Fuzzy control method for intelligently adjusting lighting time of street lamp

Abstract

The invention discloses a fuzzy control method for intelligently adjusting the lighting time of a street lamp, and the method employs a K-Fuzzy fuzzy control algorithm of a double-input single-output control architecture, carries out the fuzzy processing of two input variable vehicle numbers and vehicle speeds through the double inputs, builds a vehicle number universe and a vehicle speed universe, and achieves the intelligent adjustment of the lighting time of the street lamp based on a fuzzy control rule. And a Mamdani inference mechanism is adopted to carry out logical operation on the input parameters, a gravity center method is adopted to carry out defuzzification processing, a membership function is converted into a delay control quantity, and the delay control quantity is the lighting duration of the street lamp. The road traffic efficiency can be improved, the energy consumption is effectively reduced, and the traffic safety is improved.

Description

Technical Field

[0001] The present invention relates to the field of control technology, and particularly to a fuzzy control method for intelligently adjusting the lighting time of street lamps. Background Art

[0002] [1] Han et al. proposed an energy-saving strategy based on multi-source information fusion and edge computing. By using multi-sensors and convolutional neural networks to count the traffic flow of people and vehicles, and combining with a visual statistics system to optimize the street lamp control cycle, real-time brightness adjustment is achieved through edge computing, significantly reducing power consumption. [2] Chen et al. designed an intelligent street lamp control algorithm with high energy consumption ratio, which collects environmental light and traffic flow data in real time, and dynamically adjusts the brightness through a prediction model to achieve a balance between energy consumption and lighting effect, with the energy consumption reduced by about 35%. Although the above methods perform excellently in energy-saving efficiency, some methods are still limited by complex environmental changes and real-time requirements, and there is still room for optimization.

[0003] [1] Han Qiang. Research on the Energy-saving Control Strategy of Street Lamps Based on Multi-sensor Information Fusion and Edge Computing [J]. Internet of Things Technologies, 2020, 10(10): 43-45+49. DOI: 10.16667 / j.issn.2095-1302.2020.10.012.

[0004] [2] Chen Yingfei, Zhang Zhenpeng. Research on the Intelligent Street Lamp Energy-saving Control Algorithm with High Energy Consumption Ratio [J]. Light & Lighting, 2024, (10): 57-59.

[0005] Fuzzy control is an intelligent control method based on fuzzy set theory, fuzzy language variables, and fuzzy logical reasoning. It is an intelligent control method that mimics the fuzzy reasoning and decision-making process of humans in terms of behavior. This method first encodes the experience of operators or experts into fuzzy rules, then fuzzifies the real-time signals from sensors, uses the fuzzified signals as the input of the fuzzy rules to complete fuzzy reasoning, and adds the output obtained after reasoning to the actuator. Summary of the Invention

[0006] To further optimize energy consumption and adapt to complex environmental changes and real-time requirements, the present invention proposes a fuzzy control method for intelligently adjusting the lighting time of street lamps, which can improve the road traffic efficiency, effectively reduce energy consumption, and enhance traffic safety.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A fuzzy control method for intelligently adjusting the lighting time of street lamps, which adopts the K-Fuzzy fuzzy control algorithm with a double-input single-output control architecture. The two inputs respectively perform fuzzy processing on two input variables, namely the number of vehicles and the vehicle speed, establish the domain of the number of vehicles and the domain of vehicle speed, and based on the fuzzy control rules, use the Mamdani inference mechanism to perform logical operations on the input parameters, and use the centroid method for defuzzification processing to convert the membership function into a delay control quantity, and the delay control quantity is the lighting duration of the street lamp.

[0009] Preferably, both the domain of the number of vehicles and the domain of vehicle speed are [0, 50], and are divided into 5 fuzzy subsets. The domain of the delay control quantity is [0, 120], with the unit of seconds, and is divided into five fuzzy subsets;

[0010] The membership functions of the number of vehicles, vehicle speed, and delay control quantity all adopt triangular membership functions:

[0011] μ(x) = max(0, 1 - |x - a| / b)

[0012] Among them, a is the vertex of the triangle, b is half of the base width, and x is the input quantity.

