An MPPT-type integrated solar charging, discharging and lighting street lamp system

Abstract

The present invention provides an MPPT type solar charge-discharge light source integrated street lamp system, which relates to the technical field of solar street lamps. The specific implementation steps include: First, the actual electric energy is obtained through the solar conversion module, and the battery parameters are obtained through the battery data acquisition module; Second, the battery charging control module determines the charging mode based on the calibration parameters of different battery types and dynamically adjusts the charging current; Then, the light source demand analysis module adjusts the actual illuminance of the street lamp in combination with the adjacent street lamp spacing and meteorological data; Next, the battery discharge control module adjusts the discharge current through the non-linear mapping of the actual power demand of the light source and the discharge current; Finally, the battery state evaluation module predicts the battery health state through the battery state evaluation model and feeds it back to the battery charging control module and the battery discharge control module, effectively improving the charge-discharge efficiency of different battery types, extending the battery service life, and further improving the stability and consistency of the street lamp system.

Description

Technical Field

[0001] The present invention relates to the technical field of solar street lamps, and in particular to an MPPT type solar charging and discharging light source integrated street lamp system. Background Art

[0002] With the intensification of the global energy crisis and rising environmental awareness, solar energy, as a renewable energy source, is increasingly being adopted in various lighting systems, particularly in public facilities. However, traditional solar street lights typically use pulse-width modulation (PWM) to control the output current and voltage of solar panels. This fails to fully utilize the panels' power generation potential, resulting in low charging efficiency in poor lighting conditions and impacting overall system performance. To address this, MPPT (maximum power point tracking) technology has been introduced into solar street light systems. This technology adjusts the operating voltage and current of solar panels in real time, maintaining optimal power point operation. This significantly improves photovoltaic power generation efficiency and reduces power loss.

[0003] At present, existing MPPT solar street lights mainly control the charging and discharging process by embedding the charging and discharging curves of various types of batteries. They do not consider the impact of the loss of different types of batteries on the charging and discharging performance. As a result, different types of batteries cannot perform at their best in the street light system, and overcharging or over-discharging may occur, resulting in the battery being unable to provide stable power to the street light.

[0004] Therefore, an MPPT solar charging and discharging light source integrated street lamp system is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide an MPPT-type solar charging and discharging light source integrated street lamp system to improve the stability and consistency of the street lamp system. In order to solve the problems existing in the prior art, the present invention first obtains actual electrical energy through the solar energy conversion module and obtains battery parameters through the battery data acquisition module; secondly, the battery charging control module determines the charging mode based on the calibration parameters of different battery types and dynamically adjusts the charging current; then, the light source demand analysis module adjusts the actual illumination of the street lamp in combination with the spacing between adjacent street lamps and meteorological data; then, the battery discharge control module adjusts the discharge current through nonlinear mapping of the actual power demand of the light source and the discharge current; finally, the battery status assessment module predicts the battery health status through the battery status assessment model and feeds back to the battery charging control module and the battery discharge control module, effectively improving the charging and discharging efficiency of different battery types, extending the battery life, and further improving the stability and consistency of the street lamp system.

[0006] An MPPT solar charging and discharging light source integrated street lamp system includes the following specific implementation steps:

[0007] A solar energy conversion module that collects the voltage, current, and temperature of a solar panel in real time through sensors, and combines the conversion efficiency of the MPPT controller to obtain the actual electric energy;

[0008] A battery data acquisition module that obtains battery calibration parameters, battery health, battery capacity, and remaining battery capacity, and collects the battery charging voltage, battery charging current, battery charging temperature, battery discharging voltage, battery discharging current, and battery discharging temperature in real time according to voltage sensors, current sensors, and temperature sensors;

[0009] A battery charging control module that obtains the charging mode according to the calibration parameters of different types of batteries; adjusts the battery charging current in different charging modes through temperature regulation, non-linear mapping of charging current, and power compensation;

[0010] A light source demand analysis module that determines the on / off state of the light source according to the ambient light intensity data; combines the distance between adjacent street lamps and ambient meteorological data, and obtains the second adjusted illuminance through the inverse square law formula and the weather condition influence factor;

[0011] A battery discharging control module that obtains the discharging mode according to the remaining battery capacity and the battery discharging temperature; adjusts the discharging current in different discharging modes through the actual power demand of the light source, the remaining battery capacity, and non-linear mapping of the discharging current;

[0012] A battery state evaluation module that inputs the publicly available lithium-ion battery aging dataset into the battery state evaluation model for model training; optimizes the model through transfer learning to further accurately predict the battery state after charging and discharging of different types of batteries.

[0013] Preferably, the solar energy conversion module includes:

[0014] Obtain the instantaneous voltage, instantaneous current, and instantaneous temperature of the solar panel according to voltage sensors, current sensors, and temperature sensors; obtain the instantaneous power through the instantaneous voltage and the instantaneous current; obtain the instantaneous temperature loss of the solar panel according to the instantaneous temperature; combine the instantaneous power, instantaneous temperature loss, and the conversion efficiency of the MPPT controller to obtain the actual electric energy.

[0015] Preferably, the battery calibration parameters include calibrated capacity, battery internal resistance, battery nominal voltage, battery self-discharge rate, calibrated charging voltage, calibrated minimum discharging voltage, calibrated maximum charging current, calibrated highest working temperature, calibrated highest charging temperature, and calibrated highest discharging temperature; the battery health, the battery capacity, and the remaining battery capacity are fed back by the battery state evaluation module.

[0016] Preferably, the battery charging control module includes:

[0017] According to the battery calibration parameters, obtain the calibrated charging voltage; input the battery calibration parameters into a multi-layer perceptron network to obtain the battery type; obtain the historical charging data corresponding to the battery type, input the battery health and the historical charging data into a convolutional neural network to obtain a first predetermined state threshold and a second predetermined state threshold; obtain the battery charging state according to the percentage of the remaining battery capacity to the battery capacity; obtain the charging voltage ratio through the ratio of the battery charging voltage to the calibrated charging voltage; determine the charging self-discharge rate according to the calibrated self-discharge rate and the battery charging temperature; if the battery charging state is less than or equal to the first predetermined state threshold and the charging voltage ratio is less than or equal to a predetermined voltage ratio, the battery adopts a first charging mode; if the battery charging state is less than the second predetermined state threshold and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts a second charging mode; if the battery charging state is greater than or equal to the second predetermined state threshold, the charging self-discharge rate is greater than a predetermined self-discharge rate, and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts a third charging mode.

