Smart city lighting energy efficiency management system based on Internet of Things

Abstract

The invention relates to the technical field of lighting energy efficiency management, in particular to a smart city lighting energy efficiency management system based on the Internet of Things, and adopts the technical scheme that the smart city lighting energy efficiency management system comprises a data acquisition and analysis module, an intelligent control module, a fault detection and early warning module, an energy management module and a report and statistical analysis module, environment data and traffic flow data are acquired in real time through an illuminance sensor, a temperature sensor, a humidity sensor, a pedestrian flow sensor and a traffic flow sensor which are deployed in each region of a city. According to the invention, through multi-sensor data acquisition and a complex algorithm, the illumination demand is accurately predicted, and personalized brightness adjustment and energy saving are realized; stable operation of the system is ensured through efficient fault detection and early warning; energy management is optimized, and efficient utilization of energy is realized in combination with a genetic algorithm and LSTM prediction; and the report and statistical analysis module provides visual display and decision support, so that efficient utilization of energy and optimization of the lighting effect are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of lighting energy efficiency management, and in particular to a smart city lighting energy efficiency management system based on the Internet of Things. Background Art

[0002] In today's rapidly developing urban environment, urban lighting systems play a vital role. Traditional urban lighting control methods mainly rely on fixed schedules for switching operations. However, this method has many obvious disadvantages.

[0003] On the one hand, fixed schedules cannot adapt to the actual needs of dynamic changes. The weather conditions in cities are complex and changeable, and there are significant differences in different seasons. Relying solely on a fixed schedule, it is impossible to make precise adjustments based on key information in the weather forecast, such as sunrise and sunset times, cloud thickness, etc. For example, on cloudy days or when the clouds are thick, the actual light intensity may be much lower than on sunny days. If the lighting is still controlled according to a fixed schedule, it may result in insufficient lighting, posing a safety hazard to pedestrians and vehicles. In the process of seasonal changes, changes in the length of daylight will also make the irrationality of the fixed schedule more prominent, which may cause energy waste or poor lighting effects.

[0004] On the other hand, traditional methods cannot respond promptly to changes in pedestrian and vehicle traffic in different areas. Different areas in the city have different functions, and commercial areas, residential areas, industrial areas, etc. have very different lighting needs. Fluctuations in pedestrian and vehicle traffic directly affect lighting needs. In areas with large pedestrian and vehicle traffic, higher lighting brightness is required to ensure safety and convenience; while in areas with small pedestrian and vehicle traffic, excessive lighting brightness not only wastes energy, but may also cause light pollution. However, traditional fixed schedule control cannot be flexibly adjusted according to these actual conditions, making it difficult to achieve efficient energy utilization and optimization of lighting effects.

[0005] In summary, the traditional urban lighting control method using fixed schedules can no longer meet the needs of modern urban development. This application proposes a smart city lighting energy efficiency management system based on the Internet of Things. Summary of the invention

[0006] The purpose of the present invention is to propose a smart city lighting energy efficiency management system based on the Internet of Things to address the problem that traditional urban lighting uses a fixed schedule for on-off control and cannot be flexibly adjusted according to actual needs in the background technology.

[0007] The technical solution of the present invention is a smart city lighting energy efficiency management system based on the Internet of Things, comprising:

[0008] The data collection and analysis module collects environmental data and traffic flow data in real time through the light intensity sensors, temperature sensors, humidity sensors, pedestrian flow sensors, and vehicle flow sensors deployed in various areas of the city. The environmental data includes light intensity, temperature, humidity, pedestrian flow, and vehicle flow. The collected data is analyzed and processed using the weighted average algorithm and trend analysis algorithm to extract useful information, including sunrise and sunset times, cloud thickness, and the changing trends of pedestrian and vehicle flow. Based on the analysis results, the regression analysis algorithm is used to predict the lighting needs in different areas and at different times.

