A multi-node illumination control method based on dynamic feedback weighted fusion

Abstract

The application discloses a multi-node illumination control method based on dynamic feedback weighted fusion, and belongs to the technical field of illumination control. The method obtains node illumination values, historical illumination values and historical confidence of a plurality of illumination collection nodes, calculates reference illumination values, node current deviations and node change rates, and generates an overall environment change identifier according to the consistency of the plurality of node change rates. The current confidence is updated in combination with the node current deviation, the node change rate, the historical confidence and the overall environment change identifier. When the current confidence is lower than a failure threshold, a node weight reduction result is generated, and a fusion weight is generated. An environment fusion illumination value is obtained according to the fusion weight, a PWM dimming signal is output according to the illumination error between the environment fusion illumination value and a target illumination value, so that the influence of a local abnormal node on dimming control is reduced.

Description

Technical Field

[0001] This application belongs to the field of lighting control technology, specifically a multi-node lighting control method based on dynamic feedback weighted fusion. Background Technology

[0002] With the development of LED lighting, wireless communication, and embedded control technologies, intelligent lighting control systems are gradually evolving from single-point illuminance acquisition and single-source light source adjustment to multi-node sensing and networked control. Existing intelligent lighting systems typically acquire ambient illuminance through light sensors and adjust the LED light source brightness based on the difference between the acquired value and the target illuminance value. To improve acquisition stability, some solutions deploy multiple light sensors within the target environment and employ average filtering, median filtering, Kalman filtering, or fixed-weight fusion methods to obtain the ambient illuminance value used for dimming control.

[0003] However, existing multi-node illumination control schemes mostly rely on current sampled values ​​or fixed fusion rules as the basis for judging ambient illuminance, making it difficult to distinguish between local node anomalies and the actual changes in the overall environment. Specifically, when a light acquisition node is blocked, exposed to strong local light, or experiences a sampling anomaly, the acquired value of that node will deviate significantly from other nodes. If this abnormal acquired value is still used in the fusion process according to the average value or a fixed weight, it will skew the final ambient illuminance judgment, leading to misadjustment of the LED light source. Conversely, when the target environment is affected by cloud cover outside the window, overall movement of curtains, or changes in ambient lighting, multiple light acquisition nodes may rise or fall simultaneously. If the control system judges anomalies based solely on the difference between a single node and other nodes, it may misjudge the actual overall illumination change as a node anomaly, resulting in a delayed dimming response.

[0004] Meanwhile, existing multi-node lighting control schemes typically lack continuous evaluation of the historical operating status and continuously changing status of nodes. For acquisition nodes that are affected by local shading, local reflections, installation angle misalignment, or communication instability for a long time, abnormal data may continue to participate in the fusion calculation over multiple sampling periods, causing the ambient illuminance value to deviate from the true illuminance level of the target environment for a long period, thereby reducing the stability of the lighting closed-loop control. Therefore, it is necessary to provide a multi-node lighting control method that can reduce the impact of local abnormal nodes on the ambient illuminance fusion results in multi-node sampling scenarios and avoid misjudging overall illumination changes as node abnormalities. Summary of the Invention

[0005] To address the problems of existing multi-node illumination control schemes, such as difficulty in accurately distinguishing between local node anomalies and real changes in the overall environment, the tendency of abnormal nodes to participate in fusion calculations for extended periods and thus skew the judgment of ambient illumination, and the potential reduction in closed-loop dimming stability due to mutual blocking of data reception, communication, display refresh, and PWM output during multi-task control, this application provides a multi-node illumination control method based on dynamic feedback weighted fusion.

[0006] To achieve the above objectives, this application provides a multi-node illumination control method based on dynamic feedback weighted fusion, comprising: Obtain the node illuminance values ​​uploaded by multiple light acquisition nodes in the current sampling period, as well as the historical illuminance values ​​and historical reliability of each light acquisition node in the previous sampling period, and calculate a reference illuminance value based on the multiple node illuminance values; The current deviation of a node is obtained based on the difference between the node illuminance value and the reference illuminance value; the node change rate is obtained based on the change between the node illuminance value and the historical illuminance value; and the overall environmental change indicator is obtained based on the consistency of multiple node change rates. The current credibility of each light acquisition node is updated based on the current deviation of the node, the node change rate, the historical credibility, and the overall environmental change identifier. When the current credibility is lower than the failure threshold, a node weight reduction result is generated, and a fusion weight is generated based on the current credibility and the node weight reduction result; The illuminance values ​​of multiple nodes are weighted and fused according to the fusion weight to obtain the environmental fusion illuminance value; The light source adjustment control quantity is generated based on the illuminance error between the environmental fusion illuminance value and the target illuminance value, and a PWM dimming signal is output based on the light source adjustment control quantity. The node illuminance value is updated to the historical illuminance value of the next sampling period, and the current confidence level is updated to the historical confidence level of the next sampling period.

[0007] In one possible implementation, acquiring the node illuminance values ​​uploaded by multiple light acquisition nodes in the current sampling period includes: each light acquisition node acquiring the ambient illuminance at its location through a light sensor, combining the acquired illuminance data with a node number, a sampling period number, a transmission time identifier, and a verification field to form a node sampling data packet, and transmitting it to a central control terminal via a wireless communication module; the central control terminal reading the sampling period number, transmission time identifier, and verification field from the node sampling data packet, determining the difference between the current system reception time and the transmission time identifier as the data transmission delay; when the sampling period number is consistent with the current sampling period, the data transmission delay is not greater than a preset reception time window, and the verification field passes the verification, the corresponding illuminance data is read as the node illuminance value.

[0008] In one possible implementation, calculating a reference illuminance value based on multiple node illuminance values ​​includes: reading multiple node illuminance values ​​successfully received within the current sampling period, forming an illuminance sequence according to node numbers, and performing averaging on the node illuminance values ​​in the illuminance sequence to obtain a reference illuminance value; or, after removing the maximum and minimum node illuminance values ​​from the illuminance sequence, performing averaging on the remaining node illuminance values ​​to obtain a reference illuminance value.

[0009] In one possible implementation, obtaining an overall environmental change identifier based on the consistency of the change rates of multiple nodes includes: comparing the node illuminance value of each illumination acquisition node with historical illuminance values ​​to obtain a change direction marker for the corresponding illumination acquisition node; counting the number of illumination acquisition nodes with an upward change direction marker and a node change rate not less than a preset change rate threshold, and the number of illumination acquisition nodes with a downward change direction marker and a node change rate not less than the preset change rate threshold, and determining the larger of the two as the number of synchronously changing nodes; generating an overall environmental change identifier indicating an overall environmental change when the ratio of the number of synchronously changing nodes to the number of successfully received nodes is not less than a preset synchronization ratio threshold; otherwise, generating an overall environmental change identifier indicating a local node change.