[0018] Preferably, the battery charging control module further includes:

[0019] According to the battery calibration parameters, obtain the battery internal resistance, calibrated maximum charging temperature, calibrated maximum operating temperature, and calibrated maximum charging current; if the charging mode is the first charging mode, charge according to the calibrated maximum charging current and monitor the battery charging temperature in real time; if the battery charging temperature is greater than or equal to the calibrated maximum charging temperature, adjust the battery charging current through temperature limitation; if the charging mode is the second charging mode, obtain the equivalent capacitance of the battery according to the charging duration of the first charging mode, the difference between the battery charging current and the battery charging voltage within the charging duration; combine the battery internal resistance and the equivalent capacitance of the second charging mode to adjust the battery charging current; if the charging mode is the third charging mode, adjust the charging current according to the charging self-discharge rate, the battery charging temperature, and the battery charging voltage.

[0020] Preferably, the light source demand analysis module includes:

[0021] Collect the surrounding light intensity data in real time through a light sensor to obtain the average light intensity; if the average light intensity is less than a predetermined light threshold, turn on the light source; obtain the street lamp spacing and the standard illuminance, and obtain the first adjusted illuminance according to the inverse square law formula; combine the environmental meteorological data and the first adjusted illuminance to obtain the influence factor of different weather conditions on the illuminance, and obtain the second adjusted illuminance.

[0022] Preferably, the battery discharge control module includes:

[0023] Obtain the actual power demand of the light source through the second adjusted illuminance; if the remaining battery capacity is greater than the predetermined capacity threshold and the actual power demand is less than or equal to the rated maximum discharge power of the battery, adopt the first discharge mode; if the remaining battery capacity is less than or equal to the predetermined capacity threshold, adopt the second discharge mode; when adopting the first discharge mode or the second discharge mode, if the battery discharge voltage is less than the calibrated minimum discharge voltage or the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, stop discharging.

[0024] Preferably, the battery discharge control module further includes:

[0025] Obtain the nominal voltage and the calibrated maximum discharge temperature of the battery according to the battery calibration parameters; if the charging mode is the first discharge mode, obtain the first discharge current according to the actual power demand and the nominal voltage of the battery and perform discharging; monitor the battery discharge temperature in real time, if the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, adjust the battery charging current according to the calibrated maximum discharge temperature to obtain the second discharge current; if the charging mode is the second discharge mode, obtain the third discharge current through the non-linear mapping of the discharge current according to the first discharge current and the remaining battery capacity.

[0026] Preferably, the specific implementation process of the battery state evaluation module includes:

[0027] The data acquisition unit collects the charge and discharge historical data according to the battery data acquisition module, including the historical charge and discharge electricity, historical discharge electricity, historical charge and discharge voltage, historical charge and discharge current, historical charge and discharge temperature, and the number of charge and discharge cycles; combine the historical charge and discharge electricity and the battery capacity to obtain the historical charge depth and historical discharge depth;

[0028] The data preprocessing unit performs preprocessing operations on the data obtained by the data acquisition unit, including outlier processing, missing value processing, time alignment operation, and normalization processing, to obtain the standard historical data;

[0029] The battery state evaluation model training unit inputs the publicly available lithium-ion battery aging dataset into a convolutional neural network to obtain the charge and discharge feature sequence; according to the spatial self-attention mechanism, obtain the first enhanced feature sequence from the charge and discharge feature sequence; input the first enhanced feature sequence into a bidirectional convolutional long short-term memory network to obtain the second enhanced feature sequence; through the fully connected layer and the activation function, obtain the predicted battery health, remaining battery capacity ratio, and remaining battery life;

[0030] The battery state evaluation model inference unit performs data augmentation operations on the standard historical data to obtain augmented historical data; inputs the augmented historical data into a pre-trained battery state evaluation model, and adjusts the model through a transfer learning strategy to obtain the battery health, the battery remaining capacity ratio, and the battery remaining life after charging or discharging.

[0031] Preferably, the specific implementation process of the battery state evaluation module further includes:

[0032] Supervise the battery health and the battery remaining life predicted by the battery state evaluation model training unit according to the mean absolute loss function; constrain the battery remaining capacity ratio predicted by the battery state evaluation model training unit through the logarithmic hyperbolic cosine loss function; adjust the model weights of the battery state evaluation model inference unit through a weight update method to obtain the battery health, the battery remaining capacity ratio, and the battery remaining life after historical charging or discharging; obtain the current battery remaining capacity according to the battery remaining capacity ratio and the battery health; obtain the current battery capacity through the battery calibration parameters and the battery health; feedback the current battery remaining capacity and the current battery capacity to the battery charging control module to update the battery remaining capacity and the battery capacity.

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

[0034] 1. The above-mentioned scheme of the battery charging control module proposed by the present invention comprehensively analyzes the characteristics of different types of batteries by introducing a multi-layer perceptron network and a convolutional neural network, realizing accurate identification of battery types and optimization of charging strategies. This module intelligently switches and adjusts the charging mode, which not only ensures the safety and stability of different types of batteries in different working states, but also can accurately adjust the charging current and charging mode according to the real-time battery state, avoiding overcharging, over-discharging and battery damage caused by too high temperature; at the same time, by optimizing the charging mode switching conditions and the charging current adjustment strategy, it can effectively reduce the inconsistency caused by battery characteristic differences, improving the stability and service life of the solar street lamp system.

[0035] 2. The battery discharge control module proposed by the present invention precisely controls the remaining capacity of the battery, the battery discharge temperature and the actual power demand under different discharge modes, avoiding system instability caused by battery performance differences. Through the real-time monitoring and dynamic adjustment of the battery discharge temperature, the system can timely adjust the discharge current when the battery temperature approaches or exceeds the calibrated maximum discharge temperature, preventing battery performance degradation or safety hazards caused by too high temperature, which not only improves the adaptability of the system under different environmental temperatures, but also enhances the discharge consistency between different battery types.

[0036] 3. The battery state evaluation module proposed by the present invention uses algorithm models such as convolutional neural networks and bidirectional long short-term memory networks to perform deep learning on the historical charge and discharge data of the battery. Combining with the spatial self-attention mechanism to enhance the feature extraction ability, it accurately predicts the battery health, remaining capacity ratio, and remaining life. It not only takes into account the aging trend of the battery but also can perform personalized adjustment on the health status of different types of batteries, enabling the system to maintain consistency in diverse battery configurations. Through data augmentation and transfer learning, the system can adapt to the usage history of different batteries, timely adjust the charge and discharge strategies, prevent over-discharge or shortening of the battery life, and thus extend the service life of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flowchart of an MPPT type solar charging and discharging light source integrated street lamp system provided by an embodiment of the present invention;

[0038] Figure 2 It is a flowchart of charging mode matching provided by an embodiment of the present invention;

[0039] Figure 3 It is a schematic diagram of a solar street lamp on the road in Area A provided by an embodiment of the present invention;

[0040] Figure 4 It is a flowchart of the battery state evaluation module provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] 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.

