LED intelligent illumination control method and device based on Internet of Things

Abstract

The invention relates to an LED intelligent lighting control method and device based on the Internet of Things, and the method can collect the environment perception data of a plurality of intelligent LED lamps, and carries out the dimming calculation, and obtains an initial dimming parameter. Performing instruction construction on the initial dimming parameters in combination with historical energy consumption records and environmental perception data of the intelligent LED lamp, and generating a target dimming instruction; controlling the intelligent LED lamp through an edge gateway based on the target dimming instruction, and obtaining lamp response data; and continuously judging the intelligent LED lamp according to the lamp response data to obtain fault judgment information, and triggering an operation and maintenance control instruction. According to the invention, the operation state of the intelligent LED lamp can be accurately judged, potential faults can be found in time, and automatic operation and maintenance can be triggered. According to the invention, the operation state of the intelligent LED lamp can be accurately judged, potential faults can be found in time, and automatic operation and maintenance can be triggered.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of LED intelligent lighting, and particularly relates to an LED intelligent lighting control method and device based on Internet of Things. BACKGROUND

[0002] The development of Internet of Things technology promotes the wide application of LED intelligent lighting systems, but the existing control methods still have deficiencies. The traditional scheme relies on simple timing or ambient light adjustment, which has poor adaptability. Although intelligent dimming based on real-time data is improved, it is difficult to optimize long-term energy efficiency due to the lack of historical energy consumption analysis. In addition, the existing system has limited monitoring capability for lamp states, and fault detection relies on manual or threshold alarms, which has a lagging response, affecting the stability and operation efficiency of the system. Therefore, an intelligent lighting control method that combines environmental perception and historical data and has dynamic fault detection capability is urgently needed to improve the adaptability and reliability of the system. SUMMARY

[0003] The main purpose of the present application is to provide an LED intelligent lighting control method and device based on Internet of Things, which can accurately determine the running state of intelligent LED lamps and timely detect potential faults and trigger automated operation and maintenance.

[0004] To achieve the above purpose, the present application provides an LED intelligent lighting control method based on Internet of Things, comprising: Collecting environmental perception data of multiple intelligent LED lamps and performing dimming calculation to obtain initial dimming parameters; Combining the historical energy consumption records and environmental perception data of the intelligent LED lamps, constructing instructions for the initial dimming parameters, and generating target dimming instructions; Controlling the intelligent LED lamps based on the target dimming instructions through an edge gateway, and obtaining lamp response data; Continuously determining the intelligent LED lamps according to the lamp response data to obtain fault determination information, and triggering operation and maintenance control instructions.

[0005] Further, the collecting of environmental perception data of multiple intelligent LED lamps and the dimming calculation to obtain initial dimming parameters comprise: Collecting sensing data of sensor nodes of multiple intelligent LED lamps to obtain light intensity and node temperature and humidity data; Obtaining regional meteorological information according to the location information of multiple intelligent LED lamps; Comparing the light intensity with a preset illumination threshold level by level to generate a reference brightness compensation coefficient; Compensating the node temperature and humidity data for meteorological interference through the regional meteorological information to obtain meteorological interference parameters; The weather interference parameter is combined with the reference brightness compensation coefficient to generate the initial dimming parameter. Further, the initial dimming parameter is instructed to be constructed by combining the historical energy consumption record and the environmental perception data of the intelligent LED lamp, to generate a target dimming instruction, including: Energy consumption data of a plurality of intelligent LED lamps is acquired to perform group periodic energy consumption integration to form the historical energy consumption record. The historical energy consumption record is matched according to a preset time period division rule to obtain energy consumption data of each time period. The energy consumption feature data of each time period is associated and mapped with the environmental perception data to generate an energy consumption association data set. The initial dimming parameter is group cooperatively verified based on the energy consumption association data set to generate a group dimming parameter. The group dimming parameter is instructed to be constructed by a preset dimming rule to generate the target dimming instruction.

[0006] Further, the initial dimming parameter is group cooperatively verified based on the energy consumption association data set to generate a group dimming parameter, including: The energy consumption association data set is divided into groups to obtain a group division result. The initial dimming parameter is dynamically threshold corrected in combination with the group division result to generate a group dynamic dimming threshold. The group dynamic dimming threshold is matched with the environmental perception data to generate a group light demand matching degree. The initial dimming parameter is cooperatively optimized based on the group light demand matching degree to generate the group dimming parameter.

[0007] Further, the intelligent LED lamp is controlled by an edge gateway based on the target dimming instruction, and lamp response data is acquired, including: The target dimming instruction is interpreted by an edge communication protocol to generate a device driving signal. The device driving signal is sent to each intelligent LED lamp by the edge gateway, and hierarchical scheduling is performed to generate a real-time control queue. The intelligent LED lamp is executed by the edge gateway based on the real-time control queue to perform operation monitoring to capture lamp operation data. The lamp operation data is compound state calibrated to generate lamp response data.

[0008] Further, the edge gateway sends the device driving signal to each smart LED lamp and performs hierarchical scheduling to generate a real-time control queue, including: The edge gateway calibrates the transmission node of each smart LED lamp to generate a node communication scheduling table; Based on the node communication scheduling table, the device driving signal is signal-fragmented and packaged and hierarchical detection is performed to generate a fragmented driving data packet; The edge gateway is allocated interface broadband based on the node communication scheduling table and the fragmented driving data packet to obtain a signal transmission queue; According to the signal transmission queue, the smart LED lamp is controlled in a response queue to obtain the real-time control queue.

[0009] Further, the smart LED lamp is continuously determined based on the lamp response data to obtain fault determination information, and an operation and maintenance control instruction is triggered, including: According to a pre-set fault type database, the fault state of each smart LED lamp is analyzed based on the lamp response data to obtain lamp fault information; The lamp fault information and the historical energy consumption record are associated to obtain energy consumption anomaly information; The smart LED lamp is determined based on the lamp fault information and the energy consumption anomaly information to obtain the fault determination information; Based on the fault determination information, a pre-set operation and maintenance instruction library is triggered to construct operation and maintenance to obtain the operation and maintenance control instruction.

[0010] Further, the operation and maintenance control instruction is obtained based on the fault determination information by triggering a pre-set operation and maintenance instruction library to construct operation and maintenance, including: According to the operation and maintenance instruction library, the fault type of the fault determination information is matched to generate an initial operation and maintenance instruction set; According to the response time limit rule of the operation and maintenance instruction library, the urgency of the initial operation and maintenance instruction set is divided to generate a first operation and maintenance task queue and a second operation and maintenance task queue, wherein the first operation and maintenance task queue contains real-time operation and maintenance instructions, and the second operation and maintenance task queue contains delayed operation and maintenance instructions; Based on a pre-set real-time scheduling strategy, the first operation and maintenance task queue is optimized to obtain a first operation and maintenance instruction sequence; Based on a pre-set periodic scheduling strategy, the second operation and maintenance task queue is optimized to obtain a second operation and maintenance instruction sequence; The first operation and maintenance instruction sequence and the second operation and maintenance instruction sequence are combined in priority to execute the instructions to obtain the operation and maintenance control instruction.

[0011] The application further provides an LED intelligent lighting system control device based on the Internet of Things, applied to the LED intelligent lighting system control method based on the Internet of Things. An identification module is configured to collect environment sensing data of a plurality of intelligent LED lamps and lanterns, and perform dimming calculation to obtain initial dimming parameters. An analysis module is configured to combine historical energy consumption records and environment sensing data of the intelligent LED lamps and lanterns, and perform instruction construction on the initial dimming parameters to generate target dimming instructions. A processing module is configured to control the intelligent LED lamps and lanterns based on the target dimming instructions through an edge gateway, and acquire lamp and lantern response data. A construction module is configured to continuously determine the intelligent LED lamps and lanterns according to the lamp and lantern response data, obtain fault determination information, and trigger operation and maintenance control instructions.

