Centralized light storage and charging integrated road lighting management system and method

Abstract

The invention discloses a centralized light storage and charging integrated road lighting management system and method. The system comprises a solar power generation module, an energy storage module, a charging module, a lighting module, a central control module and a grid-connected interface module. The solar power generation module is used for storing the converted electric energy in the energy storage module; the energy storage module is used for supplying power to the charging module and the lighting module; the grid-connected interface module is connected between a municipal power grid and a system and is used for realizing bidirectional flow of electric energy; the central control module is used for collecting operation state data of each module; the central control module is also in communication connection with an environment sensor and a traffic flow sensor which are arranged beside a road, and is used for collecting environment illumination data and traffic flow data. According to the invention, efficient, convenient, stable and reliable road illumination management is realized.

Description

Technical Field

[0001] This invention relates to the field of management systems, and specifically to a centralized integrated photovoltaic, energy storage, and charging road lighting management system and method. Background Technology

[0002] Centralized integrated photovoltaic-storage-charging road lighting management integrates solar power generation, energy storage, electric vehicle charging, and municipal road lighting functions. It achieves efficient energy utilization and coordinated management of road lighting and charging services through centralized control, suitable for public scenarios such as municipal roads. Its core objective is to balance energy supply, lighting safety, and charging demand. As the core carrier for this management model, the centralized integrated photovoltaic-storage-charging road lighting management system needs to integrate multiple modules for collaborative operation. However, existing systems have significant shortcomings: a lack of dynamic lighting adjustment mechanisms based on ambient light and actual activity needs, easily leading to excessive lighting wasting energy or insufficient lighting affecting traffic safety; energy storage charging and discharging strategies do not incorporate light forecasting and electricity price fluctuations, resulting in high operating costs; power allocation lacks clear priorities, potentially causing core lighting loads to be preempted by other loads; weak fault detection and self-healing capabilities, easily causing overall service interruptions due to localized faults; control parameters rely on manual adjustment, unable to dynamically adapt to seasonal changes, traffic flow, and other scenario variations; and maintenance relies on on-site operations, resulting in low efficiency. Therefore, this paper proposes a centralized integrated photovoltaic-storage-charging road lighting management system and method. Summary of the Invention

[0003] The present invention solves the above-mentioned technical problems through the following technical solutions, the present invention comprising: Solar power generation module, energy storage module, charging module, lighting module, central control module and grid connection interface module; The output end of the solar power generation module is electrically connected to the input end of the energy storage module, which is used to store the converted electrical energy in the energy storage module; The output terminal of the energy storage module is electrically connected to the input terminals of the charging module and the lighting module, respectively, to supply power to both. The grid connection interface module is connected between the municipal power grid and the system to enable bidirectional flow of electrical energy; The central control module is communicatively connected to the solar power generation module, energy storage module, charging module, lighting module, and grid connection interface module to collect the operating status data of each module. The central control module is also communicatively connected to environmental sensors and traffic flow sensors installed along the road to collect ambient light data and traffic flow data. The central control module is configured to dynamically adjust the brightness of the lighting module based on ambient light data and traffic flow data, and simultaneously control the charging and discharging process of the energy storage module and allocate the output power of the charging module through the grid-connected interface module according to the output power of the solar power generation module and the remaining power of the energy storage module.

[0004] Furthermore, the central control module dynamically adjusts the brightness of the lighting modules based on real-time traffic flow data. This is achieved through the following calculation process: First, real-time data from traffic flow sensors is acquired and filtered to eliminate noise. Then, a brightness adjustment factor is calculated by normalizing the filtered traffic flow data and applying an exponential smoothing function. Finally, the lighting brightness is dynamically adjusted according to the brightness adjustment factor, where the formula for calculating the brightness adjustment factor is: ; in, This indicates the adjusted brightness. This indicates the preset base brightness. This represents the filtered real-time traffic flow data. This indicates the maximum traffic flow allowed by the system design. This represents the adjustment coefficient obtained based on historical data.

[0005] Furthermore, the central control module optimizes the charging and discharging strategy of the energy storage module based on ambient light prediction data. This is achieved through the following calculation process: First, weather forecast data and historical light data are acquired, and a time series analysis model is used to predict the ambient light intensity for future periods. Then, the expected value of solar power generation is calculated based on the prediction results, and combined with real-time electricity price data obtained through the grid connection interface module, an energy storage charging and discharging plan is generated. The optimization objective of the energy storage charging and discharging plan is to minimize the total system operating cost. Its objective function is obtained by weighted summation of the expected power generation, energy storage status, and electricity price data. The calculation formula is as follows: ; in, This represents the total operating cost. This represents the power purchased from the grid through the grid connection interface module at time t. This represents the electricity price at time t. This represents the charging and discharging power of the energy storage module at time t, with a positive value during charging and a negative value during discharging. The energy storage loss coefficient is obtained through regression analysis of battery cycle life data, and T represents the optimization period.

[0006] Furthermore, the central control module dynamically allocates charging power based on the load demand of the charging modules and the system's energy status. This is achieved through the following calculation process: First, it monitors the real-time load data of the charging modules, including the number of connected electric vehicles and the battery's state of charge. Then, it allocates available power based on an energy priority algorithm, with the priority order being: lighting modules first, energy storage modules second, and charging modules last. During the allocation process, the adjustment value of the charging power is obtained by proportional integration of the available power data and the load demand data. The calculation formula is as follows: ; in, This indicates the adjusted charging power value. This represents the system's available power, obtained by summing the solar power generation, the remaining energy storage power, and the power purchased through the grid-connected interface module. This indicates the power requirements of the lighting module and the base load. and The proportional and integral coefficients are determined through system simulation and experimental data calibration.

[0007] Furthermore, the central control module calculates the overall energy efficiency index of the system through an energy efficiency optimization algorithm, and adjusts the operating parameters based on this index. The specific calculation process is as follows: First, real-time energy consumption data and output data of each module are collected to calculate the system's energy efficiency ratio; then, a multivariate regression model is used to analyze the relationship between the energy efficiency ratio and the operating parameters to generate an optimization strategy; wherein, the system energy efficiency ratio is obtained by calculating the ratio of total output energy to total input energy, and the calculation formula is: ; in, Indicates the system's energy efficiency ratio. The total output energy of the lighting and charging modules is obtained by integrating the output power. The total input energy represents the energy purchased from the solar power generation module and the grid-connected interface module, which is obtained by integrating the input power. The central control module dynamically adjusts the lighting brightness curve and the energy storage charging and discharging threshold based on historical energy efficiency ratio data.

[0008] Furthermore, the central control module also includes a fault detection function, which automatically diagnoses system faults by analyzing abnormal current and voltage data. The specific calculation process is as follows: First, the current and voltage data of each module are monitored in real time, and their deviation from the rated values ​​is calculated; then, a sliding window statistical method is applied to detect continuous abnormal points, and a fault alarm is triggered when the deviation exceeds a preset threshold; wherein, the deviation value is obtained by calculating the root mean square error between the real-time data and the rated values, and the calculation formula is: ; Where D represents the deviation value. This represents the current or voltage data at the i-th sampling point. The value represents the rated value, and N represents the number of sampling points within the sliding window. The central control module performs self-healing operations based on the magnitude of the deviation value, including isolating faulty modules and switching to backup power.

[0009] Furthermore, the central control module also employs an adaptive learning algorithm to dynamically optimize control parameters. The specific calculation process is as follows: First, historical operational data is collected, including traffic flow, ambient light levels, and system energy efficiency data. Then, the control parameters are iteratively updated using gradient descent. The parameter update amount is obtained by taking the partial derivative of the historical energy consumption data and multiplying it by the learning rate. The calculation formula is as follows: ; in, and These represent the control parameters before and after the update, respectively. denoted by the learning rate, determined through cross-validation, and J representing the loss function, defined as the weighted average of the system's total energy consumption, calculated using historical energy consumption data and applied weighting coefficients.

[0010] Furthermore, the central control module also integrates remote monitoring capabilities, allowing users to view system status and adjust control strategies in real time via mobile terminals. Specifically, the central control module encrypts system operation data and uploads it to a cloud server. The cloud server then generates a user interface through data parsing and visualization processing. User adjustment commands are sent to the central control module via the cloud server, and the central control module modifies preset algorithm parameters based on the commands. The amount of parameter modification is obtained by performing fuzzy logic calculations on user input data and system security constraints.

[0011] A centralized integrated photovoltaic, energy storage, and charging road lighting management method, comprising the following steps: The central control module collects ambient light data, traffic flow data, and system operation status data. The system operation status data includes the output power of the solar power generation module, the remaining power of the energy storage module, the load demand of the charging module, and the working status of the lighting module. Based on ambient lighting data and traffic flow data, the brightness of the lighting module is dynamically adjusted through a preset intelligent control algorithm; The charging and discharging process of the energy storage module is controlled based on the output power of the solar power generation module and the remaining power of the energy storage module. The output power of the charging modules is dynamically allocated based on the overall energy status of the system. The intelligent control algorithm includes the following processing steps: The collected traffic flow data is filtered to obtain filtered traffic flow data. The brightness adjustment factor is calculated based on the filtered traffic flow data. The brightness adjustment factor is determined by comparing the filtered traffic flow data with the maximum traffic flow value allowed by the system design and combining it with the adjustment coefficient learned from historical data. Lighting control commands are generated based on the brightness adjustment factor to adjust the brightness of the lighting module.

