An LED Lamp Energy Consumption Optimization Control Method and System Using Ant Colony Algorithm

Abstract

This invention discloses an LED lamp energy consumption optimization control method and system using an ant colony algorithm, specifically relating to the field of LED lamp energy consumption optimization control. The method includes constructing a path network model of a multi-node lighting area and acquiring real-time user activity trajectory information within the lighting area through the LED lighting equipment's acquisition module, generating path usage frequency data. Based on the ant colony algorithm, pheromone optimization and enhancement processing is applied to the path usage frequency, driving the existing lighting control module to select the LED lamps corresponding to the current high-frequency path area for activation. By introducing pheromone attenuation control, negative pheromone suppression mechanisms, and virtual individual trial activation strategies into the path frequency map, outdated paths are identified and suppressed, while new user behavior paths are detected and enhanced. This achieves time-sensitive scheduling of lighting resources and real-time optimal updating of energy consumption configuration, improving the system's responsiveness to behavioral changes and energy-saving efficiency.

Description

Technical Field

[0001] This invention relates to the field of LED lamp energy consumption optimization control technology, and more specifically, to an LED lamp energy consumption optimization control method and system using ant colony algorithm. Background Technology

[0002] In current LED lighting control systems based on ant colony algorithms, the algorithm simulates the behavior of ants leaving pheromones in the environment and gathering along the optimal path to achieve self-organized scheduling and energy-minimized configuration of lighting resources.

[0003] This technology exhibits good local responsiveness and path adaptability in multi-node lighting systems (such as building lighting, urban streetlights, or exhibition lighting). However, with the increasing complexity of application scenarios and the highly nonlinear evolution of spatiotemporal behavior, the system gradually exposes the structural risk of stagnation: Since the core reinforcement mechanism of the ant colony algorithm relies on the historical accumulation of path pheromones, when the user's activity pattern changes, the original high-frequency path is not suppressed in time, while the emerging low-frequency path has not yet established sufficient pheromone support. The system still prioritizes maintaining the high brightness response of the original path, which leads to the lighting control system continuously maintaining the resource supply to the outdated path, forming a redundant energy consumption chain.

[0004] This energy consumption skew not only reduces the energy efficiency of the lighting system, but also weakens the system's behavioral adaptability in dynamic scenarios;

[0005] In summary, the core problem faced by the current technology is that the path reinforcement mechanism of the ant colony algorithm in dynamic lighting environments lacks a sensitive decay design for behavioral changes, resulting in path memory lag in lighting resource allocation. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an LED lamp energy consumption optimization control method and system using an ant colony algorithm. By introducing pheromone attenuation control, negative pheromone suppression mechanism, and virtual individual trial activation strategy into the path frequency spectrum, outdated paths are dynamically identified and suppressed, while new user behavior paths are detected and strengthened. This enables time-sensitive scheduling of lighting resources and real-time optimal updating of energy consumption configuration, thereby improving the system's responsiveness to behavioral changes and energy-saving efficiency.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an LED lamp energy consumption optimization control method using an ant colony algorithm, comprising:

[0008] S1. Construct a path network model for a multi-node lighting area, and obtain real-time user activity trajectory information within the lighting area through the acquisition module of the LED lighting equipment to generate path usage frequency data.

[0009] S2. Based on the ant colony algorithm, the path usage frequency is optimized by pheromone processing, which drives the existing lighting control module to select the LED lights corresponding to the current high-frequency path area for activation response.

[0010] S3. Set the path decay judgment threshold and detect the usage decay rate of each path within a preset time period. If a path is not triggered by user activity for several consecutive periods, mark the path as a "historical high-frequency low-efficiency path".

[0011] S4. Perform negative pheromone injection operation of ant colony algorithm on the marked path. Apply reverse inhibition factor to the pheromone value corresponding to the path through the existing lighting control module to weaken its weight in ant colony path selection.

[0012] S5. Based on the ant colony algorithm, virtual exploration individuals are introduced to periodically and tentatively activate low-frequency paths in the path network model of the lighting area, and the path priority is dynamically adjusted based on the trial response results to form an optimized control of lighting energy consumption.

[0013] In a preferred embodiment, within the lighting area of ​​S1, a node set consisting of multiple LED lighting devices is constructed, and based on the passability relationship between each LED lighting device, a path connection relationship between the nodes is established to form a path network model containing nodes and paths.

[0014] A trajectory sensing module is configured on each LED lighting device. The configured trajectory sensing module continuously collects the user's position status and path switching behavior relative to each node in the lighting area, and generates the user's original trajectory data stream.

[0015] In a preferred embodiment, in S1, the original trajectory data stream is parsed for time series analysis to extract the user's continuous movement path between adjacent LED lighting devices and divide it into multiple path crossing events according to time sequence.

