Street illumination intelligent regulation and control method based on illumination and people flow parameters

Abstract

The invention relates to the technical field of illumination control, in particular to an intelligent street illumination regulation and control method based on illumination and people flow parameters, which comprises the following steps: firstly, carrying out space-time alignment and weighted fusion on multi-source sensing data to generate fusion regulation and control parameters reflecting illumination demand urgency; then, in combination with real-time working state data of the street lamp unit, a dynamic dimming strategy giving consideration to the illumination requirement and the equipment operation health degree is generated by utilizing a self-adaptive decision-making model; on the basis, a dimming instruction is issued through a reliable communication network, execution feedback is obtained, and a regulation and control closed loop is formed; meanwhile, a prospective regulation and control plan is triggered based on a people flow time sequence prediction result, and pre-distribution and smooth transition of lighting resources are completed before people flow abnormal gathering occurs. According to the invention, the safety, comfort and energy utilization efficiency of street lighting can be effectively improved, and the service life of street lamp equipment is prolonged.

Description

Technical Field

[0001] This invention relates to the field of lighting control technology, and in particular to a method for intelligent control of street lighting based on illumination and pedestrian flow parameters. Background Technology

[0002] With the continuous advancement of urbanization, urban roads and public streets, as vital infrastructure for residents' travel and urban operations, have seen their lighting systems become an essential component in ensuring nighttime traffic safety, public order, and the city's image. In recent years, with increasingly stringent energy conservation and emission reduction requirements and the deepening development of smart city construction, street lighting systems have gradually evolved from traditional timed on / off control towards intelligent and refined systems. They are beginning to incorporate technologies such as ambient light sensing, remote centralized control, and networked management, aiming to reduce energy consumption and improve operational efficiency while meeting lighting safety requirements.

[0003] Existing intelligent street lighting control technologies still have significant shortcomings in practical applications. On the one hand, most systems adjust brightness primarily based on fixed time periods or single ambient light thresholds, making it difficult to accurately reflect the actual lighting needs of different areas at different times due to changes in pedestrian activity. This can easily lead to unreasonable allocation of lighting resources. On the other hand, existing dimming strategies often ignore the real-time operating status and health constraints of streetlight units, continuing to output higher brightness even under high load or aging conditions, which can accelerate equipment wear and tear. Furthermore, for non-periodic pedestrian flow scenarios such as the dispersal of large events or sudden gatherings at transportation hubs, existing technologies generally lack effective forward-looking prediction and early control mechanisms. They typically only respond passively after the gathering has occurred, resulting in delayed lighting adjustments that fail to meet actual needs in terms of safety and comfort. Summary of the Invention

[0004] This invention provides a smart street lighting control method based on illumination and pedestrian flow parameters, addressing the problems of existing street lighting systems such as slow response to environmental changes, crude dimming strategies, and insufficient consideration of equipment health.

[0005] A method for intelligent street lighting control based on illumination and pedestrian flow parameters includes the following steps: S1: Real-time acquisition of ambient light intensity data and dynamic pedestrian density data of the target street area, and spatiotemporal alignment and fusion processing of the two types of data to generate fusion control parameters that reflect the urgency of current lighting needs; S2: The fusion control parameters and the real-time working status data of each street light unit in the target street area are input into a preset adaptive decision model. The adaptive decision model outputs a dynamic dimming strategy that takes into account both the current lighting needs and the health of the equipment operation. The dynamic dimming strategy includes brightness adjustment instructions for each street light unit. S3: According to the dynamic dimming strategy, send a brightness adjustment command to the corresponding street light unit to control its brightness output; S4: Based on the dynamic pedestrian density data, perform short-term trend prediction. If it is predicted that pedestrian flow will unexpectedly gather in a specific sub-area, trigger the forward-looking control plan and adjust the dynamic dimming strategy of the streetlights in the relevant sub-area and upstream of the path in advance.

[0006] Optionally, S1 includes: S11: Real-time collection of raw ambient light intensity data through a network of light sensors deployed in the target street area, and real-time collection of raw dynamic pedestrian density data through pedestrian flow monitoring devices deployed in the target street area; S12: Perform spatiotemporal alignment processing on the original ambient light intensity data and the original dynamic pedestrian density data, map the data collected at the same time to a unified geographic information grid, and generate aligned ambient light intensity data and aligned dynamic pedestrian density data corresponding to each geographic grid. S13: Determine the real-time weighting coefficient of the aligned dynamic pedestrian density data in the fusion calculation according to the current preset time period judgment rule, wherein the preset time period judgment rule stipulates that the real-time weighting coefficient is assigned a higher value when the nighttime period or when the aligned ambient light intensity data is lower than a preset threshold. S14: Based on the real-time weighting coefficient, perform weighted calculations on the aligned ambient light intensity data and the aligned dynamic pedestrian density data within the same geographic grid to generate fusion control parameters that characterize the urgency of the current lighting demand for the grid.

