A centralized control system for solar street lights

Abstract

The present invention relates to the technical field of solar street lamp control, and specifically provides a centralized control system for solar street lamps. The system can collect the first environmental data, the second solar panel data, the third battery data, and the fourth street lamp operation data on the highway in real time through a data acquisition module. The available power analysis module calculates the first available power based on the first environmental data and the second solar panel data, calculates the second storable power according to the third battery data, and finally obtains the third available power. The supply-demand difference analysis module generates a first power supply-demand difference matrix by calculating the total available power and the total required power in each region. The supply-demand balance control module dynamically selects a power scheduling strategy using a DQN network model and adaptively updates the Q function according to environmental feedback to achieve the adaptive distribution of power in each region. Through this system, the power supply-demand balance in each region can be effectively achieved, and the problem of insufficient power supply for solar street lamps can be effectively avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of solar street lamp control, and specifically provides a centralized control system for solar street lamps. Background Art

[0002] With the enhancement of environmental awareness, solar street lamps are widely used in urban roads, rural lighting, park and public facility lighting and other places due to their energy-saving and environmental protection advantages. Solar street lamps mainly rely on solar panels to convert light energy into electrical energy and are powered by energy storage batteries. However, due to the volatility of solar power generation and the limitation of battery energy storage capacity, especially in rainy weather or in winter with short daylight hours, the problem of insufficient power supply often occurs, resulting in the failure of street lamp lighting or unstable performance.

[0003] Existing solar street lamp systems often adopt an independent single-lamp power supply mode, lacking a centralized control system to effectively coordinate and manage the power distribution and use among street lamps, which easily leads to the problem of insufficient power supply for solar street lamps.

[0004] Therefore, a centralized control system for solar street lamps is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a centralized control system for solar street lamps. The present invention relates to the technical field of solar street lamp control, and specifically provides a centralized control system for solar street lamps. The present invention uses a data acquisition module to collect first environmental data, second solar panel data, third battery data, and fourth street lamp working data on the highway in real time; a power supply analysis module calculates the first generable power based on the first environmental data and the second solar panel data, calculates the second storable power according to the third battery data, and finally obtains the third available power supply; a supply-demand difference analysis module generates a first power supply-demand difference matrix by calculating the total available power supply and the total required power in each area; a supply-demand balance control module dynamically selects a power scheduling strategy using a DQN network model and adaptively updates the Q function according to environmental feedback to achieve the adaptive distribution of power in each area. Through this system, the power supply-demand balance in each area can be effectively achieved, and the problem of insufficient power supply for solar street lamps can be effectively avoided.

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

[0007] A centralized control system for solar street lamps, the system comprising:

[0008] A data acquisition module, configured to collect first environmental data, second solar panel data, third battery data, and fourth street lamp working data on the highway in a preset monitoring period in real time;

[0009] A power supplyable energy analysis module, configured to obtain the first generatable energy of the solar street lamp according to the first environmental data and the second solar panel data; obtain the second storable energy of the solar street lamp according to the third battery data, and obtain the third power supplyable energy of the solar street lamp according to the first generatable energy and the second storable energy;

[0010] A supply-demand difference analysis module, configured to divide regions of the solar street lamps on the highway, calculate the second total supplyable energy of each region according to the third power supplyable energy, and obtain the first total demand energy of each region according to the fourth street lamp working data; obtain the first power supply-demand difference matrix of each region according to the first total demand energy and the second total supplyable energy;

[0011] A supply-demand balance control module, using a DQN network model, inputs the first power supply-demand difference matrix into the model, in combination with the constraint conditions of regional power distribution; the model dynamically selects a power scheduling strategy and adaptively updates the Q function according to environmental feedback to achieve adaptive distribution of power in each region.

