Lamp energy efficiency optimization method based on artificial intelligence

Abstract

The invention relates to the technical field of energy efficiency optimization, and discloses a lamp energy efficiency optimization method based on artificial intelligence, and the method comprises the steps: data collection: carrying out the all-around and multi-dimensional data collection of an environment where a lamp is located through a plurality of sensors. According to the lamp energy efficiency optimization method based on artificial intelligence, in order to make energy utilization efficient, the multi-dimensional environment and lamp operation data are obtained through the data acquisition step, optimization is carried out in combination with the adaptive learning algorithm, lamp energy efficiency control parameters can be adjusted in real time according to different environments, places, time periods and other situations, and the energy efficiency optimization efficiency is improved. The lamp can be kept in the optimal energy efficiency state in various complex and changeable environments, the energy utilization efficiency of the lamp is effectively improved, energy waste is reduced, in the scene-based energy efficiency optimization link, for different application scenes, differentiation strategies are implemented according to respective typical lighting requirements and energy-saving potentials, parameters such as lamp brightness are accurately adjusted, and the energy efficiency of the lamp is improved. And the energy utilization efficiency of the lamp in various scenes is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy efficiency optimization, and in particular to a method for optimizing lamp energy efficiency based on artificial intelligence. Background Art

[0002] With the development of the global economy and the acceleration of urbanization, energy demand continues to grow. Lighting accounts for a considerable proportion of building energy consumption. According to statistics, lighting energy consumption in commercial buildings can account for about 20%-50% of the total energy consumption. Therefore, improving the energy efficiency of lamps is of great significance for energy conservation and emission reduction and reducing energy costs. This is an important driving force for the development of such optimization methods.

[0003] In recent years, artificial intelligence technology has made rapid progress. Machine learning and deep learning algorithms have demonstrated powerful data analysis and prediction capabilities in many fields, and have mined valuable information to optimize the energy efficiency of lamps.

[0004] Traditional lighting control methods are relatively simple, usually based on timer switches or manual brightness adjustment. Timer switches cannot be flexibly adjusted according to actual lighting needs and personnel activities, which may cause the lamps to be turned on when lighting is not needed, resulting in energy waste. Manual brightness adjustment relies on people's subjective judgment, and it is difficult to achieve accurate and real-time optimization. In large venues or scenarios with multiple lamps, the workload of manual operation is huge and unrealistic. It is not possible to optimize the energy efficiency of lamps based on artificial intelligence, and energy utilization cannot be made efficient. The personalization and convenience of user experience need to be improved. There is a lack of convenient and flexible ways for users to manage the energy efficiency of lamps, so it is not possible to intelligently maintain lamps and optimize system operations. In view of this, we propose a lamp energy efficiency optimization method based on artificial intelligence. Summary of the invention

[0005] The purpose of the present invention is to provide a method for optimizing lamp energy efficiency based on artificial intelligence to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for optimizing lamp energy efficiency based on artificial intelligence, comprising the following steps:

[0008] S1. Data collection: using a variety of sensors to collect all-round and multi-dimensional data about the environment in which the lamp is located. The sensors include light intensity sensors, temperature sensors, humidity sensors, personnel activity detection sensors (using infrared thermal imaging and microwave radar) and air quality sensors to obtain environmental data including light intensity, temperature, humidity, personnel activity and air quality, and at the same time collect the operating data of the lamp itself, including current, voltage and brightness parameters;

[0009] S2. Adaptive learning algorithm optimization: the collected environmental data and lamp operation data are input into the advanced adaptive learning algorithm. The adaptive learning algorithm adjusts the lamp energy efficiency control parameters in real time based on the continuously collected new environmental data and lamp operation data, through in-depth analysis of past data and rapid learning of current new situations, so that the lamps can maintain the best energy efficiency in various complex and changeable environments, that is, meet the energy efficiency optimization goal to achieve the maximum value. Among them, the most suitable brightness adjustment curve for a specific place is automatically summarized according to the operation conditions in different seasons and different weather conditions, and the brightness adjustment expression can be obtained by regression analysis of historical data;

[0010] S3. Satisfy the personalized needs of users. Through easy-to-use mobile applications or smart terminal devices, the personalized lighting atmosphere and brightness mode parameters set by users are received. The artificial intelligence system intelligently adjusts the energy efficiency of lamps based on user settings and actual environmental conditions, achieving energy saving while meeting the lighting effects expected by users.

