Internet of Things street lamp intelligent control method and system

Abstract

The invention belongs to the technical field of Internet of Things intelligent illumination, and particularly discloses an Internet of Things street lamp intelligent control method and system, and the method comprises the steps: collecting the traffic flow, environment illumination, weather conditions, date types and specific time data, and obtaining the total power limit of a system, the rated power of a street lamp, and the accumulated operation time; identifying a traffic state based on traffic flow and determining an illumination reference value, fusing multi-source data through a dynamic weight distribution model, and calculating a dynamic target illumination demand value; calculating a lighting effect parameter based on the accumulated operation time, determining a power demand value in combination with the dynamic target illumination demand value and a preset irradiation area, and predicting a total power limit change trend of the system; taking the dynamic target illumination demand value, the power demand value and the basic safety illumination value as input, and generating a power instruction set through calculation under system constraint; the power instruction set is executed in a time-staggered or smooth gradual change mode, and energy-saving, safe and self-adaptive regulation and control of road lighting are achieved.

Description

Technical Field

[0001] This invention relates to the field of IoT smart lighting technology, and more specifically, to an IoT street light smart control method and system. Background Technology

[0002] With the development of IoT technology, road lighting control systems have gradually evolved from traditional timed control to intelligent control. Existing technologies include automatic dimming systems based on ambient light detection, which can adjust streetlight brightness according to natural light intensity; other systems employ traffic flow statistics technology to provide differentiated lighting during different traffic periods. These systems achieve energy-saving effects to a certain extent through single-parameter sensing and fixed threshold control. Their simple system structure and low deployment cost lay the foundation for the field of intelligent lighting.

[0003] However, it still has some drawbacks in practical use: 1. Insufficient ability to integrate and dynamically adjust multi-dimensional environmental data: Existing systems usually only consider a single or a few environmental parameters for decision-making, lacking a collaborative analysis and dynamic weight allocation mechanism for multi-dimensional situational factors such as date type, specific time and real-time weather conditions. This static or single-dimensional control mode cannot accurately identify complex scenarios such as severe weather combined with traffic peaks, causing lighting strategies to deviate from actual safety requirements, which not only poses driving safety hazards but also causes energy waste. 2. High-frequency data fluctuations cause system control oscillations: Systems based on ambient light acquisition generally lack effective mechanisms to suppress instantaneous fluctuations in input data. They directly respond to instantaneous measurements containing noise. When encountering brief disturbances such as passing clouds, flying birds, or vehicle lights, the system will generate unnecessary brightness adjustment commands, causing frequent fluctuations in lamp output. This not only affects driving visual comfort and road safety, but also accelerates equipment aging and increases system maintenance costs. 3. Lack of system-level resource constraints and forward-looking optimization capabilities: Existing solutions fail to establish an optimization model that integrates the prediction of the total system power limit with multiple objective constraints. They lack the ability to predict the future trend of the total system power limit. When faced with grid power fluctuations or insufficient solar power supply, they cannot generate the optimal power allocation instruction set under multiple constraints such as the rated power of a single lamp, the basic safe illuminance, the predicted value of the total system power limit, and the overall power demand. This results in the inability to achieve smooth degradation and reliable operation during energy shortages, affecting the overall stability of the system. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent control method and system for Internet of Things street lights, which solves the problems mentioned in the background art through the following solutions.

[0005] To achieve the above objectives, the present invention provides an intelligent control method for IoT streetlights, comprising: S1. Multidimensional data acquisition: Collect external environment perception data and dynamic situation factor data of the target road, and obtain the operating status parameters of the street lights. The external environment perception data includes traffic flow data and ambient light data. The dynamic situation factor data includes real-time weather data, date type data and specific time data. The operating status parameters include the total power limit of the system, the rated power of each street light and the cumulative operating time. S2. Dynamic Fusion and Demand Assessment: Based on the traffic flow data, the traffic status of the road is identified and the corresponding illuminance benchmark value is determined. Then, the dynamic context factor data and the external environment perception data are input into the dynamic weight allocation model. The illuminance benchmark value is dynamically adjusted based on the date type data, the specific time data and the real-time weather data. The dynamic target illuminance demand value is calculated by weighted fusion by suppressing the instantaneous fluctuations of the input data. S3. Lighting Capacity Assessment and Prediction: Calculate the current luminous efficacy parameters based on the cumulative operating time, and then determine the power demand value of the streetlights based on the dynamic target illuminance demand value, the preset irradiation area, and the current luminous efficacy parameters; based on the total system power limit and its historical data, predict the trend of the change of the total system power limit in a future set period through a time series model; S4. Constraint Power Instruction Set Generation: Taking the dynamic target illuminance demand value, the power demand value and the basic safe illuminance value as input, the system performs optimization calculations under constraints through the illuminance-power mapping relationship to generate a power instruction set for each street light in the area within a future set time period. S5. Execution of Regional Coordination Instruction Set: Execute the power instruction set to adjust the output power of adjacent streetlights through staggered or smooth gradual adjustment.