[0012] Preferably, the first environmental data includes light intensity and environmental temperature; the second solar panel data includes solar panel area, solar panel conversion efficiency, and solar panel tilt angle; the third battery data includes battery rated capacity, battery discharge depth, and charge-discharge efficiency; the fourth street lamp working data includes the working duration and working power of the solar street lamp.

[0013] Preferably, the first generatable energy is:

[0014]

[0015] wherein, E i produce represents the first generatable energy of the i-th solar street lamp; t_begin represents the start time of the preset monitoring period; t_end represents the end time of the preset monitoring period; S i represents the solar panel area of the i-th solar street lamp; I i (t) represents the light intensity of the i-th solar street lamp at time t; η i (t) represents the solar panel conversion efficiency of the i-th solar street lamp at time t; θ i represents the solar panel tilt angle of the i-th solar street lamp.

[0016] Preferably, the second storable energy is:

[0017] E i store =Ei rated *C i charge *DoD i ;

[0018] Among them, E i store represents the second storable electric energy of the i-th solar street lamp; E i rated represents the rated capacity of the i-th solar street lamp; C i charge represents the charge-discharge efficiency of the i-th solar street lamp; DoD i represents the depth of discharge of the storage battery of the i-th solar street lamp.

[0019] Preferably, the third available electric energy is:

[0020] E i available = Min(E i produce , E i store );

[0021] Among them, E i available represents the third available electric energy of the i-th solar street lamp; E i produce represents the first generable electric energy of the i-th solar street lamp; E i store represents the second storable electric energy of the i-th solar street lamp.

[0022] Preferably, the second total available electric energy is obtained by summing up the third available electric energies of all solar street lamps in each area; the first total required electric energy is obtained by multiplying the working hours, working power and number of solar street lamps in each area.

[0023] Preferably, the first electric energy supply-demand difference matrix is obtained by taking the difference between the second total available electric energy and the first total required electric energy in each area and constructing a matrix; the first electric energy supply-demand difference matrix is specifically:

[0024]

[0025] Among them, ESD represents the first electric energy supply-demand difference matrix; ESD j represents the electric energy supply-demand difference value of the j-th area; E j total_available represents the second total available electric energy of the j-th area; E j total_demandIt represents the total power demand of the j-th region, where j represents the number of regions.

[0026] Preferably, the DQN network model includes a state space, an action space, a reward function, and a Q function; the state space s t is represented by the first power supply-demand difference matrix; the action space represents the power distribution process between regions; the Q function reflects the long-term benefit after executing action a t in state s t .

[0027] Preferably, the reward function is:

[0028]

[0029] where R t represents the reward function; ESD i represents the power supply-demand difference value in the i-th region; j represents the number of regions.

[0030] Preferably, the updated Q function is:

[0031] Q new (s t , a t ) = Q(s t , a t ) + α * [R t + γ * max a' Q(s t+1 , a') - Q(s t , a t )];

[0032] where Q new (s t , a t ) represents the updated Q function; α represents the learning rate; Q(s t , a t ) represents the Q function before update; R t represents the reward function; γ represents the discount factor; max a' Q(s t+1 , a') represents the Q value after executing the optimal action a'.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] 1. The present invention collects the first environmental data, the second solar panel data, the third battery data, and the fourth street lamp working data of the highway in a preset monitoring period in real time; and based on the first environmental data and the second solar panel data, obtains the first generable electric energy of the solar street lamp; obtains the second storable electric energy of the solar street lamp according to the third battery data, and based on the first generable electric energy and the second storable electric energy, obtains the third available electric energy of the solar street lamp; the available electric energy provided by the solar street lamp provides a data basis for improving the power supply balance of highway solar street lamps, thereby effectively avoiding the problem of insufficient power supply of highway solar street lamps.

