Agricultural park energy storage system optimal operation method considering time-shiftable load

Abstract

The application discloses a kind of agricultural park energy storage system optimization operation methods considering time-shiftable load, it is related to new energy storage application technical field.The method includes: establishing the time-shiftable load model of agricultural park, including irrigation system load model and greenhouse system load model;With minimum daily comprehensive operation cost as target to construct objective function;Set power balance constraint, energy storage state of charge constraint, energy storage charge-discharge power constraint, photovoltaic output constraint and grid interactive power constraint;Adopt particle swarm and linear programming double-layer optimization method, outer optimization energy storage capacity and power configuration, inner layer solves daily optimal charge-discharge strategy.The application includes unified optimization framework with energy storage configuration and load scheduling, realizes the collaborative optimization of energy storage system and time-shiftable load, effectively reduces the daily comprehensive operation cost of agricultural park.

Description

Technical Field

[0001] This invention relates to the field of new energy storage application technology, specifically to an optimized operation method for an agricultural park energy storage system that takes into account time-shiftable loads. Background Technology

[0002] Agricultural parks achieve a dual improvement in agricultural production efficiency and resource utilization through large-scale planting, intensive management, and supply chain collaboration. Agricultural parks possess inherent advantages in the allocation of resources such as land, water, and energy, facilitating unified planning and centralized scheduling. This creates conditions for the introduction of intelligent energy management methods, gradually making them an important pillar of my country's agricultural supply system.

[0003] The energy consumption characteristics of agricultural parks are closely related to the needs of crop growth environments. They are not only affected by natural climate conditions but also must meet the growth requirements of crops, such as photosynthesis and temperature and humidity regulation, resulting in a complex energy consumption pattern involving multiple energy sources. Renewable energy generation, such as photovoltaic power, in these parks is affected by weather and seasons, exhibiting intermittency and fluctuation. The uncertainty of photovoltaic output and the peak-shaving contradiction with load demand lead to the dilemma of midday curtailment of solar power and peak-hour electricity purchases in the morning and evening. Energy storage systems, due to their ability to transfer energy over time and space, have attracted widespread attention in park energy systems. Optimizing the cost of energy storage systems while meeting demand has become a key scientific issue.

[0004] Several research solutions exist for optimizing energy storage configuration and load dispatching in microgrids. For example, a microgrid optimization dispatching model for photovoltaic smart agricultural greenhouses achieves local consumption of photovoltaic power through time-shifting load adjustments. However, its load dispatching strategy and energy storage configuration are independent, and the optimization process does not consider their coordination. Another example is a hybrid maximum power point tracking algorithm combining perturbation observation and fast prediction methods for low-voltage photovoltaic-storage DC microgrid systems designed for new greenhouses. This algorithm enables intelligent energy exchange between photovoltaics, batteries, and loads. However, this research focuses on system control strategies and hardware implementation, without addressing the coordinated optimization dispatching of time-shifting loads. Yet another example is an energy storage optimization configuration model based on bilevel programming theory, using load fluctuations, system cost, and energy storage state of charge (SOC) deviation as optimization objectives. An improved multi-objective particle swarm optimization algorithm is used for solving this model. However, this research targets general industrial and commercial microgrids and lacks modeling for the time-shifting loads specific to agricultural parks, making it difficult to adapt to the special requirements of agricultural production. In addition, a stochastic planning method for energy storage capacity of grid-connected microgrids is proposed based on the consideration of optimal power balance and demand response. An improved particle swarm optimization algorithm is used to optimize energy storage configuration, but its objective function only focuses on power balance and cost, without taking into account photovoltaic consumption and park economics.

[0005] Furthermore, existing optimization methods for integrated agricultural energy systems typically incorporate various devices such as photovoltaic, wind power, combined heat and power units, gas boilers, electric energy storage, and thermal energy storage into a unified scheduling model. While generally mentioning time-shiftable load constraints in the constraints, these methods do not establish refined physical models of time-shiftable loads for the specific energy consumption mechanisms of irrigation and greenhouse systems in agricultural parks. Time-shiftable loads exist only as abstract constraints in the optimization model, making it difficult to accurately reflect the actual temporal flexibility of agricultural irrigation and greenhouse supplemental lighting and temperature control. At the same time, these methods usually treat energy storage capacity configuration and intraday operation scheduling as independent problems, failing to incorporate them into a unified two-layer optimization framework, making it difficult to achieve global synergistic optimization of energy storage system capacity-power configuration and operation strategy.