[0007] Optionally, the spatiotemporal alignment process specifically includes: establishing a unified geographic information grid covering the target street area based on a geographic information system map; simultaneously, ensuring clock synchronization of all sensors and monitoring devices through a time synchronization server deployed in the network, matching the original ambient light intensity data and the original dynamic pedestrian density data to the corresponding geographic information grid according to their collection timestamps and physical location coordinates, thereby forming the aligned ambient light intensity data and aligned dynamic pedestrian density data.

[0008] Optionally, the preset time period judgment rule is as follows: based on the latitude and longitude of the target street area and the current date, the sunrise and sunset times are dynamically calculated, the nighttime period defined by the sunset time to the sunrise time is used as the first judgment condition, and the aligned ambient light intensity data being lower than the first preset light threshold is used as the second judgment condition; when either the first judgment condition or the second judgment condition is met, the real-time weight coefficient corresponding to the aligned dynamic pedestrian density data is increased.

[0009] Optionally, S2 includes: S21: The fusion control parameters generated in S1 are formatted and assembled with the real-time working status data obtained from the monitoring terminals of each street light unit in the target street area to form a complete decision input vector containing timestamps, spatial location identifiers, fusion control parameter values ​​and multiple real-time working status data values. S22: Input the complete decision input vector into the preset adaptive decision model. The adaptive decision model calculates and analyzes the input vector according to its internal mapping rules and optimization objectives to generate a preliminary brightness adjustment instruction for each specific street light unit. The internal mapping rules and optimization objectives ensure that the preliminary brightness adjustment instruction simultaneously responds to the urgency of lighting demand represented by the fusion control parameters and the equipment health constraints reflected by the real-time working status data. S23: Perform regional coordination verification and smoothing filtering on all the preliminary brightness adjustment commands output by the adaptive decision model to eliminate conflicts between commands and ensure the spatiotemporal continuity of brightness changes, and finally generate the dynamic dimming strategy that can be directly issued to each street light unit for execution and contains all the brightness adjustment commands.

[0010] Optionally, the real-time operating status data includes the current operating temperature, continuous operating time, current operating brightness percentage, and historical fault code records for each street light unit.

[0011] Optionally, S3 includes: S31: Analyze the dynamic dimming strategy generated by S2, extract the unique identifier for each target street light unit and its corresponding specific brightness adjustment instruction, and generate a series of executable dimming instruction sets containing target address and brightness setting value according to the preset instruction encoding rules. S32: Through a wireless or wired communication network deployed in the target street area, each instruction in the executable dimming instruction set is distributed in real time and reliably to the corresponding street light unit controller according to the target address it contains. S33: After receiving and verifying the executable dimming instruction set, each street light unit controller drives its internal dimming circuit to perform a brightness adjustment operation, and uploads the actual brightness value after execution as instruction execution feedback data back to the central control system, thereby completing the closed loop of the execution of the dynamic dimming strategy.

[0012] Optionally, the unique identifier is the MAC address of the street light unit, a unique IP address in the communication network, or a pre-assigned physical location code; the preset instruction encoding rule adopts JSON format or binary protocol format to encapsulate the unique identifier and the brightness adjustment instruction.

[0013] Optionally, S4 includes: S41: Based on the dynamic pedestrian density data, the time-series prediction algorithm is used to analyze the spatial movement trajectory, speed and density change trend of pedestrian flow, identify specific sub-regions where abnormal pedestrian flow will occur within a preset future time window, and determine whether the convergence is an unexpected aggregation that exceeds the normal periodic pattern. S42: Based on the predicted scale, arrival time and duration of the expected external gathering, call the preset plan library to generate a matching forward-looking control plan. The plan clearly specifies the brightness adjustment target and adjustment timetable for the streetlights involved in the specific sub-area and its upstream path of the crowd from the current moment to the end of the gathering event. S43: The forward-looking control plan is integrated in real time with the dynamic dimming strategy that is being generated or has been generated in S2 to form an enhanced dynamic dimming strategy with forward-looking brightness adjustment instructions superimposed, and submitted to S3 for execution, so as to complete the pre-allocation of lighting resources and the gradual transition of brightness before the unexpected gathering actually occurs.

