Energy storage resource scheduling optimization method and system based on digital twinning

Abstract

The invention discloses an energy storage resource scheduling optimization method and system based on digital twinning. The method comprises the following steps: step 1, constructing an energy storage resource scheduling optimization platform based on digital twinning; establishing a digital twinborn model for describing the operation state of the energy storage equipment in combination with the physical characteristics and historical data of the energy storage system; step 2, based on the constructed scheduling optimization platform, performing joint modeling on a wind power generation system, a photovoltaic system and an energy storage system, and establishing an equivalent model of a joint power generation system; 3, based on the established equivalent model of the combined power generation system, establishing a multi-objective optimization function and physical and operation constraint conditions of a scheduling algorithm, and constructing an energy storage system resource scheduling optimization problem; and step 4, based on the established scheduling optimization problem, performing problem solving and dynamic adjustment by using a differential evolution algorithm. The running state of the energy storage system is accurately reflected, and the real-time response capability is improved; the economy and the stability of the scheduling scheme are obviously improved; and the security and feasibility of the scheduling strategy are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grids, and in particular to a method and system for optimizing energy storage resource scheduling based on digital twins. Background Art

[0002] Since energy storage systems play an important role in coping with the volatility and unpredictability of new energy, their application in grid dispatching and operation optimization has received widespread attention in recent years. However, the efficient use of energy storage systems depends on reasonable resource scheduling methods, and traditional scheduling methods are often difficult to dynamically respond to real-time changes in complex power grid environments. For example, the output characteristics of wind and photovoltaic power generation are greatly affected by external factors such as weather, and their volatility may not only lead to grid voltage instability, but also affect the charging and discharging efficiency of the energy storage system. If there is a lack of accurate scheduling strategies, it may not only cause overcharging and discharging of energy storage equipment, but also increase system operating costs and losses of energy storage equipment.

[0003] In the scheduling of energy storage systems, the optimization of charging and discharging strategies is one of the key issues. Energy storage devices need to be charged when the grid load is low and release electricity when the load is peak, so as to achieve peak shaving and valley filling and stabilize grid operation. However, the effective implementation of this process depends on the accurate prediction of grid load, distributed power generation output and energy storage equipment status. Traditional methods are mostly based on static models or historical data, which makes it difficult to capture changes in the system operating status in real time, and it is also difficult to flexibly adjust the scheduling strategy to adapt to dynamic grid needs. Therefore, there is an urgent need for a technical method that can reflect the status of the energy storage system in real time, predict grid changes, and provide dynamic scheduling optimization to meet the challenges brought by the access of new energy to the grid.

[0004] Digital twin technology provides a new idea for the dynamic scheduling optimization of energy storage systems. By establishing a digital twin model of the energy storage system, two-way interaction between the virtual model and the physical device can be achieved, the operating status of the energy storage device can be reflected in real time, and future system change trends can be predicted based on the virtual model. Combined with optimization algorithms, digital twin technology can effectively guide the adjustment of the charging and discharging strategies of energy storage devices, dynamically respond to uncertain needs in complex power grid environments, and thus provide technical guarantees for the stable operation of the power grid.

[0005] The prior art discloses a digital twin method for group dispatching and group control of distributed photovoltaic systems (CN118899914A), including data collection and cluster division, real-time collection of operating data of distributed photovoltaic systems, and determination of similarities and differences of clusters; power prediction for each divided cluster; building a digital twin model for each cluster based on cluster division and power prediction results; designing collaborative control strategies between clusters, including power allocation, energy storage scheduling, etc.; calculating and analyzing the energy consumption characteristics of each cluster, and identifying energy consumption bottlenecks; applying the optimized collaborative control strategy to the actual photovoltaic system, and comparing and analyzing it with the digital twin model to form a closed-loop feedback mechanism.

[0006] Among the existing energy storage resource scheduling optimization algorithms, although some algorithms also combine digital twin technology or particle swarm optimization, there are still the following deficiencies: Insufficient real-time response capability: Traditional scheduling methods are usually based on static models or historical data, and it is difficult to dynamically capture the real-time changes of the energy storage system status and the power grid environment; Lack of effective response strategies for the impact of wind and photovoltaic power generation volatility, which may lead to grid instability and low charging and discharging efficiency of energy storage equipment; Low optimization accuracy: Although some multi-objective optimizations currently take into account system operating costs, energy storage efficiency, and peak-to-valley benefits, they do not comprehensively consider power generation costs, energy storage system investment and construction costs, operation and maintenance costs, power losses and peak-to-valley differences, lack accurate modeling of distributed energy volatility, and are difficult to capture dynamic changes in the power grid in real time; Poor adaptability: Insufficient support for volatility and uncertainty in complex power grid environments, making it difficult to dynamically adjust parameters and optimization strategies. Existing particle swarm optimization algorithms may have insufficient population diversity in a dynamic power grid environment due to excessive reliance on historical optimal solutions, making it difficult to quickly adapt to complex nonlinear changes such as wind and photovoltaic power generation fluctuations. Summary of the invention