[0013] Preferably, the 5 fuzzy subsets of the domain of the number of vehicles are respectively "very few", "few", "medium", "many", "very many" in which the number of vehicles increases in sequence. "Very few" means less than 5 vehicles, and "very many" means at least 45 vehicles. The membership function of the number of vehicles adopts an overlapping triangular design: when x ∈ [0, 5], it is defined as the rising edge function of "very small"; in the interval of x ∈ [45, 50], the falling edge function of "very large" is set, and the middle three fuzzy subsets "few", "medium", and "many" form a 20% overlapping area with an interval of 10 vehicles;

[0014] The 5 fuzzy subsets of the domain of vehicle speed are respectively "very slow", "relatively slow", "normal", "relatively fast", "very fast" in which the vehicle speed increases in sequence. "Very slow" means the vehicle speed is less than 5 km / h, and "very fast" means the vehicle speed is not less than 45 km / h. The membership function of vehicle speed is constructed with a step of 10 km / h, and a 25% overlapping band is set in the vehicle speed mutation area, and the vehicle speed mutation area includes the area centered on 5 km / h and / or 45 km / h.

[0015] Preferably, the 5 fuzzy subsets of the domain of the delay control quantity are respectively "very short", "short", "medium", "long", "very long" which increase in sequence. "Very short" means less than 15 seconds, "very long" means not less than 110 seconds, and "short", "medium", and "long" form a 20% overlapping area with an interval of 30 seconds. The overall membership function formulas of the number of vehicles, vehicle speed, and the lighting duration of the street lamp are as follows:

[0016]

[0017] Wherein:

[0018] When x < Xi-1, the membership degree is 0, which means that when the input value is less than the left endpoint, it does not belong to this fuzzy set;

[0019] When Xi-1 ≤ x < Xi, the membership degree increases linearly, from 0 to 1.

[0020] When Xi ≤ x ≤ Xi+1, the membership degree decreases linearly, from 1 to 0.

[0021] When x > Xi+1, the membership degree becomes 0 again, indicating that when the input value is greater than the right endpoint, it does not belong to this fuzzy set.

[0022] Preferably, the fuzzy control rules are as shown in the following table,

[0023]

[0024]

[0025] Preferably, the Mamdani inference mechanism is an intelligent decision-making mechanism based on multi-level fuzzy logic operations, including:

[0026] For the fuzzified vehicle quantity and vehicle speed parameters, the AND operation is performed using the minimum method to extract the triggering intensity of the rules that meet the conditions;

[0027] For the OR operation with multi-channel inputs, the multi-dimensional features are dynamically aggregated through the maximum method. The OR operation with multi-channel inputs is the OR operation performed on the vehicle quantities on multiple roads;

[0028] In the rule activation stage, the minimum "implication" operation is used to generate the output membership function of each rule, and the outputs of multiple rules are synthesized and superimposed through the maximum method to form a global output fuzzy set with probability distribution characteristics;

[0029] Defuzzification is performed using the centroid. By calculating the geometric centroid of the area under the output membership function curve, the output fuzzy quantity is converted into the lighting duration of the street lamp.

[0030] Preferably, the method for converting the output fuzzy quantity into the lighting duration of the street lamp is as follows:

[0031] When 0 < V < 5, the lighting duration of the street lamp is:

[0032] L_sum = 30 + F(V,S) + 6 + 35 = F(V,S) + 71

[0033] When V ≥ 5, the lighting duration of the street lamp is:

[0034]

[0035] When V = 0, the street lamp is turned off and L_sum = 0.

[0036] Wherein, L_sum is the lighting duration of the street lamp, with the unit of seconds, V is the number of vehicles, S is the vehicle speed, and F(V, S) is the output fuzzy quantity.

[0037] The beneficial effects of the present invention are as follows:

[0038] 1. The present invention has high and stable energy-saving efficiency, and can delay the lighting time long enough to meet the traffic needs, improving traffic safety.

