Electric power energy storage day-ahead scheduling method

Abstract

The invention relates to the technical field of electric power system optimization operation, and discloses an electric power energy storage day-ahead scheduling method, which comprises the following steps: collecting electric power data used for inputting a mixed integer programming model, the electric power data comprising next day spot market clearing electricity price and historical load data; calculating the maximum demand constraint value of the next day according to the historical load data; constructing a mixed integer programming model; solving the constructed mixed integer programming model; if the solution is successful, generating a next-day scheduling plan and sending the next-day scheduling plan to the energy storage management module; if the solution fails, giving an alarm; and the energy storage management module executes the next-day scheduling plan, monitors the real-time demand in the execution process of the next-day scheduling plan on the running day, updates the input information of the mixed integer programming model by using the real-time demand, and compares the real-time demand with the real-time demand. According to the invention, the problem that efficient and accurate energy storage scheduling is difficult to realize in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of power system optimization operation technology, specifically a day-ahead dispatching method for power energy storage. Background Technology

[0002] Industrial users typically employ a two-part electricity pricing structure, where the total electricity cost consists of an energy consumption charge calculated based on electricity usage and a basic charge calculated based on monthly maximum demand. With the development of the electricity spot market, energy consumption charges are now determined by real-time fluctuating market prices, significantly increasing the complexity of user-side cost optimization. Against this backdrop, utilizing energy storage systems for peak shaving and valley filling to reduce total electricity costs has become a core requirement and crucial tool for industrial users. However, existing energy storage system optimization and scheduling methods often suffer from numerous technical limitations, making it difficult to meet the application needs of the current complex market environment. Existing industrial user energy storage systems face three core bottlenecks when participating in the electricity spot market: insufficient response to dynamic price signals, thus limiting arbitrage opportunities; lack of coordination between demand and energy costs in day-ahead planning, making it difficult to minimize total electricity costs; and insufficient model solution efficiency and practicality under complex constraints.

[0003] Chinese patent CN118537160A, entitled "A Method and System for Energy Management of Photovoltaic-Storage-Automobile Parks Based on Maximum Demand Estimation," describes a method for energy management in such parks. This method includes: collecting historical photovoltaic (PV) data and historical load data for the park; performing cluster analysis on these data using a clustering algorithm to determine typical PV-load days; dividing these typical PV-load days into multiple preset time periods and collecting the number of electric vehicles arriving at and leaving the park during each time period; updating the V2G resource pool for the park based on the number of electric vehicles; constructing a rolling optimization model based on the typical PV-load days and the updated V2G resource pool, combined with time-of-use pricing and two-part tariffs; solving the energy management problem using the rolling optimization model; and implementing energy management based on the solution results. This method can improve renewable energy consumption, reduce park electricity costs, and generate revenue for users during idle periods without affecting the user experience. However, this method treats demand control as an insurmountable hard constraint rather than a decision variable that can be optimized by weighing electricity consumption and electricity costs. It lacks the ability to proactively and collaboratively optimize the maximum demand, making it difficult to fundamentally minimize the total cost of electricity purchase. It may fall into a suboptimal dilemma of "saving on electricity consumption and costs, but increasing the cost of demand electricity costs."

[0004] Chinese patent CN118195211A, entitled "Energy Management Method for Industrial Parks Based on Improved NSGA-II and GA-BP Combined Algorithm," presents a power prediction model based on the GA-BP algorithm, an industrial park microgrid model, and an optimized scheduling model based on the improved NSGA-II algorithm. Using models specific to industrial parks as a foundation, and combining predicted distributed energy generation and load power, the improved NSGA-II algorithm is central to achieving optimal coordinated scheduling for energy management in commercial parks. However, its optimization model is based on a price curve composed of fixed peak, flat, and valley periods. Such models cannot effectively respond to dynamic price signals in the spot market, which occur over periods of 15 minutes or even shorter. While this method addresses uncertainty to some extent, it is essentially a stochastic search algorithm, suffering from slow convergence and unstable results. It struggles to guarantee high-quality, executable day-ahead plans within the timeframes required by engineering projects, thus failing to meet the dual high demands of the spot market for decision-making speed and accuracy. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides a day-ahead dispatching method for power storage, which solves the problems of difficulty in achieving efficient and accurate energy storage dispatching in existing technologies.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A day-ahead dispatch method for power storage includes the following steps: Collect electricity data for input into the mixed-integer programming model. The electricity data includes the next day's spot market clearing price and historical load data. Calculate the maximum demand constraint value for the next day based on historical load data. , The expression is: ; In the formula, This represents the average of the highest historical demand for the same month in the current month. This represents the average maximum demand over the specified time period preceding the current month. μ Indicates the weighting coefficient; Construct a mixed-integer programming model. The input information of the mixed-integer programming model includes power data and maximum demand constraints. The objective function of the mixed-integer programming model is to minimize the total electricity consumption and electricity cost on the operating day. The expression for the maximum demand constraint is: ; In the formula, t represents the time number. T represents the total number of time periods in the scheduling cycle. This represents the power purchased by the power grid at time t. This represents the maximum demand constraint value for the next day; The constructed mixed integer programming model is solved; if the solution is successful, the scheduling plan for the next day is generated and sent to the energy storage management module; if the solution fails, an alarm is issued; wherein, the scheduling plan for the next day includes one or more of the following data: energy storage charging power at each time, energy storage discharging power at each time, grid power purchase at each time, and energy storage state of charge at each time. The energy storage management module executes the next day's scheduling plan. During the operation day, it monitors the real-time demand during the execution of the next day's scheduling plan and uses the real-time demand to update the input information of the mixed-integer programming model. It also compares the real-time demand with... Comparison: If the real-time demand is greater than the set threshold, the intervention logic is triggered to adjust the real-time demand to be less than or equal to the set threshold.