[0016] The path traversal events are statistically processed to calculate the cumulative number of traversals of each path within a unit of time, and the usage frequency data of the corresponding path is constructed.

[0017] The usage frequency data is normalized to obtain the path usage frequency weight value corresponding to each path in the path network model, and a frequency weight map containing all paths is constructed.

[0018] Based on the frequency weight map, the initial pheromone concentration associated with each path in the path network model is assigned a value, and this pheromone concentration value is written into the lighting control module to support the initial decision-making basis for the subsequent ant colony path selection process.

[0019] In a preferred embodiment, in S2, an ant colony lighting control model for LED path activation is constructed, and A is defined in the ant colony lighting control model. k (t) indicates whether the LED light fixture numbered k is activated at time t. k A value of 1 for A indicates activation. k A value of 0 for (t) indicates that the function is not activated;

[0020]

[0021] Where 1(·) is a sign function that outputs 1 when the condition inside the parentheses is true, and 0 otherwise; Φ(·) represents the summation of all paths (i,j) connected to LED lamp k; Φ(·) is the nonlinear control function for path response intensity; λ is the weighting coefficient of path usage frequency in response intensity calculation; f i,j (t) represents the normalized frequency of use of path (i,j) at time t; μ is the adjustment weight of the dynamic change term of path behavior in the overall model; δt is the time window span, which is used to calculate the rate of change of frequency.

[0022] in θ represents the frequency growth rate of path (i,j); θ is the adjustment weight of the path interference suppression factor. This represents the summation of all adjacent paths (m,n) that overlap with path (i,j); |f i,j (t)-f m,n (t) represents the difference in usage frequency between path (i,j) and its adjacent paths; Γ k Let k be the set of paths corresponding to LED lighting fixture k. This represents the set of all paths that have adjacent nodes to path (i,j);

[0023] in τ represents the response intensity; τ is the system's preset lighting activation threshold. When the response intensity exceeds this value, the luminaire is activated.

[0024] In a preferred embodiment, in S3, the behavior decay magnitude of a path is evaluated over multiple consecutive time periods by constructing a periodic decreasing function based on the path frequency difference sequence; if the path is detected to exhibit stable decay and does not meet the lower limit of the active threshold, the path is automatically marked as a "historical high-frequency inefficient path" through a logical labeling function, providing a basis for subsequent lighting strategy adjustments.

[0025] The path is constructed using an attenuation rate calculation model:

[0026]

[0027] Constructing inefficient path label functions:

[0028]

[0029] Where path (i,j) represents a directed path from node i to node j in the path network; m in the above formula represents the index variable of the historical period; t is the current judgment time; Δ is the time span of a single historical evaluation period; M is the number of evaluation periods, which is used as the time window length for decay judgment; f i,j (t) represents the frequency of path (i,j) at time t; ∈ represents a minimal positive number; ψ(∈) is the nonlinear suppression function; D i,j (t) represents the average decay rate of path (i,j) over the past M periods; ρ is the system's preset path decay judgment threshold; 1(∈) is the sign function; L i,j (t) is the inefficient path label for path (i,j). A value of 1 indicates that the path has been judged by the system as a "historical high-frequency inefficient path" at the current moment, and a value of 0 indicates that the marking condition has not been triggered. γ is the set minimum number of non-triggering cycles threshold. If the path has a usage frequency of 0 for at least γ cycles in the most recent M cycles, then the inefficient behavior accumulation criterion is met.

[0030] In a preferred embodiment, in S4, a path pheromone regulation structure is constructed to establish a mapping relationship between each path in the lighting path network and its corresponding pheromone intensity, and a pheromone update rule set is defined in the structure. The pheromone update rule set includes positive enhancement rules for enhancing the weight of the current high-frequency path and negative suppression rules for weakening the weight of historical high-frequency inefficient paths.

[0031] Extract all path indices that are identified as historical high-frequency and inefficient paths in S3 from the path label management structure, generate a path suppression target set, and load the current pheromone intensity corresponding to each path in the path suppression target set as the path suppression input sequence.

[0032] In a preferred embodiment, step S4 further includes: for each path pheromone intensity value in the path-suppressed slug intrusion sequence, constructing a pheromone suppression function based on an exponential decay mechanism, and calculating the suppressed pheromone value. The pheromone suppression function is defined as follows:

[0033]

[0034] in τ represents the pheromone intensity of path (i,j) after the negative pheromone suppression rule is applied at time t+1. i,j (t) represents the original pheromone intensity of the path at time t, η represents the global inhibition coefficient, and ψ i,j(t) represents the cumulative inactivity index of the short path in the recent period; exp is an exponential function;

[0035] The calculation result of the pheromone suppression function is written back to the path pheromone regulation structure as the updated value of the path pheromone intensity, and the structure index and pheromone state of all paths are kept consistent.