[0014] Optionally, the specific rule for determining whether it is an expected external gathering is as follows: when the predicted pedestrian density of the specific sub-region exceeds its historical benchmark value for the same period by a first preset threshold, and the rate of increase of pedestrian density exceeds a second preset threshold, it is determined to be an expected external gathering.

[0015] The beneficial effects of this invention are: This invention generates fusion control parameters that accurately reflect the urgency of lighting needs at different times across different geographic information grids by spatiotemporally aligning, dynamically adjusting weights, and fusing ambient light intensity data with dynamic pedestrian density data. This allows street lighting control to move beyond relying solely on fixed time periods or single light thresholds, comprehensively considering the synergistic effects of natural lighting conditions and pedestrian activity intensity. Furthermore, based on an adaptive decision-making model using these fusion control parameters, differentiated and refined control of the brightness of each street light unit is achieved, thereby avoiding a crude lighting pattern that is either too bright or too dim overall, and improving the targeted and scientific nature of lighting resource allocation.

[0016] This invention incorporates real-time operating status data of the street light unit, such as its current operating temperature, continuous operating time, current brightness percentage, and historical fault code records, into the decision-making process. Furthermore, it explicitly incorporates equipment health constraints into the optimization objective of the adaptive decision model through an equipment aging acceleration factor lookup table. This allows the generated dynamic dimming strategy to proactively avoid excessive brightness output under high temperature, high load, and long-term operating conditions. Combined with regional coordination verification and time-series smoothing of brightness adjustment commands, it further reduces the stress impact on dimming circuits and light source devices caused by frequent and drastic brightness changes. This reduces equipment wear at the system level, extends the service life of the street light unit, and lowers maintenance and replacement costs.

[0017] This invention, through time-series prediction and anticipated external gathering identification of dynamic pedestrian density data, can generate proactive control plans in advance before non-periodic pedestrian flow scenarios such as the dispersal of large events, sudden gatherings at transportation hubs, or crowds gathering for localized incidents. These plans are then integrated in real-time with existing dynamic dimming strategies, enabling pre-adjustment and gradual transition of streetlight brightness in specific sub-areas and along upstream paths of pedestrians before actual gatherings occur. This mechanism avoids the response lag problem caused by traditional lighting systems passively increasing brightness only after a surge in pedestrian flow, significantly improving the safety, accessibility, and pedestrian experience of nighttime public spaces, while ensuring the continuity and stability of the lighting adjustment process. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the S2 process in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0022] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0023] like Figures 1-2 As shown, a method for intelligent street lighting control based on illumination and pedestrian flow parameters includes the following steps: S1: Real-time acquisition of ambient light intensity data and dynamic pedestrian density data for the target street area, followed by spatiotemporal alignment and fusion processing of the two types of data to generate fusion control parameters reflecting the urgency of current lighting needs. The specific steps are as follows: S11: Within the target street area, a light sensor network is constructed according to a pre-defined spatial deployment plan. The light sensor network is deployed along the street direction, street light locations, and key intersections to cover the entire lighting service area of ​​the target street area. Each light sensor continuously collects raw ambient light intensity data at its corresponding installation location at a fixed acquisition frequency and uploads the collected raw ambient light intensity data to the data processing unit via wired or wireless communication.

[0024] Simultaneously, pedestrian monitoring equipment is deployed within the target street area. This equipment includes at least one of the following: high-definition cameras, infrared thermal imaging sensors, millimeter-wave radar, and mobile communication network signaling data acquisition units. Each monitoring device, according to its own operating mechanism, continuously collects information on the number of human targets, their movement trajectories, or communication connection density within its coverage area, and generates raw dynamic pedestrian density data for the corresponding time slice. This raw dynamic pedestrian density data is also transmitted to the data processing unit via the communication network for subsequent processing.

[0025] In this step, the acquisition frequency of the original ambient light intensity data is set to be higher than that of the original dynamic pedestrian density data to ensure a rapid response capability to changes in ambient light.

[0026] S12: Based on the geographic information system map, a unified geographic information grid covering the target street area is established in the data processing unit. The unified geographic information grid divides the target street area into multiple geographic information grid units with clear spatial boundaries, and each geographic information grid corresponds to a unique spatial index identifier.

