A conventional-power-off dual-purpose building cluster light storage and charging collaborative optimization method

Abstract

This invention belongs to the field of building cluster energy optimization and scheduling technology, and discloses a collaborative optimization method for building cluster photovoltaic, energy storage, charging, and utilization that is applicable to both normal and power outage conditions. The specific steps are as follows: S1, data collection and preprocessing to obtain continuous time series; S2, electricity consumption characteristic analysis and behavioral modeling; S3, source-load prediction of the building cluster photovoltaic, energy storage, charging, and utilization power system; S4, performance evaluation of the building cluster photovoltaic, energy storage, charging, and utilization power system; S5, power load-side regulation and management of the building cluster; S6, optimized scheduling of the building cluster photovoltaic, energy storage, charging, and utilization power system. The method provided by this invention can effectively coordinate various resources such as photovoltaic, energy storage, charging, and utilization, integrate advanced technologies such as mobile energy storage, power outage prediction, and smart charging for electric vehicles, and take into account multiple dimensions such as economy, resilience, and low carbon emissions, to achieve optimized scheduling of the building cluster photovoltaic, energy storage, charging, and utilization power system under both normal and power outage conditions.

Description

Technical Field

[0001] This invention relates to the field of energy optimization and scheduling technology for building complexes, specifically to a collaborative optimization method for photovoltaic, energy storage, charging, and utilization in building complexes that is applicable to both normal and power outage scenarios. Background Technology

[0002] Currently, with the gradual maturation of technologies such as renewable energy power generation and new energy vehicles, integrated power systems combining photovoltaics, energy storage, charging, and utilization have become an important form of energy systems for building complexes such as industrial parks, commercial complexes, and residential communities. These systems typically include the municipal main power grid, distributed photovoltaic systems, battery systems, and related supporting components, and encompass various energy consumers such as buildings and electric vehicles. Coordination between energy sources, storage, loads, and the grid is achieved through energy management, scheduling, and control.

[0003] For the optimized scheduling of integrated "photovoltaic-storage-charging-utilization" power systems in building complexes, engineers have proposed numerous related methods. These methods typically use the configuration, location, and operational status of various components within the energy system as decision variables, cost, carbon emissions, and renewable energy utilization rate as objective functions, and energy conservation, equipment power range, and energy storage cycle count as constraints. Solvers include physical simulation software such as HOMER, various types of mathematical programming, heuristic optimization algorithms, and reinforcement learning algorithms. Some engineers also emphasize the importance of load-side regulation and demand response, proposing strategies such as optimizing air conditioning temperature setpoints, cold and heat storage in building envelopes, intelligent charging and discharging of electric vehicles, and power management for lighting sockets.

[0004] However, with the large-scale integration of renewable energy, the extensive access of load-side flexible and adjustable resources, and the intensification of the global climate and energy crisis, building cluster power systems need to better cope with issues such as unstable power supply and demand, and even short-term power outages. Most existing optimal dispatching methods are designed only for normal grid-connected operation, with economic efficiency (such as minimizing electricity costs) as their objective. When grid faults or planned power outages occur, these methods often fail to work or require switching to completely independent emergency control strategies. There is a lack of a unified, adaptive dual-condition optimization framework to coordinate and switch the objective function and decision variables of optimal dispatching. Furthermore, existing optimal dispatching methods often fail to fully integrate rapidly iterating advanced technologies such as mobile energy storage (typically manifested as mobile energy storage vehicles, integrating mobile batteries into trucks, which can temporarily and quickly travel to the faulty area after grid failures or other power outages to provide alternative main power), power outage prediction, and smart charging for electric vehicles. Summary of the Invention

[0005] To address the problems encountered by the aforementioned methods, this invention proposes a collaborative optimization method for building clusters with photovoltaic, energy storage, charging, and utilization that is applicable in both normal and power outage scenarios. This method decouples the integrated optimization and scheduling of "photovoltaic-storage-charging-utilization" into optimizing and managing building power consumption and electric vehicle charging first, and then optimizing the joint operation of photovoltaic and battery systems based on the adjusted load. Under normal conditions, it minimizes overall costs and carbon emissions; under power outage conditions, it minimizes overall costs, including power outage losses, and ensures that resilience indicators exceed minimum limits, while integrating mobile energy storage technology and power outage prediction technology.

[0006] The technical problem to be solved by the present invention is to provide a collaborative optimization scheduling method for building clusters of light-storage-charging-utilization that takes into account multiple dimensions such as economy, resilience, and low carbon, fully coordinates the resources of the supply side, demand side, and storage side, and is applicable to both normal and power outage dual-operation conditions.

[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: a collaborative optimization method for photovoltaic, energy storage, and charging applications in building complexes that is applicable in both conventional and power outage scenarios, comprising the following steps: S1. Collect and preprocess data to obtain continuous time series; S2. Based on continuous time series, conduct electricity consumption characteristic analysis and behavior modeling to obtain electricity consumption characteristics and behavior probabilities; S3. Using electricity consumption characteristics and behavioral probabilities as inputs, predict the source load of the photovoltaic-storage-charging-utilization power system in the building complex; S4. Evaluate the performance of the photovoltaic-storage-charging power system in a building complex based on the predicted source load of the power system. S5. Using the performance of the photovoltaic, energy storage, charging and utilization power system of the building complex as the objective function and constraint, conduct power load-side regulation and management of the building complex; S6. Based on the adjusted power load of the building complex, optimize the scheduling of the photovoltaic, energy storage, charging and utilization power system of the building complex.

[0008] The collected data includes data on the relationship between power generation and consumption, indoor environmental data, and outdoor meteorological data; The power generation and consumption correlation data includes the total hourly power consumption of the building complex, the total number of people in the building per hour, the total hourly power consumption and duration of air conditioning, the total hourly power consumption and duration of lighting, the total hourly power consumption and duration of sockets, the total charging amount of electric vehicles through charging piles within the building complex and the charging time of electric vehicles, and the hourly power generation of the building complex's rooftop photovoltaic system; indoor environmental data includes indoor air temperature and indoor carbon dioxide concentration; outdoor meteorological data includes outdoor air temperature, outdoor solar radiation intensity, and outdoor cloud cover; all data are aligned by time to form a continuous time series.

[0009] The electricity consumption characteristic analysis includes the electricity consumption characteristics of building lighting and sockets; Building lighting electricity and socket electricity are divided into adjustable electricity and non-adjustable electricity; adjustable electricity is electricity that can be reduced, stopped or transferred, and non-adjustable electricity is electricity used for equipment related to safety and health.

