Perovskite light storage and charging shed and operation control method thereof

Abstract

The invention belongs to the technical field of application of perovskite photovoltaic modules, and relates to a perovskite light storage and charging shed and an operation control method thereof. The perovskite light storage and charging shed comprises a parking shed and a shed system, the shed system comprises a perovskite photovoltaic module, an energy storage module, a charging pile, a grid-connected bidirectional inverter, a data acquisition and preprocessing module, a communication module and a control cabinet, and an OPC module is arranged in the control cabinet. The OPC module dynamically predicts the generated power of the perovskite photovoltaic module and the required power of the charging pile according to the received real-time data, and generates an electric energy optimization strategy in a feasible region delimited by constraint conditions by taking the minimization of the life cycle cost of a shed system as an objective function; the control cabinet controls operation of the perovskite photovoltaic module, the energy storage module, the charging pile and the grid-connected bidirectional inverter through the communication module according to an electric energy optimization strategy. According to the invention, the shed system can maintain global optimal operation under different illumination conditions, electricity price changes and load fluctuations.

Description

Technical Field

[0001] This invention belongs to the field of application technology of perovskite photovoltaic modules, and specifically relates to a perovskite photovoltaic energy storage and charging vehicle shed and its operation control method. Background Technology

[0002] With the rapid growth in the number of new energy vehicles, the demand for electric vehicle charging is increasing exponentially. Traditional centralized charging methods that rely on the power grid face problems such as excessive power load, decreased grid stability, and increased operating costs. To alleviate these contradictions, "photovoltaic-storage-charging" carports, which combine photovoltaic power generation with energy storage systems, are gradually becoming an important direction for the next generation of green transportation infrastructure. This type of system generates electricity through photovoltaic modules on the carport roof, combined with energy storage battery packs and smart charging interfaces, to provide clean electricity driven by renewable energy for electric vehicles, which can reduce carbon emissions and alleviate grid pressure. However, existing photovoltaic-storage-charging carports still have the following technical bottlenecks and shortcomings: 1) Limited photovoltaic module structure and application: Currently, most mainstream carport photovoltaic modules are made of crystalline silicon materials. Although they have high efficiency, they are opaque and difficult to achieve the aesthetic and functional integration of building-integrated photovoltaics (BIPV). At the same time, crystalline silicon modules are heavy and have poor flexibility, which limits the lightweight and diversified design of carport structures and is not conducive to their promotion in urban public spaces and new building scenarios. 2) Single energy management strategy: Photovoltaic power generation is significantly affected by light conditions, while the charging demand of electric vehicles has strong randomness and peak-valley differences. Existing carports mostly adopt fixed power allocation or simple grid connection modes, lacking intelligent prediction and scheduling methods, resulting in low energy utilization. During some periods, there is surplus power generation that is difficult to absorb, while at other times, high-cost power purchases are required, leading to high overall operating costs. 3) Insufficient power quality and grid compatibility: When photovoltaic carports are connected to the grid, problems such as harmonics, voltage fluctuations, or low power factors often arise due to insufficient inverter control, affecting grid stability. Existing methods for power quality control still rely mainly on hardware design, lacking global energy quality regulation combined with intelligent optimization algorithms, making it difficult to adapt to different grid standards and usage environments. 4) Poor adaptability to multiple scenarios: Different regions and application scenarios (such as commercial areas, industrial parks, or rural transportation hubs) have different requirements for system operation modes, energy storage scale, and energy scheduling strategies. Existing photovoltaic-storage-charging carports mostly adopt a "one-size-fits-all" design and control approach, lacking flexible and scalable architectures and adaptive operation control methods for multiple scenarios. Summary of the Invention

[0003] This invention addresses the shortcomings of existing photovoltaic-storage-charging carports in terms of component structure, energy management, and power quality. It provides a perovskite photovoltaic-storage-charging carport and its operation control method. Through the OPC (Optimal Perovskite Carport Control) module, the power generation of the perovskite photovoltaic modules and the power demand of the charging piles are dynamically predicted. With the minimization of the carport system's life cycle cost as the objective function, an energy optimization strategy is generated within the feasible domain defined by constraints. The output power of the perovskite photovoltaic modules, the charging and discharging state of the energy storage batteries, and the interactive power of the grid-connected bidirectional inverter are adjusted in real time, enabling the carport system to maintain globally optimal operation under different lighting conditions, electricity price changes, and load fluctuations.

[0004] This invention is implemented as follows: A perovskite photovoltaic-storage-charging carport is provided, comprising a carport and a carport system. The carport system includes perovskite photovoltaic modules, energy storage modules, charging piles, a grid-connected bidirectional inverter, a data acquisition and preprocessing module, a communication module, and a control cabinet. The perovskite photovoltaic modules are installed on the outer surface of the carport. The perovskite photovoltaic modules, energy storage modules, charging piles, grid-connected bidirectional inverter, and control cabinet are interconnected via a DC bus. The energy storage module includes an energy storage battery. The grid-connected bidirectional inverter is connected to the public power grid. The data acquisition and preprocessing module is used to collect data from the perovskite photovoltaic modules, energy storage modules, charging piles, and the grid connection. The system collects real-time operating data from the bidirectional inverter and performs preprocessing on the raw data, including time alignment, missing data interpolation, and outlier removal. The preprocessed real-time data is then transmitted to the control cabinet via a communication module. An OPC module is installed in the control cabinet. Based on the received real-time data, the OPC module dynamically predicts the power generation of the perovskite photovoltaic modules and the power demand of the charging piles. With the goal of minimizing the life cycle cost of the carport system, an energy optimization strategy is generated within the feasible domain defined by the constraints. The control cabinet then controls the operation of the perovskite photovoltaic modules, energy storage modules, charging piles, and grid-connected bidirectional inverters via the communication module according to the energy optimization strategy.

[0005] Furthermore, the objective function is: Formula 1 In the formula, min Cost To minimize costs, This is the electricity purchase cost coefficient. For the power purchase capacity, This is the electricity sales cost coefficient. For electricity sales capacity, For the equivalent cost of degradation of perovskite photovoltaic modules, This represents the theoretical degradation of the perovskite photovoltaic module's health status at hour h, caused by the current charge / discharge operation and factored towards that hour. The penalty coefficient is... This represents a penalty for delayed charging or exceeding the queue time.

[0006] Furthermore, the constraints include power quality constraints, energy storage boundary constraints, power balance constraints, charging service quality constraints, and constraints for connection to the public power grid.

[0007] Furthermore, the power quality constraints are: power factor PF ≥ 0.98 and total harmonic distortion (THD) ≤ 2%.

[0008] Furthermore, the energy storage boundary constraints are as follows: Formula 2 Formula 3 In the formula, This represents the lower limit of energy storage capacity for energy storage batteries. This represents the real-time energy storage status of the energy storage battery. This represents the upper limit of energy storage capacity for energy storage batteries. This refers to the real-time discharge power of the energy storage battery. This represents the upper limit of the discharge power of the energy storage battery.

[0009] Furthermore, the power balance constraint condition is as follows: Formula 4 In the formula, For the power generation of perovskite photovoltaic modules, This refers to the discharge power of the energy storage battery. Power purchased from the public power grid, The power consumed to charge the charging station. For auxiliary systems or conventional load power, This refers to the power loss of the carport system.