[0006] In summary, the existing technologies have the following shortcomings: (1) Energy storage configuration optimization and load scheduling are independent of each other and are not included in a unified optimization framework; (2) There is a lack of accurate modeling of time-shiftable loads unique to agricultural parks (such as irrigation systems, greenhouse temperature control systems, etc.); (3) The objective function does not fully consider multi-dimensional economic factors such as energy storage investment depreciation, operation and maintenance costs, photovoltaic costs, net cost of grid power purchase and sale, and energy storage cycle life costs. Summary of the Invention

[0007] To address the shortcomings of existing technologies, such as the independent configuration of energy storage capacity and load scheduling, the lack of accurate modeling of time-shiftable loads in agricultural parks, and the incomplete consideration of economic factors, this invention provides an optimized operation method for energy storage systems in agricultural parks that takes into account time-shiftable loads. The aim is to incorporate energy storage capacity-power optimization and intraday charging and discharging scheduling into a unified optimization framework, thereby achieving synergistic optimization of energy storage system configuration and operation strategies and reducing the overall daily operating cost of agricultural parks.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] This invention provides a method for optimizing the operation of an agricultural park energy storage system that takes into account time-shiftable loads, comprising the following steps:

[0010] S1: Establish a time-shiftable load model for the agricultural park, which includes an irrigation system load model and a greenhouse system load model;

[0011] S2: Construct an objective function with the goal of minimizing the daily comprehensive operating cost. The objective function includes the energy storage investment depreciation cost, energy storage operation and maintenance cost, photovoltaic investment depreciation cost, photovoltaic operation and maintenance cost, net cost of purchasing and selling electricity from the grid, and energy storage cycle life cost.

[0012] S3: Set constraints, including power balance constraints, dynamic update constraints for energy storage state of charge, energy storage charging and discharging power constraints, energy storage state of charge constraints, photovoltaic output constraints, and grid interaction power constraints.

[0013] S4: A two-layer optimization method using particle swarm optimization and linear programming is adopted to optimize the rated capacity and rated power of the energy storage system. The outer layer uses particle swarm optimization to optimize the energy storage capacity and power configuration, while the inner layer uses linear programming to solve for the optimal charging and discharging strategy within the day.

[0014] Furthermore, in step S1, the irrigation system load model includes a water-fertilizer integrated machine load model and a water storage load model; the operating power consumption of the water-fertilizer integrated machine is determined based on the total daily water demand of the crop and the pump operating efficiency; the total daily water demand of the crop is determined based on the standard evapotranspiration and the crop coefficient; the standard evapotranspiration is calculated based on the greenhouse environment's highest temperature, lowest temperature, average temperature, environmental temperature fluctuation correction value, evapotranspiration conversion coefficient, temperature influence index, latent heat of vaporization of water, greenhouse light transmittance, and local daily average solar irradiance.

[0015] Furthermore, in the water storage load model, the water level in the reservoir is updated recursively based on the working efficiency of the water pump, the operating power of the water pump, the operating time of the water pump, the operating time of the irrigation system, and the water consumption of the irrigation system, and the water level in the reservoir returns to the initial value after a complete irrigation day.

[0016] Furthermore, in step S1, the greenhouse system load model includes an LED supplemental lighting load model and a greenhouse thermostat load model; the total power consumption of the LED supplemental lighting is determined based on the standard light intensity required for crop growth, the actual light intensity inside the greenhouse, the illuminance conversion factor, the total area of ​​the greenhouse, the luminous flux required per unit area, and the lamp-to-electric conversion efficiency of the LED supplemental lighting; the operating power of the greenhouse thermostat is determined based on the total heat load demand of the greenhouse, which is calculated based on the heat transfer coefficient of the greenhouse protective film, the greenhouse surface area, the greenhouse volume, the air density inside the greenhouse, the daily ventilation frequency of the greenhouse, the specific heat capacity of air at constant pressure, the temperature difference between inside and outside the greenhouse, the latent heat of water vaporization, and the difference in humidity between inside and outside the greenhouse.

[0017] Furthermore, in step S2, the objective function for the daily comprehensive operating cost is:

[0018]

[0019] In the formula, This is the average daily depreciation cost of energy storage investment. This is the average daily energy storage operation and maintenance cost. This is the average daily depreciation cost of photovoltaic investment. This is the average daily operation and maintenance cost of photovoltaic systems. This is the average daily net cost of electricity purchased and sold from the power grid. It is the average daily energy storage cycle life cost. The minimum value of the daily comprehensive operating cost is used to minimize the daily comprehensive cost.

[0020] Furthermore, the average daily energy storage investment depreciation cost is calculated based on the energy storage unit capacity investment cost, rated energy storage capacity, energy storage unit power investment cost, rated energy storage power, and the energy storage equivalent annual value coefficient; the average daily energy storage operation and maintenance cost is calculated based on the energy storage unit power annual operation and maintenance cost and the energy storage rated power; the average daily photovoltaic investment depreciation cost is calculated based on the photovoltaic unit capacity investment cost, photovoltaic rated installed capacity, and the photovoltaic equivalent annual value coefficient; the average daily photovoltaic operation and maintenance cost is calculated based on the photovoltaic unit capacity annual operation and maintenance cost and the photovoltaic rated installed capacity; the daily grid purchase and sale net cost is calculated by accumulating the differences between the product of the purchased power and the purchased electricity price and the product of the sold power and the sold electricity price within a day; and the average daily energy storage cycle life cost is calculated based on the total energy storage investment cost and the total throughput of energy storage throughout its entire life cycle.

[0021] Furthermore, in step S3, the power balance constraint is:

[0022]

[0023] In the formula: The actual photovoltaic output at time t; Let t be the energy storage discharge power; Let t be the power purchased by the power grid. Let t be the load power at time t; The energy storage charging power at time t; Let t be the power output of the power grid at time t.