[0009] The intelligent control module automatically adjusts the lighting on and off time according to the prediction results of the data collection and analysis module. It uses a linear adjustment algorithm to gradually adjust the lighting brightness according to the changes in light intensity after sunrise and before sunset, and uses a dynamic adjustment algorithm to dynamically adjust the lighting brightness at night according to the changes in pedestrian and vehicle traffic. It increases the lighting brightness in areas with large pedestrian and vehicle traffic, reduces the lighting brightness in areas with small pedestrian and vehicle traffic, and temporarily turns off the lighting in areas without people or cars. It also supports remote control functions, and managers can manually control the smart lamps through the management platform or mobile applications.

[0010] The fault detection and early warning module monitors the operating status of smart lamps in real time, including current, voltage, and power parameters. When a lamp fault is detected, it automatically issues an early warning message using a fault determination algorithm, and sends the fault information to the management platform and mobile application to notify maintenance personnel to handle the problem and locate the faulty lamp.

[0011] Energy management module, which statistically analyzes the energy consumption of urban lighting systems, including power consumption and power consumption time. Based on energy consumption data, energy-saving strategies are formulated using energy-saving strategy optimization algorithms. Energy-saving strategies include optimizing lighting on and off times and adjusting lighting brightness.

[0012] The report and statistical analysis module generates reports, including daily reports, monthly reports, and annual reports, which display the operation status and energy consumption of the urban lighting system, conduct statistical analysis of historical data, and provide decision support for managers.

[0013] Optionally, in the data collection and analysis module, the weighted average algorithm formula is: comprehensive index value = Σ(index value × weight);

[0014] Among them, the calculation formula for light intensity weight is W1=a×(1-e^(-b×light intensity change rate)), the calculation formula for pedestrian flow weight is W2=c×(1-e^(-d×pedestrian flow standard deviation)), and the calculation formula for vehicle flow weight is W3=e×(1-e^(-f×vehicle flow peak-to-valley ratio));

[0015] a, b, c, d, e, and f are adjustment coefficients. These coefficients are continuously adjusted to adapt to different urban environments and needs. Different weights are given to the indicator values ​​of each sensor data according to the actual situation to obtain a comprehensive environmental data evaluation result.

[0016] Optionally, in the data collection and analysis module, the trend analysis algorithm is: by performing differential calculation on the pedestrian and vehicle flow data within a period of time, if the differential result shows an upward trend, it is judged that the pedestrian and vehicle flow have an increasing trend, otherwise it is a decreasing trend, and the autoregressive integral moving average model is further introduced to predict the long-term trend of the pedestrian and vehicle flow, and the formula is:

[0017] Yt=μ+φ1Yt-1+φ2Yt-2+…+φpYt-p+εt+θ1εt-1+θ2εt-2+…+θqεt-q;

[0018] Among them, Yt is the pedestrian or vehicle flow at the current moment, μ is a constant term, φ1, φ2, etc. are autoregressive coefficients, θ1, θ2, etc. are sliding average coefficients, εt is an error term, p and q are model parameters. The model parameters are determined by fitting historical data to more accurately predict the changing trends of pedestrian and vehicle flows.

[0019] Optionally, the regression analysis algorithm adopts a multivariate linear regression model:

[0020] Lighting demand = α + β1 × illuminance + β2 × pedestrian flow + β3 × vehicle flow + β4 × seasonal factors + β5 × weather factors + … + ε;

[0021] Among them, α is a constant term, β1, β2, β3, etc. are regression coefficients, and ε is an error term. The seasonal factor is simulated by a sine function, and the formula is Where A is the amplitude, ω is the angular frequency, and t is the time (in seasons), is the phase;

[0022] Weather factors are quantified through fuzzy logic, and a comprehensive evaluation is performed based on multiple factors such as cloud thickness and precipitation probability to obtain a value between 0 and 1, which indicates the degree of impact of weather on lighting needs.

[0023] Optionally, in the intelligent control module, the linear adjustment algorithm is:

[0024] Lighting brightness = initial brightness × (1-current illuminance ÷ maximum illuminance) × g^(-h × illuminance change speed);

[0025] Among them, g and h are adjustment coefficients. The lighting brightness is gradually adjusted as the illuminance changes. At the same time, the impact of the illuminance change speed on the lighting brightness adjustment is considered. The faster the change speed, the smaller the adjustment range is, so as to avoid frequent and large changes in brightness.

