Urban landscape lighting control system based on composite output mode
By adopting data acquisition, intelligent light pollution control, regional demand forecasting and dynamic scheduling and energy management modules in the urban landscape lighting control system, the light pollution and energy waste problems in the existing system are solved, and intelligent and efficient lighting management is realized. Improves lighting effects and user experience.
Patent Information
- Application Number
- CN202410992665.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-08
AI Technical Summary
The existing urban landscape lighting control system lacks real-time environmental monitoring and dynamic adjustment capabilities, resulting in light pollution and energy waste, low intelligence level, unable to effectively use historical and real-time data for scientific prediction and optimization scheduling, and it is difficult to achieve energy consumption of lighting systems. Accurate monitoring and optimization.
Adopt the urban landscape lighting control system based on the composite output mode, including data acquisition module, intelligent light pollution control module, regional demand forecasting and dynamic scheduling module and energy management module. Through multiple sensors, environmental parameters are monitored in real time, light pollution trends are predicted, and lighting is dynamically scheduled. Resources, optimized energy consumption management, combined with deep belief networks and Bayesian dynamic linear models for demand prediction and energy consumption analysis.
It has realized intelligent, high-efficiency, low-light pollution urban lighting solutions, improved the accuracy and reliability of lighting control, optimized resource allocation and energy efficiency, reduced energy consumption costs, and improved residents' quality of life and environmental comfort.
Smart Images

Figure CN118921804A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of urban landscape lighting, and in particular to an urban landscape lighting control system based on a composite output mode. Background Art
[0002] As an important part of modern urban management, urban landscape lighting not only plays an important role in beautifying the urban environment and enhancing the city's image, but also has important significance in terms of nighttime safety and traffic management. With the acceleration of urbanization, the demand for urban landscape lighting systems continues to grow, and the requirements for lighting effects, energy efficiency management and environmental friendliness are also increasing. Traditional lighting control systems mostly rely on fixed schedules and manual adjustments, which are difficult to meet changing environmental conditions and diverse needs. A more intelligent and efficient solution is urgently needed to improve the management level of urban landscape lighting.
[0003] The existing urban landscape lighting control system has many shortcomings in practical applications. First, the traditional system lacks the ability to monitor and dynamically adjust environmental parameters in real time, resulting in the inability to optimize the lighting effect according to actual conditions, which easily causes light pollution and energy waste. Secondly, the existing system has a low level of intelligence in lighting demand prediction and resource scheduling, and cannot effectively use historical and real-time data for scientific prediction and optimal scheduling, resulting in unreasonable resource allocation and low energy efficiency. In addition, the energy consumption management and energy-saving control strategies are imperfect, making it difficult to accurately monitor and effectively optimize the energy consumption of the lighting system, and unable to meet the high requirements of modern cities for energy conservation and environmental protection.
[0004] In view of the above-mentioned deficiencies of the prior art, the present invention proposes an urban landscape lighting control system based on a composite output mode to improve energy utilization efficiency and reduce energy consumption costs, thereby providing cities with intelligent, environmentally friendly and efficient lighting solutions, effectively improving the quality of life of residents and environmental comfort. Summary of the invention
[0005] Based on the above purpose, the present invention provides an urban landscape lighting control system based on a composite output mode.
[0006] An urban landscape lighting control system based on a composite output mode includes a data acquisition module, an intelligent light pollution control module, a regional demand prediction and dynamic scheduling module, a light source control module and an energy management module, wherein;
[0007] The data acquisition module monitors environmental parameters in real time through multiple sensors, including light intensity, weather conditions, temperature and humidity;
[0008] The intelligent light pollution control module predicts light pollution trends and adjusts the light pollution level of urban lighting based on real-time monitored environmental parameters;
[0009] The regional demand prediction and dynamic scheduling module predicts the lighting demand of different areas based on the adjustment results of the light pollution level and dynamically schedules lighting resources, specifically including:
[0010] Operational data collection: collect and record historical and real-time operational data of each area, including lighting demand data, traffic and passenger flow data, and energy consumption data;
[0011] Demand forecasting: Using historical and real-time operation data, combined with environmental parameters and based on the adjustment results of light pollution levels, the lighting demand forecasting model is used to predict the lighting demand of each area in different time periods;
[0012] Dynamic scheduling: According to the results of demand forecast, lighting resources are dynamically scheduled to adjust the lighting intensity, color temperature and switch status of each area;
[0013] Resource optimization: Combine energy management strategies to optimize the allocation of lighting resources;
[0014] The light source control module is connected to a plurality of light sources of different types, and achieves diversified lighting effects by adjusting the brightness, color and switch status of the light sources;
[0015] The energy management module monitors and optimizes the energy consumption of the lighting control system, records energy consumption data and performs energy-saving control.
[0016] Optionally, the data acquisition module includes:
[0017] Light intensity monitoring: monitor the light intensity of the environment through light sensors;
[0018] Weather Condition Monitoring: Monitoring weather conditions including wind speed, rainfall, and atmospheric pressure through meteorological sensors;
[0019] Temperature monitoring: monitor the ambient temperature through temperature sensors;
[0020] Humidity monitoring: Monitor the ambient humidity through a humidity sensor.
[0021] Optionally, the intelligent light pollution control module includes:
[0022] Light pollution monitoring: receiving real-time monitored environmental parameters and transmitting them in real time;
[0023] Light pollution prediction: Using real-time and historical environmental parameters, the light pollution prediction model is used to predict future light pollution trends;
[0024] Feedback control: Based on the prediction results of light pollution, real-time monitoring of environmental parameters and user feedback is carried out to continuously optimize the light pollution control strategy.