[0024] Furthermore, in step S3, the energy storage state of charge constraint is as follows: the energy storage state of charge is maintained between the minimum allowable state of charge and the maximum allowable state of charge, and the energy storage state of charge at the beginning of the scheduling cycle is equal to the energy storage state of charge at the end of the cycle.

[0025] Furthermore, in step S3, the energy storage charging and discharging power constraints are as follows: the energy storage charging power and discharging power do not exceed the rated energy storage power, and the energy storage does not charge and discharge at the same time; the grid interaction power constraints are as follows: the power purchase power does not exceed the maximum allowable power purchase power, the power sale power does not exceed the maximum allowable power sale power, and the power purchase and sale do not occur at the same time.

[0026] Furthermore, in step S4, the specific solution process of the particle swarm optimization and linear programming two-layer optimization method is as follows: the outer particle swarm algorithm uses the rated capacity and rated power of the energy storage as decision variables to perform optimization and passes the decision variables to the inner layer; the inner linear programming uses the charging and discharging power of the energy storage at each moment as decision variables, solves the optimal charging and discharging strategy within the day under the energy storage configuration given by the outer layer, and returns the optimal objective function value to the outer layer. The outer layer updates the position and velocity of the particle swarm according to the returned objective function value until the convergence condition is met, and outputs the optimal energy storage configuration scheme and its corresponding intraday charging and discharging strategy.

[0027] This invention proposes an optimized operation method for agricultural park energy storage systems that takes into account time-shiftable loads, which has the following advantages and beneficial effects:

[0028] (1) This invention incorporates energy storage capacity-power optimization and intraday charging and discharging scheduling into a unified optimization framework. It adopts a two-layer optimization method of particle swarm optimization and linear programming to achieve coordinated optimization of energy storage configuration and operation strategy, overcoming the technical defect that the two are independent of each other in the prior art.

[0029] (2) The present invention establishes an accurate physical model for the time-transferable loads unique to agricultural parks, including the water and fertilizer integrated machine model and water storage load model of the irrigation system, the LED supplementary lighting model and the temperature controller model of the greenhouse system. The model fully considers physical constraints such as crop evapotranspiration, water storage continuity and greenhouse heat balance, so that the optimization results are more in line with the actual agricultural production.

[0030] (3) This invention constructs a multi-dimensional daily comprehensive cost model that includes energy storage investment depreciation, operation and maintenance costs, photovoltaic costs, net cost of grid electricity purchase and sale, and energy storage cycle life cost. It comprehensively quantifies the economics of energy storage system operation and achieves the unity of multiple objectives of peak shaving and valley filling, photovoltaic consumption and cost optimization.

[0031] (4) This invention uses the “low charge and high discharge” energy storage charging and discharging strategy to make arbitrage by taking advantage of the peak and valley electricity price difference, which effectively reduces the net electricity purchase cost of the park; at the same time, it prioritizes the use of photovoltaic power for load self-use and energy storage charging, and only sells electricity when the energy storage is saturated, thus realizing the complete absorption of photovoltaic power and suppressing the low-price electricity sales behavior.

[0032] (5) This invention establishes a water and fertilizer integrated machine load model based on the physical mechanism of crop evapotranspiration, a water storage tank water level recursion model based on water balance, an LED supplementary light load model based on the difference of photosynthetically effective radiation, and a temperature controller load model based on the thermal balance inside the greenhouse for irrigation and greenhouse systems unique to agricultural parks. This realizes refined physical modeling of agricultural time-transferable loads, rather than just setting general scheduling constraints for time-transferable loads. This enables the energy storage optimization results to accurately reflect the temporal flexibility of irrigation and greenhouse operations in agricultural production, and improves the feasibility of the optimization scheme in the actual operation of agricultural parks. Attached Figure Description

[0033] Figure 1 This is a daily variation curve of the energy storage charging and discharging power of the present invention;

[0034] Figure 2 This is a daily variation curve of the state of charge of the energy storage of the present invention;

[0035] Figure 3 This is a comparison chart of the energy storage charging and discharging power of the present invention with time-of-use electricity pricing;

[0036] Figure 4 This is a comparison chart of the purchased power and time-of-use electricity price according to the present invention;

[0037] Figure 5 This is a comparison chart of the electricity sales capacity and the grid connection price of the present invention;

[0038] Figure 6 This is a power balance diagram for typical days in all four seasons according to the present invention, wherein... Figure 6 (a) in the diagram is a typical daily power balance diagram in spring. Figure 6 (b) in the diagram is a typical daily power balance diagram in summer. Figure 6 (c) in the diagram is a typical daily power balance diagram for autumn. Figure 6 (d) in the diagram represents a typical daily power balance diagram during winter.

[0039] Figure 7 This is a breakdown diagram of electricity purchase costs for typical days in all four seasons according to the present invention, wherein... Figure 7 (a) in the figure is a breakdown of the electricity purchase cost for a typical day in spring. Figure 7 (b) in the diagram is a breakdown of electricity purchase costs for a typical summer day. Figure 7 (c) in the figure is a breakdown of the electricity purchase cost for a typical day in autumn. Figure 7 (d) in the figure is a breakdown of electricity purchase costs for a typical day in winter. Detailed Implementation

[0040] The present invention will be further described below with reference to the embodiments. It should be noted that these are merely examples and descriptions of the inventive concept. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the scope of the inventive concept, they should all be considered to fall within the protection scope of the present invention.