[0012] The LED intelligent lighting control method and device based on the Internet of Things have the following beneficial effects: The dimming parameter optimization by fusing environment sensing data and historical energy consumption records overcomes the limitation of traditional lighting control relying on real-time environment data only, so that the dimming strategy meets current environment requirements and conforms to long-term energy efficiency optimization targets, and the energy saving effect of the lighting system is significantly improved. The real-time control architecture based on the edge gateway combined with dynamic monitoring of lamp and lantern response data realizes accurate determination of the running state of the intelligent LED lamps and lanterns, can discover potential faults in time and trigger automatic operation and maintenance, greatly reduces the artificial inspection cost, and improves the system reliability. Through the organic combination of environment sensing, energy consumption analysis and fault warning, a complete intelligent lighting control closed loop is constructed, the self-adaptive ability of the system to different application scenarios is enhanced, the service life of the lamps and lanterns is prolonged through accurate operation and maintenance, and reliable technical support is provided for intelligent management and efficient operation of the intelligent lighting system. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 is a flow chart of the LED intelligent lighting control method based on the Internet of Things provided by the application. Figure 2 is a structural diagram of the LED intelligent lighting control device based on the Internet of Things provided by the application.

[0014] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0016] The present application will be further described below in combination with the drawings and specific embodiments.

[0017] Referring to Figure 1 The present application provides an LED intelligent lighting control method based on Internet of Things, comprising: Step S1: collecting environment perception data of a plurality of intelligent LED lamps and performing dimming calculation to obtain initial dimming parameters; Step S2: combining historical energy consumption records and environment perception data of the intelligent LED lamps to construct instructions for the initial dimming parameters, and generating target dimming instructions; Step S3: controlling the intelligent LED lamps based on the target dimming instructions through an edge gateway, and acquiring lamp response data; Step S4: continuously determining the intelligent LED lamps according to the lamp response data to obtain fault determination information, and triggering operation and maintenance control instructions.

[0018] Based on the above steps, the detailed step process is shown as follows: Step S1: Real-time physical environment data is obtained by sensor nodes (such as light sensors, temperature and humidity sensors) deployed on each intelligent LED lamp, specifically including environmental light intensity and lamp self-working temperature and humidity data. At the same time, physical location information coordinates of the intelligent LED lamps are obtained based on an Internet of Things communication module, and real-time weather parameters are extracted by associating with an access area weather data interface or a nearby weather station. When performing dimming calculation, the light intensity threshold model (including multiple brightness intervals) preset in the controller is automatically classified and compared with the environmental light intensity data: when the measured light intensity is lower than the lowest threshold level, a basic dimming instruction is generated; when the threshold is crossed level by level, the compensation coefficient is increased by a preset proportion, and the corresponding reference brightness compensation coefficient is generated.

[0019] The offset caused by weather factors in the lamp temperature and humidity data is calibrated and corrected in combination with regional weather information (such as temperature, humidity, and atmospheric transparency), and the actual influence factor of weather interference on the lamp is quantified to form a weather interference parameter. The reference brightness compensation coefficient and the weather interference parameter are jointly input into a dimming decision unit to generate an initial dimming parameter through parameter fusion calculation (such as weighted superposition), which includes the brightness percentage and color temperature value that each lamp needs to adjust, providing a baseline reference for subsequent optimization.

[0020] Step S2: Extract the periodic performance consumption log data of each intelligent LED lamp (such as daily / weekly power consumption records) from the cloud database or edge storage device, divide the historical energy consumption data into time periods according to the preset time period rules (such as peak-valley electricity price period, day-night period), and extract the characteristics of the time period energy consumption data set (such as peak period average power, valley period cumulative energy consumption). Combine the real-time environmental perception data collected in the previous step (including light intensity, temperature and humidity), dynamically bind the environmental perception data in the same period with the time period energy consumption characteristics, and establish an energy consumption correlation data set containing time stamp, location identification, environmental state and energy consumption characteristics.

[0021] Based on the data set, the initial dimming parameters are optimized in groups: according to the spatial distribution topology of the lamps (such as adjacent lamps on the road, lamps on the same floor of the building), the group is divided, and the initial dimming parameters of the lamps in the group are dynamically threshold corrected (for example, according to the overall lighting demand of the group, the single lamp parameter is adjusted up / down), and the dynamic dimming threshold range covering the group is generated.

[0022] Through the light demand matching module, the group dynamic threshold and the light intensity deviation in the environmental perception data are compared, and the actual light demand satisfaction degree of the group is quantified (for example, the matching degree in the 80%-110% interval is considered to be compliant). Finally, based on the matching degree, the initial dimming parameters are optimized in groups (for example, if the matching degree is too low, the compensation value is proportionally increased), and the group dimming parameters are generated. The parameters input the preset dimming instruction packaging rule, and finally compiled into target dimming instructions for edge gateway execution.

[0023] Step S3: The target dimming instruction input edge gateway protocol analysis engine, converted into bottom layer device driving signal (such as DALI protocol instruction, 0-10V analog signal or PWM waveform parameter). Based on the preset node communication scheduling table, the driving signal is executed in pieces: the group lamp control instruction is divided into independent data packets according to spatial proximity, and the target node identification and check code are attached to generate piece driving data packets.

[0024] The edge gateway dynamically allocates communication interface bandwidth priority, enables real-time transmission queue for data packets of high-density area lamps, and enables batch transmission mode for low-priority nodes, forming a hierarchical signal transmission queue. After receiving the driving signal, each intelligent LED lamp executes brightness / color temperature adjustment, and simultaneously transmits operation state data (including actual brightness output value, driving current, chip temperature, working voltage) through the sensor.

[0025] Edge gateway synchronously captures the above real-time running data stream, marks the timestamp and device ID, and performs composite state calibration on the original running data: for example, "voltage anomaly" is calibrated by combining current fluctuation and brightness output deviation, and "overheating risk" is calibrated according to the chip temperature curve. Finally, it is integrated into a structured luminaire response data set, including four core dimensions of device identification, execution result, electrical parameter and state label, as the basis for fault judgment.

[0026] Step S4: The actual brightness, driving current, chip temperature and working voltage parameters in the input luminaire response data are associated with the normal working reference values (such as the rated current range and standard temperature rise curve) stored in the historical energy consumption record, and fault feature comparison is performed item by item: When the deviation of the actual brightness output value of the luminaire from the target dimming instruction continuously exceeds 15%, it is determined as "light decay failure". When the driving current continuously breaks through the preset upper limit for three times, it is determined as "abnormal driving circuit". When the temperature data breaks through the preset high temperature threshold within a ten-minute period, it is marked as "heat dissipation failure".

[0027] Synchronous correlation energy consumption anomaly analysis: extract the current period energy consumption characteristics, if the single lamp power consumption increases by 30% and is accompanied by current high frequency fluctuation, it is marked as "line leakage risk". The above fault labels and energy consumption anomaly characteristics are input into the hierarchical judgment module, and are classified and disposed according to the device safety criteria: for immediate blocking faults (such as line short circuit, overheating and fuse risk), the highest priority alarm code is generated; for performance degradation problems (such as light decay exceeding the standard, driving efficiency decline), the medium level diagnostic code is generated.

[0028] Trigger operation and maintenance actions based on alarm level: the highest priority code activates the real-time blocking instruction (automatically cuts off the power supply of the fault luminaire through the edge gateway and enables the adjacent standby light source), and the medium level diagnostic code triggers the periodic operation and maintenance work order (pushes the fault device coordinates, abnormal parameter record and recommended repair measures to the operation and maintenance platform). After the execution time sequence optimization of the two types of instructions is combined, the final operation and maintenance control instruction set is formed and issued to the execution terminal.

[0029] This invention provides an IoT-based smart LED lighting control method that optimizes dimming parameters by integrating environmental perception data and historical energy consumption records. This overcomes the limitations of traditional lighting control that relies solely on real-time environmental data, ensuring that the dimming strategy meets both current environmental needs and long-term energy efficiency goals, significantly improving the energy-saving performance of the lighting system. The real-time control architecture based on an edge gateway, combined with dynamic monitoring of luminaire response data, enables precise determination of the operating status of smart LED luminaires. This allows for timely detection of potential faults and triggers automated maintenance, significantly reducing manual inspection costs and improving system reliability. By organically combining environmental perception, energy consumption analysis, and fault early warning, and constructing a complete smart lighting control closed loop, this method not only enhances the system's adaptability to different application scenarios but also extends the lifespan of luminaires through precise maintenance, providing reliable technical support for the intelligent management and efficient operation of smart lighting systems.