[0026] Optionally, the dynamic adjustment algorithm in the intelligent control module is:

[0027] Lighting brightness = basic brightness + Δ brightness;

[0028] Among them, Δ brightness = k × pedestrian flow ^ m × vehicle flow ^ n, k, m, n are adjustment coefficients. The greater the pedestrian flow and vehicle flow, the greater the Δ brightness. The coefficients are adjusted to adapt to the traffic flow characteristics of different areas.

[0029] Optionally, in the fault detection and early warning module, the fault judgment algorithm is: when the current, voltage and power parameters of the lamp exceed the normal range [lower limit, upper limit], the lamp is judged to be faulty, and the principal component analysis algorithm is further introduced to perform dimensionality reduction processing on the operating status data of the lamp.

[0030] Optionally, in the energy management module, the energy-saving strategy optimization algorithm is: taking minimization of energy consumption as the objective function, the objective function formula is Minimize E = Σ(Pi×ti), wherein E is the total energy consumption, Pi is the power of the ith lamp, ti is the turn-on time of the ith lamp, and the lighting demand and renewable energy supply are combined as constraints, and the optimal lighting turn-on and turn-off time and lighting brightness adjustment plan are solved through a genetic algorithm; in the genetic algorithm, the chromosome is represented as a set of lighting control parameters, and the chromosome is continuously optimized through selection, crossover, and mutation operations to find the optimal solution that meets the objective function and constraints.

[0031] Optionally, the report and statistical analysis module includes a data mining unit, which performs in-depth analysis on historical data, uses an association rule algorithm to mine potential associations between different environmental factors and lighting requirements, and uses an Apriori algorithm, setting the minimum support as min_sup and the minimum confidence as min_conf. Frequent item sets are found by continuously scanning the data set, and then strong association rules are generated to provide a more accurate decision-making basis for lighting energy efficiency management. The formula is expressed as follows:

[0032] If the frequency of item set X in the data set is greater than or equal to min_sup, then X is a frequent item set; if the confidence Conf(X→Y)=P(Y|X) of rule X→Y is greater than or equal to min_conf, then X→Y is a strong association rule, where P(Y|X) represents the probability of Y appearing under the condition that X appears.

[0033] Optionally, the energy management module combines the long short-term memory network (LSTM) in deep learning to predict future energy demand. Suppose the input sequence is X=(x1, x2, ..., xt), where xt represents the feature vector at time t, including power consumption, light intensity, and pedestrian flow. The LSTM unit updates the cell state Ct and the hidden state ht through the forget gate, input gate, and output gate;

[0034] The forget gate ft decides to discard the filtered information from the cell state ft = σ(Wf · [ht-1, xt] + bf)

[0035] Among them, σ is the sigmoid function, Wf is the weight matrix, and bf is the bias term;

[0036] The input gate it determines which new information is stored in the cell state it = σ(Wi · [ht-1, xt] + bi),

[0037] Candidate cell state Ct~=tanh(Wc·[ht-1,xt]+bc), cell state Ct=ftCt-1+itCt~;

[0038] The output gate ot determines the output of the filtered information, ot = σ(Wo · [ht-1, xt] + bo), and the hidden state ht = ot*tanh(Ct). By training the LSTM network, future energy demand can be predicted to better formulate energy-saving strategies.

[0039] Compared with the prior art, the present invention has at least one of the following beneficial technical effects:

[0040] The present invention can accurately control lighting demand, collect a large amount of data through multiple sensors, and use complex algorithms for analysis and processing, and can accurately predict lighting demand in different areas and at different times. For example, according to weather conditions, seasonal changes, pedestrian and vehicle traffic, etc., the lighting on and off time and brightness are automatically adjusted to ensure lighting effects while maximizing energy conservation.