[0041] 1. Time-shiftable load modeling in agricultural parks

[0042] Time-shiftable agricultural loads that significantly impact microgrid systems can be categorized into two types: irrigation and greenhouses. Irrigation systems, including fertigation units and irrigation pumps, collectively determine the timing flexibility of water and fertilizer supply. Greenhouse systems, comprising LED supplemental lighting and greenhouse temperature control equipment, collectively influence the control window for light and temperature. The following sections will model these four load units separately.

[0043] Modeling of integrated water and fertilizer machine

[0044] The operating power of the fertigation machine is directly related to the crop's water requirement, which is determined by the natural evapotranspiration characteristics of the greenhouse area. Therefore, a load model needs to be established in conjunction with environmental parameters. The expression for the operating power of the fertigation machine is:

[0045] (1)

[0046] In the formula, The power consumption of the integrated water and fertilizer machine is expressed in kW. The total daily water requirement for crops, in mm; For pump operating efficiency.

[0047] The core parameter for calculating crop water requirement is standard evapotranspiration. (mm), needs to be calculated using the Hargreaves formula, which is as follows:

[0048] (2)

[0049] In the formula, , , These represent the highest, lowest, and average temperatures in the greenhouse environment, in °C. This is a correction value for ambient temperature fluctuations; This is the conversion factor for evaporation. The temperature effect index; The latent heat of vaporization of water is expressed in MJ / kg. For greenhouse light transmittance; The local average daily solar irradiance, kW / (m²) 2 ·d).

[0050] Based on standard evapotranspiration, and considering the scale of greenhouse cultivation and crop characteristics, calculate the total daily water demand for crops:

[0051] (3)

[0052] In the formula, Let be the total daily water demand of crop type i, in mm; Let m be the planting area of ​​the k-th greenhouse. 2 n represents the total number of daily operating periods for the integrated water and fertilizer machine. Let be the crop coefficient for the i-th type of crop.

[0053] Water storage load modeling

[0054] As the water storage unit of the park, the water level change of the reservoir directly reflects the water balance between the water pumps and the irrigation system. The recursive relationship of the reservoir water level is as follows:

[0055] (4)

[0056] In the formula, and The values ​​are the water levels of the reservoir during time period t and time period t+1, respectively, in meters. To improve the working efficiency of the water storage pump; The duration of the water storage pump during time period t, in hours; Let be the operating power of the water storage pump in time period t, in kW; Let h be the runtime of the irrigation system during time period t. Let t represent the water consumption of the irrigation system during the t-th time period. To ensure the continuity of irrigation water supply in the park and the stable operation of the reservoir, the water level of the reservoir is restored to its initial value after a complete irrigation day (24 hours) to ensure that the water demand for the next day is not affected.

[0057] LED fill light modeling

[0058] LED supplemental lighting enhances crop yield and quality by precisely adjusting the red and blue light spectra to supplement photosynthetically active radiation under low-light conditions. Its operating power is determined by the difference between the actual light intensity inside the greenhouse and the standard light intensity required for crop growth; based on this, the following load model is established:

[0059] (5)

[0060] In the formula, This refers to the total power consumption of the LED fill light; The standard light intensity required for crop growth; Let t be the actual light intensity inside the greenhouse. This is the illuminance conversion factor; This refers to the total area of ​​the greenhouse. The luminous flux required per unit area; The photoelectric conversion efficiency of LED fill lights.

[0061] Greenhouse temperature controller modeling

[0062] Greenhouse thermostats can raise and stabilize the temperature inside the greenhouse during cold seasons, preventing crop frost damage and ensuring normal growth. They also maintain a suitable and uniform temperature environment, extending the growing season, promoting metabolism, and thus improving yield and quality. Their operating power is determined by the greenhouse's heat load demand, which is affected by factors such as the temperature difference between inside and outside the greenhouse, ventilation, and humidity changes. A heat load calculation model is established as follows:

[0063] (6)

[0064] In the formula, This represents the total heat load demand of the greenhouse. The heat transfer coefficient of the greenhouse protective film; The surface area of ​​the greenhouse; The volume of the greenhouse; This refers to the air density inside the greenhouse. The number of times the greenhouse is ventilated daily; The specific heat capacity of air at constant pressure; The temperature difference between inside and outside the greenhouse; It is the latent heat of water vaporization; This represents the difference in air humidity between the inside and outside of the greenhouse. The temperature difference between the inside and outside of the greenhouse is calculated from the target temperature inside the greenhouse and the outdoor ambient temperature.

[0065] (7)

[0066] In the formula, Set the target temperature inside the greenhouse; Let t be the ambient temperature outside the shed.

[0067] 2. Optimization Methods

[0068] I. Objective Function

[0069] The daily operating cost of an energy storage system in an agricultural park is expressed as follows:

[0070] (8)

[0071] In the formula, the daily comprehensive cost is minimized. Its cost is determined by the depreciation cost of energy storage investment. Energy storage operation and maintenance costs Photovoltaic investment depreciation cost Photovoltaic operation and maintenance costs Net cost of electricity purchased and sold from the power grid Energy storage cycle life cost It consists of six items added together.