[0010] Furthermore, the charging service quality constraint is as follows: for each vehicle i, the minimum energy replenishment must be satisfied before the scheduled or expected departure time di. .

[0011] Furthermore, the constraints for connecting to the public power grid include grid connection point restrictions, grid connection power restrictions, grid connection current restrictions, grid connection voltage limits, and ramp rate restrictions, in order to meet the grid connection requirements of the public power grid.

[0012] Furthermore, the carport system also includes a monitoring and feedback module. When the monitoring and feedback module detects that the difference between the actual operating data of the carport system and the predicted results exceeds the limit, it feeds the information back to the control cabinet. The OPC module then recalculates and updates the power optimization strategy, ensuring that the carport system is always in a stable and efficient operating state.

[0013] Furthermore, the energy storage module also includes a battery management system (BMS) and a bidirectional DC-DC converter.

[0014] This invention is implemented by providing an operation control method for the perovskite photovoltaic energy storage and charging vehicle shed as described above, comprising the following steps: Step A: Data Acquisition and Preprocessing The data acquisition and preprocessing module periodically collects real-time operating data from perovskite photovoltaic modules, energy storage modules, charging piles, and grid-connected bidirectional inverters, and performs preprocessing on the data, including time alignment, missing data interpolation, and outlier removal. The preprocessed real-time data is then transmitted to the control cabinet via the communication module. Step B: Short-term forecasting Predict the power generation of perovskite photovoltaic modules and the charging power demand of charging piles; Step C: Optimize the model With minimizing the life cycle cost of the carport system as the core objective function, the OPC module is used to generate an energy optimization strategy within the feasible region defined by the constraints. Step D: Execute control The control cabinet controls the operation of perovskite photovoltaic modules, energy storage modules, charging piles, and grid-connected bidirectional inverters through power generation optimization strategies via the communication module.

[0015] Furthermore, the operation control method further includes the following steps: Step E: Charging Scheduling Build a dynamic priority charging queue and update the task order regularly; monitor the completion rate of charging piles and automatically increase the charging priority of lagging vehicles when the ratio difference exceeds the threshold; when the charging power is insufficient, activate the degradation strategy: prioritize the 3kW basic charging power of all vehicles, and allocate the remaining power to fast charging piles according to priority.

[0016] Furthermore, the operation control method further includes the following steps: Step F: V2G (Vehicle to Grid) Strategy Set up a triple triggering condition of price-driven (e.g., electricity price > threshold), demand control (e.g., monthly peak warning), and user authorization. In the power optimization strategy, establish a special energy storage model for V2G vehicles, set a decay cost coefficient of 1.5 times that of conventional energy storage, and control the discharge curve through bidirectional charging piles to ensure that the return trip SoC is not lower than the user-set value.

[0017] Furthermore, the operation control method further includes the following steps: Step G: Exception Handling The steady-state priority mode in the power optimization strategy is activated when the prediction deviation is >20%, which limits the charging peak of the charging pile, improves reactive power support, and temporarily raises the lower limit of electricity purchase. The safety backoff mode is activated when the communication module is interrupted or the temperature of the charging pile exceeds the limit. The energy storage module switches to conservative power or standby, the charging pile enters the current limiting state, maintains grid connection compliance, and is processed according to the serialized process of equipment disconnection → parameter adjustment → status confirmation → alarm notification.

[0018] Furthermore, the operation control method further includes the following steps: Step H: Parameter Optimization Initially, the degradation coefficient was fitted based on the cycle life curve of perovskite photovoltaic modules. A line loss model, P_loss = a·I² + b, was established through light and heavy load tests. The model was automatically recalibrated weekly during the low-load period in the early morning, and the least squares method was used to update the model parameters. In this model, P_loss is the total power loss of the carport system, in watts (W); I is the total current flowing through the DC bus of the carport system, in amperes (A); a is the loss coefficient (variable loss coefficient) related to the square of the current, representing the loss part of the carport system that varies with the current. Its physical essence is the equivalent variable resistance of the carport system, in ohms (Ω); and b is the fixed damage coefficient independent of the current, representing the constant power loss in the carport system, in watts (W).

[0019] Furthermore, the operation control method further includes the following steps: Step I: Effectiveness Evaluation The system displays real-time KPIs such as self-consumption rate and demand reduction rate. It compares the operating data of the power optimization strategy during on / off states through A / B testing, verifies the significance of the indicator improvements using T-tests, and finally generates an economic benefit analysis report.

[0020] Compared with the prior art, the perovskite photovoltaic energy storage and charging shed and its operation control method of the present invention have the following characteristics: 1. Intelligent energy management and operational efficiency optimization Multi-terminal linkage architecture: Through the linkage of four terminals—titanium ore photovoltaic modules, energy storage modules, charging piles, and the public power grid—flexible allocation and efficient utilization of electrical energy are achieved. The intelligent operation control method of the OPC module: Based on artificial intelligence, photovoltaic power generation and charging demand prediction, combined with a multi-objective optimization model, dynamic scheduling of power generation, energy storage, charging and grid interaction is realized. With the goal of minimizing the life cycle cost of the carport system, energy utilization efficiency and economy are significantly improved while ensuring power quality.

[0021] 2. Enhanced power quality and grid interaction capabilities Active power quality regulation: By embedding constraints such as power factor and harmonic distortion rate into the OPC module and combining them with the intelligent adjustment of the grid-connected bidirectional inverter, active control of power quality is achieved, effectively suppressing problems such as harmonics and voltage fluctuations, and improving grid friendliness. Two-way energy exchange: The carport system supports two-way energy exchange with the public power grid. It can flexibly adjust the power purchase and sale strategy according to the electricity price signal and load demand, which not only reduces the operating cost of the carport system, but also provides ancillary services such as peak shaving and valley filling for the public power grid.

[0022] 3. Multi-scenario adaptability and scalability Differentiated operation strategy: For different application scenarios such as commercial areas, industrial parks, and rural transportation hubs, the carport system can automatically switch operation modes (such as cost priority, peak shaving and valley filling, and self-consumption priority), achieving a high degree of scenario adaptability. Modular and scalable design: The hardware architecture and control strategy of the carport system adopt a modular design, which makes it easy to flexibly adjust the installed capacity of perovskite photovoltaic modules, the scale of energy storage modules and the configuration of charging piles according to actual needs, and supports the rapid deployment and subsequent expansion of the carport system.

[0023] 4. Significant economic and environmental benefits. Significantly reduced operating costs: Through intelligent prediction and optimized scheduling, the carport system can make full use of the power generation of perovskite photovoltaic modules, the charging and discharging of energy storage modules, and the peak-valley electricity price difference of the public power grid, effectively reducing dependence on external power purchases and overall electricity costs; Lifecycle cost optimization: With the goal of minimizing the lifecycle cost of the carport system, multiple factors such as equipment investment, operation and maintenance, and perovskite photovoltaic module losses are taken into account to ensure long-term economic efficiency. Outstanding contributions to green and low-carbon development: By maximizing the use of renewable energy, it reduces the consumption of traditional fossil fuels and carbon emissions, meeting the requirements of green transportation and sustainable development.

