A charging system for an all-round solar LED street lamp

Abstract

The present invention relates to the technical field of LED lighting, and specifically to a charging system for an all-round solar LED street lamp, including: a charge-discharge threshold determination module obtains the current battery life cycle and the current battery health status of the all-round solar LED street lamp, and accordingly determines the charging upper limit and the discharging lower limit. A historical data acquisition module collects the historical lighting, energy consumption, environment, weather, light, and power generation data of the street lamp, and generates the historical data of the solar LED street lamp. An energy consumption and power generation prediction module predicts the energy consumption demand and the solar power generation amount of the day based on the historical data of the solar LED street lamp. A charging module obtains battery parameters during the charging period and formulates a charging strategy, and uses this strategy to charge the battery and the backup energy storage device. A discharging module obtains battery parameters during the discharging period, formulates a discharging strategy, and supplies power to the street lamp through the battery and the backup energy storage device according to this strategy. The system realizes the intelligent management of the charging and discharging of the solar LED street lamp, which helps to extend the battery life.

Description

Technical Field

[0001] The present invention relates to the technical field of LED lighting, and particularly to a charging system for an all-round solar LED street lamp. Background Art

[0002] With the development of modern cities, the construction of smart cities has gradually become the core of urban planning. As one of the infrastructure of smart cities, street lamps play a key role in efficient lighting, intelligent management, and green energy application. To reduce energy consumption and improve management efficiency, more and more cities choose to install solar LED street lamps. Solar LED street lamps are an environmentally friendly and energy-saving outdoor lighting device, widely used in roads, squares, highways and other places. These street lamps convert solar energy into electrical energy through solar panels and store it in batteries for night lighting.

[0003] The illumination angle of the solar panel of traditional solar street lamps is often fixed, which limits the utilization rate of illumination resources. Especially in urban environments with large changes in sunlight angles or more shadow effects from buildings, trees and other obstacles, it is impossible to effectively obtain solar energy. While all-round designed solar LED street lamps introduce adjustable solar panels or multi-directional solar panels, enabling them to receive sunlight from different angles, which not only improves the collection efficiency of light energy but also provides more stable electrical energy under unstable sunlight conditions.

[0004] In the context of smart cities, although some intelligent charging technologies have been introduced into the existing charge and discharge systems, in actual applications, they still cannot be accurately adjusted according to the real-time state of the battery and energy consumption requirements. The dynamic adjustment strategies of existing systems are relatively simple, lacking overall planning and coordination of the charging and discharging processes, resulting in a control split between the two, and failing to consider the mutual dependence of battery charging and discharging of solar LED street lamps, leading to frequent overcharging and over-discharging phenomena in battery management. Especially in situations where weather conditions are unstable, sunlight is insufficient, or nighttime electricity demand fluctuates greatly, the system is difficult to flexibly respond to dynamic changes in the environment and demand. This not only affects the overall energy efficiency but also accelerates battery aging and shortens the service life of the battery.

[0005] Therefore, a charging system for an all-round solar LED street lamp is proposed. Summary of the Invention

[0006] The object of the present invention is to provide a charging system for an all-round solar LED street lamp, which includes a charge and discharge threshold determination module, a historical data acquisition module, an energy consumption and power generation prediction module, a charging module, and a discharging module. The charge and discharge threshold determination module is used to obtain the current battery life cycle and the current battery health status of the solar LED street lamp, and determine the charging upper limit and the discharging lower limit based on this information. The historical data acquisition module is responsible for obtaining the historical street lamp lighting data, historical street lamp energy consumption data, historical environmental data, historical weather data, historical light data, and historical solar panel power generation data of the street lamp, so as to generate the historical data of the solar LED street lamp. The energy consumption and power generation prediction module predicts the energy consumption demand during the discharging period of the day and the solar power generation capacity during the charging period of the day according to the historical data, and obtains the predicted consumption and the predicted power generation. The charging module obtains the current parameters of the battery during the charging period, obtains the first battery parameter set, and formulates a charging strategy based on this parameter set, the predicted power generation, the predicted consumption, and the charging upper limit, and then charges the battery and the standby energy storage device. The discharging module obtains the current parameters of the battery during the discharging period, obtains the second battery parameter set, and formulates a discharging strategy based on this parameter set, the predicted consumption, and the discharging lower limit, and finally supplies power to the solar LED street lamp through the battery and the standby energy storage device. The present invention introduces a bidirectional energy flow management module to realize the bidirectional energy flow between the battery and the standby energy storage device, so that the system can flexibly adjust the charge and discharge strategies according to the real-time battery parameters, the predicted power generation, and the energy consumption demand, thereby ensuring the safety and efficiency of the charge and discharge process. By means of the charge and discharge threshold determination method according to the battery life cycle and health status, the present invention can dynamically adjust the charging upper limit and the discharging lower limit at different life cycle stages of the battery. This method not only effectively avoids overcharging and over-discharging of the battery, ensures the safety and stability during the charging process, but also greatly extends the overall service life of the battery, improves the energy management efficiency and sustainability of the system. By introducing a deep reinforcement learning algorithm, the charging and discharging strategies of the all-round solar LED street lamp are dynamically generated, realizing more intelligent and accurate energy management. The system can flexibly adjust the charge and discharge strategies according to the real-time state of the battery and the predicted power generation and consumption, effectively avoiding phenomena such as overcharging, over-discharging, and battery overheating, and ensuring the safety and service life of the battery.

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

[0008] A charging system for an all-round solar LED street lamp, comprising:

[0009] A charge and discharge threshold determination module, configured to obtain the current battery life cycle and the current battery health status of the solar LED street lamp; determine a charging upper limit and a discharging lower limit according to the current battery life cycle and the current battery health status;

[0010] A historical data acquisition module, which is used to acquire the historical street lamp lighting data, historical street lamp energy consumption data, historical environmental data, historical weather data, historical light data, and historical solar panel power generation data of the solar LED street lamp, so as to obtain the historical data of the solar LED street lamp;

[0011] An energy consumption and power generation prediction module, which is used to predict the energy consumption demand during the discharge period of the day and the solar power generation capacity during the charging period of the day according to the historical data of the solar LED street lamp, so as to obtain the predicted consumption and the predicted power generation;

[0012] A charging module, which is used to acquire the current parameters of the battery during the charging period to obtain a first set of battery parameters, and obtain a charging strategy according to the first set of battery parameters, the predicted power generation, the predicted consumption, and the charging upper limit; use the charging strategy to charge the battery and the backup energy storage device;

[0013] A discharge module, which is used to acquire the current parameters of the battery during the discharge period to obtain a second set of battery parameters, and obtain a discharge strategy according to the second set of battery parameters, the predicted consumption, and the discharge lower limit; supply power to the solar LED street lamp through the battery and the backup energy storage device according to the discharge strategy.

[0014] Furthermore, the charging system further includes: an omnidirectional solar energy collection module, which is used to collect sunlight at different angles and convert the sunlight into electric energy; it includes a plurality of solar panels with adjustable angles.

[0015] Furthermore, the charging system further includes an early warning module, and the early warning module includes:

[0016] An abnormality detection unit, which is used to generate a solar panel warning signal when it detects that the solar panel fails; generate a life cycle warning signal when it detects that the current battery life cycle is in the late stage; generate a battery status warning signal when the current battery health status is lower than a preset threshold;

[0017] An information sending unit, which is used to send the solar panel warning signal, the life cycle warning signal, and the battery status warning signal to a remote monitoring center through a communication network to obtain warning data;

[0018] A data storage unit, which is used to store the warning data in a specified format;

[0019] An early warning processing unit, which is used to analyze the warning data and generate early warning maintenance suggestions.