[0010] In one possible implementation, updating the current reliability of each illumination acquisition node based on the current node deviation, the node change rate, the historical reliability, and the overall environmental change identifier includes: generating a node stability score for each illumination acquisition node according to the rule that an increase in the current node deviation leads to a decrease in the node stability score, and an increase in the node change rate leads to a decrease in the node stability score; selecting a change rate adjustment coefficient based on the overall environmental change identifier; and performing a weighted fusion process on the historical reliability, the node stability score, and the change rate adjustment coefficient to obtain the current reliability.

[0011] In one possible implementation, when the overall environmental change indicator represents an overall environmental change, a first change rate adjustment coefficient is selected from a preset change rate adjustment coefficient table; when the overall environmental change indicator represents a local node change, a second change rate adjustment coefficient is selected from the preset change rate adjustment coefficient table, wherein the second change rate adjustment coefficient is greater than the first change rate adjustment coefficient, so that the reduction effect of the node change rate on the node stability score when the local node changes is greater than the reduction effect of the node change rate on the node stability score when the overall environment changes.

[0012] In one possible implementation, generating a node deweighting result when the current confidence level is lower than a failure threshold includes: comparing the current confidence level of each light acquisition node with the failure threshold; when the current confidence level is less than the failure threshold, marking the corresponding light acquisition node as a deweighted node and generating a node deweighting result containing a node number, current confidence level, failure threshold, and deweighting coefficient; when the current confidence level is not less than the failure threshold, marking the corresponding light acquisition node as a normal node.

[0013] In one possible implementation, generating fusion weights based on the current credibility and the node weight reduction result includes: for normal nodes, using the current credibility of the normal node as the base value for weight calculation; for weighted nodes, using the current credibility of the weighted node reduced by the weight reduction coefficient as the base value for weight calculation; summing the base values ​​for weight calculation of multiple illumination acquisition nodes to obtain a total base value; when the total base value is greater than 0, normalizing the base values ​​for weight calculation of multiple illumination acquisition nodes to obtain the fusion weight of each illumination acquisition node; when the total base value is equal to 0, reading the fusion weight of the previous sampling period as the fusion weight of the current sampling period, or assigning the same fusion weight to each illumination acquisition node that successfully received light in the current sampling period.

[0014] In one possible implementation, the fusion of multiple node illuminance values ​​according to the fusion weights to obtain an environmental fusion illuminance value includes: multiplying the node illuminance value of each light acquisition node by the corresponding fusion weight to obtain the weighted illuminance component of the corresponding light acquisition node; and summing the multiple weighted illuminance components to obtain the environmental fusion illuminance value.

[0015] In one possible implementation, generating a light source adjustment control quantity based on the illuminance error between the ambient blended illuminance value and the target illuminance value includes: reading the target illuminance value, the ambient blended illuminance value of the current sampling period, the illuminance error of the previous sampling period, and the PWM output of the previous sampling period; determining the difference between the target illuminance value and the ambient blended illuminance value as the current illuminance error; and generating a PWM dimming control quantity for the current sampling period based on the current illuminance error, the change between the current illuminance error and the illuminance error of the previous sampling period, the sampling period duration, and the PWM output of the previous sampling period, wherein the light source adjustment control quantity is the PWM dimming control quantity.

[0016] In one possible implementation, the method is executed by a real-time operating system, which configures node data reception, credibility update, dynamic weighted fusion, closed-loop dimming control, display refresh, and network communication as independent tasks. The node data reception task writes the received node illuminance values ​​into a message queue. The credibility update task reads the node illuminance values ​​from the message queue and generates the current credibility. The dynamic weighted fusion task reads the current credibility and generates an ambient fused illuminance value. The closed-loop dimming control task reads the ambient fused illuminance value, generates a PWM dimming control value, and outputs a PWM dimming signal. The display refresh task reads the ambient fused illuminance value, the current credibility, and the PWM dimming control value and refreshes the display module. The network communication task reads the ambient fused illuminance value, the current credibility, and node maintenance prompts and uploads them to a remote platform.

[0017] Compared with related technologies, this application has the following advantages: Overall, this application generates current node deviation, node change rate, and overall environmental change identifiers during multi-node illuminance sampling, and updates the current reliability of each illuminance acquisition node by combining historical reliability. This allows node reliability evaluation to simultaneously include current lateral deviation, longitudinal changes in adjacent sampling periods, and historical operating status. Environmental illuminance judgment no longer relies solely on a single sample value, simple average value, or fixed weight result. Therefore, when a illuminance acquisition node is affected by local occlusion, local strong light, reflective interference, or communication anomalies, the system can identify abnormal changes in that node relative to the overall illuminance level and its own historical status, thereby reducing the problem of environmental illuminance judgment being skewed due to abnormal nodes participating in fusion calculations for a long time.

[0018] This application generates an overall environmental change identifier based on the consistency of the change rates of multiple nodes, and incorporates this identifier into the current confidence update. This enables the control system to distinguish between a few node mutations and multiple node synchronous changes. When multiple light acquisition nodes rise or fall synchronously, they can be treated as true changes in the overall illuminance of the target environment, reducing the possibility of misjudging changes in external light, curtain movement, or overall supplementary lighting as node anomalies. When only a few nodes mutate, they can be treated as local node changes, thereby suppressing the impact of local abnormal nodes on subsequent environmental illuminance fusion results.

[0019] This application generates node weighting results when the current confidence level is below the failure threshold, generates fusion weights based on the current confidence level and the node weighting results, and then weights and fuses the illuminance values ​​of multiple nodes according to the fusion weights to obtain the environmental fusion illuminance value. This reduces the weight contribution of low-confidence nodes in the fusion calculation, allowing the sampling results of normal nodes to occupy a higher proportion in the environmental fusion illuminance value. As a result, the environmental fusion illuminance value can more stably reflect the actual illuminance state of the target environment, and generates a PWM dimming signal based on the illuminance error between the environmental fusion illuminance value and the target illuminance value, thereby improving the stability and response consistency of LED closed-loop dimming control. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments are briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of a traditional single-node intelligent lighting system. Figure 2 This is a system overall structure block diagram provided for embodiments of this application; Figure 3 The schematic diagram of the overall control terminal provided in the embodiments of this application; Figure 4 This is a design diagram of the closed-loop dimming control algorithm provided in the embodiments of this application; Figure 5 This is a circuit diagram of an LED light source adjustment module provided in an embodiment of this application; Figure 6 This is a data comparison chart showing the feedback weighting algorithm and the average value algorithm under node occlusion interference conditions provided in the embodiments of this application; Figure 7 A comparison chart of the feedback weighting algorithm and the average value algorithm when external light source interference is provided in the embodiments of this application. Detailed Implementation

[0022] The technical solution of this application will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are only some embodiments of this application, and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are all within the protection scope of this application.