[0024] 5. Improved user experience and reliability Intelligent charging service: Supports DC fast charging and AC slow charging, and can intelligently allocate charging power according to vehicle needs and carport system status, shortening user waiting time and improving the charging experience; Carport system reliability and stability: Through real-time monitoring and feedback adjustments, the carport system can quickly respond to internal and external changes, such as sudden weather changes and load fluctuations, ensuring continuous and stable operation and improving power supply reliability. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the perovskite photovoltaic energy storage and charging carport system of the present invention.

[0026] The symbols in the image are as follows: 1. Perovskite photovoltaic modules; 2. Energy storage modules; 21. Energy storage batteries; 3. Charging piles; 4. Grid-connected bidirectional inverters; 5. Data acquisition and preprocessing modules; 6. Communication modules; 7. Control cabinets; 71. OPC modules; 8. DC bus; 9. Public power grid. Detailed Implementation

[0027] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0028] Please refer to Figure 1 As shown, a preferred embodiment of the perovskite photovoltaic energy storage and charging carport of the present invention includes a carport and a carport system (not shown in the figure). The carport system includes a perovskite photovoltaic module 1, an energy storage module 2, a charging pile 3, a grid-connected bidirectional inverter 4, a data acquisition and preprocessing module 5, a communication module 6, and a control cabinet 7.

[0029] The perovskite photovoltaic module 1 is installed on the sun-facing outer surface of the parking shed. The perovskite photovoltaic module 1, energy storage module 2, charging pile 3, grid-connected bidirectional inverter 4, and control cabinet 7 are interconnected via DC bus 8. The energy storage module 2 includes an energy storage battery 21, and the grid-connected bidirectional inverter 4 is connected to the public power grid 9.

[0030] The data acquisition and preprocessing module 5 is used to acquire real-time operating data from the perovskite photovoltaic module 1, energy storage module 2, charging pile 3, and grid-connected bidirectional inverter 4, and to perform preprocessing on the acquired raw data, including time alignment, missing data interpolation, and outlier removal. The preprocessed real-time data is transmitted to the control cabinet 7 via the communication module 6.

[0031] An OPC module 71 is installed in the control cabinet 7. The OPC module 71 dynamically predicts the power generation of the perovskite photovoltaic module 1 and the power demand of the charging pile 3 based on the received real-time data. Taking the minimization of the life cycle cost of the carport system as the objective function, it generates an energy optimization strategy within the feasible domain defined by the constraints. The control cabinet 7 controls the operation of the perovskite photovoltaic module 1, the energy storage module 2, the charging pile 3 and the grid-connected bidirectional inverter 4 through the communication module 6 according to the energy optimization strategy.

[0032] Specifically, the energy storage module 2 also includes a battery management system (BMS) and a bidirectional DC-DC converter (not shown in the figure). The energy storage module 2, equipped with a bidirectional DC-DC converter, can flexibly charge and discharge. It not only absorbs excess energy when the perovskite photovoltaic module 1 generates more power, but also releases energy at night or during peak hours, connecting it to the public grid 9 via the grid-connected bidirectional inverter 4, thus alleviating the pressure on the public grid 9. The synergistic effect of the bidirectional DC-DC converter and the grid-connected bidirectional inverter 4 reduces energy conversion losses and improves the overall efficiency of the carport system.

[0033] The electricity generated by the perovskite photovoltaic module 1 can be used to charge electric vehicles via charging pile 3, to charge the energy storage battery 21 of the energy storage module 2, and to be fed into the public power grid 9 via the grid-connected bidirectional inverter 4. The energy storage module 2 exchanges energy bidirectionally with the public power grid 9 through the grid-connected bidirectional inverter 4. Through the synergistic effect of the bidirectional DC-DC converter and the grid-connected bidirectional inverter 4, the carport system has greater flexibility in the energy flow path, which can effectively reduce conversion losses and improve energy utilization efficiency.

[0034] In this embodiment, the perovskite photovoltaic module 1 is a high-efficiency, semi-transparent perovskite photovoltaic module, which not only generates electricity but also maintains a certain level of light transmittance, providing comfortable lighting for the space under the parking shed. To reduce the impact of local shading on overall power generation, the perovskite photovoltaic module 1 is connected to the DC bus 8 in a zoned manner, with each zone capable of independently outputting power. Thus, even if some of the perovskite photovoltaic modules 1 are shaded by trees or vehicles, the remaining zones can still operate normally, preventing large-scale power generation losses.

[0035] By using high-efficiency, semi-transparent perovskite photovoltaic modules 1 as the roof of the parking shed, not only is efficient power generation achieved, but the aesthetics of the building and the need for natural lighting are also met. This solves the problems of opacity and bulkiness of traditional crystalline silicon modules, significantly enhancing the potential for integrated application of parking sheds in urban public spaces and commercial buildings. The perovskite photovoltaic modules 1 are connected to the DC bus independently in different zones, effectively reducing the impact of localized shading on overall power generation efficiency, improving the stability and power generation efficiency of the perovskite photovoltaic modules 1, and enhancing their adaptability to complex environments.

[0036] In this embodiment, the grid-connected bidirectional inverter 4 not only converts DC power to AC power, but also adjusts the power factor and suppresses harmonics, thereby ensuring power quality. The carport system can decide independently whether to sell electricity to or purchase electricity from the public grid 9 based on electricity prices and demand.

[0037] Specifically, the objective function is: Formula 1 In the formula, min Cost To minimize costs, This is the electricity purchase cost coefficient. For the power purchase capacity, This is the electricity sales cost coefficient. For electricity sales capacity, For the equivalent cost of degradation of perovskite photovoltaic modules, This represents the theoretical degradation of the perovskite photovoltaic module's health status at hour h, caused by the current charge / discharge operation and factored towards that hour. The penalty coefficient is... This represents a penalty for delayed charging or exceeding the queue time.

[0038] Specifically, the constraints include power quality constraints, energy storage boundary constraints, power balance constraints, charging service quality constraints, and constraints for connection to the public power grid.

[0039] Specifically, the power quality constraints are: power factor PF ≥ 0.98 and total harmonic distortion (THD) ≤ 2%.

[0040] Specifically, the energy storage boundary constraints are as follows: Formula 2 Formula 3 In the formula, This represents the lower limit of energy storage capacity for energy storage batteries. This represents the real-time energy storage status of the energy storage battery. This represents the upper limit of energy storage capacity for energy storage batteries. This refers to the real-time discharge power of the energy storage battery. This represents the upper limit of the discharge power of the energy storage battery.

[0041] Specifically, the power balance constraint condition is as follows: Formula 4 In the formula, For the power generation of perovskite photovoltaic modules, This refers to the discharge power of the energy storage battery. Power purchased from the public power grid, The power consumed to charge the charging station. For auxiliary systems or conventional load power, This refers to the power loss of the carport system.

[0042] Specifically, the charging service quality constraint is: for each vehicle i, the minimum energy replenishment must be met before the scheduled or expected departure time di. .

[0043] Specifically, the constraints for connecting to the public power grid include grid connection point restrictions, grid connection power restrictions, grid connection current restrictions, grid connection voltage limits, and ramp rate restrictions, in order to meet the grid connection requirements of the public power grid 9.