[0020] Further, the charging system further includes a bidirectional energy flow management module, including: managing the bidirectional energy flow between the battery and the backup energy storage device; obtaining the remaining battery capacity of the first battery and the remaining battery capacity of the second battery according to the first battery parameter set and the second battery parameter set respectively; calculating the rechargeable amount according to the remaining battery capacity of the first battery and the charging upper limit; if the rechargeable amount exceeds the predicted power generation amount, storing the electric energy generated by the omnidirectional solar energy collection module into the battery; if the rechargeable amount does not exceed the predicted power generation amount, storing the electric energy generated by the omnidirectional solar energy collection module into the battery, and when the voltage of the battery reaches the charging upper limit, storing the electric energy generated by the omnidirectional solar energy collection module into the backup energy storage device; calculating the available power according to the remaining battery capacity of the first battery and the discharging lower limit; if the available power exceeds the predicted consumption amount, using the battery to supply power to the omnidirectional solar LED street lamp; if the available power does not exceed the predicted consumption amount, using the battery to supply power to the omnidirectional solar LED street lamp; when the voltage of the battery reaches the discharging lower limit, using the backup energy storage device to supply power to the omnidirectional solar LED street lamp.

[0021] Further, determining the charging upper limit and the discharging lower limit according to the current battery life cycle and the current battery health state includes:

[0022] Obtaining the historical usage time, charge and discharge times, and rated service life of the battery; determining the current battery life cycle of the battery according to the historical usage time, charge and discharge times, and rated service life of the battery;

[0023] Obtaining the charge and discharge data of multiple historical omnidirectional solar LED street lamps with completed life cycles from the historical charge and discharge database to obtain historical reference data; the historical reference data covers the full life cycle charge and discharge records of the batteries of the historical omnidirectional solar LED street lamps from initial use to end of life, and the full life cycle covers all life cycle stages, including the initial stage, the middle stage, and the late stage; selecting the historical reference data with the charge and discharge times greater than a preset threshold to obtain the charge and discharge reference data of the long-life batteries;

[0024] Preprocessing the charge and discharge reference data of the long-life batteries to obtain the charge and discharge reference data of the standard long-life batteries; the charge and discharge reference data of the standard long-life batteries includes the battery health state during the charging process and the discharging process;

[0025] Extracting the current charge and discharge times and the current usage time of each omnidirectional solar LED street lamp battery in the charge and discharge reference data of the standard long-life batteries; determining the life cycle according to the current charge and discharge times and the current usage time;

[0026] Divide the reference charge-discharge data of the longevity standard battery into N groups according to the described life cycle and the battery health state, obtaining N groups of battery state data. The battery state data includes battery charging process data and battery discharging process data; the battery charging process data includes the charging start voltage, the charging end voltage, the charging current, the charging duration, and the charging temperature change; the battery discharging process data includes the discharging start voltage, the discharging end voltage, the discharging current, the discharging duration, and the discharging temperature change.

[0027] For each group of the battery state data, extract the maximum value of the charging end voltage in the battery charging process data and the minimum value of the discharging end voltage in the battery discharging process data, obtaining the maximum charging voltage and the minimum discharging voltage.

[0028] According to the current battery life cycle and the current battery health state, screen out the groups in the N groups of battery state data where the battery health state is the same as the current battery health state and the battery life cycle is the same as the current battery life cycle, obtaining the battery state data in the same stage; obtain the maximum charging voltage and the minimum discharging voltage of the battery state data in the same stage, obtaining the charging upper limit and the discharging lower limit.

[0029] Furthermore, the energy consumption and power generation prediction module includes a street lamp energy consumption prediction model. The training process of the street lamp energy consumption prediction model includes the following steps:

[0030] Preprocess the historical data of the solar LED street lamp to obtain the standard historical data of the solar LED street lamp; the standard historical data of the solar LED street lamp includes historical standard street lighting data, historical standard street lamp energy consumption data, historical standard environmental data, and historical standard weather data; the historical standard environmental data includes image data and sound data of a specified lighting area.

[0031] Use feature engineering techniques to extract features from the standard historical data, obtaining a feature set; the feature set includes time features, lighting intensity features, energy consumption features, street lamp control features, battery features, weather features, regional environment features, and traffic flow features; the traffic flow features include the number of pedestrians and the number of vehicles extracted according to the image data and the sound data; the regional environment features include the number of faults and the fault distance of surrounding street lamps in a specified area; construct interaction features according to the feature set, and integrate the interaction features with the feature set into a first fusion feature set; the interaction features include: time and traffic flow interaction features, weather and battery interaction features, regional environment and lighting intensity interaction features, and street lamp control and energy consumption interaction features.

[0032] Taking the first fusion feature set as the input and the historical street lamp energy consumption data in the historical data of the solar LED street lamp as the label, train an XGBoost model, and obtain the street lamp energy consumption prediction model through hyperparameter optimization.

[0033] Further, according to the historical data of the solar LED street lamp, predict the solar power generation capacity during the charging period of the day to obtain the predicted power generation amount specifically as follows: preprocess the historical environmental data, the historical weather data, and the historical solar panel power generation data in the historical data of the solar LED street lamp to obtain historical standard environmental data, historical standard weather data, and historical standard solar panel power generation data; obtain historical power generation amount data from the historical standard solar panel power generation data; use feature engineering technology to extract features from the historical standard environmental data, historical standard weather data, historical standard illumination data, and historical standard solar panel power generation data and construct interaction features to obtain a second fusion feature set; take the second fusion feature set as the input and the historical power generation amount data as the label, and train a solar panel power generation prediction model according to the XGBoost model; organize the environmental data and weather forecast data of the day into the feature format during training, and gradually input them into the trained solar panel power generation prediction model according to the time window to predict the power generation amount in each time window; accumulate the power generation amounts of all the time windows to obtain the predicted power generation amount.

[0034] Further, generate the charging strategy according to the deep reinforcement learning algorithm, and the generation process includes:

[0035] Construct a state set according to the first battery parameter set; the first battery parameter set includes the voltage, current, temperature, state of charge, and health state of the battery during the charging period.

[0036] Define an action set, including adjusting the charging current of the battery, selecting a charging mode, and switching to the backup energy storage device for charging.

[0037] Through the state transition function, calculate the state transition process under different actions according to the changes in the current state set and the action set to obtain the next moment state under each action; the next moment state includes battery voltage, battery current, battery temperature, battery state of charge, battery health state, and backup energy storage device state.

[0038] The charging module evaluates the action according to the predicted power generation amount, the charging upper limit, and the result of state transition, defines rewards according to the energy utilization efficiency, battery voltage penalty, battery overheating penalty, battery health state deterioration penalty, and battery overcharging penalty to obtain a reward function.

[0039] The charging module generates the charging strategy according to the state set, the action set, the state transition function, and the reward function through the deep reinforcement learning algorithm.

[0040] Further, the process of generating the charging strategy includes:

[0041] Using a policy network to select continuous actions according to the current state; according to the reward function, using a Q-network to evaluate the long-term cumulative reward of the continuous actions in the current state; through an experience replay mechanism, the charging module repeatedly trains the policy network and the Q-network, and uses backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and the Q-network to maximize the long-term cumulative reward. When the preset convergence condition is met, the charging strategy is obtained.

