Intelligent LED illumination energy-saving optimization method based on AI edge calculation

Abstract

The invention provides an intelligent LED lighting energy-saving optimization method based on AI edge calculation, and relates to the field of LED lighting energy-saving optimization. The intelligent LED illumination energy-saving optimization method based on AI edge calculation comprises the following specific steps: step 1, technical improvement: high-purity GaN and InGaN materials should be adopted for an existing light source module; gaps of the heat dissipation module are filled with heat conduction materials, a heat dissipation system is optimized, and efficiency reduction caused by high temperature is avoided; 2, intelligent control is conducted, specifically, the lighting environment is sensed through an optical sensor, a human body infrared sensor, a microwave radar sensor and a human body existence sensor. High-purity GaN and InGaN materials are adopted to improve the luminous efficiency, and the flip chip structure is combined to reduce the optical loss; according to the heat dissipation module, gaps are filled with heat conduction materials, a heat dissipation system is optimized, efficiency reduction caused by high temperature is avoided, energy-saving use can be optimized through the structure, meanwhile, the lighting body can be optimized, and the heat dissipation efficiency and the light emitting efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving optimization of LED lighting, and specifically to an intelligent LED lighting energy-saving optimization method based on AI edge computing. Background Art

[0002] With the rapid development of intelligent technology and the wide popularization of the concept of energy conservation and emission reduction, the optimization of LED lighting energy conservation has become increasingly significant in the field of building lighting, and the requirements for its energy-saving effect and the accuracy of intelligent control have also become increasingly stringent. The intelligent LED lighting energy-saving optimization method and system based on AI computing bring new challenges and opportunities to the design and management of intelligent building lighting due to their key role in reducing lighting energy consumption and improving lighting comfort. Efficient energy-saving optimization can ensure the stable operation of the lighting system, reduce energy waste, and meet the requirements of green and low-carbon buildings.

[0003] The existing intelligent LED lighting energy-saving optimization mainly includes lighting environment monitoring, data transmission and storage, AI computing analysis, and lighting strategy formulation. Through the monitoring of the lighting environment, various types of sensors and monitoring devices are set in each area of the building, such as corridors, offices, meeting rooms, etc., to collect data such as light intensity, personnel activities, time information, and natural light changes in the lighting environment, so as to achieve extensive monitoring of the current lighting situation; through data transmission and storage, the collected data is processed using wireless transmission technology and cloud storage platforms to ensure the stability and integrity of the data during transmission and storage; through the analysis of AI computing, according to the lighting area function, personnel usage habits, and lighting demand characteristics at different time periods, specific AI computing is used to deeply analyze the data, and possible lighting energy-saving optimization models and strategy frameworks are constructed to provide a basis for determining lighting strategies; through the formulation of lighting strategies, managers can intuitively understand lighting optimization-related information and lighting system responses, achieve real-time adjustment and optimization of lighting strategies, timely correct lighting plans, and improve the scientificity and adaptability of lighting energy-saving optimization.

[0004] The LED lighting energy-saving optimization method in the prior art lacks improvement in technology during the optimization process of LED lighting, resulting in a decrease in lighting efficiency and an increase in energy consumption during use when the LED lamp body gets hot during high-efficiency energy conservation, as well as problems such as usage errors in the dimming strategy for lighting during different peak hours of electricity consumption, inaccurate collected data, and inaccurate analysis results. Summary of the Invention

[0005] Technical Problems to be Solved In view of the deficiencies of the prior art, the present invention provides an intelligent LED lighting energy-saving optimization method based on AI edge computing, which solves the problems of the lack of improvement in the prior art, resulting in high temperature when the LED lighting body is used during the process of high-efficiency energy saving, still causing a decrease in lighting efficiency and an increase in energy consumption, as well as the problem of use error in the dimming strategy for lighting during different peak hours.