[0039] 2. The present invention adopts the K-Fuzzy fuzzy control algorithm, which shows good dynamic characteristics during the response process, can effectively adapt to complex traffic environments, ensure the reasonable regulation of traffic signals, and thus provides strong technical support for urban traffic management.

[0040] 3. The present invention can effectively promote the development of intelligent control technology for smart street lamps, and has good social and economic effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a general view of the K-Fuzzy fuzzy control algorithm.

[0042] Figure 2 It is a structural schematic diagram of the K-Fuzzy fuzzy control algorithm.

[0043] Figure 3 It is a schematic diagram of the membership function of the number of vehicles.

[0044] Figure 4 It is a schematic diagram of the membership function of the vehicle speed.

[0045] Figure 5 It is a schematic diagram of the membership function of the lighting time of the street lamp output by the fuzzy controller. DETAILED DESCRIPTION OF THE INVENTION

[0046] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings.

[0047] Example 1

[0048] This embodiment discloses a fuzzy control method for intelligently adjusting the lighting time of street lamps, which is an intelligent street lamp regulation based on the K-Fuzzy fuzzy control algorithm. The core lies in realizing dynamic adjustment of the street lamp duration by collecting traffic flow and vehicle speed data in real time. This algorithm adopts a dual-input single-output control architecture. First, it performs fuzzy processing on the input variables, namely the number of vehicles (vehicles) and vehicle speed (km / h), and respectively establishes a universe of discourse of the number of vehicles composed of five fuzzy sets: "very few, few, medium, many, very many" and a universe of discourse of vehicle speed composed of five fuzzy sets: "very slow, slow, normal, fast, very fast". A fuzzy rule base containing 25 control rules is constructed through expert experience. For example, when "the number of vehicles is high and the vehicle speed is fast", the "long delay" control strategy is triggered. In the process of fuzzy reasoning, the Mamdani reasoning mechanism is used to perform logical operations on the input parameters, and then the centroid method is used for defuzzification processing to convert the membership function into an accurate delay control quantity (in seconds). The entire control process is presented completely through Figure 1 and Figure 2 shown in the system architecture diagram. Among them, the data acquisition module and the fuzzy controller form a closed-loop control system, which can automatically optimize the street lamp lighting duration according to the real-time traffic state, and achieve energy consumption optimization on the premise of ensuring driving safety.

[0049] In the K-Fuzzy fuzzy control algorithm, the input quantities are the number of vehicles and vehicle speed. Among them, the universes of discourse of the number of vehicles and vehicle speed are both [0, 50], which are divided into 5 fuzzy subsets, namely V very few / very slow, few / slow, medium, many / fast, very many / very fast.

[0050] In view of the non-linear characteristics of urban road traffic flow, the present invention proposes a design method of membership function based on five-level fuzzy division. In the fuzzy processing of the number of vehicles, according to the statistical law of actual road monitoring data, the universe of discourse is divided into five fuzzy subsets: "very small (<5 vehicles), relatively small, medium, relatively large, very large (≥45 vehicles)", and the boundary thresholds are set by fully considering the differential characteristics of low-flow roads (such as branch roads) and main roads. The vehicle speed universe of discourse constructs a five-level fuzzy set of "very slow (<5 km / h), relatively slow, normal, relatively fast, very fast (≥45 km / h)". This division scheme can effectively distinguish typical working conditions such as congested creep (<5 km / h) and speeding (≥45 km / h). In view of the non-linear distribution characteristics of traffic parameters, such as Figure 3 、 Figure 4 shown, the present invention adopts a triangular membership function with high calculation efficiency and easy parameter adjustment, and its mathematical expression is:

[0051] μ(x) = max(0, 1 - |x - a| / b)

[0052] where a is the vertex of the triangle, b is half of the base width, and x is the input quantity.