[0042] Further, obtaining the discharge strategy according to the second battery parameter set, the predicted consumption, and the discharge lower limit includes:

[0043] Generating the discharge strategy according to the deep reinforcement learning algorithm, and the generation process includes:

[0044] Constructing a state set according to the second battery parameter set; the second battery parameter set includes the voltage, current, temperature, state of charge, and health state of the battery during the discharge period;

[0045] Defining an action set, including adjusting the charging current of the battery, selecting a discharge mode, and switching to the standby energy storage device for discharging;

[0046] Through the state transition function, according to the changes in the current state set and the action set, calculating the state transition process under different actions, and obtaining the next moment state under each action; the next moment state includes battery voltage, battery current, battery temperature, battery state of charge, battery health state, and standby energy storage device state;

[0047] The discharge module evaluates the actions according to the predicted consumption, the discharge lower limit, and the result of state transition, defines rewards according to energy utilization efficiency, battery voltage penalty, battery overheating penalty, battery health state deterioration penalty, and battery over-discharge penalty, and obtains a reward function;

[0048] The discharge module generates the discharge strategy according to the state set, the action set, the state transition function, and the reward function through the deep reinforcement learning algorithm;

[0049] The policy network is used to select continuous actions according to the current state; according to the reward function, the Q network is used to evaluate the long-term cumulative reward of the continuous actions in the current state; through the experience replay mechanism, the discharge module repeatedly trains the policy network and the Q network through the experience replay mechanism, and uses backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and the Q network to maximize the long-term cumulative reward. When the preset convergence condition is met, the discharge strategy is obtained.

[0050] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0051] 1. The all-round solar LED street lamp charging system proposed by the present invention realizes the bidirectional energy flow between the battery and the backup energy storage device through the bidirectional energy flow management module, enabling the system to flexibly adjust the charge and discharge strategies according to real-time battery parameters, predicted power generation, and predicted consumption, thereby ensuring the safety and efficiency of the charge and discharge process. When the battery is in a low power state, the system will intelligently control the charging process according to the actual consumption demand to avoid battery loss caused by overcharging; when the battery voltage reaches the charging upper limit, the system will transfer the excess electric energy to the backup energy storage device to prevent overcharging and ensure that the battery operates within a safe range. At the same time, during the discharge process, the system can real-time monitor the remaining power and health status of the battery, reasonably set the discharge lower limit, and avoid battery loss caused by deep discharge. Through this dynamic management mechanism, not only the charge and discharge safety of the battery is greatly improved, but also the battery life can be effectively extended, the risks of overuse and over-discharge of the battery are reduced, and the long-term reliability and stability of the system are improved.

[0052] 2. The present invention proposes a method for determining the charge and discharge threshold according to the battery life cycle and health status, which can dynamically adjust the charging upper limit and discharge lower limit at different life cycle stages of the battery. By obtaining the charge and discharge data covering the entire life cycle from the historical charge and discharge database and combining the historical usage time, charge and discharge times, and health status of the battery, the system can accurately identify the current battery life cycle stage of the battery. Selecting data with a life greater than the preset threshold provides more representative and reference-worthy charge and discharge behaviors, ensuring that the system can extract the most appropriate charging and discharging voltage ranges according to best practices during the battery management and optimization process. This method not only effectively avoids overcharging and over-discharging of the battery, ensures the safety and stability during the charging process, but also greatly extends the overall service life of the battery, improves the energy management efficiency and sustainability of the system. This charge and discharge strategy based on big data analysis and historical experience enables the system to flexibly adapt to different environmental conditions and achieve efficient and stable operation.

[0053] 3. By integrating the deep reinforcement learning algorithm, the present invention generates intelligent charging and discharging strategies respectively, achieving dynamic and precise management of the battery of the solar LED street lamp. During the charging process, the system can monitor the voltage, current, temperature, state of charge and health state of the battery in real time, and flexibly adjust the charging current and mode by combining the predicted power generation and the set charging upper limit, effectively improving the energy utilization efficiency, avoiding overcharging and energy waste, and extending the battery life. During the discharging process, the system also optimizes the discharging strategy according to the battery state, predicted consumption and discharging lower limit, reasonably distributes the discharging tasks of the battery and the backup energy storage device, avoids over-discharging, and ensures the stable operation of the system. The intelligent adjustment of the charging and discharging strategies greatly improves the overall energy efficiency of the street lamp system, extends the service life of the battery, enhances the safety and stability of the charging and discharging processes, and enables the system to flexibly respond to complex environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 FIG. is a structural diagram of a charging system of an all-round solar LED street lamp provided by an embodiment of the present invention;

[0055] Figure 2 FIG. is a complete structural diagram of a charging system of an all-round solar LED street lamp provided by an embodiment of the present invention;

[0056] Figure 3 FIG. is a decision framework diagram of a charging strategy based on deep reinforcement learning provided by an embodiment of the present invention;

[0057] Figure 4 FIG. is a decision framework diagram of a discharging strategy based on deep reinforcement learning provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the protection scope of the present invention.

[0059] Charging and discharging are two basic working processes of the battery. Among them, charging means that the all-round solar panel converts the collected solar energy into electrical energy and stores it in the storage battery. Discharging means that during the lighting period, the storage battery supplies the previously stored electrical energy to the LED street lamp for lighting. The specific solutions of the present invention will be introduced in detail below through two embodiments.

[0060] Embodiment 1

[0061] On the highway in City A, there are omnidirectional solar LED street lights of the same model. The rechargeable batteries of this batch of omnidirectional solar LED street lights are all 3.2V lithium iron phosphate batteries (storage batteries), and its backup energy storage device is the grid backup power supply system. This street light has an advanced light adaptive adjustment function, and it does not remain constantly lit throughout the night lighting period. It can flexibly adjust the brightness or on-off state of the light according to the real-time ambient light intensity and traffic conditions. Due to its flexible lighting method, it brings higher dynamic management requirements for charge and discharge control.

[0062] The omnidirectional solar LED street light B on the highway applies a charging system for omnidirectional solar LED street lights, as Figure 1 shown, including:

[0063] A charge and discharge threshold determination module, which is used to obtain the current battery life cycle and the current battery health status of the solar LED street light; determine the charging upper limit and the discharging lower limit according to the current battery life cycle and the current battery health status; specifically, obtain the current battery life cycle of the solar LED street light B - early stage, and obtain the current battery health status of the solar LED street light B - 95%; determine the charging upper limit and the discharging lower limit according to the current battery life cycle and the current battery health status;

[0064] The omnidirectional solar LED street light B includes a solar panel and a storage battery. The solar panel is responsible for obtaining energy from sunlight and converting light energy into electrical energy. The storage battery is responsible for storing the electrical energy generated by the solar panel and powering the LED street light at night or when sunlight is insufficient. Together, they constitute the core of the omnidirectional solar LED street light system. The batteries in this embodiment all refer to lithium iron phosphate batteries (rechargeable battery storage batteries).

[0065] A historical data acquisition module, which is used to obtain the historical street light lighting data, historical street light energy consumption data, historical environmental data, historical weather data, historical light data, and historical solar panel power generation data of the solar LED street light, and obtain the historical data of the solar LED street light;

[0066] An energy consumption and power generation prediction module, which is used to predict the energy consumption demand during the discharging period of the day and the solar power generation capacity during the charging period of the day according to the historical data of the solar LED street light, and obtain the predicted consumption and the predicted power generation; in this embodiment, this prediction is carried out before the start of the charging period of the day.

[0067] A charging module, which is used to obtain the current parameters of the battery during the charging period to obtain a first set of battery parameters, and obtain a charging strategy according to the first set of battery parameters, the predicted power generation, the predicted consumption, and the charging upper limit; use the charging strategy to charge the battery and the backup energy storage device; wherein, the charging period is obtained according to the sunrise time and sunset time of the location of the street lamp.