[0006] Technical solution To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent LED lighting energy-saving optimization method based on AI edge computing includes the following specific methods: Step 1: Technical improvement: The existing light source module should use high-purity GaN and InGaN materials; The heat dissipation module uses heat-conducting materials to fill the gaps and optimize the heat dissipation system to avoid a decrease in efficiency caused by high temperature: Step 2: Intelligent control: The lighting environment is sensed through a light sensor, a human infrared sensor, a microwave radar sensor, and a human presence sensor, and data collection is completed; Step 3: System management: The intelligent detection uses a photoresistor / photodiode to collect the ambient light intensity in the range of 0 - 2000 lux in real time, and a sliding window filter is used to eliminate instantaneous interference to achieve ambient light monitoring; The environmental perception uses a PIR sensor to capture moving targets, and a 24GHz millimeter-wave radar to identify the breathing signals of stationary humans; The scene management uses spatial dimension division to divide functional areas based on a GIS map; Step 4: AI calculation: The LED data analysis is used to analyze the collected target data, filter out the noise, and analyze the human activity signals; The LED data evaluation is used to measure the input power factor through a power quality analyzer and evaluate the power conversion efficiency; The behavior prediction module can use an LSTM neural network to learn the historical movement trajectory, predict the use demand of the meeting room 10 - 15 minutes in advance, and turn on the lighting; AI suggestion: AI will give real-time energy efficiency optimization and environmental adaptive adjustment. Based on the current / voltage sensor data, it will automatically recommend the best driving power, such as reducing the output by 30% during the off-peak hours at night. Combining with the time-of-use electricity price model to generate a dimming strategy, it will preferentially turn off unnecessary lighting during the peak electricity price stage, and dynamically recommend the regional lighting level by identifying the crowd density through the human presence sensor.

[0007] Preferably, in step 3, the spatial dimension division divides functional areas such as the office area, corridor, and parking lot based on a GIS map, and sets different illumination standards.

[0008] Preferably, in step 4, the LED data evaluation is used to record the voltage / current fluctuation range and evaluate the output stability of the drive power supply; the energy consumption of traditional lighting and the experimental group of LEDs is monitored synchronously, and the power saving rate is calculated.

[0009] Preferably, the function of the human presence sensor in step 2 is as follows: in the meeting room, the presence sensor is used to identify the "false unoccupied state", such as the sedentary participants, to avoid mis-turning off the lights.

[0010] An intelligent LED lighting energy-saving optimization system based on AI edge computing includes technical improvement, intelligent control, system management, and AI calculation; The technical improvement is used to improve the luminous efficiency by using high-purity GaN and InGaN materials for the existing light source module, and reduce the light loss by combining with the flip-chip structure; The intelligent control: includes a collection module, and light, human body infrared, microwave radar, and human presence sensors. Each sensor is used to complete the intelligent street lamp and garage lighting in different environments. Through the combination of illuminance + infrared + temperature sensors, a comprehensive energy saving of 40%-60% is achieved; The system management specifically further includes an intelligent monitoring module, an environmental perception module, and a scene management module, which are used to collect the light intensity in the range of 0-2000 lux in real time by using a photoresistor / photodiode, and eliminate instantaneous interference by using a sliding window filter to achieve environmental light monitoring; The AI calculation includes an LED data analysis module, an LED data evaluation module, a behavior prediction module, and AI suggestions, which are used to analyze the data collected by the LED, separate the human activity data from the noise, and through the evaluation of the data, measure the input power factor by using a power quality analyzer, evaluate the power conversion efficiency, and complete the behavior prediction function through the behavior prediction module, and give an automatic recommendation of the best drive power through AI, such as reducing the output by 30% during the non-peak period at night, and generating a dimming strategy in combination with the time-of-use electricity price model.

[0011] Beneficial effects The present invention provides an intelligent LED lighting energy-saving optimization method based on AI edge computing. It has the following beneficial effects: The present invention is provided with a light source module and a heat dissipation module. For the existing light source module, high-purity GaN and InGaN materials should be used to improve the luminous efficiency, and the flip-chip structure is combined to reduce the light loss; the heat dissipation module uses a heat-conducting material to fill the gap and optimize the heat dissipation system to avoid the efficiency decline caused by high temperature. Through this structure, while optimizing the energy-saving use, the lighting body can also be optimized to improve the heat dissipation and luminous efficiency.

[0012] The present invention is provided with: an intelligent detection module, an environmental perception module, and a scene management module. The intelligent detection uses a photoresistor / photodiode to collect the ambient light intensity in the range of 0 - 2000 lux in real time, and uses a sliding window filter to eliminate instantaneous interference to achieve ambient light monitoring. The environmental perception uses a PIR sensor to capture moving targets, and a 24GHz millimeter-wave radar to identify the respiration signal of stationary humans. The scene management uses spatial dimension division to divide functional areas based on a GIS map, such as office areas, corridors, and parking lots, and sets differential illuminance standards, which can improve the energy saving of LED lighting to achieve the optimal optimization method.