[0012] Compared with existing technologies, this invention has the following advantages: The centralized integrated photovoltaic-storage-charging road lighting management system and method dynamically adjusts the brightness of lighting modules through a central control module combined with ambient light data and traffic flow data, meeting lighting needs in different scenarios while achieving energy conservation; it optimizes the charging and discharging strategy of energy storage modules based on ambient light prediction data and real-time electricity prices, minimizing the total system operating cost; it dynamically allocates charging power according to the priority order of lighting modules, followed by energy storage modules, and finally charging modules, ensuring stable power supply to critical loads and meeting the charging needs of electric vehicles; it calculates the system's energy efficiency ratio and adjusts operating parameters through an energy efficiency optimization algorithm, improving the overall energy utilization efficiency of the system; it has fault detection and self-healing functions, promptly diagnosing faults by monitoring current and voltage deviations and performing operations such as isolating faulty modules and switching backup power supplies, ensuring stable system operation; it uses an adaptive learning algorithm to iteratively update control parameters, making system control more precise and adaptable to actual operating scenarios; it integrates remote monitoring functions, allowing users to view system status and adjust control strategies in real time via mobile terminals, improving ease of use; and it enables bidirectional power flow between the system and the municipal power grid through a grid-connected interface module, improving renewable energy utilization and reducing dependence on the power grid in conjunction with solar power generation modules. Attached Figure Description

[0013] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0014] The embodiments of the present invention are described in detail below. These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. However, the scope of protection of the present invention is not limited to the following embodiments.

[0015] like Figure 1 As shown, this embodiment provides a technical solution: a centralized integrated photovoltaic, energy storage, and charging road lighting management system, comprising: Solar power generation module, energy storage module, charging module, lighting module, central control module and grid connection interface module; The output end of the solar power generation module is electrically connected to the input end of the energy storage module, which is used to store the converted electrical energy in the energy storage module; The output terminal of the energy storage module is electrically connected to the input terminals of the charging module and the lighting module, respectively, to supply power to both. The grid connection interface module is connected between the municipal power grid and the system to enable bidirectional flow of electrical energy; The central control module is communicatively connected to the solar power generation module, energy storage module, charging module, lighting module, and grid connection interface module to collect the operating status data of each module. The central control module is also communicatively connected to environmental sensors and traffic flow sensors installed along the road to collect ambient light data and traffic flow data. The central control module is configured to dynamically adjust the brightness of the lighting module based on ambient light data and traffic flow data, and simultaneously control the charging and discharging process of the energy storage module and allocate the output power of the charging module through the grid-connected interface module according to the output power of the solar power generation module and the remaining power of the energy storage module.

[0016] The central control module dynamically adjusts the brightness of the lighting modules based on real-time traffic flow data. This is achieved through the following calculation process: First, real-time data from traffic flow sensors is acquired and filtered to eliminate noise. Then, a brightness adjustment factor is calculated by normalizing the filtered traffic flow data and applying an exponential smoothing function. Finally, the lighting brightness is dynamically adjusted according to the brightness adjustment factor, where the formula for calculating the brightness adjustment factor is: ; in, This indicates the adjusted brightness. This indicates the preset base brightness. This represents the filtered real-time traffic flow data. This indicates the maximum traffic flow allowed by the system design. This represents the adjustment coefficient obtained based on historical data; For municipal road lighting scenarios, filtering real-time dynamic data around municipal roads can effectively eliminate noise interference in the municipal environment, ensuring that the data accurately reflects the actual activity needs of the road and avoiding misadjustment of municipal road lighting brightness due to data fluctuations. By combining the preset basic brightness of municipal road lighting, the maximum non-traffic dynamic data value of municipal roads designed by the system, and the adjustment coefficient β obtained from long-term operation data of municipal roads, the brightness adjustment factor can be accurately calculated through a formula. This enables the adaptation of municipal road lighting brightness to actual activity needs, increasing brightness when there is no heavy traffic to ensure pedestrian crossing safety, non-motorized vehicle avoidance safety, and temporary municipal operation safety, while reducing brightness when there is little traffic to reduce municipal lighting energy consumption. At the same time, the adjustment coefficient β makes the brightness adjustment conform to the activity patterns of municipal roads at different times, improving the safety and economy of lighting control.

[0017] Example of a secondary arterial road during the morning non-traffic peak hours (6:30-8:00 AM, this period is the peak time for residents to exercise in the morning, students to go to school, and commuters to work by walking / cycling. There are many pedestrians and non-motorized vehicles, and a small number of traffic assistants are on-site to guide traffic): Set the system's default base brightness. ; Maximum non-traffic dynamic data value allowed by system design for municipal roads (Based on historical statistics of this secondary arterial road during the morning hours over the past year, the maximum total activity frequency of pedestrians crossing the street, non-motorized vehicles coming and going, and traffic assistants working is taken to match the characteristics of pedestrian flow on municipal secondary arterial roads during the morning). The adjustment coefficient β = 0.65 (obtained by analyzing the correlation data of "non-traffic activity frequency - lighting brightness - pedestrian safety complaint rate" of the municipal road over the past 6 months, meaning that for every 10% increase in non-traffic activity frequency towards the maximum value, the brightness correspondingly increases by 6.5% of the base brightness, aligning with the high requirements of municipal roads for pedestrian safety); During peak hours, the millimeter-wave radar sensors (with strong anti-lighting interference capabilities and adapted to the traffic lighting environment of municipal roads) installed on the lampposts on both sides of the road collect real-time non-traffic activity frequency data of 72, 75, 73, 76, and 74 times / 15 minutes (due to the reflection of some vehicles using high beams in the morning and the swaying of roadside trees in the wind, the data has fluctuation noise of ±3). A 5-point mean filter is applied to this set of data to calculate the filtered real-time non-traffic activity frequency data. ; Substitute the parameters into the formula , can be obtained ; At this brightness level, pedestrians can clearly identify the edges of zebra crossings and gaps in road guardrails (avoiding accidental entry into motor vehicle lanes), cyclists can promptly spot small potholes (≤10cm in diameter), and traffic assistants can guide pedestrians without needing to carry additional high-intensity flashlights. Compared to a base brightness of 500cd, the risk of pedestrian collisions on this road during this period is reduced by 90%, and complaints about non-motorized vehicle bumps are reduced by 82%, fully meeting the safety needs of secondary municipal roads during the morning when there is less traffic activity.

[0018] Example of sparse non-traffic activity on municipal branch roads late at night (0:00-4:00, during which there are no regular pedestrians and non-motorized vehicles, only occasional municipal sanitation vehicles sweeping the road and road inspectors checking manhole covers / streetlight malfunctions): Using the above system parameters, i.e. (Meets the basic lighting standards for municipal branch roads) ; During the late-night hours, the millimeter-wave radar sensor collected non-traffic activity frequency data in real time at 1, 0, 2, 1, 0 times / 15 minutes (only a small amount of signals generated by sanitation vehicle operations and patrol personnel walking around; the data fluctuates slightly due to wind noise interference in the late-night environment). After 5-point mean filtering, ; Substituting into the formula, we get: ; This brightness is only slightly higher than the baseline brightness, which is sufficient for sanitation vehicle drivers to clearly see fallen leaves and gravel on the road, and also allows inspection personnel to clearly check whether manhole covers are loose and whether street light bases are intact (without needing to turn on high-intensity light equipment). At the same time, it significantly reduces municipal lighting energy consumption. It is known that this municipal branch road has 25 street lights installed with this system. The brightness and power of the lights are linearly related (according to actual measurements by the municipal department, a brightness of 500 cd corresponds to a power of 60W, and a brightness of 800.5 cd corresponds to a power of 96.1W). If a traditional municipal lighting scheme (maintaining a fixed brightness of 550 cd at night) were used... The brightness of the lights (corresponding to a power of 66W) is compared to 503.25 cd at night (corresponding to a power of 60.4W). The monthly energy savings are: (66W-60.4W)×25 lights×4 hours / day×30 days=5.6W×25×120h=16800Wh=16.8kWh. Based on the unified municipal lighting electricity price of 0.75 yuan / kWh, the monthly electricity cost savings are 12.6 yuan. Moreover, due to the long-term low power operation of the lights, the failure rate is reduced from 15% / year in the traditional solution to 5% / year, reducing the later maintenance costs and workload of the municipal department.