[0068] A discharging module, which is used to obtain the current parameters of the battery during the discharging period to obtain a second set of battery parameters, and obtain a discharging strategy according to the second set of battery parameters, the predicted consumption, and the discharging lower limit; supply power to the solar LED street lamp through the battery and the backup energy storage device according to the discharging strategy.

[0069] Further, the charging system further includes: an omnidirectional solar collection module, which is used to collect sunlight at different angles and convert the sunlight into electric energy; the omnidirectional solar collection module includes a plurality of solar panels with adjustable angles.

[0070] The omnidirectional solar collection module lays a foundation for obtaining more comprehensive historical solar panel power generation data subsequently. Since the solar panels with omnidirectional adjustable angles can be dynamically adjusted to the optimal angle to collect solar energy, the system can accumulate more dimensional and richer power generation data under different lighting conditions. Through these data, the system can more accurately predict the future power generation capacity, thereby further optimizing the charging strategy.

[0071] Further, the charging system further includes a warning module, and the warning module includes:

[0072] An abnormality detection unit, which is used to generate a solar panel warning signal when it detects that a solar panel fails; generate a life cycle warning signal when it detects that the current battery life cycle is in the late stage; generate a battery status warning signal when the current battery health status is lower than a preset threshold;

[0073] An information sending unit, which is used to send the solar panel warning signal, the life cycle warning signal, and the battery status warning signal to a remote monitoring center through a communication network to obtain warning data;

[0074] A data storage unit, which is used to store the warning data in a specified format, as shown in Table 1;

[0075] A warning processing unit, which is used to analyze the warning data and generate warning maintenance suggestions.

[0076] Table 1. Example table of warning data

[0077]

[0078] Through the warning module, the present invention realizes the intelligent monitoring and preventive maintenance of the solar LED street lamp system, which can improve the operation and maintenance efficiency of the system, extend the service life of the equipment, reduce the maintenance cost, and ensure the long-term stability and reliability of the solar LED street lamp system.

[0079] Furthermore, the charging system further includes a bidirectional energy flow management module, as Figure 2 shown, including: managing the bidirectional energy flow between the battery and the backup energy storage device; obtaining the remaining battery power of the first battery and the remaining battery power of the second battery according to the first battery parameter set and the second battery parameter set respectively; calculating the rechargeable amount according to the remaining battery power of the first battery and the charging upper limit; if the rechargeable amount exceeds the predicted power generation amount, storing the electric energy generated by the omnidirectional solar collection module in the battery; if the rechargeable amount does not exceed the predicted power generation amount, storing the electric energy generated by the omnidirectional solar collection module in the battery, and when the voltage of the battery reaches the charging upper limit, storing the electric energy generated by the omnidirectional solar collection module in the backup energy storage device; calculating the available power according to the remaining battery power of the first battery and the discharge lower limit; if the available power exceeds the predicted consumption amount, using the battery to supply power to the omnidirectional solar LED street lamp; if the available power does not exceed the predicted consumption amount, using the battery to supply power to the omnidirectional solar LED street lamp; when the voltage of the battery reaches the discharge lower limit, using the backup energy storage device to supply power to the omnidirectional solar LED street lamp.

[0080] Through the bidirectional energy flow management module, the present invention improves the energy management efficiency and power supply stability of the solar LED street lamp system. This module can dynamically manage the bidirectional energy flow between the battery and the backup energy storage device to ensure the reasonable distribution and storage of electric energy in different situations of the system. According to the calculation of the remaining battery power and the charging upper limit, when the battery reaches the charging upper limit, the system will automatically transfer the excess solar electric energy to the backup energy storage device to avoid overcharging the battery and make full use of solar energy resources. At the same time, by calculating the available battery power and the discharge lower limit, when the battery power reaches the discharge lower limit, the system will intelligently switch to the backup energy storage device for power supply to ensure the uninterrupted power supply of the street lamp. While ensuring the battery life, this module further improves the system's response ability when the power supply is unstable or the environmental conditions are poor, provides a more continuous and reliable power supply solution, and realizes the efficient operation and safety guarantee of the system.

[0081] Furthermore, determining the charging upper limit and the discharge lower limit according to the current battery life cycle and the current battery health status includes:

[0082] Obtain the historical usage time, charge and discharge times, and rated service life of the battery; determine the current battery life cycle of the battery according to the historical usage time, charge and discharge times, and rated service life of the battery.

[0083] Obtain the charge and discharge data of multiple historical all-round solar LED street lights with completed life cycles from the historical charge and discharge database to obtain historical reference data; the historical reference data covers the full life cycle charge and discharge records of the batteries of the historical all-round solar LED street lights from initial use to end of life, and the full life cycle covers all life cycle stages, including the initial stage, the middle stage, and the late stage.

[0084] Table 2. Charge and Discharge Data Table of a Historical All-round Solar LED Street Light with a Completed Life Cycle

[0085]

[0086] The charge and discharge data of a certain all-round solar LED street light with a completed life cycle can be referred to Table 2. The charge and discharge times in Table 2 refer to charge and discharge cycles, that is, the battery completes a complete charging and discharging process. A total of 5000 charge and discharge processes were carried out during the entire life cycle of this all-round solar LED street light.

[0087] Select the historical reference data with charge and discharge times greater than a preset threshold to obtain charge and discharge reference data for long-life batteries; in this embodiment, select the historical reference data with charge and discharge times greater than the preset threshold (5000 times) in the historical all-round solar LED street lights as the charge and discharge reference data for long-life batteries.

[0088] Preprocess the charge and discharge reference data for long-life batteries to obtain charge and discharge reference data for standard long-life batteries; the charge and discharge reference data for standard long-life batteries includes the battery health status during the charging process and the discharging process.

[0089] Extract the current charge and discharge times and current usage time of each all-round solar LED street light battery in the charge and discharge reference data for standard long-life batteries; determine the life cycle according to the current charge and discharge times and the current usage time.

[0090] Divide the charge and discharge reference data for standard long-life batteries into N groups according to the life cycle and the battery health status to obtain N groups of battery status data, and the battery status data includes battery charging process data and battery discharging process data; in this embodiment, the value of N is 8.

[0091] Specifically, according to the service life and charging and discharging times of the battery, the battery life cycle is divided into different stages. The life cycle of a typical lithium iron phosphate battery is divided into the initial stage (0 - 2 years, charging and discharging times: 0 - 1000 times), the middle stage (2 - 6 years, charging and discharging times: 1000 - 3000 times), and the late stage (6 - 10 years, charging and discharging times: more than 3000 times); each life cycle stage can be further subdivided according to the battery health state. It is divided by the percentage of the battery health state (SOC, State of Charge);

[0092] The battery charging process data includes the charging start voltage, charging end voltage, charging current, charging duration, and charging temperature change; the battery discharging process data includes the discharging start voltage, discharging end voltage, discharging current, discharging duration, and discharging temperature change;

[0093] Table 3. Lithium Iron Phosphate Battery Life Cycle and Health State Grouped Charging and Discharging Management Table

[0094]

[0095] For each group of the battery state data, extract the maximum value of the charging end voltage in the battery charging process data and the minimum value of the discharging end voltage in the battery discharging process data to obtain the maximum charging voltage and the minimum discharging voltage, as shown in Table 3.