[0072] Average daily energy storage investment depreciation cost

[0073] The initial investment costs for the purchase, installation, and construction of the energy storage system are amortized over daily costs based on its design life and discount rate, reflecting the asset depreciation and capital occupation costs of the energy storage equipment per unit operating day. Therefore, the total investment is converted into daily depreciation costs and included in the objective function, as shown in the following formula:

[0074] (9)

[0075] In the formula: This represents the average daily depreciation cost of energy storage investment. The cost of energy storage per unit capacity; This refers to the rated capacity of the energy storage. The cost of energy storage per unit power; Rated power of energy storage; 365 represents the annual value coefficient for energy storage, etc.; 365 represents the number of days in a year, used to convert the annual average cost into the daily average cost.

[0076] Daily energy storage operation and maintenance cost

[0077] The costs incurred by an energy storage system during daily operation, including routine maintenance, repairs, monitoring, and auxiliary facility operation, are converted into daily costs, reflecting the ongoing expenses required to maintain the system's normal operating status. As follows:

[0078] (10)

[0079] In the formula: This represents the average daily energy storage operation and maintenance cost; Annual operation and maintenance cost per unit power of energy storage; This refers to the rated power of the energy storage; 365 is used to convert the annual average operation and maintenance cost into the daily average cost.

[0080] Average daily photovoltaic investment depreciation cost

[0081] Photovoltaic systems are long-term investment equipment. The daily depreciation cost, calculated by discounting the initial investment in photovoltaic modules, inverters, mounting systems, and power distribution equipment over their entire lifecycle, is used to quantify the fixed investment costs amortized into daily operation of the photovoltaic system. As follows:

[0082] (11)

[0083] In the formula: This represents the average daily depreciation cost of photovoltaic investment. This refers to the investment cost per unit capacity of photovoltaic power. This refers to the rated installed capacity of photovoltaic power. 365 is the annual value coefficient for photovoltaics, etc.; 365 is used to convert the annual average cost into the daily average cost.

[0084] Daily photovoltaic operation and maintenance cost

[0085] This cost covers the daily expenses incurred in routine work such as photovoltaic array cleaning, equipment inspection, line maintenance, troubleshooting, and safety monitoring. It is the necessary operating cost to ensure the long-term stable and efficient power generation of the photovoltaic system. See below:

[0086] (12)

[0087] In the formula: This represents the average daily operation and maintenance cost of photovoltaic systems. The annual operation and maintenance cost per unit capacity of photovoltaic power generation; This refers to the rated installed capacity of photovoltaic power; 365 is used to convert the annual average operation and maintenance cost into the daily average cost.

[0088] Net cost of electricity purchase and sale from the grid

[0089] The difference between the fees paid by the industrial park to the power grid during the dispatch cycle and the revenue from selling surplus electricity to the grid directly reflects the actual economic income and expenditure generated by the interaction of electricity between the park and the distribution network, and is a key factor affecting the economic efficiency of operation. As follows:

[0090] (13)

[0091] In the formula: The net cost of buying and selling electricity from the daily grid; t is the time number within a day, ranging from 1 to 24; Let t be the power purchased by the power grid. The electricity purchase price at time t; Let t be the power output of the power grid at time t; Let t be the electricity price at time t.

[0092] Daily average energy storage cycle life cost

[0093] Based on the impact of energy storage charge / discharge cycles, depth of charge / discharge, and actual operating conditions on battery lifespan, lifespan loss is converted into corresponding daily economic costs to reflect the impact of different operating strategies on the health status and equivalent usage costs of energy storage. The following is a summary:

[0094] (14)

[0095] In the formula: This represents the average daily energy storage cycle life cost; the first item in parentheses represents the total energy storage investment cost. The energy storage charging power at time t; Let t be the energy storage discharge power; The total throughput over the entire lifecycle of energy storage is calculated as follows:

[0096] (15)

[0097] In the formula: The total throughput of energy storage throughout its entire lifecycle; The rated cycle life of the energy storage; This refers to the rated depth of discharge for energy storage. This refers to the rated capacity of the energy storage.

[0098] II. Constraints

[0099] Power balance constraints

[0100] The system must maintain energy supply and demand balance at all times to ensure normal power supply to the load. The photovoltaic power generation, the power stored or released by the energy storage system, the load demand, and the grid-connected power should satisfy the system's power-energy conservation law, as shown in the following formula:

[0101] (16)

[0102] In the formula: The actual photovoltaic output at time t; Let t be the energy storage discharge power; Let t be the power purchased by the power grid. Let t be the load power at time t; The energy storage charging power at time t; Let t be the power output of the power grid at time t.

[0103] Energy Storage SOC Dynamic Update Constraints

[0104] The stored energy capacity changes dynamically with charging and discharging behavior, reflecting the time-series relationship of the energy. As follows:

[0105] (17)

[0106] In the formula: To improve energy storage charging efficiency; This refers to the energy storage and discharge efficiency.