[0006] The present invention also provides an IoT street light intelligent control system, comprising: The multi-dimensional data acquisition module is used to collect external environmental perception data and dynamic context factor data of the target road, and to obtain the operating status parameters of the streetlights; The illuminance demand fusion module is used to identify the traffic status of the road based on the traffic flow data and determine the corresponding illuminance benchmark value, and dynamically adjust the illuminance benchmark value through a dynamic weight allocation model, and calculate the dynamic target illuminance demand value by suppressing the instantaneous fluctuation of the input data through weighted fusion. The power demand assessment module is used to calculate the current luminous efficacy parameters based on the cumulative running time, and then determine the power demand value of the street light based on the dynamic target illuminance demand value, the preset irradiation area and the current luminous efficacy parameters. The power limit prediction module is used to predict the trend of the total power limit of the system within a future set period based on the total power limit of the system and its historical data, using a time series model. The optimization instruction generation module is used to take the dynamic target illuminance demand value, the power demand value and the basic safe illuminance value as input, and generate a set of power instructions for each street light in the area within a future set time period through optimization calculation under system constraints. The regional collaborative execution module is used to execute the power instruction set and adjust the output power of adjacent streetlights in a staggered or smooth manner.

[0007] The technical effects and advantages of this invention are as follows: 1. Multi-dimensional data fusion enhances adaptability to complex scenarios: This invention integrates multi-dimensional data such as traffic flow, ambient light, weather, date and time through a dynamic weight allocation model, and dynamically adjusts the illuminance benchmark value based on date type and weather conditions. This effectively solves the shortcomings of existing technologies in the collaborative analysis of multi-dimensional situational factors, enabling the system to accurately identify complex scenarios such as severe weather superimposed with traffic peaks, and achieve precise matching between lighting output and actual safety needs. This significantly improves energy efficiency while enhancing road safety. 2. Fluctuation suppression mechanism eliminates system control oscillation: This invention sets up an input data instantaneous fluctuation suppression mechanism in the dynamic weight allocation model. By smoothing and filtering high-frequency data such as ambient light and performing anti-interference processing, it fundamentally solves the control oscillation problem caused by direct response noise in existing systems. It effectively avoids erroneous adjustments caused by instantaneous interference such as passing clouds, ensures the stability of lamp output, and significantly extends the service life of equipment while improving driving comfort. 3. System-level resource optimization for reliable power supply management: This invention uses a time-series model to predict the trend of changes in the total power limit of the system, and establishes an optimization model with the rated power of each street light, the predicted value of the total power limit of the system, and the basic safe illuminance as constraints. This overcomes the lack of optimization capability of existing solutions under multiple constraints, enabling the system to generate the optimal power allocation strategy when the power grid supply fluctuates or solar energy is insufficient, achieving stable degraded operation and ensuring the overall reliability of the system. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0009] Figure 2 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0010] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] refer to Figure 1 The method for intelligent control of IoT streetlights includes multi-dimensional data acquisition, dynamic fusion and demand assessment, lighting capacity assessment and prediction, generation of constraint power instruction sets, and execution of regional collaborative instruction sets.

[0012] refer to Figure 2 The IoT street light intelligent control system shown includes a multi-dimensional data acquisition module, an illuminance demand fusion module, a power demand assessment module, a power limit prediction module, an optimization instruction generation module, and a regional collaborative execution module.

[0013] The multidimensional data acquisition module is connected to the illuminance demand fusion module, the power demand assessment module, and the power limit prediction module. The illuminance demand fusion module, the power demand assessment module, and the power limit prediction module are each connected to the optimization instruction generation module. The optimization instruction generation module is connected to the regional collaborative execution module.