[0035] 2. The present invention divides the solar street lamps on the highway into regions, calculates the second total available electric energy of each region according to the third available electric energy, and obtains the first total required electric energy of each region according to the fourth street lamp working data; according to the first total required electric energy and the second total available electric energy, obtains the first electric energy supply-demand difference matrix of each region; this matrix clearly represents the electric energy supply-demand difference values of each region, provides a good data basis for the later power supply balance of highway solar street lamps, effectively promotes the power supply balance of each region on the highway, and thereby effectively avoids the problem of insufficient power supply of highway solar street lamps.

[0036] 3. The present invention introduces a DQN network model, inputs the first electric energy supply-demand difference matrix into the DQN network model, combines the constraint conditions of regional electric energy distribution, and the DQN network model dynamically selects an electric energy scheduling strategy and adaptively updates the Q function according to environmental feedback to achieve the adaptive distribution of electric energy in each region, which effectively promotes the power supply balance of each region on the highway, and thereby effectively avoids the problem of insufficient power supply of highway solar street lamps. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of a centralized control system for a solar street lamp provided by an embodiment of the present invention;

[0038] Figure 2 It is a schematic diagram of the acquisition process of a first electric energy supply-demand difference matrix provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] Embodiment 1

[0041] Highway A has applied a centralized control system for solar street lights to effectively avoid the problem of insufficient power supply for solar street lights;

[0042] Refer to Figure 1 , which is a schematic diagram of a centralized control system for solar street lights;

[0043] A centralized control system for solar street lights, the system includes:

[0044] A data acquisition module, which is used to collect the first environmental data, the second solar panel data, the third battery data, and the fourth street light working data of Highway A within a preset monitoring period in real time;

[0045] In this embodiment, the rule for determining the preset monitoring period is: if it is spring and autumn, it is set from 6:00 am to 6:00 pm in a day; if it is summer, it is set from 5:00 am to 7:30 pm, and if it is winter, it is set from 7:30 am to 5:00 pm;

[0046] Furthermore, the first environmental data includes light intensity and environmental temperature; the second solar panel data includes solar panel area, solar panel conversion efficiency, and solar panel tilt angle; the third battery data includes battery rated capacity, battery discharge depth, and charge and discharge efficiency; the fourth street light working data includes the working duration and working power of the solar street light;

[0047] The solar panel area is obtained by measuring the surface area of the solar panel; the solar panel conversion efficiency is represented according to the data provided by the solar panel manufacturer; the solar panel tilt angle is obtained by measuring with an inclinometer; the battery rated capacity is represented according to the fixed parameters provided by the battery manufacturer; the battery discharge depth and charge and discharge efficiency are monitored and obtained through a battery management system (BMS); the working duration of the solar street light is represented by the total remaining time outside the preset monitoring period in a day; the working power is represented according to the data provided by the solar panel manufacturer;

[0048] The light intensity is obtained through a light sensor; the environmental temperature is obtained through a temperature sensor; the

[0049] A power supply analysis module, which is used to obtain the first generable electric energy of the solar street light according to the first environmental data and the second solar panel data; obtain the second storable electric energy of the solar street light according to the third battery data, and obtain the third power supply electric energy of the solar street light according to the first generable electric energy and the second storable electric energy;

[0050] Furthermore, the first generable electric energy is:

[0051]

[0052] Among them, E i produce represents the first available electric energy of the i-th solar street lamp; t_begin represents the start time of the preset monitoring period; t_end represents the end time of the preset monitoring period; S i represents the area of the solar panel of the i-th solar street lamp; I i (t) represents the illumination intensity of the i-th solar street lamp at time t; η i (t) represents the conversion efficiency of the solar panel of the i-th solar street lamp at time t; θ i represents the tilt angle of the solar panel of the i-th solar street lamp.

[0053] Furthermore, the second storable electric energy is:

[0054] E i store = E i rated * C i charge * DoD i ;

[0055] Among them, E i store represents the second storable electric energy of the i-th solar street lamp; E i rated represents the rated capacity of the i-th solar street lamp; C i charge represents the charge-discharge efficiency of the i-th solar street lamp; DoD i represents the depth of discharge of the battery of the i-th solar street lamp.