[0027] By deploying a time synchronization server in the network, the light sensor network and the pedestrian flow monitoring equipment are synchronized with a unified clock, enabling all types of equipment to use the same time reference when collecting raw ambient light intensity data and raw dynamic pedestrian flow density data. The data processing unit maps the raw ambient light intensity data and raw dynamic pedestrian flow density data to the corresponding geographic information grids based on the collection timestamp information and physical location coordinate information carried in each data record.

[0028] To address the issue that the original ambient light intensity data collection frequency is higher than the original dynamic pedestrian density data collection frequency, after spatial mapping, downsampling is performed on the original ambient light intensity data to ensure that the data point intervals in the downsampling data are consistent with the original dynamic pedestrian density data in the time dimension. This results in a one-to-one correspondence between aligned ambient light intensity data and aligned dynamic pedestrian density data within each geographic information grid and at each unified time slice.

[0029] S13: In the data processing unit, the sunrise and sunset times of the target street area are dynamically calculated based on the latitude and longitude information and the current date information, and the period from sunset to sunrise of the next day is determined as the nighttime period. A preset time period judgment rule is constructed, using whether it is during the nighttime period as the first judgment condition and whether the aligned ambient light intensity data is lower than a first preset light threshold as the second judgment condition.

[0030] Within each time slice, the data processing unit assigns a real-time weighting coefficient to the aligned dynamic pedestrian density data in the fusion calculation based on whether the current time meets the first judgment condition and whether the aligned ambient light intensity data corresponding to the current geographic information grid meets the second judgment condition.

[0031] If neither the first nor the second criterion is met, the real-time weighting coefficient will be set to the base value. 0; When only the first judgment condition or only the second judgment condition is met, the real-time weighting coefficient is set to... 1; When both the first and second judgment conditions are met simultaneously, the real-time weighting coefficient is set to... 2, of which, 2 greater than 1, 1 greater than 0.

[0032] The above method enables real-time dynamic changes in weight coefficients over time and with environmental conditions.

[0033] S14: After the real-time weight coefficients are determined, the data processing unit performs normalization processing on the aligned ambient light intensity data and the aligned dynamic pedestrian density data for each geographic information grid within the corresponding time slice. The normalization processing maps data of different dimensions to a unified numerical range through a normalization function.

[0034] Subsequently, the data processing unit calculates the weighted formula. ; in, Indicates at time Next, the The fusion control parameters corresponding to each geographic information grid are used to characterize the urgency of the lighting demand for that geographic information grid at the current moment. This indicates the geographic information grid index number obtained by dividing the target street area according to the geographic information system map, used to distinguish geographic information grids of different spatial locations. This indicates the current time under a unified time base, and this time is consistent with the timestamps of the aligned ambient light intensity data and the aligned dynamic pedestrian density data. Indicates at time Next, the Aligned ambient light intensity data corresponding to each geographic information grid is obtained by spatiotemporal alignment and necessary downsampling processing of the original ambient light intensity data. Indicates at time Next, the Aligned dynamic pedestrian density data corresponding to each geographic information grid is obtained by spatiotemporal alignment of the original dynamic pedestrian density data. Indicates at time The real-time weighting coefficient is used to reflect the weight ratio of the aligned dynamic pedestrian density data in the fusion calculation. Its value is dynamically determined according to the preset time period judgment rules and satisfies the following: ; Weighted fusion calculations are performed on normalized ambient light intensity data and normalized dynamic pedestrian density data within the same geographic information grid to obtain the corresponding data for that geographic information grid. Timing fusion control parameters The fusion control parameters are used to characterize the urgency of the current lighting needs of the geographic information grid and serve as one of the input data for subsequent adaptive decision-making models.

[0035] S2: The integrated control parameters and real-time operating status data of each street light unit in the target street area are input into the preset adaptive decision model. The adaptive decision model outputs a dynamic dimming strategy that takes into account both current lighting needs and equipment operating health. The dynamic dimming strategy includes brightness adjustment instructions for each street light unit. The specific steps are as follows: S21: In the data processing platform, real-time operating status data of each street light unit is acquired through monitoring terminals deployed in each street light unit within the target street area. This real-time operating status data includes the current operating temperature, continuous operating time, current brightness percentage, and historical fault code records for each street light unit. This data is collected by the status monitoring module within each street light unit and then transmitted to the data processing platform according to a unified communication protocol.

[0036] Meanwhile, ambient auxiliary temperature data is collected from the ambient temperature sensor deployed at the location of the street light unit, and this ambient auxiliary temperature data is bound to the current operating temperature for subsequent calculation of the street light unit's own temperature rise data.