[0041] At the same time, personalized adjustments are made according to the characteristics of different areas. The lighting brightness is increased in areas with large pedestrian and vehicle traffic to ensure safety and convenience; the lighting brightness is reduced in areas with small pedestrian and vehicle traffic to reduce energy waste and light pollution; the lighting is temporarily turned off in areas where there are no people and no cars to further improve energy utilization efficiency.

[0042] Further use of efficient fault detection and early warning technology to monitor the operating status of smart lamps in real time can detect faults in time and automatically issue early warning information to notify maintenance personnel to handle them. At the same time, the faulty lamps can be located to facilitate maintenance personnel to quickly find the fault point, improve maintenance efficiency, and ensure the stable operation of the urban lighting system.

[0043] Further optimize energy management, statistically analyze the energy consumption of urban lighting systems, use energy-saving strategy optimization algorithms to formulate reasonable energy-saving strategies, and combine renewable energy supply to achieve optimal energy utilization. Genetic algorithms are used to solve the optimal lighting on and off time and lighting brightness adjustment plan, with the goal of minimizing energy consumption, reducing the energy cost of urban lighting systems.

[0044] In addition, the use of long short-term memory networks (LSTM) to predict future energy demand can enable energy-saving strategies to be formulated in advance and better respond to changes in energy demand.

[0045] The report and statistical analysis module generates various reports to display the operation status and energy consumption of the urban lighting system, providing managers with intuitive data display. At the same time, statistical analysis and data mining are performed on historical data to explore the potential relationship between different environmental factors and lighting needs, providing managers with more accurate decision support and helping to optimize the layout and management of the urban lighting system.

[0046] In summary, the present invention uses multi-sensor data collection and complex algorithms to accurately predict lighting needs, achieve personalized brightness adjustment and energy saving; efficient fault detection and early warning ensure stable system operation; optimize energy management, combine genetic algorithms and LSTM prediction to achieve efficient energy utilization; report and statistical analysis modules provide intuitive display and decision support. It comprehensively improves urban lighting energy efficiency, reduces costs, helps cities to achieve sustainable development, and achieves efficient energy utilization and optimization of lighting effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a block diagram of a smart city lighting energy efficiency management system based on the Internet of Things. DETAILED DESCRIPTION

[0048] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0049] Example

[0050] like Figure 1 As shown, the present invention proposes a smart city lighting energy efficiency management system based on the Internet of Things, which includes a data acquisition and analysis module, an intelligent control module, a fault detection and early warning module, an energy management module, and a report and statistical analysis module. Each module is described in detail below.

[0051] 1. Data collection and analysis module

[0052] Light intensity sensors, temperature sensors, humidity sensors, pedestrian flow sensors, and vehicle flow sensors are deployed in various areas of the city to collect real-time environmental data and traffic flow data including light intensity, temperature, humidity, pedestrian flow, vehicle flow, etc.

[0053] The collected data are analyzed and processed using the weighted average algorithm, and the weighted average algorithm formula is comprehensive index value = Σ(index value × weight). Among them, the calculation formula for light intensity weight is W1 = a × (1-e^(-b × light intensity change rate)), the calculation formula for pedestrian flow weight is W2 = c × (1-e^(-d × pedestrian flow standard deviation)), and the calculation formula for vehicle flow weight is W3 = e × (1-e^(-f × vehicle flow peak-to-valley ratio)). By continuously adjusting the adjustment coefficients a, b, c, d, e, and f to adapt to different urban environments and needs, and assigning different weights to the indicator values ​​of each sensor data according to actual conditions, a comprehensive environmental data evaluation result is obtained.

[0054] The trend analysis algorithm is used to determine the trend by performing differential calculation on the pedestrian and vehicle flow data within a period of time. If the differential result shows an upward trend, it is judged that the pedestrian and vehicle flow have an increasing trend, otherwise it is a decreasing trend. The autoregressive integrated moving average model is further introduced to predict the long-term trend of pedestrian and vehicle flow. The formula is Yt=μ+φ1Yt-1+φ2Yt-2+…+φpYt-p+εt+θ1εt-1+θ2εt-2+…+θqεt-q. Among them, Yt is the pedestrian or vehicle flow at the current moment, μ is a constant term, φ1, φ2, etc. are autoregressive coefficients, θ1, θ2, etc. are moving average coefficients, εt is an error term, p and q are model parameters, and the model parameters are determined by fitting historical data to more accurately predict the changing trend of pedestrian and vehicle flow.