[0044] Based on OPC module 71, this invention also proposes a scalable multi-scenario adaptation strategy, enabling the carport system to operate differently according to various application environments. In commercial area scenarios, the carport system prioritizes utilizing peak-valley electricity price differences to reduce electricity costs; in industrial park scenarios, the operating logic focuses on peak shaving and valley filling to improve load balance; and in rural transportation hub scenarios, the carport system emphasizes increasing the self-consumption ratio of electricity generated by perovskite photovoltaic modules. Through this multi-scenario adaptive operation control, the carport system of this invention not only possesses the capabilities of efficient power generation and intelligent energy management but also ensures power quality and economic efficiency, demonstrating promising prospects for widespread application.

[0045] In the charging section of charging station 3, the carport system's charging stations 3 support both DC fast charging and AC slow charging, meeting the needs of different vehicles and users. Fast charging can replenish a large amount of electrical energy in a short time, while slow charging is suitable for vehicles parked for extended periods. All charging stations 3 are connected to the control cabinet via communication module 6, ensuring that the charging needs of each vehicle can be identified and managed.

[0046] Specifically, the carport system also includes a monitoring and feedback module (not shown in the figure). When the monitoring and feedback module detects that the difference between the actual operating data and the predicted results of the carport system exceeds a certain limit, it feeds this information back to the control cabinet 7. The OPC module 71 then recalculates the optimization and updates the power optimization strategy, ensuring that the carport system always operates in a stable and efficient state. In this way, the carport system can maintain good adaptability under various climate and load conditions, truly achieving efficient power generation, intelligent charging, and green grid connection.

[0047] This invention also discloses an operation control method for the perovskite photovoltaic energy storage and charging canopy as described above, comprising the following steps: Step A: Data Acquisition and Preprocessing The data acquisition and preprocessing module 5 periodically collects real-time operating data from the perovskite photovoltaic module 1, energy storage module 2, charging pile 3, and grid-connected bidirectional inverter 4, and performs preprocessing on the data, including time alignment, missing data interpolation, and outlier removal; the preprocessed real-time data is then transmitted to the control cabinet 7 via the communication module 6.

[0048] Specifically, the data acquisition and preprocessing module 5 periodically collects data such as environmental irradiance Gt, ambient temperature Tt, back temperature of perovskite photovoltaic module 1, DC bus measurement points, energy storage module SoC / SoH, electric vehicle (EV) arrival / queue information, real-time electricity price, and demand. It also performs preprocessing on abnormal data, including missing data imputation and outlier removal, and time alignment to generate a feature sequence within a [tk,t] window. The preprocessed real-time data is then output.

[0049] The hardware layer employs a multi-protocol adaptation architecture, communicating with grid-connected bidirectional inverter 4, energy storage module 2, charging pile 3, and other devices via Modbus (a serial communication protocol), OPC UA (OPC Unified Architecture), and MQTT (Message Queuing Telemetry Transport) protocols. An NTP (Network Time Protocol) time server is deployed to achieve millisecond-level clock synchronization. Electrical quantity data is collected at a frequency of 1-2 seconds, while slowly varying parameters such as temperature and SoC are collected at a frequency of 10 seconds. The software layer establishes a data cleaning pipeline, filling missing data within 30 seconds using linear interpolation or Last Observation Carried Forward (LOCF) and removing outliers using a rolling Z-Score method based on a 5-minute window (±3σ, where σ is the standard deviation). Feature engineering extracts instantaneous features, 15-minute statistical features (mean / standard deviation / extreme values), and derived features calculated from power ramp-up rates, etc.

[0050] Step B: Short-term forecasting Predict the power generation of perovskite photovoltaic module 1 and the charging power demand of charging pile 3.

[0051] Specifically, the power generation prediction of perovskite photovoltaic module 1 adopts a Seq2Seq (sequence-to-sequence) model or TCN (Temporal Convolutional Network), inputting feature sequences such as historical irradiance, cloud cover forecast, temperature, and perovskite photovoltaic module back temperature, and outputting photovoltaic power predictions for the next 1-4 hours. The power demand prediction of charging pile 3 uses an XGBoost (Extreme Gradient Boosting) / LightGBM (Light Gradient Boosting Machine) tree model, integrating features such as historical charging curves, real-time queuing information, and BMS (Battery Management System) data to predict electric vehicle charging demand. The model employs a sliding window update mechanism, retaining the most recent 30 days of training data, and triggering online re-prediction when the prediction deviation exceeds 20% three times consecutively.

[0052] Step C: Optimize the model With minimizing the life cycle cost of the carport system as the core objective function, the OPC module 71 is used to generate an energy optimization strategy within the feasible region defined by the constraints.

[0053] Specifically, the objective function is shown in Formula 1.

[0054] Specifically, the constraints include power quality constraints, energy storage boundary constraints, power balance constraints, charging service quality constraints, and grid connection constraints. The energy storage boundary constraints are shown in Formulas 2 and 3, and the power balance constraints are shown in Formula 4. The power quality constraints are: power factor PF ≥ 0.98 and total harmonic distortion (THD) ≤ 2%. The charging service quality constraints are: for each vehicle i, minimum energy replenishment must be met before the scheduled or expected departure time di. .

[0055] Specifically, the constraints for connecting to the public power grid include grid connection point restrictions, grid connection power restrictions, grid connection current restrictions, grid connection voltage limits, and ramp rate restrictions, in order to meet the grid connection requirements of the public power grid 9.

[0056] Among them, a) Grid connection point restriction: The grid connection point is the connection point between the microgrid and the main grid. The grid connection point restriction is the most critical and is usually stipulated by the public grid company or determined by the equipment capacity.

[0057] Among them, b) grid-connected power limit and grid-connected current limit: the power purchased from the grid, P_grid(h), cannot exceed an upper limit value, P_grid_max, and the power transmitted back to the grid, P_PV→grid(h), cannot exceed an upper limit value (sometimes this value is smaller). Under fixed voltage conditions, the limited power corresponds to the limited current. The purpose is to prevent overloading of connecting cables, meet the agreed capacity with the power grid company, and avoid impacting the upstream power grid.

[0058] c) Grid connection voltage limit: The voltage at the grid connection point must be within the range specified by national standards. For example, the nominal voltage in China is 380V, and the normal range is within ±7%. The purpose is to ensure power quality and prevent damage to equipment or impact on other users due to excessively high or low voltage.

[0059] Among them, d) ramp rate limit: the rate of change of power absorbed from or injected into the grid cannot be too rapid. The purpose is to avoid power surges impacting the grid (similar to "flexible start") and to meet the smoothness requirements of grid dispatch. This is particularly important for systems that include intermittent photovoltaics and stochastic charging.

[0060] In addition, the power generation optimization strategy should also consider decision variables and the power allocation of each charging station. The decision variables are... , In the formula, The discharge / charge power of the energy storage module, Power purchased from the power grid, To provide charging power for electric vehicles to the photovoltaic system. The power supplied to the grid by the photovoltaic system.

[0061] Furthermore, the relationship between the decision variable and the power allocation of each charging station is one of upper-level optimization and lower-level allocation. The decision variable P_PV→EV(h) is an aggregate variable, representing the sum of power allocated by the carport system to all electric vehicles currently charging within the time period h. P_PV→EV(h)=Σ(P_ev_i(h)), where P_ev_i(h) is the actual power of the i-th charging station. This variable is determined in the system-level optimization, with the goal of maximizing economic efficiency and maximizing the use of green electricity.