[0025] Optionally, the light pollution prediction model adopts a gradient boosting tree model, and the gradient boosting tree model includes:
[0026] Feature engineering: Generate time features (year, month, day, hour, minute), lag features (light pollution data for the past few days / hours), differential features (the difference between the current value and the previous value), and combine environmental parameters, expressed as:
[0027] X t ={t,t-1,t-2,...,tp,Δy t ,env_parameters};
[0028] Among them, t is the current time feature, tp is the lag feature of the past p moments, Δy t is the differential feature, env_parameters is the environmental parameter;
[0029] Initialize model: Initialize the model and set the initial prediction value to the mean of the target value, expressed as:
[0030]
[0031] Among them, y i is the target value of the i-th sample, and N is the total number of samples;
[0032] Build a tree model: Iteratively train the gradient boosting tree, fit the residual each iteration, and update the model prediction value, expressed as:
[0033] r im =y i -F m-1 (x i );
[0034] Among them, r im is the residual in the mth iteration, F m-1 (x i ) is the predicted value of the m-1th iteration;
[0035] F m (x) = F m-1 (x)+η·h m (x);
[0036] Where η is the learning rate, h m (x) is the predicted value of the mth tree;
[0037] Introducing regularization terms: Adding regularization terms controls model complexity and prevents overfitting, expressed as:
[0038]
[0039] Where l is the mean square error loss function, Ω(h m ) is the complexity regularization term of the mth tree, F M (x i ) is the model in the i-th sample x i The predicted value on M is the final model containing M trees, and λ is the regularization parameter;
[0040] Model prediction: Use the trained model to make predictions, expressed as:
[0041]
[0042] in, is the predicted value of the ith sample.
[0043] Optionally, the feedback control includes:
[0044] Real-time monitoring of environmental parameters: Real-time monitoring and receiving of environmental parameters through multiple sensors;
[0045] User feedback collection: Collect user feedback on lighting effects and light pollution levels, including satisfaction with lighting brightness, lighting color, perceived light pollution, lighting coverage, the impact of light pollution on the living environment, and the impact of light pollution on sleep;
[0046] Control strategy optimization: Based on the predicted results of light pollution and user feedback, the brightness, color and switch status of the light source are dynamically adjusted to continuously optimize the light pollution control strategy.
[0047] Optionally, the operation data collection includes:
[0048] Lighting demand data collection: The lighting demand data of each area is recorded in real time through the lighting control system, including lighting switch time, brightness setting and lighting duration;
[0049] Traffic and pedestrian flow data collection: Traffic sensors and pedestrian flow monitoring equipment installed in various areas monitor and record traffic flow and pedestrian flow data in real time;
[0050] Energy consumption data collection: Real-time monitoring and recording of lighting control system energy consumption data in each area, including power consumption and energy efficiency.
[0051] Optionally, the lighting demand prediction model adopts a deep belief network (DBN) model, and the deep belief network (DBN) model includes:
[0052] Feature collection: Generate time features, lag features, differential features, and combine environmental parameters and operational data, expressed as:
[0053] X t ={t,t-1,t-2,...,tp,Δy t ,env_parameters,operation_data,pollution_adjustments};
[0054] Among them, X t is the feature vector of time step t, t is the current time feature, tp is the time feature of the past p moments, Δy t is the differential feature, env_parameters is the environmental parameter, operation_data is the operation data, and pollution_adjustments is the adjustment result of the light pollution level;
[0055] Introducing the time series processing module: Introducing time series processing, the time series features are input into the deep belief network (DBN) model, expressed as:
[0056]
[0057] Among them, y t-p Represents the light pollution data of the past p moments;
[0058] Construct RBM layers: Each layer of RBM consists of a visible layer and a hidden layer. The characteristics of the data are learned by training each layer of RBM, which is expressed as:
[0059] Energy function of RBM: E(v,h) = -a T vb T Hv T Wh;
[0060] Probability distribution of RBM:
[0061] Hidden layer activation probability: P(h j =1|v)=σ(b j +∑ i v i W ij );
[0062] Where v is the visible layer (input data), h is the hidden layer, a and b are the biases of the visible layer and the hidden layer respectively, W is the connection weight matrix, P(v,h) is the joint probability distribution of the visible layer and the hidden layer, Z is the normalization constant (partition function), P(h j =1|v) is the probability of the jth node in the hidden layer being activated given the visible layer v, σ is the Sigmoid activation function, b j is the bias of the jth node in the hidden layer, v iis the value of the ith node in the visible layer, W ij is the weight of the i-th row and j-th column in the connection weight matrix;
[0063] Stacked RBM layers: Multiple layers of RBM are stacked, and the output of each layer is used as the input of the next layer to gradually extract features, which can be expressed as:
[0064] h (l) =σ(W (l) h (l-1) +b (l) );
[0065] Among them, h (l) is the hidden layer output of layer l, W (l) and b (l) are the weight and bias of the lth layer respectively;
[0066] Fully connected layer and output layer: Add a fully connected layer and an output layer on the last layer of RBM to predict lighting demand, which is expressed as:
[0067] Fully connected layer: h (L) =σ(W (L) h (L-1) +b (L) );
[0068] Output layer: y t =W out ·h (L) +b out ;
[0069] Among them, y t is the predicted value (lighting demand) at time step t, W out is the weight matrix of the output layer, b out is the bias vector of the output layer, h (L) is the output of the fully connected layer.