[0007] The beneficial effects of this invention are: This invention embeds dynamic spot electricity prices as the core signal into the objective function, enabling the objective function to accurately respond to price fluctuations and deeply explore arbitrage potential. By innovatively introducing adaptive demand constraints and integrating them as rigid conditions into the objective function's constraints, it ensures that electricity savings do not come at the cost of higher demand costs, achieving forward-looking synergistic optimization of electricity and demand costs during the decision-making stage. The invention effectively overcomes the problem of solving complex constraints by employing a mixed-integer linear programming framework, significantly improving the efficiency, stability, and engineering practicality of the solution. Furthermore, by deeply integrating market signals, demand management, and energy storage physical characteristics, this invention establishes an integrated optimization scheduling framework, generating day-ahead scheduling plans that combine optimal economics, operational safety, and physical feasibility, while ensuring the efficiency and stability of the optimization solution process. This provides users with a precise, reliable, and practical decision support tool for participating in the electricity market.

[0008] Based on the above technical solution, the present invention can be further improved as follows.

[0009] As a preferred technical solution, the objective function of the mixed-integer programming model is expressed as follows: ; In the formula, Indicates time interval, This represents the day-ahead clearing price of the spot market at time t.

[0010] The beneficial effects of adopting the above-mentioned preferred technical solution are: Based on the next-day clearing price of the spot market, multi-objective optimization is achieved through maximum demand constraints, and forward-looking synergistic optimization of electricity and demand costs during the decision-making stage.

[0011] As a preferred technical solution, the input information of the mixed integer programming model also includes one or more of the following: the predicted load power value for the next day. Forecast value of photovoltaic power generation for the next day Rated energy storage capacity Maximum charging power of energy storage Maximum discharge power of energy storage Energy storage charging efficiency Energy storage and discharge efficiency Minimum State of Charge for Energy Storage Energy storage highest state of charge .

[0012] The beneficial effects of adopting the above-mentioned preferred technical solution are: With further enriched load and photovoltaic forecasts, spot electricity prices, energy storage parameters, and historical load data used to calculate demand constraints, it is more conducive to achieving accurate responses to market prices and in-depth exploration of arbitrage potential.

[0013] As a preferred technical solution, the input information of the mixed-integer programming model also includes power balance constraints, the expression of which is: ; In the formula, This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t.

[0014] The beneficial effects of adopting the above-mentioned preferred technical solution are: Power balance constraints require that the total power generation of the system must always be equal to the total power consumption, that is, to maintain instantaneous power balance and ensure that the power system can operate safely and stably.

[0015] As a preferred technical solution, the input information of the mixed-integer programming model also includes energy storage dynamic constraints, the expression of which is: ; In the formula, This represents the state of charge of the stored energy at the end of time t; in: The expression for the boundary conditions is: , ; In the formula, Indicates the initial state of charge of the energy storage system. Indicates the state of charge at the end of the scheduling cycle; The expression for the update equation is: ; In the formula, Indicates time interval, This represents the state of charge of the stored energy at the end of time t-1. This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t.

[0016] The beneficial effects of adopting the above-mentioned preferred technical solution are: Dynamic constraints on energy storage ensure that the energy storage state conforms to physical laws during the optimization process, preventing equipment from being overcharged or over-discharged and operating at excessive power.

[0017] As a preferred technical solution, the input information of the mixed-integer programming model also includes ramp power constraints, the expression of which is: ; In the formula, Indicates the maximum uphill gradient. Indicates the maximum downhill gradient. This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t. This represents the energy storage discharge power at time t-1. This represents the energy storage charging power at time t-1.

[0018] The beneficial effects of adopting the above-mentioned preferred technical solution are: The ramp power constraint limits the rate of power change of an energy storage system between two adjacent moments.

[0019] As a preferred technical solution, the input information of the mixed integer programming model also includes a maximum charge-discharge cycle count constraint, the expression of which is: ; In the formula, This represents the initial variable of the charge-discharge cycle at time t. , This indicates the maximum number of complete state transitions for energy storage charge and discharge that are allowed per day.

[0020] The beneficial effects of adopting the above-mentioned preferred technical solution are: The maximum charge-discharge cycle count constraint limits the maximum number of complete energy storage charge-discharge state transitions within one operating cycle.

[0021] As a preferred technical solution, the input information of the mixed-integer programming model also includes charging and discharging mutual exclusion constraints, the expression of which is: , ; In the formula, Represents the binary decision variables of energy storage operation status. , This represents the energy storage discharge power at time t. Let t represent the energy storage charging power at time t, and M represent the variable coefficient.