[0010] The behavioral modeling includes building air conditioning usage behavior modeling and electric vehicle charging behavior modeling; The building air conditioning usage behavior modeling is based on the strong influence of indoor and outdoor air temperature on people's use of air conditioning, and the probability of usage behavior is calculated through air conditioning usage time data; Based on indoor air temperature data, the probability of air conditioning usage behavior is fitted according to the following formula. : The probability of heating air conditioner usage behavior is fitted using the following formula. : in, The threshold for triggering air conditioning usage behavior. The threshold for activating heating air conditioner usage behavior. Indoor air temperature k , l , τ All of these are fitting parameters, representing the sensitivity of the personnel, the dimensionless constant, and the time scale parameter, respectively; The electric vehicle charging behavior modeling is based on fitting the overall number of charging cycles within the building complex using the following formula: in, N This represents the total number of charging cycles within the building complex. The fitting parameters for the total number of charging cycles. n This represents the number of peaks in a multimodal Gaussian distribution. t The moment when electric vehicles begin charging. , and These are the fitting parameters corresponding to a single peak, representing the amplitude, mean, and standard deviation, respectively.

[0011] The power source and load of the building complex's photovoltaic, energy storage, and charging system include electric vehicle charging demand, building electricity demand, photovoltaic output, and battery charging and discharging. The electric vehicle charging demand is predicted based on the Monte Carlo stochastic simulation method; the actual number of electric vehicles and battery performance parameters are input; the average daily mileage of each electric vehicle is randomly generated using the following formula based on a log-normal distribution: in, The average daily driving distance of electric vehicles x The corresponding probability density function, x It is the average daily mileage of electric vehicles. and These are the expected value of the average daily mileage of electric vehicles and the standard deviation of the average daily mileage of electric vehicles, respectively. The charging start time for each electric vehicle is randomly generated using the following formula: (Multimodal Gaussian distribution is used) in, It is the moment when electric vehicle charging begins. t The corresponding probability density function, and These are the expected value and the standard deviation of the electric vehicle charging start time, respectively. Assuming a constant charging power, the charging time for each electric vehicle is calculated based on the electricity consumption, charging power, and charging efficiency corresponding to the average daily mileage of each electric vehicle. The charging situation of each electric vehicle is simulated one by one through a Monte Carlo cycle to obtain the final total charging demand of the electric vehicles. in, Charging power for electric vehicles, This is the standard charging power of the charging station. For charging duration, M For the number of electric vehicles, For the first i The charging status of an electric vehicle, during charging =1, when not charging =0, It refers to the state of charge of an electric vehicle battery. It refers to charging efficiency; The building's electricity demand is predicted using a long short-term memory network time-series forecasting method. The input data includes the previous hour's historical energy consumption, time characteristics, indoor air temperature, indoor carbon dioxide concentration, outdoor air temperature, outdoor solar radiation intensity, outdoor cloud cover, and number of people in the building. The long short-term memory network time-series forecasting method is used to predict the building's electricity demand and output the building's electricity consumption. The photovoltaic output is calculated according to the following formula: in, This represents the real-time solar radiation intensity. Photovoltaic power generation capacity, It is the output power of a photovoltaic power generation system under standard conditions. =25℃, =1000W / m 2 , It is the temperature power coefficient. It's the temperature of the photovoltaic panel. It refers to the ambient temperature when the photovoltaic panels are working. This refers to the rated operating temperature of the photovoltaic panel. It is the derating factor of a photovoltaic power generation system, which is the percentage decrease in the amount of electricity generated in the current year compared to the amount of electricity generated in the first year; The charging and discharging of the storage battery is calculated using a kinetic energy battery model, as shown in the following formula: in, and They represent t Battery capacity at any time and ( t -1) Battery capacity at time point and Let represent the battery charging power at time t and the battery discharging power at time t, respectively. and These represent inverter efficiency and battery efficiency, respectively. and These represent the battery's charging efficiency and discharging efficiency, respectively. This indicates the operating status of the battery pack. It is 1 when charging, 0 when neither charging nor discharging, and -1 when discharging.

[0012] The performance of the photovoltaic, energy storage, charging and utilization power system of the building complex includes investment cost, energy cost, carbon emissions and environmental protection cost, penalties and rewards, and system resilience; The investment cost is shown in the following formula: in, It is the average annual discounted cost of the total investment. and These correspond to the average annual depreciated investment costs of photovoltaic power generation systems, electric vehicle charging piles, battery packs, and mobile energy storage vehicles, respectively. Represents the annualized interest rate. Represents the total lifespan in years; Energy costs include the cost of purchasing electricity from the grid for the building complex's photovoltaic, energy storage, and charging power system, and the cost of purchasing electricity from mobile energy storage vehicles for the same system. The cost of purchasing electricity from the grid for the building complex's photovoltaic, energy storage, and charging power system is calculated using the following formula: in, It is the average daily cost of purchasing electricity through the power grid. and These refer to the electricity price that buildings pay for electricity purchased from the grid and the electricity price that electric vehicle charging stations pay for electricity purchased from the grid, respectively. and These refer to the electricity purchased by buildings from the grid and the electricity purchased by electric vehicle charging stations from the grid, respectively. and These are the start and end times of the measurement, respectively. The cost of purchasing electricity via mobile energy storage vehicles for a building complex's photovoltaic, energy storage, and charging power system is calculated using the following formula: in, This is the average daily cost of purchasing electricity through mobile energy storage vehicles. and These refer to the unit price of electricity purchased through mobile energy storage vehicles and the total amount of electricity purchased through mobile energy storage vehicles, respectively. and These refer to the hourly labor cost and the hourly hardware cost associated with deploying mobile energy storage vehicles. This refers to the time when the mobile energy storage vehicle is deployed; Carbon emissions and environmental protection costs are the associated costs of treating pollutants and carbon dioxide emissions from building complex photovoltaic-storage-charging power systems that purchase electricity from non-clean energy power plants. These costs are calculated using the following formula: in, This is the average daily cost used for pollutant and carbon dioxide treatment. For the first q The cost of treating these emissions It is the first q Emission coefficients of various pollutants This refers to the total amount of electricity purchased through non-clean energy power plants. q Refers to the types of emissions; The penalties and rewards are calculated by offsetting renewable energy rewards and power outage losses for the photovoltaic-storage-charging power system of the building complex, using the following formula: in, It is the average daily cost after deducting rewards and penalties. This refers to the amount of renewable energy generated that is awarded as a reward. The incentive coefficient for renewable energy generation. This refers to the amount of electricity lost during a power outage. This is the power outage penalty coefficient; System resilience is calculated using the following formula: in, R It is the system power supply resilience index. These refer to photovoltaic power generation, battery charging and discharging, grid power supply, mobile energy storage vehicle power supply, building power supply, and electric vehicle charging power, respectively.