[0023] In the description of this application, the terms "first," "second," etc., are used only to distinguish different objects and do not indicate a specific order or degree of importance. The terms "comprising," "having," and any variations thereof are intended to cover non-exclusive inclusion. Unless otherwise expressly defined, method steps may be performed in the order shown herein, or may be performed in parallel or in a different order without creating technical inconsistencies.

[0024] In this application, a light acquisition node refers to a data acquisition unit deployed at different locations in a target environment, capable of acquiring the ambient illuminance at that location and uploading the node's illuminance value to a central control unit. Each light acquisition node may include a microcontroller, a light sensor, and a wireless communication module. The target environment may be a classroom, office, greenhouse, laboratory, or other space requiring stable illuminance.

[0025] In this application, the node illuminance value refers to the illuminance data collected and uploaded by the light acquisition node in the current sampling period. The historical illuminance value refers to the node illuminance value saved by the same light acquisition node in the previous sampling period. The historical reliability refers to the reliability of the same light acquisition node saved in the previous sampling period. The "current reliability" in this application refers to the node reliability evaluation value updated based on the node's current deviation, node change rate, historical reliability, and overall environmental change identifier in the current sampling period.

[0026] In this application, the overall environmental change identifier refers to an identifier used to characterize whether multiple illumination acquisition nodes exhibit consistent illuminance changes within the current sampling period. When multiple nodes change in the same direction with an amplitude not less than a preset change rate threshold, the overall environmental change identifier indicates an overall environmental change; when only one or a few nodes have a node change rate not less than a preset change rate threshold and a synchronous change ratio less than a preset synchronous ratio threshold, the overall environmental change identifier indicates a local node change.

[0027] In this application, PWM is an abbreviation for Pulse Width Modulation; LED is an abbreviation for Light Emitting Diode; IIC is an abbreviation for Inter-Integrated Circuit; RTOS is an abbreviation for Real Time Operating System; UCOSIII is an embedded real-time operating system; and WiFi is a wireless local area network communication technology.

[0028] like Figure 1As shown, a traditional multi-collection-point intelligent lighting system typically consists of multiple collection points, a central control terminal, adjustment points, LED drivers, LED light sources, and an infrared remote controller. Multiple collection points upload ambient light data to the central control terminal via wireless communication. The central control terminal generates a dimming control signal based on the received light data and a preset target illuminance value, and sends it wirelessly to the adjustment points. The adjustment points then adjust the brightness of the LED light sources, such as LED beads, via LED drivers. The infrared remote controller is used to input control commands to the central control terminal. While this structure can utilize multiple collection points to collect ambient light, it typically still performs averaging or fixed-weighting based on the received data. If a collection point is obstructed, partially illuminated, or its installation location is not representative, it is easy to mistake local anomalies for overall ambient illuminance changes, resulting in incorrect dimming.

[0029] like Figure 2 As shown in the embodiments of this application, the system includes multiple light acquisition nodes, a central control terminal, adjustment points, and a wireless communication network. The multiple light acquisition nodes are distributed at different locations in the target environment. Each light acquisition node acquires the ambient illuminance at its location and sends the node illuminance value to the central control terminal via the wireless communication network. After receiving the illuminance values ​​from multiple nodes, the central control terminal performs dynamic evolution of node reliability, dynamic weight fusion, and closed-loop dimming control. The adjustment point is a light source adjustment module, which receives the PWM dimming signal output from the central control terminal and drives the brightness of the LED light source to change.

[0030] like Figure 3 As shown, the main control unit may include a main controller, a wireless communication module, a display module, and a network communication module. The wireless communication module is used to receive node sampling data packets uploaded by multiple light acquisition nodes; the main controller is used to parse the node sampling data packets and run a dynamic feedback weighted fusion algorithm; the display module is used to display the current ambient fusion illuminance value, target illuminance value, node confidence level, and PWM dimming control value; the network communication module is used to connect to the network via WiFi, enabling users to remotely view ambient illuminance data, set target illuminance values, or execute remote dimming control.

[0031] In one embodiment, each illumination acquisition node includes an STM32F103 microcontroller, a TSL2561 optical-to-digital converter, and an NRF24L01 radio frequency communication module. The TSL2561 optical-to-digital converter acquires the illumination intensity at its location and outputs illuminance data. The STM32F103 microcontroller reads this illuminance data via the IIC bus and assembles a node sampling data packet containing the node number, sampling period number, illuminance data, and transmission time identifier. This data packet is then sent to the central control terminal via the NRF24L01 radio frequency communication module. After receiving the node sampling data packet, the central control terminal writes the illuminance data into the data buffer of the corresponding node according to the node number, forming multiple node illuminance values ​​for the current sampling period.

[0032] The multi-node illumination control method provided in this application embodiment can be executed according to the following steps.

[0033] Step S101: Obtain the node illuminance values ​​uploaded by multiple light acquisition nodes in the current sampling period, as well as the historical illuminance values ​​and historical reliability of each light acquisition node in the previous sampling period, and calculate the reference illuminance value based on the multiple node illuminance values.

[0034] Specifically, the main control unit listens for node sampling data packets received by the wireless communication module within the current sampling period. For each node sampling data packet, the main control unit first reads the node number, then reads the sampling period number. When the sampling period number matches the current sampling period, the illuminance data in that node sampling data packet is identified as the node illuminance value corresponding to that node number. When the sampling period number is not the current sampling period, the node sampling data packet is discarded as expired data. After completing the reception, the main control unit obtains the illuminance values ​​of multiple nodes successfully received within the current sampling period.

[0035] In one implementation, the node sampling data packet includes a node number, a sampling period number, a node illuminance value, a transmission time identifier, and a check field. The node number is used to distinguish different illuminance acquisition nodes; the sampling period number is used to determine whether the node's sampling data packet belongs to the current sampling period; the transmission time identifier is used to determine whether the data packet exceeds a preset reception time window; and the check field is used to determine whether data errors have occurred during wireless communication.

[0036] The central control unit first reads the node number and sampling period number, then determines whether the sampling period number is equal to the current sampling period number. The central control unit also reads the transmission time identifier and the current system reception time, determining the difference between the current system reception time and the transmission time identifier as the data transmission delay. When the sampling period number equals the current sampling period number, the data transmission delay is not greater than a preset reception time window, and the verification field passes the verification, the node illuminance value in the node's sampling data packet is written to the corresponding node's data buffer. The preset reception time window is determined by the sampling period duration, the maximum allowable delay of wireless communication, and the real-time operating system task scheduling cycle, and the preset reception time window is less than or equal to one sampling period duration. If the sampling period numbers are inconsistent, the data transmission delay is greater than the preset reception time window, or the verification field fails the verification, the node's sampling data packet is discarded. This generates a set of node illuminance values ​​for the current sampling period for use in calculating reference illuminance values.