[0070] Optionally, the dynamic scheduling includes:
[0071] Resource scheduling: Receive demand forecast results, dynamically schedule lighting resources, and adjust the lighting intensity, color temperature, and switch status of each area;
[0072] Lighting intensity adjustment: According to the demand forecast results, adjust the brightness of lighting equipment in each area to provide corresponding lighting intensity in different time periods, including gradual brightness adjustment and regional adjustment;
[0073] Color temperature adjustment: According to the results of demand forecast, adjust the color temperature of lighting equipment in each area to meet the needs of different environments and time periods, including time period color temperature adjustment and regional color temperature adjustment;
[0074] Switch state control: According to the results of demand forecasting, dynamically control the switch state of lighting equipment in each area, and turn off the equipment when lighting is not needed, including time period control, regional control and event-driven control.
[0075] Optionally, the resource optimization includes:
[0076] Energy consumption analysis: Analyze the collected energy consumption data, identify high energy consumption areas and low energy consumption areas, and generate energy consumption reports, expressed as:
[0077]
[0078] Among them, E i is the power consumption of the ith region;
[0079] Energy-saving strategy formulation: Based on the energy consumption analysis results, energy-saving strategies are formulated, including adjusting the operating time, brightness and color temperature of lighting equipment and optimizing the switching status of lighting equipment, expressed as:
[0080]
[0081] Among them, E i is the power consumption of the ith region, η i is the energy efficiency of the ith region, L i is the output light energy of the ith region;
[0082] Optimize lighting resource allocation: Combined with energy consumption management strategy, dynamically adjust the allocation of lighting resources according to actual lighting needs and energy-saving strategies, expressed as:
[0083]
[0084] Among them, R i is the resource allocation of the ith region, D i is the lighting demand of the ith area, W i The weight of the i-th region (determined based on energy consumption analysis and energy-saving strategies);
[0085] Feedback control: Real-time monitoring of the operating status and energy consumption of the lighting control system, and continuous optimization of lighting resource allocation based on feedback data, expressed as:
[0086]
[0087] Among them, ΔR i is the adjustment amount of resource allocation in the ith region, K p is the proportional gain, K i is the integral gain, K d is the differential gain.
[0088] Optionally, the energy management module includes:
[0089] Energy consumption monitoring: real-time monitoring of the power consumption and energy efficiency of the lighting control system;
[0090] Energy consumption record: record and store the monitored energy consumption data to form a historical energy consumption database, expressed as:
[0091] D={E t ,η t ,t};
[0092] Among them, D is the energy consumption data record, E t is the power consumption at time step t, η t is the energy efficiency at time step t, where t is the timestamp;
[0093] Energy consumption prediction: The Bayesian dynamic linear model is used to predict future energy consumption, which is expressed as:
[0094] θ t =F t θ t-1 +v t ;
[0095] y t =G t θ t +∈ t ;
[0096] Among them, θ t is the state vector, F t and G t is the state transfer matrix and the observation matrix, v t and ∈ t is the noise term;
[0097] Energy-saving optimization control: According to the results of energy consumption forecast, dynamically adjust the operating status of lighting equipment, including brightness, color temperature and switch status.
[0098] Beneficial effects of the present invention:
[0099] The present invention realizes an intelligent, high-efficiency, and low-light-pollution urban lighting solution through the collaborative work of multiple modules. First, the data acquisition module provides accurate basic data by real-time monitoring of environmental parameters. The intelligent light pollution control module uses these data to predict light pollution trends and makes dynamic adjustments to ensure that the impact of light pollution is minimized under different environmental conditions. Through real-time monitoring and prediction, the system can comprehensively consider environmental and operational data to improve the accuracy and reliability of lighting control.
[0100] The present invention predicts the lighting demand of each area through a deep belief network model and dynamically schedules lighting resources. By collecting and analyzing historical and real-time operation data, the system can achieve optimal resource allocation and maximize energy efficiency. Combined with intelligent dynamic scheduling, it adjusts the lighting intensity, color temperature and switch status of each area to ensure the most suitable lighting effect in different time periods, thereby improving the overall efficiency of urban lighting and user experience.
[0101] The present invention monitors and optimizes the energy consumption of the lighting control system, records energy consumption data and performs energy consumption analysis and prediction, formulates energy-saving optimization control strategies, further reduces energy consumption costs, uses a Bayesian dynamic linear model to predict future energy consumption, and dynamically adjusts the operating status of the lighting equipment according to the prediction results, including brightness, color temperature and switch status. This not only improves energy utilization efficiency, but also ensures optimal energy efficiency under different environments and usage conditions, providing cities with intelligent, environmentally friendly and efficient lighting solutions, and effectively improving the quality of life of residents and environmental comfort. BRIEF DESCRIPTION OF THE DRAWINGS
[0102] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0103] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention;
[0104] Figure 2 Schematic diagram of a regional demand forecasting and dynamic scheduling module according to an embodiment of the present invention. DETAILED DESCRIPTION
[0105] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.
[0106] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).
[0107] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.