[0030] In one embodiment, environmental perception data from multiple smart LED lights is collected and dimming calculations are performed to obtain initial dimming parameters, including: This process relies on the sensing unit and edge data processor integrated into the luminaire to capture environmental parameters. Each smart LED luminaire is equipped with a multispectral illuminance sensor and a temperature and humidity composite probe: the illuminance sensor generates an analog voltage signal corresponding to the ambient light intensity based on the photoelectric conversion principle, which is then quantized by the analog-to-digital conversion module to form a standard digital quantity of light intensity; the temperature and humidity sensor adopts capacitive measurement technology, and simultaneously obtains the surface temperature of the heat sink and the relative humidity of the surrounding air through a temperature-sensitive element and a humidity-sensitive capacitor.

[0031] The data acquisition unit employs a time-division multiplexing-based node polling mechanism to dynamically allocate communication time slots for each sensor, achieving conflict-free transmission. Raw sensor data undergoes error correction processing: illuminance data is compensated for spectral response deviations based on the standard human visual function, while temperature and humidity data are processed to eliminate measurement errors introduced by thermal radiation and sensor hysteresis. Standardized data is encapsulated in a structured format, with key fields including device identifier, acquisition time stamp, light intensity value, temperature value, and relative humidity value. Data packets are stored in the high-speed cache of the edge nodes, forming a time-seriesd environmental perception basic dataset.

[0032] The geographic coordinate information of each node is extracted from the lighting fixture registration database: latitude and longitude coordinates output by a satellite positioning module, or a pre-set municipal facility grid coding system. The location dataset is input into the spatial computing unit to generate a geometric model of a compact envelope region covering the target lighting fixture group.

[0033] Call the standardized meteorological data service interface, submit query parameters including target area geographical boundary descriptor, time retrieval range, meteorological element list; the element list at least contains atmospheric temperature, humidity parameter, aerosol concentration and cloud optical thickness index. The returned original meteorological data performs quality verification, and the core verification dimensions cover data timeliness threshold, field integrity standard and regional continuity constraint.

[0034] The effective data is reconstructed into a regional meteorological parameter matrix, the matrix row index corresponds to the lamp equipment identifier, and the column dimension includes longitude, latitude, reference atmospheric temperature, humidity correction coefficient and atmospheric transmittance attenuation factor. The matrix is associated with the device identifier and the time marker based on the environment perception data in substep 1 to establish an association mapping, and the spatial consistency is guaranteed by a geographical coordinate conversion algorithm, meeting the spatial reference requirements of meteorological compensation calculation.

[0035] A preset multi-level illumination threshold sequence is stored in the non-volatile memory of the edge computing node, and the sequence is arranged in ascending order of environmental illumination intensity and divided into multiple discrete intervals. The real-time collected illumination intensity value is input into the threshold matching engine, and an iterative comparison operation is performed: initializing the compensation coefficient to zero value, activating the basic compensation mode when the measured illumination intensity is lower than the first level threshold, and generating an initial compensation amount; the illumination intensity crosses a level threshold interval, and the compensation coefficient is increased by a preset rule.

[0036] The compensation coefficient generation process introduces a smooth transition mechanism, and a linear interpolation algorithm is used in the adjacent threshold boundary area to prevent brightness jump. The output reference brightness compensation coefficient is a dimensionless scalar, and the numerical range represents the target brightness compensation demand intensity. The coefficient and the environmental illumination intensity are in a non-linear negative correlation, and the core realizes the smooth switching control logic of daylight supplemental lighting and night basic lighting.

[0037] The reference atmospheric temperature, humidity correction coefficient and atmospheric transmittance attenuation factor are extracted from the regional meteorological parameter matrix. The meteorological normalization processing is performed on the lamp local temperature and humidity sensor data: the temperature data is separated from the lamp self-heating and external meteorological heat radiation influence by calculating the environmental heat balance equation, and the meteorological interference temperature component is output; the humidity data uses a dew point temperature conversion model to eliminate the influence of regional atmospheric humidity gradient, and generates a humidity compensation factor.

[0038] The atmospheric transmittance attenuation factor acts on the illumination compensation calibration link, and quantifies the weakening effect of aerosol and cloud on natural light. The final fusion processing outputs the meteorological interference parameter as a multi-dimensional vector data, and the vector elements respectively represent the thermal environment interference intensity, humidity deviation index and light attenuation compensation demand, which constitute the comprehensive quantitative index of environmental interference.

[0039] The reference brightness compensation coefficient is input to the main control unit as a basic lighting demand reference value. The meteorological interference parameters participate in the dimming decision through a weight distribution mechanism: the thermal environment interference intensity is mapped to a lamp heat dissipation efficiency correction table to automatically reduce the maximum allowed brightness under high-temperature working conditions; the humidity deviation index triggers the anti-condensation protection strategy to constrain the minimum working temperature threshold; the light attenuation compensation demand is superimposed on the reference brightness compensation coefficient to enhance the lighting compensation intensity in natural light deficient scenes.

[0040] The multi-dimensional compensation factors are dynamically weighted and fused to output initial dimming parameters as structured control instructions, which mainly include three control variables: target brightness percentage, color temperature adjustment amount, and drive current limit value. The control instructions are written into the device drive interface register to establish an executable basic operation instruction set for subsequent optimization control.

[0041] This embodiment realizes atmospheric transmission attenuation compensation and thermal humidity interference correction by fusing multi-node light intensity data and regional meteorological parameters, so that the initial dimming parameters accurately match the real lighting demand under complex meteorological conditions. Combined with historical energy consumption characteristics, a periodized dimming model is constructed, and relying on the group cooperation mechanism, the single lamp overcompensation phenomenon is inhibited, which significantly reduces the overall lighting energy consumption while ensuring the lighting quality. Based on real-time lamp response data, composite state calibration and fault feature matching are performed to quickly identify hidden dangers such as drive abnormalities and light decay failures, and predictive maintenance is realized through hierarchical operation and maintenance instructions.

[0042] In one embodiment, in combination with the historical energy consumption records and environmental perception data of intelligent LED lamps, initial dimming parameters are constructed to generate target dimming instructions, including: The built-in electric energy metering chip in each intelligent LED lamp monitors the working voltage, current, and power factor in real time to generate raw energy consumption time series data with a granularity of minutes. The raw data is transmitted to the edge aggregation node through the Modbus-RTU protocol, and the node divides the adjacent lamps into logical groups according to the spatial topology rules: in the road lighting scene, the lamps in the same power supply loop are classified into basic groups; in the building cluster scene, cooperative groups are defined according to floors or functional partitions.

[0043] Periodic data compression is performed within the group: the average power of single lamp is aggregated in units of hours to calculate the cumulative value of group total energy consumption; the peak power characteristic value of the group is extracted in units of days. The integrated historical energy consumption records are stored as a multi-dimensional data table structure, the primary key field includes group identifier, date label, and time period granularity identifier, and the data field covers total energy consumption, average power value, power range, and typical load curve code, forming an energy consumption basic database with spatial and time dimensions.

[0044] The self-configuration register is loaded based on a preset period division rule, and the rule type includes a fixed period division and a dynamic event response mode: the fixed period is divided into a morning and evening transition period, a full night lighting period, and a deep night power reduction period according to the day-night rhythm; the dynamic period responds to holiday events, extreme weather warning signals, and special activity arrangements. The historical energy consumption record input period matching processor performs data slicing according to the time window defined by the rule: taking the fixed period mode as an example, the matching processor retrieves the UTC timestamp of the energy consumption record and attributes it to the corresponding morning / evening / full night period block.

[0045] After slicing, the data is executed for period feature extraction, three core indicators of group total energy consumption average, power fluctuation rate, and load duration ratio in each period are extracted, and a structured period energy consumption dataset is output. The dataset is synchronized to the collaborative processing unit through the time sequence alignment interface, maintaining millisecond-level time sequence synchronization accuracy with real-time environmental data.