[0055] Useful information is extracted, including the changing trends of sunrise and sunset times, cloud thickness, and pedestrian and vehicle traffic. Based on the analysis results, a multivariate linear regression model is used to predict lighting needs in different regions and at different times: lighting needs = α + β1 × illuminance + β2 × pedestrian traffic + β3 × vehicle traffic + β4 × seasonal factors + β5 × weather factors + … + ε. α is a constant term, β1, β2, β3, etc. are regression coefficients, and ε is an error term. Seasonal factors are simulated using a sine function, and the formula is: Where A is the amplitude, ω is the angular frequency, and t is the time (in seasons), The weather factor is quantified through fuzzy logic, and a comprehensive evaluation is performed based on multiple factors such as cloud thickness and precipitation probability to obtain a value between 0 and 1, which indicates the degree of influence of weather on lighting demand.

[0056] 2. Intelligent control module

[0057] According to the prediction results of the data acquisition and analysis module, the lighting on and off time is automatically adjusted. After sunrise and before sunset, the linear adjustment algorithm is used to gradually adjust the lighting brightness according to the change in illuminance. Lighting brightness = initial brightness × (1-current illuminance ÷ maximum illuminance) × g^(-h × illuminance change speed). Among them, g and h are adjustment coefficients. The lighting brightness is gradually adjusted as the illuminance changes. At the same time, the influence of the illuminance change speed on the lighting brightness adjustment is considered. The faster the change speed, the smaller the adjustment range is, so as to avoid frequent and large changes in brightness.

[0058] At night, the lighting brightness is adjusted dynamically according to the changes in pedestrian and vehicle traffic using a dynamic adjustment algorithm. Lighting brightness = basic brightness + Δ brightness, where Δ brightness = k × pedestrian flow ^ m × vehicle flow ^ n, k, m, n are adjustment coefficients. The greater the pedestrian and vehicle flow, the greater the Δ brightness. The coefficients are adjusted to adapt to the traffic flow characteristics of different areas. Increase lighting brightness in areas with large pedestrian and vehicle traffic, reduce lighting brightness in areas with small pedestrian and vehicle traffic, and temporarily turn off lighting in areas where there are no people or vehicles.

[0059] Support remote control function, managers can manually control smart lamps through the management platform or mobile applications.

[0060] By collecting a large amount of data through a variety of sensors and applying complex algorithms for analysis and processing, it is possible to accurately predict the lighting needs of different areas and at different times. For example, the lighting on and off time and brightness are automatically adjusted according to weather conditions, seasonal changes, pedestrian and vehicle traffic, etc., to maximize energy conservation while ensuring the lighting effect. Personalized adjustments are made according to the characteristics of different areas, increasing the lighting brightness in areas with large pedestrian and vehicle traffic to ensure safety and convenience; reducing the lighting brightness in areas with small pedestrian and vehicle traffic to reduce energy waste and light pollution; and temporarily turning off the lighting in areas with no people or cars to further improve energy efficiency.

[0061] 3. Fault detection and early warning module

[0062] Real-time monitoring of the operating status of smart lamps, including current, voltage, power and other parameters. When a lamp fault is detected, a fault judgment algorithm is used to automatically issue a warning message, and the fault information is sent to the management platform and mobile application to notify maintenance personnel to handle and locate the faulty lamp. The fault judgment algorithm is: when the current, voltage, power and other parameters of the lamp exceed the normal range [lower limit, upper limit], the lamp is judged to be faulty. The principal component analysis algorithm is further introduced to reduce the dimension of the lamp's operating status data. Real-time monitoring of the operating status of smart lamps can detect faults in a timely manner and automatically issue warning information to notify maintenance personnel to handle. At the same time, the faulty lamp is located to facilitate maintenance personnel to quickly find the fault point, improve maintenance efficiency, and ensure the stable operation of the urban lighting system.