[0037] The central control unit reads the historical illuminance values ​​and historical reliability of each light acquisition node from the historical data table based on the node number. The historical data table includes at least the node number, the historical illuminance value of the previous sampling period, and the historical reliability of the previous sampling period. If a node is participating in control for the first time, its historical illuminance value can be set to the illuminance value of the node it acquired for the first time, and its historical reliability can be set to a preset initial reliability. The preset initial reliability can be set to a value between 0 and 1, such as 0.8 or 1.0. The specific value can be determined based on the node's factory calibration results, installation and acceptance results, or historical operation records.

[0038] The main control unit reads the illuminance values ​​of multiple nodes within the current sampling period and calculates the reference illuminance value according to the following formula: Formula 1: in, This indicates the current sampling period or current sampling time, obtained from the sampling period count at the main control terminal; This indicates the sequence number of the light acquisition node, which is obtained by mapping from the node number; This indicates the number of illumination acquisition nodes participating in the calculation of the reference illuminance value within the current sampling period, which is determined by the number of nodes that successfully received the data. Indicates the first Each light acquisition node is in The node illuminance value at any given time is derived from the parsing results of the node's sampled data packets, and the unit can be lux; express The reference illuminance value at time t, in lux. The output of Formula 1 is the reference illuminance value, which is used in step S102 to calculate the current node deviation.

[0039] In another optional implementation, when the number of light-collecting nodes in the target environment is not less than three, the central control terminal can first sort the illuminance values ​​of multiple nodes according to their numerical values, remove the highest and lowest illuminance values, and then perform an average calculation on the remaining node illuminance values ​​to obtain a reference illuminance value. This method is suitable for calculating the reference illuminance value when a single node is subjected to localized strong light or shading. If the number of nodes involved in the calculation is less than three, the average calculation method corresponding to Formula 1 can be used to calculate the reference illuminance value.

[0040] Step S102: Obtain the current deviation of the node based on the difference between the node illuminance value and the reference illuminance value; obtain the node change rate based on the change between the node illuminance value and the historical illuminance value; and obtain the overall environmental change indicator based on the consistency of the change rates of multiple nodes.

[0041] Specifically, for the first For each light acquisition node, the central control terminal calculates the current node deviation using the following formula: Formula 2: in, Indicates the first Each light acquisition node is in The current deviation of the node at time t, in lux; Indicates the first Each light acquisition node is in The nodal illuminance value at a given time; express The reference illuminance value at time 1. The output of Formula 2 is the current node deviation, which is used in step S103 to calculate the node stability score.

[0042] For the For each illumination acquisition node, the central control terminal calculates the node change rate using the following formula: Formula 3: in, Indicates the first Each light acquisition node is in The rate of change of the nodes at time t is a dimensionless value; Indicates the first The current illuminance value of each light acquisition node; Indicates the first The historical illuminance values ​​of each light acquisition node in a sampling period are derived from the historical data table. This represents a very small positive number to avoid a denominator of zero, and can be set based on the minimum effective illuminance reading of the light sensor. The output of Formula 3 is the node change rate, which is used in step S103 to calculate the node stability score and in the current step to generate the overall environmental change indicator.

[0043] The central control unit generates an overall environmental change indicator based on the consistency of the change rates across multiple nodes. Specifically, the central control unit compares the current illuminance value of each node with its historical illuminance values. When the current illuminance value is greater than the historical illuminance value, and the absolute value of the difference is not less than a preset small change threshold, the node's change direction is marked as increasing; when the current illuminance value is less than the historical illuminance value, and the absolute value of the difference is not less than the preset small change threshold, the node's change direction is marked as decreasing; when the absolute value of the difference is less than the preset small change threshold, the node's change direction is marked as stable. The preset small change threshold can be determined based on the resolution of the light sensor, sampling noise, and the natural fluctuation range of the target environment.

[0044] The central control unit counts the number of light-collecting nodes whose change direction is marked as upward and whose node change rate is not less than a preset change rate threshold, and the number of light-collecting nodes whose change direction is marked as downward and whose node change rate is not less than a preset change rate threshold. The larger of these two values ​​is determined as the number of synchronously changing nodes. The number of synchronously changing nodes is compared with the number of currently successfully received nodes to obtain the synchronization ratio. When the synchronization ratio is not less than a preset synchronization ratio threshold, an overall environmental change indicator representing an overall environmental change is generated; when the synchronization ratio is less than the preset synchronization ratio threshold, an overall environmental change indicator representing a local node change is generated. The preset synchronization ratio threshold can be set to, for example, 0.6, 0.75, or 0.8, or it can be calibrated based on the number and installation location distribution of light-collecting nodes within the target environment.

[0045] For example, in a classroom scenario, four light-collecting nodes are positioned in the lectern area, the window-side area, the back row area, and the corridor-side area. When the light-collecting node in the window-side area is partially obscured by a curtain, only the node illuminance value of the window-side node decreases, while the node illuminance values ​​of the other nodes remain stable. In this case, the synchronous change ratio is less than a preset synchronous ratio threshold, and the central control unit generates an overall environmental change indicator representing the change in the local node. When clouds outside the window obscure the sun, the node illuminance values ​​of all four light-collecting nodes decrease, and the direction of change is consistent. In this case, the synchronous change ratio is not less than a preset synchronous ratio threshold, and the central control unit generates an overall environmental change indicator representing the overall environmental change. The former indicator is used to reduce the credibility of locally abnormal nodes, while the latter indicator is used to reduce the penalty for real-world environmental changes.

[0046] In one optional implementation, to avoid misjudgments caused by wireless jitter within a single sampling period, the central control unit can continuously read the overall environmental change identifier for two or more sampling periods. Only when the synchronization change ratio is not less than a preset synchronization ratio threshold for multiple consecutive sampling periods is the overall environmental change confirmed. This method can reduce the impact of short-term communication fluctuations or instantaneous interference on the identification of overall environmental changes.

[0047] Step S103: Update the current credibility of each illumination acquisition node based on the current node deviation, node change rate, historical credibility, and overall environmental change identifier.

[0048] The central control unit can first calculate the node stability score based on the current node deviation and the node change rate, and then calculate the current reliability based on the historical reliability and the node stability score.