[0019] The central control module also optimizes the charging and discharging strategy of the energy storage module based on ambient light prediction data. This is achieved through the following calculation process: First, weather forecast data and historical light data are acquired, and a time series analysis model is used to predict the ambient light intensity for future periods. Then, the expected value of solar power generation is calculated based on the prediction results, and a charging and discharging plan for the energy storage module is generated by combining this with real-time electricity price data obtained through the grid connection interface module. The optimization objective of the energy storage charging and discharging plan is to minimize the total system operating cost. Its objective function is obtained by weighted summation of the expected power generation, energy storage status, and electricity price data. The calculation formula is as follows: ; in, This represents the total operating cost. This represents the power purchased from the grid through the grid connection interface module at time t. This represents the electricity price at time t. This represents the charging and discharging power of the energy storage module at time t, with a positive value during charging and a negative value during discharging. The energy storage loss coefficient is obtained through regression analysis of battery cycle life data, where T represents the optimization period. By combining weather forecast data with historical sunshine data and using time series analysis models to predict future ambient sunshine intensity, we can anticipate the changing trends of solar power generation, providing a basis for planning energy storage charging and discharging, and avoiding energy waste or insufficient power supply to core loads caused by blind charging and discharging. By combining real-time electricity price data obtained from the grid connection interface module to formulate energy storage charging and discharging plans, we can achieve peak-shifting electricity use, purchasing electricity from the grid to supplement energy storage during off-peak hours and prioritizing solar or energy storage power supply during peak hours, significantly reducing the cost of purchasing high-priced electricity. The objective function incorporates an energy storage loss coefficient λ obtained from battery cycle life data regression, which can accurately quantify the loss cost during the energy storage charging and discharging process, making the total operating cost calculation more realistic and avoiding cost misjudgment caused by ignoring losses. The optimized setting of the cycle T can adapt to changes in energy demand over a single day or multiple days, improving the flexibility and targeting of the charging and discharging strategy, and ultimately minimizing the total operating cost of the system while ensuring the stable operation of the municipal road lighting system.

[0020] For example, the application of photovoltaic-storage-charging systems on municipal main roads during summer weekdays (sunny days in June): Taking a solar-energy-storage-charging integrated lighting system on a main urban road as an example, the core parameters of the system are uniformly set as follows: energy storage module capacity 200kWh; energy storage loss coefficient λ obtained from regression analysis of 5000 cycles of lithium battery life data is 0.2 yuan / kWh (0.2 yuan loss cost per kWh of charging and discharging); optimization cycle T=24 hours (based on daily optimization); municipal power grid peak and off-peak electricity price standard: 1.2 yuan / kWh during peak hours (8:00-22:00) and 0.5 yuan / kWh during off-peak hours (22:00-8:00 the next day); rated power of solar power generation module 50kW; total power of lighting module 80kW (80 1kW municipal streetlights, turned on from 20:00 to 6:00 the next day, meeting the lighting standard of the main road); charging module is roadside electric vehicle charging pile, with an average daily load of 30kW (daily fluctuation range ≤5%).

[0021] Step 1: Ambient Sunlight Forecast and Calculation of Expected Solar Power Generation: Using a time series analysis model (ARIMA model), and inputting the historical sunshine data (hourly sunshine intensity and sunshine duration) for the past 30 days for this road section, along with the next day's weather forecast (clear skies), the predicted effective sunshine period for the next 24 hours is 8:00-18:00. The specific expected solar power generation for each period is as follows: 8:00-10:00 (2 hours): Early morning sunlight is weak, average power generation is 15kW, total power generation = 15kW × 2h = 30kWh; 10:00-16:00 (6 hours): Peak solar radiation at noon, average power generation 33.3kW, total power generation = 33.3kW × 6h ≈ 200kWh; 16:00-18:00 (2 hours): Evening sunlight weakens, average power generation is 35kW, total power generation = 35kW × 2h = 70kWh; 18:00-8:00 the next day (12 hours): No effective sunlight, solar power generation = 0kWh.

[0022] Step 2: Develop an energy storage charging and discharging plan (aiming to minimize total operating costs): Based on the difference between power generation forecasts and electricity prices, energy is allocated according to time periods (priority: solar energy → energy storage → grid, with priority given to supplementing energy storage during off-peak hours). The specific operating status for each time period is as follows: 0:00-1:48 (approximately 1.8 hours): Both the lighting module (80kW) and the charging module (30kW) are operating, with a total load of 80kW + 30kW = 110kW. There is no solar power supply. The energy storage module discharges at -110kW (negative discharge power). In 1.8 hours, it discharges a total of 110kW × 1.8h = 198kWh (close to the energy storage capacity of 200kWh, to avoid over-discharge). During this period, no electricity needs to be purchased from the grid. .

[0023] 1:48-6:00 (approximately 4.2 hours): Energy storage is nearly depleted, total load remains at 110kW, no solar power, switch to purchasing electricity during off-peak hours, purchased power... =110kW, total electricity purchased in 4.2 hours =110kW×4.2h=462kWh, electricity purchase cost is calculated based on off-peak electricity price of 0.5 yuan / kWh.

[0024] 6:00-8:00 (2 hours): Lighting module off, total load only includes charging module 30kW, no solar power. To avoid high electricity prices during peak hours, electricity is purchased from the grid during off-peak hours: part for charging load (30kW), and the other part for charging energy storage (170kW, requiring an additional 200kWh to reach full capacity). Therefore, the purchased power... =30kW+170kW=200kW, total purchased electricity in 2 hours =200kW×2h=400kWh, energy storage is charged at +170kW power (charging power is positive), and after 2 hours the energy storage is restored to full capacity of 200kWh.

[0025] 8:00-10:00 (2 hours): Average solar power generation is 15kW, total load is 30kW. Solar power is insufficient to cover the load; the difference = 30kW - 15kW = 15kW. Electricity will be purchased from the grid during peak hours. =15kW, total electricity purchased in 2 hours =15kW×2h=30kWh, electricity purchase cost is calculated based on peak electricity price of 1.2 yuan / kWh, energy storage has no charging or discharging (P_battery(t)=0).

[0026] 10:00-16:00 (6 hours): Average solar power generation is 33.3kW, total load is 30kW, solar surplus = 33.3kW - 30kW = 3.3kW, surplus electricity is used to charge energy storage, charging power... =+3.3kW, total charging capacity in 6 hours = 3.3kW × 6h = 19.8kWh. Since the energy storage is already at full capacity (200kWh), it is only charged to full capacity, and the excess 19.8kWh is discarded. No electricity needs to be purchased during this stage. =0).

[0027] 16:00-18:00 (2 hours): Average solar power generation is 35kW, total load is 30kW, solar surplus = 35kW - 30kW = 5kW. Since energy storage is still at full capacity, the surplus of 5kW × 2h = 10kWh of electricity is wasted. Energy storage has no charging or discharging (P_battery(t) = 0), so no electricity purchase is required. =0).

[0028] 18:00-20:00 (2 hours): Lighting module not turned on, total load 30kW, no solar power, energy storage discharges at -30kW to supply the load, total discharge in 2 hours = 30kW × 2h = 60kWh, remaining energy storage capacity = 200kWh - 60kWh = 140kWh, no need to purchase electricity. =0).

[0029] 20:00-22:00 (2 hours): Lighting module is on, total load = 80kW + 30kW = 110kW, no solar power, energy storage discharges at -110kW power: for the first 1.27 hours (140kWh ÷ 110kW ≈ 1.27h) the energy storage is used to supply power, the discharge amount is 140kWh; for the next 0.73 hours the energy storage is emptied, and electricity is purchased from the grid during peak hours, the purchased power P_grid(t) = 110kW, the purchased electricity amount = 110kW × 0.73h ≈ 80.3kWh, the electricity purchase cost is calculated at 1.2 yuan / kWh.

[0030] 22:00-0:00 the next day (2 hours): Total load 110kW, no solar power, electricity price is low, electricity is purchased from the grid: part to power the load (110kW), and the other part to charge the energy storage (100kW, requiring a supplement of 200kWh), therefore the purchased power is... =110kW+100kW=210kW, total purchased electricity in 2 hours =210kW×2h=420kWh, energy storage is charged at +100kW power, and after 2 hours the energy storage is restored to full capacity of 200kWh.

[0031] Step 3: Calculate the total operating cost ; According to the objective function formula Substitute the data from each time period into the calculation: Off-peak electricity purchase cost: Total electricity purchased during off-peak hours = 462kWh (1:48-6:00) + 400kWh (6:00-8:00) + 420kWh (22:00-0:00 the next day) = 1282kWh, cost = 1282kWh × 0.5 yuan / kWh = 641 yuan; Peak electricity purchase cost: Total electricity purchased during peak hours = 30kWh (8:00-10:00) + 80.3kWh (20:00-22:00) = 110.3kWh, cost = 110.3kWh × 1.2 yuan / kWh ≈ 132.36 yuan; Energy storage loss cost: First calculate the sum of the absolute values ​​of energy storage charging and discharging power in each time period = 198kWh (discharging from 0:00 to 1:48) + 340kWh (charging from 6:00 to 8:00, 170kW×2h) + 19.8kWh (charging from 10:00 to 16:00) + 60kWh (discharging from 18:00 to 20:00) + 140kWh (discharging from 20:00 to 21:16) + 200kWh (charging from 22:00 to 0:00 the next day, 100kW×2h) = 957.8kWh, loss cost = 957.8kWh×0.2 ​​yuan / kWh≈191.56 yuan; Total operating costs: Yuan.