[0007] In order to solve the deficiencies in the prior art, the present invention provides a method for optimizing energy storage resource scheduling based on digital twins.

[0008] The present invention adopts the following technical solution.

[0009] The first aspect of the present invention provides a method for optimizing energy storage resource scheduling based on digital twins, comprising:

[0010] Step 1: Build a digital twin-based energy storage resource scheduling optimization platform; combine the physical characteristics and historical data of the energy storage resource scheduling optimization platform to establish a digital twin model that describes the operating status of the energy storage equipment;

[0011] Step 2: Based on the digital twin model of the scheduling optimization platform constructed in step 1, jointly model the wind power generation, photovoltaic system and energy storage system to establish an equivalent model of the joint power generation system;

[0012] Step 3: Based on the equivalent model of the combined power generation system established in step 2, establish the physical and operational constraints of the multi-objective optimization function and the scheduling algorithm, and construct the energy storage system resource scheduling optimization problem;

[0013] Step 4: Based on the scheduling optimization problem established in step 3, the differential evolution algorithm is used to solve the problem and dynamically adjust it to achieve the optimal energy storage resource scheduling.

[0014] Preferably, in step 1, the energy storage resource scheduling optimization platform based on digital twins includes: a measurement system, a modeling and simulation system, and an operation optimization system;

[0015] The measurement system is used to provide the basic data required for energy storage system modeling and optimization;

[0016] The modeling and simulation system is used to simulate the dynamic operation of the system to provide support for the site selection and sizing of the energy storage system;

[0017] The operation optimization system is used to optimize the scheduling of the energy storage system using simulation results and real-time data.

[0018] Preferably, in step 2, wind power generation is modeled using a PQ node, and the reactive power output by the wind turbine is:

[0019]

[0020] Among them, Q W is the reactive power output by the wind turbine; U and P W are the voltage and active power of the wind turbine respectively; x 1 and x 2 They represent the stator leakage reactance and rotor leakage reactance of the wind turbine respectively; x m is the excitation reactance of the wind turbine;

[0021] The photovoltaic system adjusts the output voltage through the voltage source control inverter. The reactive power output by the photovoltaic system is:

[0022]

[0023] Among them, Q P is the reactive power output by the photovoltaic system; P P is the active power output by the photovoltaic system, I is the effective value of the output current of the photovoltaic system, e is the output voltage factor of the inverter control module, and f is the reactive power adjustment factor of the photovoltaic system;

[0024] The energy storage system describes its impact on the grid by injecting power. The injection power formula is:

[0025] P GC =-Pe -i*Q e

[0026] P GD =P e +i*Q e

[0027] Among them, P GC and P GD are the power flows of the energy storage system in charging and discharging states respectively; P e and Q e They are the active power and reactive power output by the energy storage system respectively.

[0028] Preferably, in step 2, the wind power generation and the photovoltaic system constitute a wind-solar power generation system, and the active power and reactive power of the wind-solar power generation system can be expressed as:

[0029] P IDG =P W +P P

[0030] Q IDG =Q W +Q P

[0031] Among them, P IDG and Q IDG are the active power and reactive power of the wind and solar power generation systems respectively; P W and Q W are the active power and reactive power of the wind turbine generator respectively; P P and Q P They are the active power and reactive power output by the photovoltaic system respectively.