[0049] Node stability score is calculated using the following formula: Formula 4: in, Indicates the first Each light acquisition node is in The node stability score at time step is a dimensionless value. Represents the natural constant; This represents the deviation adjustment coefficient. Not less than 0, used to control the impact of the node's current deviation on the node's stability score; Indicates the first Each light acquisition node is in The current deviation of the node at any given moment; This represents the rate of change adjustment coefficient. Not less than 0, used to control the impact of the node change rate on the node stability score; Indicates the first Each light acquisition node is in The node change rate at time t. The output of Equation 4 is the node stability score, which is used for the current confidence calculation.

[0050] The current credibility is calculated according to the following formula: Formula 5: in, Indicates the first Each light acquisition node is in The current confidence level at any given moment is a dimensionless value. Indicates the first The historical reliability of a sampling period at a light acquisition node is derived from the historical data table. Represents the historical weighting factor, and This is used to control the proportion of historical credibility in current credibility; Indicates the first Each light acquisition node is in The node stability score at time step S104. The output of Formula 5 is the current credibility score, which is used in step S104 for node weight reduction judgment and in step S105 for generating fusion weights.

[0051] , and This is determined during the system installation and commissioning phase. Specifically, under unobstructed target environment conditions and stable LED light source output, nodal illuminance values ​​are collected for multiple sampling periods to obtain unobstructed calibration samples. Nodal illuminance values ​​are also collected under conditions of single-node occlusion, single-node localized strong light, and overall environmental illuminance variation to obtain abnormal calibration samples and overall variation calibration samples. Candidate samples are selected sequentially from the main control terminal or commissioning terminal. Candidates and candidates The system calculates the current confidence level for each calibration sample and selects parameter sets that meet the following conditions: In unobstructed calibration samples and overall change calibration samples, the current confidence level of most nodes is not lower than the failure threshold; in single-node occlusion and single-node local strong calibration samples, the current confidence level of the harassed node is lower than the current confidence level of the undisturbed node. Parameter sets that meet the above conditions are written into the system parameter table for use in subsequent sampling periods when calculating the current confidence level.

[0052] When the overall environmental change indicator signifies a change in the overall environment, it means that multiple nodes are rising or falling simultaneously. This change is more likely due to a change in the overall illuminance of the target environment. In this case, the central control unit can reduce the rate of change adjustment coefficient to reduce the penalty for the node's rate of change, preventing a true environmental abrupt change from being misjudged as multiple nodes simultaneously being abnormal. When the overall environmental change indicator signifies a change in local nodes, it means that only a few nodes are experiencing abrupt changes in illuminance. In this case, the central control unit can maintain or increase the rate of change adjustment coefficient to reduce the current confidence level of nodes corresponding to local shading, local strong light exposure, or reflective interference.

[0053] In one embodiment, the preset rate of change adjustment coefficient table can be set as follows: When an overall environmental change is identified as such, the criterion is that the proportion of synchronous changes is not less than a preset synchronous proportion threshold; the change rate adjustment coefficient is selected by choosing the first change rate adjustment coefficient. The output is used to calculate node stability scores.

[0054] When an overall environmental change is identified as a local node change, the criterion is that the proportion of synchronous change is less than a preset synchronous proportion threshold; the change rate adjustment coefficient is selected by choosing the second change rate adjustment coefficient. ,and The output is used to calculate node stability scores.

[0055] in, The rate of change adjustment coefficient represents the overall change in the environment, and its value can be found in the preset rate of change adjustment coefficient table. This represents the rate of change adjustment coefficient when local nodes change; its value is obtained from a preset rate of change adjustment coefficient table. Greater than This relationship means that changes in local nodes have a greater effect on reducing the rate of change of nodes than changes in the overall environment.

[0056] The preset rate of change adjustment coefficient table can be generated during the system installation and commissioning phase. Specifically, firstly, under unobstructed conditions and with stable LED light source output, node illuminance values ​​for multiple sampling periods are collected to obtain normal fluctuation samples. Then, overall shading, overall supplemental lighting, single-node shading, and single-node strong light interference conditions are set separately, and the corresponding node illuminance values, current node deviations, and node rate of change are collected. The parameter group that ensures most nodes remain above the failure threshold under overall shading and overall supplemental lighting conditions, and that the confidence level of the interfered node is lower than that of the normal node under single-node shading and single-node strong light interference conditions, is written into the preset rate of change adjustment coefficient table.

[0057] In another optional implementation, the node stability score can be generated using a segmented scoring method. Specifically, the central control unit compares the current node deviation with a preset deviation segment table and the node change rate with a preset change rate segment table to obtain a deviation score and a change rate score. The deviation score and the change rate score are then weighted and summed to obtain the node stability score. The preset deviation segment table and the preset change rate segment table can be formed using calibration samples of no occlusion, single-node occlusion, and overall illuminance changes. This method is suitable for embedded control scenarios with low processor computing power and where exponential operations are inconvenient.

[0058] Step S104: When the current credibility is lower than the failure threshold, generate node weight reduction results and generate fusion weights based on the current credibility and node weight reduction results.

[0059] The central control unit compares the current confidence level of each illumination acquisition node with a failure threshold. When the current confidence level of a node is lower than the failure threshold, the central control unit identifies that node as a downweighted node and generates a node downweighting result. The node downweighting result includes at least the node number, current confidence level, failure threshold, and downweighting coefficient. The downweighting coefficient is a value between 0 and 1, used to reduce the impact of the corresponding node's participation in the fusion calculation. When the current confidence level is not lower than the failure threshold, the central control unit identifies that node as a normal node.

[0060] The failure threshold can be determined based on the unobstructed calibration data after the light acquisition nodes are installed. Specifically, under normal lighting conditions in the target environment without local obstruction, the node reliability is continuously collected for multiple sampling periods. The lowest quantile value of each node's reliability, or the value obtained by subtracting a preset safety margin from the lowest quantile value, is used as the failure threshold. The weighting coefficient can be determined based on the local obstruction calibration data. Specifically, under single-node obstruction conditions, the weighting coefficient is adjusted so that the difference between the weighted fused environmental illuminance value and the mean value of the unobstructed nodes is less than the preset fusion error upper limit, and the weighting coefficient that meets this condition is written into the system parameter table.

[0061] The fusion weights are calculated using a normalized method based on the baseline weight calculation values. In one embodiment, for normal nodes, the node's current credibility is used as the baseline weight calculation value; for deweighted nodes, the result of multiplying the node's current credibility by the deweighting coefficient is used as the baseline weight calculation value. The baseline weight calculation values ​​of all participating nodes are summed to obtain the total baseline value. The fusion weight of each node is obtained by dividing its baseline weight calculation value by the total baseline value.

[0062] The base value for weight calculation is determined according to the following formula: Formula 6: When the When each light acquisition node is a normal node, the base value for weight calculation is determined according to the above formula.