[0037] Temperature rise data is obtained by numerically differentiating the current operating temperature with the ambient auxiliary temperature data, and is used as part of the real-time operating status data in subsequent decision-making.

[0038] The data processing platform formats and assembles the fusion control parameters generated in S1 with the spatial location identifiers of the corresponding streetlight units and the aforementioned real-time operating status data. The formatting and assembly process follows a predefined data structure order, sequentially encapsulating the timestamp, spatial location identifier, fusion control parameter values, current operating temperature, ambient auxiliary temperature data, temperature rise data, continuous operating time, current operating brightness percentage, and historical fault code records to form a complete decision input vector with a fixed structure and complete fields. Each complete decision input vector uniquely corresponds to the operating and demand status of a streetlight unit at a specific timestamp.

[0039] S22: After constructing the complete decision input vector, the data processing platform inputs the complete decision input vector into the pre-set adaptive decision model. The adaptive decision model is a neural network model trained based on a deep reinforcement learning algorithm. Its input layer structure corresponds one-to-one with the field structure of the complete decision input vector, ensuring that each fused control parameter and real-time working status data can be effectively perceived by the model.

[0040] During the training phase, the adaptive decision model learns its internal mapping rules and optimization objectives by minimizing the comprehensive loss function. In actual operation, the model performs forward inference calculations on the complete decision input vector based on the learned internal mapping rules, and outputs preliminary brightness adjustment instructions for the corresponding street light unit.

[0041] The comprehensive loss function includes three types of constraints: One type is a lighting satisfaction penalty term that is negatively correlated with the fusion control parameters, used to constrain the output of excessively low brightness adjustment results when the lighting demand is high. One type is the equipment loss penalty term, which is positively correlated with the parameters representing equipment load in real-time operating status data, and is used to limit excessive brightness output under high equipment load conditions; Another type is a smoothness penalty term related to the brightness adjustment range, which is used to suppress drastic changes in brightness adjustment commands.

[0042] In the calculation of the equipment wear penalty, the data processing platform queries a pre-set equipment aging acceleration factor lookup table based on the current operating temperature and the continuous running time in the real-time operating status data to obtain the theoretical lifespan depreciation coefficient corresponding to the current operating status, and inputs the theoretical lifespan depreciation coefficient as the weight of the equipment wear penalty into the comprehensive loss function.

[0043] Through the above mechanism, it is ensured that the initial brightness adjustment command output by the model meets the lighting requirements while also complying with the equipment health constraints.

[0044] S23: After obtaining the initial brightness adjustment instructions output by the adaptive decision model for all street light units, the data processing platform first performs regional coordination verification.

[0045] The regional coordination verification is performed by traversing any two adjacent or visually related street light units, calculating the brightness difference value determined by their respective initial brightness adjustment commands, and determining whether the brightness difference value exceeds the preset maximum allowable brightness step value.

[0046] When the brightness difference is detected to exceed the maximum allowable brightness step value, the initial brightness adjustment command with higher brightness is reduced and corrected according to the preset rules until the brightness difference between adjacent street light units meets the constraint condition of the maximum allowable brightness step value.

[0047] After completing the regional coordination verification, for each street light unit, the data processing platform performs smoothing filtering on the time series dimension for its corresponding preliminary brightness adjustment command.

[0048] Smoothing filtering is achieved by applying a first-order low-pass digital filter to the brightness adjustment command time series of the street light unit, so that the changes in brightness adjustment commands within continuous time slices have a smooth transition characteristic, thereby avoiding the step jump phenomenon in brightness output.

[0049] After regional coordination verification and smoothing filtering, a dynamic dimming strategy is finally formed that includes brightness adjustment commands for all street light units.

[0050] The dynamic dimming strategy is a control result that can be directly sent to each street light unit for execution, and serves as the basis for street light brightness control in subsequent steps.

[0051] S3: Based on the dynamic dimming strategy, send a brightness adjustment command to the corresponding street light unit to control its brightness output. The specific steps are as follows: S31: After receiving the dynamic dimming strategy generated by S2, the central control system parses and processes the dynamic dimming strategy. The parsing and processing reads the brightness adjustment instructions generated for each target street light unit in the dynamic dimming strategy one by one according to the preset data structure rules, and simultaneously extracts the unique identifier corresponding to each brightness adjustment instruction.