[0056] Furthermore, the third available electric energy is:

[0057] E i available = Min(E i produce , E i store );

[0058] Among them, E i available represents the third available electric energy of the i-th solar street lamp; E i produce represents the first available electric energy of the i-th solar street lamp; E i store represents the second storable electric energy of the i-th solar street lamp.

[0059] In this embodiment, the first environmental data, the second solar panel data, the third battery data, and the fourth street lamp working data of the highway are collected in real time within a preset monitoring period; and based on the first environmental data and the second solar panel data, the first generable electric energy of the solar street lamp is obtained; the second storable electric energy of the solar street lamp is obtained according to the third battery data, and based on the first generable electric energy and the second storable electric energy, the third available electric energy of the solar street lamp is obtained; the available electric energy provided by the solar street lamp provides a data basis for improving the power supply balance of the highway solar street lamp, thereby effectively avoiding the problem of insufficient power supply of the highway solar street lamp.

[0060] The supply-demand difference analysis module is used to divide the solar street lamps on Highway A. The division method is as follows: on both sides of Highway A, every 5 solar street lamps are used as a division unit, and the street lamps on both sides are combined into one area. Specifically, every time 5 consecutive solar street lamps on the left side and 5 consecutive solar street lamps on the right side of Highway A form a complete area. Therefore, each area contains a total of 10 solar street lamps, that is, 5 on the left side and 5 on the right side. And calculate the second total available electric energy of each area according to the third available electric energy, and obtain the first total demand electric energy of each area according to the fourth street lamp working data; obtain the first electric energy supply-demand difference matrix of each area according to the first total demand electric energy and the second total available electric energy;

[0061] Further, the second total available electric energy is obtained by summing the third available electric energy of all solar street lamps in each area; the first total demand electric energy is obtained by multiplying the working hours, working power, and number of solar street lamps in each area.

[0062] Further, the first electric energy supply-demand difference matrix is obtained by taking the difference between the second total available electric energy and the first total demand electric energy of each area and constructing a matrix; the first electric energy supply-demand difference matrix is specifically:

[0063]

[0064] Among them, ESD represents the first electric energy supply-demand difference matrix; ESD j represents the electric energy supply-demand difference value of the jth area; E j total_available represents the second total available electric energy of the jth area; E j total_demand represents the first total demand electric energy of the jth area, and j represents the number of areas.

[0065] To sum up, the schematic diagram of the acquisition process of the first electric energy supply-demand difference matrix is specifically as Figure 2 shown;

[0066] In the first power supply-demand difference matrix, if an element is positive, it indicates that there is a surplus of electric energy in that area; if an element is negative, it indicates that there is a shortage of power supply in that area.

[0067] In this embodiment, the solar street lights on the highway are divided into regions, the second total available power of each region is calculated according to the third available power, and the first total required power of each region is obtained according to the fourth street light operation data; according to the first total required power and the second total available power, the first power supply-demand difference matrix of each region is obtained; this matrix clearly represents the power supply-demand difference values of each region, providing a good data basis for the later power supply balance of the highway solar street lights, effectively promoting the power supply balance of each region on the highway, and thus effectively avoiding the problem of insufficient power supply for the highway solar street lights.

[0068] The supply-demand balance control module uses the DQN network model to input the first power supply-demand difference matrix into the model and combines the constraints of regional power distribution; the model dynamically selects a power scheduling strategy and adaptively updates the Q function according to the environmental feedback to achieve the adaptive distribution of power in each region.

[0069] Further, the DQN network model includes a state space, an action space, a reward function, and a Q function; the state space s t is represented by the first power supply-demand difference matrix; the action space is defined as the set of all possible inter-regional power distribution schemes. For example, the operation of allocating a certain amount of power from region i to region j is an action in the action space; the Q function reflects the long-term benefit after executing action a t under state s t .