[0063] 4. Energy management module

[0064] Statistical analysis of the energy consumption of urban lighting systems, including electricity consumption and electricity consumption time. Based on energy consumption data, energy-saving strategies are formulated using energy-saving strategy optimization algorithms, including optimizing lighting on and off time, adjusting lighting brightness, etc. The energy-saving strategy optimization algorithm is: taking energy consumption minimization as the objective function, the objective function formula is Minimize E = Σ(Pi×ti), where E is the total energy consumption, Pi is the power of the i-th lamp, and ti is the on time of the i-th lamp. Combining lighting demand and renewable energy supply as constraints, the genetic algorithm is used to solve the optimal lighting on and off time and lighting brightness adjustment plan. In the genetic algorithm, chromosomes are represented as a set of lighting control parameters. Chromosomes are continuously optimized through selection, crossover, and mutation operations to find the optimal solution that meets the objective function and constraints. Statistical analysis of the energy consumption of urban lighting systems, the energy-saving strategy optimization algorithm is used to formulate reasonable energy-saving strategies, and combined with renewable energy supply to achieve optimal energy utilization. The genetic algorithm is used to solve the optimal lighting on and off time and lighting brightness adjustment plan, with the goal of minimizing energy consumption, to reduce the energy cost of the urban lighting system.

[0065] Combined with the long short-term memory network (LSTM) in deep learning, the future energy demand is predicted. Let the input sequence be X = (x1, x2, ..., xt), where xt represents the feature vector at time t, including power consumption, light intensity, and pedestrian flow. The LSTM unit updates the cell state Ct and hidden state ht through the forget gate, input gate, and output gate. The forget gate ft decides to discard the filtered information from the cell state, ft = σ(Wf · [ht-1, xt] + bf), where σ is the sigmoid function, Wf is the weight matrix, and bf is the bias term; the input gate it decides which new information is stored in the cell state, it = σ(Wi · [ht-1, xt] + bi), the candidate cell state Ct ~ = tanh(Wc · [ht-1, xt] + bc), the cell state Ct = ftCt-1 + itCt ~; the output gate ot decides the output of the filtered information, ot = σ(Wo · [ht-1, xt] + bo), the hidden state ht = ot*tanh(Ct). By training the LSTM network, the future energy demand can be predicted in order to better formulate energy-saving strategies. Using the long short-term memory network (LSTM) to predict future energy demand can make energy-saving strategies in advance and better cope with changes in energy demand.

[0066] 5. Report and statistical analysis module

[0067] Generate daily, monthly, and annual reports to display the operation status and energy consumption of the city lighting system.

[0068] Perform statistical analysis on historical data to provide decision support for managers. Set up a data mining unit to conduct in-depth analysis of historical data, use association rule algorithm to mine the potential association between different environmental factors and lighting needs, use Apriori algorithm, set the minimum support as min_sup, and the minimum confidence as min_conf, and continuously scan the data set to find frequent item sets, and then generate strong association rules to provide more accurate decision-making basis for lighting energy efficiency management. The formula is expressed as follows: If the frequency of item set X in the data set is greater than or equal to min_sup, then X is a frequent item set; if the confidence of rule X→Y Conf(X→Y)=P(Y|X) is greater than or equal to min_conf, then X→Y is a strong association rule, where P(Y|X) represents the probability of Y appearing under the condition that X appears. The report and statistical analysis module generates various reports to display the operation and energy consumption of the urban lighting system, providing intuitive data display for managers. At the same time, statistical analysis and data mining are performed on historical data to mine the potential association between different environmental factors and lighting needs, providing more accurate decision support for managers, and helping to optimize the layout and management of urban lighting systems.