[0013] The present invention is provided with: AI will give real-time energy efficiency optimization and environmental adaptive adjustment. Based on the current / voltage sensor data, it will automatically recommend the best driving power, such as reducing the output by 30% during off-peak hours at night, and generate a dimming strategy in combination with the time-of-use electricity price model. During peak electricity price periods, non-essential lighting will be turned off first. The occupancy density of people is identified through a human presence sensor, and the regional lighting level is dynamically recommended. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flowchart of an intelligent LED lighting energy-saving optimization system based on AI edge computing proposed by the present invention; Figure 2 It is an AI calculation flowchart in an intelligent LED lighting energy-saving optimization system based on AI edge computing proposed by the present invention. SPECIFIC EMBODIMENTS

[0015] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0016] Embodiment 1: The embodiment of the present invention provides an intelligent LED lighting energy-saving optimization method based on AI edge computing, including the following specific methods: Step 1: Technical improvement: High-purity GaN and InGaN materials should be used for the existing light source module; The heat dissipation module uses a heat-conducting material to fill the gap and optimize the heat dissipation system to avoid efficiency decline caused by high temperature: Step 2: Intelligent control: The lighting environment is sensed through a light sensor, a passive infrared sensor, a microwave radar sensor, and a human presence sensor, and data collection is completed. The function of the human presence sensor is: in the meeting room, the presence sensor is used to identify the "pseudo-unoccupied state", such as sitting participants, to avoid mis-turning off the lights; Step 3: System Management: The intelligent detection uses a photoresistor / photodiode to collect the light intensity in the range of 0 - 2000 lux in real time, and uses a sliding window filter to eliminate instantaneous interference to achieve ambient light monitoring; The environmental perception uses a PIR sensor to capture moving targets, and a 24GHz millimeter-wave radar to identify the breathing signals of stationary humans; The scene management divides the functional areas such as the office area, corridor, and parking lot based on the GIS map in terms of spatial dimensions, and sets different illumination standards; Step 4: AI Calculation: The LED data analysis is used to analyze the collected target data, filter out the noise, and analyze the human activity signals; The LED data evaluation is used to measure the input power factor through a power quality analyzer, evaluate the power conversion efficiency, and the LED data evaluation is used to record the voltage / current fluctuation range and evaluate the output stability of the drive power supply; The energy consumption of traditional lighting and the experimental group LED is monitored synchronously to calculate the power saving rate; The behavior prediction module can use the LSTM neural network to learn the historical movement trajectory, predict the usage demand of the meeting room 10 - 15 minutes in advance and turn on the lighting; AI Suggestions: AI will give real-time energy efficiency optimization and environmental adaptive adjustment. Based on the current / voltage sensor data, it will automatically recommend the best drive power, such as reducing the output by 30% during the off-peak period at night. Combining the time-of-use electricity price model to generate a dimming strategy, giving priority to turning off unnecessary lighting during the peak electricity price stage. Identifying the crowd density through the human presence sensor and dynamically recommending the regional lighting level. The machine learning model analyzes the historical energy consumption data and automatically generates the optimal brightness curve. For example, the tunnel lighting system realizes segmented dimming through AI edge computing to avoid the energy waste of the traditional "fully on and fully off".

[0017] Embodiment 2: Such as Figure 1-2 As shown, the difference between this embodiment and Embodiment 1 is that an intelligent LED lighting energy-saving optimization system based on AI edge computing includes technical improvement, intelligent control, system management, and AI calculation; The technical improvement is used to improve the luminous efficiency by using high-purity GaN and InGaN materials for the existing light source module, and combining the flip-chip structure to reduce light loss; The intelligent control: includes a collection module and sensors such as light, human infrared, microwave radar, and human presence sensors. For different environmental systems of smart street lights and garage lighting, through the combination of illuminance + infrared + temperature sensors, comprehensive energy saving of 40% - 60% is achieved; The system management specifically further includes an intelligent monitoring module, an environmental perception module, and a scenario management module, which are used to collect the light intensity in the range of 0-2000 lux in real time by using a photoresistor / photodiode, and eliminate instantaneous interference by using a sliding window filter to achieve ambient light monitoring; The AI calculation includes an LED data analysis module, an LED data evaluation module, a behavior prediction module, and AI suggestions, which are used to analyze the data collected by the LED, separate the human activity data from the noise, and use the evaluation of the data to measure the input power factor by a power quality analyzer, evaluate the power conversion efficiency, and complete the prediction of behaviors through the behavior prediction module, and give an automatic recommendation for the best driving power through AI, such as reducing the output by 30% during the off-peak period at night, and generating a dimming strategy in combination with the time-of-use electricity price model.