[0062] The power allocation for each charging station is a refined allocation performed at the charging control layer after the total P_PV→EV(h) is determined by the upper-level optimization. The specific allocation criteria include the following: 1. Vehicle requirements: Remaining battery level, target battery level, and estimated dwell time for each vehicle; 2. User priority / contract: For example, VIP users or users with high fees are given priority; 3. Principle of fairness: Ensure that every vehicle can receive the most basic charging power; 4. Hardware limitations: Each charging station has its own independent power limit P_ev_i_max.

[0063] In addition, the main content of generating power optimization strategies is to minimize total electricity costs and maximize electricity sales revenue, which specifically includes the following steps: a) Prediction An optimization model is established based on the current state of the carport system (such as the SOC of the energy storage battery, the status of the vehicles being charged, and the current electricity price) and the forecast data for a future period (the forecast time domain, such as the next hour).

[0064] The forecast data includes: perovskite photovoltaic module power generation forecast (P_PV_forecast(h)), electric vehicle charging demand forecast (e.g., how many cars will come, how much electricity will be charged, how long will they stay, etc.), and public grid electricity price forecast (if the electricity price is variable); corresponding to the objective function of Formula 1 above, the goal is to minimize the total electricity cost and maximize electricity sales revenue.

[0065] b) Correction Since predictions cannot be completely accurate, after the carport system has actually run for one cycle (e.g., 5 minutes), the actual measured value will be compared with the predicted value of the previous cycle, and the optimization for the next cycle will be corrected.

[0066] The specific steps for correction include: ①. Scrolling time domain: Optimize the window to move forward one cycle (e.g., discard the furthest moment and add the latest moment).

[0067] ②. State Update: Update the initial state of the optimization model with the latest measured data. The most important thing is to update the SoC of the energy storage battery and the remaining charge of the electric vehicle.

[0068] ③. Feedback correction: The deviation between the previous cycle's forecast and the actual situation is treated as a disturbance term and taken into account in the new optimization, so that the new plan is more in line with the actual situation.

[0069] c) Hierarchical decomposition Break down complex large-scale optimization problems into more manageable subproblems.

[0070] Upper layer: Energy management optimization is performed with a coarser time resolution (e.g., 15 minutes) and aggregated variables (e.g., total charging power P_PV→EV) to determine long-term energy scheduling strategies.

[0071] Lower layer: Real-time power allocation with finer time resolution (e.g., 5 minutes or 1 minute) and specific variables (power of each charging gun) to quickly respond to real-time changes.

[0072] In summary, within each solution cycle, minimizing the objective function (economic efficiency) is the core objective. Within the feasible region defined by constraints (safety, physical laws, service quality, etc.), the optimal decision variables (system-level power allocation) are solved. These system-level decision variables are then refined into specific power commands for each charging pile 3 based on the allocation criteria at the lower level. This continuous iteration and optimization of the OPC module 71's algorithm model ultimately achieves efficient, economical, safe, and user-satisfactory optimized control of the carport system. The prediction, correction, and hierarchical decomposition process of the solution strategy continuously optimizes the model. For example, at a certain time, 60% of the electricity generated by the perovskite photovoltaic module 1 is allocated to electric vehicle charging, 30% is stored in the energy storage battery 21, and 10% is connected to the public power grid 9.

[0073] Step D: Execute control Control cabinet 7 controls the operation of perovskite photovoltaic module 1, energy storage module 2, charging pile 3 and grid-connected bidirectional inverter 4 through the power generation optimization strategy under the communication module.

[0074] Specifically, the energy storage battery 21 will charge and discharge according to the set power, the grid-connected bidirectional inverter 4 will adjust the output to maintain power quality, and the charging pile 3 will allocate power according to vehicle demand and priority. The whole process is automated. Users only need to park their car under the parking shed, plug in the charging gun, and they can enjoy intelligently allocated clean energy.

[0075] Specifically, P_bat_ref, Q_grid_ref, and other setpoints are sent in real-time via EtherCAT Ethernet. The energy storage battery 21 employs a dual-closed-loop PI (Proportional-Integral Control) system with an outer power loop and an inner current loop. The grid-connected bidirectional inverter 4 achieves PQ decoupling control through Park conversion and uses a proportional resonant PR controller to track current commands. The charging pile 3 interacts with the vehicle's BMS according to the ISO15118 protocol, taking the minimum value between the setpoint and the BMS limit to perform constant current (CC) / constant voltage (CV) charging. A three-level safety linkage mechanism is established, with hardware protection achieving millisecond-level action and software logic triggering standardized SOP (Standard Operating Procedure) procedures.

[0076] Step E: Charging Scheduling Build a dynamic priority charging queue and update the task order regularly; monitor the completion rate of charging pile 3, and automatically increase the charging priority of lagging vehicles when the ratio difference exceeds the threshold; when the charging power is insufficient, start the degradation strategy: prioritize the 3kW basic charging power of all vehicles, and allocate the remaining power to fast charging piles according to priority.

[0077] Step F: V2G (Vehicle to Grid) Strategy A triple trigger condition is set up: price-driven (e.g., electricity price > threshold), demand control (e.g., monthly peak warning), and user authorization. A special energy storage model is established for V2G vehicles in the power optimization strategy, with a degradation cost coefficient of 1.5 times that of conventional energy storage. The discharge curve is controlled through bidirectional charging piles to ensure that the return trip SoC is not lower than the user-set value. When the electricity price is high / demand control is triggered, and the vehicle allows it, V2G external speaker is enabled.

[0078] Step G: Exception Handling The steady-state priority mode in the power optimization strategy is activated when the prediction deviation is >20%, which limits the charging peak of the charging pile, improves reactive power support, and temporarily raises the lower limit of electricity purchase. The safety backoff mode is activated when the communication module is interrupted or the temperature of the charging pile exceeds the limit. The energy storage module switches to conservative power or standby, the charging pile enters the current limiting state, maintains grid connection compliance, and is processed according to the serialized process of equipment disconnection → parameter adjustment → status confirmation → alarm notification.

[0079] Step H: Parameter Optimization Initially, the degradation coefficient was fitted based on the cycle life curve of perovskite photovoltaic modules. A line loss model, P_loss = a·I² + b, was established through light and heavy load tests. The model was automatically recalibrated weekly during the low-load period in the early morning, and the least squares method was used to update the model parameters. In this model, P_loss is the total power loss of the carport system, in watts (W); I is the total current flowing through the DC bus of the carport system, in amperes (A); a is the loss coefficient (variable loss coefficient) related to the square of the current, representing the loss part of the carport system that varies with the current. Its physical essence is the equivalent variable resistance of the carport system, in ohms (Ω); and b is the fixed damage coefficient independent of the current, representing the constant power loss in the carport system, in watts (W).

[0080] Step I: Effectiveness Evaluation The Energy Management System (EMS) dashboard displays real-time KPIs such as self-consumption rate, demand reduction rate, cost per service trip, power consumption / total demand (PF / THD) compliance rate, and average user waiting time. A / B testing is used to compare the operational data of the energy optimization strategy during on / off operation, and a T-test is used to verify the significance of the indicator improvements. Finally, an economic benefit analysis report is generated.

[0081] Specifically, in the step A to step B, data-driven prediction is employed, with step A forming the foundation for step B. The "historical feature sequences" (such as historical irradiance, power, and temperature) required for step B depend entirely on the data collection, cleaning, and generation in step A.