[0053] Specifically, the membership function of the vehicle quantity adopts an overlapping triangular design: when \(x\in[0,5]\), it is defined as a rising-edge function of "very small"; in the interval \(x\in[45,50]\), a falling-edge function of "very large" is set, and the middle three fuzzy sets form a 20% overlapping area at intervals of 10 vehicles. The membership function of the vehicle speed is constructed in steps of 10 km / h, and a 25% overlapping band is set at the speed mutation area (such as the critical points of 5 km / h and 45 km / h) to enhance the robustness of the system. This design ensures the real-time calculation while achieving the smoothness of state transition through the overlapping areas of adjacent fuzzy sets, effectively avoiding the step problem of the control quantity caused by the traditional rectangular membership function.

[0054] In the present invention, the output of the fuzzy controller, that is, the lighting time of the street lamp, is set within the domain range of [0, 120] seconds, and a five-level fuzzy division strategy is adopted, dividing it into five fuzzy subsets of "very short (<15 seconds), short, medium, long, very long (≥110 seconds)". This division method fully considers the actual lighting needs of urban roads: when the lighting time is less than 15 seconds, it can meet the needs of sporadic vehicles passing quickly; when the lighting time exceeds 110 seconds, it is suitable for continuous lighting needs during peak traffic hours or special weather conditions. Considering the non-linear mapping relationship between the lighting time and the traffic flow, Figure 5 As shown, the present invention adopts a triangular membership function with high calculation efficiency and flexible parameter adjustment. Specifically, clear boundaries of 15 seconds and 110 seconds are set in the two extreme regions (very short and very long) of the lighting time, and the middle three fuzzy subsets (short, medium, long) form a 20% overlapping area at intervals of 30 seconds. This design not only ensures the accuracy of control but also realizes the smoothness of state transition. Through the combination of this five-level fuzzy division and the triangular membership function, the system can effectively handle the randomness and uncertainty of traffic flow while ensuring real-time performance, realizing the intelligent adjustment of the lighting time of street lamps.

[0055] The overall membership function formulas of vehicle quantity, vehicle speed, and the lighting time of street lamps are expressed as follows:

[0056]

[0057] The explanations of the function formulas are as follows:

[0058] When \(x < X_{i - 1}\), the membership degree is 0, which means that when the input value is less than the left endpoint, it does not belong to this fuzzy set;

[0059] When \(X_{i - 1}\leq x < X_{i}\), the membership degree increases linearly from 0 to 1.

[0060] When \(X_{i}\leq x\leq X_{i + 1}\), the membership degree decreases linearly from 1 to 0.

[0061] When x > Xi+1, the membership degree becomes 0 again, indicating that when the input value is greater than the right endpoint, it also does not belong to the fuzzy set.

[0062] The K-Fuzzy fuzzy control algorithm disclosed by the present invention adopts a fuzzy rule base based on expert experience, as shown in the following table:

[0063]

[0064]

[0065] By analyzing the coupling relationship between the number of vehicles and the vehicle speed to dynamically adjust the lighting time of street lights, the fuzzy rule base contains 25 control rules, and uses the form of "if... then..." to describe the non-linear relationship between input and output variables. For example, when it is detected that "the number of vehicles is small and the vehicle speed is fast", the rule of "very short" lighting duration of street lights is triggered, which is applicable to the sparse traffic period at night; when "the number of vehicles is large and the vehicle speed is slow" appears, the rule of "very long" lighting duration of street lights is activated to meet the lighting needs during the traffic peak. The intermediate states respectively correspond to three lighting durations of street lights, namely "short", "medium" and "long", according to different combinations of the number of vehicles and the vehicle speed. This rule design based on traffic flow characteristics not only conforms to the basic principle of traffic engineering that "the traffic flow density is negatively correlated with the vehicle speed", but also can effectively adapt to the road lighting needs at different times, achieving energy-saving optimization while ensuring driving safety.