[0042] MPPT type solar street lamps have high-efficient energy conversion capabilities and can track the maximum power point of solar panels in real time, thus maximizing the utilization of solar energy resources under various environmental conditions. However, different types of batteries such as lead-acid batteries, nickel-metal hydride batteries, gel batteries, and lithium batteries have significant differences in charge and discharge characteristics, life, and performance. If not properly managed, it may lead to a decrease in system efficiency or battery loss. Therefore, automatically adjusting the charge and discharge voltage and current of MPPT type solar street lamps according to the characteristics of different types of batteries can ensure that the battery is always in the best working state, not only extending the battery life but also improving the stability and consistency of the overall system.

[0043] The present invention proposes an MPPT-type solar charging and discharging light source integrated street lamp system to automatically adjust the charging and discharging processes of different types of energy storage batteries to ensure stable power supply for different batteries. To illustrate the effectiveness of the system proposed by the present invention, specific descriptions will be given in combination with the drawings of this embodiment and the following two embodiments.

[0044] Embodiment 1

[0045] The embodiment of the present application discloses an MPPT-type solar charging and discharging light source integrated street lamp system to achieve efficient lighting of roads in Area A under sunny conditions. Refer to Figure 1 , the system proposed by the present invention includes: a solar conversion module, a battery data acquisition module, a battery charging control module, a light source demand analysis module, a battery discharging control module, and a battery status evaluation module.

[0046] Furthermore, during the day, the solar conversion module obtains the instantaneous voltage, instantaneous current, and instantaneous temperature of the solar panel according to the voltage sensor, current sensor, and temperature sensor; obtains the instantaneous power through the instantaneous voltage and the instantaneous current; and obtains the instantaneous temperature loss of the solar panel according to the instantaneous temperature.

[0047] L temp i =V i ·I i ·η ref ·(1 - α·|T i -T ref |);

[0048] Among them, L temp i represents the instantaneous temperature loss at the i sampling point; V i , I i respectively represent the instantaneous voltage and instantaneous current at the i sampling point; η ref represents the conversion efficiency of the solar panel at the reference temperature; T i , T ref respectively represent the instantaneous temperature and reference temperature at the i sampling point; α represents the temperature sensitivity coefficient of the solar panel, which is obtained through a large amount of experimental data;

[0049] Combining the instantaneous power, instantaneous temperature loss, and the conversion efficiency of the MPPT controller, the actual electric energy is obtained. The specific calculation formula is:

[0050]

[0051] Among them, E in represents the actual electric energy; η zhq represents the conversion efficiency of the MPPT controller; P irepresents the instantaneous power at the i-th sampling point; L temp i represents the instantaneous temperature loss at the i-th sampling point; N represents the number of samplings during the sunshine time; Δt represents the sampling time.

[0052] Under sunny conditions, the sampling time and the number of samplings should be reasonably set according to the speed of change of solar radiation, so as to ensure that the solar radiation changes slowly during the sampling time, that is, the corresponding light intensity changes slowly, and the rapid changes of environmental factors such as solar irradiance and temperature can be captured; generally, the light changes slowly on sunny days, making the instantaneous power obtained by the solar panel change slowly within a certain period of time, so the sampling time is generally taken as 5 to 10 seconds.

[0053] By considering the influence of temperature loss and controller conversion efficiency on the solar energy conversion efficiency, the embodiments of the present application can accurately calculate the actual electric energy, effectively avoiding the electric energy fluctuation caused by the temperature change of the battery panel. This real-time adjustment and compensation mechanism helps to maintain the stability of the battery charging process and reduce the system performance fluctuation caused by the environmental temperature fluctuation.

[0054] Further, the battery data acquisition module obtains battery calibration parameters, battery capacity and remaining battery capacity, and real-time collects battery charging voltage, battery charging current, battery charging temperature, battery discharging voltage, battery discharging current and battery discharging temperature according to voltage sensors, current sensors and temperature sensors; the battery calibration parameters include calibrated capacity, battery internal resistance, battery nominal voltage, battery self-discharge rate, calibrated charging voltage, calibrated minimum discharging voltage, calibrated maximum charging current, calibrated highest working temperature, calibrated highest charging temperature and calibrated highest discharging temperature; the battery health, the battery capacity and the remaining battery capacity are fed back by the battery state evaluation module.

[0055] By using the battery calibration parameters and the data such as the real-time collected charging and discharging voltages, currents and temperatures, the embodiments of the present application can accurately evaluate the battery capacity, intelligently adjust the charging and discharging process, thereby improving the charging and discharging efficiency and reducing the battery self-discharge loss. In addition, by real-time temperature monitoring and calibrated temperature parameters, potential safety hazards such as overheating can be effectively prevented, ensuring the stable operation of the battery and providing data support for subsequent charging and discharging adjustment.

[0056] Further, the battery charging control module combines the battery calibration parameters, the remaining battery capacity, the battery charging voltage and the battery charging current to obtain the charging mode of the battery; the specific implementation process includes:

[0057] According to the battery calibration parameters, obtain the calibrated charging voltage; input the battery calibration parameters into a multi-layer perceptron network to obtain the battery type; obtain the historical charging data corresponding to the battery type, input the battery health and the historical charging data into a convolutional neural network to obtain a first predetermined state threshold and a second predetermined state threshold; the first predetermined state threshold is less than or equal to the second predetermined state threshold; obtain the battery charging state according to the percentage of the remaining battery capacity to the battery capacity; obtain the charging voltage ratio through the ratio of the battery charging voltage to the calibrated charging voltage; determine the charging self-discharge rate according to the calibrated self-discharge rate and the charging temperature, and the specific calculation formula is:

[0058]

[0059] where SDR cha represents the charging self-discharge rate; SDR represents the calibrated self-discharge rate at the reference temperature; T cha represents the battery charging temperature; T r represents the reference temperature, generally 25°C; k represents the influence factor of temperature on the self-discharge rate, which is obtained through a large amount of experimental data;

[0060] Refer to Figure 2 , if the battery charging state is less than or equal to the first predetermined state threshold and the charging voltage ratio is less than or equal to the predetermined voltage ratio, the battery adopts the first charging mode; if the battery charging state is less than the second predetermined state threshold and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts the second charging mode; if it satisfies that the battery charging state is greater than or equal to the second predetermined state threshold, the charging self-discharge rate is greater than the predetermined self-discharge rate, and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts the third charging mode.