[0022] The beneficial effects of adopting the above-mentioned preferred technical solution are: The strict mutual exclusion constraint between charging and discharging ensures the mutual exclusion of the charging and discharging states of the energy storage system, thus ensuring the physical feasibility of the scheduling plan.

[0023] As a preferred technical solution, the input information of the mixed-integer programming model also includes power constraint, the expression of which is: , ; In the formula, This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t.

[0024] The beneficial effects of adopting the above-mentioned preferred technical solution are: Power limitation constraints stipulate that the charging and discharging power of an energy storage system at any given time must not exceed its maximum permissible power.

[0025] As a preferred technical solution, the constructed mixed-integer programming model is solved; if the solution is successful, a scheduling plan is generated and sent to the energy storage management module; if the solution fails, an alarm is issued; including the following steps: The mathematical programming solver is invoked to solve the constructed mixed integer programming model; the maximum solution time and optimal gap threshold of the mathematical programming solver are set before solving. The mathematical programming solver branches according to the solution status: If the mathematical programming solver successfully obtains a feasible solution within a solution time ≤ the maximum solution time and within a gap ≤ the optimal gap threshold, then the optimal solution is extracted from the feasible solution, a scheduling plan for the next day is generated, and sent to the energy storage management module. If the mathematical programming solver takes longer than the maximum solve time and fails to find a feasible solution, an alarm will be issued.

[0026] The beneficial effects of adopting the above-mentioned preferred technical solution are: By setting the maximum solution time and the optimal gap threshold, the calculation process is ensured to be completed within a controllable time, meeting the timeliness requirements of engineering decision-making and avoiding indefinite waiting or system unresponsiveness. By providing clear alarms and diagnostic guidance for solution failures, the reliability and usability of the system are enhanced, ensuring that users can respond and adjust quickly when encountering problems, and guaranteeing the continuous and stable operation of the optimization system.

[0027] Compared with the prior art, the present invention has the following advantages: (1) This invention embeds the dynamic spot electricity price as the core signal into the objective function, enabling the objective function to accurately respond to price fluctuations and deeply explore arbitrage potential; by innovatively introducing adaptive demand constraints and incorporating them as rigid conditions into the constraints of the objective function, it ensures that the saved electricity costs will not come at the cost of higher demand costs, thus achieving forward-looking synergistic optimization of electricity and demand costs in the decision-making stage; by adopting a mixed integer linear programming framework, this invention effectively overcomes the problem of solving under complex constraints, significantly improving the efficiency, stability and engineering practicality of the solution; by deeply integrating market signals, demand management and energy storage physical characteristics, this invention establishes an integrated optimization scheduling framework, generating a day-ahead scheduling plan that combines optimal economy, operational safety and physical feasibility, while ensuring the efficiency and stability of the optimization solution process, thereby providing users with an accurate, reliable and practical decision support tool for participating in the electricity market; (2) Based on the next day's clearing spot market electricity price and the maximum demand constraint, multi-objective optimization is achieved, and forward-looking collaborative optimization of electricity and demand costs is achieved in the decision-making stage; (3) Based on further enriched load and photovoltaic forecasts, spot electricity prices, energy storage parameters and historical load data used to calculate demand constraints, it is more conducive to achieving accurate response to market prices and in-depth exploration of arbitrage potential; (4) The power balance constraint restricts the total power generation of the system to be equal to the total power consumption at all times, that is, to maintain instantaneous power balance and ensure that the power system can operate safely and stably; (5) Dynamic constraints on energy storage ensure that the energy storage state conforms to physical laws during the optimization process, and prevent the equipment from being overcharged or over-discharged and operating at excessive power; (6) The ramp power constraint limits the rate of power change of the energy storage system between two adjacent moments; (7) The maximum number of charge-discharge cycles is limited to the maximum number of complete state transitions of energy storage charge and discharge within one operating cycle. (8) The mutual exclusion constraint of charging and discharging strictly ensures the mutual exclusion of the charging and discharging states of the energy storage system, thus ensuring the physical feasibility of the scheduling plan; (9) Power limitation constraints limit the charging and discharging power of the energy storage system at any given time to not exceed its maximum allowable power; (10) By setting the maximum solution time and the optimal gap threshold, the calculation process is ensured to be completed within a controllable time, which meets the timeliness requirements of engineering decision-making and avoids indefinite waiting or system unresponsiveness; by providing clear alarms and diagnostic guidance for solution failure, the reliability and practicality of the system are enhanced, ensuring that users can respond and adjust quickly when encountering problems, and ensuring the continuous and stable operation of the optimization system. Attached Figure Description

[0028] Figure 1 This is a schematic diagram illustrating the steps of a day-ahead dispatching method for power storage proposed in this invention; Figure 2 This is a diagram of the user-side optical-storage collaborative system architecture of the present invention; Figure 3 This is a data interaction diagram of the present invention; Figure 4 A power balance analysis diagram provided for an embodiment of the present invention; Figure 5 A state-of-charge curve of an energy storage system provided in an embodiment of the present invention; Figure 6 This is a diagram showing the relationship between time-of-use electricity pricing and electricity purchase cost, provided for an embodiment of the present invention. Detailed Implementation

[0029] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0030] The principles and features of the present invention are described below. The embodiments given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0031] Example 1 like Figures 1 to 6 As shown, to address the problems of existing technologies, this invention provides a day-ahead optimal scheduling method for industrial user energy storage systems based on mixed integer programming. This method establishes an integrated optimal scheduling framework by deeply integrating market signals, demand management, and energy storage physical characteristics. It employs an efficient mathematical programming solver and sets intelligent solution strategies to generate day-ahead scheduling plans that combine optimal economics, operational safety, and physical feasibility. Simultaneously, it ensures the efficiency and stability of the optimization solution process, thereby providing industrial users with an accurate, reliable, and practical decision support tool for participating in the electricity market.