[0011] S4, scenario-based energy efficiency optimization, automatically identifying the application scenarios of the lamps, including office spaces and commercial retail, and implementing differentiated energy efficiency optimization strategies based on the typical lighting needs and energy-saving potential of the identified scenarios;

[0012] S5. Predictive maintenance of lamp failures: using artificial intelligence data analysis and prediction capabilities to continuously track the current, voltage, and temperature parameters of lamps, and compare and analyze them with historical normal operation data. When abnormal fluctuations in key lamp parameters are detected and exceed the normal range, combined with relevant environmental factors, the potential fault hazards of lamps are accurately predicted, and maintenance personnel are promptly reminded to perform maintenance to avoid energy waste caused by lamp failures and the impact on normal lighting needs;

[0013] S6. Energy efficiency and maintenance are coordinated and optimized. When it is predicted that a lamp is about to enter a state requiring maintenance, in addition to issuing a maintenance reminder, the working parameters of the lamp are intelligently adjusted according to the remaining useful life of the lamp and the current energy efficiency, including but not limited to appropriately reducing the brightness to extend its service life while maintaining a certain energy efficiency level, or reasonably arranging the rotation of lamps without affecting the overall lighting effect, so that the entire lamp system can maintain high-efficiency lighting services and reduce energy consumption and equipment replacement costs through reasonable maintenance arrangements;

[0014] S7, distributed intelligent control, adopts a distributed intelligent architecture, giving each lamp a certain degree of independent data processing and decision-making capabilities. Each lamp can independently make preliminary energy efficiency optimization adjustments based on its own local environment data, such as the light intensity and human activities around it, as well as some information received from adjacent lamps, to achieve collaborative optimization of lamps in the local area, reduce dependence on the centralized control center, and improve the response speed and flexibility of the system;

[0015] S8. Cloud platform big data analysis and remote control. The massive local operating data of the lighting system is uploaded to the cloud platform for big data analysis. The cloud platform uses computing resources and artificial intelligence algorithms to summarize, analyze and mine data from different locations and different types of lamps, extract more universal and guiding energy efficiency optimization strategies, and remotely send these strategies to each lighting system for updating and optimization. At the same time, the cloud platform allows users to remotely view the operating status and energy efficiency index information of the lamps, and perform remote control, such as remotely adjusting the brightness mode of the lamps at home through mobile phones.

[0016] Preferably, in the data collection step, the personnel activity detection sensor is configured by combining a plurality of sensors with different detection principles to improve the accuracy and comprehensiveness of personnel activity detection. Specifically, an infrared thermal imaging sensor and a microwave radar sensor may be combined. The personnel activity detected by the infrared thermal imaging sensor is: ir (quantified by the number of people or activity frequency, the unit can be determined according to the specific quantification method), the personnel activities detected by the microwave radar sensor are: P radar (quantified by the number of personnel or activity frequency, the unit can be determined according to the specific quantification method), through the weighted summation formula: P = αP ir +(1-α)P radar (where α is a weighting coefficient, and its value range is 0<α<1) to comprehensively determine the personnel activity situation P, so as to improve the accuracy and comprehensiveness of personnel activity detection.

[0017] Preferably, in the adaptive learning algorithm optimization step, the adaptive learning algorithm adopts a recurrent neural network (RNN) or its variants in a deep learning algorithm, such as a long short-term memory network (LSTM) and a gated recurrent unit (GRU), to better process time series data and achieve accurate optimization of lamp energy efficiency.

[0018] Preferably, in the step of satisfying the personalized needs of the user, the mobile application or smart terminal device provides a visual lighting effect preview function so that the user can set the personalized parameters more intuitively. The specific implementation method can be to establish a lighting effect preview model to generate a visual lighting effect preview image or video for the user to refer to when setting the personalized parameters, so that the user can set the personalized parameters more intuitively.

