component is used as the input for the calculation of light-on judgment, light-off judgment, and constant illuminance dimming requirements. Divide its own illuminance contribution component by the product of the current dimming setting value and the power synchronously collected by the electrical metering module to obtain the current actual luminous efficacy. Then compare the current actual luminous efficacy with the luminous efficacy reference value to determine the current light decay factor of this lamp. Based on the light decay factor, the target illuminance value, and the current time period operation constraints determined by the real-time clock and timing strategy, the compensation release amount is determined within the reserved dimming upper limit, and a dimming output is generated and executed by the dimming execution module; wherein, the self-illuminance contribution component is used to obtain the real ambient light component and determine the light decay factor. The light decay factor is obtained as follows: a ratio sequence is constructed by repeatedly obtaining the ratio of the current actual luminous efficacy to the luminous efficacy reference quantity. The monotonically slowly varying component in the long time window and the fluctuating component in the short time window are extracted respectively. The monotonically slowly varying component is associated with the luminous efficacy decay of the lamps, and the fluctuating component is associated with the change in road surface reflectivity. The light decay factor is updated only with the monotonically slowly varying component. The long time window is an observation window for capturing the monotonically slowly varying current actual luminous efficacy ratio sequence on a monthly or yearly scale, and the short time window is an observation window for capturing the current actual luminous efficacy ratio sequence that fluctuates on a daily scale. The luminous efficacy attenuation trend index is determined by the product of the percentage of duration with a negative downward slope and a sustained sign in the ratio sequence within a long time window and the slope amplitude. A sample of the current actual luminous efficacy sequence within a short time window is taken, fitted, and the trend term within that short time window is subtracted to obtain a detrended residual sequence. The reflection disturbance fluctuation index is determined by the magnitude of the detrended residual in the ratio sequence within the short time window, and the absolute value of the normalized correlation coefficient between the detrended residual sequence and the rainfall indicator sequence, temperature and humidity state sequence, or daily periodic template. The daily periodic template is formed from the detrended residuals of the same lamp position at the same historical moment, and the normalized correlation coefficient is calculated from the detrended residual sequence aligned by sampling time and the rainfall indicator sequence, temperature and humidity state sequence, or daily periodic template. The luminous efficacy attenuation trend index is updated when the ratio of the luminous efficacy attenuation trend index to the reflection disturbance fluctuation index is greater than 1; otherwise, the luminous efficacy attenuation factor is frozen and updated. The compensation release amount is obtained as follows: based on the power collected by the electrical metering module and the ambient temperature obtained from the meteorological data sent by the drive power supply casing temperature acquisition terminal or remote server, the junction temperature or cumulative thermal stress of the light-emitting diode is estimated, and the magnitude or step size of the compensation release amount is based on the estimated junction temperature or thermal stress contraction. The release of the reserved dimming upper limit is a graded release, initially locked at the first upper limit, and gradually released to the second upper limit higher than the first upper limit as the light decay factor decreases. The first upper limit is the power upper limit corresponding to the initial dimming output locked when the target illuminance value is met during the initialization stage, and the second upper limit is the highest allowable power upper limit formed by the first upper limit plus the reserved dimming upper limit. The second upper limit does not exceed the rated output power of the lamp and the allowable power boundary of the junction temperature or cumulative thermal stress.

2. The method for coordinated control of constant illuminance and time-sharing dimming of smart streetlights according to claim 1, characterized in that, The method for obtaining the luminous efficacy reference quantity is as follows: During the installation or initialization phase, the total steady-state ambient illuminance and synchronous power are collected at at least two different dimming settings. The difference between the total illuminance values ​​of adjacent dimming settings is divided by the difference between adjacent dimming settings to obtain the illuminance contribution response per unit dimming. Then, the illuminance contribution response per unit dimming is normalized to 1 by the average value or steady-state representative value of adjacent sampled power to obtain the illuminance contribution per unit dimming and per unit power, which is stored locally as the luminous efficacy reference quantity.

3. The method for coordinated control of constant illuminance and time-sharing dimming of smart streetlights according to claim 1, characterized in that, The method for obtaining the self-illuminance contribution component is as follows: the total steady-state ambient illuminance value before and after the step is annotated with the acquisition time using a real-time clock, and the ambient light natural gradual change envelope is obtained by extrapolating the ambient light natural change rate within the adjacent short window outside the step window; when the ambient light gradual change between two acquisition times exceeds the ambient light natural gradual change envelope, the corresponding sample is discarded and then differential is performed; when the ambient illuminance acquisition module supports synchronous sampling with the dimming phase, the known amplitude step is formed by periodic micro-step disturbance, and the single lamp controller extracts the steady-state representative value of the synchronously acquired illuminance sequence according to the high phase segment and low phase segment of the periodic micro-step disturbance, respectively. The difference between the steady-state representative value of the high phase segment and the steady-state representative value of the low phase segment is divided by the difference between the dimming setting value of the high phase segment and the dimming setting value of the low phase segment to obtain the self-illuminance contribution response per unit dimming, and the self-illuminance contribution response per unit dimming is multiplied by the current dimming setting value to obtain the self-illuminance contribution component.

4. The method for coordinated control of constant illuminance and time-sharing dimming of smart streetlights according to claim 1, characterized in that, The remote communication module reports a running data frame to the remote server. The running data frame includes the lamp identifier, cumulative running time, light decay factor, current actual luminous efficacy, and compensation release amount. The cumulative running time is used as a maintenance reference field, and the light decay factor and current actual luminous efficacy are used as the main fields for lamp-by-lamp aging sorting and anomaly warning. The module also receives the time control strategy frame and target illuminance value sent by the server.

5. A single-lamp controller for performing the method as described in claim 1, characterized in that, include: The system includes an ambient illuminance acquisition module, a real-time clock, an electrical metering module, a dimming execution module, a remote communication module, and a processor. The processor is configured as follows: Store the luminous efficacy reference value of this lamp. The luminous efficacy reference value characterizes the relationship between the lamp's own illuminance contribution in the initial state and the dimming output and power consumption. The dimming execution module applies a step of known amplitude to the dimming setpoint. The ambient illuminance acquisition module collects the steady-state total ambient illuminance before and after the step. The difference between the two steady-state total values ​​and the difference between the dimming setpoint are used to determine the current self-illuminance contribution component of the lamp. The self-illuminance contribution component is then subtracted from the total ambient illuminance to obtain the true ambient light component. When the strategy issued by the real-time clock and remote communication module determines that the current period is in the self-sensing constant illuminance mode, the real ambient light component is used as the input for the calculation of the light-on judgment, light-off judgment, and constant illuminance dimming requirements. The actual luminous efficacy is obtained by dividing the lamp's own illuminance contribution by the product of the current dimming setting and the power synchronously collected by the electrical metering module. The actual luminous efficacy is then compared with the luminous efficacy reference value to determine the lamp's current light decay factor. Based on the light decay factor, the target illuminance value, and the current time period operation constraints determined by the real-time clock and timing strategy, the compensation release amount is determined within the reserved dimming upper limit, and dimming output is generated and executed by the dimming execution module; among them, the self-illuminance contribution component is used to obtain the real ambient light component and determine the light decay factor.

6. A smart street light constant illuminance and time-sharing dimming coordinated control system, used to implement the method as described in claim 1, characterized in that, It includes the single-lamp controller as described in claim 5 and a server connected to each single-lamp controller via remote communication; each single-lamp controller reports its light decay factor and current actual luminous efficacy to the server, and the server generates a lamp-by-lamp aging sort based on multiple light decay factors and sends a time control strategy and target illuminance value to each single-lamp controller.

Images