[0066] In the design of the inference mechanism of the K-Fuzzy fuzzy control algorithm, an intelligent decision-making system based on multi-level fuzzy logic operations is constructed. This intelligent decision-making system realizes an accurate mapping from input to output through three stages: First, for the fuzzified vehicle quantity ({"very few", "few", "medium", "many", "very many"}) and vehicle speed ({"very slow", "slow", "normal", "fast", "very fast"}) parameters, the "AND" operation is performed using the minimum method to accurately extract the triggering intensity of the composite condition rules; for the "OR" operation with multi-channel inputs, the multi-dimensional features are dynamically aggregated through the maximum method to enhance the adaptability of the system to complex traffic scenarios. In the rule activation stage, the minimum "implication" operation is used to generate the output membership function of each rule to ensure that the inference process conforms to the causal constraints of fuzzy logic. Subsequently, the outputs of multiple rules are synthesized and superimposed through the maximum method to form a global output fuzzy set with probability distribution characteristics, effectively retaining the contribution information of each rule. Finally, defuzzification is performed using the centroid. By calculating the geometric centroid of the area under the output membership function curve, the fuzzy quantity is converted into an accurate lighting time value. Through the collaboration of the minimum-maximum operator combination and centroid calculation, this mechanism not only solves the modeling problem of the non-linear coupling relationship between traffic parameters but also realizes the smooth transition of the control quantity. Its unique multi-level inference architecture significantly improves the anti-interference ability of the system, enabling the lighting duration regulation to have both dynamic response speed and steady-state accuracy, providing a theoretically complete and engineering-feasible solution for intelligent lighting in complex traffic environments.

[0067] This embodiment defines the following symbol description table for the K-Fuzzy fuzzy control algorithm:

[0068] .Symbol Description Table for K-Fuzzy Algorithm

[0069]

[0070] F(V, S) is the lighting time output by the fuzzy controller, that is:

[0071] F(V,S) = evalfis(fuzzyController[u,s])

[0072] When 0 < V < 5 (the vehicle quantity is greater than 0 and less than 5),

[0073] the lighting time output by the fuzzy controller is F(V, S), and the actual lighting time (including the time compensation for the vehicle approaching and the time compensation for the vehicle passing by the street lamp) is

[0074] L_sum = 30 + F(V, S) + 6 + 35 = F(V, S) + 71

[0075] The energy consumption calculation for this period is (converting seconds to hours):

[0076]

[0077] When V ≥ 5 (the number of vehicles is greater than or equal to 5),

[0078] The output of the fuzzy controller for the lighting time is still denoted as F(V, S). At this time, the actual lighting time is calculated as

[0079]

[0080] Since the numerator 140 + 20 + 20v can be combined into 160 + 20v, and the denominator s / 3.6 is used to convert km / h to m / s, the formula can be converted to

[0081]

[0082] The corresponding energy consumption is

[0083] Y

[0084] When V = 0 (the number of vehicles is 0),

[0085] That is, no vehicles pass by and the street lights are turned off. So, we have:

[0086] L_sum = 0

[0087] The corresponding energy consumption is:

[0088] E_period = 0

[0089] There are N sampling periods in the entire experimental process. Therefore, the total energy consumption Etotal is:

[0090]

[0091] The energy consumption of always-on:

[0092] Ealways_on = Ttotal × P

[0093] Energy saved:

[0094] E_saved = Ealways_on × Etotal

[0095] Energy saving rate:

[0096]

[0097] Example 2

[0098] Based on Example 1, this example discloses a verification instance of the fuzzy control method for intelligently adjusting the lighting time of street lights. Experiments are carried out with different numbers of vehicles and the number of lights in each area to verify the energy-saving efficiency of the system. The results are shown in the following table:

[0099]

[0100]

[0101]