[0013] Power load regulation in building complexes is divided into load regulation under normal operating conditions and load regulation under power outage conditions; Under normal operating conditions, load regulation is based on time-of-use peak and valley electricity pricing, and parameter adjustments are made to electric vehicles, air conditioners, some lighting, and some sockets, including: adjusting the charging start time of electric vehicles, optimizing the start threshold of cooling or heating air conditioner usage behavior, and transferring some available power. Load regulation during power outages is based on safety rules and power supply resilience. It involves adjusting parameters for electric vehicles, air conditioners, some lighting, and some sockets, including: ceasing charging of electric vehicles, further optimizing the start threshold for air conditioner cooling or heating, and further diverting available power.

[0014] The optimized scheduling of the photovoltaic, energy storage, charging, and utilization power system for building complexes specifically includes: S6.1 Optimize the construction of scheduling constraints; The constraint equations for the optimal scheduling of the photovoltaic, energy storage, charging, and power system of the building complex are constructed, including: the state of charge of the batteries and electric vehicle batteries is between 10% and 90% throughout the entire process; the photovoltaic power generation is less than the rated maximum value; the electric vehicle charging power is less than the rated maximum charging power of the charging pile; the power supply of the mobile energy storage vehicle is less than the rated maximum value; the charging and discharging power of the battery pack is less than the rated maximum value; the total charging and discharging amount of the battery every 24 hours is less than twice the maximum battery capacity; and system resilience. R Always greater than or equal to 0; S6.2 Optimize the construction of the scheduling solver; A solver for the optimal scheduling of the photovoltaic-storage-charging-utilization power system of the building complex is constructed. The jellyfish search algorithm is selected, and the population is initialized using the Logistic chaotic mapping. An adversarial learning strategy is introduced to generate inverse solutions to expand the search space. Levy flight is incorporated to enhance the global exploration capability and allow for sporadic large step updates. S6.3 Optimized scheduling under normal operating conditions; The optimized scheduling process under normal operating conditions involves load adjustment under normal operating conditions, followed by hourly optimization of the photovoltaic-battery operation based on the adjusted load. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, and grid power supply. The objective function is the daily total depreciated cost, calculated as follows: in, This is the total daily discounted cost under normal operating conditions; S6.4 Optimized scheduling under power outage conditions with power outage prediction; The optimized scheduling process under power outage conditions with outage prediction is as follows: First, load adjustment is performed under the power outage condition. During the off-peak electricity price period before the expected power outage time, the battery is fully charged and not restarted. Hourly photovoltaic operation optimization is performed based on the adjusted load, and the battery pack capacity is released at the time of the power outage. If the minimum electricity demand is still insufficient, an external mobile energy storage vehicle is needed to supplement the power. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, grid power supply, and mobile energy storage vehicle power supply. The objective function is the converted total operating cost on the power outage day, calculated as follows: in, It is the total operating cost per day of power outage under power outage conditions with power outage forecasting; S6.5 Optimized scheduling under power outage conditions without power outage prediction; The optimized scheduling process under power outage conditions without power outage prediction is as follows: First, load adjustment is performed under the power outage conditions. Then, hourly photovoltaic-battery-mobile energy storage vehicle operation optimization is performed based on the adjusted load. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, grid power supply, and mobile energy storage vehicle power supply. The objective function of the optimization is the converted total operating cost on the power outage day, calculated as follows: in, It is the total operating cost per day of power outage under power outage conditions with power outage forecasting.

[0015] This method can better coordinate various resources such as photovoltaic, energy storage, charging, and utilization, integrate advanced technologies such as mobile energy storage, power outage prediction, and smart charging of electric vehicles, and take into account multiple dimensions such as economy, resilience, and low carbon, so as to achieve optimized scheduling of photovoltaic, energy storage, charging, and utilization power systems in building clusters under both normal and power outage conditions.

[0016] The beneficial effects of the present invention are as follows: The present invention provides a method for the coordinated optimization of photovoltaic, energy storage, charging and utilization power systems in building complexes that is applicable to both normal and power outage conditions. It can better coordinate various resources such as photovoltaic, energy storage, charging and utilization, integrate advanced technologies such as mobile energy storage, power outage prediction and smart charging of electric vehicles, and take into account multiple dimensions such as economy, resilience and low carbon, so as to realize the optimized scheduling of photovoltaic, energy storage, charging and utilization power systems in building complexes under both normal and power outage conditions. Attached Figure Description

[0017] Figure 1Flowchart of a collaborative optimization method for photovoltaic, energy storage, and charging systems in building complexes, applicable to both normal and power outage scenarios.

[0018] Figure 2 A schematic diagram of the power system for photovoltaic energy storage and charging in the building complex.

[0019] Figure 3 This is a schematic diagram of power system resilience assessment indicators.

[0020] Figure 4 This is a diagram illustrating the adjustment of air conditioner usage behavior.

[0021] Figure 5 This is a schematic diagram illustrating the electricity consumption characteristics of an air conditioner in a real-world case.

[0022] Figure 6 This is a schematic diagram illustrating the time-series characteristics of electric vehicle charging times in a real-world case.

[0023] Figure 7 This is a schematic diagram illustrating the timing characteristics of power consumption for lighting sockets in a real-world case.

[0024] Figures 8-11 This is a schematic diagram illustrating the optimized dispatching results of the power system under normal operating conditions on a certain day in autumn. Figure 8 The result is the result of running without any adjustments. Figure 9 This refers to the operational results when only load regulation and management are performed. Figure 10 This refers to the operational results when only photovoltaic-battery optimization scheduling is performed. Figure 11 To simultaneously consider the operational results when load regulation and management, as well as the optimized scheduling of photovoltaic-battery systems.