[0046] The period energy consumption dataset is input into the space-time alignment module, which establishes a joint index according to the UTC timestamp and group identifier: the index time dimension matching error is controlled within ±1 minute, and the space dimension is associated with the lamp coordinate set through the group topology table. Environmental perception data is mounted to the corresponding group through the sensor identifier, and the associated field includes the illumination average and temperature median of the group geographic center point, and the humidity extreme value. After space-time alignment, heterogeneous data is executed for feature concatenation: the power fluctuation rate in the period energy consumption feature and the environmental temperature median establish a dynamic coupling, the load duration ratio generates a correlation vector with the illumination average, and the total energy consumption average combines with the humidity extreme value to construct a conditional constraint.

[0047] The output energy consumption correlation dataset is a tensor data structure, the tensor row dimension corresponds to the group-period joint primary key, and the column dimension includes twelve associated feature values: six energy consumption-derived features (energy consumption average / fluctuation rate / duration, etc.), six environmental-derived parameters (illumination / temperature / humidity-derived indicators), and explicit association markers between features.

[0048] The energy consumption correlation dataset is loaded into the edge collaborative processor, which calls the collaborative verification rule library according to the group topology table. The initial dimming parameter is input into the group verification engine, which performs three verification operations according to the associated data features: based on the power fluctuation rate and environmental temperature coupling feature, the thermal stability of the initial brightness compensation coefficient is verified, and when the compensation brightness causes the power range to exceed the group safety threshold, dynamic limiting is triggered; the color temperature parameter is corrected in combination with the load duration and illumination correlation vector to suppress the risk of lamp aging acceleration caused by short-time high-intensity dimming; the drive current rise slope is adjusted according to the total energy consumption average and humidity constraint condition to avoid the risk of arc breakdown in high-humidity environments.

[0049] A group balancing mechanism is introduced in the verification process: the dimming parameters of spatially adjacent luminaires are subjected to gradient smoothing processing to eliminate illumination discontinuity; functionally complementary luminaire groups (such as main roads and auxiliary roads) adopt a master-slave following strategy to achieve coordinated power reduction. The output group dimming parameters are a structured group control instruction set, including the independent brightness target value of each luminaire in the group, the color temperature allowable range, the current soft start time window, and the group cascade control identifier.

[0050] The group dimming parameter input instruction compiling unit has a preset rule library containing multi-protocol dimming logic: the DALI protocol encoder converts the brightness target value into a 16-bit address instruction frame; the PWM dimming engine calculates the duty cycle gradient curve based on the current soft start time window; the 0-10V analog output module maps the color temperature range to a voltage control signal. The compiling process performs physical constraint adaptation: the drive current limit is written into the constant current source configuration register; the linkage control identifier triggers the master-slave synchronization instruction of the RS-485 bus.

[0051] The output target dimming instruction is a heterogeneous instruction set, including: digital dimming protocol data packets (DALI / DSI format), analog control voltage waveform parameters, device register configuration code, and bus synchronization trigger pulse four types of control entities. The instruction set is compiled by a hardware description language and written into the drive cache area of the edge gateway, and the group device MAC address list and execution timing label are added to the instruction header to complete the construction of the executable control chain for physical devices.

[0052] This embodiment is based on the dynamic coupling of periodized energy consumption characteristics and real-time environmental data, which drives the group collaborative dimming mechanism to effectively suppress single-lamp overcompensation, while maintaining the uniformity of regional lighting and significantly reducing the overall system energy consumption. The multi-source data fusion mechanism synchronously processes meteorological parameters, historical energy consumption characteristics, and luminaire response characteristics, so that the target dimming instruction automatically adapts to complex working condition environmental changes such as day and night alternation and extreme weather. Through the dynamic verification of thermal stability and color temperature correction strategy, the problem of drive circuit over-limit operation and stroboscopic aging is avoided, and the light decay rate and failure rate of the luminaire are greatly reduced. The hierarchical instruction construction mechanism generates a standardized protocol instruction set, which is compatible with mainstream dimming interfaces to achieve millisecond-level group control response and reduce the frequency of manual intervention. The current soft start logic and humidity constraint condition automatically avoid the risk of electrical breakdown in high-humidity environments, ensuring the safe operation of the lighting system.

[0053] In one embodiment, the initial dimming parameters are group collaboratively verified based on an energy consumption associated data set to generate group dimming parameters, including: The energy consumption association dataset is loaded into the group division engine, which parses the geographic coordinate matrix and functional attribute label in the dataset. The spatial dimension uses the triangulation algorithm to generate the device adjacency topology graph, and sets the spatial distance threshold to automatically aggregate adjacent lamps to form physical groups; the functional dimension matches the attribute label through the preset rule library: the road lighting scene is divided into chain groups according to the lane direction, and the square lighting is defined as ring groups according to the landscape partition. The division process introduces an overlapping group processing mechanism to assign master-slave group identifiers to lamps located at the intersection of multiple regions.

[0054] The output group division result is a structured group relationship table, which contains the master group ID, group type code, member device ID list, spatial boundary coordinates, functional weight coefficient, and priority label in the mixed group scene. The division result is written into the distributed storage unit of the edge node for subsequent collaborative operation calls.

[0055] The initial dimming parameter is input into the dynamic correction processor, which loads the functional weight coefficient and spatial boundary constraint in the group relationship table. Three-level correction operations are performed: the first level calculates the collaborative attenuation factor based on the number of group members, and compresses the maximum brightness limit of single lamp according to the inverse square law; the second level scales the color temperature adjustment range according to the functional weight coefficient, and expands the adjustment bandwidth for high-weight groups; the third level generates a position compensation mapping using the spatial boundary coordinates, and additionally increases the brightness compensation margin for boundary lamps.

[0056] The correction process monitors the environmental temperature change rate in real time, and triggers a dynamic derating strategy when the temperature rise rate exceeds the preset warning value, automatically reducing the upper limit of the brightness threshold. The output group dynamic dimming threshold is a multi-dimensional vector set, and the vector elements include the group reference brightness upper limit, the color temperature allowed interval, the maximum driving current value, and the response delay tolerance window. The vector dimension matches the ID index of the group relationship table.

[0057] The group dynamic dimming threshold is input into the demand matching unit, which synchronously accesses the environmental perception data stream. The matching operation is performed within the group geographic boundary: the illuminance monitoring point data of the area covered by the group centroid point coordinates and radius is extracted, and the area illuminance uniformity index is calculated; the brightness demand satisfaction coefficient is generated by comparing the group reference brightness upper limit and the environmental illuminance median value; the color temperature adaptation parameter is generated by analyzing the color temperature allowed interval and the deviation of the environmental spectral color coordinates.

[0058] The matching process introduces a time decay function to apply a smoothing weighting process to the matching results during the dawn and dusk transition period. The output group lighting demand matching degree is a quantitative evaluation matrix, with rows corresponding to group IDs and columns containing brightness matching coefficients (0-1 dimensionless), color temperature adaptation indexes (percentage), and uniformity compliance markers, as well as additional environmental interference correction markers.

[0059] The group light illumination requirement matching degree matrix is loaded to the optimization engine, the engine calls the preset rule library to perform triple control strategy fusion: the brightness matching coefficient is input to a nonlinear gain adjuster, is mapped to a target brightness percentage according to an exponential curve relationship, and is synchronously superimposed with ambient temperature attenuation compensation; the color temperature adaptation index drives a lookup table module to match a preset color temperature-voltage conversion curve to generate a target color temperature control amount; the uniformity standard mark triggers an edge compensation mechanism to automatically add a brightness gradient correction amount of adjacent lamps to a non-uniform group.

[0060] The optimization process monitors the driving current safety boundary in real time, and when the brightness compensation causes the current prediction value to break through the group dynamic dimming threshold, an adaptive brightness-color temperature replacement algorithm is started to keep the total optical output constant. The output group dimming parameters are a structured control instruction table, and the table fields include four core control variables, namely, the lamp physical address, the target brightness value (0-100% dimensionless), the target color temperature value (Kelvin scale), the maximum allowed current (ampere unit), and the compensation enable mark. The instruction table is stored in the edge gateway instruction queue according to the group ID.