[0019] Preferably, in the scenario-based energy efficiency optimization step, for the home environment scenario, the brightness of lamps in different areas of the bedroom and living room are dynamically adjusted according to the daily activity habits of family members, such as getting up, sleeping, and watching TV. Specifically, a family member activity schedule can be established, and the brightness of lamps in different areas of the bedroom and living room can be dynamically adjusted in combination with environmental data to improve the energy utilization efficiency of lamps in home environment scenarios.

[0020] Preferably, in the lamp fault predictive maintenance step, the prediction of the potential fault of the lamp includes making a comprehensive judgment based on the change of the luminous efficiency of the lamp in combination with the use time, assuming that the luminous efficiency of the lamp is: η (unit: lumen per watt, lm / W), and the initial luminous efficiency is η 0 , the usage time is (unit: hours, h), through the expression: Y fault (η,η 0 ,t use ) to comprehensively judge whether the lamp has hidden dangers of failure. fault (η,η 0 ,t use )〉threshold (determined according to actual experience or experiments), it is determined that the lamp has potential fault and a fault reminder is issued.

[0021] Preferably, in the energy efficiency and maintenance collaborative optimization step, when adjusting the lamp operating parameters, the human activity density in the area where the lamp is located is also taken into consideration. If the human activity density is low, the brightness can be further reduced to achieve better energy-saving effects.

[0022] Preferably, in the distributed intelligent control step, information transmission between the lamps adopts wireless communication, and the communication protocol adopts Bluetooth Low Energy (BLE) or ZigBee standard protocol to reduce energy consumption and ensure communication stability.

[0023] Preferably, in the cloud platform big data analysis and remote control step, when the cloud platform analyzes the data, cluster analysis and association rule mining data mining methods are used to more effectively extract energy efficiency optimization strategies.

[0024] Preferably, in the cloud platform big data analysis and remote control step, the remote control function also includes setting the timing switch and group control function of the lamps to facilitate users to manage the energy efficiency of the lamps more flexibly.

[0025] Compared with the prior art, the present invention provides a method for optimizing lamp energy efficiency based on artificial intelligence, which has the following beneficial effects:

[0026] 1. In order to make energy utilization more efficient, this artificial intelligence-based lamp energy efficiency optimization method obtains multi-dimensional environment and lamp operation data through data collection steps, and combines adaptive learning algorithm optimization to adjust lamp energy efficiency control parameters in real time according to different environments, locations and time periods, so that lamps can maintain the best energy efficiency in various complex and changeable environments, effectively improve the energy utilization efficiency of lamps, and reduce energy waste. In the scenario-based energy efficiency optimization link, for different application scenarios, differentiated strategies are implemented according to their respective typical lighting needs and energy-saving potentials, and parameters such as lamp brightness are accurately adjusted to improve the energy utilization efficiency of lamps in various scenarios.

[0027] 2. This AI-based lamp energy efficiency optimization method, in order to enhance the personalization and convenience of user experience, allows users to set personalized lighting atmosphere, brightness mode and other parameters through simple and easy-to-use mobile applications or smart terminal devices, and the device can provide a visual lighting effect preview function, which is convenient for users to set intuitively. The AI ​​system will intelligently adjust the lamp energy efficiency according to user settings and actual environmental conditions, while meeting the user's expected lighting effects and achieving energy saving, greatly enhancing the user's personalized experience of the lighting environment. With the help of cloud platform big data analysis and remote control functions, users can not only remotely view the lamp operation status and energy efficiency index information, but also perform remote control, providing users with a convenient and flexible way to manage lamp energy efficiency.

[0028] 3. This method of optimizing lamp energy efficiency based on artificial intelligence, in order to make lamp maintenance intelligent and optimize system operation, uses the data analysis and prediction capabilities of artificial intelligence. In the predictive maintenance step of lamp failures, it continuously tracks the key parameters of lamps and compares and analyzes them with historical data. It combines multiple factors to accurately predict potential faults, and promptly reminds maintenance personnel to inspect and repair them, so as to avoid energy waste caused by lamp failures and the impact on normal lighting needs. It realizes the intelligent maintenance of lamps. When it is predicted that a lamp needs maintenance, in addition to reminders, it can also intelligently adjust the lamp working parameters according to the remaining usable life and current energy efficiency. At the same time, distributed intelligent control gives lamps independent processing and decision-making capabilities, realizes collaborative optimization through wireless communication, reduces dependence on centralized control centers, improves system response speed and flexibility, and optimizes the operation of the lamp system as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1This is a flow chart of the method for optimizing the energy efficiency of lamps based on artificial intelligence of the present invention;