[0014] The acquisition of multidimensional data includes the following steps: S101: Environmental Sensing Data Acquisition The multi-dimensional data acquisition module continuously collects traffic flow data of the target road through the traffic flow sensor and records the number of vehicles passing by per unit time; the ambient light sensor monitors the light intensity on the road surface in real time and collects ambient light data.

[0015] S102: Acquisition of Dynamic Context Factors The multi-dimensional data acquisition module provides accurate date and time data through a real-time clock circuit, distinguishing date data into weekdays, weekends, and holidays; weather data can be obtained in real time by connecting to the meteorological station API, distinguishing between normal weather and severe weather conditions.

[0016] S103: Obtaining Operating Status Parameters The multi-dimensional data acquisition module reads the total power limit of the system from the municipal power grid monitoring system or the solar inverter; and reads the cumulative operating time of each street light from the street light controller.

[0017] S104: Data Preprocessing and Transmission The multidimensional data acquisition module performs preliminary verification and formatting on the collected raw data, and then transmits the integrated data packets to the corresponding illuminance demand fusion module, power demand assessment module, and power limit prediction module.

[0018] The dynamic fusion and demand assessment includes the following steps: S201: Traffic Condition Recognition and Illumination Reference Value Determination The illuminance demand fusion module receives traffic flow data from the multi-dimensional data acquisition module and sets a first traffic flow threshold Q1 and a second traffic flow threshold Q2, where Q1 < Q2. The specific values ​​of the two traffic flow thresholds are set based on the design capacity of the target road and the service level classification in traffic flow theory. Based on the traffic flow data, traffic state identification is performed. When the traffic flow is lower than Q1, it is identified as an idle state, and a basic safe illuminance value E is set for this state. b When the traffic flow is between Q1 and Q2, it is identified as a sparse flow state, and an enhanced illuminance value E is set for this state. e And E e >E b When the traffic flow is higher than Q2, it is identified as a continuous flow state, and a uniform illuminance value E for the entire road section is set for this state. u And E u >E e The specific values ​​of the illuminance reference values ​​for each state are determined comprehensively based on the values ​​specified in the national road lighting design standards and in conjunction with the safety requirements of the target road; the illuminance reference value E determined in this step is... r As the input benchmark for the dynamic weight allocation model.

[0019] S202: Dynamic Weight Allocation Model Construction and Data Preprocessing The illuminance demand fusion module establishes a dynamic weight allocation model, which defines the rules for weighting three contextual factors: date type, specific time, and weather conditions. Dynamic contextual factor data transmitted from the multi-dimensional data acquisition module is input into this model. Simultaneously, a filtering algorithm is applied to the input ambient light data to suppress instantaneous fluctuations; and anti-interference processing is used on the traffic flow data to eliminate abnormal counting interference.

[0020] S203: Weighting Allocation and Illuminance Demand Integration Calculation The illumination demand fusion module dynamically determines the weight coefficients of each factor based on the input context factors. Specifically, the date type weight is differentiated according to weekdays, weekends and holidays, and a relatively higher base weight is assigned to weekdays; the specific time weight is determined by identifying peak traffic periods and significantly increasing the weight coefficient during those periods; and the weather condition weight is determined by distinguishing between normal weather and severe weather, and its weight coefficient is significantly increased under severe weather conditions.

[0021] Based on the weights of each data item output by the dynamic weight allocation model, a weighted fusion calculation is performed on the filtered data. The calculation formula is as follows: , in, This represents the dynamic target illuminance requirement value. The established illuminance reference value; , , These are weighting coefficients for weather conditions, specific time, and date type, used to quantify the real-time impact of each situational factor, and are dynamically determined based on real-time data; , , These are the adjustment coefficients corresponding to the aforementioned context factors, used to control the magnitude of each factor's influence in the fusion calculation. The adjustment coefficients are preset constants, and their specific values ​​depend on the system design requirements and the verification and optimization of actual road lighting scenarios.

[0022] S204: Output of dynamic target illuminance demand value The illuminance demand fusion module will calculate the dynamic target illuminance demand value. The value is transmitted to the power demand assessment module, where it is limited to an illuminance upper limit determined according to the glare limit specified in the national road lighting design standard, while also including the basic safe illuminance value E. b The data is transmitted to the optimization instruction generation module, providing accurate input parameters for subsequent power demand assessment and optimization instruction generation.