[0061] The embodiment based on the group dynamic division mechanism of the spatial topology eliminates the brightness jump of adjacent lamps, and realizes seamless lighting transition of the road and the building area. The group dynamic dimming threshold fuses meteorological interference correction and thermal attenuation compensation to ensure the stability of the lighting parameters in extreme temperature and humidity environments. The illumination requirement matching degree quantization model drives accurate brightness compensation to avoid the invalid energy consumption caused by excessive dimming in traditional schemes. The current limit parameter and the color temperature boundary value are double-checked to actively avoid the risk of driving circuit overload and LED chip spectrum shift.

[0062] In one embodiment, the intelligent LED lamps are controlled based on target dimming instructions through an edge gateway, and lamp response data is obtained, including: The target dimming instruction is input to an edge gateway protocol interpretation unit, and the unit calls a preset communication protocol mapping relationship library: the digital dimming protocol frame structure (such as the DALI address segment and the brightness instruction bit) is parsed, and is reconstructed into a pulse width modulation waveform feature parameter; the analog dimming instruction voltage amplitude feature is extracted to generate a digital-to-analog conversion control code.

[0063] The interpretation process performs multi-level hardware adaptation: a clock synchronization unit calibrates the timing references of the gateway and the terminal device; a signal integrity detection module verifies the instruction check code and the transmission identifier; and an electrical compatibility component matches different lamp driving chip interface specifications.

[0064] The output device driving signal includes three types of entities, namely, a digital pulse sequence, an analog control amount, and a register configuration code, each of which carries a target device address identifier and an execution timing mark, and the signal output precision meets the industrial error tolerance standard. The driving signal is stored in a dual-port buffer memory for electrical isolation processing.

[0065] The device driving signal is input to a multi-stage scheduling engine, which plans a transmission path according to a spatial topology structure: a trunk road lighting scene adopts a chain routing strategy to define relay nodes and terminal nodes; a building cluster scene is divided into star transmission domains according to the positions of switching devices.

[0066] The driving signal implements intelligent frame processing: a digital signal is packaged into a low-power wireless communication protocol frame structure, and a payload area carries a brightness control parameter and a gradual change timing; an analog signal is encoded into an industrial real-time Ethernet protocol data unit, and a device logical address is added to a header.

[0067] A dynamic bandwidth allocation strategy is adopted in a transmission process: an emergency control instruction occupies main bandwidth resources preferentially; a regular instruction is sent based on a channel quality detection mechanism. A real-time control queue forms a space-time two-dimensional index structure: a time dimension is divided into execution cycles according to a preset time window; a space dimension records a target device address sequence, a signal format identifier, and a data check feature in each cycle. A forward error correction mechanism is embedded in a queue header to guarantee transmission reliability.

[0068] The real-time control queue is input to an edge monitoring engine, which initiates a state polling according to a device address sequence in the queue: a gateway sends a device self-check instruction to a target lamp through an industrial real-time bus, and the instruction triggers a lamp internal diagnosis circuit to collect driving chip working state register data. Running data capture covers multiple physical dimensions: an electric current detection unit feeds back an actual output current ripple and an effective value; a temperature sensing circuit returns an LED chip junction temperature and a heat dissipation substrate temperature rise curve; a light feedback module outputs an actual luminous intensity and a spectral distribution feature; and a voltage monitoring unit records power supply fluctuation characteristics.

[0069] Data flow is marked with a time stamp and bound to a device address to form an original running data stream. A priority control strategy is implemented in a capture process: a high-speed sampling mode is enabled for a lamp involved in a real-time dimming instruction, and a regular monitoring point adopts a timing polling mechanism. The original data packet is submitted to a data preprocessing channel to perform transmission noise filtering and range normalization processing.

[0070] The preprocessed running data is input to a state calibration engine, which calls a preset fault feature rule library to perform parallel analysis: an electric current analysis module extracts a ripple coefficient and an effective value offset feature, and combines a rated current range to calibrate an overload risk level; a temperature analysis unit identifies heat dissipation abnormalities according to a junction temperature-substrate temperature rise gradient, and refers to a thermal resistance model to mark a cooling failure index; a light output analysis component compares a target brightness and an actual intensity deviation, and determines a light decay fault category according to a decay rate model; and a voltage fluctuation analysis unit detects a sudden drop / surge event feature, and associates historical records to calibrate a power supply abnormal type.

[0071] The calibration process introduces a composite decision mechanism: activate the priority arbitration strategy when multiple abnormal features are concurrent; mark the intermittent fault as an observed state. The output luminaire response data is a structured diagnostic report, and the report field includes the device physical address, running state code (normal / warning / fault), abnormal type label, key parameter deviation value, and recommended treatment code. The report tail adds feature data fingerprints for traceability verification.

[0072] The real-time control queue input edge monitoring engine initiates state polling based on the device address sequence in the queue: the gateway sends a device self-check instruction to the target luminaire through the industrial real-time bus, which triggers the luminaire internal diagnostic circuit to collect the driving chip working state register data. The running data capture covers multiple physical dimensions: the current detection unit feeds back the actual output current ripple and effective value; the temperature sensing circuit returns the LED chip junction temperature and heat dissipation substrate temperature rise curve; the light feedback module outputs the actual luminous intensity and spectral distribution characteristics; the voltage monitoring unit records the power supply fluctuation characteristics.

[0073] The data stream is time-stamped and bound to the device address, forming the original running data stream. The capture process implements a priority control strategy, with the luminaire enabled in high-speed sampling mode for real-time dimming instructions, and a regular monitoring point using a timed polling mechanism. The original data packet is submitted to the data preprocessing channel for transmission noise filtering and range normalization processing.

[0074] The preprocessed running data is input into the state calibration engine, which calls the pre-set fault feature rule library for parallel analysis: the current analysis module extracts the ripple coefficient and effective value offset feature, and combines the rated current range to calibrate the overload risk level; the temperature analysis unit identifies the heat dissipation abnormality based on the junction temperature- substrate temperature rise gradient, and refers to the thermal resistance model to mark the cooling failure index; the light output analysis component compares the target brightness with the actual intensity deviation, and determines the light decay fault category based on the decay rate model; the voltage fluctuation analysis unit detects the surge event features, and correlates the historical records to calibrate the power supply abnormal type. The calibration process introduces a composite decision mechanism: activate the priority arbitration strategy when multiple abnormal features are concurrent; mark the intermittent fault as an observed state.

[0075] The output luminaire response data is structured data, including the specific luminaire device physical address, running state code (normal / warning / fault), abnormal type label, key parameter deviation value, and recommended treatment code. The report tail adds feature data fingerprints for traceability verification.

[0076] The embodiment realizes accurate conversion of digital instructions to device driving signals through the multi-protocol interpretation capability of the edge gateway, adapts to various industrial lighting interface standards, and greatly reduces the hardware modification cost of traditional control systems. The hierarchical scheduling strategy generates a real-time control queue optimized in space and time in combination with the spatial topology of the lamps, effectively eliminates signal conflicts and transmission delays in a large-scale networking environment, and guarantees millisecond-level synchronous execution of regional lighting control instructions. The operation monitoring mechanism captures multi-dimensional real-time data such as current, temperature, light intensity, and voltage, and establishes a panoramic device state view covering electrical characteristics and optical performance. The composite state calibration technology prioritizes abnormal features through a fault rule library, accurately identifies the hazard level of driving overload, heat dissipation failure, light decay, and other hidden dangers, and provides an operable fault tracing basis for predictive maintenance.

[0077] In one embodiment, device driving signals are sent to each intelligent LED lamp through the edge gateway, and hierarchical scheduling is performed to generate a real-time control queue, including: This process realizes the spatial topology registration and communication parameter definition of network devices. The edge gateway calls the dynamic node discovery protocol to broadcast topology detection frames to all intelligent LED lamps in the networking range. The built-in communication coprocessor of the lamp responds to the detection frame and returns four core parameters: device physical address, signal reception strength indicator, parent node path loss coefficient, and hardware interface type.

[0078] The gateway performs hierarchical node calibration according to spatial position constraints and signal quality thresholds: in the main road area, the lamp with the lowest path loss is selected as the regional routing relay node; in the building cluster, the lamp with dual network ports is selected as the subnetwork aggregation point. The calibration process detects the channel interference index in real time and automatically adds a backup routing path to high-interference areas.