[0061] Specifically, if the battery charging state is greater than the first predetermined state threshold and less than the second predetermined state threshold, then the charging voltage ratio must be greater than the predetermined voltage ratio. The first charging mode is a fast charging mode, whose purpose is to quickly supplement the battery power when the battery power is low, thereby improving the charging efficiency; the second charging mode is a stable charging mode, which can stably charge the battery when the battery power is high to prevent battery loss caused by too high a battery charging voltage; the third charging mode can continuously charge the battery with a small current when the battery is close to the full charge state, so that the battery remains fully charged. The calibration parameters of the battery can directly reflect the characteristics of different types of batteries, and the charging self-discharge rate can effectively distinguish whether the third charging mode needs to be adopted for the battery. For example, the calibrated self-discharge rate of a lithium battery is low and the influence of temperature on the self-discharge rate is small. By setting a predetermined self-discharge rate, it can be made not to adopt the third charging mode to prevent overcharging; while the calibrated self-discharge rate of a lead-acid battery is high and the influence of temperature on the self-discharge rate is large. The third charging mode can effectively make up for the energy loss of the battery caused by self-discharge. Therefore, the selection process of this charging mode does not require the identification of the battery type, and the charging mode can be automatically adjusted for different types of batteries as long as the battery calibration parameters and charging parameters are based on.

[0062] To further illustrate the functions of the different charging modes proposed by the present invention, the charging modes of several solar street lights on the roads in Area A under sunny conditions are compared here by way of example. Refer to Table 1 for the charging modes of different solar street lights; among them, the predetermined voltage ratio is 0.94; the predetermined self-discharge rate is 10%; the battery health is 100% for all.

[0063] Table 1. Charging Modes of Different Solar Street Lights on the Roads in Area A

[0064]

[0065] In the embodiments of the present application, different state thresholds are set for different types of batteries; for example, the charging state of the DC01 battery of the SL01 street lamp is 0.92, the charging voltage ratio is 0.37, and the charging self-discharge rate is 2.18%. Through the first charging mode, it can be ensured that the battery can maintain a good charging state at a higher voltage and a lower self-discharge rate; for the DC02 battery of the SL02 street lamp, although the charging voltage ratio is relatively high, the charging state is 0.76, so it is in the second charging mode, which can effectively control the battery health during the charging process and avoid battery loss caused by too high charging voltage; while the self-discharge rate and charging voltage ratio of the DC03 battery of the SL06 street lamp are both relatively high. By setting the predetermined state threshold, it can work in the third charging mode, which can effectively reduce the battery power loss caused by too high self-discharge rate, thereby improving the charging efficiency. Therefore, by learning different predetermined state thresholds through the network according to different types of batteries, the charging process can be accurately controlled, the charging efficiency can be improved, and the battery performance can be optimized.

[0066] In the embodiments of the present application, by automatically adjusting the charging mode according to the calibration parameters and charging state of different types of batteries, it can be ensured that each type of battery works under the most suitable charging conditions, avoiding the charging inconsistency problem caused by different battery characteristics, and improving the charging consistency of the entire system; in addition, the monitoring of the charging self-discharge rate further improves the accuracy and adaptability of the charging process by adjusting the charging mode, reducing the power loss caused by self-discharge.

[0067] Further, the battery charging control module also combines the charging mode and battery calibration parameters to adjust the charging current of the battery in real time according to the charging temperature; the specific implementation process includes:

[0068] According to the battery calibration parameters, obtain the battery internal resistance, calibrated maximum charging temperature, calibrated maximum operating temperature, and calibrated maximum charging current; if the charging mode is the first charging mode, charge according to the maximum charging current, and monitor the battery charging temperature in real time; if the battery charging temperature is greater than or equal to the calibrated maximum charging temperature, adjust the battery charging current through temperature limitation, and the specific calculation formula is:

[0069]

[0070] Wherein, represents the adjusted charging current; I max represents the maximum charging current; T cha represents the battery charging temperature; T work represents the calibrated maximum operating temperature; T max_cha represents the calibrated maximum charging temperature and T work >T max_cha; δ represents the influence factor of temperature on current, which is obtained from a large amount of experimental data;

[0071] If the charging mode is the second charging mode, according to the charging duration of the first charging mode, the battery charging current, and the difference in the battery charging voltage during the charging duration, the equivalent capacitance of the battery is obtained; combining the internal resistance of the battery and the equivalent capacitance of the second charging mode, the battery charging current is obtained. The specific calculation formula is:

[0072]

[0073] Wherein, represents the charging current of the second charging mode; I0 represents the initial charging current in the second charging mode; R represents the equivalent resistance of the battery; C represents the equivalent capacitance; t represents the current charging duration of the second charging mode;

[0074] If the charging mode is the third charging mode, the charging current is adjusted according to the charging self-discharge rate, the battery charging temperature, and the battery charging voltage. The specific formula is expressed as:

[0075]

[0076] Wherein, represents the charging current of the third charging mode; CA represents the current battery capacity; SDR cha represents the charging self-discharge rate; V cha represents the battery charging voltage.

[0077] In the embodiment of the present application, the battery charging temperature is monitored in real time in the first charging mode. When the temperature exceeds the calibration threshold, the current is adjusted through the formula to avoid instability caused by overheating; in the second charging mode, the equivalent capacitance and internal resistance of the battery are calculated to dynamically adjust the charging current to ensure precise control of the charging process; and in the third charging mode, the charging current is further optimized according to the self-discharge rate, charging temperature, and voltage to avoid the influence of overcharging or temperature fluctuations. The above adjustment method enables different types of batteries to be stably charged within their characteristic ranges, ensures the consistency of the charging process, and reduces the influence of temperature and current fluctuations on battery performance, thereby improving the overall stability and reliability of the street lamp system.