[0032] This invention constructs a day-ahead optimization framework consisting of a data management module (DM), a demand calculation module (MC), a day-ahead optimization module (DA), and an energy storage management module (ESS). This framework, through the organic coordination of these modules, integrates market signal response, demand constraint settings, energy storage physical characteristics, and efficient solution strategies, ultimately achieving the goal of economical and efficient scheduling of the energy storage system.

[0033] A day-ahead optimization scheduling method for industrial user energy storage systems based on mixed-integer programming includes the following steps: The S1 Data Management module is responsible for collecting and integrating all input data required for the next day's optimization, including 96-point load and photovoltaic power generation forecast data, spot electricity prices, energy storage parameters, and historical load data used to calculate demand constraints. Addressing the limitations of existing technologies in responding to market signals, this module strengthens the collection and processing of dynamic spot electricity prices—a core economic signal. This price data will serve as a key input driving the "low storage, high generation" arbitrage strategy of the energy storage system, directly embedded in the objective function of the subsequent optimization model to achieve accurate market price response and in-depth exploitation of arbitrage potential. Subsequently, the collected multi-source data is cleaned, aligned, and validated to ensure data quality and consistency, laying the foundation for building a highly reliable optimization model.

[0034] S2. Maximum Demand Constraint Setting and Collaborative Optimization Mechanism. The demand calculation module calculates the maximum demand constraint value for the next day based on historical data. The calculation method for the maximum demand constraint value for the next day is as follows: in, This is the average of the highest demand for the same month in the historical database, which is the same as the current month. This represents the average of the highest demand over the past three months. μ These are the weighting coefficients. This method incorporates recent electricity consumption trends ( ) and seasonal electricity consumption patterns ( The weighting coefficient μ is adjusted to ensure that the target demand value is no longer a static empirical value, providing flexible boundary conditions for subsequent optimization scheduling. The day-ahead optimization module will calculate the... As a rigid constraint, it is fully integrated into the subsequent mixed-integer programming (MILP) model, and the decision variables of the optimization problem are defined accordingly, including the energy storage charging and discharging power and the power purchased by the grid at each time. Demand constraints are proactively incorporated into the optimization model to achieve coordinated optimization of electricity cost and demand cost in the day-ahead decision-making stage.

[0035] S3. Construct a Mixed Integer Programming (MILP) model with the objective of minimizing electricity purchase costs. This stage is completed independently by the day-ahead optimization module. Based on the input data of S1 and the demand constraints and decision variables of S2, the day-ahead optimization module constructs a MILP model with the objective of minimizing electricity purchase costs. This model uses minimizing the total electricity consumption and cost on the operating day as the objective function and integrates multiple conditions, including power balance constraints, maximum demand constraints, energy storage system dynamic constraints, ramp power constraints, maximum charge / discharge cycle constraints, charge / discharge mutual exclusion constraints, and power limit constraints. By collaboratively solving this model, it is ensured that the generated scheduling plan, while pursuing economic optimization, strictly meets all physical operational constraints and safety requirements.

[0036] Power balance constraints require the total generating power of the system to always equal the total power consumption, maintaining instantaneous power balance and ensuring the safe and stable operation of the power system. Maximum demand constraints achieve joint optimization of electricity cost and demand cost, ensuring that energy savings do not come at the expense of higher demand costs. Energy storage dynamic constraints ensure that the energy storage state conforms to physical laws during optimization, preventing overcharging, over-discharging, and over-power operation. Ramp-up power constraints limit the rate of power change of the energy storage system between two adjacent moments. Maximum charge / discharge cycle constraints limit the maximum number of complete charge / discharge state transitions within one operating cycle. Charge / discharge mutual exclusion constraints use the Big-M method and introduce binary variables, mathematically guaranteeing the mutual exclusion of the energy storage system's charge / discharge states and ensuring the physical feasibility of the scheduling plan. Power limitation constraints restrict the charging and discharging power of the energy storage system at any given moment from exceeding its maximum allowable power.

[0037] S4. The optimization module calls the mathematical programming solver to solve the mixed integer programming model constructed in S3. If successful, it generates the optimal 96-point scheduling plan; if it fails, it alerts the user.

[0038] By setting a maximum solution time and an optimal gap threshold, the computation process is ensured to be completed within a controllable timeframe, meeting the timeliness requirements of engineering decisions and avoiding indefinite waiting or system unresponsiveness. Providing clear alarms and diagnostic guidance for solution failures enhances the system's reliability and usability, ensuring users can respond and adjust quickly when encountering problems, and guaranteeing the continuous and stable operation of the optimization system.