[0070] The constraint condition for regional power distribution is that when performing inter-regional power allocation, the total amount of power available for allocation cannot exceed the sum of all positive elements in the first supply-demand difference matrix.

[0071] Further, the reward function is:

[0072]

[0073] where R t represents the reward function; ESD i represents the power supply-demand difference value in the i-th region; j represents the number of regions.

[0074] Further, the updated Q function is:

[0075] In this embodiment, by introducing a reward function, it is possible to achieve guiding the optimal allocation of electric energy. The centralized control system of solar street lights can provide immediate feedback based on the current result of electric energy allocation, enabling the system to evaluate whether a certain scheduling strategy effectively improves the situation of insufficient regional electric energy. By continuously adjusting and optimizing the electric energy scheduling strategy, the reward function guides the system to gradually achieve the dynamic balance of electric energy supply and demand, thereby reducing the areas with insufficient power supply and making the operation of the entire centralized control system of solar street lights more efficient.

[0076] Q new (s t ,a t ) = Q(s t ,a t ) + α * [R t + γ * max a' Q(s t+1 ,a') - Q(s t ,a t )];

[0077] Among them, Q new (s t ,a t ) represents the updated Q function; α represents the learning rate; Q(s t ,a t ) represents the Q function before update; R t represents the reward function; γ represents the discount factor; max a' Q(s t+1 ,a') represents the Q value after executing the optimal action a'.

[0078] In this embodiment, by introducing a DQN network model, the first electric energy supply-demand difference matrix is input into the DQN network model. Combining with the constraint conditions of regional electric energy allocation, the DQN network model dynamically selects an electric energy scheduling strategy and adaptively updates the Q function according to the environmental feedback. The Q function will be adaptively updated according to the real-time feedback of the system, which enables the system to maintain efficient scheduling under the changing electric energy supply and demand situations, thereby achieving the adaptive allocation of electric energy in each region. This effectively promotes the power supply balance in each region on the highway, thus effectively avoiding the problem of insufficient power supply for highway solar street lights.

[0079] In this embodiment, the data acquisition module collects the first environmental data, the second solar panel data, the third battery data, and the fourth street lamp working data on the highway in real time; the available power analysis module calculates the first available power based on the first environmental data and the second solar panel data, calculates the second storable power based on the third battery data, and finally obtains the third available power; the supply-demand difference analysis module generates the first power supply-demand difference matrix by calculating the total available power and the total demand power in each area; the supply-demand balance control module dynamically selects the power scheduling strategy using the DQN network model and adaptively updates the Q function according to the environmental feedback to achieve the adaptive distribution of power in each area. Through this system, the power supply-demand balance in each area can be effectively achieved, and the problem of insufficient power supply for solar street lamps can be effectively avoided.

[0080] Embodiment 2

[0081] In order to effectively avoid the problem of insufficient power supply for solar street lamps, Highway B applies a centralized control system for solar street lamps;

[0082] Refer to Figure 1 , which is a schematic diagram of a centralized control system for solar street lamps;

[0083] A centralized control system for solar street lamps, the system includes:

[0084] A data acquisition module, configured to collect the first environmental data, the second solar panel data, the third battery data, and the fourth street lamp working data on Highway B within a preset monitoring period in real time;

[0085] In this embodiment, the rule for determining the preset monitoring period is: if it is spring and autumn, it is set from 6:00 am to 6:00 pm in a day; if it is summer, it is set from 5:00 am to 7:30 pm, and if it is winter, it is set from 7:30 am to 5:00 pm;

[0086] Furthermore, the first environmental data includes light intensity and environmental temperature; the second solar panel data includes solar panel area, solar panel conversion efficiency, and solar panel tilt angle; the third battery data includes battery rated capacity, battery discharge depth, and charge-discharge efficiency; the fourth street lamp working data includes the working duration and working power of the solar street lamp;