[0078] Furthermore, real-time light intensity data is obtained through an ambient light sensor to determine the on or off state of the light source; combining the distance between adjacent street lamps and ambient meteorological data, the second adjusted illuminance is obtained through the inverse square law formula and the weather state influence factor. The light source demand analysis module includes:

[0079] Collect the ambient light intensity data in real time through a light sensor to obtain the average light intensity. If the average light intensity is less than a predetermined light threshold, turn on the light source. Obtain the street lamp spacing and standard illuminance, and calculate the first adjusted illuminance according to the inverse square law formula. The specific formula is expressed as:

[0080]

[0081] Wherein, represents the first adjusted illuminance; represents the standard illuminance; Refer to Figure 3 , h, d respectively represent the street lamp height and the spacing between adjacent street lamps; θ represents the street lamp tilt angle and 0° < θ ≤ 90°;

[0082] Combine the environmental meteorological data and the first adjusted illuminance, obtain the influence factor of different weather conditions on the illuminance, and get the second adjusted illuminance. The specific calculation formula is:

[0083]

[0084] W t = 1 - (β1·CC + β2·RI + β3·SI + β4·HC + β5·(1 - SD));

[0085] Wherein, represents the second adjusted illuminance; W t represents the weather condition influence factor, which is used to evaluate the influence degree of different weathers such as cloud cover, precipitation, snowfall and haze on the actual illuminance; represents the average light intensity; represents the environmental light intensity corresponding to the standard illuminance; CC, RI, SI, HC, SD respectively represent the cloud cover parameter, precipitation intensity parameter, snowfall intensity parameter, haze concentration parameter and sunshine duration parameter. These weather condition parameters are obtained by normalizing the corresponding original meteorological parameters; β1, β2, β3, β4, β5 represent the weight factors of different weather condition parameters and

[0086] Specifically, when different types of battery packs are used for electrical energy storage in the street lamp system of A urban area roads, the performance and charging characteristics of each battery pack may be different. Therefore, obtaining the actual power demand of the power supply can ensure that the output power of each battery pack is the same, thus avoiding the inconsistency problems caused by the performance differences between battery packs. At the same time, obtaining the actual power demand of the power supply can also help analyze the usage and lifespan of the battery packs, timely detect and handle possible battery failures or performance degradation, thereby improving the reliability and stability of the system.

[0087] Embodiments of the present application adjust the illumination requirements of street lamps according to environmental light conditions and weather factors, reducing the additional battery load. By calculating the first adjusted illumination using the inverse square law formula and combining it with the weather impact factor obtained from meteorological data such as cloud cover, precipitation, snowfall, and haze, the second adjusted illumination can be further optimized. This can avoid over-reliance on battery power supply when the light is insufficient and reduce the discharge load of the battery. At the same time, the on-off state of the light source is dynamically adjusted according to the light conditions, avoiding frequent charging and discharging of the battery caused by illumination fluctuations, thus keeping the battery operating in a more stable state, effectively improving the charging consistency of different battery types, reducing the negative impact of environmental changes on battery life, and further enhancing the overall stability of the system.

[0088] Furthermore, the battery discharge control module dynamically selects the discharge mode according to the remaining battery capacity and the battery discharge temperature. The specific implementation process includes:

[0089] Obtain the actual power demand of the light source through the second adjusted illumination. If the remaining battery capacity is greater than the predetermined capacity threshold and the actual power demand is less than or equal to the rated maximum discharge power of the battery, adopt the first discharge mode. If the remaining battery capacity is less than or equal to the predetermined capacity threshold, adopt the second discharge mode. When adopting the first discharge mode or the second discharge mode, if the battery discharge voltage is less than the calibrated minimum discharge voltage or the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, stop discharging.

[0090] Specifically, the first discharge mode is a constant power discharge mode, whose purpose is to provide a stable light output when the battery power is relatively high. The second discharge mode is a continuous discharge mode, which can continuously discharge at a relatively small discharge power when the battery power is low to extend the lighting time.

[0091] Embodiments of the present application ensure that the discharge strategy is reasonably adjusted according to the remaining battery capacity and power demand under different discharge modes through the states and working conditions of different batteries. When the remaining battery capacity is high, the first discharge mode is adopted to ensure the normal power supply of the street lamp. When the remaining capacity is low, the system switches to the second discharge mode to extend the battery life and optimize the power output. At the same time, the discharge voltage and temperature of the battery are monitored in real time to stop discharging in time to prevent over-discharge or overheating damage to the battery. This process ensures that the system can operate efficiently under different power output conditions, effectively improving the stability and consistency of the solar street lamp system.

[0092] Furthermore, the battery discharge control module also adjusts the discharge current of different discharge modes through the non-linear mapping of the actual power demand of the light source, the remaining battery capacity, and the discharge current. The specific implementation process includes:

[0093] According to the battery calibration parameters, obtain the nominal voltage of the battery and the calibrated maximum discharge temperature; if the charging mode is the first discharge mode, obtain the first discharge current based on the actual power demand and the nominal voltage of the battery, and perform discharge; monitor the battery discharge temperature in real time. If the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, adjust the battery charging current according to the calibrated maximum discharge temperature to obtain the second discharge current. The specific calculation formula is:

[0094]

[0095] Wherein, represents the second discharge current; represents the first discharge current; T dis represents the battery discharge temperature; T max_dis represents the calibrated maximum discharge temperature; γ is the temperature coefficient, representing the change rate of the discharge current when the temperature changes by 1°C;

[0096] If the charging mode is the second discharge mode, obtain the third discharge current through non-linear mapping of the discharge current based on the first discharge current and the remaining capacity of the battery. The specific formula is expressed as:

[0097]

[0098] Wherein, represents the third discharge current; ln() represents the natural logarithm function; CA dis , CA set respectively represent the remaining capacity of the battery and the predetermined capacity threshold; μ represents the control parameter, which is obtained through non-linear fitting of a large amount of battery data, and the values of the control parameters corresponding to different types of batteries are different.

[0099] In the embodiment of the present application, the battery discharge temperature is monitored in real time, and the battery charging current is adjusted according to the calibrated maximum discharge temperature to ensure that the battery operates within the optimal temperature range, effectively preventing performance degradation or safety hazards caused by overheating. This not only improves the service life of the battery but also maintains the stable operation of the system; at the same time, the non-linear mapping model is used to adjust the discharge current according to the remaining capacity of the battery, making the discharge of the battery more stable under different remaining capacities. This not only effectively improves the stability under different battery working states but also ensures the consistency of various types of batteries during the entire discharge process.