[0032] Compared to traditional methods (no light prediction, random charging and discharging): Traditional solutions do not perform solar illumination forecasting and directly purchase electricity from the grid to supply the entire load during peak hours (8:00-22:00), resulting in a total daily operating cost of approximately 1160 yuan. This solution, through solar illumination forecasting and electricity price optimization, saves approximately 195.08 yuan per day (1160 yuan - 964.92 yuan ≈ 195.08 yuan). Based on a 30-day month, this translates to monthly cost savings of approximately 5852.4 yuan, significantly reducing the operating expenses of municipal road lighting systems.

[0033] Applications of photovoltaic-storage-charging systems on main municipal roads during winter weekdays (mostly cloudy in December): Using the same parameters for the main municipal road system, and considering the shorter daylight hours and weaker sunlight intensity in winter, the specific operating conditions are adjusted as follows: Sunlight forecast results: Using the ARIMA model, inputting historical data and the next day's weather forecast (cloudy), the predicted effective sunshine period for the next 24 hours is only 10:00-16:00 (6 hours). The expected solar power generation for each period is as follows: 10:00-12:00 (2 hours) average power generation 10kW, total 20kWh; 12:00-14:00 (2 hours) average power generation 25kW, total 50kWh; 14:00-16:00 (2 hours) average power generation 15kW, total 30kWh; power generation for the remaining periods is 0kWh.

[0034] Adjustments to charging and discharging plans: Due to reduced solar power generation, it is necessary to increase electricity purchases during off-peak hours to supplement energy storage, and appropriately increase electricity purchases during peak hours to ensure load protection. 0:00-1:48: Energy storage discharges 198kWh to supply a 110kW load. ; 1:48-6:00: Purchased 462kWh of electricity during off-peak hours to supply the load. =110kW; 6:00-8:00: Purchase 400kWh of electricity during off-peak hours (30kW for load supply + 170kW for energy storage). =200kW; 8:00-10:00: No solar power available; peak electricity purchase of 60kWh (30kW x 2h) to supply the load. =30kW; 10:00-12:00: Solar power 10kW, load 30kW, peak electricity purchase 40kWh (20kW x 2h), difference to be paid. =20kW; 12:00-14:00: Solar power 25kW, load 30kW, peak electricity purchase of 10kWh (5kW×2h) with payment difference. =5kW; 14:00-16:00: Solar power 15kW, load 30kW, peak electricity purchase of 30kWh (15kW×2h), difference to be paid. =15kW; 16:00-20:00: No solar power; 120kWh (30kW×4h) of stored energy is discharged to supply the load. =0; 20:00-22:00: Energy storage discharges 80kWh (0.73h), followed by peak electricity purchase of 140kWh to supply load in the next 1.27h. =110kW; 22:00-0:00 the next day: Purchase 420kWh of electricity during off-peak hours (110kW for load supply + 100kW for energy storage). =210kW.

[0035] Total operating cost calculation: Off-peak electricity purchase cost = (462 + 400 + 420) kWh × 0.5 yuan / kWh = 641 yuan; Peak electricity purchase cost = (60+40+10+30+140) kWh × 1.2 yuan / kWh = 280 kWh × 1.2 yuan / kWh = 336 yuan; Energy storage loss cost = (198 + 340 + 120 + 80 + 200) kWh × 0.2 yuan / kWh = 938 kWh × 0.2 yuan / kWh = 187.6 yuan; Total operating costs Yuan.

[0036] Compared to traditional solutions: The traditional solution has a total daily operating cost of approximately 1350 yuan in winter. After optimization through light prediction, this solution saves approximately 185.4 yuan per day (1350 yuan - 1164.6 yuan). Even in winter when sunlight is insufficient, it can still effectively reduce operating costs, demonstrating the adaptability and economy of the solution.

[0037] The central control module also dynamically allocates charging power based on the load demand of the charging modules and the system energy status. This is achieved through the following calculation process: First, it monitors the real-time load data of the charging modules, including the number of connected electric vehicles and the battery state of charge. Then, it allocates available power based on an energy priority algorithm, with the priority order being: lighting modules first, energy storage modules second, and charging modules last. During the allocation process, the adjustment value of the charging power is obtained by proportional integration of the available power data and the load demand data. The calculation formula is as follows: ; in, This indicates the adjusted charging power value. This represents the system's available power, obtained by summing the solar power generation, the remaining energy storage power, and the power purchased through the grid-connected interface module. This indicates the power requirements of the lighting module and the base load. and The proportional and integral coefficients are determined through system simulation and experimental data calibration. By monitoring the charging module load data in real time and determining the available power based on the system's energy status, energy is allocated according to the priority order of "lighting modules first, energy storage modules second, and charging modules last." This prioritizes ensuring a stable power supply for the core public safety load of municipal road lighting and meeting the energy reserve needs of energy storage modules, preventing charging loads from monopolizing critical energy and causing lighting interruptions or insufficient energy storage. At the same time, by calculating the charging power adjustment value through the proportional-integral (PI) algorithm, the available power and charging load demand can be accurately and smoothly matched, reducing power allocation fluctuations. This avoids energy waste (full charging when energy is sufficient) and prevents energy overload (reasonable charging limitation when energy is scarce), achieving an efficient balance between public lighting safety, energy storage reserves, and electric vehicle charging services, and improving the orderliness and practicality of the system's energy utilization.

[0038] For example, during midday (12:00-13:00, when sunlight and energy are plentiful), the photovoltaic-storage-charging system can be applied to main municipal roads. Unified municipal main road system parameters: Solar power generation module rated power 50kW (peak midday sunlight, actual output 33.3kW); Energy storage module capacity 200kWh (currently full capacity, remaining power 0kW, no need for charging and discharging); The grid-connected interface module operates during the midday peak electricity price (1.2 yuan / kWh, prioritizing solar power, with a purchased power capacity of 0kW); the lighting module has a total power of 80kW (off at midday, with a load of 0kW). The charging module is currently connected to 6 electric vehicles (3 with 30% charge requiring fast charging, and 3 with 70% charge requiring slow charging). The total load requirement is... =40kW); proportional-integral coefficients Kp=0.8, Ki=0.2 (calibrated through system simulation (energy distribution model built with Simulink) and 1 month of experimental data to ensure no overshoot in power adjustment and a response speed ≤10s).

[0039] Collect system energy status and load demand data: Available power of computing system According to the definition, =Solar power generation + surplus energy storage power + grid-connected power generation Substituting the data, we get =33.3kW (solar energy) + 0kW (energy storage at full capacity, no remaining available power) + 0kW (sufficient solar energy at midday, no need to purchase electricity) = 33.3kW; Determine basic load requirements Based on priority, the lighting module takes the lead (off at midday, load 0kW), followed by the energy storage module (full capacity, no need for replenishment, load 0kW). Therefore, the basic load P_load = lighting load + energy storage replenishment load = 0kW + 0kW = 0kW. Charging module load demand: According to the monitoring of the charging module controller, the total demand of 6 electric vehicles =40kW (of which fast charging requires 25kW and slow charging requires 15kW).

[0040] Step 2: Allocate available power according to energy priority: because =33.3kW> =0kW, the remaining available power can be allocated to the charging module, but the adjustment value needs to be calculated through the PI algorithm to ensure stable power matching (avoiding current fluctuations in the charging module caused by directly supplying power at 33.3kW).

[0041] Calculate the charging power adjustment value using the proportional-integral formula. : According to the formula: ; Where the integration interval t = 1 hour (the current calculation cycle is 1 hour to ensure consistency with the energy data acquisition frequency), substitute the data for calculation: Calculation of the proportional term: =0.8×(33.3kW-0kW)=26.64kW; Integral term calculation: (The unit of the integral result is energy. Since the calculation period is fixed at 1 hour, it can be simplified to the power integral value of 33.3kW.) Therefore, Ki × integral result = 0.2 × 33.3kW = 6.66kW; Charging power adjustment value: =26.64kW+6.66kW=33.3kW (consistent with available power, no overshoot).

[0042] Actual charging operation and effect The system supplies power to the charging module at 33.3kW. The charging controller allocates power according to the electric vehicle's state of charge: 25kW for fast-charging vehicles (meeting core charging needs) and 8.3kW for slow-charging vehicles (the remaining power is reasonably allocated, and the charging efficiency of slow-charging vehicles is not affected). Compared with the traditional "fixed charging power of 30kW" solution, this solution makes full use of the surplus energy at noon (33.3kW vs. traditional 30kW), charging an additional 3.3kWh per hour. The average charging time for 6 electric vehicles is reduced by 15%, and lighting and energy storage are not affected, achieving efficient energy utilization.