[0069] The above specific embodiments are only several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant inspirations of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A smart city lighting energy efficiency management system based on the Internet of Things, characterized in that: include: The data collection and analysis module collects environmental data and traffic flow data in real time through the light intensity sensors, temperature sensors, humidity sensors, pedestrian flow sensors, and vehicle flow sensors deployed in various areas of the city. The environmental data includes light intensity, temperature, humidity, pedestrian flow, and vehicle flow. The collected data is analyzed and processed using the weighted average algorithm and trend analysis algorithm to extract useful information, including sunrise and sunset times, cloud thickness, and the changing trends of pedestrian and vehicle flow. Based on the analysis results, the regression analysis algorithm is used to predict the lighting needs in different areas and at different times. The intelligent control module automatically adjusts the lighting on and off time according to the prediction results of the data collection and analysis module. It uses a linear adjustment algorithm to gradually adjust the lighting brightness according to the changes in illumination after sunrise and before sunset. It uses a dynamic adjustment algorithm to dynamically adjust the lighting brightness according to the changes in pedestrian and vehicle traffic at night. It increases the lighting brightness in areas with large pedestrian and vehicle traffic, reduces the lighting brightness in areas with small pedestrian and vehicle traffic, and temporarily turns off the lighting in areas without people or cars. It also supports remote control functions, and managers can manually control the intelligent lamps through the management platform or mobile applications. The fault detection and early warning module monitors the operating status of smart lamps in real time, including current, voltage, and power parameters. When a lamp fault is detected, it automatically issues an early warning message using a fault determination algorithm, and sends the fault information to the management platform and mobile application to notify maintenance personnel to handle the problem and locate the faulty lamp. Energy management module, which statistically analyzes the energy consumption of urban lighting systems, including power consumption and power consumption time. Based on energy consumption data, energy-saving strategies are formulated using energy-saving strategy optimization algorithms. Energy-saving strategies include optimizing lighting on and off times and adjusting lighting brightness. The report and statistical analysis module generates reports, including daily reports, monthly reports, and annual reports, which display the operation status and energy consumption of the urban lighting system, conduct statistical analysis of historical data, and provide decision support for managers.

2. According to the IoT-based smart city lighting energy efficiency management system of claim 1, it is characterized in that: In the data collection and analysis module, the weighted average algorithm formula is: comprehensive index value = Σ (index value × weight); Among them, the calculation formula for light intensity weight is W1=a×(1-e^(-b×light intensity change rate)), the calculation formula for pedestrian flow weight is W2=c×(1-e^(-d×pedestrian flow standard deviation)), and the calculation formula for vehicle flow weight is W3=e×(1-e^(-f×vehicle flow peak-to-valley ratio)); a, b, c, d, e, and f are adjustment coefficients. These coefficients are continuously adjusted to adapt to different urban environments and needs. Different weights are given to the indicator values ​​of each sensor data according to the actual situation to obtain a comprehensive environmental data evaluation result.

3. According to the IoT-based smart city lighting energy efficiency management system of claim 1, it is characterized in that: In the data collection and analysis module, the trend analysis algorithm is: by performing differential calculation on the pedestrian and vehicle flow data within a period of time, if the differential result shows an upward trend, it is judged that the pedestrian and vehicle flow have an increasing trend, otherwise it is a decreasing trend. The autoregressive integral moving average model is further introduced to predict the long-term trend of the pedestrian and vehicle flow. The formula is: Yt=μ+φ1Yt-1+φ2Yt-2+…+φpYt-p+εt+θ1εt-1+θ2εt-2+…+θqεt-q; Among them, Yt is the pedestrian or vehicle flow at the current moment, μ is a constant term, φ1, φ2, etc. are autoregressive coefficients, θ1, θ2, etc. are sliding average coefficients, εt is an error term, p and q are model parameters. The model parameters are determined by fitting historical data to more accurately predict the changing trends of pedestrian and vehicle flows.

4. According to the IoT-based smart city lighting energy efficiency management system of claim 1, it is characterized in that: The regression analysis algorithm uses a multiple linear regression model: Lighting demand = α + β1 × illuminance + β2 × pedestrian flow + β3 × vehicle flow + β4 × seasonal factors + β5 × weather factors + … + ε; Among them, α is a constant term, β1, β2, β3, etc. are regression coefficients, and ε is an error term. The seasonal factor is simulated by a sine function, and the formula is Where A is the amplitude, ω is the angular frequency, and t is the time (in seasons), is the phase; Weather factors are quantified through fuzzy logic, and a comprehensive evaluation is performed based on multiple factors such as cloud thickness and precipitation probability to obtain a value between 0 and 1, which indicates the degree of impact of weather on lighting needs.