[0025] Figures 12-16 This is a schematic diagram illustrating the optimized power system dispatch results on a certain autumn day under power outage conditions. Figure 12 The result is the result of running without any adjustments. Figure 13 This refers to the operational results when only load regulation and management are performed. Figure 14 To simultaneously consider the operational results of load regulation and management, and photovoltaic-battery optimal scheduling, Figure 15 To simultaneously consider the operational results of load regulation and management, photovoltaic-battery optimized scheduling, and mobile energy storage integration, Figure 16 To simultaneously consider the operational results when load regulation and management, photovoltaic-battery optimized scheduling, mobile energy storage system access, and power outage prediction. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the specific embodiments of this invention are described in detail below with reference to the invention content and accompanying drawings. In this invention, a building complex with 70 different buildings is used as an example, including 400 electric vehicles and sufficient charging piles, a rooftop photovoltaic system (covering the flat roofs of all buildings), a battery pack (rated power 3 MW, capacity 6 MWh), and a mobile energy storage vehicle (rated power 500 kW, capacity 1 MWh). It is assumed that there is a power outage period from 16:00 to 19:00. This invention mainly includes the following steps: S1. Collect and preprocess relevant data; S1.1 Collect data related to power generation and consumption; Collect data on the total hourly electricity consumption of the building complex, the total hourly electricity consumption of air conditioning and the duration (in minutes) of air conditioning use, the total hourly electricity consumption of lighting and the duration (in minutes) of lighting use, the total hourly electricity consumption of sockets and the duration (in minutes) of socket use, the total charging amount and charging time (in minutes) of electric vehicles through charging piles within the building complex, and the hourly power generation of the rooftop photovoltaic system of the building complex. S1.2, Collect indoor environmental data; Collect hourly indoor air temperature and carbon dioxide concentration in the building; S1.3, Collect outdoor meteorological data; Collect hourly outdoor air temperature, net solar radiation intensity, and cloud cover at the building's location; S1.4 Data preprocessing; Align all data by time to form a continuous time series; S2. Electricity consumption characteristic analysis and behavioral modeling; S2.1 Building air conditioning usage behavior modeling; People's use of air conditioning is strongly influenced by indoor and outdoor air temperatures. The probability of air conditioning use can be calculated using the air conditioning usage duration data collected in S1.1. Combined with indoor temperature data, the probability of air conditioning use can be fitted using the following formula: The probability of heating air conditioner usage behavior is fitted using the following formula. : in, The threshold for triggering air conditioning usage behavior. The threshold for triggering heating air conditioner usage behavior refers to the temperature at which the behavior will be triggered when the temperature is above or below this threshold (for example, the cooling threshold for air conditioners can be set to 26°C). For temperature, k , l , τ All of these are fitting parameters, representing the sensitivity of the personnel, the dimensionless constant, and the time scale parameter, respectively; S2.2 Modeling of electric vehicle charging behavior; The charging frequency of electric vehicles by individuals is often influenced by multiple factors such as electricity prices, work and rest schedules, and battery performance, exhibiting a multi-peak Gaussian distribution with temporal characteristics within a day. The following formula is used to fit the overall charging frequency within the building complex: in, N This represents the total number of charging cycles within the building complex. The fitting parameters for the total number of charging cycles. n This represents the number of peaks in a multimodal Gaussian distribution. , and These are the fitting parameters corresponding to a single peak, representing the amplitude, mean, and standard deviation, respectively. In this specific implementation case, the three actual peak charging start times (the times corresponding to the peak charging counts) were 15.09 (approximately 15:05), 19.62 (approximately 19:37), and 11.84 (approximately 11:50), respectively, and the proportions of the three peak times were 7.70%, 47.52%, and 44.79%, respectively. S2.3 Analysis of the characteristics of electricity consumption for building lighting and sockets; The electricity used for lighting and sockets in buildings is divided into adjustable electricity and non-adjustable electricity. Adjustable electricity includes electricity that can be reduced, stopped, or transferred (such as electricity used for washing machines), while non-adjustable electricity includes electricity that is strongly related to safety and health and is not suitable for adjustment (such as electricity used for safety exit indicator lights). In this specific implementation case, taking the lighting socket system of a residential building as an example, the actual adjustable power consumption accounts for 15%-70% of the total power consumption, which is greatly affected by the time of day. However, when only the period from 9:00 to 21:00 is counted, the proportion is concentrated in the range of 50%-70%. S3. Source load prediction or simulation of power system for photovoltaic, energy storage and charging in building complexes; S3.1 Simulation of electric vehicle charging demand; Electric vehicle charging load modeling is based on the Monte Carlo stochastic simulation method; First, input basic information such as the actual number of electric vehicles and battery performance parameters; Secondly, the average daily mileage of each electric vehicle is randomly generated using the following formula based on a log-normal distribution: in, The average daily driving distance of electric vehiclesx The corresponding probability density function, x It is the average daily mileage of electric vehicles. and These are the expected value and standard deviation of the average daily mileage of electric vehicles, respectively. Next, the charging start time for each electric vehicle is randomly generated using the following formula, employing a multi-peak Gaussian distribution: in, It is the moment when electric vehicle charging begins. t The corresponding probability density function, and These are the expected value and the standard deviation of the electric vehicle charging start time, respectively. Finally, assuming a constant charging power, the charging time of a single electric vehicle is calculated based on the power consumption, charging power, and charging efficiency corresponding to the average daily mileage of each electric vehicle. The charging situation of each electric vehicle is simulated one by one through a large number of Monte Carlo cycles to obtain the final total charging demand of electric vehicles. in, Charging power for electric vehicles, This is the standard charging power of the charging station. For charging duration, M For the number of electric vehicles, For the first i The charging status of an electric vehicle, during charging =1, when not charging =0, It refers to the state of charge of an electric vehicle battery. It refers to charging efficiency; In this specific implementation case, the coefficient of determination for the charging demand simulation is 0.9742; S3.2 Building electricity demand forecast; Building electricity demand forecasting adopts the Long Short-Term Memory (LSTM) time series forecasting method. The input data includes the historical energy consumption of the previous hour, time characteristics, indoor air temperature, indoor carbon dioxide concentration, outdoor air temperature, outdoor solar radiation intensity, outdoor cloud cover, and number of people in the building. The data is trained using the LSTM algorithm to output the building electricity consumption. In this specific implementation case, the mean absolute percentage error (MAPE) of a single building ranges from 4.50% to 10.31%, the root mean square error coefficient of variation (CV-RMSE) ranges from 0.0617 to 0.1466, and the coefficient of determination ranges from 0.8932 to 0.9254. After integrating all the individual buildings into a building group, the overall MAPE is 3.15%. S3.3 Calculation and verification of photovoltaic output; The power generation capacity of a photovoltaic panel is related to its own performance and environmental characteristics such as solar radiation intensity and operating temperature. The power generation capacity of a single photovoltaic panel can be calculated using the following formula: in, This represents the real-time solar radiation intensity. Photovoltaic power generation capacity, It is a photovoltaic power generation system under standard conditions (i.e. =25℃, =1000W / m 2 ) output power, It is the temperature power coefficient. It's the temperature of the photovoltaic panel. It refers to the ambient temperature when the photovoltaic panels are working. This refers to the rated operating temperature of the photovoltaic panel. It is the derating factor of a photovoltaic power generation system, which is the percentage decrease in the amount of electricity generated in the current year compared to the amount of electricity generated in the first year. In this specific implementation case, the rooftop photovoltaic panel is 1.1m long, installed at a south-facing angle of 32.75°, the solar altitude angle of the area is 27.75°, and the total roof area is 126,573.18m². 2 The usable flat roof area is 56,843.82 m². 2 (Approximately 44.91%) Output power of photovoltaic systems under standard operating conditions It has a power coefficient of 0.2 kW and a temperature power coefficient. The derating factor for a photovoltaic power generation system is 0.0047 / ℃. It is 80%; S3.4 Calculation and verification of battery charging and discharging; The charging and discharging of the battery is calculated using the KiBaM model (kinetic energy battery model), as shown in the following formula: in, and They represent t Time and ( t -1) Battery capacity at time point and These represent the charging and discharging power of the battery at time t, respectively. and These represent inverter efficiency and battery efficiency, respectively. and These represent the battery's charging efficiency and discharging efficiency, respectively. This indicates the operating status of the battery system (1 when charging, 0 when neither charging nor discharging, and -1 when discharging). S4. Performance evaluation of the photovoltaic, energy storage and charging power system for building complexes; S4.1 Investment cost assessment; The average annual cost of the total investment in the photovoltaic, energy storage, charging, and power system