[0108] like Figure 1-Figure 2 As shown, an urban landscape lighting control system based on a composite output mode includes a data acquisition module, an intelligent light pollution control module, a regional demand prediction and dynamic scheduling module, a light source control module and an energy management module, wherein;
[0109] The data acquisition module monitors environmental parameters in real time through multiple sensors, including light intensity, weather conditions, temperature and humidity;
[0110] The intelligent light pollution control module predicts light pollution trends based on real-time monitored environmental parameters and adjusts the light pollution level of urban lighting to ensure that the impact of light pollution is minimized under different environmental conditions;
[0111] The regional demand prediction and dynamic scheduling module predicts the lighting demand of different regions based on the adjustment results of light pollution levels, and dynamically schedules lighting resources to achieve optimal resource allocation and maximize energy efficiency. Specifically, it includes:
[0112] Operational data collection: collect and record historical and real-time operational data of each area, including lighting demand data, traffic and passenger flow data, and energy consumption data;
[0113] Demand forecasting: Using historical and real-time operation data, combined with environmental parameters and based on the adjustment results of light pollution levels, the lighting demand forecasting model is used to predict the lighting demand of each area in different time periods;
[0114] Dynamic scheduling: According to the results of demand forecast, lighting resources are dynamically scheduled to adjust the lighting intensity, color temperature and switch status of each area;
[0115] Resource optimization: Combined with energy management strategies, optimize the allocation of lighting resources to ensure maximum energy efficiency;
[0116] The light source control module connects multiple light sources of different types and achieves diversified lighting effects by adjusting the brightness, color and switch status of the light sources;
[0117] The energy management module monitors and optimizes the energy consumption of the lighting control system, records energy consumption data and performs energy-saving control;
[0118] Through the above content, environmental parameters and light pollution levels are monitored in real time, and operational data is used for demand forecasting and dynamic scheduling to achieve optimal resource allocation and maximize energy efficiency. At the same time, through diversified light source control and effective energy management, it is ensured that efficient, energy-saving, and low-light-pollution intelligent lighting solutions are provided under different environmental conditions.
[0119] The data acquisition module includes:
[0120] Light intensity monitoring: monitor the light intensity of the environment through light sensors;
[0121] Weather Condition Monitoring: Monitoring weather conditions including wind speed, rainfall, and atmospheric pressure through meteorological sensors;
[0122] Temperature monitoring: monitor the ambient temperature through temperature sensors;
[0123] Humidity monitoring: monitor the ambient humidity through humidity sensors;
[0124] Through the above content, lighting strategies can be adjusted in real time to achieve intelligent lighting control with high efficiency and low light pollution, improving lighting effects while optimizing energy consumption management.
[0125] Intelligent light pollution control module includes:
[0126] Light pollution monitoring: receiving real-time monitored environmental parameters and transmitting them in real time to ensure the accuracy and timeliness of environmental parameters;
[0127] Light pollution prediction: Using real-time and historical environmental parameters, the light pollution prediction model is used to predict future light pollution trends;
[0128] Feedback control: Based on the prediction results of light pollution, real-time monitoring of environmental parameters and user feedback, continuous optimization of light pollution control strategies to ensure that the impact of light pollution is minimized under different environmental conditions;
[0129] Through the above content, it is possible to comprehensively consider a variety of environmental and operational data, provide high-precision prediction and control, ensure the optimization of lighting effects under different environmental conditions, reduce the impact of light pollution on residents' lives and the ecological environment, and improve energy utilization efficiency.
[0130] The light pollution prediction model uses a gradient boosting tree model, which includes:
[0131] Feature engineering: Generate time features (year, month, day, hour, minute), lag features (light pollution data for the past few days / hours), differential features (the difference between the current value and the previous value), and combine environmental parameters, expressed as:
[0132] X t ={t,t-1,t-2,...,tp,Δy t ,env_parameters};
[0133] Among them, t is the current time feature, tp is the lag feature of the past p moments, Δy t is the differential feature, env_parameters is the environmental parameter;
[0134] Initialize model: Initialize the model and set the initial prediction value to the mean of the target value, expressed as:
[0135]
[0136] Among them, y i is the target value of the i-th sample, and N is the total number of samples;
[0137] Build a tree model: Iteratively train the gradient boosting tree, fit the residual each iteration, and update the model prediction value, expressed as:
[0138] r im =y i -F m-1 (x i );
[0139] Among them, r im is the residual in the mth iteration, F m-1 (x i ) is the predicted value of the m-1th iteration;
[0140] F m (x) = F m-1 (x)+η·h m (x);
[0141] Where η is the learning rate, h m (x) is the predicted value of the mth tree;
[0142] Introducing regularization terms: Adding regularization terms controls model complexity and prevents overfitting, expressed as:
[0143]
[0144] Where l is the mean square error loss function, Ω(h m ) is the complexity regularization term of the mth tree, F M (x i ) is the model in the i-th sample x i The predicted value on M is the final model containing M trees, and λ is the regularization parameter;
[0145] Model prediction: Use the trained model to make predictions, expressed as:
[0146]
[0147] in, is the predicted value of the i-th sample;
[0148] Through the above content, it is possible to fully capture the dynamic changes of light pollution and improve the accuracy and reliability of predictions. At the same time, it has strong predictive capabilities and efficient computing performance, and provides good model interpretability, which helps to identify and understand the impact of various characteristics on light pollution levels.
[0149] Feedback control includes:
[0150] Real-time monitoring of environmental parameters: Real-time monitoring and receiving of environmental parameters through multiple sensors;
[0151] User feedback collection: Collect user feedback on lighting effects and light pollution levels, including satisfaction with lighting brightness, lighting color, perceived light pollution, lighting coverage, the impact of light pollution on the living environment, and the impact of light pollution on sleep;
[0152] Control strategy optimization: According to the light pollution prediction results and user feedback, the brightness, color and switch status of the light source are dynamically adjusted, and the light pollution control strategy is continuously optimized to ensure that the impact of light pollution is minimized under different environmental conditions;
[0153] Through the above steps, it is possible to monitor environmental changes and user needs in real time based on the prediction results of light pollution, continuously improve and optimize light pollution control strategies, and provide efficient and low-light-pollution lighting effects.