[0100] Further, the battery state evaluation module inputs the publicly available lithium-ion battery aging dataset into the battery state evaluation model for model training; optimizes the model through transfer learning to further accurately predict the battery health, the remaining capacity ratio of the battery, and the remaining battery life after charging and discharging; refer to Figure 4, the specific implementation process of the battery state evaluation module includes:

[0101] A data acquisition unit, according to the battery data acquisition module, acquires charge and discharge historical data, including historical charge and discharge power, historical discharge power, historical charge and discharge voltage, historical charge and discharge current, historical charge and discharge temperature, and charge and discharge cycle times; combines the historical charge and discharge power and the battery capacity to obtain the historical charge depth and the historical discharge depth;

[0102] A data preprocessing unit performs preprocessing operations on the data acquired by the data acquisition unit, including outlier processing, missing value processing, time alignment operation, and normalization processing, to obtain standard historical data;

[0103] A battery state evaluation model training unit inputs the publicly available lithium-ion battery aging dataset into a convolutional neural network to obtain a charge and discharge feature sequence; according to the spatial self-attention mechanism, obtains a first enhanced feature sequence from the charge and discharge feature sequence; inputs the first enhanced feature sequence into a bidirectional convolutional long short-term memory network to obtain a second enhanced feature sequence; through a fully connected layer and an activation function, obtains the predicted battery health, battery remaining capacity ratio, and battery remaining life;

[0104] A battery state evaluation model inference unit performs data augmentation operations on the standard historical data to obtain augmented historical data; inputs the augmented historical data into a pre-trained battery state evaluation model, and adjusts the model through a transfer learning strategy to obtain the battery health, battery remaining capacity ratio, and battery remaining life after charging or discharging.

[0105] In the embodiment of the present application, this module ensures the accurate evaluation of the battery health status through comprehensive data acquisition and preprocessing; combined with the spatial self-attention mechanism, bidirectional convolutional long short-term memory network, and transfer learning strategy, it can accurately predict the battery health, remaining capacity, and remaining life. This accurate prediction improves the battery management ability, prevents overcharging and over-discharging phenomena, and provides accurate battery capacity data for subsequent charging and discharging processes, ensuring the accuracy and efficiency of different battery charging and discharging controls, so that the solar street lamp system can still maintain stable performance and consistent lighting effects under various environmental conditions.

[0106] Further, the battery state evaluation model training unit supervises the battery health and the battery remaining life predicted by the battery state evaluation model training unit according to the mean absolute loss function; constrains the battery remaining capacity ratio predicted by the battery state evaluation model training unit through the logarithmic hyperbolic cosine loss function; the specific calculation formula of the loss function is:

[0107]

[0108] Among them, L lc represents the logarithmic hyperbolic cosine loss function; MAE() represents the mean absolute loss function; SOH respectively represent the predicted battery health and the actual battery health; RUL respectively represent the predicted remaining battery life and the actual remaining battery life; n represents the number of samples; log() represents the logarithmic function; respectively represent the predicted remaining battery capacity ratio and the actual remaining battery capacity ratio of the i-th sample;

[0109] Refer to Figure 1 , input historical charge and / or discharge data into the battery state evaluation model, and the inference unit of the battery state evaluation model adjusts the weights of the battery state evaluation model through a weight update method to obtain the battery health, the battery remaining capacity ratio, and the battery remaining life after historical charge and / or discharge; according to the battery remaining capacity ratio and the battery health, obtain the current battery remaining capacity; through the battery calibration parameters and the battery health, obtain the current battery capacity; feedback the battery health, the current battery remaining capacity, and the current battery capacity to the battery charging control module to update the battery health, the battery remaining capacity, and the battery capacity.

[0110] Through the optimized regulation of the training and inference unit of the battery state evaluation model in the embodiments of the present application, the health status of different types of batteries can be accurately predicted, and combined with the battery calibration parameters, the real-time evaluation of the current battery capacity can be realized. This process ensures high precision and consistency of the prediction results and effectively constrains the deviation during the battery charging and discharging process by introducing the mean absolute loss function and the logarithmic hyperbolic cosine loss function; through the dynamic update and feedback mechanism of the battery charging control module, the stability and consistency of the solar street lamp system in various battery situations are improved.

[0111] Through the collaborative work of the battery charging control module, the discharge control module, and the battery state evaluation module in the embodiments of the present application, the efficient charge and discharge management and state prediction of different types of batteries are realized. The specific implementation process mainly includes the following steps: (1) controlling the charging current according to different battery characteristics; (2) adaptive discharge control; (3) battery health state evaluation. The present invention proposes flexible charging control, stable discharge control, and accurate health prediction for the above three processes respectively; the charging control realizes efficient charging for different battery types, avoids overcharging, and improves battery compatibility and service life; the discharge control further improves the stability of the light source output and reduces energy loss; the battery health state prediction further improves the adaptability of the system to different battery states, extends the service life of the entire system, and improves the overall reliability of the system.

[0112] Embodiment 2

[0113] In Embodiment 1, the system of the present invention realizes efficient lighting for the roads in Area A under sunny conditions. In the embodiment of this application, the system proposed by the present invention will be described again to achieve efficient lighting for the roads in Mountainous Area B under continuous cloudy conditions. The specific implementation process is as follows:

[0114] During the day, the solar energy conversion module obtains the instantaneous voltage, instantaneous current, and instantaneous temperature of the solar panel according to the voltage sensor, current sensor, and temperature sensor; obtains the instantaneous power through the instantaneous voltage and the instantaneous current; obtains the instantaneous temperature loss of the solar panel based on the instantaneous temperature; and combines the instantaneous power, instantaneous temperature loss, and the conversion efficiency of the MPPT controller to obtain the actual electrical energy.

[0115] In the embodiment of this application, due to the influence of cloud cover under continuous cloudy conditions, the solar radiation obtained by the solar street lamp system on the roads in Mountainous Area B is greatly reduced; combined with the maximum power point tracking function of the MPPT controller, the electrical energy output can be dynamically adjusted under continuous cloudy conditions to obtain the maximum conversion efficiency and actual electrical energy.

[0116] Furthermore, according to the battery calibration parameters, the calibrated charging voltage is obtained; the battery calibration parameters are input into the multi-layer perceptron network to obtain the battery type; the historical charging data corresponding to the battery type is obtained, and the battery health and the historical charging data are input into the convolutional neural network to obtain the first predetermined state threshold and the second predetermined state threshold; the battery charging state is obtained according to the percentage of the remaining battery capacity to the battery capacity; the charging voltage ratio is obtained through the ratio of the battery charging voltage to the calibrated charging voltage; the charging self-discharge rate is determined according to the calibrated self-discharge rate and the battery charging temperature; if the battery charging state is less than or equal to the first predetermined state threshold and the charging voltage ratio is less than or equal to the predetermined voltage ratio, the battery adopts the first charging mode; if the battery charging state is less than the second predetermined state threshold and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts the second charging mode; if the battery charging state is greater than or equal to the second predetermined state threshold, the charging self-discharge rate is greater than the predetermined self-discharge rate, and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts the third charging mode; and the battery charging current in the three charging modes is adjusted respectively through temperature regulation, non-linear mapping of the charging current, and power compensation.