[0052] The unique identifier is the MAC address of the street light unit, the unique IP address in the communication network, or the pre-assigned physical location code, used to uniquely identify the target street light unit during communication and control.

[0053] After extracting the instruction content and unique identifier, the central control system encapsulates the instruction according to preset instruction encoding rules. The instruction encoding rules adopt JSON format or binary protocol format, combining and encoding the unique identifier field, brightness adjustment instruction field, and control field used for instruction verification to form a structured executable dimming instruction.

[0054] The central control system generates corresponding executable dimming commands sequentially for all street light units included in the dynamic dimming strategy, thereby forming an executable dimming command set containing multiple independent command units.

[0055] S32: After the executable dimming instruction set is generated, the central control system sends each executable dimming instruction in the executable dimming instruction set to the street light unit controller pointed to by its target address through a wireless or wired communication network deployed in the target street area.

[0056] The wireless or wired communication network includes one or more combinations of LoRaWAN network, NB-IoT network, 4G / 5G cellular network, and power line carrier communication network. The communication network is uniformly configured according to the deployment conditions and communication capabilities of the street light unit.

[0057] During the instruction distribution process, the central control system adopts a reliable transmission protocol with a response and retransmission mechanism to track and monitor the transmission status of each executable dimming instruction.

[0058] If no confirmation response is received from the target street light unit controller within the preset time window, the central control system will resend the corresponding executable dimming command according to the preset retransmission rules until a valid response is received.

[0059] The above methods ensure that all commands in the executable dimming command set can be accurately and reliably delivered to the corresponding street light unit controller.

[0060] S33: After receiving the executable dimming command distributed by the central control system, each street light unit controller first performs integrity and legality checks on the executable dimming command, confirming that the unique identifier contained in the command is consistent with the unique identifier configured in itself, and confirming that the brightness setting value is within the allowable working range.

[0061] After the verification is passed, the street light unit controller drives its internal dimming circuit to perform a brightness adjustment operation according to the brightness setting value in the executable dimming command.

[0062] The dimming circuit is either a pulse width modulation dimming circuit or a constant current adjustable drive circuit. For street light units using pulse width modulation dimming circuits, the street light unit controller calculates the corresponding PWM duty cycle based on the brightness setpoint and outputs a PWM control signal that matches the duty cycle to drive the light source; For street light units using a constant current adjustable drive circuit, the street light unit controller generates a corresponding analog voltage control signal based on the brightness setpoint, and the drive current adjustment module outputs the target drive current, thereby achieving the target brightness output.

[0063] After completing the brightness adjustment operation, the street light unit controller collects and generates command execution feedback data in real time. This feedback data includes at least the actual brightness value, current operating current, voltage, and controller status code. The street light unit controller then uploads this feedback data to the central control system via a communication network.

[0064] After receiving the instruction execution feedback data, the central control system compares the actual brightness value with the corresponding expected brightness set value in the dynamic dimming strategy, generates a strategy execution deviation report, and uses the strategy execution deviation report as one of the input data for optimizing the adaptive decision model, thus forming a closed-loop control process for the execution of the dynamic dimming strategy.

[0065] S4: Based on dynamic pedestrian density data, short-term trend prediction is performed. If it is predicted that pedestrian flow will unexpectedly gather in a specific sub-area, a forward-looking control plan is triggered to adjust the dynamic dimming strategy of streetlights in the relevant sub-area and upstream of the path in advance. The specific steps are as follows: S41: In the central control system, based on the dynamic pedestrian density data acquired and aligned by S1, the changes in pedestrian density of each geographic information grid within a continuous time slice are organized according to a unified time series format.

[0066] Dynamic pedestrian density data is input into a time-series prediction model based on a long short-term memory network in chronological order. For each geographic information grid, the time-series prediction model outputs predicted pedestrian density values ​​corresponding to multiple consecutive time slices within a preset future time window.

[0067] The central control system will compare the predicted pedestrian density with the historical baseline values ​​for the corresponding geographic information grid. The historical baseline values ​​are obtained by statistically calculating pedestrian density data for the same historical time period and are used to characterize the pedestrian distribution level of the geographic information grid under normal periodic conditions.

[0068] If the predicted pedestrian density in a specific sub-region exceeds its historical benchmark value for the same period and reaches a first preset threshold, and the rate of increase in pedestrian density in that sub-region exceeds a second preset threshold, then the central control system determines that the specific sub-region will experience unexpected gathering within a preset future time window.

[0069] The above methods enable the early identification and spatial location of anticipated external gatherings.