[0063] When the When a light acquisition node is a deweighted node, the base value for weight calculation is determined according to the above formula.

[0064] The fusion weights are determined according to the following formula: Formula 7: in, Indicates the first Each light acquisition node is in The base value for calculating the weight at any given time; Indicates the first Each light acquisition node is in The current credibility of a given moment; Indicates the first The weighting coefficient corresponding to each illumination acquisition node ; Indicates the first Each light acquisition node is in The fusion weights at each time step are dimensionless values. Indicates the node number in the summation process; This indicates the number of nodes participating in the fusion weight normalization calculation; Indicates the first Each light acquisition node is in The weight calculation base value at time. The output of Formula 7 is the fusion weight, which is used in step S105 to calculate the environmental fusion illuminance value.

[0065] When the sum of the base values ​​of the weights of all participating nodes is 0, the central control unit does not perform division normalization, but instead reads the fusion weight from the previous sampling period as the fusion weight for the current sampling period. If there is no fusion weight from the previous sampling period, the same fusion weight is assigned to each light acquisition node that successfully receives light in the current sampling period. This avoids the denominator being 0 during the fusion weight normalization process. This fusion weight continues to be used in step S105 to calculate the environmental fusion illuminance value.

[0066] In one optional implementation, when the current confidence level of a certain illumination acquisition node falls below the failure threshold for multiple consecutive sampling periods, the central control unit can temporarily remove the node from the fusion calculation and generate a node maintenance prompt. The node maintenance prompt may include the node number, the number of consecutive low confidence periods, the most recent node illuminance value, the most recent node change rate, and suggested inspection items. Suggested inspection items may include whether there is obstruction, sensor surface contamination, power supply abnormality, wireless communication abnormality, or installation angle misalignment.

[0067] Step S105: Weighted fusion of multiple node illuminance values ​​according to fusion weights to obtain the environmental fusion illuminance value.

[0068] The central control unit reads the illuminance value and fusion weight of each illumination acquisition node, multiplies the illuminance value of each node by its corresponding fusion weight, and then sums the results to obtain the environmental fusion illuminance value. The environmental fusion illuminance value is calculated according to the following formula: Formula 8: in, express The ambient illuminance value at any given time, in lux; Indicates the first Each light acquisition node is in Moment-level fusion weights; Indicates the first Each light acquisition node is in The nodal illuminance value at time t. The output of Formula 8 is the environmental fusion illuminance value, which is used in step S106 to calculate the illuminance error.

[0069] Since the fusion weights are derived from the current confidence level, and the current confidence level is determined by the current bias, node change rate, historical confidence level, and overall environmental change indicators, the environmental fusion illuminance value can reduce the impact of local abnormal nodes on the overall illuminance estimation, while retaining the contribution of overall environmental changes to the fusion illuminance.

[0070] Step S106: Generate a light source adjustment control quantity based on the illuminance error between the ambient fusion illuminance value and the target illuminance value, and output a PWM dimming signal based on the light source adjustment control quantity to adjust the brightness of the LED light source.

[0071] like Figure 4 and Figure 5 As shown, the main control terminal reads the target illuminance value and the ambient blended illuminance value, and calculates the illuminance error according to the following formula: Formula 9: in, This indicates the current illuminance error, and the unit can be lux; This represents the target illuminance value, which can be in lux and can be sent via local buttons, a local display interface, or a remote cloud platform. This represents the current ambient blended illuminance value. The output of Formula 9 is the current illuminance error, which is used for calculating the PWM dimming control value. When the ambient blended illuminance value is less than the target illuminance value, the illuminance error is positive, indicating that the LED light source brightness needs to be increased; when the ambient blended illuminance value is greater than the target illuminance value, the illuminance error is negative, indicating that the LED light source brightness needs to be decreased.

[0072] The main control terminal generates the PWM dimming control quantity according to the following formula: Formula 10: in, This indicates the PWM dimming control amount for the current sampling period. This indicates the PWM dimming control amount in the previous sampling period; This represents the proportional coefficient, used to control the impact of the current illuminance error on the PWM dimming control quantity; This represents the differential coefficient, used to control the impact of the change in illuminance error between adjacent sampling periods on the PWM dimming control quantity; This indicates the current illuminance error for the current sampling period; This indicates the illuminance error in the previous sampling period; This indicates the sampling period duration. The output of Formula 10 is the PWM dimming control quantity, which is used to output the PWM dimming signal after being limited.

[0073] The main control unit limits the calculated PWM dimming control quantity within a preset PWM output range, such as 0% to 100% duty cycle, and then outputs the PWM dimming signal to the LED driver. The LED driver adjusts the conduction duty cycle of the LED light source according to the PWM dimming signal, thereby achieving continuous adjustment of the light source brightness.

[0074] The proportional and differential coefficients can be determined during the LED light source commissioning phase. Specifically, the target illuminance value is set as a step change value, and the response time, maximum overshoot, and steady-state error required for the ambient blended illuminance value to reach the target illuminance value are recorded. When the response time exceeds the preset upper limit of the response time, the proportional coefficient is increased; when the maximum overshoot exceeds the preset upper limit of the overshoot, the proportional coefficient is decreased or the differential coefficient is increased; when the steady-state error exceeds the preset upper limit of the steady-state error, the proportional coefficient is readjusted. The combination of coefficients that meets the requirements of response time, maximum overshoot, and steady-state error is used as the dimming control parameters for the current scene.

[0075] In one optional implementation, the main control unit can generate the PWM dimming control quantity using a dimming control method that only includes a proportional term. This method is suitable for scenarios where the LED light source has a fast response and the target environment allows for small fluctuations in illuminance. In another optional implementation, the main control unit can add output limiting and rate of change limiting to the proportional and differential terms, ensuring that the PWM change in adjacent sampling periods does not exceed a preset PWM change upper limit, thereby reducing the flickering of the light source.

[0076] Step S107: Update the node illuminance value to the historical illuminance value of the next sampling period, and update the current confidence level to the historical confidence level of the next sampling period.

[0077] Specifically, after the PWM dimming signal output for the current sampling period is completed, the central control unit writes the current illuminance value of each illumination acquisition node into the historical illuminance value field of the historical data table, writes the current reliability of that node into the historical reliability field of the historical data table, and records the current sampling period number. At the start of the next sampling period, the central control unit reads this data from the historical data table to calculate the new node change rate and the new current reliability. Through this closed-loop feedback method, the node reliability can dynamically evolve with the long-term operating status of the node.

[0078] In one embodiment, if a light acquisition node fails to communicate during the current sampling period, the central control unit may not update the node's historical illuminance value, but may reduce the node's historical reliability; or the node's current reliability may be set as the product of its historical reliability and a preset communication penalty coefficient. The communication penalty coefficient can be determined based on the allowable number of consecutive packet losses, wireless communication quality, and node importance.