[0154] Operational data collection includes:
[0155] Lighting demand data collection: The lighting demand data of each area is recorded in real time through the lighting control system, including lighting switch time, brightness setting and lighting duration;
[0156] Traffic and pedestrian flow data collection: Traffic sensors and pedestrian flow monitoring equipment installed in various areas monitor and record traffic flow and pedestrian flow data in real time;
[0157] Energy consumption data collection: Real-time monitoring and recording of lighting control system energy consumption data in each area, including power consumption and energy efficiency, including:
[0158] Power consumption is calculated by measuring current, voltage, and usage time and is expressed as:
[0159] E = P × t;
[0160] Where E is the electricity consumption in kilowatt-hours (kWh), P is the power in kilowatts (kW), and t is the time in hours (h);
[0161]
[0162] Where V is voltage in volts (V), I is current in amperes (A), is the power factor, which is used to take into account the phase difference between current and voltage;
[0163] Energy efficiency measures the ratio of output light energy to input electrical energy, expressed as:
[0164]
[0165] Where η is the energy efficiency, L is the output light energy in lumens (Im), and E is the input electrical energy in watts (W);
[0166] The above content will help to make accurate demand forecasts and optimize light pollution control strategies, improve the operating efficiency of lighting systems, reduce energy consumption, and ensure that urban lighting meets actual needs, thereby improving residents' quality of life and environmental comfort.
[0167] The lighting demand prediction model uses a deep belief network (DBN) model, which includes:
[0168] Feature collection: Generate time features, lag features, differential features, and combine environmental parameters and operational data, expressed as:
[0169] X t =
[0170] {t,t-1,t-2,...,tp,Δy t ,env_parameters,operation_data,pollution_adjustments};
[0171] Among them, X t is the feature vector of time step t, t is the current time feature, tp is the time feature of the past p moments, Δy t is the differential feature, that is, the difference between the current value and the previous value, env_parameters is the environmental parameters, operation_data is the operation data, and pollution_adjustments is the adjustment result of the light pollution level;
[0172] Introducing the time series processing module: In order to capture time dependency, time series processing is introduced, and the time series features are input into the deep belief network (DBN) model, which is expressed as:
[0173]
[0174] Among them, y t-p Represents the light pollution data of the past p moments;
[0175] Construct RBM layers: Each layer of RBM consists of a visible layer and a hidden layer. The characteristics of the data are learned by training each layer of RBM, which is expressed as:
[0176] Energy function of RBM: E(v,h) = -a T vb T Hv T Wh;
[0177] Probability distribution of RBM:
[0178] Hidden layer activation probability: P(h j =1|v)=σ(b j +∑ i v i W ij );
[0179] Where v is the visible layer (input data), h is the hidden layer, a and b are the biases of the visible layer and the hidden layer respectively, W is the connection weight matrix, P(v,h) is the joint probability distribution of the visible layer and the hidden layer, Z is the normalization constant (partition function), P(h j =1|v) is the probability of the jth node in the hidden layer being activated given the visible layer v, σ is the Sigmoid activation function, b j is the bias of the jth node in the hidden layer, v i is the value of the ith node in the visible layer, W ij is the weight of the i-th row and j-th column in the connection weight matrix;
[0180] Stacked RBM layers: Multiple layers of RBM are stacked, and the output of each layer is used as the input of the next layer to gradually extract features, which can be expressed as:
[0181] h (l) =σ(W (l) h (l-1) +b (l) );
[0182] Among them, h (l) is the hidden layer output of layer l, W (l) and b (l) are the weight and bias of the lth layer respectively;
[0183] Fully connected layer and output layer: Add a fully connected layer and an output layer on the last layer of RBM to predict lighting demand, which is expressed as:
[0184] Fully connected layer: h (L) =σ(W (L) h (L-1) +b (L) );
[0185] Output layer: y t =W out · (L) +b out ;
[0186] Among them, y t is the predicted value (lighting demand) at time step t, W out is the weight matrix of the output layer, b out is the bias vector of the output layer, h (L) is the output of the fully connected layer;
[0187] Through the above content, the introduction of multivariate features and time series processing can capture the time dependence and complex high-order features in the data, improve the accuracy and robustness of lighting demand prediction, and enable the model to accurately predict lighting demand under different environmental conditions, optimize light pollution control strategies, and provide efficient and accurate lighting solutions.
[0188] Dynamic scheduling includes:
[0189] Resource scheduling: Receive demand forecast results, dynamically schedule lighting resources, and adjust the lighting intensity, color temperature, and switch status of each area to meet the predicted lighting needs;
[0190] Lighting intensity adjustment: According to the demand forecast results, adjust the brightness of lighting equipment in each area to provide corresponding lighting intensity in different time periods, including gradual brightness adjustment and regional adjustment;
[0191] Color temperature adjustment: According to the results of demand forecast, adjust the color temperature of lighting equipment in each area to meet the needs of different environments and time periods, including time period color temperature adjustment and regional color temperature adjustment;
[0192] Switch state control: According to the demand forecast results, dynamically control the switch state of lighting equipment in each area, and turn off the equipment when lighting is not needed to save energy, including time period control, regional control and event-driven control;
[0193] Gradually adjust the brightness Based on the refined results of demand forecasting, gradually adjust the brightness of lighting equipment from low to high or from high to low to provide lighting intensity that adapts to the current environment and activity needs, including:
[0194] During the day: During the day when there is plenty of natural light, reduce the brightness of artificial lighting to save energy;
[0195] Evening and early morning: During the period when light gradually changes, adjust the lighting brightness appropriately to synchronize it with the changes in natural light;
[0196] Nighttime: Provide adequate lighting brightness to ensure safety during peak activity hours at night, and reduce brightness to reduce energy consumption during late night when there are fewer people.