[0079] The output node communication scheduling table is a multi-dimensional data matrix, with the matrix row index corresponding to the device physical address and the column parameters including: node level mark (terminal / relay / aggregation), transmission time slot allocation proportion, maximum allowed transmission delay, communication protocol type identification, and parent node address mapping relationship. The scheduling table is dynamically refreshed every 5 minutes and stored in the non-volatile storage area of the edge gateway.

[0080] The device driving signal input signal slicing engine analyzes the protocol type identification in the scheduling table: the DALI digital signal is divided into 128-byte payload units after 8:1 data compression, and a relay hop count field is added; the 0-10V analog signal is converted to a digital sequence using incremental encoding modulation and divided into 16-byte data blocks.

[0081] The encapsulation process implements a three-layer detection mechanism: the physical layer detects and verifies that the signal amplitude meets the target lamp interface voltage range; the data link layer adds a cyclic redundancy check code and a sequence number identifier; and the network layer marks the target node level and relay path.

[0082] The output fragment driving data packet has a composite data structure, and the packet header field is fixed at 32 bytes, including a source address, a target address, and a total fragment number mark; the payload field includes a signal type label (digital / analog / register configuration) and a fragment data entity; and the check field includes a 16-bit error correction code and a channel state fingerprint. When the data packet is written into the transmission buffer queue, it is automatically matched with the maximum transmission delay parameter defined by the scheduling table.

[0083] The node communication scheduling table and the fragment driving data packet are input into the wideband allocation engine, and the engine analyzes the target node level mark and the relay hop number parameter in the packet header field.

[0084] Based on the transmission time slot allocation ratio and the maximum delay constraint recorded by the scheduling table, the interface bandwidth dynamic splitting operation is performed: the gateway physical interface is virtualized into multiple logical channels, a high-priority channel is allocated to the backbone routing node with a fixed bandwidth quota, and a polling variable bandwidth slot is allocated to the terminal node.

[0085] The allocation process implements a two-level load balancing strategy: the first level dynamically adjusts the physical layer symbol transmission rate according to the channel state fingerprint; and the second level adaptively allocates the error correction resource proportion according to the length of the fragment data packet check field. The output signal transmission queue is a space-time composite structure, the time axis is divided into a sequence of continuous transmission windows, the space axis records the communication interface identifier, the bandwidth proportion value, the data transmission mode identifier and the delay tolerance threshold allocated in each window, and the tail of the queue is marked with the gateway resource occupation state for real-time rescheduling decision.

[0086] The signal transmission queue is input into the response controller, which activates the response strategy according to the data transmission mode identifier in the queue: for the fixed bandwidth channel, a hardware flow control mechanism is used to generate a strict time sequence control sequence based on clock synchronization; and for the variable bandwidth slot, an adaptive response protocol is enabled to automatically compensate for transmission jitter according to the lamp signal reception strength.

[0087] The response control process implements three-dimensional optimization: the spatial dimension constructs a star / tree response path topology according to the node level mark; the time dimension aligns the transmission window boundary to generate a microsecond-level trigger pulse; and the protocol dimension analyzes the data packet check field to generate a response retransmission rule library.

[0088] The output real-time control queue integrates control and monitoring dual-channel instructions: the control channel records the physical transmission path of the fragment data packet and the execution timestamp; the monitoring channel reserves a lamp response data collection time slot, and the time slot length matches the maximum delay parameter defined by the scheduling table. The queue header is embedded with a topology self-healing protocol code, which automatically rebuilds a backup transmission path when the network link is interrupted.

[0089] The embodiment generates a communication scheduling table of adaptive topology through dynamic node calibration, effectively overcomes the networking challenge brought by heterogeneous distribution of lamps in complex lighting scenes, and significantly improves the wireless signal coverage of large-scale lighting clusters. Signal fragmentation packaging combined with hierarchical detection technology realizes decoupled transmission of multi-protocol driven instructions, eliminates data conflict and packet loss risk under traditional centralized control, and guarantees the end-to-end transmission reliability of key dimming instructions. The intelligent bandwidth allocation strategy based on node attributes accurately divides the gateway interface resources, so that the backbone routing nodes obtain deterministic transmission delay, and ensure the synchronization execution of regional dimming instructions within hundreds of milliseconds.

[0090] In one embodiment, the intelligent LED lamps are continuously determined according to the lamp response data, fault determination information is obtained, and operation and maintenance control instructions are triggered, including: Through the fault diagnosis rule, the running state code and key parameter deviation value field in the lamp response data are analyzed. The diagnosis process calls the matching rule of the fault type database: the current ripple coefficient exceeding the limit and accompanied by voltage drop mark "abnormal driving power supply"; the LED chip junction temperature gradient exceeding the safety threshold and the light intensity output decay persistently determine "heat dissipation system failure"; the driving current effective value deviates from the rated range and superimposes the light spectrum color coordinate offset triggers "LED chip light decay" alarm; the voltage fluctuation characteristics conform to the transient pulse model and there is no accompanying temperature anomaly marked "power surge event".

[0091] The analysis process implements a double confirmation mechanism: the primary diagnosis conclusion is executed to perform historical fault feature backtracking, and when the same type of fault occurs more than the trigger value within the preset observation period, the alarm confidence level is improved; the lamp fault information is output as a structured diagnosis report, and the report field includes the device physical address, fault classification code, confidence weight value, feature parameter snapshot, and suggested diagnosis review mark.

[0092] The lamp fault information is input into the energy consumption correlator, the correlator extracts the device physical address and time stamp mark in the fault information, and indexes the energy consumption feature data of the corresponding device in the fault window period in the historical energy consumption record. Multi-dimensional anomaly detection is performed: the average power change rate in the period before the fault occurs is calculated, and when the change rate exceeds the allowed fluctuation range defined by the fault type database, the "energy consumption trend anomaly" mark is generated; the power factor difference between the fault period and the historical same period is compared, and when the deviation exceeds the preset threshold, the "power quality degradation" alarm is marked; the energy consumption distribution characteristic variation of the fault device in the power peak and valley period is identified, and the "load mode deviation" index is generated.

[0093] The energy consumption anomaly information is output as an extended diagnosis matrix, the matrix row inherits the lamp fault information field, and the new column dimension includes the average power change deviation mark (percentage), power factor degradation index (dimensionless), load mode deviation coefficient (0-1 scale), and associated confidence label.

[0094] The luminaire fault information and energy consumption abnormal information input state determinator determines the correlation mapping relationship between the fault classification code and the energy consumption abnormality mark: for the equipment with both "drive power abnormality" fault mark and "power quality degradation" alarm, the determination confidence weight is promoted to the highest level; when "heat dissipation system failure" fault is accompanied by "load mode deviation coefficient" continuously exceeding the standard, the composite fault mark is activated.

[0095] The determination process applies a preset hierarchical threshold mechanism: three-level determination threshold (warning / fault / emergency) is adopted for electrical faults, two-level determination standard (degradation / failure) is set for optical faults, and temperature-time integral determination model is implemented for thermal faults. The output fault determination information is an enhanced diagnostic report, which adds state determination code (0-9 severity level), disposal time limit mark (minute level / hour level / day level), associated equipment impact range list, and determination logic trace fingerprint for audit verification on the basis of original fault information field.

[0096] The fault determination information input operation and maintenance instruction compiler calls the rule mapping table of the operation and maintenance instruction library: the state determination code is matched with the pre-stored disposal template (such as fault code A01 is mapped to drive board reset instruction), and the disposal time limit mark triggers the corresponding response protocol channel. The construction process performs priority arbitration: emergency electrical fault automatically generates device power-off protection code; optical degradation alarm is compiled into a gradual dimming and load reduction sequence; heat failure triggers heat dissipation fan enhanced control pulse. The output operation and maintenance control instruction contains four operation entities: hardware operation code (register read-write instruction), device control parameter (voltage / current set value), linkage strategy identification (influencing device cooperative action), and audit log record item, all of which are attached with time limit validity label.

[0097] The luminaire fault information and energy consumption abnormal information input state determinator determines the correlation mapping relationship between the fault classification code and the energy consumption abnormality mark: for the equipment with both "drive power abnormality" fault mark and "power quality degradation" alarm, the determination confidence weight is promoted to the highest level; when "heat dissipation system failure" fault is accompanied by "load mode deviation coefficient" continuously exceeding the standard, the composite fault mark is activated.