[0087] The area of the solar panel is obtained by measuring the surface area of the solar panel; the conversion efficiency of the solar panel is represented according to the data provided by the solar panel manufacturer; the tilt angle of the solar panel is obtained by measuring with an inclinometer; the rated capacity of the storage battery is represented according to the fixed parameters provided by the storage battery manufacturer; the depth of discharge and charge-discharge efficiency of the storage battery are monitored and obtained by the battery management system (BMS); the working duration of the solar street lamp is represented by the remaining total time outside the preset monitoring period in a day; the working power is represented according to the data provided by the solar panel manufacturer;

[0088] The light intensity is obtained by a light sensor; the ambient temperature is obtained by a temperature sensor; the

[0089] The available power analysis module is configured to obtain the first available electric energy of the solar street lamp according to the first environmental data and the second solar panel data; obtain the second storable electric energy of the solar street lamp according to the third storage battery data, and obtain the third available electric energy of the solar street lamp according to the first available electric energy and the second storable electric energy;

[0090] Further, the first available electric energy is:

[0091]

[0092] where, E i produce represents the first available electric energy of the i-th solar street lamp; t_begin represents the start time of the preset monitoring period; t_end represents the end time of the preset monitoring period; S i represents the area of the solar panel of the i-th solar street lamp; I i (t) represents the light intensity of the i-th solar street lamp at time t; η i (t) represents the conversion efficiency of the solar panel of the i-th solar street lamp at time t; θ i represents the tilt angle of the solar panel of the i-th solar street lamp.

[0093] Further, the second storable electric energy is:

[0094] E i store =E i rated *C i charge *DoD i ;

[0095] where, E i store represents the second storable electric energy of the i-th solar street lamp; Ei rated represents the rated capacity of the i-th solar street lamp; C i charge represents the charge-discharge efficiency of the i-th solar street lamp; DoD i represents the depth of discharge of the battery of the i-th solar street lamp.

[0096] Furthermore, the third available electric energy is:

[0097] E i available = Min(E i produce , E i store );

[0098] wherein, E i available represents the third available electric energy of the i-th solar street lamp; E i produce represents the first available generated electric energy of the i-th solar street lamp; E i store represents the second storable electric energy of the i-th solar street lamp.

[0099] Supply-demand difference analysis module, which is used to divide the solar street lamps on Highway B into regions. The division method is: on both the left and right sides of Highway B, every 5 solar street lamps are used as a division unit, and the street lamps on both sides are combined into one region. Specifically, whenever 5 consecutive solar street lamps on the left side of Highway B and 5 consecutive solar street lamps on the right side form a complete region. Therefore, each region contains a total of 10 solar street lamps, that is, 5 on each of the left and right sides. And calculate the second total available electric energy of each region according to the third available electric energy, and obtain the first total demand electric energy of each region according to the fourth street lamp working data; according to the first total demand electric energy and the second total available electric energy, obtain the first electric energy supply-demand difference matrix of each region;

[0100] Furthermore, the second total available electric energy is obtained by summing up the third available electric energy of all solar street lamps in each region; the first total demand electric energy is obtained by multiplying the working hours, working power and number of solar street lamps in each region;

[0101] Furthermore, the first electric energy supply-demand difference matrix is obtained by taking the difference between the second total available electric energy and the first total demand electric energy of each region and constructing a matrix; the first electric energy supply-demand difference matrix is specifically:

[0102]

[0103] Among them, ESD represents the first electric energy supply-demand difference matrix; ESD j represents the electric energy supply-demand difference value of the j-th region; E j total_available represents the second total available electric energy of the j-th region; E j total_demand represents the first total demanded electric energy of the j-th region, and j represents the number of regions.