[0082] Step B to Step C: Prediction-Supported Optimization; Step B is a prerequisite for Step C. The outer optimization model (Step C) addresses a "forward-looking" problem: how to optimally allocate energy over a future period. Therefore, it relies on the predicted future power generation of perovskite photovoltaic modules and the electricity demand of charging piles, P_PV_forecast and P_EV_forecast, provided by Step B, as its core inputs.

[0083] Step C → Step D: The plan guides execution; Step C is the instruction source for Step D. The optimal plan derived in Step C is directly issued to the execution layer of Step D. The mission of Step D is to quickly and accurately track these setpoints and translate optimization decisions into actual action.

[0084] Step D → Step A: Execution feedback sensing; Step D is the data loop of Step A. During the execution of Step D, new real-time data reflecting the actual state of the system is generated. This data is collected again by Step A, thus initiating a new control cycle and forming a closed-loop feedback. This enables the carport system to perform rolling optimization based on the actual state and correct deviations in a timely manner.

[0085] Steps E, F, G, and H belong to the function enhancement and security module. These steps interact with the core closed loop, providing enhanced functionality and security guarantees. Step E (charging scheduling) is a refinement and supplement to step C. It receives the total charging power command from step C and allocates it fairly and efficiently to specific vehicles. The allocation results are then directly passed to step D for execution.

[0086] Step F (V2G strategy) is an extension of step C. It adds new schedulable resources (V2G vehicles) and triggering conditions to the optimization model, directly affecting the optimization decision in step C, and is implemented by controlling the bidirectional charging piles through step D.

[0087] Step G (Abnormal Handling) acts as the "safety guardian" and "stabilizer" of the carport system. It monitors the operational status of steps A, B, and D. When excessive prediction deviation is detected (abnormality in step B) or equipment failure (abnormality in step D), it directly intervenes and corrects the optimization strategy of step C or the execution logic of step D to ensure the safe and stable operation of the carport system under abnormal conditions.

[0088] Step H (parameter optimization) is the "self-learning" module of the carport system. It uses the long-term operating data collected in step A to periodically update the key cost model and system model in step C, making the optimization decisions closer to the actual system characteristics and achieving continuous performance improvement.

[0089] Step I (Effect Evaluation): A horizontal comparison and analysis of the key data collected in Step A under the two modes of "OPC enabled" (i.e., the entire process including steps A to H) and "OPC disabled" (traditional control) is conducted to quantitatively verify the comprehensive benefits of the entire OPC module algorithm.

[0090] The following specific embodiments further illustrate the perovskite photovoltaic energy storage and charging shed and its operation control method of the present invention.

[0091] Example 1: Perovskite Photovoltaic Storage and Charging Carport in Commercial Area This embodiment addresses the application scenario of a parking lot in a city's commercial complex by constructing a multifunctional perovskite photovoltaic energy storage and charging carport. The carport possesses photovoltaic power generation, energy storage regulation, fast and slow charging of electric vehicles, and bidirectional grid interaction functions, and utilizes the perovskite photovoltaic energy storage and charging carport and its operation control method as described in this invention.

[0092] Regarding the construction of the parking shed system, the configuration capacity of the parking shed was first determined based on the scale and electricity demand of the commercial area's parking lot. The parking lot averages approximately 200 vehicles parked simultaneously daily, with 20% requiring fast charging and 40% suitable for slow charging. Based on the site area and electricity characteristics, the installed capacity of perovskite photovoltaic module 1 was designed to be 200kWp. High-efficiency semi-transparent perovskite photovoltaic modules were used, with a single module power of 300W and a module efficiency of 20.2%, totaling 670 modules. Perovskite photovoltaic module 1 was divided into eight electrical zones, each with independently led-out positive and negative terminals connected to DC bus 8, ensuring both aesthetics and reducing the impact of localized shading on power generation.

[0093] The installed capacity of perovskite photovoltaic module 1 is calculated using the following formula: Photovoltaic capacity .

[0094] In the formula, For the capacity of perovskite photovoltaic modules, The number of perovskite photovoltaic modules, This refers to the nominal maximum power of a single publicly disclosed photovoltaic module. This represents the total loss factor of a perovskite photovoltaic module, including temperature rise, dust, line loss, and inverter efficiency reduction.

[0095] This parking shed uses high-efficiency, semi-transparent perovskite photovoltaic (PV) modules as the rooftop power generation units. The perovskite PV modules 1 are connected to the DC bus 8 via a zoned busbar system, allowing each zone to independently output power. Compared to traditional crystalline silicon modules, the perovskite PV modules 1 offer advantages in lightweight design and light transmission, achieving both high-efficiency power generation and aesthetic appeal while maintaining good natural lighting, making them particularly suitable for building-integrated applications. The zoned busbar structure further reduces the impact of localized shading on overall power generation efficiency, enhancing system stability.

[0096] Energy storage module 2 is equipped with a 400kW / 600kWh lithium iron phosphate battery. The operating SoC (System-on-Chips) range of energy storage battery 21 is 20%~90%, capable of supporting approximately 2~3 hours of full-power discharge. It can be used for peak shaving and valley filling, and can also provide clean energy for vehicles at night. Energy storage module 2 is also equipped with a bidirectional DC-DC converter with a rated power of 400kW and an efficiency of 96%. The capacity of energy storage battery 21 should be matched to peak shaving needs, backup time, and the peak power of the charging pile, with a recommended starting energy / power (E / P) ratio of 1~2 hours. The operating SoC window of energy storage battery 21 is 20~90%.

[0097] The rated voltage of DC bus 8 is preferably in the range of 750~1000Vdc, which is compatible with mainstream DC fast charging and PCS / inverter.

[0098] In terms of charging infrastructure, the facility is equipped with two 120kW DC fast charging piles, capable of providing fast charging services for four vehicles simultaneously, and six 11kW AC slow charging piles. The fast charging piles primarily serve vehicles with short-term parking, replenishing 60-80kWh of energy within 30-40 minutes; the slow charging piles provide low-power charging for vehicles parked for extended periods. All charging piles are interconnected with the control cabinet in the parking shed via the OCPP 1.6J protocol.

[0099] The grid-connected section uses a single 250kVA grid-connected bidirectional inverter 4, equipped with LCL filtering and reactive power regulation functions, meeting the requirements of a power factor ≥0.98 and total harmonic distortion ≤2% at the grid connection point. The grid-connected bidirectional inverter 4 can feed back electricity to the public grid 9 during peak electricity price periods and purchase electricity from the public grid 9 to supplement energy storage during off-peak electricity price periods, achieving economic optimization.

[0100] In terms of operational model control, the OPC module 71 proposed in this invention is used for optimization calculations. Data such as light intensity, ambient temperature, vehicle charging demand, energy storage battery SoC, and real-time electricity price are collected in real time through the data acquisition and preprocessing module 5. Using an artificial intelligence prediction model, the carport system can predict the power curve of the perovskite photovoltaic module 1 and the charging load of the charging pile 3 one hour in advance. For example, under sunny weather conditions, the output of the perovskite photovoltaic module 1 is expected to reach 180kW at noon; while based on vehicle reservations and historical data, the midday charging load is predicted to be 100kW. At night, the power generation of the perovskite photovoltaic module is essentially zero, with a peak charging load of 220kW.