[0036] Based on the updated path pheromone regulation structure, the path probability allocation process is re-executed, the ant colony path selection probability update module is called, and the path selection probability distribution between node pairs is reconstructed according to the current pheromone intensity of all paths, ensuring that the guidance probability of suppressed paths in ant colony path selection is explicitly weakened.

[0037] The path selection probability distribution is synchronously transmitted to the lighting control module, which then adjusts the activation priority of LED lights based on the latest path guidance probability, thereby suppressing and controlling the lighting resources corresponding to historically high-frequency and inefficient paths.

[0038] In a preferred embodiment, in S5, a virtual exploration individual generation module is constructed. The virtual exploration individual generation module generates a set of virtual exploration individuals at the beginning of each cycle based on a preset periodic scheduling mechanism. The set of virtual exploration individuals is used to perform exploratory path guidance in the path network without relying on the user's real behavior input.

[0039] In the path usage frequency data structure, paths that do not reach the high-frequency path determination threshold are selected to form a low-frequency path set. The low-frequency path set serves as the path guidance target set for the virtual exploration individual set in the current period and as the limiting range of the exploratory guidance path.

[0040] The virtual exploration individual set is controlled to perform path selection operation based on the current pheromone distribution probability model in the path network. After each path selection is completed, the lighting control module performs instantaneous tentative activation of the LED lights corresponding to the selected path. The instantaneous tentative activation process does not affect the user control flow and meets the set energy consumption protection constraints.

[0041] In a preferred embodiment, in S5, the response behavior of the instantaneous tentative activation process is observed and collected, and it is determined in the lighting behavior feedback data structure whether the target path triggers subsequent path access behavior of the user after activation, and a path tentative response tag is generated based on the determination result.

[0042] Adjust path priorities and perform positive correction or maintain the original value operation for the pheromone value of each path in the low-frequency path set based on the path trial response label results;

[0043] The path selection probability distribution model is recalculated based on the updated pheromone value. The updated path selection probability distribution model is then fed into the lighting control module to drive the activation priority sorting strategy of the LED lights in the next cycle and to achieve periodic performance optimization control of low-frequency paths in the path network structure.

[0044] An LED light energy consumption optimization control system using ant colony algorithm includes a trajectory construction module, a response enhancement module, an attenuation determination module, a weight suppression module, and an activation module.

[0045] The trajectory construction module is used to build a path network model of a multi-node lighting area and to acquire the user's activity trajectory information in the lighting area in real time through the acquisition module of the LED lighting equipment, and generate path usage frequency data.

[0046] The response enhancement module uses pheromone optimization processing based on the ant colony algorithm to optimize the frequency of path usage, and drives the existing lighting control module to select the LED lights corresponding to the current high-frequency path area for activation response;

[0047] The attenuation judgment module is used to set the path attenuation judgment threshold and detect the usage attenuation rate of each path within a preset time period. If a path is not triggered by user activity for several consecutive periods, the path is marked as a "historical high-frequency low-efficiency path".

[0048] The weight suppression module is used to perform negative pheromone injection operation of ant colony algorithm on the marked path. It applies a reverse suppression factor to the pheromone value corresponding to the path through the existing lighting control module to weaken its weight in ant colony path selection.

[0049] The activation module introduces virtual exploration individuals based on the ant colony algorithm. It periodically and tentatively activates low-frequency paths in the path network model of the lighting area and dynamically adjusts the path priority based on the trial response results, thereby forming an optimized control of lighting energy consumption.

[0050] The technical effects and advantages of this invention are as follows:

[0051] 1. This solution addresses the path memory lag problem in existing ant colony algorithms by constructing a path decay mechanism and a negative pheromone suppression process; by performing usage frequency judgment and weight reduction on paths that have not been used for a long time, it quickly responds to user behavior shifts, thereby eliminating redundant energy consumption chains of the original paths and improving the dynamic adaptability and overall energy-saving efficiency of lighting resources.

[0052] 2. The scheme introduces a window sliding statistics strategy in path status judgment, constructs a path usage frequency sequence and performs attenuation assessment accordingly, realizes the identification and labeling management of low-frequency paths, enables the control system to have the ability to respond to path status changes in a periodic manner, and enhances the time sensitivity and adjustable scheduling capability of the lighting strategy.

[0053] 3. By performing negative pheromone injection operations in the path frequency map, the failure path can be actively suppressed. The path selection probability can be dynamically reconstructed without external intervention, thereby enhancing the convergence stability and behavior correction capability of the internal path selection mechanism of the system.

[0054] 4. The system constructs virtual exploration individuals to explore the responsiveness of low-frequency paths. Through a three-step closed loop of activation, monitoring and feedback, potential paths are tentatively activated and user responses are evaluated in real time, thereby achieving automatic differentiation between "pseudo-low-frequency paths" and "real cold paths" and improving the structural discrimination capability of the control strategy.