[0032] Preferably, in step 3, the multi-objective optimization function is:

[0033]

[0034] Among them, f ESS is a multi-objective optimization function, For the cost of electricity generation; The investment and construction costs of the energy storage system; is the operation and maintenance cost; LOSS Cost of power loss reduction; f LS The capacity to absorb distributed generation;

[0035] The cost of electricity generation is:

[0036]

[0037] Among them, P Gi refers to the power generated by the i-th generator; ai ,b i ,c i They represent the power generation cost coefficient respectively;

[0038] S gen is a collection of generators;

[0039] The investment and construction cost of the energy storage system is:

[0040]

[0041] Among them, K E Refers to the investment cost per unit capacity; S E Indicates the rated capacity of the energy storage system; T E is the life cycle of the energy storage system;

[0042] The operation and maintenance costs are:

[0043]

[0044] Among them, K F ,K V Represent the operating cost per unit capacity and per unit charge and discharge respectively; S E Indicates the rated capacity of the energy storage system; T E is the life cycle of the energy storage system; T C is the charging and discharging time of the energy storage system; P C Indicates the actual charging and discharging power;

[0045] The cost of the reduced power loss is:

[0046]

[0047] Among them, P LOSS Represents the power loss before the energy storage system is connected; P LOSS,E Indicates the power loss after the energy storage system is connected; E L Indicates the unit electricity price;

[0048] The benefits of peak shaving and valley filling are:

[0049] f LS =|E L (ΔP S,E -ΔP)|

[0050] Among them, ΔP represents the peak-to-valley difference before the energy storage system is connected; ΔP S,E Represents the peak-to-valley difference after the energy storage system is connected; E L Indicates the unit electricity price.

[0051] Preferably, in step 3, the operating constraints include power balance constraints, voltage constraints, thermal stability constraints, charging state constraints, output power constraints and energy storage device quantity constraints;

[0052] The power balance constraint is:

[0053]

[0054] Among them, P Iinj and Q Iinj Respectively represent the active and reactive injection power of the node; P IDG and Q IDG are the active power and reactive power of the wind and solar power generation systems respectively; P GC and P GD are the power flows of the energy storage system in the charging and discharging states respectively; P ILOAD and Q ILOAD Respectively represent the active power and reactive power energy consumption of the load;

[0055] The voltage constraint is:

[0056] V min ≤V k,t ≤V max

[0057] Among them, V k,t represents the working voltage of node k at time t; V min ,V max Respectively represent the minimum and maximum voltage limit values;

[0058] The thermal stability constraint is:

[0059] I L,T ≤I R

[0060] Among them, I L,T ,I R Respectively represent the actual current and maximum current of L line;

[0061] The charge state constraints are:

[0062]

[0063] Among them, B soc,i (t) is the SOC value of the energy storage system; SOC min and SOC max They represent the upper and lower limits of the SOC of the energy storage system respectively;

[0064] B soc,i (t=1)=B soc,i (t=24)

[0065] Among them, Bsoc,i (t = 1) and B soc,i (t=24) represents the SOC value at the start and end time of the energy storage system respectively;

[0066] The output power constraint is:

[0067] -P E,out ≤P e ≤P E,in

[0068] Among them, P e Represents the active power output of the energy storage system; P E,out and P E,in They are respectively the limit values ​​of the active power output by the energy storage system;

[0069] The number of energy storage devices is constrained as follows:

[0070] N E ≤N E,max

[0071] Among them, N E and N E,max are the actual number and permitted number of new energy storage systems, respectively.

[0072] Preferably, the active and reactive power P of the system node Iinj and Q Iinj for:

[0073]

[0074] Among them, V k and V j are the voltage amplitudes of two different nodes k and j respectively; G kj is the conductance between nodes; θ kj is the phase difference between nodes; B kj is the inter-node susceptance;

[0075] Preferably, the calculation formula of the SOC value of the energy storage system is:

[0076]

[0077] Among them, E B,i Indicates rated capacity; B soc,i (t) and B soc,i (t-1) represents the SOC value of the energy storage system at different times; P i represents the charge and discharge power; Δt is the time interval.

[0078] Preferably, in step 4, the differential evolution algorithm includes: performing encoding and population initialization, setting N initial solutions, each solution x iis a D-dimensional vector, where D represents the dimension of the optimization variable;

[0079] Define the fitness function to measure the quality of each individual. The calculation formula is:

[0080]

[0081] Among them, w 1 ,w 2 ,…are the weights of each optimization objective;

[0082] Perform differential mutation to generate new candidate solutions:

[0083] V i =X r1 +F·(X r2 -X r3 )

[0084] Among them, V i is a candidate solution, X r1 ,X r2 ,X r3 are three different individuals randomly selected from the population; F is the scaling factor, which takes the value F∈[0.5,1];

[0085] Generate trial solutions via crossover:

[0086]

[0087] Among them, U i is the experimental solution; CR is the crossover probability, j rand is a randomly selected index, ensuring that at least one dimension changes;

[0088] Solve the test i and the current solution x i Compare and retain individuals with higher fitness:

[0089]

[0090] Determine the convergence conditions, which include: the optimal value of the fitness function remains unchanged in multiple iterations; the maximum number of iterations k is reached max ;

[0091] If one of the above conditions is met, the iteration stops and the optimal solution is output; otherwise, the iteration continues.