[0101] OPC module 71 calculates the optimal power optimization strategy: During midday, of the predicted 180kW of power from the perovskite photovoltaic modules, 120kW is allocated to fast charging pile 3, 20kW to slow charging pile 3, and 40kW is stored in energy storage battery 21. Once energy storage battery 21 is full, any excess power is uploaded to the public grid 9. At night, energy storage battery 21 prioritizes power supply, with 120kW allocated to fast charging pile 3, 60kW to slow charging pile 3, and 40kW supplied by the public grid 9. During off-peak electricity prices, the power supply share of the public grid 9 is increased. Considering factors such as surplus energy in the energy storage battery and electricity costs, the power supply ratio between the public grid 9 and energy storage battery 21 is rationally allocated to get through the night. Through this allocation method, the carport system achieves the goals of prioritizing the consumption of perovskite photovoltaic modules, providing auxiliary compensation from energy storage batteries, and minimizing pressure on the public grid.

[0102] During operation, the OPC module 71 will re-predict and optimize every 10 minutes, and update the scheduling plan on a rolling basis. For example, if the weather changes suddenly in the afternoon and the output of the perovskite photovoltaic module 1 drops to 80kW, the carport system will immediately adjust its strategy, reduce the power of some slow charging piles 3, prioritize the output of fast charging piles 3, and at the same time schedule the energy storage battery 21 to discharge 100kW to balance the load.

[0103] After six months of operational data analysis, the self-consumption rate of the perovskite photovoltaic modules in the commercial parking sheds increased to 73%, peak electricity purchases decreased by 32%, and operating costs decreased by approximately 35%. The power factor at the grid connection point of the parking shed system remained consistently above 0.985, and the total harmonic distortion stabilized at around 1.8%, fully meeting power quality requirements.

[0104] In summary, this embodiment demonstrates that the carport system can not only achieve efficient power generation and clean charging in commercial areas, but also realizes intelligent scheduling and economical operation of multi-source energy through the OPC module 71, greatly improving the practicality and promotion value of the integrated photovoltaic, energy storage and charging carport.

[0105] Example 2: Perovskite Photovoltaic Storage and Charging Vehicle Shed in Industrial Park This embodiment addresses the application needs of a parking lot in a large industrial park by constructing a multi-functional perovskite photovoltaic energy storage and charging shed. The goal is to utilize the difference between the high photovoltaic power generation during the day and the low electricity price for charging at night to achieve peak shaving and valley filling of electricity and optimize overall costs.

[0106] Regarding the construction of the parking shed system, the overall configuration of the parking sheds was determined based on the scale of the park's parking lots and its electricity consumption characteristics. During peak weekday hours, the park can accommodate up to 500 vehicles, with approximately 25% requiring fast charging and the remaining 50% accepting slow charging. Considering the differences in daytime and nighttime electricity load within the park, the installed capacity of the perovskite photovoltaic module 1 was designed to be 500kWp. Large-area, high-efficiency perovskite photovoltaic modules were used, with each module having a power of 310W and an efficiency of 20.5%, totaling approximately 1610 modules. The perovskite photovoltaic module 1 was divided into twelve electrical zones, each independently connected to DC bus 8, achieving zoned power output and improving overall shading resistance and flexibility.

[0107] Energy storage module 2 uses a 1MW / 2MWh lithium iron phosphate battery system. The operating SoC range of energy storage battery 21 is 15%~90%, and it can continuously discharge for more than 2 hours. Energy storage module 2 is connected to DC bus 8 through a bidirectional DC-DC converter with a rated power of 1MW and a conversion efficiency of 97%. The purpose of configuring a large-capacity energy storage is to reduce the amount of electricity purchased during peak industrial loads during the day, and to charge the battery using public grid electricity during off-peak hours at night, preparing for operation the next day.

[0108] In terms of charging infrastructure, the facility is equipped with six 150kW DC fast charging piles and twenty 11kW AC slow charging piles, with a total charging capacity exceeding 1.1MW. The fast charging piles are used to quickly replenish the energy of vehicles belonging to senior management, logistics vehicles, and vehicles making short-term stops. The slow charging piles serve employee vehicles parked for extended periods, capable of replenishing 40-60kWh of energy during 6-8 hours of parking. All charging piles are connected to a control cabinet through a unified energy management system, enabling load allocation and priority scheduling.

[0109] The grid-connected section uses a 1.5MVA grid-connected bidirectional inverter 4, equipped with LCL filtering and low-voltage ride-through functions. It can flexibly choose to sell electricity to or purchase electricity from the public grid 9 according to the park's peak-valley electricity price. The power factor of the carport system at the grid connection point is maintained at ≥0.985, and the total harmonic distortion is ≤2.0%, meeting the stringent power quality requirements of the industrial power grid.

[0110] In terms of operational model control, OPC module 71 is used for optimization calculations. The carport system first collects real-time data from the park through data acquisition and preprocessing module 5, including solar irradiance, ambient temperature, energy storage battery SoC, grid electricity price, and vehicle charging demand. Based on the AI ​​prediction model, the carport system can predict the output of perovskite photovoltaic module 1 and vehicle charging load for the next 1-4 hours. For example, during the midday period on a sunny summer day, the predicted power of perovskite photovoltaic module 1 is 480kW, while the peak charging demand of charging pile 3 in the park reaches 600kW.

[0111] OPC module 71 combines the prediction results and optimization objectives to generate the optimal power optimization strategy. In the above scenario, the carport system is scheduled as follows: the output of perovskite photovoltaic module 1 is 480kW, of which 450kW is directly supplied to fast charging pile 3, and 30kW is allocated to slow charging pile 3. At the same time, the energy storage battery 21 discharges 120kW to meet the remaining charging load, ensuring that the vehicle can complete the recharging within the specified time without having to purchase additional electricity from the public grid 9.

[0112] When the industrial park enters off-peak electricity pricing hours (e.g., 11:00 PM to 6:00 AM), the perovskite photovoltaic module 1 has no output. The OPC module 71 automatically schedules the energy storage battery 21 to enter charging mode, purchasing 500kW of electricity from the public grid 9 at a low price for energy storage replenishment. In this way, before the peak industrial electricity consumption arrives the next day, the energy storage battery 21 is already in a high SoC state, able to output power during peak hours, thus playing a peak-shaving role.

[0113] Throughout the operation, the OPC module 71 updates the prediction and optimization results every 10 minutes and monitors the power factor, harmonic content, and charging pile queuing status in real time. If there is a sudden change in lighting or a concentration of vehicles arriving at the station, the carport system will automatically adjust the allocation strategy, such as temporarily reducing the power of some slow charging piles 3 to prioritize the charging speed of fast charging vehicles, while simultaneously scheduling the energy storage battery 21 to respond quickly and smooth out fluctuations.