[0055] 5. Based on the path traversal data generated by the trajectory perception module, the system can continuously update the path frequency map and perform pheromone adaptive regulation, enabling the entire lighting control mechanism to have continuous learning capabilities without increasing the perception burden, and realizing long-term tracking of user behavior evolution and iterative optimization of energy consumption configuration. Attached Figure Description

[0056] Figure 1 This is a flowchart of the method steps of the present invention.

[0057] Figure 2 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Refer to the instruction manual appendix Figure 1-2 An embodiment of the present invention provides an LED lamp energy consumption optimization control method using an ant colony algorithm, comprising:

[0060] S1. Construct a path network model for a multi-node lighting area, and obtain real-time user activity trajectory information within the lighting area through the acquisition module of the LED lighting equipment to generate path usage frequency data.

[0061] S2. Based on the ant colony algorithm, the path usage frequency is optimized and enhanced with pheromone, which drives the existing lighting control module to select the LED lights corresponding to the current high-frequency path area for activation response.

[0062] S3. Set the path decay judgment threshold and detect the usage decay rate of each path within a preset time period. If a path is not triggered by user activity for several consecutive periods, mark the path as a "historical high-frequency low-efficiency path".

[0063] S4. Perform negative pheromone injection operation of ant colony algorithm on the marked path. Apply reverse inhibition factor to the pheromone value corresponding to the path through the existing lighting control module to weaken its weight in ant colony path selection and avoid continuous resource allocation.

[0064] S5. Based on the ant colony algorithm, virtual exploration individuals are introduced to periodically and tentatively activate low-frequency paths in the path network model of the lighting area, and the path priority is dynamically adjusted based on the trial response results to form an optimized control of lighting energy consumption.

[0065] Within the lighting area of ​​S1, a node set consisting of multiple LED lighting devices is constructed, and based on the passability relationship between each LED lighting device, a path connection relationship between the nodes is established to form a path network model containing nodes and paths.

[0066] A trajectory sensing module is configured on each LED lighting device. The configured trajectory sensing module continuously collects the user's position status and path switching behavior relative to each node in the lighting area, and generates the user's original trajectory data stream.

[0067] In S1, the original trajectory data stream is parsed over time to extract the user's continuous movement path between adjacent LED lighting devices and divide it into multiple path crossing events in chronological order.

[0068] The path traversal events are statistically processed to calculate the cumulative number of traversals of each path within a unit of time, and the usage frequency data of the corresponding path is constructed.

[0069] The usage frequency data is normalized to obtain the path usage frequency weight value corresponding to each path in the path network model, and a frequency weight map containing all paths is constructed.

[0070] Based on the frequency weight map, the initial pheromone concentration associated with each path in the path network model is assigned a value, and this pheromone concentration value is written into the lighting control module to support the initial decision-making basis for the subsequent ant colony path selection process.

[0071] In S2, an ant colony lighting control model for LED path activation is constructed, and A is defined in the ant colony lighting control model. k (t) indicates whether the LED light fixture numbered k is activated at time t. k A value of 1 for A indicates activation. kA value of 0 for (t) indicates that the function is not activated;

[0072]

[0073] Where 1(·) is a sign function that outputs 1 when the condition inside the parentheses is true, and 0 otherwise; This represents summing and traversing all paths (i,j) connected to LED lamp k; Φ(∈) is a nonlinear control function for path response intensity, used to enhance stability and avoid extreme values ​​dominating system behavior; λ is the weighting coefficient of path usage frequency in response intensity calculation; f i,j (t) represents the normalized frequency of use of path (i,j) at time t; μ is the adjustment weight of the dynamic change term of path behavior in the overall model; δt is the time window span, which is used to calculate the rate of change of frequency.

[0074] in The frequency growth rate of path (i,j) is used to perceive behavioral trends; θ is the adjustment weight of the path interference suppression factor. This represents the summation of all adjacent paths (m,n) that overlap with path (i,j); |f i,j (t)-f m,n (t)| represents the difference in usage frequency between path (i,j) and its adjacent paths, used to measure the degree of mutual interference; Γ k Let k be the set of paths corresponding to LED lighting fixture k. This represents the set of all paths that have adjacent nodes to path (i,j);

[0075] in τ represents the response intensity; τ is the system's preset lighting activation threshold. When the response intensity exceeds this value, the luminaire is activated.

[0076] In S3, a periodic decreasing function based on the path frequency difference sequence is constructed to evaluate the magnitude of its behavior decay over multiple consecutive time periods. If the path is detected to exhibit stable decay and does not meet the lower limit of the active threshold, the path is automatically marked as a "historical high-frequency inefficient path" through a logical labeling function, providing a basis for subsequent lighting strategy adjustments.