[0098] The determination process applies a preset hierarchical threshold mechanism: three-level determination threshold (warning / fault / emergency) is adopted for electrical faults, two-level determination standard (degradation / failure) is set for optical faults, and temperature-time integral determination model is implemented for thermal faults. The output fault determination information is an enhanced diagnostic report, which adds state determination code (0-9 severity level), disposal time limit mark (minute level / hour level / day level), associated equipment impact range list, and determination logic trace fingerprint for audit verification.

[0099] The fault determination information is input into an operation and maintenance instruction compiler, the compiler calls a rule mapping table of an operation and maintenance instruction library: the state determination code is matched with a pre-stored treatment template (for example, the fault code A01 is mapped to a drive board reset instruction), and a treatment time limit label triggers a corresponding response protocol channel. A process execution priority arbitration is constructed: an emergency electrical fault automatically generates a device power-off protection code; an optical decay type alarm is compiled into a gradual dimming load reduction sequence; and a heat fault triggers a heat dissipation fan enhanced control pulse.

[0100] The output operation and maintenance control instruction includes four operation entities: a hardware operation code (a register read-write instruction), a device control parameter (a voltage / current set value), a linkage strategy identifier (a device cooperative action), and an audit log record item, and all instruction units are attached with a time limit validity label.

[0101] Compared with a traditional threshold alarm mode, the embodiment significantly reduces a false alarm rate through rule matching of a fault type database and a lamp response data. An energy consumption correlation analysis technology fuses historical load characteristics and real-time fault parameters, effectively identifies potential hidden dangers such as drive power aging and heat dissipation efficiency degradation, and makes a predictive maintenance decision have multi-dimensional data support. A double-channel operation and maintenance instruction construction mechanism combines an emergency degree weight division strategy, provides a hardware level real-time blocking capability for high-risk faults such as short circuit power-off, simultaneously reasonably schedules non-emergency tasks such as lamp light decay correction to a low power consumption period for execution, and maximally reduces the influence of operation and maintenance actions on lighting continuity.

[0102] In one embodiment, based on the fault determination information, a preset operation and maintenance instruction library is triggered to perform operation and maintenance construction, to obtain an operation and maintenance control instruction, including: The fault determination information is input into an operation and maintenance instruction mapping engine, the engine analyzes a state determination code and a treatment time limit label field, and retrieves a pre-stored fault type-instruction mapping tree in the operation and maintenance instruction library. The mapping rule is realized based on multi-level indexes: a first-level index matches a fault classification code (for example, "E01" represents a drive power short circuit), a second-level index is associated with a state determination severity level (1-9 levels), and a third-level index is bound to a treatment time limit label (minute / hour / day level).

[0103] A dynamic template adaptation mechanism is implemented in the matching process: when a new type of fault code is encountered, a transition instruction is generated by automatically retrieving an existing instruction template with a similarity exceeding a preset threshold, and a manual review label is added. The output initial operation and maintenance instruction set is a structured operation set, each instruction includes five types of basic operation units: device positioning coordinates, target operation type code (such as reset / power-off / load reduction), hardware control parameter threshold range, allowed retry number, and associated device linkage identifier. A unique traceability fingerprint is added to the operation unit for version control, and an abnormal matching event triggers an instruction library incremental update mechanism.

[0104] The initial operation and maintenance instruction set is input into the time efficiency decision maker, and the decision maker loads the response time efficiency rule matrix built in the operation and maintenance instruction library: the row dimension of the matrix defines the fault scene (such as electrical fire risk), and the column dimension sets the geographical weight coefficient. The decision logic adopts a three-dimensional criterion: the time dimension forcibly classifies the minute-level response demand into the real-time disposal queue; the safety dimension implements a one-vote veto upgrade mechanism for high-risk scenes such as short circuit and ignition; and the space dimension adjusts the response level boundary value according to the geographical weight coefficient.

[0105] The division process introduces a device state buffer mechanism: offline devices automatically reduce the response priority and increase the standby disposal window. The output first operation and maintenance task queue (real-time) contains hardware-level operation instructions that need to be responded within 300 seconds, and the second operation and maintenance task queue (delayed) integrates performance optimization instructions that allow delayed disposal. The double queues each carry environmental constraint labels: the real-time queue is marked with the maximum allowed operation delay, and the delayed queue is bound to the best execution time window parameter.

[0106] The first operation and maintenance task queue is input into the real-time scheduling engine, which calls the hardware resource pre-allocation strategy: a dedicated communication channel bandwidth is reserved for short-circuit power-off instructions; and a interruptible execution time slot is allocated for dimming and load reduction instructions. The optimization process implements a three-fold reinforcement mechanism: electrical safety instructions are embedded in hardware watchdog circuits to achieve microsecond-level response; a three-fold redundant retransmission protocol is automatically triggered when data verification fails; and high-priority instructions are forced to execute within a time window.

[0107] The output first operation and maintenance instruction sequence is a timestamp-labeled atomic operation chain, and each instruction in the sequence contains four types of elements: operation type identifier (pass-through type / interrupt type), hardware interface address, upper limit of execution time, and failure rollback path encoding, and the key operation unit is additionally added with electrostatic discharge protection instruction code.

[0108] The second operation and maintenance task queue is input into the periodic scheduler, which loads the electricity price period database and device geographical clustering rules: light decay correction instructions are bound to the valley period electricity price execution period; and heat maintenance tasks are implemented in batches according to the device location coordinates.

[0109] The optimization process implements dynamic constraint detection: when multiple devices operate in parallel, the peak power distribution load is predicted, and when the load exceeds the safety threshold, the task package is automatically split; and the offline device maintenance task is additionally added with a standby power supply activation instruction.

[0110] The output second operation and maintenance instruction sequence is a spatiotemporal coupled task package set, and the task package structure contains device cluster geographical hash value, maintenance time window (start UTC timestamp + time length), maximum number of parallel devices allowed, and backup energy switching flag bit, and the task package header is added with an energy efficiency optimization label for energy consumption audit tracking.

[0111] The first operation and maintenance instruction sequence and the second operation and maintenance instruction sequence are input into a dynamic merging engine, and the engine establishes a three-dimensional scheduling model according to an execution time line and a space coordinate. Real-time instructions are given the highest time sequence priority, and a dedicated transmission channel is locked on a hardware control bus, so that an emergency power-off operation is ensured to preempt execution resources in an interruption-free manner; periodic operation and maintenance tasks are executed in transmission gaps of the real-time instructions, and an idle time slot detection mechanism is used to automatically capture a time window margin.

[0112] Space dimension implementation area mutual exclusion lock control is used to generate a sequential execution marker chain for spatially overlapped operation and maintenance actions by using a geographic grid algorithm, so as to avoid operation and maintenance team job conflicts. The merging process integrates a double safety mechanism: a strong electric control instruction implants a hardware level anti-misoperation trigger pulse sequence, so as to ensure that non-target devices are not disturbed. A light school correction type instruction superimposes an anti-dazzling gradual change protocol, so as to constrain the brightness mutation amplitude to be less than a human eye adaptation threshold.

[0113] The output operation and maintenance control instruction is a binary executable code stream, and the code stream adopts a hierarchical encapsulation structure: an outer layer header carries a space-time constraint matrix and a failure safety rollback pointer; an inner layer control frame contains four types of core control quantities, namely, a device physical address index, an operation instruction code, a driving voltage / current setting value and a maximum operation time length parameter. A code stream tail integrates a password hash check block and an electrostatic protection code, and is directly driven to an execution unit via a hardware instruction decoder of an edge gateway, so as to realize millisecond-level accurate response and fault safety isolation of operation and maintenance actions.

[0114] The present application significantly improves the diagnostic accuracy of typical faults such as driving circuit abnormalities and light decay failures through a multi-level matching mechanism of a fault rule library and device responses, effectively avoiding false positives caused by single threshold determination in traditional schemes. The double-channel task queue architecture realizes intelligent grading of operation and maintenance instructions according to response time characteristics, provides hardware-level real-time blocking capability for emergency electrical faults, and reasonably schedules non-critical maintenance tasks to low load periods for execution, thereby maximizing the continuity of lighting services. The hardware acceleration strategy of the real-time instruction sequence ensures that high-risk operations obtain priority execution resources, and the space mutual exclusion control mechanism completely avoids on-site maintenance job conflicts.