[0030] Figure 2 This is a data collection flow chart of the present invention;

[0031] Figure 3 This is a schematic diagram of a scene where the lamp of the present invention is located;

[0032] Figure 4 This is a flowchart of the steps of predictive maintenance of lamp failures of the present invention;

[0033] Figure 5 A flow chart of the steps of energy efficiency and maintenance collaborative optimization of the present invention;

[0034] Figure 6 This is a flow chart of the steps of distributed intelligent control of the present invention;

[0035] Figure 7 This is a flow chart of the cloud platform big data analysis and remote control steps of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] See also Figure 1 - Figure 7 , the present invention provides a technical solution:

[0038] A method for optimizing lamp energy efficiency based on artificial intelligence, comprising the following steps:

[0039] S1. Data collection: Use a variety of sensors to collect all-round and multi-dimensional data about the environment in which the lamp is located. The sensors include light intensity sensors, temperature sensors, humidity sensors, personnel activity detection sensors (using infrared thermal imaging and microwave radar) and air quality sensors to obtain environmental data including light intensity, temperature, humidity, personnel activity and air quality, and collect the operating data of the lamp itself, including current, voltage and brightness parameters;

[0040] S2, adaptive learning algorithm optimization, input the collected environmental data and lamp operation data into the advanced adaptive learning algorithm. The adaptive learning algorithm adjusts the lamp energy efficiency control parameters in real time based on the continuously collected new environmental data and lamp operation data, through in-depth analysis of past data and rapid learning of current new situations. Suppose the lamp energy efficiency control parameter vector is:

[0041] Y=〔y1,y2,…,yn〕

[0042] (where n is the number of control parameters, and the specific parameters are determined according to the actual algorithm, such as brightness adjustment coefficient and switching time interval). The adaptive learning algorithm is based on the expression:

[0043] F=(X,t)

[0044] (where t is a time variable, indicating different moments of data collection) to update Y, so that the lamps can maintain the best energy efficiency in various complex and changeable environments, that is, to meet the energy efficiency optimization goal and reach the maximum value. Among them, the most suitable brightness adjustment curve for a specific place is automatically summarized according to the operation conditions in different seasons and different weather conditions. The brightness adjustment expression can be obtained by regression analysis of historical data: L adj (t,S)

[0045] (where S represents a set of factors related to a specific site, such as site area and layout);

[0046] S3. Satisfy the personalized needs of users. Through easy-to-use mobile applications or smart terminal devices, the personalized lighting atmosphere and brightness mode parameters set by users are received. The artificial intelligence system intelligently adjusts the energy efficiency of lamps based on user settings and actual environmental conditions, achieving energy saving while meeting the lighting effects expected by users.

[0047] S4, scenario-based energy efficiency optimization, automatically identifies the application scenarios of the lamps, including office spaces and commercial retail, and the identified scenario categories are: C,

[0048] Based on the typical lighting requirements and energy saving potential of the identified scenes, differentiated energy efficiency optimization strategies are implemented as follows:

[0049] For the office scene (C = "Office"), the focus is on sufficient lighting for office desks and safe lighting for public passages. The minimum light intensity required for office desks is: I desk (Unit: lx), the minimum light intensity required for public channels is: I channel (Unit: lx), optimize the use of lamps according to the activity patterns of personnel at different time periods, such as during work, lunch break, and after get off work. Assume that the probability distribution expression of personnel's activities in different time periods is: P office (t) (where t is a time variable), through the expression: c, to adjust the lamp energy efficiency control parameters to improve the energy efficiency of the lamp in the office scene;

[0050] For the commercial retail scenario (C = "commercial retail"), the brightness of lamps in different areas of the display area and cashier area is dynamically adjusted based on the store's business hours and customer flow distribution factors. Suppose the store's business hours are T open To T close , the customer flow distribution expression is: F customer (x, y, t) (x, y are the spatial coordinates in the store, t is the time variable), through the expression: N(C, F customer (x,y,t),E), to adjust the lamp energy efficiency control parameters to improve the energy efficiency of lamps in commercial retail scenarios;