[0117] In the embodiments of the present application, the solar street lamp system on the roads in Mountain B faces the problem of insufficient light, which leads to a decrease in the charging efficiency of the battery and a reduction in the remaining capacity, thus affecting the overall stability and lighting effect of the system. In this case, the embodiments of the present application automatically adjust the charging mode according to the characteristics of different types of battery packs: for example, in the first charging mode, the charging temperature of the battery is monitored in real time. Since the ambient temperature fluctuates greatly under cloudy conditions, the street lamp system will adjust the charging current according to the temperature limit to prevent the battery from overheating, thereby prolonging the battery life; in the second charging mode, the battery charging current is calculated by combining the battery internal resistance and equivalent capacitance to cope with the unstable battery capacity under low light conditions and avoid the problem of insufficient charging caused by too large or too small current during the charging process of the battery; in the case of continuous cloudy days, since the battery may have an increased self-discharge rate due to being in a low light environment for a long time, the charging current will be dynamically adjusted according to the battery self-discharge rate, charging temperature and charging voltage in the third charging mode to avoid power supply shortage caused by the decline of battery performance, thus ensuring the lighting consistency and system stability on the roads in Mountain B under bad weather conditions.

[0118] Further, the light source demand analysis module obtains real-time light intensity data through an ambient light sensor to determine the on or off state of the light source; combines the adjacent street lamp spacing and ambient meteorological data, and obtains the second adjusted illuminance through the inverse square law formula and the weather state influence factor.

[0119] Further, the battery discharge control module obtains the battery calibration parameters, battery capacity and remaining battery capacity, and real-time collects the battery discharge voltage, battery discharge current and battery discharge temperature according to the voltage sensor, current sensor and temperature sensor; if the remaining battery capacity is greater than the predetermined capacity threshold and the actual power demand is less than or equal to the rated maximum discharge power of the battery, the first discharge mode is adopted; if the remaining battery capacity is less than or equal to the predetermined capacity threshold, the second discharge mode is adopted; when the first discharge mode or the second discharge mode is adopted, if the battery discharge voltage is less than the calibrated minimum discharge voltage or the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, the discharge is stopped; and the discharge current of different discharge modes is adjusted through the non-linear mapping of the actual power demand of the light source, the remaining battery capacity and the discharge current.

[0120] Under continuous cloudy conditions, the actual power demand of the light source in the solar street lamp system on the roads in Mountain B increases, while the charging efficiency of the solar panels decreases, which may lead to insufficient battery energy storage. In this case, when the remaining battery capacity is relatively high, the system adopts the first discharge mode to ensure that the street lamps can still meet the lighting requirements of the road when the light is insufficient; as continuous cloudy days continue, the remaining battery capacity gradually decreases, and the system switches to the second discharge mode, taking into account both the remaining capacity and the actual power demand to extend the working time of the battery, so that the lighting requirements of the roads in Mountain B can be met with limited energy reserves.

[0121] Further, the battery state evaluation module inputs the lithium-ion battery aging public data set into the battery state evaluation model for model training; optimizes the model according to transfer learning, historical charging and / or discharging data to further accurately predict the battery health, the battery remaining capacity ratio, and the battery remaining life after battery charging and / or discharging; obtains the current battery remaining capacity according to the battery remaining capacity ratio and the battery health; obtains the current battery capacity through the battery calibration parameters and the battery health; and feeds back the current battery remaining capacity and the current battery capacity to the battery charging control module to update the battery remaining capacity and the battery capacity.

[0122] In the embodiment of the present application, the solar street lamp system on the roads in Mountain B may face the problem of low power generation efficiency of the solar photovoltaic panels, resulting in insufficient battery charging and affecting the continuous power supply ability of the street lamp system. By real-time monitoring and updating the battery health and the remaining capacity ratio, the available power of the current battery can be accurately evaluated, which helps to take preventive measures in advance when the battery health deteriorates or the remaining capacity is insufficient, such as reducing the street lamp brightness, adjusting the working duration, or intermittently turning on the light source, to extend the battery usage time. Feeding back the data of the battery remaining capacity and the actual capacity to the charging and / or discharging control module, the system precisely regulates the charging and / or discharging process according to the actual battery power, effectively preventing overcharging and over-discharging of the battery, thus ensuring the consistency and stability of the system during continuous cloudy days.

[0123] 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. An MPPT type solar charge-discharge light source integrated street lamp system, characterized in that, Including: A solar energy conversion module that collects the voltage, current, and temperature of a solar panel in real time through sensors, and obtains the actual electric energy by combining the conversion efficiency of the MPPT controller; A battery data acquisition module that obtains battery calibration parameters, battery health, battery capacity, and remaining battery capacity, and collects the battery charging voltage, battery charging current, battery charging temperature, battery discharging voltage, battery discharging current, and battery discharging temperature in real time according to voltage sensors, current sensors, and temperature sensors; A battery charging control module that obtains a charging mode according to the calibration parameters of different types of batteries; Adjust the battery charging current in different charging modes through temperature regulation, non-linear mapping of charging current, and power compensation; The battery charging control module includes: Obtain the calibrated charging voltage according to the battery calibration parameters; input the battery calibration parameters into a multi-layer perceptron network to obtain the battery type; obtain the historical charging data corresponding to the battery type, input the battery health and the historical charging data into a convolutional neural network to obtain a first predetermined state threshold and a second predetermined state threshold; obtain the battery charging state according to the percentage of the remaining battery capacity to the battery capacity; obtain the charging voltage ratio through the ratio of the battery charging voltage to the calibrated charging voltage; determine the charging self-discharge rate according to the calibrated self-discharge rate and the battery charging temperature; if the battery charging state is less than or equal to the first predetermined state threshold and the charging voltage ratio is less than or equal to the predetermined voltage ratio, the battery adopts the first charging mode; if the battery charging state is less than the second predetermined state threshold and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts the second charging mode; if the battery charging state is greater than or equal to the second predetermined state threshold, the charging self-discharge rate is greater than the predetermined self-discharge rate, and the charging voltage ratio is greater than the predetermined voltage ratio, the battery adopts the third charging mode; A light source demand analysis module that determines the on / off state of the light source according to the ambient light intensity data; combines the distance between adjacent street lamps and the ambient meteorological data, and obtains the second adjusted illuminance through the inverse square law formula and the weather state influence factor; A battery discharging control module that obtains a discharging mode according to the remaining battery capacity and the battery discharging temperature; adjusts the discharging current in different discharging modes through the actual power demand of the light source, the remaining battery capacity, and non-linear mapping of the discharging current; A battery state evaluation module that inputs the publicly available lithium-ion battery aging dataset into the battery state evaluation model for model training; optimizes the model through transfer learning to predict the battery state after charging and discharging of different types of batteries.