[0077] The path is constructed using an attenuation rate calculation model:

[0078]

[0079] Constructing inefficient path label functions:

[0080]

[0081] Where path (i,j) represents a directed path from node i to node j in the path network; m in the above formula represents the index variable of the historical period; t is the current judgment time; Δ is the time span of a single historical evaluation period, which can be in minutes or hours; M is the number of evaluation periods, and the number of evaluation periods is used for the time window length of decay judgment; f i,j (t) represents the frequency of use of path (i,j) at time t, where the frequency of use of path (i,j) at time t represents the cumulative number of times the user traverses this path within that time period; ∈ represents a minimal positive number, used in the above formula to avoid a zero denominator and ensure computational stability; ψ(∈) is a nonlinear suppression function, used to enhance the significance of the decay trend, its form includes ψ(x)=tanh(1-x), used to strengthen the decay characteristics, where x is the input variable; D i,j (t) represents the average decay rate of path (i,j) over the past M periods, which is the overall measure of the decay trend; ρ is the system's preset path decay judgment threshold. If the average decay rate of a path is lower than this threshold, it is judged as continuously decaying; 1(∈) is the sign function used for logical judgment, returning a value of 0 or 1; L i,j (t) is the inefficient path label for path (i,j). A value of 1 indicates that the path has been judged by the system as a "historical high-frequency inefficient path" at the current moment, and a value of 0 indicates that the marking condition has not been triggered. γ is the set minimum number of non-triggering cycles threshold. If the path has a usage frequency of 0 for at least γ cycles in the most recent M cycles, then the inefficient behavior accumulation criterion is met.

[0082] In S4, a path pheromone regulation structure is constructed to establish a mapping relationship between each path in the lighting path network and its corresponding pheromone intensity. A pheromone update rule set is defined in this structure. The pheromone update rule set includes positive enhancement rules for enhancing the weight of the current high-frequency path and negative suppression rules for weakening the weight of historical high-frequency inefficient paths.

[0083] Extract all path indices that are identified as historical high-frequency and inefficient paths in S3 from the path label management structure, generate a path suppression target set, and load the current pheromone intensity corresponding to each path in the path suppression target set as the path suppression input sequence.

[0084] S4 also includes: for each path pheromone intensity value in the path-suppressed slug intrusion sequence, constructing a pheromone suppression function based on an exponential decay mechanism, and calculating the suppressed pheromone value. The pheromone suppression function is defined as follows:

[0085]

[0086] in τ represents the pheromone intensity of path (i,j) after the negative pheromone suppression rule is applied at time t+1. i,j (t) represents the original pheromone intensity of the path at time t, η represents the global inhibition coefficient, and ψ i,j (t) represents the cumulative inactivity index of the short path in the recent period; exp is an exponential function;

[0087] The calculation result of the pheromone suppression function is written back to the path pheromone regulation structure as the updated value of the path pheromone intensity, and the structure index and pheromone state of all paths are kept consistent.

[0088] Based on the updated path pheromone regulation structure, the path probability allocation process is re-executed, the ant colony path selection probability update module is called, and the path selection probability distribution between node pairs is reconstructed according to the current pheromone intensity of all paths, ensuring that the guidance probability of suppressed paths in ant colony path selection is explicitly weakened.

[0089] The path selection probability distribution is synchronously transmitted to the lighting control module, which then adjusts the activation priority of LED lights based on the latest path guidance probability, thereby suppressing and controlling the lighting resources corresponding to historically high-frequency and inefficient paths.

[0090] In S5, a virtual exploration individual generation module is constructed. The virtual exploration individual generation module is triggered by a preset periodic scheduling mechanism. At the beginning of each period, a set of virtual exploration individuals is generated. The set of virtual exploration individuals is used to perform exploratory path guidance in the path network without relying on the user's real behavior input.

[0091] In the path usage frequency data structure, paths that do not reach the high-frequency path determination threshold are selected to form a low-frequency path set. The low-frequency path set serves as the path guidance target set for the virtual exploration individual set in the current period and as the limiting range of the exploratory guidance path.

[0092] The virtual exploration individual set is controlled to perform path selection operation based on the current pheromone distribution probability model in the path network. After each path selection is completed, the lighting control module performs instantaneous tentative activation of the LED lights corresponding to the selected path. The instantaneous tentative activation process does not affect the user control flow and meets the set energy consumption protection constraints.

[0093] In S5, the response behavior of the instantaneous tentative activation process is observed and collected, and the target path is determined in the lighting behavior feedback data structure to determine whether the target path triggers the user's subsequent path access behavior after activation. Based on the determination result, a path tentative response tag is generated.

[0094] The path priority is adjusted. Based on the path trial response label results, the pheromone value of each path in the low-frequency path set is positively corrected or maintained at its original value. The pheromone correction operation does not cause a path level jump, but is only used to adjust the response intensity of the path in the path selection probability model.