[0039] S5. The day-ahead optimization module distributes the optimization plan to the energy storage management module for execution. Real-time monitoring and adjustments are performed during the operating day, and operational analysis is conducted at the end of the day, storing the actual operational feedback in the data management module. By constructing a two-layer collaborative control architecture of "day-ahead planning and real-time correction," a demand exceedance risk warning and dynamic adjustment mechanism based on real-time power and SOC data is added to the execution of the economical day-ahead plan. Once a risk of real-time demand exceeding a set threshold is detected, intervention logic is immediately triggered to adjust subsequent charging and discharging power, ensuring that demand does not exceed limits. This ensures the safety and reliability of the system in complex operating environments (if real-time demand and SOC data are within limits). If the difference between the real-time demand and the set threshold is greater than the threshold, the intervention logic is triggered to adjust the real-time demand so that it is equal to the threshold value. (The difference is ≤ a set threshold). After the end of the day, the system automatically performs operational analysis and archives data such as actual load, photovoltaic power, electricity price, and SOC curve into the historical database in the data management module. This data is used to optimize the accuracy of subsequent forecasts and decision parameters, enabling the system to have continuous self-learning and adaptive optimization capabilities.

[0040] This invention embeds dynamic spot electricity prices as the core signal into the objective function, enabling the objective function to accurately respond to price fluctuations and deeply explore arbitrage potential. By innovatively introducing adaptive demand constraints and incorporating them as rigid conditions into the objective function's constraints, it achieves forward-looking synergistic optimization of electricity consumption and demand costs during the decision-making stage. Furthermore, by adopting a mixed-integer linear programming framework, integrating efficient solvers, and setting intelligent solution strategies, it effectively overcomes the problem of solving under complex constraints, significantly improving the efficiency, stability, and engineering applicability of the solution.

[0041] Example 2 like Figures 1 to 6 As shown, based on Example 1, this example provides a more detailed implementation method.

[0042] A day-ahead optimization scheduling method for industrial user energy storage systems based on mixed-integer programming includes the following steps: The S1 data management module is responsible for collecting and integrating all input data required for the next day's optimization, including 96-point load and photovoltaic forecasts, spot electricity prices, energy storage parameters, and historical load data used to calculate demand constraints. The specific steps are as follows: The S1.1 data management module collects all the input data required for optimization at 96 points the following day (with a time resolution of 15 minutes) from data sources such as the park's energy management system, power trading platform, and historical databases. Specifically, this includes the load power forecast values ​​for the next 96 points. Photovoltaic power generation forecast The next day at 96 points, the day-ahead spot market clearing electricity price forecast sequence Rated capacity of energy storage system Maximum charging / discharging power Charge / discharge efficiency And the upper and lower limits of the State of Charge (SOC) operation.

[0043] To address the limitations of existing technologies in responding to market signals, this module provides the day-ahead optimization module with a 96-point day-ahead spot market electricity price sequence for the following day. This electricity price data is the most critical input driving the energy storage system of this invention to perform "low-storage, high-generation" arbitrage and achieve economic optimization.

[0044] The S1.2 data management module performs unified cleaning, timestamp alignment, unit conversion, and validity verification on all collected multi-source heterogeneous data, eliminating data contradictions and errors, and ensuring a high degree of consistency of all data in terms of time and logic, thereby providing a solid data foundation for building a highly reliable and accurate optimization model.

[0045] S2. Maximum Demand Constraint Setting and Collaborative Optimization Mechanism. The demand calculation module calculates the maximum demand constraint value for the next day based on historical data. The specific implementation steps are as follows: The demand calculation module retrieves historical load data from the data management module, including historical data for the same month and recent operational data. Based on this data, the module determines the maximum demand constraint value for the next day, calculated using the following formula: in, This is the average of the highest demand for the same month in the historical database, which is the same as the current month. This represents the average of the highest demand over the past three months. μ These are the weighting coefficients. This method incorporates recent electricity consumption trends ( ) and seasonal electricity consumption patterns ( The weighting coefficient μ is adjusted to ensure that the target demand value is no longer a static empirical value, providing flexible boundary conditions for subsequent optimization scheduling. The day-ahead optimization module will calculate the... As a rigid constraint, it is fully integrated into the subsequent mixed integer programming (MILP) model, and the decision variables of the optimization problem are defined accordingly, including the energy storage charging and discharging power and the power purchased by the grid at each time. Demand constraints are proactively incorporated into the optimization model, thereby achieving coordinated optimization of electricity consumption and demand costs in the day-ahead decision-making stage, ensuring the minimization of the total electricity cost of the park.

[0046] S3. Construct a mixed-integer programming (MILP) model with the objective of minimizing electricity purchase costs. This stage is completed independently by the day-ahead optimization module. Its core task is based on the input data provided in stage S1 and the demand constraints set in stage S2. A mixed-integer linear programming (MILP) model is constructed with the objective of minimizing electricity purchase costs. This model integrates multiple constraints to ensure that the generated scheduling plan is economically optimal while strictly meeting all physical operational limitations and safety requirements. The specific implementation steps are as follows: S3.1. Define the objective function: With the optimization objective of minimizing the total electricity consumption and cost per operating day, its mathematical expression is: in The time interval is set to 0.25 hours (i.e., 15 minutes). It is the day-ahead spot market clearing price (yuan / kWh) at time t. It is the power purchased by the power grid at time t (kW).