for the building complex can be expressed by the following formula: in, It is the average annual discounted cost of the total investment. and The average annual depreciated costs of investment for photovoltaic systems, electric vehicle charging stations, battery systems, and mobile energy storage systems (including trucks, on-board batteries, power transmission interfaces, and labor costs) are respectively. Represents the annualized interest rate. Represents the total lifespan in years; S4.2 Energy Cost Assessment; The cost of purchasing electricity from the grid for a building complex's photovoltaic, energy storage, and charging power system can be calculated using the following formula: in, It is the average daily cost of purchasing electricity through the power grid. and These refer to the electricity prices that buildings and electric vehicle charging stations purchase from the grid. and These refer to the electricity purchased from the grid by buildings and electric vehicle charging stations, respectively. and These are the start and end times of the measurement, respectively. The cost of purchasing electricity via mobile energy storage vehicles for a building complex's photovoltaic, energy storage, and charging power system can be calculated using the following formula: in, This is the average daily cost of purchasing electricity through mobile energy storage vehicles. and These refer to the unit price and total amount of electricity purchased through mobile energy storage vehicles. and These refer to the hourly labor cost and hardware cost (including battery wear and tear, truck fuel consumption, etc.) incurred by the deployment of mobile energy storage vehicles. and These are the start and end times of the measurement, respectively. This refers to the time when the mobile energy storage vehicle is deployed; S4.3 Assessment of carbon emissions and environmental protection costs; When a building complex's photovoltaic, energy storage, and charging power system purchases electricity from non-clean energy power plants, it generates pollutants and carbon dioxide emissions. The associated remediation costs can be calculated using the following formula: in, This is the average daily cost used for pollutant and carbon dioxide treatment. For the first q The cost of treating a type of emission (referring to the average cost of treating each gram of pollutant). It is the first q The emission factor of a type of emission (referring to the average weight of emissions generated per kilowatt-hour of non-clean energy electricity). This refers to the total amount of electricity purchased through non-clean energy power plants. q Refers to the types of emissions (usually including CO2, CO, NO) x SO2, soot). and These are the start and end times of the measurement, respectively. S4.4, Evaluation of Punishments and Rewards; The renewable energy incentives and power outage losses for a building complex's photovoltaic, energy storage, and charging power system can be calculated using the following formula: in, It is the average daily cost after deducting rewards and penalties. This refers to the renewable energy generation that can be rewarded. The incentive factor for renewable energy generation (the incentive amount per kilowatt-hour). This refers to the amount of electricity lost during a power outage. This is the power outage penalty coefficient (the actual penalty and loss incurred per kilowatt-hour). and These are the start and end times of the measurement, respectively. S4.5 System resilience assessment; The resilience of a building complex's photovoltaic, energy storage, charging, and power system can be calculated using the following formula: in, It is the system power supply resilience index. These refer to photovoltaic power generation, battery charging and discharging (charging is a negative value, discharging is a positive value), grid power supply, mobile energy storage vehicle power supply, building power supply, and electric vehicle charging power, respectively. and These are the start and end times of the measurement, respectively. S5. Power load regulation and management of building complexes; S5.1 Load regulation under normal operating conditions; Load regulation under normal operating conditions is mainly based on time-of-use peak and off-peak electricity pricing, adjusting parameters for electric vehicles, air conditioners, some lighting, and some socket loads. This includes adjusting the charging start time of electric vehicles (shifting the charging start time from peak electricity price periods to off-peak periods), optimizing the activation threshold for cooling or heating air conditioner usage (this adjustment should not significantly affect thermal comfort and operating efficiency; based on existing research, the activation threshold for cooling air conditioner usage is adjusted to 24℃, and the activation threshold for heating air conditioner usage is adjusted to 22℃), and transferring some available power from peak electricity price periods to off-peak periods. In this specific implementation case, the air conditioning cooling start temperature was adjusted from the measured value (all between 18℃ and 22℃) to 22℃-24℃. The first electric vehicle charging start time at 15:05 was moved forward to 12:30, the second charging start time at 19:37 was postponed to 0:00, and the third charging start time at 11:50 was not adjusted. In addition, 20% of the adjustable electricity consumption for lighting and sockets during the period from 19:00 to 21:00 was transferred from the peak electricity price period to the off-peak period. S5.2 Load regulation under power outage conditions; Load regulation during power outages is primarily based on safety regulations and power supply resilience, adjusting parameters for electric vehicles, air conditioners, and some lighting and socket loads. This includes ceasing charging of electric vehicles, further optimizing the activation thresholds for cooling or heating air conditioner usage (this adjustment must not significantly impact human health; based on existing research, the activation threshold for cooling air conditioner usage is adjusted to 26°C, and for heating air conditioner usage to 20°C), and further shifting adjustable electricity consumption for lighting and sockets (shifting 50% of adjustable electricity consumption for lighting and sockets from peak electricity price periods to off-peak periods). In this specific implementation case, the air conditioning cooling start temperature was further adjusted from the measured value (both between 18℃ and 22℃) to 26℃, electric vehicle charging was stopped, and 50% of the adjustable power consumption for lighting and sockets during the power outage period was transferred from peak electricity price periods to off-peak periods. S6. Optimized scheduling of power systems for photovoltaic, energy storage, and charging in building complexes; S6.1 Optimize the construction of scheduling constraints; The constraint equations for the optimal scheduling of the photovoltaic, energy storage, charging, and power system of the building complex are constructed. These include ensuring that the state of charge of the batteries and electric vehicle batteries is between 10% and 90% throughout the entire process, the photovoltaic power generation is less than the rated maximum value, the electric vehicle charging power is less than the rated maximum charging power of the charging pile, the power supply of the mobile energy storage vehicle is less than the rated maximum value, the charging and discharging power of the battery pack is less than the rated maximum value, the total battery charging and discharging amount every 24 hours is less than twice the maximum battery capacity, and ensuring the system power supply resilience index in S4.5. R Always greater than or equal to 0; S6.2 Optimize the construction of the scheduling solver; A solver for the optimal scheduling of the photovoltaic-storage-charging-utilization power system of a building complex is constructed. The algorithm selected is the jellyfish search algorithm, and the population is initialized using the Logistic chaotic mapping. An adversarial learning strategy is introduced to generate inverse solutions to expand the search space. Levy flight is incorporated to enhance the global exploration capability and allow for sporadic large step updates. S6.3 Optimized scheduling under normal operating conditions; The optimized scheduling process under normal operating conditions is as follows: first, load adjustment is performed according to S5.1; then, hourly operation optimization of the photovoltaic-battery system is performed based on the adjusted load. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, and grid power supply. The objective function is the daily total depreciated cost, calculated as follows: in, This is the total daily discounted cost under normal operating conditions; S6.4 Optimized scheduling under power outage conditions with power outage prediction; The optimized scheduling process under power outage conditions with outage prediction is as follows: First, load adjustment is performed according to S5.2. Then, the battery is fully charged during the off-peak electricity price period before the expected outage time on the outage day and is not restarted. Hourly photovoltaic operation optimization is performed based on the adjusted load, and the battery pack capacity is released at the outage time. If the minimum electricity demand is still insufficient, external mobile energy storage vehicles are needed to supplement the power. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, grid power supply, and mobile energy storage power supply. The objective function is the converted total operating cost on the outage day, calculated as follows: in, It is the total operating cost per day of power outage under power outage conditions with power outage forecasting; S6.5 Optimized scheduling under power outage conditions without power outage prediction; The optimized scheduling process under power outage conditions with outage prediction first involves load adjustment under power outage conditions according to S5.2, and then hourly optimization of the operation of photovoltaic-battery-mobile energy storage based on the adjusted load. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, grid power supply, and mobile energy storage power supply. The objective function of the optimization is the equivalent total operating cost on the power outage day, calculated as follows: in, It is the total operating cost per day of power outage under power outage conditions without power outage prediction; S7. Performance Evaluation and Analysis; The proposed method is evaluated for carbon emissions, total cost, and system power supply resilience index. In this specific implementation case, Table 1 lists the optimized scheduling results under normal operating conditions, specifically comparing whether the load-side adjustment and management recorded in S5.1 was carried out, whether the photovoltaic-battery operation optimization recorded in S6.3 was carried out, and comparing different seasons; Table 1 Optimized scheduling results under normal operating conditions