[0095] The path selection probability distribution model is recalculated based on the updated pheromone value. The updated path selection probability distribution model is then fed into the lighting control module to drive the activation priority sorting strategy of the LED lights in the next cycle and to achieve periodic performance optimization control of low-frequency paths in the path network structure.

[0096] An LED light energy consumption optimization control system using ant colony algorithm includes a trajectory construction module, a response enhancement module, an attenuation determination module, a weight suppression module, and an activation module.

[0097] The trajectory construction module is used to build a path network model of a multi-node lighting area and to acquire the user's activity trajectory information in the lighting area in real time through the acquisition module of the LED lighting equipment, and generate path usage frequency data.

[0098] The response enhancement module uses the ant colony algorithm to perform pheromone optimization and enhancement processing on the path usage frequency, driving the existing lighting control module to select the LED lights corresponding to the current high-frequency path area for activation response;

[0099] The attenuation judgment module is used to set the path attenuation judgment threshold and detect the usage attenuation rate of each path within a preset time period. If a path is not triggered by user activity for several consecutive periods, the path is marked as a "historical high-frequency low-efficiency path".

[0100] The weight suppression module is used to perform negative pheromone injection operation of ant colony algorithm on the marked path. It applies reverse suppression factor to the pheromone value corresponding to the path through the existing lighting control module to weaken its weight in ant colony path selection and avoid continuous resource allocation.

[0101] The activation module introduces virtual exploration individuals based on the ant colony algorithm. It periodically and tentatively activates low-frequency paths in the path network model of the lighting area and dynamically adjusts the path priority based on the trial response results, thereby forming an optimized control of lighting energy consumption.

[0102] It should be noted that this solution is designed around "LED lamp energy consumption optimization control method and system using ant colony algorithm". The process began with an in-depth analysis of the energy waste problem of existing lighting control systems. Especially in multi-node lighting areas, traditional systems often cannot dynamically adjust the lighting response path according to changes in user activity trajectories, resulting in a large number of low-frequency paths consuming power continuously, while high-frequency paths also lack optimized control in response delay.

[0103] To this end, this solution first constructs a path network model covering all LED lighting devices, mapping the lighting space structure with node connection relationships. Then, the trajectory sensing module deployed on each LED device collects the user's behavioral trajectory in the lighting area in real time, generating a raw trajectory data stream. This data is then parsed within the system into user traversal events between different paths, forming quantifiable path usage frequency data. This mechanism aims to visualize the originally discrete user behavior as path-level frequency data, providing a foundation for subsequent intelligent path selection.

[0104] Based on this frequency data, the solution introduces the heuristic pheromone mechanism of the ant colony algorithm to assign frequency weights to each path, construct a frequency weight map, and drive the activation behavior of LED lights accordingly. In the high-frequency path area, the corresponding lighting equipment is lit first, realizing the initial control logic of response enhancement.

[0105] As the frequency of path usage changes dynamically over time, the system introduces a path decay determination mechanism. Based on a periodic decreasing function, it evaluates the frequency degradation trend of each path. If a path is not triggered by the user for several consecutive time periods and the decay rate reaches a preset threshold, the system automatically marks it as a "historical high-frequency inefficient path," that is, a path that was once frequently used but is currently inefficient. Subsequently, the system performs a reverse suppression operation of pheromone weight for such paths, weakening their priority in the path selection process through a negative pheromone mechanism, preventing the system resources from being ineffectively allocated to areas that do not currently require a response, thereby suppressing redundant energy consumption.

[0106] To further enhance the model's perception capabilities and energy consumption control flexibility on low-frequency paths, the solution introduces the concept of "virtual exploration individuals." This involves generating a set of virtual guides at preset intervals to actively explore paths, without relying on real user behavior. These virtual individuals select low-frequency paths based on the pheromone distribution in the current path network and perform instantaneous lighting activation, observing whether this triggers a subsequent response from real user behavior. This dynamically determines whether the path should be prioritized or maintained. This strategy addresses the problem of conventional systems abandoning potential needs due to unused paths, enabling the entire control system to possess predictive and exploratory capabilities. Ultimately, this forms a closed-loop control architecture based on user behavior, driven by path status, and centered on pheromone regulation.

[0107] This design approach has good adaptability and scalability in actual deployment, and can be applied to various lighting scenarios with node structures, such as office buildings, smart parks, and underground parking garages, to achieve energy saving and consumption reduction while ensuring the timeliness and accuracy of lighting response.

[0108] Overall, the design of this scheme is based on a full perception and feedback of the dynamic characteristics of path usage, and deeply integrates swarm intelligence algorithms with lighting behavior models.