[0104] In summary, the schematic diagram of the acquisition process of the first electric energy supply-demand difference matrix is specifically as Figure 2 shown;

[0105] In the first electric energy supply-demand difference matrix described above, if the element is positive, it means there is an electric energy surplus in this region; if the element is negative, it means there is a power supply shortage in this region;

[0106] The supply-demand balance control module uses the DQN network model to input the first electric energy supply-demand difference matrix into the model and combines the constraint conditions of regional electric energy distribution; the model realizes the adaptive distribution of electric energy in each region by dynamically selecting electric energy scheduling strategies and adaptively updating the Q function according to environmental feedback.

[0107] Furthermore, the DQN network model includes a state space, an action space, a reward function, and a Q function; the state space s t is represented by the first electric energy supply-demand difference matrix; the action space is defined as the set of all possible inter-regional electric energy distribution schemes. For example, the operation of allocating a certain amount of electric energy from region i to region j is an action in the action space; the Q function reflects the long-term benefit after executing the action a t under the state s t .

[0108] The constraint condition for regional electric energy distribution is that when performing inter-regional electric energy allocation, the total amount of electric energy available for allocation cannot exceed the sum of all positive elements in the first supply-demand difference matrix;

[0109] Furthermore, the reward function is:

[0110]

[0111] where, R t represents the reward function; ESD i represents the electric energy supply-demand difference value in the i-th region; j represents the number of regions.

[0112] Furthermore, the updated Q function is:

[0113] In this embodiment, by introducing a reward function, the guided optimal allocation of electric energy can be achieved. The centralized control system of solar street lights can provide immediate feedback based on the current result of electric energy allocation, enabling the system to evaluate whether a certain scheduling strategy effectively improves the regional power shortage. By continuously adjusting and optimizing the electric energy scheduling strategy, the reward function guides the system to gradually achieve the dynamic balance of power supply and demand, thereby reducing the areas with power supply shortages and making the entire centralized control system of solar street lights operate more efficiently.

[0114] Q new (s t ,a t )=Q(s t ,a t )+α*[R t +γ*max a' Q(s t+1 ,a')-Q(s t ,a t )];

[0115] Among them, Q new (s t ,a t ) represents the updated Q function; α represents the learning rate; Q(s t ,a t ) represents the Q function before update; R t represents the reward function; γ represents the discount factor; max a' Q(s t+1 ,a') represents the Q value after performing the optimal action a'.

[0116] In this embodiment, by introducing a DQN network model, the first power supply-demand difference matrix is input into the DQN network model. Combining with the constraint conditions of regional power allocation, the DQN network model dynamically selects a power scheduling strategy and adaptively updates the Q function according to environmental feedback. The Q function will be adaptively updated according to the real-time feedback of the system, which enables the system to maintain efficient scheduling under changing power supply and demand conditions, thereby achieving the adaptive allocation of power in each region, effectively promoting the power supply balance in each region on the highway, and effectively avoiding the problem of insufficient power supply for highway solar street lights.

[0117] In this embodiment, the data acquisition module collects the first environmental data, the second solar panel data, the third battery data, and the fourth street lamp working data on the highway in real time; the available power analysis module calculates the first available power based on the first environmental data and the second solar panel data, calculates the second storable power according to the third battery data, and finally obtains the third available power; the supply-demand difference analysis module generates the first power supply-demand difference matrix by calculating the total available power and the total demand power in each region; the supply-demand balance control module dynamically selects the power scheduling strategy using the DQN network model and adaptively updates the Q function according to the environmental feedback to achieve the adaptive allocation of power in each region. Through this system, the power supply-demand balance in each region can be effectively achieved, and the problem of insufficient power supply for solar street lamps can be effectively avoided.