[0023] The lighting capability assessment and prediction includes the following steps: S301: Power Demand Calculation The power demand assessment module receives the dynamic target illuminance demand value from the illuminance demand fusion module. It also receives cumulative running time data from the multidimensional data acquisition module.

[0024] Based on the received cumulative running time, the current luminous efficacy parameters are calculated using a preset luminous decay curve model. The calculation formula is: ,in, These are the initial luminous efficacy parameters for the streetlights. This is the light decay function based on the cumulative running time.

[0025] Then, the power demand value is calculated. The calculation formula is: ,in, The preset illumination area is determined based on the installation height, light distribution type, and spacing of the streetlights.

[0026] Parallel calculations are performed on each street light within the area to generate a complete dataset containing the power requirements of all street lights.

[0027] S302: Preparation of Historical System Total Power Limit Data The power limit prediction module obtains historical data on the total system power limit from the multi-dimensional data acquisition module, including records of the total system power limit over the past few days or weeks. The historical data is cleaned and preprocessed, and the processed historical total system power limit data is arranged in chronological order to construct a time series dataset. The periodic characteristics of the historical data are analyzed to identify regular change patterns such as daily and weekly cycles.

[0028] S303: Prediction of the Trend of Change in System Total Power Limit The power limit prediction module uses the Autoregressive Integral Moving Average (ARIMA) model as the core prediction algorithm to perform rolling predictions of the total system power limit within a set future time period, generating a sequence of predicted total system power limit values.

[0029] S304: Prediction Results Output The power limit prediction module transmits the predicted total power limit value of the system to the optimization instruction generation module, providing system-level power constraints for constraint optimization calculations. At the same time, it transmits the power demand value to the optimization instruction generation module as a reference input for optimization calculations.

[0030] The generation of the constrained power instruction set includes the following steps: S401: Optimize Input and Constraint Initialization The optimization instruction generation module receives and integrates all input parameters, setting the dynamic target illuminance demand as the primary objective of this optimization round. It sets the basic safe illuminance value as the lower limit constraint that the entire optimization problem must adhere to, ensuring that any solution meets the minimum safe lighting requirements for roads. Simultaneously, two types of power constraints are applied: one is a hardware constraint, namely the rated power of each streetlight, serving as the upper limit for each decision variable; the other is a system-level constraint, namely the predicted total system power limit for future periods provided by the power limit prediction module, serving as the upper limit for the sum of the power of all streetlights. Finally, the power demand value calculated by the power demand assessment module is used as the initial reference solution for the optimization calculation to accelerate the convergence process.

[0031] S402: Constrained Optimization Calculation and Solution The optimization command generation module constructs a constrained nonlinear optimization problem based on the initialized objective and constraints. The solution process relies on a preset illuminance-power mapping relationship, which is determined by the optical characteristics of the streetlights, mapping the power command to the actual illuminance distribution on the road surface. The optimization algorithm uses this mapping relationship as its basis and iteratively solves the problem with the objective function of minimizing the overall deviation between the dynamic target illuminance demand and the predicted achievable illuminance.

[0032] S403: Power instruction set generation The instruction generation module optimizes the output and generates the final instructions based on the solution results. This process includes two output modes: 1. Standard Power Command Set Generation: When a feasible solution satisfying all constraints is successfully found, the solution is converted into precise power commands for each street light in the area within a future set time period, generating a standard power command set. This command set aims to achieve dynamic target illuminance requirements under ideal conditions.

[0033] 2. Degraded Power Instruction Set Generation: When the predicted total system power limit is too low or other constraints are too stringent, making it impossible to find a feasible solution that meets the primary objective, a degradation processing mechanism is triggered. In this case, the system automatically switches the optimization objective to "meeting the basic safe illuminance value" and re-optimizes the solution. The solution generated in this mode is defined as the degraded power instruction set. This instruction set prioritizes ensuring that all areas reach the lower limit of safe illuminance and allocates power within the remaining power allowance, reflecting the system's resilience design and fault-tolerant principles under resource constraints.

[0034] S404: Output of optimization results The optimization instruction generation module packages the final power instruction set into a structured data packet. This data packet is then transmitted to the regional collaborative execution module as its direct control basis.