[0107] Energy storage charging and discharging power constraints

[0108] The charging and discharging power of energy storage cannot exceed the rated power, and charging and discharging cannot occur simultaneously. The formula is as follows:

[0109] (18)

[0110] In the formula: The charging power at time t; Let t be the discharge power at time t; This is the rated power of the energy storage.

[0111] Energy storage state of charge (SOC) constraints

[0112] The state of charge (SOC) of energy storage must be maintained within safe upper and lower limits to prevent overcharging and over-discharging. Furthermore, to ensure continuous operation during the dispatch cycle, the initial and final daily charge levels must remain consistent. The formula is as follows:

[0113] (19)

[0114] In the formula: The energy storage state of charge at time t; This represents the minimum permissible state of charge for energy storage. This represents the maximum permissible state of charge for energy storage. The state of charge (SOC) is the energy storage at the beginning of the scheduling cycle; The SOC is the energy storage state at the end of the scheduling cycle.

[0115] Photovoltaic output constraints

[0116] Photovoltaic power output cannot exceed its rated installed capacity, as it is limited by sunlight conditions.

[0117] (20)

[0118] Power grid interaction constraints

[0119] The power exchange between the industrial park and the power distribution network is limited by the rated capacity of the transformer at the grid connection point and the safe operation specifications of the power distribution network. To avoid overload and exceeding limits of the power exchange with the grid and to ensure the safe and stable operation of the system, upper limits must be set on the power purchased and sold by the grid. Simultaneously, power cannot be purchased from and sold to the grid at the same time, thus meeting the actual operating rules. As follows:

[0120] (twenty one)

[0121] In the formula: Let t be the power purchased at time t; Maximum permitted power purchase capacity; Let t be the power output at time t; This represents the maximum permissible power output.

[0122] 3. Two-level optimization solution method

[0123] This invention employs a two-layer optimization method combining particle swarm optimization and linear programming to solve the aforementioned optimization model. The outer layer uses particle swarm optimization to optimize energy storage capacity and power configuration, while the inner layer uses linear programming to solve for the optimal intraday charging and discharging strategy.

[0124] Outer Particle Swarm Optimization (PSO) Algorithm: Initialize the position and velocity of the particle swarm using the rated energy storage capacity (Eess) and rated power (Pess) as decision variables. The position vector of each particle is X = [Eess, Pess]. The decision variables of each particle are then passed to the inner linear programming model.

[0125] Inner-layer linear programming: Given the energy storage configuration (Eess, Pess) in the outer layer, the charging and discharging power Pess_c and Pess_d at each time of energy storage are used as decision variables. Under the premise of satisfying the constraints (16)-(26), the minimum value of the objective function (8) is solved and the optimal objective function value is returned to the outer layer.

[0126] Outer layer update: Calculate the fitness of each particle based on the objective function value returned by the inner layer, update the individual optimal position and the global optimal position, and then update the particle swarm according to the velocity and position update formula of the particle swarm algorithm. Iterate until the convergence condition is met (such as reaching the maximum number of iterations or the change in global optimal fitness is less than a set threshold), and output the optimal energy storage configuration scheme and its corresponding intraday charging and discharging strategy.

[0127] 4. Verification of Implementation Examples

[0128] To verify the effectiveness of the energy storage optimization model of the "photovoltaic-storage-parallel-off-grid" system proposed in this invention, the optimization solution was carried out based on the typical daily data of a certain agricultural experimental park in four seasons, and the system's full-process operation results were obtained.

[0129] (1) Energy storage charging and discharging strategies and SOC changes

[0130] Based on typical daily load and photovoltaic output characteristics, the charge and discharge power curves of energy storage in each season under optimal configuration are as follows: Figure 1 As shown (positive value for discharging, negative value for charging). Charging activity is highly concentrated between 11:00 AM and 2:00 PM in all seasons, when photovoltaic output reaches its peak and the net load is negative, allowing energy storage to quickly absorb excess electricity at near-rated power. Discharging activity is concentrated in two periods: 7:00 AM to 9:00 AM and 6:00 PM to 10:00 PM, which coincide with peak load and peak electricity prices. Discharging power is highest and lasts longest in winter because of high evening peak load and insufficient photovoltaic power; in summer, discharging is relatively concentrated, with midday charging only needed to support a shorter peak period.

[0131] The intraday variation curves of energy storage state of charge (SOC) for each season are as follows: Figure 2 As shown, the SOC (State of Charge) for all seasons starts from an initial value of 0.25, rapidly rises to an upper limit of 0.90 during the midday charging period, and then drops to a lower limit of 0.20 during the morning and evening discharging periods. The SOC accurately returns to 0.25 at the end of each day, satisfying the intraday energy conservation constraint. In winter, the SOC fluctuates fully between 0.2 and 0.9, indicating that energy storage is fully utilized. In summer, the SOC fluctuation is relatively small (0.25-0.85) because solar power is abundant, and only partial discharge after midday charging is needed to meet the evening peak demand. The SOC is consistently strictly limited between 0.20 and 0.90, avoiding damage to battery life caused by overcharging and over-discharging.