[0047] S3.2. Set constraints: To ensure the safe operation of the system and the physical feasibility of the solution, the model must meet the following constraints: S3.2.1. Power balance constraint: in, It represents the power purchased by the power grid at time t. It is the predicted photovoltaic output at time t. It is the energy storage discharge power at time t. It is the predicted load demand at time t. This represents the energy storage charging power at time t. This constraint ensures that at any given time t, the system's total power supply (grid purchase, photovoltaic output, and energy storage discharge) must be balanced with its total power consumption (load demand and energy storage charging). This is the most fundamental constraint to guarantee the stable operation of the system.

[0048] S3.2.2. Maximum Demand Constraint: in, It represents the power purchased by the power grid at time t. This is the maximum demand constraint for the following day. This constraint ensures that the grid's power purchases at any given time do not exceed the demand limit determined by stage S2. .

[0049] S3.2.3. Dynamic constraints of energy storage systems: a) SOC update equation: This equation describes the continuous variation of the energy storage state of charge (SOC). Among them, The energy storage state of charge (kWh) at the end of time t. , These are the charging efficiency and discharging efficiency of energy storage, respectively. The rated capacity (kWh) of the energy storage.

[0050] b) Boundary conditions: in, This is the initial state of charge of the stored energy. This represents the minimum state of charge for energy storage. Initial conditions are set to optimize the initial energy storage state. Terminal conditions ensure that the energy storage has enough power at the end of the day to meet the load demand or emergencies at the beginning of the next day.

[0051] c) SOC upper and lower limit constraints: in, This is the lowest state of charge for energy storage. This represents the highest state of charge (SOC) for energy storage. This constraint limits the SOC of energy storage to a safe operating range, preventing overcharging and over-discharging, and extending battery life.

[0052] S3.2.4. Climbing Power Constraint: This constraint limits the rate of power change of the energy storage system between two adjacent moments, preventing sudden power fluctuations, reducing the impact on the energy storage converter and the power grid, and improving the smoothness and safety of system operation. in and These are the maximum uphill gradient and the maximum downhill gradient (kW), respectively.

[0053] S3.2.5. Maximum charge / discharge cycle count constraint: This constraint aims to extend the lifespan of energy storage batteries by limiting the maximum number of complete state transitions between charge and discharge states within an operating cycle, thus preventing damage to battery health from excessively frequent charge-discharge cycles.

[0054] Introducing auxiliary binary variables This is used to identify the start of a charge / discharge cycle (e.g., the transition from discharging to charging). The constraint can be expressed as: in This represents the maximum number of complete cycles allowed per day.

[0055] S3.2.6. Charge-discharge mutual exclusion constraint (Big-M method): Introducing binary variables (Binary decision variables for energy storage operation status) and variable coefficients (a positive real number whose value is greater than or equal to the feasible solution of the energy storage system's charging power and discharging power, such as...) Construct the following constraints: in, The maximum charging power (kW) for energy storage. This represents the maximum discharge power (kW) of the energy storage system. This set of constraints ensures that the energy storage system can operate at any given time. t It can only be in one of three states: charging, discharging, or standby, strictly avoiding the situation of charging and discharging simultaneously, which violates the physical characteristics.

[0056] S3.2.7. Power Limitation Constraints: in, The maximum charging power (kW) for energy storage. This is the maximum discharge power (kW) of the energy storage. This constraint limits the charging and discharging power of the energy storage to within the rated capacity of the device.

[0057] By collaboratively solving the MILP model, the day-ahead optimization module can generate a complete 96-point scheduling plan that, while pursuing the minimization of electricity purchase costs, strictly adheres to all physical operational constraints and safety requirements of the system.

[0058] S4. The optimization module calls the mathematical programming solver to solve the mixed integer programming model constructed in S3. If successful, it generates the optimal 96-point scheduling plan; if it fails, it alerts the user. The specific implementation steps are as follows: S4.1. The day-ahead optimization module calls the integrated high-performance mathematical programming solver (such as Gurobi, CPLEX, etc.) to solve the constructed MILP model. To meet the stringent timeliness and reliability requirements of practical engineering applications, key parameters of the solver need to be configured before solving: the maximum solution time is set according to the real-time requirements of the engineering application. This ensures that the optimization calculation process is completed within a controllable time, meeting the timeliness requirements of day-ahead decision-making. An acceptable optimal gap (MIP Gap) threshold is set. This strategy strikes a balance between solution accuracy and computational efficiency, allowing the solver to terminate the calculation after finding a feasible solution that meets the accuracy requirement, avoiding excessive time spent pursuing the theoretically optimal solution.

[0059] S4.2. Subsequently, the system performs branching processing based on the solution status: If the solver successfully finds a feasible solution within the set time and optimal interval, the day-ahead optimization module extracts the optimal solution from the solution results and generates a detailed scheduling plan for 96 time periods the following day. This plan includes the energy storage charging power, discharging power, grid power purchase, and SOC state trajectory for each time period, and is immediately prepared to be sent to the Energy Storage Management Module (ESS) for execution.