[0027] Table 2 lists the optimized scheduling results under power outage conditions, specifically comparing whether the load-side adjustment and management recorded in S5.2 were carried out, whether the photovoltaic-battery operation optimization recorded in S6.4 or S6.5 was carried out, and comparing different seasons, whether power outage prediction was available, and whether mobile energy storage technology was available. Table 2 Optimized Dispatch Results under Power Outage Conditions

[0028] The results show that the proposed synergistic optimization method for photovoltaic-storage-charging-utilization in building clusters, applicable to both conventional and outage scenarios, has a positive effect on carbon emission reduction, cost reduction, and system power supply resilience improvement. The method is also affected by different strategies and seasons, as detailed below: Under normal operating conditions, the proposed synergistic optimization method for photovoltaic, energy storage, charging, and utilization in building clusters, applicable to both normal and power outage conditions, can reduce total costs by 9.70%-23.87%, CO2 emissions by 9.74-22.09 tons / day, and pollutant emissions (CO, NO) by [missing information]. x SO2 and particulate matter emissions will be reduced by 0.46-1.05 tons / day. In power outage conditions, the proposed synergistic optimization method for photovoltaic, energy storage, charging, and utilization in building clusters, applicable to both normal and power outage scenarios, can maintain resilience while avoiding critical load losses and reduce total costs by 50.55%-58.84%. Within the entire methodology, load-side regulation and management can reduce daily total cost by 5.68%-7.03% under normal circumstances and reduce daily operating cost by 26.54%-28.40% under power outage conditions. In the entire methodology, photovoltaic-battery cloud optimization can reduce the total daily cost by 3.29%-18.99% under normal circumstances, and reduce the daily operating cost by 18.78%-43.79% under power outage conditions. Compared with load-side regulation, it is more affected by seasonal and other factors, and the benefits are unstable. In the entire methodology, mobile energy storage technology can reduce costs by up to 14.96% while ensuring user well-being, and requires 12-26 deployments per year to achieve cost recovery; while effective power outage prediction can guide the system to reserve battery power in advance, and in some cases (such as when the potential of photovoltaics is large in summer), it may not be necessary to meet the minimum load requirements under power outage conditions without external equipment such as mobile energy storage.

Claims

1. A collaborative optimization method for photovoltaic, energy storage, and charging utilization in building complexes, applicable to both conventional and power outage scenarios, characterized in that: The steps are as follows: S1. Collect and preprocess data to obtain continuous time series; S2. Based on continuous time series, conduct electricity consumption characteristic analysis and behavior modeling to obtain electricity consumption characteristics and behavior probabilities; S3. Using electricity consumption characteristics and behavioral probabilities as inputs, predict the source load of the photovoltaic-storage-charging-utilization power system in the building complex; S4. Evaluate the performance of the photovoltaic-storage-charging power system in a building complex based on the predicted source load of the power system. S5. Using the performance of the photovoltaic, energy storage, charging and utilization power system of the building complex as the objective function and constraint, conduct power load-side regulation and management of the building complex; S6. Based on the adjusted power load of the building complex, optimize the scheduling of the photovoltaic, energy storage, charging and utilization power system of the building complex.