[0051] S5. Predictive maintenance of lamp failures: using artificial intelligence data analysis and prediction capabilities to continuously track the current, voltage, and temperature parameters of lamps, and compare and analyze them with historical normal operation data. When abnormal fluctuations in key lamp parameters are detected and exceed the normal range, combined with relevant environmental factors, the potential fault hazards of lamps are accurately predicted, and maintenance personnel are promptly reminded to perform maintenance to avoid energy waste caused by lamp failures and the impact on normal lighting needs;

[0052] S6. Energy efficiency and maintenance are coordinated and optimized. When it is predicted that a lamp is about to enter a state requiring maintenance, in addition to issuing a maintenance reminder, the working parameters of the lamp are intelligently adjusted according to the remaining useful life of the lamp and the current energy efficiency, including but not limited to appropriately reducing the brightness to extend its service life while maintaining a certain energy efficiency level, or reasonably arranging the rotation of lamps without affecting the overall lighting effect, so that the entire lamp system can maintain high-efficiency lighting services and reduce energy consumption and equipment replacement costs through reasonable maintenance arrangements;

[0053] S7, distributed intelligent control, adopts a distributed intelligent architecture, giving each lamp a certain degree of independent data processing and decision-making capabilities. Each lamp can independently make preliminary energy efficiency optimization adjustments based on its own local environment data, such as the light intensity and human activities around it, as well as some information received from adjacent lamps, to achieve collaborative optimization of lamps in the local area, reduce dependence on the centralized control center, and improve the response speed and flexibility of the system;

[0054] S8. Cloud platform big data analysis and remote control. The massive local operating data of the lighting system is uploaded to the cloud platform for big data analysis. The cloud platform uses computing resources and artificial intelligence algorithms to summarize, analyze and mine data from different locations and different types of lamps, extract more universal and guiding energy efficiency optimization strategies, and remotely send these strategies to each lighting system for updating and optimization. At the same time, the cloud platform allows users to remotely view the operating status and energy efficiency index information of the lamps, and perform remote control, such as remotely adjusting the brightness mode of the lamps at home through mobile phones.

[0055] In one embodiment of the present invention, in the data collection step, the personnel activity detection sensor is configured by combining a plurality of sensors with different detection principles to improve the accuracy and comprehensiveness of personnel activity detection. Specifically, an infrared thermal imaging sensor and a microwave radar sensor may be combined. The personnel activity detected by the infrared thermal imaging sensor is: ir (quantified by the number of people or activity frequency, the unit can be determined according to the specific quantification method), the personnel activities detected by the microwave radar sensor are: P radar (quantified by the number of personnel or activity frequency, the unit can be determined according to the specific quantification method), through the weighted summation formula: P = αP ir +(1-α)P radar (where α is a weighting coefficient with a value range of 0<α<1) to comprehensively determine the personnel activity situation P, so as to improve the accuracy and comprehensiveness of personnel activity detection. In addition, in the optimization step of the adaptive learning algorithm, the adaptive learning algorithm adopts a recurrent neural network (RNN) or its variants in the deep learning algorithm, such as a long short-term memory network (LSTM) and a gated recurrent unit (GRU), to better process time series data and achieve accurate optimization of the energy efficiency of lamps. In addition, in the step of satisfying user personalized needs, the mobile application or smart terminal device provides a visual lighting effect preview function so that users can set personalized parameters more intuitively. The specific implementation method can be achieved by establishing a lighting effect preview model to generate a visual lighting effect preview image or video for users to refer to when setting personalized parameters, so that users can set personalized parameters more intuitively.

[0056] In one embodiment of the present invention, in the scenario-based energy efficiency optimization step, for the home environment scene, the brightness of lamps in different areas of the bedroom and living room is dynamically adjusted according to the daily activity habits of family members, such as getting up, sleeping, and watching TV. Specifically, a family member activity schedule can be established, as shown in the following table:

[0057] Family members Wake-up time Bedtime TV Watching Time Member 1 <![CDATA[t 1r ]]> <![CDATA[t 1s ]]> <![CDATA[t 1t ]]> Member 2 <![CDATA[t 2r ]]> <![CDATA[t 2s ]]> <![CDATA[t 2t ]]> ... ... ... ...