[0096] According to the current battery life cycle and the current battery health state, screen out the groups in the N groups of the battery state data where the battery health state is the same as the current battery health state and the battery life cycle is the same as the current battery life cycle to obtain the same-stage battery state data; obtain the maximum charging voltage and the minimum discharging voltage of the same-stage battery state data to obtain the charging upper limit and the discharging lower limit.

[0097] The charging system of this all-round solar LED street lamp dynamically determines the charging upper limit and the discharging lower limit by analyzing the historical usage data and the current health state of the battery. At the same time, by combining the historical data of the optimized life battery, the system can extract the reference data matching the current battery health state to achieve targeted adjustment and avoid overcharging and deep discharging. This adaptive management method can effectively extend the service life of the battery and prevent the battery from premature aging caused by improper charging and discharging. Secondly, dynamically adjusting the charging upper limit and the discharging lower limit, optimizing the charging and discharging process according to different life cycle stages, improving the efficiency and safety of the battery, ensuring a stable power supply can still be provided when the battery health deteriorates, thereby improving the overall stability and reliability of the system.

[0098] Furthermore, the energy consumption and power generation prediction module includes a street lamp energy consumption prediction model, and the training process of the street lamp energy consumption prediction model includes the following steps:

[0099] Preprocess the historical data of the solar LED street lamp to obtain the standard historical data of the solar LED street lamp; the standard historical data of the solar LED street lamp includes historical standard street lighting data, historical standard street lamp energy consumption data, historical standard environmental data, and historical standard weather data; the historical standard environmental data includes image data and sound data of a specified lighting area;

[0100] Use feature engineering techniques to extract features from the standard historical data to obtain a feature set; the feature set includes time features, lighting intensity features, energy consumption features, street lamp control features, battery features, weather features, regional environment features, and traffic flow features; the traffic flow features include the number of pedestrians and vehicles extracted according to the image data and the sound data; the regional environment features include the number of faults and fault distances of surrounding street lamps in a specified area;

[0101] Construct interaction features based on the feature set, and integrate the interaction features with the feature set into a first fusion feature set; the interaction features include: time and traffic flow interaction features, weather and battery interaction features, regional environment and lighting intensity interaction features, and street lamp control and energy consumption interaction features;

[0102] Use the first fusion feature set as the input and the historical street lamp energy consumption data in the historical data of the solar LED street lamp as the label to train the XGBoost model, and obtain the street lamp energy consumption prediction model through hyperparameter optimization.

[0103] By preprocessing and feature engineering of multi-dimensional historical data, multi-dimensional factors such as time, lighting intensity, weather, traffic flow, and regional environment are comprehensively considered, thereby improving the accuracy and stability of energy consumption prediction. The introduction of interaction features enables the street lamp energy consumption prediction model to more effectively capture the complex correlations between these features. Especially in special situations such as high traffic flow or extreme weather, it can accurately predict energy consumption fluctuations and achieve refined energy consumption management. At the same time, accurate same-day energy consumption prediction can lay a data foundation for the formulation of same-day charging and discharging strategies.

[0104] Further, according to the historical data of the solar LED street lamp, predict the solar power generation capacity during the charging period of the current day, and the specific predicted power generation amount is obtained as follows: preprocess the historical environmental data, the historical weather data, and the historical solar panel power generation data in the historical data of the solar LED street lamp to obtain historical standard environmental data, historical standard weather data, and historical standard solar panel power generation data; obtain historical power generation amount data from the historical standard solar panel power generation data; use feature engineering technology to extract features from the historical standard environmental data, historical standard weather data, historical standard illumination data, and historical standard solar panel power generation data and construct interaction features to obtain a second fusion feature set; use the second fusion feature set as the input and the historical power generation amount data as the label, and train a solar panel power generation prediction model according to the XGBoost model;

[0105] Organize the environmental data and weather forecast data of the current day into the feature format during training, and gradually input them into the trained solar panel power generation prediction model according to time windows to predict the power generation amount within each time window; accumulate the power generation amounts of all the time windows to obtain the predicted power generation amount.

[0106] By using feature engineering technology and the construction of interaction features, the system can capture the complex correlations among the environment, weather, and power generation data, improving the accuracy of power generation prediction. Secondly, by training the historical data using the XGBoost model, the system can make fine-grained predictions for each time period based on real-time environmental and weather forecast data, achieving a more stable power generation plan. Finally, after accumulating the power generation amounts of each time period, the system can obtain an accurate all-day power generation prediction, optimize the charging strategy of the battery, avoid energy waste, and ensure that the solar LED street lamp can still operate efficiently under changing weather conditions.

[0107] Further, generate the charging strategy according to the deep reinforcement learning algorithm, and the generation process includes:

[0108] Construct a state set according to the first battery parameter set; the first battery parameter set includes the voltage, current, temperature, state of charge, and health state of the battery during the charging period;

[0109] Define an action set, including adjusting the charging current of the battery, selecting a charging mode, and switching to the backup energy storage device for charging;

[0110] Through a state transition function, calculate the state transition process under different actions according to the changes of the current state set and the action set to obtain the next moment state under each action; the next moment state includes battery voltage, battery current, battery temperature, battery state of charge, battery health state, and backup energy storage device state;

[0111] The charging module evaluates actions based on the predicted power generation, the charging upper limit, and the result of state transition, defines rewards according to energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery overcharge penalty, and obtains a reward function;

[0112] The charging module generates the charging strategy according to the state set, the action set, the state transition function, and the reward function through the deep reinforcement learning algorithm.

[0113] Further, as Figure 3 shown, the process of generating the charging strategy includes:

[0114] Using a policy network to select continuous actions according to the current state; evaluating the long-term cumulative reward of the continuous actions in the current state using a Q network according to the reward function; through an experience replay mechanism, the charging module repeatedly trains the policy network and the Q network, and optimizes and updates the weights of the policy network and the Q network using backpropagation and gradient descent algorithms to maximize the long-term cumulative reward. When the preset convergence condition is met, the charging strategy is obtained.

[0115] In this embodiment, it specifically includes the following steps:

[0116] S1: Construct a state set ;

[0117] During the charging period, the current state set of the system is defined as:

[0118] ;

[0119] Among them, represents the voltage of the battery at the current time t; represents the current of the battery at the current time t; represents the temperature of the battery at the current time t; represents the state of charge (State of Charge) of the battery at the current time t, that is, the percentage of the current remaining power; represents the health state of the battery at the current time t; represents the state of the backup energy storage device of the battery at the current time t, that is, the charging situation of the current backup energy storage device; represents the predicted consumption for the day; represents the predicted power generation for the day; represents the voltage of the battery at the current time t;

[0120] During the calculation of the charging strategy, the predicted power generation and the predicted consumption have been calculated in advance based on the environmental conditions of the day, historical data, and the battery health status, while the charging upper limit is dynamically set according to the current health status and life cycle of the battery. Therefore, within a day's charging cycle, these parameters do not change further with time.

[0121] S2: Define the action set; in each state the system can take a series of optional actions, define the action set :

[0122] ;

[0123] Among them, represents adjusting the battery charging current, for example, selecting different charging rates; represents selecting a charging mode, including constant current charging or constant voltage charging; represents switching to a backup energy storage device for charging.

[0124] S3: Construct the state transition function; the state transition function defines how the state of the system transfers to the next moment state after executing the action :

[0125] ;

[0126] In this embodiment, the state transition function is derived through a physical model and an empirical formula.