[0102] As can be seen from the above table, when there are relatively few vehicles (e.g., 1 - 4 vehicles) and the vehicle speed is relatively fast (e.g., 30 km / h, 40 km / h, where the energy-saving efficiency is maintained at about 63% in the case of 30 km / h and about 66% in the case of 40 km / h), the K-Fuzzy algorithm shows a relatively high energy-saving efficiency. When the number of streetlights is 2 and the vehicle speed is 40 km / h, as the number of vehicles ranges from 1 to 4, the energy-saving efficiency is maintained at about 66%, and the energy consumption loss is maintained at about 0.473 kw / h. It can be seen that at low traffic flow and high vehicle speed, the fuzzy controller can dynamically adjust the streetlight on-time (about 29 seconds) to effectively reduce redundant energy consumption. However, when the number of vehicles reaches the threshold of 5 set by the K-Fuzzy algorithm, the energy consumption of the streetlights will increase significantly (e.g., when the vehicle speed is 20 km / h, the energy consumption is about 0.564 kw / h), so that the fuzzy controller can delay the on-time long enough to meet the traffic needs in the case of high traffic flow. Therefore, the energy-saving rate drops to about 60%.

[0103] With the increase in the number of lights, it will lead to a significant increase in energy consumption, but the energy-saving efficiency remains stable. The K-Fuzzy algorithm has obvious advantages under medium traffic flow (or low traffic flow) and high vehicle speed (or medium vehicle speed) conditions, with high and stable energy-saving efficiency.

[0104] The K-Fuzzy fuzzy control algorithm exhibits good dynamic characteristics during the response process, with a rise time of 5 minutes. This characteristic enables the system to quickly respond to vehicle inputs in the initial stage, ensuring that the output lighting time can rapidly reach the expected target, which is 110 seconds. This rapid response ability is crucial for traffic management, capable of effectively coping with sudden traffic flow changes and enhancing road traffic efficiency. After a stable time of up to 420 minutes, which almost covers the entire simulation cycle, the system demonstrates good stability. At this stage, the steady-state error is only 1.565 seconds, indicating that at the end of the simulation, the gap between the system output and the target value is very small, showing a high final control accuracy. This accuracy level stems from three aspects of optimized design: First, the precise integral operation of the geometric characteristics of the output membership function when using the centroid method to defuzzify; Second, retaining the effective control quantity components through the maximum superposition method during the rule synthesis stage; Third, the asymmetric distribution characteristics formed after optimizing the parameters of the triangular membership function by the particle swarm algorithm. The combination of this rapid response and high-precision maintenance ability not only meets the immediate lighting requirements under sudden traffic flow conditions but also achieves energy-saving optimization under normal traffic conditions, fully demonstrating the unique advantages of fuzzy control in dealing with nonlinear time-varying systems. This precise control ability not only improves the intelligent adjustment level of streetlights but also provides strong support for the smooth operation of traffic flow. In summary, the K-Fuzzy fuzzy control algorithm performs excellently in terms of response speed, stability, and control accuracy, can effectively adapt to complex traffic environments, ensure the reasonable regulation of traffic signals, and thus provide strong technical support for urban traffic management.

[0105] Of course, the present invention can also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and deformations according to the present invention, but these corresponding changes and deformations should all fall within the protection scope of the appended claims of the present invention.

Claims

1. A fuzzy control method for intelligently adjusting the lighting time of street lamps, characterized in that: The K-Fuzzy fuzzy control algorithm with a dual-input single-output control architecture. The two inputs respectively perform fuzzy processing on two input variables, namely the number of vehicles and the vehicle speed, establish the domain of discourse of the number of vehicles and the domain of discourse of the vehicle speed. Based on the fuzzy control rules, the Mamdani inference mechanism is used to perform logical operations on the input parameters, and the centroid method is used for defuzzification processing to convert the membership function into a delay control quantity, and the delay control quantity is the lighting duration of the street lamp.

2. The fuzzy control method for intelligently adjusting the lighting time of street lamps according to claim 1, wherein: The domain of discourse of the number of vehicles and the domain of discourse of the vehicle speed are both [0, 50], and are divided into 5 fuzzy subsets. The domain of discourse of the delay control quantity is [0, 120], with the unit of seconds, and is divided into five fuzzy subsets; The membership functions of the number of vehicles, the vehicle speed, and the delay control quantity all adopt triangular membership functions: μ(x) = max(0, 1 - |x - a| / b) Among them, a is the vertex of the triangle, b is half of the base width, and x is the input quantity.