[0109] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An LED light energy consumption optimization control method using ant colony algorithm, including: S1. Construct a path network model for a multi-node lighting area, and obtain real-time user activity trajectory information within the lighting area through the acquisition module of the LED lighting equipment to generate path usage frequency data. Its features are: S2. Based on the ant colony algorithm, the path usage frequency is optimized by pheromone processing, which drives the existing lighting control module to select the LED lights corresponding to the current high-frequency path area for activation response. S3. Set the path decay judgment threshold and detect the usage decay rate of each path within a preset time period. If a path is not triggered by user activity for several consecutive periods, mark the path as a "historical high-frequency low-efficiency path". S4. Perform negative pheromone injection operation of ant colony algorithm on the marked path. Apply reverse inhibition factor to the pheromone value corresponding to the path through the existing lighting control module to weaken its weight in ant colony path selection. S5. Based on the ant colony algorithm, virtual exploration individuals are introduced to periodically and tentatively activate low-frequency paths in the path network model of the lighting area, and the path priority is dynamically adjusted based on the trial response results to form an optimized control of lighting energy consumption.

2. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 1, characterized in that: Within the lighting area of ​​S1, a node set consisting of multiple LED lighting devices is constructed, and based on the passability relationship between each LED lighting device, a path connection relationship between the nodes is established to form a path network model containing nodes and paths. A trajectory sensing module is configured on each LED lighting device. The configured trajectory sensing module continuously collects the user's position status and path switching behavior relative to each node in the lighting area, and generates the user's original trajectory data stream.

3. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 2, characterized in that: In S1, the original trajectory data stream is parsed over time to extract the user's continuous movement path between adjacent LED lighting devices and divide it into multiple path crossing events in chronological order. The path traversal events are statistically processed to calculate the cumulative number of traversals of each path within a unit of time, and the usage frequency data of the corresponding path is constructed. The usage frequency data is normalized to obtain the path usage frequency weight value corresponding to each path in the path network model, and a frequency weight map containing all paths is constructed. Based on the frequency weight map, the initial pheromone concentration associated with each path in the path network model is assigned a value, and this pheromone concentration value is written into the lighting control module to support the initial decision-making basis for the subsequent ant colony path selection process.

4. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 3, characterized in that: In S2, an ant colony lighting control model for LED path activation is constructed, and A is defined in the ant colony lighting control model. k (t) indicates whether the LED light fixture numbered k is activated at time t. k A value of 1 for A indicates activation. k A value of 0 for (t) indicates that the function is not activated; A k (t)=1 Where 1(·) is a sign function that outputs 1 when the condition inside the parentheses is true, and 0 otherwise; Φ(·) represents summing and traversing all paths (i,j) connected to LED lamp k; Φ(·) is the nonlinear control function for path response intensity; λ is the weighting coefficient of path usage frequency in response intensity calculation. f i,j (t) represents the normalized frequency of use of path (i,j) at time t; μ is the adjustment weight of the path behavior dynamic change term in the overall model; δt represents the time window span, which is used to calculate the rate of change of frequency. in θ represents the frequency growth rate of path (i,j); θ is the adjustment weight of the path interference suppression factor. This represents the summation of all adjacent paths (m,n) that overlap with path (i,j); |f i,j (t)-f m,n (t) represents the difference in usage frequency between path (i,j) and its adjacent paths; Γ k Let k be the set of paths corresponding to LED lighting fixture k. This represents the set of all paths that have adjacent nodes to path (i,j); in τ represents the response intensity; τ is the system's preset lighting activation threshold. When the response intensity exceeds this value, the luminaire is activated.

5. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 4, characterized in that: In S3, a periodic decreasing function based on the path frequency difference sequence is constructed to evaluate the magnitude of its behavior decay over multiple consecutive time periods. If the path is detected to exhibit stable decay and does not meet the lower limit of the active threshold, the path is automatically marked as a "historical high-frequency inefficient path" through a logical labeling function, providing a basis for subsequent lighting strategy adjustments. The path is constructed using an attenuation rate calculation model: Constructing inefficient path label functions: Where path (i,j) represents a directed path from node i to node j in the path network; m in the above formula represents the index variable of the historical period; t is the current judgment time; Δ is the time span of a single historical evaluation period; M is the number of evaluation periods, which is used as the time window length for decay judgment; f i,j (t) represents the frequency of path (i,j) at time t; ∈ represents a minimal positive number; ψ(·) is the nonlinear suppression function; D i,j (t) represents the average decay rate of path (i,j) over the past M periods; ρ is the system's preset path attenuation threshold; 1(·) is the sign function; L i,j (t) is the inefficient path label for path (i,j). A value of 1 indicates that the path has been judged by the system as a "historical high-frequency inefficient path" at the current time, and a value of 0 indicates that the marking condition has not been triggered. γ is the set minimum number of non-triggering cycles threshold. If the path has a usage frequency of 0 for at least γ cycles in the most recent M cycles, then the inefficient behavior accumulation criterion is met.

6. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 5, characterized in that: In S4, a path pheromone regulation structure is constructed to establish a mapping relationship between each path in the lighting path network and its corresponding pheromone intensity. A pheromone update rule set is defined in this structure. The pheromone update rule set includes positive enhancement rules for enhancing the weight of the current high-frequency path and negative suppression rules for weakening the weight of historical high-frequency inefficient paths. Extract all path indices that are identified as historical high-frequency and inefficient paths in S3 from the path label management structure, generate a path suppression target set, and load the current pheromone intensity corresponding to each path in the path suppression target set as the path suppression input sequence.

7. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 6, characterized in that: S4 also includes: for each path pheromone intensity value in the path-suppressed slug intrusion sequence, constructing a pheromone suppression function based on an exponential decay mechanism, and calculating the suppressed pheromone value. The pheromone suppression function is defined as follows: in τ represents the pheromone intensity of path (i,j) after the negative pheromone suppression rule is applied at time t+1. i,j (t) represents the original pheromone intensity of the path at time t, η represents the global inhibition coefficient, and ψ i,j (t) represents the cumulative inactivity index of the short path in the recent period; exp is an exponential function; The calculation result of the pheromone suppression function is written back to the path pheromone regulation structure as the updated value of the path pheromone intensity, and the structure index and pheromone state of all paths are kept consistent. Based on the updated path pheromone regulation structure, the path probability allocation process is re-executed, the ant colony path selection probability update module is called, and the path selection probability distribution between node pairs is reconstructed according to the current pheromone intensity of all paths, ensuring that the guidance probability of suppressed paths in ant colony path selection is explicitly weakened. The path selection probability distribution is synchronously transmitted to the lighting control module, which then adjusts the activation priority of LED lights based on the latest path guidance probability, thereby suppressing and controlling the lighting resources corresponding to historically high-frequency and inefficient paths.

8. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 7, characterized in that: In S5, a virtual exploration individual generation module is constructed. The virtual exploration individual generation module is triggered by a preset periodic scheduling mechanism. At the beginning of each period, a set of virtual exploration individuals is generated. The set of virtual exploration individuals is used to perform exploratory path guidance in the path network without relying on the user's real behavior input. In the path usage frequency data structure, paths that do not reach the high-frequency path determination threshold are selected to form a low-frequency path set. The low-frequency path set serves as the path guidance target set for the virtual exploration individual set in the current period and as the limiting range of the exploratory guidance path. The virtual exploration individual set is controlled to perform path selection operation based on the current pheromone distribution probability model in the path network. After each path selection is completed, the lighting control module performs instantaneous tentative activation of the LED lights corresponding to the selected path. The instantaneous tentative activation process does not affect the user control flow and meets the set energy consumption protection constraints.

9. The LED lamp energy consumption optimization control method using ant colony algorithm according to claim 8, characterized in that: In S5, the response behavior of the instantaneous tentative activation process is observed and collected, and the target path is determined in the lighting behavior feedback data structure to determine whether the target path triggers the user's subsequent path access behavior after activation. Based on the determination result, a path tentative response tag is generated. Adjust path priorities and perform positive correction or maintain the original value operation for the pheromone value of each path in the low-frequency path set based on the path trial response label results; The path selection probability distribution model is recalculated based on the updated pheromone value. The updated path selection probability distribution model is then fed into the lighting control module to drive the activation priority sorting strategy of the LED lights in the next cycle and to achieve periodic performance optimization control of low-frequency paths in the path network structure.

10. An LED light energy consumption optimization control system using ant colony algorithm, comprising a trajectory construction module, a response enhancement module, an attenuation determination module, a weight suppression module, and an activation module, characterized in that: The trajectory construction module is used to build a path network model of a multi-node lighting area and to acquire the user's activity trajectory information in the lighting area in real time through the acquisition module of the LED lighting equipment, and generate path usage frequency data. The response enhancement module uses pheromone optimization processing based on the ant colony algorithm to optimize the frequency of path usage, and drives the existing lighting control module to select the LED lights corresponding to the current high-frequency path area for activation response; The attenuation judgment module is used to set the path attenuation judgment threshold and detect the usage attenuation rate of each path within a preset time period. If a path is not triggered by user activity for several consecutive periods, the path is marked as a "historical high-frequency low-efficiency path". The weight suppression module is used to perform negative pheromone injection operation of ant colony algorithm on the marked path. It applies a reverse suppression factor to the pheromone value corresponding to the path through the existing lighting control module to weaken its weight in ant colony path selection. The activation module introduces virtual exploration individuals based on the ant colony algorithm. It periodically and tentatively activates low-frequency paths in the path network model of the lighting area and dynamically adjusts the path priority based on the trial response results, thereby forming an optimized control of lighting energy consumption.

Images