[0127] S4: Define the reward function; the system uses the reward function to evaluate the effect of each action and defines the corresponding reward and punishment mechanisms:

[0128] ;

[0129]

[0130] Among them, represents the energy utilization efficiency, that is, the ratio of the energy effectively stored in the battery to the total input energy; is the battery voltage penalty. If the battery voltage exceeds the charging upper limit , the system is punished; is the battery overheat penalty. If the battery temperature exceeds the safe temperature range, the system is punished; is the battery health status deterioration penalty. If the SOH drops too fast or the SOH is too low, the system is punished; is the rechargeable amount calculated based on the remaining battery power and the charging upper limit; represents the predicted power generation for the day; is the battery overcharge penalty, indicating that when the predicted power generation exceeds the rechargeable amount of the battery, the excess energy is still stored in the battery and the system is penalized; 、 、 、 and are the weight coefficients of the energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery overcharge penalty respectively; () is the function to find the maximum value;

[0131] Calculate the immediate reward at time t through the reward function ;

[0132] S5: The deep reinforcement learning algorithm generates a charging strategy;

[0133] Using the deep reinforcement learning algorithm, the process of the system generating a charging strategy is as follows:

[0134] Select an action using the policy network: According to the current state , through the policy network select an optimal action , that is:

[0135] ;

[0136] Evaluate the long-term cumulative reward of the action: Through the Q network evaluate the long-term cumulative reward of the selected action in the current state. The Q network update formula is:

[0137] ;

[0138] Among them, are the weight parameters of the Q network, and the weights of the network will be continuously optimized during the training process; represents at the current time t, given the state and the action , through the Q network parameters the Q value evaluated (i.e., the estimated value of the cumulative reward); is the learning rate; is the discount factor, which controls the degree of emphasis on future rewards; represents the maximum Q value after selecting the optimal action at the next time t + 1 state .

[0139] Experience replay mechanism: The system saves each state-action-reward-new state quadruple Store in the experience replay buffer and repeatedly train the policy network and Q-network through the experience replay mechanism.

[0140] Backpropagation and gradient descent: Use backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and Q-network to maximize the long-term cumulative reward.

[0141] Stop training when any of the following convergence conditions are met:

[0142] The change in Q-value is less than the preset threshold, or the change in cumulative reward is less than the preset threshold.

[0143] Therefore, the final charging strategy is the optimal action selected in the current state ;

[0144] ;

[0145] where denotes selecting the action that can make the function reach the maximum value .

[0146] S6: Execute the charging strategy; According to the generated charging strategy, the system adjusts the charging current, charging mode or switches to the backup energy storage device for charging when necessary to optimize energy utilization and reduce adverse phenomena such as overcharging and overheating.

[0147] The charging system of this all-round solar LED street lamp generates a charging strategy through a deep reinforcement learning algorithm, which can flexibly adjust the charging mode and charging current according to multiple parameters such as the voltage, current, temperature, state of charge and health state of the battery, and switch to the backup energy storage device when necessary. This charging strategy based on multi-dimensional state parameters and dynamic prediction greatly improves the energy utilization efficiency, reduces problems such as overcharging, overheating and over-discharging of the battery, and effectively extends the service life of the battery. At the same time, the system ensures the stability and safety of the battery during long-term operation through mechanisms such as monitoring the health state of the battery and over-discharge penalty, realizing intelligent and precise management of the battery charging process. At the same time, by combining the policy network and Q-network in deep reinforcement learning, it can select the optimal continuous charging action according to the current battery state, and continuously optimize the strategy through the experience replay mechanism. During the repeated training process, the system uses backpropagation and gradient descent algorithms to optimize the weights of the policy network and Q-network to ensure that the charging strategy can maximize the long-term cumulative reward in different states. This mechanism enables the system to achieve dynamic and adaptive charging control, greatly improving the energy utilization efficiency of the battery, reducing the risk of overcharging and battery aging, effectively extending the service life of the battery, and maintaining efficient and stable operation in complex environments.

[0148] Further, obtaining the discharge strategy according to the second battery parameter set, the predicted consumption, and the discharge lower limit includes:

[0149] Generating the discharge strategy according to the deep reinforcement learning algorithm, as Figure 4 shown, and the generation process includes:

[0150] Constructing a state set according to the second battery parameter set; the second battery parameter set includes the voltage, current, temperature, state of charge, and health state of the battery during the discharge period;

[0151] Defining an action set, including adjusting the charging current of the battery, selecting a discharge mode, and switching to the standby energy storage device for discharge;

[0152] Through the state transition function, according to the changes in the current state set and the action set, calculating the state transition process under different actions to obtain the next moment state under each action; the next moment state includes battery voltage, battery current, battery temperature, battery state of charge, battery health state, and standby energy storage device state;

[0153] The discharge module evaluates the actions according to the predicted consumption, the discharge lower limit, and the result of state transition, defines rewards according to energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery over-discharge penalty, and obtains a reward function;

[0154] The discharge module generates the discharge strategy through the deep reinforcement learning algorithm according to the state set, the action set, the state transition function, and the reward function;

[0155] Using the policy network to select continuous actions according to the current state; according to the reward function, using the Q network to evaluate the long-term cumulative reward of the continuous actions in the current state; through the experience replay mechanism, the discharge module repeatedly trains the policy network and the Q network through the experience replay mechanism, and uses backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and the Q network to maximize the long-term cumulative reward. When the preset convergence condition is met, the discharge strategy is obtained.

[0156] The discharge module of this all-round solar LED street lamp uses a deep reinforcement learning algorithm to generate an optimal discharge strategy using a policy network and a Q-network. It can dynamically adjust the discharge process according to real-time parameters such as the voltage, current, temperature, state of charge, and health state of the battery. Through the state transition function and the reward function, the system can not only efficiently manage the energy output during the discharge process but also avoid problems such as over-discharge and overheating of the battery. At the same time, the experience replay mechanism and backpropagation optimization enable the system to continuously update the discharge strategy to maximize the long-term cumulative reward. This mechanism effectively improves the energy utilization efficiency of the discharge process, extends the battery life, ensures that the system can operate efficiently and stably in complex and variable environments, and further optimizes the intelligent level of the overall power management.

[0157] The present invention realizes intelligent charge and discharge management of the solar LED street lamp system by integrating a charge and discharge threshold determination module, a historical data acquisition module, an energy consumption and power generation prediction module, and a charge and discharge module. The system can accurately determine the charge upper limit and discharge lower limit according to the current battery life cycle and health state of the battery, ensuring that the battery works in the best state and extending its service life. Through the acquisition and analysis of historical data, the system can accurately predict the energy consumption demand and power generation capacity of the day, thereby formulating a more reasonable charge and discharge strategy, adapting to the variability of the discharge behavior, and avoiding over-discharge or overcharge phenomena. The charge and discharge processes are interdependent, ensuring stable power supply through a backup energy storage device when power generation is insufficient, improving the energy utilization efficiency and reliability of the system, and enabling the solar LED street lamp to operate continuously and efficiently in the complex environment of an intelligent city. This system ensures the stability and efficient operation of the street lamp system in complex and variable environments through advance prediction and dynamic adjustment.