[0018] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent LED lighting energy-saving optimization method based on AI edge computing, characterized in that: It includes the following specific methods: Step 1: Technical improvement: High-purity GaN and InGaN materials should be used for the existing light source module; The heat dissipation module uses heat-conducting materials to fill the gaps and optimize the heat dissipation system to avoid efficiency reduction caused by high temperature: Step 2: Intelligent control: The lighting environment is sensed through a light sensor, a human infrared sensor, a microwave radar sensor, and a human presence sensor, and data collection is completed; Step 3: System management: Intelligent detection uses a photoresistor / photodiode to collect the illumination intensity in the range of 0-2000 lux in real time, and a sliding window filter is used to eliminate instantaneous interference to achieve ambient light monitoring; Environmental perception uses a PIR sensor to capture moving targets, and a 24GHz millimeter-wave radar to identify the breathing signals of stationary humans; Scene management uses spatial dimension division to divide functional areas based on a GIS map; Step 4: AI calculation: LED data analysis is used to analyze the collected target data, filter out noise, and analyze human activity signals; LED data evaluation is used to measure the input power factor through a power quality analyzer and evaluate the power conversion efficiency; The behavior prediction module can use an LSTM neural network to learn the historical movement trajectory, predict the meeting room usage requirements 10-15 minutes in advance, and turn on the lighting; AI suggestions: AI will give real-time energy efficiency optimization and environmental adaptive adjustment. Based on the current / voltage sensor data, it will automatically recommend the best drive power, such as reducing the output by 30% during off-peak hours at night. Combining with the time-of-use electricity price model to generate a dimming strategy, it will give priority to turning off unnecessary lighting during peak electricity price periods, and dynamically recommend the regional lighting level by identifying the crowd density through the human presence sensor.

2. The intelligent LED lighting energy-saving optimization method based on AI edge computing according to claim 1, wherein: In the said Step 3, the spatial dimension division divides functional areas such as the office area, corridor, and parking lot based on a GIS map, and sets different illumination standards.

3. An intelligent LED lighting energy-saving optimization method based on AI edge computing according to claim 1, characterized in that: In the said Step 4, the LED data evaluation is used to record the voltage / current fluctuation range and evaluate the output stability of the drive power supply; The energy consumption of traditional lighting and the experimental group of LEDs is monitored synchronously, and the power saving rate is calculated.

4. An intelligent LED lighting energy-saving optimization method based on AI edge computing according to claim 1, characterized in that: The function of the human presence sensor in the said Step 2 is: In the meeting room, the presence sensor is used to identify the "pseudo-unoccupied state", such as sitting participants, to avoid mis-turning off the lights.

5. An intelligent LED lighting energy-saving optimization system based on AI edge computing, based on the method for optimizing energy saving of intelligent LED lighting based on AI edge computing according to any one of claims 1-4, characterized in that: It includes technical improvement, intelligent control, system management, and AI calculation; The said technical improvement is used to improve the luminous efficiency by using high-purity GaN and InGaN materials for the existing light source module, and combining with a flip-chip structure to reduce light loss; The said intelligent control: includes a collection module and light, human infrared, microwave radar, and human presence sensors. For each sensor, different environmental systems such as intelligent street lights and garage lighting are completed through a combination of illuminance + infrared + temperature sensors, achieving a comprehensive energy saving of 40%-60%; The said system management specifically further includes an intelligent monitoring module, an environmental perception module, and a scene management module, which are used to use a photoresistor / photodiode to collect the illumination intensity in the range of 0-2000 lux in real time, and use a sliding window filter to eliminate instantaneous interference to achieve ambient light monitoring; The AI calculation includes an LED data analysis module, an LED data evaluation module, a behavior prediction module, and AI suggestions, which are used to analyze the data collected by the LED, separate the human activity data from the noise, and use the data evaluation to measure the input power factor through a power quality analyzer, evaluate the power conversion efficiency, and complete the prediction of behavior through the behavior prediction module, and give an automatic recommendation of the best driving power through AI, such as reducing the output by 30% during the non-peak period at night, and generating a dimming strategy in combination with the time-of-use electricity price model.

Images