[0118] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A centralized control system for solar street lights, characterized in that, The system includes: A data acquisition module, configured to collect in real time first environmental data, second solar panel data, third battery data, and fourth street lamp working data of a highway within a preset monitoring period; A power generation analysis module, configured to obtain a first generatable power of a solar street lamp according to the first environmental data and the second solar panel data; the first generatable power is: Among them, E i produce represents the first generated electric energy of the i-th solar street lamp; t_begin represents the start time of the preset monitoring period; t_end represents the end time of the preset monitoring period; S i represents the area of the solar panel of the i-th solar street lamp; I i (t) represents the light intensity of the i-th solar street lamp at time t; η i (t) represents the conversion efficiency of the solar panel of the i-th solar street lamp at time t; θ i represents the tilt angle of the solar panel of the i-th solar street lamp; Obtain a second storable power of the solar street lamp according to the third battery data, and obtain a third power supplyable power of the solar street lamp according to the first generatable power and the second storable power; A supply-demand difference analysis module, configured to divide regions of the solar street lamps on the highway, calculate a second total supplyable power of each region according to the third power supplyable power, and obtain a first total demand power of each region according to the fourth street lamp working data; obtain a first power supply-demand difference matrix of each region according to the first total demand power and the second total supplyable power; the first power supply-demand difference matrix is obtained by taking the difference between the second total supplyable power and the first total demand power of each region and constructing a matrix; the first power supply-demand difference matrix is specifically: Among them, ESD represents the first electric energy supply-demand difference matrix; ESD j represents the electric energy supply-demand difference value of the j-th region; E j total_available represents the second total available electric energy of the j-th region; E j total_demand represents the first total demand electric energy of the j-th region, and j represents the number of regions; A supply-demand balance control module, using a DQN network model, inputting the first power supply-demand difference matrix into the model, and combining the constraint conditions of regional power distribution; the model adaptively distributes the power of each region by dynamically selecting a power scheduling strategy and adaptively updating the Q function according to environmental feedback; the updated Q function is: Q new (s t ,a t ) = Q(s t ,a t ) + α * [R t + γ * max a' Q(s t+1 , a') - Q(s t ,a t )]; Among them, Q new (s t , a t ) represents the updated Q function; α represents the learning rate; Q(s t , a t ) represents the Q function before update; R t represents the reward function; γ represents the discount factor; max a' Q(s t+1 , a') represents the Q value after executing the optimal action a'.

2. The centralized control system of a solar street lamp according to claim 1, characterized in that, The first environmental data includes light intensity and environmental temperature; the second solar panel data includes solar panel area, solar panel conversion efficiency, and solar panel tilt angle; the third battery data includes battery rated capacity, battery discharge depth, and charge-discharge efficiency; the fourth street lamp working data includes the working duration and working power of the solar street lamp.

3. The centralized control system of a solar street lamp according to claim 1, characterized in that The second storable power is: E i store = E i rated * C i charge * DoD i ; Among them, E i store represents the second storable electrical energy of the i-th solar street lamp; E i rated represents the rated capacity of the i-th solar street lamp; C i charge represents the charge-discharge efficiency of the i-th solar street lamp; DoD i represents the depth of discharge of the battery of the i-th solar street lamp.

4. The centralized control system of a solar street lamp according to claim 1, characterized in that, The third power supplyable power is: E i available = Min(E i produce , E i store ); Among them, E i available represents the third available power of the i-th solar street lamp; E i produce represents the first generable power of the i-th solar street lamp; E i store represents the second storable power of the i-th solar street lamp.

5. The centralized control system of a solar street lamp according to claim 1, characterized in that, The second total supplyable power is obtained by summing the third power supplyable powers of all solar street lamps in each region; the first total demand power is obtained by multiplying the working duration, working power, and number of solar street lamps in each region.

6. The centralized control system of a solar street lamp according to claim 1, characterized in that, The DQN network model includes a state space, an action space, a reward function, and a Q function; the state space s t is represented by the first electric energy supply-demand difference matrix; the action space represents the electric energy distribution process between regions; the Q function reflects the long-term benefit after executing the action a t under the state s t ​ 7. The centralized control system of a solar street lamp according to claim 6, characterized in that, The reward function is: Among them, R t represents the reward function; ESD i represents the power supply-demand difference value in the i-th region; j represents the number of regions.

Images