[0197] Regional adjustment adjusts the brightness of each area according to the lighting needs of different areas to achieve local optimization, including:
[0198] Commercial area: Provide higher brightness during the period of busy commercial activities, and reduce the brightness after the commercial activities are over;
[0199] Residential areas: During rest time in residential areas, reduce brightness to reduce light pollution, and during peak travel hours in the morning and evening, increase brightness to improve safety;
[0200] Parks and attractions: Provide adequate brightness during peak visitor hours and turn off or reduce the brightness of lighting equipment during closing hours.
[0201] Time period color temperature adjustment Adjust the color temperature according to different time periods of the day to adapt to the changes in natural light and human physiological rhythms, including:
[0202] Morning and daytime: Use cool white light with a higher color temperature (such as 5000K-6500K) to simulate natural light, which helps improve concentration and work efficiency;
[0203] Evening and night: Use warm white light with a lower color temperature (such as 2700K-3500K) to provide a comfortable light environment to help you relax and prepare for sleep.
[0204] Regional color temperature adjustment adjusts the color temperature of each area according to the functional requirements of different areas to provide the best visual effects and environmental atmosphere, including:
[0205] Commercial areas: In shops and office areas, use medium to high color temperature (4000K-5000K) to create a bright and clear environment;
[0206] Residential areas: In public spaces in residential areas, use warm light with a lower color temperature to create a warm and comfortable atmosphere;
[0207] Parks and attractions: Choose the appropriate color temperature based on the landscape and activity requirements to enhance the landscape effect and visual appeal.
[0208] Time period control sets the on / off time of lighting equipment in each area according to the predicted lighting demand and usage time, including:
[0209] Daytime shutdown: Turn off unnecessary artificial lighting during the daytime hours when there is plenty of natural light;
[0210] Turn on at night: Turn on lighting equipment during peak activity hours at night to provide a safe lighting environment; turn off or partially turn off lighting equipment when the flow of people decreases late at night.
[0211] Regional control controls the on / off status of lighting equipment in each area according to the usage of different areas to achieve local optimization, including:
[0212] Commercial areas: turn off unnecessary lighting after commercial activities are over; keep necessary lighting on during cleaning and restocking hours;
[0213] Residential areas: turn off or reduce the brightness of public lighting equipment during rest time in residential areas; turn on necessary lighting equipment during peak travel hours in the morning and evening;
[0214] Parks and attractions: Turn off all landscape lighting during closing hours; gradually turn on lighting during opening hours to provide an appropriate light environment.
[0215] Event-driven control dynamically adjusts the on / off status of lighting equipment according to the needs of special events or activities (such as festivals, emergencies, etc.), including:
[0216] Festivals: During festivals, turn on more lighting equipment to enhance the festive atmosphere; after the event, turn off unnecessary lighting in a timely manner;
[0217] Emergency: In emergency situations (such as sudden accidents, disasters, etc.), turn on or off lighting equipment as needed to provide necessary light environment support.
[0218] Resource optimization includes:
[0219] Energy consumption analysis: Analyze the collected energy consumption data, identify high energy consumption areas and low energy consumption areas, and generate energy consumption reports, expressed as:
[0220]
[0221] Among them, E i is the power consumption of the ith region;
[0222] Energy-saving strategy formulation: Based on the energy consumption analysis results, energy-saving strategies are formulated, including adjusting the operating time, brightness and color temperature of lighting equipment and optimizing the switching status of lighting equipment, expressed as:
[0223]
[0224] Among them, E i is the power consumption of the ith region, η i is the energy efficiency of the ith region, L i is the output light energy of the ith region;
[0225] Optimize lighting resource allocation: Combined with energy consumption management strategies, dynamically adjust the allocation of lighting resources according to actual lighting needs and energy-saving strategies to ensure maximum energy efficiency, expressed as:
[0226]
[0227] Among them, R i is the resource allocation of the ith region, D i is the lighting demand of the ith area, W i The weight of the i-th region (determined based on energy consumption analysis and energy-saving strategies);
[0228] Feedback control: Real-time monitoring of the operating status and energy consumption of the lighting control system, continuous optimization of lighting resource allocation based on feedback data, and ensuring optimal energy efficiency under different environments and usage conditions, expressed as:
[0229]
[0230] Among them, ΔR i is the adjustment amount of resource allocation in the ith region, K p is the proportional gain, K i is the integral gain, K d is the differential gain;
[0231] Through the above steps, it is possible to combine energy management strategies to intelligently optimize the allocation of lighting resources, improve energy efficiency, reduce energy costs, and ensure optimal energy efficiency under different environments and usage conditions.