[0035] The execution of the regional collaborative instruction set includes the following steps: S501: Instruction Set Parsing and Traffic State Matching The regional collaborative execution module first deconstructs the received power command set data packet, extracting the target brightness level for each street light and its corresponding effective timestamp. Simultaneously, it obtains the current road traffic status identification results provided by the illuminance demand fusion module in real time through a data interface. This real-time traffic scenario is then matched with the brightness command to be executed, providing a decision-making basis for subsequently selecting the most suitable brightness adjustment strategy.

[0036] S502: Collaborative Execution Model: Decision-Making and Execution The regional collaborative execution module dynamically selects and activates the corresponding brightness adjustment execution mode based on real-time traffic conditions. The specific modes and execution logic are as follows: When the regional cooperative execution module identifies a sparse flow state, it activates staggered adjustment. This module links with vehicle detection sensors to obtain the precise location and speed of the target vehicle on the road in real time. Based on this, the regional cooperative execution module implements a "following" lighting strategy: it controls the streetlights within a predetermined distance in front of the vehicle to sequentially and smoothly increase to the target brightness required by the power command set at preset safety time intervals, thereby creating a continuous "light corridor" for the moving vehicle. The streetlights behind the vehicle gradually reduce their brightness to maintain a basic safe illuminance or enter a dormant state after the vehicle passes, using a similar delay strategy. This mode achieves on-demand lighting and energy saving while ensuring driving safety visual requirements.

[0037] When the regional collaborative execution module identifies a continuous flow state or an idle state, it activates the smooth transition mode. In this mode, the module ignores the position of individual vehicles and instead controls all relevant streetlights within the area to transition uniformly and slowly from the current brightness level to the target brightness set by the power command set at a uniform, preset safe rate of change. This global, slow change is almost imperceptible to the human eye, effectively avoiding visual persistence and glare discomfort that may be caused by sudden changes in brightness. It is particularly suitable for continuous flow states with heavy traffic or for adjusting basic lighting throughout the day, ensuring the overall comfort and stability of the lighting environment.

[0038] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. 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 smart control method for IoT streetlights, characterized in that, include: S1. Multidimensional data acquisition: Collect external environment perception data and dynamic situation factor data of the target road, and obtain the operating status parameters of the street lights. The external environment perception data includes traffic flow data and ambient light data. The dynamic situation factor data includes real-time weather data, date type data and specific time data. The operating status parameters include the total power limit of the system, the rated power of each street light and the cumulative operating time. S2. Dynamic Fusion and Demand Assessment: Based on the traffic flow data, the traffic status of the road is identified and the corresponding illuminance benchmark value is determined. Then, the dynamic context factor data and the external environment perception data are input into the dynamic weight allocation model. The illuminance benchmark value is dynamically adjusted based on the date type data, the specific time data and the real-time weather data. The dynamic target illuminance demand value is calculated by weighted fusion by suppressing the instantaneous fluctuations of the input data. S3. Lighting Capacity Assessment and Prediction: Calculate the current luminous efficacy parameters based on the cumulative operating time, and then determine the power demand value of the streetlights based on the dynamic target illuminance demand value, the preset irradiation area, and the current luminous efficacy parameters; based on the total system power limit and its historical data, predict the trend of the change of the total system power limit in a future set period through a time series model; S4. Constraint Power Instruction Set Generation: Taking the dynamic target illuminance demand value, the power demand value and the basic safe illuminance value as input, the system performs optimization calculations under constraints through the illuminance-power mapping relationship to generate a power instruction set for each street light in the area within a future set time period. S5. Execution of Regional Coordination Instruction Set: Execute the power instruction set to adjust the output power of adjacent streetlights through staggered or smooth gradual adjustment.

2. The IoT street light intelligent control method according to claim 1, characterized in that, In S2, identifying the road traffic conditions and determining the corresponding illuminance reference value based on the traffic flow data includes: Set a first traffic flow threshold and a second traffic flow threshold, wherein the first traffic flow threshold is less than the second traffic flow threshold; When the traffic flow is lower than the first traffic flow threshold, it is identified as an idle state, and a basic safe illuminance value is set for the idle state. When the traffic flow is between the first traffic flow threshold and the second traffic flow threshold, it is identified as a sparse flow state, and a road segment enhanced illuminance value is set for the sparse flow state, wherein the road segment enhanced illuminance value is higher than the basic safe illuminance value. When the traffic flow is higher than the second traffic flow threshold, it is identified as a continuous flow state, and a uniform illuminance value for the entire road segment is set for the continuous flow state. The uniform illuminance value for the entire road segment is higher than the enhanced illuminance value for the road segment. The illuminance benchmark value serves as the input benchmark for the dynamic weight allocation model.