[0070] S42: After determining the expected external aggregation, the central control system extracts the predicted scale, arrival time and duration of the expected external aggregation, and inputs the above-mentioned prediction features as search conditions into the preset plan library.

[0071] The contingency plan database contains multiple template contingency plans that match different scales and types of gathering events. The types of template contingency plans include large-scale event dispersal type, sudden gathering type at transportation hub type, and local event observation type.

[0072] The central control system performs matching calculations based on the predicted characteristics and the predefined triggering conditions in each template plan, and selects the template plan with the highest matching degree as the basic template for generating the forward-looking control plan.

[0073] After selecting the basic template, the central control system interpolates and calibrates the predefined brightness adjustment schedule in the template plan based on the expected arrival time and duration of external aggregation, so that the start time, peak duration and recovery time of brightness enhancement correspond to the actual prediction process of the expected external aggregation.

[0074] Through the above-mentioned measures, a forward-looking control plan was developed, which clearly defines the target and timetable for adjusting the brightness of streetlights in specific sub-areas and their upstream pedestrian paths.

[0075] S43: After generating the forward-looking control plan, the central control system will perform real-time fusion processing of the forward-looking control plan with the dynamic dimming strategy that is being generated or has been generated in S2.

[0076] The fusion processing takes the street light unit as the smallest control object and compares the brightness settings for the same street light unit in the forward-looking control plan and the dynamic dimming strategy one by one.

[0077] For street light units that already have a brightness setting in the dynamic dimming strategy, when they are within the effective period specified in the forward-looking control plan, the central control system replaces the brightness setting value of the street light unit in the dynamic dimming strategy with a higher target brightness value specified in the forward-looking control plan that matches the expected external concentration.

[0078] Meanwhile, before and after the proactive control plan takes effect, the central control system configures a smooth transition zone for the street light unit. By gradually adjusting the brightness setpoint, the brightness change is made to transition continuously over time, avoiding abrupt changes.

[0079] After the fusion process is completed, the central control system generates an enhanced dynamic dimming strategy that superimposes forward-looking brightness adjustment instructions, and submits the enhanced dynamic dimming strategy to S3 for execution, thereby completing the pre-allocation of lighting resources and brightness gradient transition control before the expected gathering actually occurs.

[0080] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0081] 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 principle 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 intelligent control of street lighting based on illumination and pedestrian flow parameters, characterized in that, Includes the following steps: S1: Real-time acquisition of ambient light intensity data and dynamic pedestrian density data of the target street area, and spatiotemporal alignment and fusion processing of the two types of data to generate fusion control parameters that reflect the urgency of current lighting needs; S2: The fusion control parameters and the real-time working status data of each street light unit in the target street area are input into a preset adaptive decision model. The adaptive decision model outputs a dynamic dimming strategy that takes into account both the current lighting needs and the health of the equipment operation. The dynamic dimming strategy includes brightness adjustment instructions for each street light unit. S3: According to the dynamic dimming strategy, send a brightness adjustment command to the corresponding street light unit to control its brightness output; S4: Based on the dynamic pedestrian density data, perform short-term trend prediction. If it is predicted that pedestrian flow will unexpectedly gather in a specific sub-area, trigger the forward-looking control plan and adjust the dynamic dimming strategy of the streetlights in the relevant sub-area and upstream of the path in advance.

2. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 1, characterized in that, S1 includes: S11: Real-time collection of raw ambient light intensity data through a network of light sensors deployed in the target street area, and real-time collection of raw dynamic pedestrian density data through pedestrian flow monitoring devices deployed in the target street area; S12: Perform spatiotemporal alignment processing on the original ambient light intensity data and the original dynamic pedestrian density data, map the data collected at the same time to a unified geographic information grid, and generate aligned ambient light intensity data and aligned dynamic pedestrian density data corresponding to each geographic grid. S13: Determine the real-time weighting coefficient of the aligned dynamic pedestrian density data in the fusion calculation according to the current preset time period judgment rule, wherein the preset time period judgment rule stipulates that the real-time weighting coefficient is assigned a higher value when the nighttime period or when the aligned ambient light intensity data is lower than a preset threshold. S14: Based on the real-time weighting coefficient, perform weighted calculations on the aligned ambient light intensity data and the aligned dynamic pedestrian density data within the same geographic grid to generate fusion control parameters that characterize the urgency of the current lighting demand for the grid.

3. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 2, characterized in that, The spatiotemporal alignment process specifically includes: establishing a unified geographic information grid covering the target street area based on a geographic information system map; simultaneously, ensuring clock synchronization of all sensors and monitoring devices through a time synchronization server deployed in the network, matching the original ambient light intensity data and the original dynamic pedestrian density data to the corresponding geographic information grid according to their collection timestamps and physical location coordinates, thereby forming the aligned ambient light intensity data and aligned dynamic pedestrian density data.

4. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 2, characterized in that, The preset time period judgment rule is as follows: based on the latitude and longitude of the target street area and the current date, the sunrise and sunset times are dynamically calculated, the nighttime period defined from sunset to sunrise is used as the first judgment condition, and the aligned ambient light intensity data is lower than the first preset light threshold as the second judgment condition. When either the first or the second judgment condition is met, the real-time weight coefficient corresponding to the aligned dynamic pedestrian density data is increased.

5. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 2, characterized in that, S2 includes: S21: The fusion control parameters generated in S1 are formatted and assembled with the real-time working status data obtained from the monitoring terminals of each street light unit in the target street area to form a complete decision input vector containing timestamps, spatial location identifiers, fusion control parameter values ​​and multiple real-time working status data values. S22: Input the complete decision input vector into the preset adaptive decision model. The adaptive decision model calculates and analyzes the input vector according to its internal mapping rules and optimization objectives to generate a preliminary brightness adjustment instruction for each specific street light unit. S23: Perform regional coordination verification and smoothing filtering on all the preliminary brightness adjustment commands output by the adaptive decision model to eliminate conflicts between commands and ensure the spatiotemporal continuity of brightness changes, and finally generate the dynamic dimming strategy that can be directly issued to each street light unit for execution and contains all the brightness adjustment commands.

6. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 5, characterized in that, The real-time operating status data includes the current operating temperature, continuous operating time, current operating brightness percentage, and historical fault code records for each street light unit.

7. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 5, characterized in that, S3 includes: S31: Analyze the dynamic dimming strategy generated by S2, extract the unique identifier for each target street light unit and its corresponding specific brightness adjustment instruction, and generate a series of executable dimming instruction sets containing target address and brightness setting value according to the preset instruction encoding rules. S32: Through a wireless or wired communication network deployed in the target street area, each instruction in the executable dimming instruction set is distributed in real time and reliably to the corresponding street light unit controller according to the target address it contains. S33: After receiving and verifying the executable dimming instruction set, each street light unit controller drives its internal dimming circuit to perform a brightness adjustment operation, and uploads the actual brightness value after execution as instruction execution feedback data back to the central control system, thereby completing the closed loop of the execution of the dynamic dimming strategy.

8. The intelligent street lighting control method based on illumination and pedestrian flow parameters according to claim 7, characterized in that, The unique identifier is the MAC address of the street light unit, a unique IP address in the communication network, or a pre-assigned physical location code; the preset instruction encoding rule uses JSON format or binary protocol format to encapsulate the unique identifier and the brightness adjustment instruction.

9. A method for intelligent control of street lighting based on illumination and pedestrian flow parameters according to claim 7, characterized in that, S4 includes: S41: Based on the dynamic pedestrian density data, the time-series prediction algorithm is used to analyze the spatial movement trajectory, speed and density change trend of pedestrian flow, identify specific sub-regions where abnormal pedestrian flow will occur within a preset future time window, and determine whether the convergence is an unexpected aggregation that exceeds the normal periodic pattern. S42: Based on the predicted scale, arrival time and duration of the expected external gathering, call the preset plan library to generate a matching forward-looking control plan. The plan clearly specifies the brightness adjustment target and adjustment timetable for the streetlights involved in the specific sub-area and its upstream path of the crowd from the current moment to the end of the gathering event. S43: The forward-looking control plan is integrated in real time with the dynamic dimming strategy that is being generated or has been generated in S2 to form an enhanced dynamic dimming strategy with forward-looking brightness adjustment instructions superimposed, and submitted to S3 for execution, so as to complete the pre-allocation of lighting resources and the gradual transition of brightness before the unexpected gathering actually occurs.

10. A method for intelligent control of street lighting based on illumination and pedestrian flow parameters according to claim 9, characterized in that, The specific rule for determining whether it is an expected external gathering is as follows: when the predicted population density of the specific sub-region exceeds its historical benchmark value for the same period by a first preset threshold, and the rate of increase of population density exceeds a second preset threshold, it is determined to be an expected external gathering.

Images