[0232] The energy management module includes:
[0233] Energy consumption monitoring: real-time monitoring of the power consumption and energy efficiency of the lighting control system;
[0234] Energy consumption record: record and store the monitored energy consumption data to form a historical energy consumption database, expressed as:
[0235] D={E t ,η t ,t};
[0236] Among them, D is the energy consumption data record, E t is the power consumption at time step t, η tis the energy efficiency at time step t, where t is the timestamp;
[0237] Energy consumption prediction: The Bayesian dynamic linear model is used to predict future energy consumption, which is expressed as:
[0238] θ t =F t θ t-1 +v t ;
[0239] y t =G t θ t +∈ t ;
[0240] Among them, θ t is the state vector, F t and G t is the state transfer matrix and the observation matrix, v t and ∈ t is the noise term;
[0241] Energy-saving optimization control: According to the results of energy consumption prediction, dynamically adjust the operating status of lighting equipment, including brightness, color temperature and switch status, to optimize energy consumption and achieve energy-saving goals;
[0242] Through the above content, not only the energy utilization efficiency is improved and the energy consumption cost is reduced, but also the optimal energy efficiency is ensured under different environments and usage conditions, providing the city with a more intelligent, environmentally friendly and efficient lighting solution.
[0243] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.
[0244] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. An urban landscape lighting control system based on a composite output mode, characterized in that: It includes data acquisition module, intelligent light pollution control module, regional demand forecasting and dynamic scheduling module, light source control module and energy management module, among which; The data acquisition module monitors environmental parameters in real time through multiple sensors, including light intensity, weather conditions, temperature and humidity; The intelligent light pollution control module predicts light pollution trends and adjusts the light pollution level of urban lighting based on real-time monitored environmental parameters; The regional demand prediction and dynamic scheduling module predicts the lighting demand of different areas based on the adjustment results of the light pollution level and dynamically schedules lighting resources, specifically including: Operational data collection: collect and record historical and real-time operational data of each area, including lighting demand data, traffic and passenger flow data, and energy consumption data; Demand forecasting: Using historical and real-time operation data, combined with environmental parameters, and based on the adjustment results of light pollution levels, the lighting demand forecasting model is used to predict the lighting demand of each area in different time periods; Dynamic scheduling: According to the results of demand forecast, lighting resources are dynamically scheduled to adjust the lighting intensity, color temperature and switch status of each area; Resource optimization: Combine energy management strategies to optimize the allocation of lighting resources; The light source control module is connected to a plurality of light sources of different types, and achieves diversified lighting effects by adjusting the brightness, color and switch status of the light sources; The energy management module monitors and optimizes the energy consumption of the lighting control system, records energy consumption data and performs energy-saving control.
2. The urban landscape lighting control system based on the composite output mode according to claim 1 is characterized in that: The data acquisition module comprises: Light intensity monitoring: monitor the light intensity of the environment through light sensors; Weather Condition Monitoring: Monitoring weather conditions including wind speed, rainfall, and atmospheric pressure through meteorological sensors; Temperature monitoring: monitor the ambient temperature through temperature sensors; Humidity monitoring: Monitor the ambient humidity through a humidity sensor.
3. The urban landscape lighting control system based on a composite output mode according to claim 1, characterized in that: The intelligent light pollution control module comprises: Light pollution monitoring: receiving real-time monitored environmental parameters and transmitting them in real time; Light pollution prediction: Using real-time and historical environmental parameters, the light pollution prediction model is used to predict future light pollution trends; Feedback control: Based on the prediction results of light pollution, real-time monitoring of environmental parameters and user feedback is carried out to continuously optimize the light pollution control strategy.
4. The urban landscape lighting control system based on the composite output mode according to claim 3 is characterized in that: The light pollution prediction model adopts a gradient boosting tree model, and the gradient boosting tree model includes: Feature engineering: Generate time features, lag features, differential features and combine environmental parameters, expressed as: X t ={t,t-1,t-2,...,t-p,Δy t ,env_parameters}; Among them, t is the current time feature, tp is the lag feature of the past p moments, Δy t is the differential feature, env_parameters is the environmental parameter; Initialize the model: Initialize the model and set the initial prediction value to the mean of the target value, expressed as: Among them, y i is the target value of the i-th sample, and N is the total number of samples; Build a tree model: Iteratively train the gradient boosting tree, fit the residual each iteration, and update the model prediction value, expressed as: r im =y i -F m-1 (x i ); Among them, r im is the residual in the mth iteration, F m-1 (x i ) is the predicted value of the m-1th iteration; F m (x)=F m-1 (x)+η·h m (x); Where η is the learning rate, h m (x) is the predicted value of the mth tree; Introducing regularization terms: Adding regularization terms controls model complexity and prevents overfitting, expressed as: Where l is the mean square error loss function, Ω(h m ) is the complexity regularization term of the mth tree, F M (x i ) is the model in the i-th sample x i The predicted value on M is the final model containing M trees, and λ is the regularization parameter; Model prediction: Use the trained model to make predictions, expressed as: in, is the predicted value of the ith sample.
5. The urban landscape lighting control system based on the composite output mode according to claim 4 is characterized in that: The feedback control comprises: Real-time monitoring of environmental parameters: Real-time monitoring and receiving of environmental parameters through multiple sensors; User feedback collection: Collect user feedback on lighting effects and light pollution levels, including satisfaction with lighting brightness, lighting color, perceived light pollution, lighting coverage, the impact of light pollution on the living environment, and the impact of light pollution on sleep; Control strategy optimization: Based on the predicted results of light pollution and user feedback, the brightness, color and switch status of the light source are dynamically adjusted to continuously optimize the light pollution control strategy.
6. The urban landscape lighting control system based on a composite output mode according to claim 1, characterized in that: The operational data collection includes: Lighting demand data collection: The lighting control system records the lighting demand data of each area in real time, including lighting switch time, brightness setting and lighting duration; Traffic and pedestrian flow data collection: Traffic sensors and pedestrian flow monitoring equipment installed in various areas monitor and record traffic flow and pedestrian flow data in real time; Energy consumption data collection: Real-time monitoring and recording of lighting control system energy consumption data in each area, including power consumption and energy efficiency.