[0043] Application of photovoltaic-storage-charging systems on municipal main roads at night (9:00 PM - 10:00 PM, no solar energy, energy shortage): Using the same system parameters: Solar power generation 0kW (no sunlight at night); Energy storage module current remaining capacity 80kWh (remaining power -80kW, currently discharging at 80kW to supply lighting); Grid connection interface module nighttime peak electricity price (1.2 yuan / kWh, to control costs, purchased power 0kW); Lighting module on, load... =80kW; The charging module is currently connected to 4 electric vehicles (2 with 20% charge requiring fast charging, and 2 with 60% charge requiring slow charging, total load requirement is...). =30kW); the proportional-integral coefficients remain Kp=0.8 and Ki=0.2.

[0044] Collect system energy status and load demand data Available power of computing system : =Solar power generation + Remaining energy storage power + Grid-connected power = 0kW (no solar power) + (-80kW) (energy storage discharge for lighting, no remaining available power) + 0kW (no power purchased) = -80kW (the negative sign indicates that the current energy is only enough to maintain the basic load, with no surplus). Determine basic load requirements Prioritized by order, the lighting module takes first place (80kW load), followed by the energy storage module (current remaining power is 80kWh, enough for only 1 hour of lighting, no additional replenishment required, 0kW load). =80kW + 0kW = 80kW; Charging module load requirement: The total load requirement is monitored to be P_load_charge = 30kW.

[0045] Allocate available power according to energy priority: because =-80kW< =80kW, the current energy is only enough to maintain lighting, and there is no surplus power to be allocated to the charging module. It is necessary to calculate the adjustment value through the PI algorithm to reasonably limit the charging power (to avoid charging taking away lighting energy).

[0046] Calculate the charging power adjustment value using the proportional-integral formula. : Substitute into the formula The integration interval is t = 1 hour. Calculation of the proportional term: =0.8×(-80kW-80kW)=0.8×(-160kW)=-128kW (The negative sign indicates that there is no available power and charging needs to be limited). Integral term calculation: Therefore, the integral result of Ki × = 0.2 × (-160kW) = -32kW; Charging power adjustment value: =-128kW+(-32kW)=-160kW (The actual charging power cannot be negative. The system limits the charging power to 0kW according to the "minimum guarantee" principle, only keeping the charging module in standby mode and not charging).

[0047] Actual charging operation and results: The system limits the charging power to 0kW, prioritizing a stable 80kW power supply to the lighting modules (the energy storage continues to discharge at 80kW, with 0kWh remaining after 1 hour, after which it switches to grid-connected power purchase). Compared to the traditional "fixed charging power of 20kW" solution (which forces charging despite energy shortages, resulting in a drop in lighting power to 60kW, failing to meet municipal lighting standards), this solution dynamically allocates power to avoid insufficient lighting brightness, ensuring road safety at night. After the energy storage replenishes energy in the early morning (e.g., during off-peak electricity prices after 10 PM), the charging power is restored (20kW charging can be allocated between 2 AM and 4 AM), achieving a balance between safety and service.

[0048] The central control module also calculates the overall system energy efficiency index using an energy efficiency optimization algorithm and adjusts operating parameters based on this index. The specific calculation process is as follows: First, real-time energy consumption and output data from each module are collected to calculate the system's energy efficiency ratio (EER). Then, a multivariate regression model is used to analyze the relationship between the EER and operating parameters to generate optimization strategies. The system EER is calculated by dividing the total output energy by the total input energy, using the following formula: ; in, Indicates the system's energy efficiency ratio. The total output energy of the lighting and charging modules is obtained by integrating the output power. The total input energy, representing the energy purchased from the solar power generation module and the grid-connected interface module, is obtained by integrating the input power. The central control module dynamically adjusts the lighting brightness curve and the energy storage charging and discharging threshold based on historical energy efficiency ratio data. By collecting real-time energy consumption and output data from each module, the system's energy efficiency ratio (η) is calculated as the ratio of total output energy to total input energy. This allows for a direct quantification of the system's energy utilization efficiency, avoiding blind operation and energy waste due to the inability to assess energy efficiency. Utilizing a multivariate regression model to analyze the correlation between the energy efficiency ratio and operating parameters accurately identifies key factors affecting energy efficiency, preventing unfounded parameter adjustments. Dynamically optimizing the lighting brightness curve and energy storage charging / discharging thresholds based on historical energy efficiency ratio data reduces ineffective energy consumption while ensuring the safety of municipal road lighting and the charging needs of electric vehicles. This leads to long-term improvements in system energy utilization efficiency, reduced operating costs, and optimization strategies that align with the system's actual operating patterns, resulting in greater stability and economic efficiency.

[0049] For example, the application of photovoltaic-storage-charging systems on municipal main roads during sunny midday hours (12:00-13:00, energy is plentiful, initial energy efficiency needs optimization). Unified system parameters Solar power generation module: rated power 5kW, actual output power during midday peak sunlight. ; Energy storage module: 20kWh capacity, currently at full capacity (remaining capacity 20kWh, no charging or discharging). ); Grid connection interface module: Peak midday electricity price (1.2 yuan / kWh, priority given to solar power, purchased power capacity) ); Lighting module: Total power 8kW (80 100W municipal streetlights, turned off at midday) ); Charging module: Currently connected to 2 electric vehicles (30% charge state, total load requirement) ); Initial operating parameters: The lighting brightness curve is "fixed at 500 cd (power 8kW) from 20:00 at night to 6:00 the next day", and the energy storage charging and discharging threshold is 20% (forced charging when the remaining power is ≤2kWh).

[0050] Calculate the initial energy efficiency ratio η using collected data: According to the formula ,in: Total output energy Lighting module output (off at midday, 0kW × 1h = 0kWh) + charging module output (3kW × 1h = 3kWh), i.e. ; Total input energy Solar power input (3.3kW × 1h = 3.3kWh) + grid-connected electricity purchase input (0kW × 1h = 0kWh), i.e. ; Initial energy efficiency ratio: (There is 3.3-3=0.3kWh of solar power wasted, and the energy efficiency is not optimal).

[0051] Step 2: Analyze the factors affecting energy efficiency using a multivariate regression model; Collect system runtime data from the past 7 days during midday (12:00-13:00) and set variables: Independent variable: Solar power (Unit: kW) Charging power (Unit: kW), Energy storage charge and discharge threshold (unit:%); Dependent variable: Energy efficiency ratio η.

[0052] The fitted model was obtained through multivariate regression analysis: ; Analysis conclusion: Currently (The threshold is too low, which can easily lead to additional losses due to charging when the battery is low), and the solar energy surplus at midday is 0.3 kWh. If the energy storage threshold is increased and the surplus of charging power is fine-tuned, η can be improved.

[0053] Step 3: Dynamically adjust operating parameters: Based on the regression analysis results, optimize the parameters: The energy storage charge / discharge threshold has been increased from 20% (2kWh) to 40% (8kWh) to avoid low-energy charging losses; The charging power has been adjusted from 3kW to 3.3kW (utilizing the midday solar surplus, with no additional energy consumption).

[0054] Step 4: Calculate the optimized energy efficiency ratio and its effect: Optimized data for the first hour (12:00-13:00): ; ; Optimized energy efficiency ratio: (No solar power wasted, energy efficiency improved by 9.9%).

[0055] Long-term effect: Based on 1 hour of electricity consumption per day at noon, the monthly (30 days) electricity waste can be reduced by 0.3kWh×30=9kWh, which is equivalent to 9kWh×1.2 yuan / kWh=10.8 yuan in electricity costs.

[0056] Application of photovoltaic-storage-charging systems on municipal main roads during cloudy evenings (6:00 PM - 7:00 PM, when there is no solar energy and energy is scarce): Unified system parameters: Solar power generation module: No sunlight in the evening, ; Energy storage module: Current remaining power 6kWh (discharge power) (for partial load) Grid connection interface module: Evening peak electricity price (1.2 yuan / kWh, purchased power) ); Lighting module: On, initial power 8kW (500cd); Charging module: Connects to one electric vehicle (20% state of charge, requirement) ).

[0057] Calculate the initial energy efficiency ratio η using collected data: Calculated based on a 1-hour cycle: Total output energy Lighting output (8kW×1h=8kWh) + charging output (3kW×1h=3kWh), i.e. ; Total input energy : Solar input (0) + Grid-connected power purchase input (7kW×1h=7kWh) (4kWh of energy storage discharge is historical input and is not included in the current calculation) ),Right now ; Initial energy efficiency ratio: (The lighting power is fixed at 8kW, but the actual number of pedestrians decreases in the evening, resulting in excessive lighting).

[0058] Step 2: Multivariate Regression Model Analysis Collect data from the past 7 evenings (18:00-19:00), and set the independent variable as: lighting power. (kW), purchased power (kW), remaining energy storage capacity (kWh), dependent variable η, fitted model: ; Analysis conclusion: Currently (Lighting power is too high) Pedestrian / vehicle traffic decreases in the evening. If the lighting brightness is reduced to 400 cd (power 6.4 kW), If η decreases, then η can increase.