3. The IoT street light intelligent control method according to claim 1, characterized in that, In S2, dynamically adjusting the illuminance reference value based on the date type data, the specific time data, and the real-time weather condition data includes: The date type data is divided into weekdays, weekends and holidays. Traffic peak periods are identified based on the date type data and specific time data. During the traffic peak periods, the traffic flow data is assigned a relatively higher weight. The weather condition data is divided into normal weather and severe weather. Under severe weather conditions, the illuminance reference value is adaptively increased.

4. The IoT street light intelligent control method according to claim 1, characterized in that, In S3, the power requirement of streetlights is calculated as follows: Power requirement value = (Dynamic target illuminance requirement value × preset irradiation area) / current luminous efficacy parameter; The preset illumination area is determined based on the installation height, light distribution type, and arrangement spacing of the streetlights. The current luminous efficacy parameter is based on the preset initial luminous efficacy parameter and is calculated by combining the cumulative operating time of the streetlights with the preset light decay curve model.

5. The IoT street light intelligent control method according to claim 1, characterized in that, In S3, the time series model is an autoregressive integral moving average (ARIMA) model, whose input is historical system total power limit data and output is the predicted system total power limit value for the next few hours.

6. The IoT street light intelligent control method according to claim 1, characterized in that, In S4, the optimization calculations under system constraints include: The optimization objective is to use the dynamic target illuminance demand value as the optimization objective, the basic safe illuminance value to ensure road safety as the lower limit of illuminance, and the rated power of each street light and the predicted total power limit of the system as power constraints. The constraint optimization calculation is performed by referring to the power demand value through the illuminance-power mapping relationship.

7. The IoT street light intelligent control method according to claim 1, characterized in that, In S4, the power command set for each street light in the area within a future set time period includes: When a solution exists that satisfies all constraints, output a power instruction set corresponding to the dynamic target illuminance requirement value; If no solution satisfies all constraints, the basic safe illuminance value is used as the optimization target to resolve the problem and output a degraded power instruction set.

8. The IoT street light intelligent control method according to claim 7, characterized in that, The streetlights include conventional streetlights powered by the municipal power grid and solar streetlights powered by solar energy; The conventional streetlights, due to their stable total system power limit, mainly operate in a state that outputs a power instruction set corresponding to the dynamic target illuminance demand value; The solar streetlights, whose total system power limit fluctuates due to weather conditions, mainly operate in a state of outputting the degraded power instruction set.

9. The IoT street light intelligent control method according to claim 1, characterized in that, In S5, adjusting the output power of adjacent streetlights through staggered timing or smooth transitions includes: When the sparse flow state is identified, a staggered adjustment method is adopted to control the street lights in front of the vehicle to adjust their brightness sequentially at safe time intervals according to a preset order. When the system is identified as being in a continuous flow state or an idle state, a smooth and gradual transition method is used to control all streetlights to transition to the target brightness at a safe rate.

10. An IoT street light intelligent control system according to any one of claims 1-9, characterized in that, include: The multi-dimensional data acquisition module is used to collect external environmental perception data and dynamic context factor data of the target road, and to obtain the operating status parameters of the streetlights; The illuminance demand fusion module is used to identify the traffic status of the road based on the traffic flow data and determine the corresponding illuminance benchmark value, and dynamically adjust the illuminance benchmark value through a dynamic weight allocation model, and calculate the dynamic target illuminance demand value by suppressing the instantaneous fluctuation of the input data through weighted fusion. The power demand assessment module is used to calculate the current luminous efficacy parameters based on the cumulative running time, and then determine the power demand value of the street light based on the dynamic target illuminance demand value, the preset irradiation area and the current luminous efficacy parameters. The power limit prediction module is used to predict the trend of the total power limit of the system within a future set period based on the total power limit of the system and its historical data, using a time series model. The optimization instruction generation module is used to take the dynamic target illuminance demand value, the power demand value and the basic safe illuminance value as input, and generate a set of power instructions for each street light in the area within a future set time period through optimization calculation under system constraints. The regional collaborative execution module is used to execute the power instruction set and adjust the output power of adjacent streetlights in a staggered or smooth gradual manner.

Images