[0115] Referring to Figure 2 The present application also provides an LED intelligent lighting system control device based on the Internet of Things, which is applied to the LED intelligent lighting system control method based on the Internet of Things in any of the above aspects, and comprises: A recognition module is configured to collect environment perception data of a plurality of intelligent LED lamps and perform dimming calculation to obtain initial dimming parameters. An analysis module is configured to combine historical energy consumption records and environment perception data of the intelligent LED lamps to perform instruction construction on the initial dimming parameters and generate target dimming instructions. The processing module is configured to control the intelligent LED lamp through the edge gateway based on the target dimming instruction, and obtain lamp response data; The constructing module is configured to continuously determine the intelligent LED lamp according to the lamp response data, obtain fault determination information, and trigger operation and maintenance control instructions.

[0116] The application provides an LED intelligent lighting control device based on the Internet of Things, which optimizes the dimming parameters by fusing environmental perception data and historical energy consumption records, overcomes the limitations of traditional lighting control relying only on real-time environmental data, makes the dimming strategy meet the current environmental demand and comply with the long-term energy efficiency optimization goal, and significantly improves the energy saving effect of the lighting system. The real-time control architecture based on the edge gateway combines the dynamic monitoring of the lamp response data, realizes accurate determination of the operation state of the intelligent LED lamp, can timely discover potential faults and trigger automatic operation and maintenance, greatly reduces the artificial inspection cost, and improves the system reliability. By organically combining environmental perception, energy consumption analysis and fault warning, a complete intelligent lighting control closed loop is constructed, the self-adaptive ability of the system to different application scenarios is enhanced, the lamp service life is prolonged through accurate operation and maintenance, and reliable technical support is provided for intelligent management and efficient operation of the intelligent lighting system.

[0117] It should be noted that, for the convenience and brevity of description, the specific working processes of the system and each module described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0118] The above only describes the preferred embodiments of the application, and does not limit the patent scope of the application, and any equivalent structure or equivalent process transformation based on the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the application.

Claims

1. An Internet of Things-based LED intelligent lighting control method, characterized in that, The method comprises the following steps: Collecting environmental sensing data of a plurality of intelligent LED lamps and performing dimming calculation to obtain initial dimming parameters; Combining historical energy consumption records and environmental sensing data of the intelligent LED lamps, constructing instructions for the initial dimming parameters, and generating target dimming instructions; Controlling the intelligent LED lamps based on the target dimming instructions through an edge gateway and obtaining lamp response data; Continuously determining the intelligent LED lamps according to the lamp response data to obtain fault determination information and triggering operation and maintenance control instructions. 2.The IoT-based LED smart lighting system control method of claim 1, wherein, The method comprises the following steps: Collecting sensing data of a plurality of intelligent LED lamps to obtain light intensity and node temperature and humidity data; Obtaining regional meteorological information according to the location information of a plurality of intelligent LED lamps; Comparing the light intensity with a preset light intensity threshold value to generate a reference brightness compensation coefficient; Compensating the node temperature and humidity data based on the regional meteorological information to obtain meteorological interference parameters; Joint dimming of the meteorological interference parameters and the reference brightness compensation coefficient to generate the initial dimming parameters. 3.The IoT-based LED smart lighting system control method of claim 1, wherein, The method comprises the following steps: Obtaining energy consumption data of a plurality of intelligent LED lamps, periodically integrating the energy consumption of groups to form the historical energy consumption records; Matching the historical energy consumption records according to a preset time period division rule to obtain energy consumption data of each time period; Associating and mapping the energy consumption feature data of each time period with the environmental sensing data to generate an energy consumption association data set; Group collaborative verification of the initial dimming parameters based on the energy consumption association data set to generate group dimming parameters; Constructing instructions for the group dimming parameters through a preset dimming rule to generate the target dimming instructions. 4.The IoT-based LED smart lighting system control method of claim 3, wherein, The method comprises the following steps: Dividing the energy consumption association data set into groups to obtain group division results; Combining the group division results to dynamically correct the initial dimming parameters to generate group dynamic dimming thresholds; Matching the group dynamic dimming thresholds with the environmental sensing data to generate group light demand matching degrees; Collaborative optimization of the initial dimming parameters based on the group light demand matching degrees to generate the group dimming parameters. 5.The IoT-based LED smart lighting system control method of claim 1, wherein, The method comprises the following steps: Interpreting the target dimming instructions through an edge communication protocol to generate device driving signals; Sending the device driving signals to each intelligent LED lamp through the edge gateway and performing hierarchical scheduling to generate a real-time control queue; Performing operation monitoring on the intelligent LED lamps based on the real-time control queue through the edge gateway to capture lamp operation data; The lamp operation data is compound state calibrated to generate lamp response data. 6.The IoT-based LED smart lighting system control method of claim 5, wherein, The device driving signal is sent to each smart LED lamp through the edge gateway, and hierarchical scheduling is performed to generate a real-time control queue, including: Transmission node calibration of each smart LED lamp is performed through the edge gateway to generate a node communication scheduling table; Based on the node communication scheduling table, signal fragmentation packaging and hierarchical detection are performed on the device driving signal to generate fragmented driving data packets; Interface wideband allocation of the edge gateway is performed through the node communication scheduling table and the fragmented driving data packets to obtain a signal transmission queue; Response queue control of the smart LED lamp is performed according to the signal transmission queue to obtain the real-time control queue. 7.The IoT-based LED smart lighting system control method of claim 1, wherein, The smart LED lamp is continuously determined according to the lamp response data to obtain fault determination information, and an operation and maintenance control instruction is triggered, including: Fault state analysis of each smart LED lamp is performed through the lamp response data according to a preset fault type database to obtain lamp fault information; Energy consumption anomaly information is obtained by associatively analyzing the lamp fault information and the historical energy consumption record; State determination of the smart LED lamp is performed in combination of the lamp fault information and the energy consumption anomaly information to obtain the fault determination information; Based on the fault determination information, a preset operation and maintenance instruction library is triggered for operation and maintenance construction to obtain the operation and maintenance control instruction. 8.The IoT-based LED smart lighting system control method of claim 7, wherein, Based on the fault determination information, a preset operation and maintenance instruction library is triggered for operation and maintenance construction to obtain the operation and maintenance control instruction, including: An initial operation and maintenance instruction set is generated by matching the fault determination information with the operation and maintenance instruction library according to the response time limit rules of the operation and maintenance instruction library; The emergency degree of the initial operation and maintenance instruction set is divided according to the response time limit rules of the operation and maintenance instruction library to generate a first operation and maintenance task queue and a second operation and maintenance task queue, wherein the first operation and maintenance task queue contains real-time operation and maintenance instructions, and the second operation and maintenance task queue contains delayed operation and maintenance instructions; An initial operation and maintenance instruction set is generated by matching the fault determination information with the operation and maintenance instruction library according to the response time limit rules of the operation and maintenance instruction library; An initial operation and maintenance instruction set is generated by matching the fault determination information with the operation and maintenance instruction library according to the response time limit rules of the operation and maintenance instruction library; The first operation and maintenance instruction sequence and the second operation and maintenance instruction sequence are combined in priority according to the execution of the instructions to obtain the operation and maintenance control instruction.

9. An Internet of Things-based LED intelligent lighting system control device, characterized by, The LED intelligent lighting system control method based on the Internet of Things is applied to any one of the above claims 1-8, including: An identification module is used to collect environmental perception data of a plurality of smart LED lamps and perform dimming calculation to obtain initial dimming parameters; An analysis module is used to combine the historical energy consumption record and environmental perception data of the smart LED lamp to perform instruction construction on the initial dimming parameters to generate target dimming instructions; A processing module is used to control the smart LED lamp through the edge gateway based on the target dimming instructions and obtain lamp response data; A building module is configured to continuously determine the smart LED lamp according to the lamp response data, obtain fault determination information, and trigger operation and maintenance control instructions.

Images