[0132] like Figure 3 As shown, the charging period (negative power) corresponds to the flat or low electricity price (0.32-0.62 yuan / kWh), while the discharging period (positive power) corresponds to the peak or high electricity price (1.04-1.22 yuan / kWh). By utilizing the "low charging, high discharging" model, the energy storage system can effectively profit from the peak-valley price difference, thereby reducing the net electricity purchase cost.

[0133] (2) Grid interaction and power balance

[0134] After configuring energy storage, the purchased power ( Figure 4 The main periods of electricity purchases are concentrated during off-peak hours at night (11:00 PM to 6:00 AM the next day) and brief gaps in afternoon photovoltaic power supply. During peak load periods in the morning and evening, the purchased electricity capacity is significantly reduced, sometimes even reaching zero in the summer evenings, as energy storage discharge completely replaces the high-priced electricity purchases. Purchased electricity capacity is higher during off-peak hours (around 0.32 yuan / kWh), allowing energy storage systems to actively purchase and charge electricity at night to prepare for the next day's peak; during peak hours, purchased electricity capacity approaches zero.

[0135] Electricity sales power ( Figure 5 The electricity generated is concentrated only during the midday period when photovoltaic and energy storage systems are fully charged, with almost zero electricity sold during other times. Due to the economic constraint that the grid-connected electricity price is much lower than the purchase price, the model prioritizes the use of photovoltaic power for self-consumption and energy storage charging, and is only forced to sell electricity when the energy storage is saturated. This achieves full utilization of photovoltaic power while effectively suppressing the practice of selling electricity at low prices.

[0136] Figure 6 The diagram illustrates a power balance comparison across the four seasons. Taking summer as an example, during midday, photovoltaic (PV) output far exceeds load, with excess power initially used for charging energy storage (negative region), and then partially sold after charging (negative region). In the morning and evening, PV output is insufficient, so energy storage discharges (positive region) to fill the gap, and the remaining load is met by grid power purchases (positive region). The power balance equation is satisfied at any time of day, proving the physical feasibility of the optimization results. The power balance characteristics on typical days in spring and autumn are similar to those in summer; on typical days in winter, PV output decreases, midday surplus power decreases, the scale of power sales shrinks, and the system relies more on grid power purchases and energy storage discharges to meet load demands.

[0137] (3) Economic analysis

[0138] Figure 7The daily cost composition for optimized energy storage configuration in each season is as follows: Electricity purchase cost accounts for the largest proportion in all seasons, ranging from 51% to 56%, making it the main component of the park's daily costs; photovoltaic investment cost accounts for 23% to 29%; energy storage investment cost accounts for 15% to 18%; photovoltaic operation and maintenance cost is stable at 4%, while energy storage operation and maintenance cost accounts for only 1%. In terms of seasonal differences, winter has the highest proportion of electricity purchase cost (56%), and the corresponding net electricity purchase cost is also the largest (3955.4 yuan); summer has the lowest proportion of electricity purchase cost (51%), with a net electricity purchase cost of only 1396.2 yuan; spring and autumn are in the middle. Summer has the highest proportion of photovoltaic investment cost (29%), while winter has a relatively higher proportion of energy storage investment cost (18%), reflecting the optimization logic of increasing photovoltaic configuration when photovoltaic resources are abundant and relying on energy storage for adjustment when photovoltaic output is insufficient.

[0139] The above is an exemplary description of the invention. Obviously, the specific implementation of the invention is not limited to the above-described manner. Any non-substantial improvement made using the inventive concept and technical solution of the invention, or the direct application of the inventive concept and technical solution to other situations without modification, is within the protection scope of the invention.

Claims

1. A method for optimizing the operation of an agricultural park energy storage system that takes into account time-shiftable loads, characterized in that, Includes the following steps: S1: Establish a time-transferable load model for the agricultural park, which includes an irrigation system load model and a greenhouse system load model; the irrigation system load model includes a water and fertilizer integrated machine load model and a water storage load model; the greenhouse system load model includes an LED supplemental lighting load model and a greenhouse thermostat load model. S2: Construct an objective function with the goal of minimizing the daily comprehensive operating cost. The objective function includes the energy storage investment depreciation cost, energy storage operation and maintenance cost, photovoltaic investment depreciation cost, photovoltaic operation and maintenance cost, net cost of purchasing and selling electricity from the grid, and energy storage cycle life cost. S3: Set constraints, including power balance constraints, dynamic update constraints for energy storage state of charge, energy storage charging and discharging power constraints, energy storage state of charge constraints, photovoltaic output constraints, and grid interaction power constraints. S4: A two-layer optimization method using particle swarm optimization and linear programming is employed to optimize the rated capacity and rated power of the energy storage system. The outer layer uses particle swarm optimization with rated capacity and rated power as decision variables to find the optimal solution. The inner layer uses linear programming with charging and discharging power at each moment as decision variables to solve for the optimal charging and discharging strategy within the day. The outer layer updates the position and velocity of the particle swarm based on the objective function value returned by the inner layer until the convergence condition is met, and outputs the optimal energy storage configuration scheme and its corresponding intraday charging and discharging strategy.

2. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S1, the power consumption of the integrated water and fertilizer machine is directly proportional to the total daily water demand of the crop and inversely proportional to the pump operating efficiency; the total daily water demand of the crop is directly proportional to the product of the standard evapotranspiration and the crop coefficient; the standard evapotranspiration is calculated by the Hargreaves formula based on the greenhouse environment's highest temperature, lowest temperature, average temperature, environmental temperature fluctuation correction value, evapotranspiration conversion coefficient, temperature influence index, latent heat of vaporization of water, greenhouse light transmittance, and local daily average solar irradiance.

3. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 2, characterized in that, In the water storage load model, the water level of the reservoir is updated recursively according to the following relationship: the water level of the reservoir at the next moment is equal to the water level of the reservoir at the current moment plus the water storage volume corresponding to the product of the working efficiency of the water pump, the operating power of the water pump, and the operating time of the water pump, minus the water intake volume corresponding to the product of the operating time of the irrigation system and the water consumption of the irrigation system; and the water level of the reservoir returns to the initial value after a complete irrigation day.

4. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S1, the total power consumption of the LED supplemental light is directly proportional to the difference between the standard light intensity and the actual light intensity required for crop growth, the total area of ​​the greenhouse, and the light flux required per unit area, and inversely proportional to the lamp-to-electric conversion efficiency of the LED supplemental light. The operating power of the greenhouse thermostat is directly proportional to the total heat load demand of the greenhouse. The total heat load demand of the greenhouse is obtained by summing the heat transfer term of the enclosure structure corresponding to the product of the heat transfer coefficient of the greenhouse protective film and the surface area of ​​the greenhouse and the temperature difference between the inside and outside of the greenhouse, plus the ventilation heat transfer term corresponding to the product of the air density inside the greenhouse, the greenhouse volume, the number of daily ventilations, the specific heat capacity of air at constant pressure, and the temperature difference between the inside and outside of the greenhouse, plus the latent heat heat transfer term corresponding to the latent heat of water vaporization and the difference in humidity between the inside and outside of the greenhouse.

5. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S2, the objective function for the daily comprehensive operating cost is: In the formula, This is the average daily depreciation cost of energy storage investment. This is the average daily energy storage operation and maintenance cost. This is the average daily depreciation cost of photovoltaic investment. This is the average daily operation and maintenance cost of photovoltaic systems. This is the average daily net cost of electricity purchased and sold from the power grid. It is the average daily energy storage cycle life cost. The minimum value of the daily comprehensive operating cost is used to minimize the daily comprehensive cost.

6. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 5, characterized in that, The average daily energy storage investment depreciation cost is proportional to the product of the energy storage unit capacity investment cost and the rated energy storage capacity, plus the product of the energy storage unit power investment cost and the rated energy storage capacity, and then multiplied by the energy storage equivalent annual value factor; the average daily energy storage operation and maintenance cost is proportional to the product of the energy storage unit power annual operation and maintenance cost and the rated energy storage capacity; the average daily photovoltaic investment depreciation cost is proportional to the product of the photovoltaic unit capacity investment cost, the photovoltaic rated installed capacity, and the photovoltaic equivalent annual value factor; the average daily photovoltaic operation and maintenance cost is proportional to the product of the photovoltaic unit capacity annual operation and maintenance cost and the photovoltaic rated installed capacity; the average daily grid purchase and sale net cost of electricity is equal to the sum of the products of the purchased power and the purchased electricity price at each time point minus the sum of the products of the sold power and the sold electricity price at each time point; the average daily energy storage cycle life cost is proportional to the total energy storage investment cost and inversely proportional to the total throughput of the energy storage throughout its entire life cycle, and is calculated based on the daily charge and discharge throughput.

7. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S3, the power balance constraint is: In the formula: The actual photovoltaic output at time t; Let t be the energy storage discharge power; Let t be the power purchased by the power grid. Let t be the load power at time t; The energy storage charging power at time t; Let t be the power output of the power grid at time t.

8. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S3, the energy storage state of charge constraint is as follows: the energy storage state of charge is maintained between the minimum allowable state of charge and the maximum allowable state of charge, and the energy storage state of charge at the beginning of the scheduling cycle is equal to the energy storage state of charge at the end of the cycle.

9. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S3, the energy storage charging and discharging power constraints are: the energy storage charging power and discharging power do not exceed the rated energy storage power, and the energy storage does not charge and discharge at the same time; the grid interaction power constraints are: the power purchase power does not exceed the maximum allowable power purchase power, the power sale power does not exceed the maximum allowable power sale power, and the power purchase and sale do not occur at the same time.

10. The method for optimizing the operation of an agricultural park energy storage system considering time-shiftable loads as described in claim 1, characterized in that, In step S4, the convergence condition of the particle swarm algorithm is to reach a preset maximum number of iterations or for the global optimal fitness to change by a preset number of consecutive generations less than a set threshold. The inner-layer linear programming, given the rated capacity and rated power of the energy storage in the outer layer, aims to minimize the daily comprehensive operating cost. It solves for the energy storage charging and discharging power at each time point while satisfying constraints such as power balance, dynamic update of energy storage state of charge, energy storage charging and discharging power, energy storage state of charge, photovoltaic output, and grid interaction power.

Images