3. The fuzzy control method for intelligently adjusting the lighting time of street lamps according to claim 2, characterized in that: The 5 fuzzy subsets of the domain of discourse of the number of vehicles are respectively "very few", "few", "medium", "many", "very many" with the number of vehicles increasing in sequence. "Very few" means less than 5 vehicles, and "very many" means at least 45 vehicles. The membership function of the number of vehicles adopts an overlapping triangular design: when x ∈ [0, 5], it is defined as the rising edge function of "very small"; in the interval x ∈ [45, 50], the falling edge function of "very large" is set, and the middle three fuzzy subsets "few", "medium", "many" form a 20% overlapping area with an interval of 10 vehicles; The 5 fuzzy subsets of the domain of discourse of the vehicle speed are respectively "very slow", "slow", "normal", "fast", "very fast" with the vehicle speed increasing in sequence. "Very slow" means the vehicle speed is less than 5 km / h, and "very fast" means the vehicle speed is not less than 45 km / h. The membership function of the vehicle speed is constructed with a step of 10 km / h, and a 25% overlapping band is set in the vehicle speed mutation area, and the vehicle speed mutation area includes the area centered on 5 km / h and / or 45 km / h.

4. The fuzzy control method for intelligently adjusting the lighting time of street lamps according to claim 3, characterized in that: The 5 fuzzy subsets of the domain of discourse of the delay control quantity are respectively "very short", "short", "medium", "long", "very long" increasing in sequence. "Very short" means less than 15 seconds, and "very long" means not less than 110 seconds. "Short", "medium", "long" form a 20% overlapping area with an interval of 30 seconds. The overall membership function formulas of the number of vehicles, the vehicle speed, and the lighting duration of the street lamp are as follows: Among them: When x < Xi - 1, the membership degree is 0, which means that when the input value is less than the left endpoint, it does not belong to this fuzzy set; When Xi - 1 ≤ x < Xi, the membership degree increases linearly, from 0 to 1. When Xi ≤ x ≤ Xi + 1, the membership degree decreases linearly, from 1 to 0. When x > Xi + 1, the membership degree becomes 0 again, indicating that when the input value is greater than the right endpoint, it does not belong to this fuzzy set.

5. The fuzzy control method for intelligently adjusting the lighting time of street lamps according to claim 4, characterized in that: The fuzzy control rules are as shown in the following table, 6. The fuzzy control method for intelligently adjusting the lighting time of street lamps according to claim 5, characterized in that: The Mamdani inference mechanism is an intelligent decision-making mechanism based on multi-level fuzzy logic operations, including: For the fuzzified vehicle quantity and vehicle speed parameters, the AND operation is performed using the minimum method to extract the trigger intensity that meets the conditional rules; For the OR operation with multi-channel inputs, the multi-dimensional features are dynamically aggregated by the maximum method. The OR operation with multi-channel inputs is the OR operation performed on the vehicle quantities on multiple roads; In the rule activation stage, the minimum "implication" operation is used to generate the output membership function of each rule, and the outputs of multiple rules are synthesized and superimposed by the maximum method to form a global output fuzzy set with probability distribution characteristics; Defuzzification is performed using the centroid. By calculating the geometric centroid of the area under the output membership function curve, the output fuzzy quantity is converted into the lighting duration of the street lamp.

7. The fuzzy control method for intelligently adjusting the lighting time of street lamps according to claim 6, characterized in that: The method for converting the output fuzzy quantity into the lighting duration of the street lamp is as follows: When 0 < V < 5, the lighting duration of the street lamp is: L_sum = 30 + F(V, S) + 6 + 35 = F(V, S) + 71 When V ≥ 5, the lighting duration of the street lamp is: When V = 0, the street lamp is turned off, L_sum = 0, where L_sum is the lighting duration of the street lamp, with the unit of seconds, V is the vehicle quantity, S is the vehicle speed, and F(V, S) is the output fuzzy quantity.

Images