[0059] Dynamically adjust the lighting brightness curve: Optimize the lighting brightness curve: "400 cd (6.4 kW) from 18:00 to 20:00 in the evening, 500 cd (8 kW) from 20:00 to 22:00 at night, and 350 cd (5.6 kW) from 22:00 to 6:00 the next day".

[0060] Step 4: Calculate the optimized energy efficiency ratio and its effect: Optimized data for the first hour (18:00-19:00): The lighting meets the needs of evening traffic, and charging is unaffected. (Purchased power = Total load power - Energy storage discharge, i.e. 9.4 - 4 = 5.4kW); Optimized energy efficiency ratio: (10.8% improvement over the initial value).

[0061] Long-term effects: One hour of lighting per day in the evening reduces electricity purchases by 7-5.4=1.6kWh per month, resulting in monthly electricity savings of 1.6×30×1.2=57.6 yuan. At the same time, the lighting brightness meets actual needs and poses no safety hazards.

[0062] The central control module also includes a fault detection function, which automatically diagnoses system faults by analyzing abnormal current and voltage data. The specific calculation process is as follows: First, it monitors the current and voltage data of each module in real time and calculates their deviation from the rated values. Then, it applies a sliding window statistical method to detect continuous abnormal points, triggering a fault alarm when the deviation exceeds a preset threshold. The deviation value is calculated by performing a root mean square error calculation on the real-time data and the rated values. The calculation formula is as follows: ; Where D represents the deviation value. This represents the current or voltage data at the i-th sampling point. The value represents the rated value, and N represents the number of sampling points within the sliding window. The central control module performs self-healing operations based on the magnitude of the deviation value, including isolating faulty modules and switching to backup power. By monitoring the current and voltage data of each module in real time and calculating the deviation value D using the root mean square error formula, the deviation between the actual operating data and the rated value can be accurately quantified, avoiding misjudgment based solely on single data fluctuations. Combined with the sliding window statistical method to detect continuous anomalies, false alarms caused by accidental factors such as instantaneous electromagnetic interference and sporadic sensor errors can be eliminated, improving the accuracy of fault identification. When the deviation exceeds a preset threshold, a fault alarm can be triggered promptly, and a self-healing operation can be automatically performed, effectively curbing the spread of the fault while ensuring uninterrupted access to core public services such as municipal road lighting and electric vehicle charging. This reduces road safety risks and economic losses caused by faults, significantly improving the stability and reliability of system operation.

[0063] For example, a single lamp short-circuit fault in a municipal road lighting module (21:00-21:05 at night, during peak lighting hours): Unified system basic parameters: Lighting module: 80 1kW municipal LED streetlights (total power 80kW), single lamp rated voltage Rated current of a single lamp ; Fault detection parameters: Sliding window sampling number N=5 (sampled once every 10 seconds, covering 50 seconds of continuous data), short circuit fault deviation threshold. (Based on system security testing settings, exceeding this value is considered a fault). Self-healing configuration: Roadside backup emergency lighting battery (capacity 5kWh, output power 3kW, can simultaneously power 3 street lights in the faulty section, maintaining a brightness of 500cd, in line with municipal lighting standards).

[0064] Real-time monitoring and data collection: The central control module collects real-time current data from the target street light (the 15th light) via a current sensor: Before the fault (21:00:00-21:00:30): Sampled values (Close to the rated value, data is normal); Fault occurred (21:00:30, short circuit in the street light's internal wiring): Subsequent sampled values (Sudden increase in current, abnormalities at 5 consecutive sampling points).

[0065] Calculate the deviation value D (root mean square error): Take 5 consecutive sampling points of anomalies within the sliding window ( ), substitute into the deviation formula: ; in, This is the real-time current sampling value. N=5, the calculation process is as follows: Calculate the square of the difference between each sampling point and the nominal value: ; Summing and taking the average: ; Calculate the deviation value: .

[0066] Fault diagnosis and self-healing operation: Fault diagnosis: The central control module immediately triggers an audible and visual alarm (pushing a notification to the maintenance personnel's mobile app, marking the fault location as "15th street light"); Self-healing operation: 1. Disconnect the independent power supply circuit of the 15th street light (isolate the short circuit fault to prevent the fault from spreading to the distribution box and causing the entire section of 20 street lights to trip); 2. Start the backup emergency battery to supply power to the 3 street lights in the section where the 15th street light is located (street lights 14-16) and maintain a brightness of 500cd.

[0067] Comparison of troubleshooting effectiveness: Traditional solution: A short circuit in a single lamp will cause the circuit breaker in the distribution box to trip, turning off all 20 streetlights in the section. Maintenance personnel need to investigate on-site (averaging 30 minutes). During this time, the dark road section is prone to pedestrian falls and vehicle collisions. This solution isolates only one faulty light, the backup power supply starts within 10 seconds, the road lighting remains uninterrupted, maintenance personnel can accurately locate the fault based on the alarm and complete the repair within 15 minutes, with no safety accidents or service interruption losses.

[0068] Energy storage module discharge overcurrent fault (18:30-18:35, during the dual load period of lighting and charging): Unified system basic parameters: Energy storage module: Lithium battery pack (capacity 200kWh, rated discharge voltage) Rated discharge current (corresponding to a discharge power of 100kW). Fault detection parameters: Sliding window sampling number N=4 (sampled once every 5 seconds, covering 20 seconds of continuous data), overcurrent fault deviation threshold. ; Self-healing configuration: Grid connection interface module (automatically switches to municipal power grid to ensure power supply to the load, with a power purchase response time of ≤15 seconds); Current load: Lighting module on (40kW, evening brightness 400cd) + charging module (20kW, fast charging for 2 electric vehicles), total load 60kW.

[0069] Real-time monitoring and data collection: The central control module collects the discharge current of the energy storage module through a DC current sensor. Normal discharge (18:30:00-18:30:20): Sampled values (Close to rated value, matching 60kW load requirements); Fault occurred (18:30:20, a single battery cell in the energy storage module short-circuited): Subsequent sampled values (Current far exceeds rated value, risk of overcurrent).

[0070] Calculate the deviation value D (root mean square error): Take 4 consecutive abnormal sampling points within the sliding window ( ), substitute into the deviation formula: ; in, For real-time discharge current, N=4, the calculation process is as follows: Squared difference: ; average value: ; Deviation value: .

[0071] Fault diagnosis and self-healing operation: Fault diagnosis: This immediately triggers a fault alarm (pushing a notification to maintenance personnel, indicating "energy storage overcurrent"); Self-healing operation: 1. Disconnect the discharge circuit of the energy storage module (isolate the fault through the DC contactor to avoid overcurrent causing the battery pack to burn out or catch fire); 2. The grid connection interface module automatically switches to the municipal power grid to purchase electricity, supplying 60kW of power to the lighting (40kW) + charging (20kW) loads to maintain normal operation.

[0072] Comparison of troubleshooting effectiveness: Traditional solution: Overcurrent in energy storage can lead to thermal runaway of the battery pack (80% probability of burnout), and repair requires 24 hours. During this period, lighting and charging services are completely interrupted (charging service loses an average of 500 yuan per day, and the interruption of lighting causes traffic hazards). In this case: fault isolation and grid connection were completed within 15 seconds, the load power supply was uninterrupted, the charging service was normal, the lighting brightness was stable, only the faulty battery cell needed to be replaced (repaired in 2 hours), and there were no major safety accidents or economic losses.

[0073] The central control module also employs an adaptive learning algorithm to dynamically optimize control parameters. The specific calculation process is as follows: First, historical operational data is collected, including traffic flow, ambient light levels, and system energy efficiency data. Then, the control parameters are iteratively updated using gradient descent. The parameter update amount is obtained by taking the partial derivative of the historical energy consumption data and multiplying it by the learning rate. The calculation formula is as follows: ; in, and These represent the control parameters before and after the update, respectively. The learning rate is determined through cross-validation, and J represents the loss function, which is defined as the weighted average of the total energy consumption of the system and is calculated by applying weight coefficients to historical energy consumption data. By collecting historical operational data such as traffic flow, ambient light, and system energy efficiency in municipal road scenarios, the gradient descent method is used to iteratively update control parameters (such as brightness adjustment coefficient and energy storage charging and discharging threshold). The parameter update amount is calculated based on the partial derivative of the loss function (weighted average of total system energy consumption) and the learning rate determined by cross-validation. This allows the control parameters to dynamically adapt to actual operating conditions such as seasonal changes, traffic flow fluctuations, and differences in light intensity, avoiding the lag and subjectivity of manual parameter adjustments and improving the accuracy of system control. At the same time, the adaptive learning algorithm can continuously optimize parameters, reduce ineffective energy consumption caused by parameter mismatch (such as excessive lighting and inefficient charging and discharging of energy storage), ensure the safety of municipal road lighting and the stable satisfaction of electric vehicle charging needs, improve system operating efficiency and economy in the long term, and reduce the workload of parameter adjustment for maintenance personnel.