2. The method for coordinated optimization of photovoltaic, energy storage, charging, and utilization in building clusters applicable to both conventional and power outage scenarios as described in claim 1, characterized in that: The collected data includes data on the relationship between power generation and consumption, indoor environmental data, and outdoor meteorological data; The power generation and consumption correlation data includes the total hourly power consumption of the building complex, the total number of people in the building per hour, the total hourly power consumption and duration of air conditioning, the total hourly power consumption and duration of lighting, the total hourly power consumption and duration of sockets, the total charging amount of electric vehicles through charging piles within the building complex and the charging time of electric vehicles, and the hourly power generation of the building complex's rooftop photovoltaic system; indoor environmental data includes indoor air temperature and indoor carbon dioxide concentration; outdoor meteorological data includes outdoor air temperature, outdoor solar radiation intensity, and outdoor cloud cover; all data are aligned by time to form a continuous time series.

3. The collaborative optimization method for photovoltaic, energy storage, and charging applications in building complexes, applicable to both conventional and power outage scenarios, as described in claim 1, is characterized in that... The electricity consumption characteristic analysis includes the electricity consumption characteristics of building lighting and sockets; Building lighting electricity and socket electricity are divided into adjustable electricity and non-adjustable electricity; The adjustable power supply refers to the power that can be reduced, stopped, or transferred, while the non-adjustable power supply refers to the power supply of equipment related to safety and health.

4. The collaborative optimization method for photovoltaic, energy storage, charging, and utilization in building clusters applicable to both conventional and power outage scenarios as described in claim 1, characterized in that: The behavioral modeling includes building air conditioning usage behavior modeling and electric vehicle charging behavior modeling; The building air conditioning usage behavior modeling is based on the strong influence of indoor and outdoor air temperature on people's use of air conditioning, and the probability of usage behavior is calculated through air conditioning usage time data; Based on indoor air temperature data, the probability of air conditioning usage behavior is fitted according to the following formula. : The probability of heating air conditioner usage behavior is fitted using the following formula. : in, The threshold for triggering air conditioning usage behavior. The threshold for activating heating air conditioner usage behavior. Indoor air temperature k , l , τ All of these are fitting parameters, representing the sensitivity of the personnel, the dimensionless constant, and the time scale parameter, respectively; The electric vehicle charging behavior modeling is based on fitting the overall number of charging cycles within the building complex using the following formula: in, N This represents the total number of charging cycles within the building complex. The fitting parameters for the total number of charging cycles. n This represents the number of peaks in a multimodal Gaussian distribution. t The moment when electric vehicles begin charging. , and These are the fitting parameters corresponding to a single peak, representing the amplitude, mean, and standard deviation, respectively.

5. The collaborative optimization method for photovoltaic, energy storage, and charging applications in building complexes applicable to both conventional and power outage scenarios as described in claim 1, characterized in that: The power source and load of the building complex's photovoltaic, energy storage, and charging system include electric vehicle charging demand, building electricity demand, photovoltaic output, and battery charging and discharging. The electric vehicle charging demand is predicted based on the Monte Carlo stochastic simulation method; the actual number of electric vehicles and battery performance parameters are input; the average daily mileage of each electric vehicle is randomly generated using the following formula based on a log-normal distribution: in, The average daily driving distance of electric vehicles x The corresponding probability density function, x It is the average daily mileage of electric vehicles. and These are the expected value of the average daily mileage of electric vehicles and the standard deviation of the average daily mileage of electric vehicles, respectively. The charging start time for each electric vehicle is randomly generated using the following formula: (Multimodal Gaussian distribution is used) in, It is the moment when electric vehicle charging begins. t The corresponding probability density function, and These are the expected value and the standard deviation of the electric vehicle charging start time, respectively. Assuming a constant charging power, the charging time for each electric vehicle is calculated based on the electricity consumption, charging power, and charging efficiency corresponding to the average daily mileage of each electric vehicle. The charging situation of each electric vehicle is simulated one by one through a Monte Carlo cycle to obtain the final total charging demand of the electric vehicles. in, Charging power for electric vehicles, This is the standard charging power of the charging station. For charging duration, M For the number of electric vehicles, For the first i The charging status of an electric vehicle, during charging =1, when not charging =0, It refers to the state of charge of an electric vehicle battery. It refers to charging efficiency; The building's electricity demand is predicted using a long short-term memory network time-series forecasting method. The input data includes the previous hour's historical energy consumption, time characteristics, indoor air temperature, indoor carbon dioxide concentration, outdoor air temperature, outdoor solar radiation intensity, outdoor cloud cover, and number of people in the building. The long short-term memory network time-series forecasting method is used to predict the building's electricity demand and output the building's electricity consumption. The photovoltaic output is calculated according to the following formula: in, This represents the real-time solar radiation intensity. Photovoltaic power generation capacity, It is the output power of a photovoltaic power generation system under standard conditions. =25℃, =1000W / m 2 , It is the temperature power coefficient. It's the temperature of the photovoltaic panel. It refers to the ambient temperature when the photovoltaic panels are working. This refers to the rated operating temperature of the photovoltaic panel. It is the derating factor of a photovoltaic power generation system, which is the percentage decrease in the amount of electricity generated in the current year compared to the amount of electricity generated in the first year; The charging and discharging of the storage battery is calculated using a kinetic energy battery model, as shown in the following formula: in, and They represent t Battery capacity at any time and ( t -1) Battery capacity at time [time]. and Let represent the battery charging power at time t and the battery discharging power at time t, respectively. and These represent inverter efficiency and battery efficiency, respectively. and These represent the battery's charging efficiency and discharging efficiency, respectively. This indicates the operating status of the battery pack. It is 1 when charging, 0 when neither charging nor discharging, and -1 when discharging.