[0079] In another embodiment, if the current confidence level of a certain light acquisition node is lower than the failure threshold for multiple consecutive sampling cycles, the main control terminal outputs a node maintenance prompt according to the node maintenance prompt generation rule in step S104. The display module can display the node maintenance prompt, and the network communication module can also send the node maintenance prompt to a remote cloud platform.

[0080] In the experimental environment, four light acquisition nodes can be evenly arranged at equal intervals. Each light acquisition node is connected to the central control terminal via an NRF24L01 wireless communication module. Each light acquisition node periodically collects ambient light data and uploads it to the central control terminal. After receiving the data, the central control terminal first calculates the current ambient reference illuminance value, and then calculates the current node deviation and node change rate of each node. The system updates the current reliability based on the current node deviation, node change rate, and historical reliability, and dynamically adjusts the fusion weight of each node based on the current reliability. The central control terminal generates a PWM dimming signal based on the error between the fused ambient illuminance value and the target illuminance value, and dynamically adjusts the brightness of the LED light source.

[0081] In node occlusion interference scenarios, when a node is obstructed by books, curtains, people, or other objects, the collected value of that node is lower than that of other nodes, and the difference between the collected value of that node and the reference illuminance value is not less than a preset deviation threshold. According to traditional averaging algorithms, this abnormally low value still participates in the averaging calculation, easily lowering the final illuminance estimate and causing the system to unnecessarily increase the brightness of the LED light source. In this embodiment, the current deviation of the obstructed node is large, and the node's rate of change relative to its historical illuminance value is large. Since other nodes have not decreased synchronously, the overall environmental change indicator represents a local node change. The central control terminal increases its effect on reducing the rate of change, reducing the current credibility of the obstructed node and lowering its fusion weight. Therefore, the environmental fusion illuminance value will not decrease due to a single obstructed node, which is not in line with the actual environmental change. Figure 6 As shown, under the condition of node occlusion interference, the feedback weighting algorithm has a more stable fusion output than the average value algorithm.

[0082] In scenarios with external light source interference, when a localized strong light or reflection occurs near a node, the value collected by that node is higher than that of other nodes, and the difference between the collected value and the reference illuminance value is not less than a preset deviation threshold. Traditional averaging algorithms can lead to an overestimation of the final environmental illuminance, potentially reducing the brightness of the LED light source and resulting in insufficient illuminance in the actual dark areas of the target environment. In this embodiment, the node with the localized strong light has a large current deviation, and its rate of change is inconsistent with that of other nodes, resulting in decreased current reliability and a reduced fusion weight. Therefore, the impact of its abnormally high value on the fused environmental illuminance value is suppressed. Figure 7 As shown, when there is external light source interference, the feedback weighting algorithm can reduce the bias of local strong light on the fused illumination compared to the average value algorithm.

[0083] In scenarios involving overall environmental changes, such as clouds blocking the sun outside a window, curtains being drawn all the way in a room, or supplemental lighting being turned on, the illuminance values ​​of multiple light acquisition nodes will rise or fall synchronously. At this time, the central control unit, upon detecting that the synchronous change ratio is not less than a preset synchronous ratio threshold, generates an overall environmental change identifier representing the overall environmental change and reduces the change rate adjustment coefficient. This ensures that the current confidence level of multiple nodes does not simultaneously fall below the failure threshold due to synchronous changes. Therefore, the system can retain the actual environmental changes in the environmental fusion illuminance value, enabling the LED light source to dim in response to actual environmental changes.

[0084] In one embodiment, the method of this application is executed by a central control terminal based on the UCOSIII real-time operating system. The central control terminal can be configured with node data receiving tasks, credibility update tasks, dynamic weighted fusion tasks, closed-loop dimming control tasks, display refresh tasks, and network communication tasks. The node data receiving task has a higher priority than the display refresh task, and is used to promptly receive node sampling data packets uploaded by each illumination acquisition node, and write the received node illuminance values ​​into a message queue; the credibility update task reads the node illuminance values ​​from the message queue and performs calculations of the node's current deviation, node change rate, overall environmental change identifier, and current credibility; the dynamic weighted fusion task reads the current credibility and generates fusion weights and environmental fusion illuminance values; the closed-loop dimming control task reads the environmental fusion illuminance values, generates PWM dimming control values, and updates the PWM register; the display refresh task reads the environmental fusion illuminance values, current credibility, and PWM dimming control values ​​and refreshes the display module; the network communication task reads the environmental fusion illuminance values, current credibility, and node maintenance prompts and uploads them to the cloud platform via a WiFi module. By splitting tasks and using message queues, the problem of mutual blocking between communication reception, algorithm calculation, and display refresh in a single bare-metal loop can be reduced.

[0085] The main challenges in implementing this embodiment lie in ensuring the representativeness of the light acquisition node installation locations, calibrating node reliability parameters, distinguishing between local anomalies and overall environmental changes, and tuning PWM dimming control parameters. To reduce errors caused by node installation locations, installers should distribute multiple light acquisition nodes across different representative areas of the target environment, avoiding the concentration of multiple nodes in the same localized illumination area. To ensure the effectiveness of node reliability updates, installation and commissioning personnel should collect calibration data under conditions of no obstruction, single-node obstruction, localized strong light interference, and overall illumination changes, and accordingly set preset change rate thresholds, preset synchronization ratio thresholds, failure thresholds, weighting coefficients, ratio coefficients, and differential coefficients. During operation, when a node is marked as a weighted node for multiple consecutive sampling periods, the system outputs a node maintenance prompt, prompting maintenance personnel to check whether the node is obstructed, whether sensor contamination has occurred, communication abnormalities have occurred, or installation misalignment has occurred.

[0086] The aforementioned manual intervention steps do not alter the main chain of automatic control in this application. Specifically, manual intervention is primarily used for system installation, threshold calibration, target illuminance setting, and maintenance of abnormal nodes. During normal operation, node illuminance value acquisition, reference illuminance value calculation, current node deviation calculation, node change rate calculation, overall environmental change identifier generation, current confidence update, fusion weight generation, environmental fusion illuminance value calculation, and PWM dimming signal output are all automatically executed by the central control unit.