[0074] Optimize brightness adjustment factor β for summer weekdays (sufficient sunlight, stable traffic flow): Target optimization parameters: Brightness adjustment coefficient θ=β, corresponding to the core parameters of brightness adjustment, initial parameters. Based on the initial experimental data; The learning rate γ = 0.003 was determined through 5 sets of cross-validation to avoid overshooting due to excessively rapid parameter updates or slow convergence. Loss function J: Defined as the weighted average of the system's total daily energy consumption, with weighting coefficients allocated according to: daily energy consumption 0.6, energy storage loss 0.3, and charging energy consumption 0.1. ; Historical data: Collects data on daily traffic flow over the past 30 summer workdays. Vehicles per minute (average from 8:00 to 22:00), ambient light intensity averages 80,000 lux at midday, system energy consumption data is daily average lighting energy consumption. Energy storage loss Charging energy consumption ; Other related parameters: Base brightness The system is designed for maximum traffic flow. Vehicles per minute.

[0075] Calculate the loss function J and its partial derivatives. : Initial loss function J calculation: Substituting historical energy consumption data, J = 0.6 × 640 + 0.3 × 80 + 0.1 × 200 = 384 + 24 + 20 = 428, where the unit is weighted energy consumption value, and the energy consumption is uniformly expressed in kWh; Partial derivatives Calculation: β indirectly affects lighting energy consumption by influencing lighting brightness; a reasonable relationship is obtained by combining historical data. .

[0076] When β=0.6 It perfectly matches historical data; when β=0.5 , ; When β=0.7 , J=0.6×660+0.3×80+0.1×200=396+24+20=440.

[0077] Taking Δβ = 0.1, the partial derivatives .

[0078] Update the parameter θ using gradient descent: Update the formula based on the parameters: Substitute the data into the calculation: The β value was optimized to 0.36, which is within the reasonable range of 0.3-0.8.

[0079] Optimized performance: Brightness adjustment: ; Substitute the data to get ; Before optimization: when β=0.6 After optimization, the brightness is reduced by 60 cd, but it still meets the summer municipal road traffic lighting standard of ≥500 cd; Energy consumption optimization: Lighting energy consumption This reduces the Wh consumption by 48 kWh compared to before optimization. The loss function J = 0.6 × 592 + 0.3 × 80 + 0.1 × 200 = 355.2 + 24 + 20 = 399.2, which is 28.8% lower than the initial value of 428. Long-term benefits: Based on a 30-day month, the monthly energy savings for lighting is 48 × 30 = 1440 kWh. At the municipal electricity price of 0.75 yuan / kWh, the monthly electricity cost savings are 1440 × 0.75 = 1080 yuan, with no safety hazards.

[0080] Optimize the energy storage charging start-up threshold θ during winter workdays (insufficient sunlight, fluctuating traffic flow); Target optimization parameters: Energy storage charging start threshold θ; grid charging is initiated when the remaining energy is below θ. This corresponds to 60kWh of a 200kWh energy storage capacity; With a learning rate γ=0.02, cross-validation was performed to adapt to the parameter fluctuation characteristics in winter. Loss function J: As before, ; Historical data: Data collected from the past 30 winter workdays, with 4 hours of effective sunshine per day, 80 kWh of solar power generation, and traffic flow of 70 vehicles / minute during the morning rush hour and 5 vehicles / minute during the night. The system energy consumption data is the average daily lighting energy consumption. Energy storage loss Charging energy consumption ; Other parameters: The peak-valley electricity price of the municipal power grid is 1.2 yuan / kWh during peak hours and 0.5 yuan / kWh during off-peak hours.

[0081] Calculate the loss function J and its partial derivatives. ; Initial loss function J calculation: Substitute historical energy consumption data, ; Partial derivatives Calculation: An excessively high θ will lead to frequent low-power charging of energy storage during periods of insufficient sunlight in winter, increasing energy loss. Based on historical data... .

[0082] When θ=30% Exact match; When θ=35% , .

[0083] Take Δθ = 5%, partial derivatives The unit is kWh / percentage.

[0084] Update the parameter θ using gradient descent, according to the formula: ; Substitute the data into the calculation: .

[0085] Optimized performance: Energy storage operation: The charging start threshold is optimized to 31.2%, which corresponds to 62.4kWh of the energy storage capacity of 200kWh. When there is insufficient sunlight in winter, the energy storage capacity will start off-peak charging before the remaining energy level drops to a low level, thus reducing deep discharge loss. Energy consumption optimization: energy storage loss The loss function J = 0.6 × 640 + 0.3 × 117.6 + 0.1 × 180 = 384 + 35.28 + 18 = 437.28, which is 0.72 lower than the initial value of 438. Long-term benefits: Based on a 30-day month, the monthly energy storage loss is reduced by 2.4 × 30 = 72 kWh, which is equivalent to 72 × 0.5 = 36 yuan in electricity costs (off-peak charging). At the same time, it reduces the deep cycling of the battery, extends the battery life by about 15%, and reduces maintenance and replacement costs.

[0086] The central control module also integrates remote monitoring capabilities, allowing users to view system status and adjust control strategies in real time via mobile terminals. Specifically, the central control module encrypts system operation data and uploads it to a cloud server. The cloud server then generates a user interface through data parsing and visualization processing. User adjustment commands are sent to the central control module via the cloud server. The central control module modifies preset algorithm parameters based on these commands, with the parameter modification amount obtained through fuzzy logic calculations of user input data and system security constraints.

[0087] Uploading system operation data to the cloud server via encryption ensures data transmission security and prevents data leakage or tampering. Users can view the real-time operating status of each module, energy consumption data, load status, and fault information via mobile terminals, gaining a comprehensive understanding of the system without on-site inspections, significantly improving maintenance convenience. Remote adjustment commands are supported, and fuzzy logic is used to calculate parameter modifications, ensuring adjustments comply with system security constraints. This allows for rapid response to unexpected needs in municipal road lighting scenarios, avoiding the lag of manual on-site adjustments. Cloud server data parsing and visualization make information presentation more intuitive, helping users make accurate decisions. Remote operation also reduces the workload of maintenance personnel on-site, lowers labor costs, and ensures the continuity and stability of municipal road lighting and electric vehicle charging services.

[0088] Remotely adjust the lighting brightness curve during summer afternoons (14:00-14:30, sudden change in light intensity). Remote monitoring configuration: The central control module uses the AES-256 encryption algorithm to upload data, and the cloud server supports mobile APP (compatible with iOS / Android), which visually displays the module's operating status, energy consumption curves, and parameter configuration interface; System core parameters: Total power of lighting modules: 80kW; Current brightness curve: 300cd (48kW power) maintained from 14:00 to 16:00; Base brightness: The safety limit for brightness range is 200cd-600cd; Fuzzy logic parameters: Parameter modification amount calculation weights ω=0.7 (user demand weight), 1-ω=0.3 (safety constraint weight), user input target brightness. Current brightness Optimal value of safety constraints (Meets afternoon road lighting standards); Other parameters: The solar power generation module currently outputs 25kW, the energy storage module has a remaining capacity of 150kWh, and the charging module has a load of 30kW.

[0089] Remotely check system status: Maintenance personnel can log in to the cloud server via a mobile app to obtain system data in real time. Environmental data: A sudden thunderstorm occurred in the afternoon, and the light intensity dropped sharply from 80,000 lux to 20,000 lux; Operating status: The current brightness of the lighting module is 300 cd and the power is 48 kW, which cannot meet the road traffic needs under low light conditions; Energy status: Solar power generation capacity is 25kW, energy storage capacity is 150kWh, and the system has sufficient available power.

[0090] Initiate remote adjustment command: In the APP parameter configuration interface, the maintenance personnel enter the target brightness of 400cd and click "send adjustment" command. The command is transmitted to the central control module via cloud server in an encrypted manner.

[0091] Parameter modification amount for fuzzy logic calculation: After receiving the instruction, the central control module calculates the parameter modification amount using fuzzy logic, with the following formula: ; Substitute the data into the calculation: The final adjusted brightness is 300cd + 85cd = 385cd (which meets the safety constraints).

[0092] Execution adjustments and status feedback: Based on the calculation results, the central control module modified the lighting brightness curve parameters, adjusting the brightness from 14:00 to 16:00 to 385 cd, corresponding to a power of (385 / 500) × 80 kW = 61.6 kW; The adjusted data is encrypted and uploaded to the cloud server. The APP provides real-time feedback of "brightness adjustment successful" and displays the new brightness curve and power consumption trend, while simultaneously updating the energy distribution status of solar energy and energy storage. On-site effect: The brightness of 385cd meets the road traffic needs in thunderstorm weather, there is no safety hazard of pedestrians and vehicles not being able to see clearly, no maintenance personnel are required to work on-site, and the entire adjustment process takes ≤2 minutes.