[0060] If the solver times out or fails to find a feasible solution, the DA module will immediately trigger an alarm mechanism, sending a clear fault alarm message to the user interface. This alarm message not only indicates the failure status but also provides preliminary diagnostic guidance and adjustment suggestions (such as checking the validity of input data, relaxing certain constraints, or adjusting demand constraint values). This enhances the reliability and usability of the system, ensuring that users can respond and adjust quickly when encountering problems, effectively guaranteeing the continuous and stable operation of the optimization system.

[0061] S5. The day-ahead optimization module distributes the optimization plan to the energy storage management module for execution. Real-time monitoring and adjustments are performed during the operating day, and operational analysis is conducted at the end of the day. Actual operational feedback is stored in the data management module. This stage is the final implementation and continuous optimization phase of the optimized scheduling scheme. Its core task is to ensure the safe execution of the day-ahead plan, guarantee system reliability through real-time monitoring and dynamic adjustments during operation, and ultimately form a closed-loop learning mechanism for continuous improvement through end-of-day analysis. Specific implementation steps are as follows: S5.1. Plan Issuance and Execution. The day-ahead optimization module issues the 96-point optimal scheduling plan generated in stage S4 to the energy storage management module. After receiving the plan, the energy storage management module converts it into specific control commands, driving the energy storage converter (PCS) to strictly execute the charging, discharging, or standby operations at each time point on the operating day according to the plan.

[0062] During the execution of the plan, S5.2 activates a real-time monitoring and dynamic correction mechanism, constructing a two-layer collaborative control architecture of "day-ahead planning and real-time correction." The energy storage management module feeds back the real-time operating status of the energy storage to the day-ahead optimization module at a frequency of minutes, mainly including the state of charge and actual charging and discharging power. Based on the feedback data, the day-ahead optimization module calculates the average power purchased within the rolling time window in real time and compares it with the demand constraint value set in stage S2. Once it detects that the real-time demand > the set threshold (e.g., 0.95), the system will take action. When the demand peak is reached, the optimization module will immediately trigger the intervention logic and send a dynamic adjustment command to the ESS module (such as increasing the discharge power in advance or limiting the charging power) to actively suppress the demand peak and adjust the real-time demand to ≤ the set threshold to ensure that the demand does not exceed the limit in actual operation.

[0063] S5.3. After the day's operation ends, the system automatically enters the end-of-day analysis phase, completing the data loop. The system compares and analyzes the day-ahead plan with the actual operating results, calculating key performance indicators such as real-time peak demand, total electricity purchase cost, and the deviation rate between the plan and the actual demand. Subsequently, all actual operating data, including actual load curves, photovoltaic output curves, spot electricity price sequences, and SOC trajectories, are fully archived into the historical database of the data management module.

[0064] To fully disclose the content of this invention and enable those skilled in the art to understand and implement it, the following detailed description of the park energy management method based on mixed integer programming described in this application is provided through a specific exemplary embodiment. It should be particularly noted that this embodiment is only used to assist in illustrating the technical solution and beneficial effects of this invention, and its specific parameters, scenarios, and some simplifications should not be construed as any limitation on the scope of protection of this invention.

[0065] This embodiment uses the Python programming language and the Pyomo optimization modeling library to fully implement the aforementioned day-ahead optimization scheduling method. To focus on verifying the effectiveness of the method in the core collaborative optimization mechanism of day-ahead market arbitrage and demand management, and to reduce the complexity of the model in the early stages of verification, this embodiment intentionally relaxes constraints such as ramp power constraints and maximum charge-discharge cycle count constraints.

[0066] This embodiment strictly follows Figure 2 The system architecture shown implements the following core classes using object-oriented programming: The `DataManager` class corresponds to the data management module, responsible for the generation, verification, and management of sample data. The `EnergyStorageSystem` class corresponds to the energy storage system, encapsulating various parameters of the energy storage devices. The `DayAheadOptimization` class corresponds to the day-ahead optimization module, which is the core of this embodiment and is responsible for constructing and solving the mixed-integer linear programming (MILP) model. This class explicitly defines the objective function and constraints (including power balance constraints, maximum demand constraints, and energy storage dynamic constraints) as described in the claims. The `ResultVisualizer` class is responsible for the visualization and analysis of the optimization results.

[0067] After executing the optimization method, the system successfully solved and generated an optimized scheduling plan. Figures 4 to 6 The results of applying the method of this invention on a typical day are shown. Figure 4 The power balance time-series curves shown indicate that the grid's power purchase is strictly limited to the preset maximum demand constraint of 800kW. Meanwhile, the energy storage system charges during off-peak hours and discharges during peak load and electricity price periods, effectively achieving the optimization goal of "peak shaving and valley filling". Figure 5This further demonstrates that the State of Charge (SOC) of the energy storage system remains within a safe range throughout the entire daily scheduling cycle, verifying the physical feasibility and operational sustainability of the scheduling plan. Figure 6 This correlates the dynamic electricity price sequence with the time-of-use electricity purchase cost, intuitively revealing the economic driving logic of optimization decisions based on price signals. The results in the accompanying figures collectively demonstrate that the method described in this invention works effectively, achieving significant economic optimization results while ensuring the safe operation of the system.

[0068] As described above, the present invention can be implemented well.

[0069] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0070] In the description of this invention, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0071] In the description of this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0072] In the description of this invention, although embodiments of the invention have been shown and described herein, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this invention.