2. The MPPT type solar charge-discharge light source integrated street lamp system according to claim 1, characterized in that, The solar energy conversion module includes: Obtain the instantaneous voltage, instantaneous current, and instantaneous temperature of the solar panel according to voltage sensors, current sensors, and temperature sensors; obtain the instantaneous power through the instantaneous voltage and the instantaneous current; obtain the instantaneous temperature loss of the solar panel according to the instantaneous temperature; combine the instantaneous power, instantaneous temperature loss, and the conversion efficiency of the MPPT controller to obtain the actual electric energy.

3. The MPPT type solar charge-discharge light source integrated street lamp system according to claim 1, characterized in that, The battery calibration parameters include calibrated capacity, battery internal resistance, battery nominal voltage, battery self-discharge rate, calibrated charging voltage, calibrated minimum discharge voltage, calibrated maximum charging current, calibrated maximum operating temperature, calibrated maximum charging temperature, and calibrated maximum discharge temperature; the battery health, the battery capacity, and the remaining battery capacity are fed back by the battery state evaluation module.

4. The MPPT type solar charging and discharging light source integrated street lamp system according to claim 1, characterized in that, The battery charging control module further includes: According to the battery calibration parameters, obtain the battery internal resistance, calibrated maximum charging temperature, calibrated maximum operating temperature, and calibrated maximum charging current; if the charging mode is the first charging mode, charge according to the calibrated maximum charging current, and monitor the battery charging temperature in real time; if the battery charging temperature is greater than or equal to the calibrated maximum charging temperature, adjust the battery charging current through temperature limitation; if the charging mode is the second charging mode, obtain the equivalent capacitance of the battery according to the charging duration of the first charging mode, the battery charging current, and the difference in the battery charging voltage within the charging duration; combine the battery internal resistance and the equivalent capacitance of the second charging mode to adjust the battery charging current; if the charging mode is the third charging mode, adjust the charging current according to the charging self-discharge rate, the battery charging temperature, and the battery charging voltage.

5. The MPPT type solar charging and discharging light source integrated street lamp system according to claim 1, characterized in that, The light source demand analysis module includes: Collect the surrounding light intensity data in real time through a light sensor to obtain the average light intensity; if the average light intensity is less than a predetermined light threshold, turn on the light source; obtain the street lamp spacing and standard illuminance, and obtain the first adjusted illuminance according to the inverse square law formula; combine the environmental meteorological data and the first adjusted illuminance to obtain the influence factor of different weather conditions on the illuminance, and obtain the second adjusted illuminance.

6. The MPPT type solar charge-discharge light source integrated street lamp system according to claim 1, characterized in that The battery discharge control module includes: Obtain the actual power demand of the light source through the second adjusted illuminance; if the remaining battery capacity is greater than a predetermined capacity threshold and the actual power demand is less than or equal to the rated maximum discharge power of the battery, adopt the first discharge mode; if the remaining battery capacity is less than or equal to the predetermined capacity threshold, adopt the second discharge mode; when adopting the first discharge mode or the second discharge mode, if the battery discharge voltage is less than the calibrated minimum discharge voltage or the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, stop discharging.

7. An MPPT type solar charging and discharging light source integrated street lamp system according to claim 6, characterized in that, The battery discharge control module further includes: According to the battery calibration parameters, obtain the battery nominal voltage and the calibrated maximum discharge temperature; if the charging mode is the first discharge mode, obtain the first discharge current according to the actual power demand and the battery nominal voltage and discharge; monitor the battery discharge temperature in real time, if the battery discharge temperature is greater than or equal to the calibrated maximum discharge temperature, adjust the battery charging current according to the calibrated maximum discharge temperature to obtain the second discharge current; if the charging mode is the second discharge mode, obtain the third discharge current through the non-linear mapping of the discharge current according to the first discharge current and the remaining battery capacity.

8. An MPPT type solar charging and discharging light source integrated street lamp system according to claim 1, characterized in that, The specific implementation process of the battery state evaluation module includes: The data acquisition unit collects charge and discharge historical data according to the battery data acquisition module, including historical charge and discharge power, historical discharge power, historical charge and discharge voltage, historical charge and discharge current, historical charge and discharge temperature, and charge and discharge cycle times; combines the historical charge and discharge power and the battery capacity to obtain the historical charge depth and the historical discharge depth; The data preprocessing unit performs preprocessing operations on the data obtained by the data acquisition unit, including outlier processing, missing value processing, time alignment operation, and normalization processing, to obtain standard historical data; The battery state evaluation model training unit inputs the publicly available lithium-ion battery aging dataset into a convolutional neural network to obtain a charge and discharge feature sequence; obtains a first enhanced feature sequence from the charge and discharge feature sequence according to the spatial self-attention mechanism; inputs the first enhanced feature sequence into a bidirectional convolutional long short-term memory network to obtain a second enhanced feature sequence; obtains the predicted battery health, battery remaining capacity ratio, and battery remaining life through a fully connected layer and an activation function; The battery state evaluation model inference unit performs data augmentation operations on the standard historical data to obtain augmented historical data; inputs the augmented historical data into a pre-trained battery state evaluation model, and adjusts the model through a transfer learning strategy to obtain the battery health, battery remaining capacity ratio, and battery remaining life after charging or discharging.

9. The MPPT type solar charge-discharge light source integrated street lamp system according to claim 8, characterized in that, The specific implementation process of the battery state evaluation module further includes: Supervises the battery health and the battery remaining life predicted by the battery state evaluation model training unit according to the mean absolute loss function; constrains the battery remaining capacity ratio predicted by the battery state evaluation model training unit through the logarithmic hyperbolic cosine loss function; adjusts the model weights of the battery state evaluation model inference unit through a weight update method to obtain the battery health, the battery remaining capacity ratio, and the battery remaining life after historical charging or discharging; obtains the current battery remaining capacity according to the battery remaining capacity ratio and the battery health; obtains the current battery capacity through the battery calibration parameters and the battery health; feeds back the current battery remaining capacity and the current battery capacity to the battery charging control module to update the battery remaining capacity and the battery capacity.

Images