[0093] In case of sudden surges in charging load during the night (22:00-22:10), remote optimization of charging power allocation can be implemented. Remote monitoring configuration: As before, the APP additionally supports charging module load warning and power distribution adjustment functions; System core parameters: Energy storage module capacity 200kWh, current remaining power 80kWh, initial charging power allocation parameters. , The safety-constrained charging power range is 10kW-50kW; Fuzzy logic parameters: Calculation weights for parameter modification amount are ω=0.6 and 1-ω=0.4, with the user inputting the target charging power. Current charging power Optimal value of safety constraints (Matching the current energy storage status); Other parameters: Lighting module operating power 80kW (brightness 500cd), solar power generation power 0kW, grid-connected electricity price during off-peak hours 0.5 yuan / kWh.

[0094] Remotely receive early warning information: The cloud server detected a sudden increase in the charging module load (3 new electric vehicles were added for fast charging, with a total demand of 45kW), triggering a load warning. This warning was pushed to the maintenance personnel's mobile phones via the app and displayed simultaneously. The charging module is currently under a load of 45kW, which exceeds the initial allocated power by 30kW. The energy storage module has a remaining capacity of 80kWh and a discharge power of 80kW (for lighting), which can also support additional charging loads. Off-peak electricity prices during grid connection periods can supplement electricity purchases to ensure load protection.

[0095] Remotely adjust charging power parameters: After viewing the warning details on the APP, the maintenance personnel input the target charging power of 40kW (taking into account both load demand and energy supply) and issue an adjustment command.

[0096] Parameter modification amount for fuzzy logic calculation: The central control module calculates the modification amount using the following formula: : Substitute the data into the calculation: The final adjusted charging power is 30kW + 8kW = 38kW (within the safety constraints).

[0097] Implementation adjustments and feedback: The central control module updates the charging power allocation parameters, supplying power to the charging module at 38kW, while coordinating with the energy storage module to increase the discharge power to 88kW (80kW for lighting + 8kW for supplementary power), and purchasing 2kW of electricity from the grid to ensure the stability of the total load; The APP provides real-time feedback on the adjustment results: the charging module power is 38kW, all electric vehicles are charging normally, the lighting brightness is maintained at 500cd, and the remaining energy storage power consumption rate is stable. Performance Comparison: Traditional solutions require on-site adjustments by maintenance personnel, taking at least one hour round trip, during which charging overload may cause tripping; this solution allows for remote adjustments in just one minute with no service interruption, ensuring normal operation of nighttime road lighting and charging services.

[0098] A centralized integrated photovoltaic, energy storage, and charging road lighting management method, comprising the following steps: The central control module collects ambient light data, traffic flow data, and system operation status data. The system operation status data includes the output power of the solar power generation module, the remaining power of the energy storage module, the load demand of the charging module, and the working status of the lighting module. Based on ambient lighting data and traffic flow data, the brightness of the lighting module is dynamically adjusted through a preset intelligent control algorithm; The charging and discharging process of the energy storage module is controlled based on the output power of the solar power generation module and the remaining power of the energy storage module. The output power of the charging modules is dynamically allocated based on the overall energy status of the system. The intelligent control algorithm includes the following processing steps: The collected traffic flow data is filtered to obtain filtered traffic flow data. The brightness adjustment factor is calculated based on the filtered traffic flow data. The brightness adjustment factor is determined by comparing the filtered traffic flow data with the maximum traffic flow value allowed by the system design and combining it with the adjustment coefficient learned from historical data. Lighting control commands are generated based on the brightness adjustment factor to adjust the brightness of the lighting module.

[0099] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0100] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0101] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A centralized integrated photovoltaic, energy storage, and charging road lighting management system, characterized in that, include: Solar power generation module, energy storage module, charging module, lighting module, central control module and grid connection interface module; Solar power generation modules are used to store the converted electrical energy in energy storage modules; The energy storage module is used to power the charging module and the lighting module; The grid connection interface module is connected between the municipal power grid and the system to enable bidirectional flow of electrical energy; The central control module is used to collect the operating status data of each module; The central control module also communicates with environmental sensors and traffic flow sensors installed along the roadside to collect ambient light data and traffic flow data. The central control module is also used to dynamically adjust the brightness of the lighting module based on ambient light data and traffic flow data. At the same time, based on the output power of the solar power generation module and the remaining power of the energy storage module, it controls the charging and discharging process of the energy storage module and allocates the output power of the charging module through the grid connection interface module.

2. The centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 1, characterized in that: The central control module dynamically adjusts the brightness of the lighting module based on real-time traffic flow data. Specifically, this is achieved through the following calculation process: First, real-time data from traffic flow sensors is acquired, and the data is filtered to eliminate noise. Then, the brightness adjustment factor is calculated. The brightness adjustment factor is obtained by normalizing the filtered traffic flow data and applying an exponential smoothing function. Finally, the lighting brightness is dynamically adjusted according to the brightness adjustment factor.

3. The centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 2, characterized in that: The central control module also optimizes the charging and discharging strategy of the energy storage module based on ambient light prediction data. Specifically, it achieves this through the following calculation process: First, it acquires weather forecast data and historical light data, and then uses a time series analysis model to predict the ambient light intensity for future periods. Then, the expected value of solar power generation is calculated based on the prediction results, and an energy storage charging and discharging plan is generated by combining the real-time electricity price data obtained through the grid connection interface module. The optimization objective of the energy storage charging and discharging plan is to minimize the total operating cost of the system. Its objective function is obtained by weighted summation of expected power generation, energy storage status, and electricity price data.

4. A centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 3, characterized in that: The central control module also dynamically allocates charging power based on the load demand of the charging module and the system energy status. Specifically, this is achieved through the following calculation process: First, it monitors the real-time load data of the charging module, including the number of connected electric vehicles and the battery state of charge; then, it allocates available power based on an energy priority algorithm, with the priority order being lighting module first, energy storage module second, and charging module last; during the allocation process, the adjustment value of the charging power is obtained by proportional integration calculation of available power data and load demand data.

5. A centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 4, characterized in that: The central control module also calculates the overall energy efficiency index of the system through an energy efficiency optimization algorithm, and adjusts the operating parameters based on the index. The specific calculation process is as follows: First, real-time energy consumption data and output data of each module are collected to calculate the system energy efficiency ratio; then, a multivariate regression model is used to analyze the relationship between the energy efficiency ratio and the operating parameters to generate an optimization strategy; wherein, the system energy efficiency ratio is obtained by calculating the ratio of total output energy to total input energy. The central control module dynamically adjusts the lighting brightness curve and energy storage charging and discharging thresholds based on historical energy efficiency ratio data.

6. A centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 5, characterized in that: The central control module also includes a fault detection function, which automatically diagnoses system faults by analyzing abnormal current and voltage data. The specific calculation process is as follows: First, the current and voltage data of each module are monitored in real time, and their deviation from the rated value is calculated; then, a sliding window statistical method is applied to detect continuous abnormal points, and a fault alarm is triggered when the deviation exceeds a preset threshold; the deviation value is obtained by calculating the root mean square error between the real-time data and the rated value. The central control module performs self-healing operations based on the magnitude of the deviation, including isolating faulty modules and switching to backup power.

7. A centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 6, characterized in that: The central control module also uses an adaptive learning algorithm to dynamically optimize control parameters. The specific calculation process is as follows: First, historical operating data is collected, including traffic flow, ambient light, and system energy efficiency data; then, the gradient descent method is used to iteratively update the control parameters; the parameter update amount is obtained by taking the partial derivative of the historical energy consumption data and multiplying it by the learning rate.

8. A centralized integrated photovoltaic, energy storage, and charging road lighting management system according to claim 1, characterized in that: The central control module also integrates remote monitoring capabilities, allowing users to view system status and adjust control strategies in real time via mobile terminals. Specifically, the central control module encrypts system operation data and uploads it to a cloud server. The cloud server then generates a user interface through data parsing and visualization processing. User adjustment commands are sent to the central control module via the cloud server. The central control module modifies preset algorithm parameters based on these commands, with the parameter modification amount obtained through fuzzy logic calculations of user input data and system security constraints.

9. A centralized integrated photovoltaic, energy storage, and charging road lighting management method, wherein the method is applied to the system described in any one of claims 1-8, characterized in that: The method includes the following steps: The central control module collects ambient light data, traffic flow data, and system operation status data. The system operation status data includes the output power of the solar power generation module, the remaining power of the energy storage module, the load demand of the charging module, and the working status of the lighting module. Based on ambient lighting data and traffic flow data, the brightness of the lighting module is dynamically adjusted through a preset intelligent control algorithm; The charging and discharging process of the energy storage module is controlled based on the output power of the solar power generation module and the remaining power of the energy storage module. The output power of the charging modules is dynamically allocated based on the overall energy status of the system. The intelligent control algorithm includes the following processing steps: The collected traffic flow data is filtered to obtain filtered traffic flow data. The brightness adjustment factor is calculated based on the filtered traffic flow data. The brightness adjustment factor is determined by comparing the filtered traffic flow data with the maximum traffic flow value allowed by the system design and combining it with the adjustment coefficient learned from historical data. Lighting control commands are generated based on the brightness adjustment factor to adjust the brightness of the lighting module.

Images