[0073] In the description of this invention, all features disclosed in all embodiments of this specification, or steps in all methods or processes implied in the disclosure, may be combined and / or extended or replaced in any way, except for mutually exclusive features and / or steps.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A day-ahead dispatching method for power storage, characterized in that, Includes the following steps: Collect electricity data for input into the mixed-integer programming model. The electricity data includes the next day's spot market clearing price and historical load data. Calculate the maximum demand constraint value for the next day based on historical load data. , The expression is: ； In the formula, This represents the average of the highest historical demand for the same month in the current month. This represents the average maximum demand over the specified time period preceding the current month. μ Indicates the weighting coefficient; Construct a mixed-integer programming model. The input information of the mixed-integer programming model includes power data and maximum demand constraints. The objective function of the mixed-integer programming model is to minimize the total electricity consumption and electricity cost on the operating day. The expression for the maximum demand constraint is: ； In the formula, t represents the time number. T represents the total number of time periods in the scheduling cycle. This represents the power purchased by the power grid at time t. This represents the maximum demand constraint value for the next day; The constructed mixed integer programming model is solved; if the solution is successful, the scheduling plan for the next day is generated and sent to the energy storage management module; if the solution fails, an alarm is issued; wherein, the scheduling plan for the next day includes one or more of the following data: energy storage charging power at each time, energy storage discharging power at each time, grid power purchase at each time, and energy storage state of charge at each time. The energy storage management module executes the next day's scheduling plan. During the operation day, it monitors the real-time demand during the execution of the next day's scheduling plan and uses the real-time demand to update the input information of the mixed-integer programming model. It also compares the real-time demand with... Comparison: If the real-time demand is greater than the set threshold, the intervention logic is triggered to adjust the real-time demand to be less than or equal to the set threshold.

2. The day-ahead dispatching method for power storage according to claim 1, characterized in that, The objective function of the mixed-integer programming model is expressed as: ； In the formula, Indicates time interval, This represents the day-ahead clearing price of the spot market at time t.

3. The day-ahead dispatching method for power storage according to claim 2, characterized in that, The input information for the mixed-integer programming model also includes one or more of the following: the next day's load power forecast. Forecast value of photovoltaic power generation for the next day Rated energy storage capacity Maximum charging power of energy storage Maximum discharge power of energy storage Energy storage charging efficiency Energy storage and discharge efficiency Minimum State of Charge for Energy Storage Energy storage highest state of charge .

4. The day-ahead dispatching method for power storage according to claim 3, characterized in that, The input information for the mixed-integer programming model also includes power balance constraints, the expression of which is: ； In the formula, This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t.

5. The day-ahead dispatching method for power storage according to claim 3, characterized in that, The input information for the mixed-integer programming model also includes dynamic constraints on energy storage, the expression of which is: ； In the formula, This represents the state of charge of the stored energy at the end of time t; in: The expression for the boundary conditions is: ， ； In the formula, Indicates the initial state of charge of the energy storage system. Indicates the state of charge at the end of the scheduling cycle; The expression for the update equation is: ； In the formula, Indicates time interval, This represents the state of charge of the stored energy at the end of time t-1. This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t.

6. The day-ahead dispatching method for power storage according to claim 3, characterized in that, The input information for the mixed-integer programming model also includes ramp power constraints, the expression for which is: ； In the formula, Indicates the maximum uphill gradient. Indicates the maximum downhill gradient. This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t. This represents the energy storage discharge power at time t-1. This represents the energy storage charging power at time t-1.

7. The day-ahead dispatching method for power storage according to claim 3, characterized in that, The input information for the mixed-integer programming model also includes a maximum charge-discharge cycle count constraint, the expression for which is: ； In the formula, This represents the initial variable of the charge-discharge cycle at time t. , This indicates the maximum number of complete state transitions for energy storage charge and discharge that are allowed per day.

8. The day-ahead dispatching method for power storage according to claim 3, characterized in that, The input information for the mixed-integer programming model also includes charge-discharge mutual exclusion constraints, the expression for which is: ， ； Represents the binary decision variables of energy storage operation status. , This represents the energy storage discharge power at time t. Let t represent the energy storage charging power at time t, and M represent the variable coefficient.

9. A day-ahead dispatching method for power storage according to claim 3, characterized in that, The input information for the mixed-integer programming model also includes power constraint, the expression for which is: ， ； In the formula, This represents the energy storage discharge power at time t. This represents the energy storage charging power at time t.

10. A day-ahead dispatching method for power storage according to any one of claims 1 to 9, characterized in that, Solve the constructed mixed-integer programming model; if the solution is successful, generate a scheduling plan and send it to the energy storage management module. If the solution fails, an alarm will be issued; including the following steps: The mathematical programming solver is invoked to solve the constructed mixed integer programming model; the maximum solution time and optimal gap threshold of the mathematical programming solver are set before solving. The mathematical programming solver branches according to the solution status: If the mathematical programming solver successfully obtains a feasible solution within a solution time ≤ the maximum solution time and within a gap ≤ the optimal gap threshold, then the optimal solution is extracted from the feasible solution, a scheduling plan for the next day is generated, and sent to the energy storage management module. If the mathematical programming solver takes longer than the maximum solve time and fails to find a feasible solution, an alarm will be issued.

Images