7. The urban landscape lighting control system based on the composite output mode according to claim 6 is characterized in that: The lighting demand prediction model adopts a deep belief network model, and the deep belief network model includes: Feature collection: Generate time features, lag features, differential features, and combine environmental parameters and operational data, expressed as: X t ={t,t-1,t-2,...,t-p,Δy t ,env_parameters,operation_data,pollution_adjustments}; Among them, X t is the feature vector of time step t, t is the current time feature, tp is the time feature of the past p moments, Δy t is the differential feature, env_parameters is the environmental parameter, operation_data is the operation data, and pollution_adjustments is the adjustment result of the light pollution level; Introducing the time series processing module: Introducing time series processing, the time series features are input into the deep belief network model, expressed as: Among them, y t-p Represents the light pollution data of the past p moments; Construct RBM layers: Each layer of RBM consists of a visible layer and a hidden layer. The characteristics of the data are learned by training each layer of RBM, which is expressed as: Energy function of RBM: Probability distribution of RBM: Hidden layer activation probability: P(h j =1|v)=σ(b j +∑ i v i W ij ); Where v is the visible layer, h is the hidden layer, a and b are the biases of the visible layer and the hidden layer respectively, W is the connection weight matrix, P(v,h) is the joint probability distribution of the visible layer and the hidden layer, Z is the normalization constant, and P(h j =1|v) is the probability of the jth node in the hidden layer being activated given the visible layer v, σ is the Sigmoid activation function, b j is the bias of the jth node in the hidden layer, v i is the value of the ith node in the visible layer, W ij is the weight of the i-th row and j-th column in the connection weight matrix; Stacked RBM layers: Multiple layers of RBM are stacked, and the output of each layer is used as the input of the next layer to gradually extract features, which can be expressed as: h (l) =σ(W (l) h (l-1) +b (l) ); Among them, h (l) is the hidden layer output of layer l, W ( l ) and b ( l ) are the weight and bias of the lth layer respectively; Fully connected layer and output layer: Add a fully connected layer and an output layer on the last layer of RBM to predict lighting demand, which is expressed as: Fully connected layer: h (L) =σ(W (L) h (L-1) +b (L) ); Output layer: y t =W out ·h (L) +b out ; Among them, y t is the predicted value at time step t, W out is the weight matrix of the output layer, b out is the bias vector of the output layer, h (L) is the output of the fully connected layer.
8. The urban landscape lighting control system based on the composite output mode according to claim 7 is characterized in that: The dynamic scheduling includes: Resource scheduling: Receive demand forecast results, dynamically schedule lighting resources, and adjust the lighting intensity, color temperature, and switch status of each area; Lighting intensity adjustment: According to the demand forecast results, adjust the brightness of lighting equipment in each area, including gradual brightness adjustment and regional adjustment; Color temperature adjustment: According to the results of demand forecast, adjust the color temperature of lighting equipment in each area, including time period color temperature adjustment and regional color temperature adjustment; Switch state control: According to the results of demand forecasting, dynamically control the switch state of lighting equipment in each area, including time period control, regional control and event-driven control.
9. The urban landscape lighting control system based on the composite output mode according to claim 8 is characterized in that: The resource optimization includes: Energy consumption analysis: Analyze the collected energy consumption data, identify high energy consumption areas and low energy consumption areas, and generate energy consumption reports, expressed as: Among them, E i is the power consumption of the ith region; Energy-saving strategy formulation: Based on the energy consumption analysis results, energy-saving strategies are formulated, including adjusting the operating time, brightness and color temperature of lighting equipment and optimizing the switching status of lighting equipment, expressed as: Among them, E i is the power consumption of the ith region, η i is the energy efficiency of the ith region, L i is the output light energy of the ith region; Optimize lighting resource allocation: Combined with energy consumption management strategy, dynamically adjust the allocation of lighting resources according to actual lighting needs and energy-saving strategies, expressed as: Among them, R i is the resource allocation of the ith region, D i is the lighting demand of the ith area, W i The weight of the i-th region; Feedback control: Real-time monitoring of the operating status and energy consumption of the lighting control system, and continuous optimization of lighting resource allocation based on feedback data, expressed as: Among them, ΔR i is the adjustment amount of resource allocation in the ith region, K p is the proportional gain, K i is the integral gain, K d is the differential gain.
10. The urban landscape lighting control system based on a composite output mode according to claim 1, characterized in that: The energy management module comprises: Energy consumption monitoring: real-time monitoring of the power consumption and energy efficiency of the lighting control system; Energy consumption record: record and store the monitored energy consumption data to form a historical energy consumption database, expressed as: D={E t ,η t ,t}; Among them, D is the energy consumption data record, E t is the power consumption at time step t, η t is the energy efficiency at time step t, where t is the timestamp; Energy consumption prediction: The Bayesian dynamic linear model is used to predict future energy consumption, which is expressed as: i t =F t i t-1 +v t ; y t =G t i t +∈ t ; Among them, θ t is the state vector, F t and G t is the state transfer matrix and the observation matrix, v t and ∈ t is the noise term; Energy-saving optimization control: According to the results of energy consumption forecast, dynamically adjust the operating status of lighting equipment, including brightness, color temperature and switch status.
Citation Information
Cited By
Intelligent scene linkage edge computing gateway
CN119865515A