5. According to the IoT-based smart city lighting energy efficiency management system of claim 1, it is characterized in that: In the intelligent control module, the linear adjustment algorithm is: Lighting brightness = initial brightness × (1-current illuminance ÷ maximum illuminance) × g^(-h × illuminance change speed); Among them, g and h are adjustment coefficients. The lighting brightness is gradually adjusted as the illuminance changes. At the same time, the influence of the illuminance change speed on the lighting brightness adjustment is considered. The faster the change speed, the smaller the adjustment range.

6. According to the IoT-based smart city lighting energy efficiency management system of claim 1, it is characterized in that: The dynamic adjustment algorithm in the intelligent control module is: Lighting brightness = basic brightness + Δ brightness; Among them, Δ brightness = k × pedestrian flow ^ m × vehicle flow ^ n, k, m, n are adjustment coefficients. The greater the pedestrian flow and vehicle flow, the greater the Δ brightness.

7. The smart city lighting energy efficiency management system based on the Internet of Things according to claim 1 is characterized in that: In the fault detection and early warning module, the fault judgment algorithm is: when the current, voltage and power parameters of the lamp exceed the normal range [lower limit, upper limit], the lamp is judged to be faulty, and the principal component analysis algorithm is further introduced to perform dimensionality reduction processing on the operating status data of the lamp.

8. The smart city lighting energy efficiency management system based on the Internet of Things according to claim 1 is characterized in that: In the energy management module, the energy-saving strategy optimization algorithm is: taking minimization of energy consumption as the objective function, the objective function formula is Minimize E = Σ(Pi×ti), where E is the total energy consumption, Pi is the power of the i-th lamp, and ti is the turn-on time of the i-th lamp. Combining lighting demand and renewable energy supply as constraints, the optimal lighting turn-on and turn-off time and lighting brightness adjustment plan are solved through a genetic algorithm.

9. The smart city lighting energy efficiency management system based on the Internet of Things according to claim 1 is characterized in that: The report and statistical analysis module includes a data mining unit, which conducts in-depth analysis of historical data and uses association rule algorithms to mine potential associations between different environmental factors and lighting requirements. The Apriori algorithm is used, with the minimum support as min_sup and the minimum confidence as min_conf. Frequent item sets are found by continuously scanning the data set, and strong association rules are generated to provide a more accurate decision-making basis for lighting energy efficiency management. The formula is expressed as follows: If the frequency of item set X in the data set is greater than or equal to min_sup, then X is a frequent item set; if the confidence Conf(X→Y)=P(Y|X) of rule X→Y is greater than or equal to min_conf, then X→Y is a strong association rule, where P(Y|X) represents the probability of Y appearing under the condition that X appears.

10. The smart city lighting energy efficiency management system based on the Internet of Things according to claim 1, characterized in that: The energy management module combines the long short-term memory network in deep learning to predict future energy demand. Suppose the input sequence is X=(x1, x2, ..., xt), where xt represents the feature vector at time t, including power consumption, light intensity, and pedestrian flow. The LSTM unit updates the cell state Ct and hidden state ht through the forget gate, input gate, and output gate. The forget gate ft determines the information to be removed from the cell state ft = σ(Wf · [ht-1, xt] + bf) Among them, σ is the sigmoid function, Wf is the weight matrix, and bf is the bias term; The input gate it determines the filtered information to be stored in the cell state it = σ(Wi · [ht-1, xt] + bi), Candidate cell state Ct~=tanh(Wc·[ht-1,xt]+bc), cell state Ct=ftCt-1+itCt~; The output gate ot determines the filtered information to be output, ot = σ(Wo·[ht-1,xt]+bo), and the hidden state ht = ot*tanh(Ct).

Images