[0087] It should be noted that, in this document, the terms "comprising," "including," and any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Specific examples have been used in this document to illustrate the principles and implementation methods of the technical solutions of this application. The above examples are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are merely preferred embodiments of this application. It should be pointed out that, due to the limitations of written expression and the objective existence of infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes, or combinations, or the direct application of the concept and technical solutions of this application to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A multi-node illumination control method based on dynamic feedback weighted fusion, characterized in that, include: Obtain the node illuminance values ​​uploaded by multiple light acquisition nodes in the current sampling period, as well as the historical illuminance values ​​and historical reliability of each light acquisition node in the previous sampling period, and calculate a reference illuminance value based on the multiple node illuminance values; The current deviation of a node is obtained based on the difference between the node illuminance value and the reference illuminance value; the node change rate is obtained based on the change between the node illuminance value and the historical illuminance value; and the overall environmental change indicator is obtained based on the consistency of multiple node change rates. The current credibility of each light acquisition node is updated based on the current deviation of the node, the node change rate, the historical credibility, and the overall environmental change identifier. When the current credibility is lower than the failure threshold, a node weight reduction result is generated, and a fusion weight is generated based on the current credibility and the node weight reduction result; The illuminance values ​​of multiple nodes are weighted and fused according to the fusion weight to obtain the environmental fusion illuminance value; The light source adjustment control quantity is generated based on the illuminance error between the environmental fusion illuminance value and the target illuminance value, and a PWM dimming signal is output based on the light source adjustment control quantity. The node illuminance value is updated to the historical illuminance value of the next sampling period, and the current confidence level is updated to the historical confidence level of the next sampling period.

2. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, Obtain the node illuminance values ​​uploaded by multiple illumination acquisition nodes in the current sampling period, including: Each of the light acquisition nodes collects the ambient illuminance at its location through a light sensor, combines the collected illuminance data with the node number, sampling period number, transmission time identifier and verification field to form a node sampling data packet, and sends it to the central control terminal through a wireless communication module. The central control terminal reads the sampling period number, transmission time identifier, and verification field from the node's sampling data packet, and determines the difference between the current system reception time and the transmission time identifier as the data transmission delay. When the sampling period number is consistent with the current sampling period, the data transmission delay is not greater than the preset receiving time window, and the verification field passes the verification, the corresponding illuminance data is read as the node illuminance value.

3. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, Calculating a reference illuminance value based on multiple node illuminance values ​​includes: Read the illuminance values ​​of multiple nodes successfully received within the current sampling period, form an illuminance sequence according to the node number, and perform mean processing on the illuminance values ​​of the nodes in the illuminance sequence to obtain a reference illuminance value; Alternatively, after removing the maximum and minimum illuminance values ​​from the illuminance sequence, the remaining illuminance values ​​are averaged to obtain a reference illuminance value.

4. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, The node change rate is obtained based on the change between the node illuminance value and the historical illuminance value, including: For each light acquisition node, the absolute value of the difference between the node illuminance value of the light acquisition node and the reference illuminance value is determined as the current deviation of the node; The absolute value of the difference between the node illuminance value of the light acquisition node and the corresponding historical illuminance value is determined as the node change amount, and the node change rate is obtained according to the proportional relationship between the node change amount and the historical illuminance value.

5. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, The overall environmental change identifier is obtained based on the consistency of the change rates of multiple nodes, including: The node illuminance value of each light acquisition node is compared with the historical illuminance value to obtain the change direction mark of the corresponding light acquisition node; The number of light-collecting nodes whose change direction is marked as upward and whose node change rate is not less than a preset change rate threshold are counted separately, and the number of light-collecting nodes whose change direction is marked as downward and whose node change rate is not less than the preset change rate threshold are counted separately, and the larger of the two is determined as the number of synchronously changing nodes. When the ratio of the number of synchronized change nodes to the number of successfully received nodes is not less than a preset synchronization ratio threshold, an overall environment change identifier representing an overall environmental change is generated; otherwise, an overall environment change identifier representing a local node change is generated.

6. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, The current reliability of each light acquisition node is updated based on the current node deviation, the node change rate, the historical reliability, and the overall environmental change identifier, including: Based on the rule that a node stability score decreases when the current deviation of a node increases, and a node stability score decreases when the node change rate increases, a node stability score is generated for each illumination acquisition node. Select the rate of change adjustment coefficient based on the overall environmental change indicator; The historical credibility, the node stability score, and the change rate adjustment coefficient are weighted and fused to obtain the current credibility. The change rate adjustment coefficient is used to adjust the weight ratio of the node stability score in the weighted fusion. When the overall environmental change indicator indicates an increase in the magnitude of environmental change, the change rate adjustment coefficient is increased to enhance the influence weight of the node stability score. When the overall environmental change indicator indicates a decrease in the magnitude of environmental change, the change rate adjustment coefficient is decreased to enhance the proportion of the historical credibility.

7. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, When the current credibility is lower than the failure threshold, a node deweighting result is generated, including: Compare the current confidence level of each light acquisition node with the failure threshold; When the current confidence level is less than the failure threshold, the corresponding light acquisition node is marked as a deweighted node, and a node deweighting result containing the node number, current confidence level, failure threshold and deweighting coefficient is generated; When the current confidence level is not less than the failure threshold, the corresponding light acquisition node is marked as a normal node.

8. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 7, characterized in that, A fusion weight is generated based on the current credibility and the node weight reduction result, including: For normal nodes, the current credibility of the normal node is used as the base value for weight calculation; For nodes that have been downgraded, their current credibility is reduced by the downgrade coefficient and then used as the base value for weight calculation. The basic values ​​for weight calculation of multiple light acquisition nodes are summed to obtain the total basic value; when the total basic value is greater than 0, the basic values ​​for weight calculation of multiple light acquisition nodes are normalized to obtain the fusion weight of each light acquisition node. When the sum of the base values ​​equals 0, the fusion weight of the previous sampling period is read as the fusion weight of the current sampling period, or the same fusion weight is assigned to each light acquisition node that successfully receives light in the current sampling period.

9. The multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, The illuminance values ​​of multiple nodes are weighted and fused according to the fusion weight to obtain the environmental fusion illuminance value, including: The node illuminance value of each light acquisition node is multiplied by the corresponding fusion weight to obtain the weighted illuminance component of the corresponding light acquisition node; the multiple weighted illuminance components are summed to obtain the environmental fusion illuminance value.

10. A multi-node illumination control method based on dynamic feedback weighted fusion according to claim 1, characterized in that, The light source adjustment control quantity is generated based on the illuminance error between the environmental fusion illuminance value and the target illuminance value, including: Read the target illuminance value, the ambient fusion illuminance value of the current sampling period, the illuminance error of the previous sampling period, and the PWM output of the previous sampling period; The difference between the target illuminance value and the ambient blended illuminance value is determined as the current illuminance error; Based on the current illuminance error, the change between the current illuminance error and the illuminance error of the previous sampling period, the sampling period duration, and the PWM output of the previous sampling period, a PWM dimming control quantity for the current sampling period is generated, and the light source adjustment control quantity is the PWM dimming control quantity.

Images