[0058] Suppose the brightness adjustment expression of lamps in different areas is: Lhome (t, r) (where t is the time variable and r is the area identifier, such as bedroom, living room), according to the activity time and area of ​​family members, through the expression: X home (t, r, E) is combined with environmental data to dynamically adjust the brightness of lamps in different areas of the bedroom and living room to improve the energy efficiency of lamps in home environments. In addition, in the predictive maintenance step of lamp failures, the prediction of potential faults of lamps includes comprehensive judgment based on the changes in the luminous efficiency of the lamps and the length of use. Assume that the luminous efficiency of the lamp is: η (unit: lumen per watt, lm / W), and the initial luminous efficiency is η 0 , the usage time is (unit: hours, h), through the expression: Y fault (η,η 0 ,t use ) to comprehensively judge whether the lamp has hidden dangers of failure. fault (η,η 0 ,t use )〉threshold (determined based on actual experience or experiments), it is determined that the lamp has a potential fault and a fault reminder is issued. In addition, in the energy efficiency and maintenance collaborative optimization step, when adjusting the lamp working parameters, the density of human activities in the area where the lamp is located is also taken into account. If the density of human activities is low, the brightness can be further reduced to achieve better energy-saving effects.

[0059] In one embodiment of the present invention, in the distributed intelligent control step, information transmission between lamps adopts wireless communication, and the communication protocol adopts low-power Bluetooth (BLE) or ZigBee standard protocol to reduce energy consumption and ensure the stability of communication. In addition, in the cloud platform big data analysis and remote control step, the cloud platform uses clustering analysis and association rule mining data mining methods when analyzing data to more effectively extract energy efficiency optimization strategies. In addition, in the cloud platform big data analysis and remote control step, the remote control function also includes setting the timing switch and group control function of the lamp to facilitate users to manage the energy efficiency of the lamp more flexibly.

[0060] The above generally describes the present invention in detail, but it is obvious to a person skilled in the art that some modifications or improvements can be made to the present invention. Therefore, modifications or improvements that do not depart from the spirit of the present invention are within the scope of protection of the present invention.

Claims

1. A method for optimizing lamp energy efficiency based on artificial intelligence, characterized in that: The following steps are involved: S1. Data collection: using a variety of sensors to collect all-round and multi-dimensional data about the environment in which the lamp is located. The sensors include light intensity sensors, temperature sensors, humidity sensors, personnel activity detection sensors and air quality sensors to obtain environmental data including light intensity, temperature, humidity, personnel activity and air quality, and at the same time collect the operating data of the lamp itself, including current, voltage and brightness parameters; S2. Adaptive learning algorithm optimization: the collected environmental data and lamp operation data are input into the advanced adaptive learning algorithm. The adaptive learning algorithm adjusts the lamp energy efficiency control parameters in real time based on the continuously collected new environmental data and lamp operation data, through in-depth analysis of past data and rapid learning of current new situations, so that the lamps can maintain the best energy efficiency in various complex and changing environments, that is, meet the energy efficiency optimization goal to achieve the maximum value. Among them, the most suitable brightness adjustment curve for a specific place is automatically summarized according to the operation conditions in different seasons and different weather conditions; S3. Satisfy the personalized needs of users. Through easy-to-use mobile applications or smart terminal devices, the personalized lighting atmosphere and brightness mode parameters set by users are received. The artificial intelligence system intelligently adjusts the energy efficiency of lamps based on user settings and actual environmental conditions, achieving energy saving while meeting the lighting effects expected by users. S4, scenario-based energy efficiency optimization, automatically identifying the application scenarios of the lamps, including office and retail, and implementing differentiated energy efficiency optimization strategies based on the typical lighting needs and energy-saving potential of the identified scenarios to improve the energy efficiency of the lamps in office and retail scenarios; S5. Predictive maintenance of lamp failures: using artificial intelligence data analysis and prediction capabilities to continuously track the current, voltage, and temperature parameters of lamps, and compare and analyze them with historical normal operation data. When abnormal fluctuations in key lamp parameters are detected and exceed the normal range, combined with relevant environmental factors, the potential fault hazards of lamps are accurately predicted, and maintenance personnel are promptly reminded to perform maintenance to avoid energy waste caused by lamp failures and the impact on normal lighting needs; S6. Collaborative optimization of energy efficiency and maintenance. When it is predicted that a lamp is about to enter a state requiring maintenance, in addition to issuing a maintenance reminder, the working parameters of the lamp are intelligently adjusted according to the remaining useful life of the lamp and the current energy efficiency, including but not limited to appropriately reducing the brightness to extend its service life while maintaining a certain energy efficiency level, or reasonably arranging the rotation of lamps without affecting the overall lighting effect, so that the entire lamp system can maintain high-efficiency lighting services; S7, distributed intelligent control, adopts a distributed intelligent architecture, giving each lamp a certain degree of independent data processing and decision-making capabilities. Each lamp independently performs preliminary energy efficiency optimization adjustments based on its own local environmental data, such as the light intensity and human activities around it, as well as some information received from adjacent lamps, to achieve collaborative optimization of lamps in the local area; S8. Cloud platform big data analysis and remote control. The massive local operating data of the lighting system is uploaded to the cloud platform for big data analysis. The cloud platform uses computing resources and artificial intelligence algorithms to summarize, analyze and mine data from different locations and different types of lamps, extract more universal and guiding energy efficiency optimization strategies, and remotely send these strategies to each lighting system for updating and optimization.

2. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the data collection step, the personnel activity detection sensor is configured by combining a plurality of sensors with different detection principles to improve the accuracy and comprehensiveness of personnel activity detection. Specifically, an infrared thermal imaging sensor and a microwave radar sensor may be combined. The personnel activity detected by the infrared thermal imaging sensor is: ir (quantified by the number of people or activity frequency, the unit can be determined according to the specific quantification method), the personnel activities detected by the microwave radar sensor are: P radar (quantified by the number of personnel or activity frequency, the unit can be determined according to the specific quantification method), through the weighted summation formula: P = αP ir +(1-α)P radar (where α is a weighting coefficient, and its value range is 0<α<1) to comprehensively determine the personnel activity situation P, so as to improve the accuracy and comprehensiveness of personnel activity detection.

3. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the adaptive learning algorithm optimization step, the adaptive learning algorithm adopts a recurrent neural network (RNN) or its variants in a deep learning algorithm, such as a long short-term memory network (LSTM) and a gated recurrent unit (GRU).

4. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the step of satisfying the personalized needs of the user, the mobile application or smart terminal device provides a visual lighting effect preview function so that the user can set the personalized parameters more intuitively. The specific implementation method can be to establish a lighting effect preview model to generate a visual lighting effect preview image or video for the user to refer to when setting the personalized parameters.

5. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the scenario-based energy efficiency optimization step, for the home environment scenario, the brightness of lamps in different areas of the bedroom and living room are dynamically adjusted according to the daily activity habits of family members, such as getting up, going to bed, and watching TV.

6. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the lamp fault predictive maintenance step, the prediction of the potential faults of the lamp includes making a comprehensive judgment based on the change of the luminous efficiency of the lamp and the use time. Assuming that the luminous efficiency of the lamp is: η (unit: lumen per watt, lm / W), the initial luminous efficiency is η0, and the use time is (unit: hour, h), through the expression: Y fault (η,η0,t use ) to comprehensively judge whether the lamp has hidden dangers of failure. fault (η,η0,t use )〉threshold (determined according to actual experience or experiments), it is determined that the lamp has potential fault and a fault reminder is issued.

7. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the energy efficiency and maintenance collaborative optimization step, when adjusting the operating parameters of the lamp, the density of human activities in the area where the lamp is located is also taken into consideration. If the density of human activities is low, the brightness can be further reduced to achieve better energy-saving effects.

8. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the distributed intelligent control step, information transmission between the lamps is carried out by wireless communication, and the communication protocol adopts Bluetooth Low Energy (BLE) or ZigBee standard protocol.

9. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the cloud platform big data analysis and remote control step, the cloud platform uses cluster analysis and association rule mining data mining methods when analyzing data.

10. The method for optimizing lamp energy efficiency based on artificial intelligence according to claim 1, characterized in that: In the cloud platform big data analysis and remote control steps, the remote control function also includes setting the timing switch and group control function of the lamps.

Images