[0114] After six months of operation and testing, the self-consumption rate of the perovskite photovoltaic modules 1 in the industrial park carport system increased to 78%, peak-hour power purchase decreased by approximately 40%, and total electricity expenditure in the park decreased by 28%. The power quality at the grid connection point met national standards, with the power factor remaining stable at around 0.99 and total harmonic distortion at approximately 1.6%. These data indicate that the carport system in this embodiment significantly improves energy utilization, reduces electricity costs, and enhances grid friendliness in an industrial park setting.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A perovskite photovoltaic energy storage and charging vehicle shed, comprising a parking shed and a shed system, characterized in that, The carport system includes perovskite photovoltaic modules, energy storage modules, charging piles, grid-connected bidirectional inverters, data acquisition and preprocessing modules, communication modules, and control cabinets. The perovskite photovoltaic modules are installed on the outer surface of the parking shed. The perovskite photovoltaic modules, energy storage modules, charging piles, grid-connected bidirectional inverters and control cabinets are interconnected by a DC bus. The energy storage modules include energy storage batteries, and the grid-connected bidirectional inverters are connected to the public power grid. The data acquisition and preprocessing module is used to collect real-time operating data from perovskite photovoltaic modules, energy storage modules, charging piles, and grid-connected bidirectional inverters, and to perform preprocessing on the collected raw data, including time alignment, missing data interpolation, and outlier removal; the preprocessed real-time data is transmitted to the control cabinet via the communication module. An OPC module is installed in the control cabinet. The OPC module dynamically predicts the power generation of the perovskite photovoltaic module and the power demand of the charging pile based on the received real-time data. Taking the minimization of the life cycle cost of the carport system as the objective function, the module generates an energy optimization strategy within the feasible domain defined by the constraints. The control cabinet controls the operation of the perovskite photovoltaic module, energy storage module, charging pile and grid-connected bidirectional inverter through the communication module according to the energy optimization strategy.

2. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 1, characterized in that, The objective function is: Formula 1 In the formula, min Cost To minimize costs, This is the electricity purchase cost coefficient. For the power purchase capacity, This is the electricity sales cost coefficient. For electricity sales capacity, For the equivalent cost of degradation of perovskite photovoltaic modules, This represents the theoretical degradation of the perovskite photovoltaic module's health status at hour h, caused by the current charge / discharge operation and factored towards that hour. The penalty coefficient is... This represents a penalty for delayed charging or exceeding the queue time.

3. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 1, characterized in that, The constraints include power quality constraints, energy storage boundary constraints, power balance constraints, charging service quality constraints, and constraints for connection to the public power grid.

4. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 3, characterized in that, The power quality constraints are: power factor PF ≥ 0.98 and total harmonic distortion (THD) ≤ 2%.

5. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 3, characterized in that, The energy storage boundary constraints are as follows: Formula 2 Formula 3 In the formula, This represents the lower limit of energy storage capacity for energy storage batteries. This represents the real-time energy storage status of the energy storage battery. This represents the upper limit of energy storage capacity for energy storage batteries. This refers to the real-time discharge power of the energy storage battery. This represents the upper limit of the discharge power of the energy storage battery.

6. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 3, characterized in that, The power balance constraint condition is: Formula 4 In the formula, For the power generation of perovskite photovoltaic modules, This refers to the discharge power of the energy storage battery. Power purchased from the public power grid, The power consumed to charge the charging station. For auxiliary systems or conventional load power, This refers to the power loss of the carport system.

7. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 3, characterized in that, The charging service quality constraint is: for each vehicle i, the minimum energy replenishment must be satisfied before the scheduled or expected departure time di. .

8. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 3, characterized in that, The constraints for connecting to the public power grid include grid connection point restrictions, grid connection power restrictions, grid connection current restrictions, grid connection voltage limits, and ramp rate restrictions.

9. The perovskite photovoltaic energy storage and charging vehicle shed as described in claim 1, characterized in that, The carport system also includes a monitoring and feedback module. When the monitoring and feedback module detects that the difference between the actual operating data of the carport system and the predicted results exceeds the limit, it feeds the information back to the control cabinet. The OPC module then recalculates and updates the power optimization strategy.

10. A method for operating and controlling a perovskite photovoltaic energy storage and charging vehicle shed as described in any one of claims 1 to 9, characterized in that, Includes the following steps: Step A: Data Acquisition and Preprocessing The data acquisition and preprocessing module periodically collects real-time operating data from perovskite photovoltaic modules, energy storage modules, charging piles, and grid-connected bidirectional inverters, and performs preprocessing on the data, including time alignment, missing data interpolation, and outlier removal. The preprocessed real-time data is then transmitted to the control cabinet via the communication module. Step B: Short-term forecasting Predict the power generation of perovskite photovoltaic modules and the charging power demand of charging piles; Step C: Optimize the model With minimizing the life cycle cost of the carport system as the core objective function, the OPC module is used to generate an energy optimization strategy within the feasible region defined by the constraints. Step D: Execute control The control cabinet controls the operation of perovskite photovoltaic modules, energy storage modules, charging piles, and grid-connected bidirectional inverters through power generation optimization strategies via the communication module.

11. The operation control method for the perovskite photovoltaic energy storage and charging vehicle shed as described in claim 9, characterized in that, The operation control method further includes the following steps: Step E: Charging Scheduling Build a dynamic priority charging queue and update the task order regularly; monitor the completion rate of charging piles and automatically increase the charging priority of lagging vehicles when the ratio difference exceeds the threshold; when the charging power is insufficient, activate the degradation strategy: prioritize the 3kW basic charging power of all vehicles, and allocate the remaining power to fast charging piles according to priority.

12. The operation control method for the perovskite photovoltaic energy storage and charging vehicle shed as described in claim 9, characterized in that, The operation control method further includes the following steps: Step F: V2G Strategy By setting up a triple triggering condition of price-driven, demand-controlled, and user-authorized charging, a special energy storage model is established for V2G vehicles in the power optimization strategy. The attenuation cost coefficient is set to 1.5 times that of conventional energy storage. The discharge curve is controlled by bidirectional charging piles to ensure that the SoC of the return trip is not lower than the user-set value.

13. The operation control method for the perovskite photovoltaic energy storage and charging vehicle shed as described in claim 9, characterized in that, The operation control method further includes the following steps: Step G: Exception Handling The steady-state priority mode in the power optimization strategy is activated when the prediction deviation is >20%, which limits the charging peak of the charging pile, improves reactive power support, and temporarily raises the lower limit of electricity purchase. The safety backoff mode is activated when the communication module is interrupted or the temperature of the charging pile exceeds the limit. The energy storage module switches to conservative power or standby, and the charging pile enters the current limiting state to maintain grid connection compliance.

14. The operation control method for the perovskite photovoltaic energy storage and charging vehicle shed as described in claim 9, characterized in that, The operation control method further includes the following steps: Step H: Parameter Optimization Initially, the degradation coefficient was fitted based on the cycle life curve of perovskite photovoltaic modules. A line loss model, P_loss = a·I² + b, was established through light and heavy load tests. The model was automatically recalibrated weekly during the low-load period in the early morning, and the least squares method was used to update the model parameters. In this model, P_loss is the total power loss of the carport system, in watts (W); I is the total current flowing through the DC bus of the carport system, in amperes (A); a is a loss coefficient related to the square of the current, representing the loss part of the carport system that varies with the current. Its physical essence is the equivalent variable resistance of the carport system, in ohms (Ω); and b is a fixed damage coefficient independent of the current, representing the constant power loss in the carport system, in watts (W).

15. The operation control method for the perovskite photovoltaic energy storage and charging vehicle shed as described in claim 9, characterized in that, The operation control method further includes the following steps: Step I: Effectiveness Evaluation The system displays real-time KPIs such as self-consumption rate and demand reduction rate. It compares the operating data of the power optimization strategy during on / off states through A / B testing, verifies the significance of the indicator improvements using T-tests, and finally generates an economic benefit analysis report.

Images