6. The collaborative optimization method for photovoltaic, energy storage, charging, and utilization in building clusters applicable to both conventional and power outage scenarios as described in claim 1, characterized in that, The performance of the photovoltaic, energy storage, charging and utilization power system of the building complex includes investment cost, energy cost, carbon emissions and environmental protection cost, penalties and rewards, and system resilience; The investment cost is shown in the following formula: in, It is the average annual discounted cost of the total investment. and These correspond to the average annual depreciated investment costs of photovoltaic power generation systems, electric vehicle charging piles, battery packs, and mobile energy storage vehicles, respectively. Represents the annualized interest rate. Represents the total lifespan in years; Energy costs include the cost of purchasing electricity from the grid for the building complex's photovoltaic, energy storage, and charging power system, and the cost of purchasing electricity from mobile energy storage vehicles for the same system. The cost of purchasing electricity from the grid for the building complex's photovoltaic, energy storage, and charging power system is calculated using the following formula: in, It is the average daily cost of purchasing electricity through the power grid. and These refer to the electricity price that buildings pay for electricity purchased from the grid and the electricity price that electric vehicle charging stations pay for electricity purchased from the grid, respectively. and These refer to the electricity purchased by buildings from the grid and the electricity purchased by electric vehicle charging stations from the grid, respectively. and These are the start and end times of the measurement, respectively. The cost of purchasing electricity via mobile energy storage vehicles for a building complex's photovoltaic, energy storage, and charging power system is calculated using the following formula: in, This is the average daily cost of purchasing electricity through mobile energy storage vehicles. and These refer to the unit price of electricity purchased through mobile energy storage vehicles and the total amount of electricity purchased through mobile energy storage vehicles, respectively. and These refer to the hourly labor cost and the hourly hardware cost associated with deploying mobile energy storage vehicles. This refers to the time when the mobile energy storage vehicle is deployed; Carbon emissions and environmental protection costs are the associated costs of treating pollutants and carbon dioxide emissions from building complex photovoltaic-storage-charging power systems that purchase electricity from non-clean energy power plants. These costs are calculated using the following formula: in, This is the average daily cost used for pollutant and carbon dioxide treatment. For the first q The cost of treating these emissions It is the first q Emission coefficients of various pollutants This refers to the total amount of electricity purchased through non-clean energy power plants. q Refers to the types of emissions; The penalties and rewards are calculated by offsetting renewable energy rewards and power outage losses for the photovoltaic-storage-charging power system of the building complex, using the following formula: in, It is the average daily cost after deducting rewards and penalties. This refers to the amount of renewable energy generated that is awarded as a reward. The incentive coefficient for renewable energy generation. This refers to the amount of electricity lost during a power outage. This is the penalty coefficient for power outages; System resilience is calculated using the following formula: in, R It is the system power supply resilience index. These refer to photovoltaic power generation, battery charging and discharging, grid power supply, mobile energy storage vehicle power supply, building power supply, and electric vehicle charging power, respectively.

7. The collaborative optimization method for photovoltaic, energy storage, and charging applications in building complexes applicable to both conventional and power outage scenarios as described in claim 1, characterized in that, Power load regulation in building complexes is divided into load regulation under normal operating conditions and load regulation under power outage conditions; Under normal operating conditions, load regulation is based on time-of-use peak and valley electricity pricing, and parameter adjustments are made to electric vehicles, air conditioners, some lighting, and some sockets, including: adjusting the charging start time of electric vehicles, optimizing the start threshold of cooling or heating air conditioner usage behavior, and transferring some available power. Load regulation during power outages is based on safety rules and power supply resilience. It involves adjusting parameters for electric vehicles, air conditioners, some lighting, and some sockets, including: ceasing charging of electric vehicles, further optimizing the start threshold for air conditioner cooling or heating, and further diverting available power.

8. The collaborative optimization method for photovoltaic, energy storage, charging, and utilization in building clusters applicable to both conventional and power outage scenarios as described in claim 1, characterized in that, The optimized scheduling of the photovoltaic, energy storage, charging, and utilization power system for building complexes specifically includes: S6.1 Optimize the construction of scheduling constraints; The constraint equations for the optimal scheduling of the photovoltaic, energy storage, charging, and power system of the building complex are constructed, including: the state of charge of the batteries and electric vehicle batteries is between 10% and 90% throughout the entire process; the photovoltaic power generation is less than the rated maximum value; the electric vehicle charging power is less than the rated maximum charging power of the charging pile; the power supply of the mobile energy storage vehicle is less than the rated maximum value; the charging and discharging power of the battery pack is less than the rated maximum value; the total charging and discharging amount of the battery every 24 hours is less than twice the maximum battery capacity; and system resilience. R Always greater than or equal to 0; S6.2 Optimize the construction of the scheduling solver; A solver for the optimal scheduling of the photovoltaic-storage-charging-utilization power system of the building complex is constructed. The jellyfish search algorithm is selected, and the population is initialized using the Logistic chaotic mapping. An adversarial learning strategy is introduced to generate inverse solutions to expand the search space. Levy flight is incorporated to enhance the global exploration capability and allow for sporadic large step updates. S6.3 Optimized scheduling under normal operating conditions; The optimized scheduling process under normal operating conditions involves load adjustment under normal operating conditions, followed by hourly optimization of the photovoltaic-battery operation based on the adjusted load. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.

2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, and grid power supply. The objective function is the daily total depreciated cost, calculated as follows: in, This is the total daily discounted cost under normal operating conditions; S6.4 Optimized scheduling under power outage conditions with power outage prediction; The optimized scheduling process under power outage conditions with outage prediction is as follows: First, load adjustment is performed under the power outage condition. During the off-peak electricity price period before the expected power outage time, the battery is fully charged and not restarted. Hourly photovoltaic operation optimization is performed based on the adjusted load, and the battery pack capacity is released at the time of the power outage. If the minimum electricity demand is still insufficient, an external mobile energy storage vehicle is needed to supplement the power. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.

2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, grid power supply, and mobile energy storage vehicle power supply. The objective function is the converted total operating cost on the power outage day, calculated as follows: in, It is the total operating cost per day of power outage under power outage conditions with power outage forecasting; S6.5 Optimized scheduling under power outage conditions without power outage prediction; The optimized scheduling process under power outage conditions without power outage prediction is as follows: First, load adjustment is performed under the power outage conditions. Then, hourly photovoltaic-battery-mobile energy storage vehicle operation optimization is performed based on the adjusted load. The optimization constraints are constructed according to S6.1, and the optimization solver is constructed according to S6.

2. The optimization decision variables are hourly photovoltaic power supply, battery charging and discharging, grid power supply, and mobile energy storage vehicle power supply. The objective function of the optimization is the converted total operating cost on the power outage day, calculated as follows: in, It is the total operating cost per day of power outage under power outage conditions with power outage forecasting.

Images