[0158] Embodiment 2

[0159] Company C produced a batch of all-round solar LED street lamps, and all of these street lamps are applied with a charging system for an all-round solar LED street lamp, including:

[0160] A charge and discharge threshold determination module, which is used to obtain the current battery life cycle and the current battery health state of the solar LED street lamp; and determine the charge upper limit and the discharge lower limit according to the current battery life cycle and the current battery health state;

[0161] A historical data acquisition module, which is used to obtain the historical street lamp lighting data, historical street lamp energy consumption data, historical environmental data, historical weather data, historical light data, and historical solar panel power generation data of the solar LED street lamp to obtain the historical data of the solar LED street lamp;

[0162] The energy consumption and power generation prediction module is used to predict the energy consumption demand during the discharge period of the current day and the solar power generation capacity during the charging period of the current day based on the historical data of the solar LED street lamp, so as to obtain the predicted consumption and the predicted power generation;

[0163] The charging module is used to obtain the current parameters of the battery during the charging period to obtain a first set of battery parameters, and obtain a charging strategy according to the first set of battery parameters, the predicted power generation, the predicted consumption, and the charging upper limit; charge the battery and the backup energy storage device using the charging strategy;

[0164] The discharge module is used to obtain the current parameters of the battery during the discharge period to obtain a second set of battery parameters, and obtain a discharge strategy according to the second set of battery parameters, the predicted consumption, and the discharge lower limit; supply power to the solar LED street lamp through the battery and the backup energy storage device according to the discharge strategy.

[0165] Specifically, the discharge module evaluates the actions according to the predicted consumption, the discharge lower limit, and the result of state transition, defines rewards according to the energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery over-discharge penalty, and obtains a reward function;

[0166] Further, obtaining the discharge strategy according to the second set of battery parameters, the predicted consumption, and the discharge lower limit includes:

[0167] S1: Construct a state set ;

[0168] During the discharge period, the current state set of the system is defined as:

[0169] ;

[0170] Among them, represents the voltage of the battery at the current time t (during the discharge period); represents the current of the battery at the current time t; represents the temperature of the battery at the current time t; represents the state of charge (State of Charge) of the battery at the current time t, that is, the percentage of the current remaining power; represents the health state of the battery at the current time t; represents the state of the backup energy storage device of the battery at the current time t, that is, the current discharge situation of the backup energy storage device; represents the voltage of the battery at the current time t;

[0171] predicted consumption is a fixed value within the day. During the calculation of the discharge strategy, the predicted consumption has been calculated in advance based on the environmental conditions, historical data, and battery health status of the day, while the discharge lower limit is dynamically set according to the current health status and life cycle of the battery. Therefore, within a one-day discharge cycle, these parameters do not change further over time.

[0172] S2: Define the action set; At each state the system can take a series of optional actions, and define the action set :

[0173] ;

[0174] Among them, represents adjusting the battery discharge current, for example, selecting different discharge rates; represents selecting a discharge mode, including constant current discharge or constant voltage discharge; represents switching to the standby energy storage device for discharge.

[0175] S3: Construct the state transition function; The state transition function defines how the state of the system transfers to the next moment state after executing the action :

[0176] ;

[0177] In this embodiment, the state transition function is derived through a physical model and empirical formula.

[0178] S'4: Define the reward function; The system uses the reward function to evaluate the effect of each action and defines the corresponding reward and penalty mechanisms:

[0179] ;

[0180]

[0181] Among them, represents the energy utilization efficiency, that is, the proportion of the energy effectively stored in the battery to the total input energy; is the battery voltage penalty. If the battery voltage is less than the discharge lower limit , the system is penalized; is the battery overheat penalty. If the battery temperature exceeds the safe temperature range, the system is penalized; is the battery health status deterioration penalty. If the SOH drops too fast or the SOH is too low, the system is penalized; is the available power calculated based on the remaining battery power and the lower discharge limit; represents the predicted consumption for the current day; is the battery over-discharge penalty, which is used to evaluate whether the battery is overused (not switched to the backup charging device for discharge in time) when the predicted consumption exceeds the available battery power, and impose corresponding penalties; 、 、 、 and are the weight coefficients of the energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery over-discharge penalty respectively; is the function to find the maximum value;

[0182] Calculate the immediate reward at time t through the reward function ;

[0183] S5: The deep reinforcement learning algorithm generates a discharge strategy;

[0184] Using the deep reinforcement learning algorithm, the process of the system generating a discharge strategy is as follows:

[0185] Select an action using the policy network: According to the current state , through the policy network select an optimal action , that is:

[0186] ;

[0187] Evaluate the long-term cumulative reward of the action: Through the Q network evaluate the long-term cumulative reward of the selected action in the current state. The Q network update formula is:

[0188] ;

[0189] Among them, are the weight parameters of the Q network, and the weights of the network will be continuously optimized during the training process; represents at the current time t, given the state and the action , through the Q network parameters the Q value evaluated (i.e., the estimated value of the cumulative reward); is the learning rate; is the discount factor, which controls the degree of emphasis on future rewards; represents the maximum Q value after selecting the optimal action at the next time t + 1 state .

[0190] Experience replay mechanism: The system stores each state-action-reward-new state quadruple in the experience replay buffer and repeatedly trains the policy network and Q network through the experience replay mechanism.

[0191] Backpropagation and gradient descent: Use backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and Q network to maximize the long-term cumulative reward.

[0192] Stop training when any of the following convergence conditions are met:

[0193] The change in Q value is less than a preset threshold or the change in cumulative reward is less than a preset threshold.

[0194] Therefore, the final discharge strategy is the optimal action selected in the current state ;

[0195] ;

[0196] where denotes selecting the action that can make the function reach the maximum value .

[0197] S6: Execute the discharge strategy; According to the generated discharge strategy, the system adjusts the discharge current, discharge mode, and switches to the standby energy storage device for discharge when necessary to optimize energy utilization and reduce adverse phenomena such as over-discharge and overheating.

[0198] 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. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A charging system for an all-round solar LED street lamp, characterized in that Including: A charge and discharge threshold determination module, configured to obtain the current battery life cycle and the current battery health status of a solar LED street lamp; Determine the charging upper limit and the discharging lower limit according to the current battery life cycle and the current battery health status; The charging upper limit and the discharging lower limit include: Obtain the charge and discharge data of multiple historical all-round solar LED street lamps with completed life cycles from a historical charge and discharge database to obtain historical reference data; the historical reference data covers the charge and discharge records of the batteries of the historical all-round solar LED street lamps from initial use to end of life, and the full life cycle covers all life cycle stages, including the initial stage, the middle stage, and the late stage; select the historical reference data with the number of charge and discharge times greater than a preset threshold to obtain the charge and discharge reference data of long-life batteries; preprocess the charge and discharge reference data of long-life batteries to obtain the charge and discharge reference data of standard long-life batteries; the charge and discharge reference data of standard long-life batteries includes the battery health status during the charging process and the discharging process; Extract the current number of charge and discharge times and the current usage time of each battery of the solar LED street lamp in the charge and discharge reference data of standard long-life batteries; determine the life cycle according to the current number of charge and discharge times and the current usage time; Divide the charge and discharge reference data of standard long-life batteries into N groups according to the life cycle and the battery health status to obtain N groups of battery status data, where the battery status data includes battery charging process data and battery discharging process data; the battery charging process data includes the charging start voltage and the charging end voltage; the battery discharging process data includes the discharging start voltage and the discharging end voltage; For each group of the battery status data, extract the maximum value of the charging end voltage in the battery charging process data and the minimum value of the discharging end voltage in the battery discharging process data to obtain the maximum charging voltage and the minimum discharging voltage; According to the current battery life cycle and the current battery health status, screen out the groups that meet the specified conditions from the N groups of battery status data to obtain the battery status data in the same stage; obtain the maximum charging voltage and the minimum discharging voltage of the battery status data in the same stage to obtain the charging upper limit and the discharging lower limit; the specified condition is that the battery health status is the same as the current battery health status and the battery life cycle is the same as the current battery life cycle; A historical data acquisition module, configured to obtain the historical street lamp lighting data, historical street lamp energy consumption data, historical environmental data, historical weather data, historical light data, and historical solar panel power generation data of the solar LED street lamp to obtain the historical data of the solar LED street lamp; An energy consumption and power generation prediction module, configured to predict the energy consumption demand during the discharging period of the day and the solar power generation capacity during the charging period of the day according to the historical data of the solar LED street lamp to obtain the predicted consumption and the predicted power generation; The charging module is used to obtain the current parameters of the battery during the charging time period, obtain the first battery parameter set, and obtain a charging strategy based on the first battery parameter set, the predicted power generation amount, the predicted consumption amount, and the charging upper limit; use the charging strategy to charge the battery and the backup energy storage device; The discharging module is used to obtain the current parameters of the battery during the discharging time period, obtain the second battery parameter set, and obtain a discharging strategy based on the second battery parameter set, the predicted consumption amount, and the discharging lower limit; supply power to the solar LED street lamp through the battery and the backup energy storage device according to the discharging strategy.

2. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, The charging system further includes: an omnidirectional solar energy collection module, which is used to collect sunlight at different angles and convert the sunlight into electric energy; it includes a plurality of solar panels with adjustable angles.

3. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, The charging system further includes a warning module, including: generating a panel warning signal when it detects that a solar panel fails; generating a life cycle warning signal when it detects that the current battery life cycle is in the late stage; generating a battery state warning signal when the current battery health state is lower than a preset threshold.

4. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, The charging system further includes a bidirectional energy flow management module, including: managing the bidirectional energy flow between the battery and the backup energy storage device; obtaining the first remaining battery power and the second remaining battery power respectively according to the first battery parameter set and the second battery parameter set; calculating the chargeable amount according to the first remaining battery power and the charging upper limit; if the chargeable amount exceeds the predicted power generation amount, storing the electric energy generated by the omnidirectional solar energy collection module into the battery; if the chargeable amount does not exceed the predicted power generation amount, storing the electric energy generated by the omnidirectional solar energy collection module into the battery, and when the voltage of the battery reaches the charging upper limit, storing the electric energy generated by the omnidirectional solar energy collection module into the backup energy storage device; Calculating the available power according to the first remaining battery power and the discharging lower limit; if the available power exceeds the predicted consumption amount, using the battery to supply power to the omnidirectional solar LED street lamp; if the available power does not exceed the predicted consumption amount, using the battery to supply power to the omnidirectional solar LED street lamp; when the voltage of the battery reaches the discharging lower limit, using the backup energy storage device to supply power to the omnidirectional solar LED street lamp.

5. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, The energy consumption and power generation prediction module includes a street lamp energy consumption prediction model, and the training process of the street lamp energy consumption prediction model includes the following steps: Preprocessing the historical data of the solar LED street lamp to obtain the standard historical data of the solar LED street lamp; the standard historical data of the solar LED street lamp includes historical standard street lamp lighting data, historical standard street lamp energy consumption data, historical standard environmental data, and historical standard weather data; the historical standard environmental data includes image data and sound data of a specified lighting area; Feature extraction is performed on the standard historical data using feature engineering techniques to obtain a feature set; the feature set includes time features, lighting intensity features, energy consumption features, street lamp control features, battery features, weather features, regional environment features, and traffic flow features; the traffic flow features include the number of pedestrians and vehicles extracted from the image data and the sound data; the regional environment features include the number of faults and the fault distance of surrounding street lamps within a specified area; interaction features are constructed based on the feature set, and the interaction features and the feature set are integrated into a first fusion feature set; the interaction features include: time and traffic flow interaction features, weather and battery interaction features, regional environment and lighting intensity interaction features, and street lamp control and energy consumption interaction features; Taking the first fusion feature set as input and the historical street lamp energy consumption data in the historical data of solar LED street lamps as labels, an XGBoost model is trained, and the street lamp energy consumption prediction model is obtained through hyperparameter optimization.

6. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, Based on the historical data of solar LED street lamps, the solar power generation capacity during the charging period of the current day is predicted, and the predicted power generation amount is specifically: preprocessing the historical environmental data, the historical weather data, and the historical solar panel power generation data in the historical data of solar LED street lamps to obtain historical standard environmental data, historical standard weather data, and historical standard solar panel power generation data; obtaining historical power generation data from the historical standard solar panel power generation data; using feature engineering techniques to perform feature extraction and construct interaction features on the historical standard environmental data, historical standard weather data, historical standard lighting data, and historical standard solar panel power generation data to obtain a second fusion feature set; taking the second fusion feature set as input and the historical power generation data as labels, training a solar panel power generation prediction model according to the XGBoost model; organizing the environmental data and weather forecast data of the current day into the feature format during training, and gradually inputting them into the trained solar panel power generation prediction model according to time windows to predict the power generation amount within each time window; accumulating the power generation amounts of all the time windows to obtain the predicted power generation amount.

7. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, Generating the charging strategy according to the deep reinforcement learning algorithm, and the generation process includes: Constructing a state set according to the first battery parameter set; the first battery parameter set includes the voltage, current, temperature, state of charge, and health state of the battery during the charging period; Defining an action set, including adjusting the charging current of the battery, selecting a charging mode, and switching to the backup energy storage device for charging; Through a state transition function, according to the changes in the current state set and the action set, calculating the state transition process under different actions to obtain the next moment state under each action; the next moment state includes battery voltage, battery current, battery temperature, battery state of charge, battery health state, and backup energy storage device state; The charging module evaluates actions based on the predicted power generation, the charging upper limit, and the result of state transition, defines rewards according to energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery overcharge penalty, and obtains a reward function. The charging module generates the charging strategy through the deep reinforcement learning algorithm according to the state set, the action set, the state transition function, and the reward function.

8. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, The process of generating the charging strategy includes: using a policy network to select continuous actions based on the current state; using a Q network to evaluate the long-term cumulative reward of the continuous actions in the current state according to the reward function; through an experience replay mechanism, the charging module repeatedly trains the policy network and the Q network, and uses backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and the Q network to maximize the long-term cumulative reward. When the preset convergence condition is met, the charging strategy is obtained.

9. The charging system of an all-round solar LED street lamp according to claim 1, characterized in that, Obtaining the discharging strategy according to the second battery parameter set, the predicted consumption, and the discharging lower limit includes: Generating the discharging strategy according to the deep reinforcement learning algorithm, and the generating process includes: constructing a state set according to the second battery parameter set; the second battery parameter set includes the voltage, current, temperature, state of charge, and health state of the battery during the discharging period; defining an action set, including adjusting the charging current of the battery, selecting a discharging mode, and switching to the standby energy storage device for discharging; through a state transition function, calculating the state transition process under different actions according to the changes of the current state set and the action set, and obtaining the next moment state under each action; the next moment state includes battery voltage, battery current, battery temperature, battery state of charge, battery health state, and standby energy storage device state; the discharging module evaluates actions based on the predicted consumption, the discharging lower limit, and the result of state transition, defines rewards according to energy utilization efficiency, battery voltage penalty, battery overheat penalty, battery health state deterioration penalty, and battery over-discharge penalty, and obtains a reward function; the discharging module generates the discharging strategy through the deep reinforcement learning algorithm according to the state set, the action set, the state transition function, and the reward function; using a policy network to select continuous actions based on the current state; using a Q network to evaluate the long-term cumulative reward of the continuous actions in the current state according to the reward function; through an experience replay mechanism, the discharging module repeatedly trains the policy network and the Q network, and uses backpropagation and gradient descent algorithms to optimize and update the weights of the policy network and the Q network to maximize the long-term cumulative reward. When the preset convergence condition is met, the discharging strategy is obtained.

Images