[0092] The optimal solution output is the optimal location, sizing and scheduling strategy for the energy storage system.

[0093] The second aspect of the present invention provides a resource scheduling optimization system, which adopts the above-mentioned energy storage resource scheduling optimization method based on digital twins, including

[0094] Platform construction module, equivalent model building module, optimization problem construction module and problem solving and dynamic adjustment module;

[0095] Among them, the platform construction module is used to build a digital twin-based energy storage resource scheduling optimization platform;

[0096] The equivalent model building module is used to model wind power generation, photovoltaic systems and energy storage systems, and equate them to different nodes in the power grid;

[0097] The optimization problem building module is used to set the objective function for optimizing the configuration of the energy storage system;

[0098] The problem solving and dynamic adjustment module is used to solve optimization problems using differential evolution algorithm.

[0099] The beneficial effect of the present invention is that, compared with the prior art, the present invention has significant advantages over the prior art by combining digital twin technology with an optimization scheduling algorithm. The present invention combines digital twin technology to realize real-time interaction between virtual and physical systems through a digital twin model, accurately reflects the operating status of the energy storage system, and predicts future system change trends, thereby improving real-time response capabilities. In addition, multi-objective optimization is achieved, and a multi-objective optimization function that comprehensively considers power generation costs, energy storage system investment and construction costs, operation and maintenance costs, power losses, and peak-to-valley differences is established, significantly improving the economy and stability of the scheduling scheme. The energy storage resource scheduling optimization platform established by the present invention also has dynamic adjustment capabilities and introduces a differential evolution algorithm. The differential evolution algorithm has more global search capabilities and population diversity through differential mutation and crossover operations, and can show better robustness in complex nonlinear constraints and mixed variable optimization problems. Differential evolution is more sensitive to environmental changes and can quickly generate new adaptive solutions. The safety and feasibility of the scheduling strategy are also ensured through a variety of constraints (such as power balance, voltage constraints, charging state constraints, etc.).

[0100] First of all, the present invention is based on the real-time interaction between the digital twin model and the physical energy storage system, which can accurately reflect the operating status of the energy storage system, and combine the dynamic load demand and distributed power generation output to achieve high-precision resource scheduling optimization, effectively improving the system's real-time response capability and regulation accuracy. By introducing digital twin technology, real-time dynamic capture of the energy storage system status and power grid environment is achieved. The digital twin combines physical models, real-time measurement data and historical information to build a virtual mirror of the power grid, allowing the platform to monitor the charging and discharging behavior of the energy storage system and the fluctuating output of distributed energy, and perform dynamic optimization based on real-time data.

[0101] Secondly, the present invention adopts sequential power flow calculation and multi-objective optimization algorithm, comprehensively considers system operation cost, energy storage efficiency and peak-shaving and valley-filling benefits, and realizes optimal configuration of energy storage resources through iterative solution to maximize economic benefits. In view of the volatility and uncertainty in complex power grid environment, the present invention supports dynamic parameter adjustment and scheduling strategy optimization, significantly enhances the robustness and adaptability of the system, and ensures the stable operation of the energy storage system under variable working conditions.

[0102] In addition, the present invention can realize the refined scheduling of energy storage resources, balance the supply and demand relationship of the power grid, reduce power loss, improve the absorption capacity of distributed energy, and further support the deep integration of smart grid and renewable energy. Overall, the present invention effectively improves the operating efficiency, stability and flexibility of the energy storage system, and provides an efficient and intelligent energy storage resource scheduling solution for smart grids. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] Figure 1 It is a schematic diagram of the composition of the resource optimization platform of the present invention;

[0104] Figure 2 is an equivalent model diagram of the combined power generation system of the present invention;

[0105] Figure 3 It is the overall flow chart of the present invention;

[0106] Figure 4 The figure is a flow chart of the solution process of the optimization method of the present invention. DETAILED DESCRIPTION

[0107] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only embodiments of a part of the present invention, rather than all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the protection scope of the present invention.

[0108] like Figure 1-4 As shown, an embodiment of the present invention provides a method for optimizing energy storage resource scheduling based on digital twins, comprising the following steps:

[0109] Step 1: Build a digital twin-based energy storage resource scheduling optimization platform; combine the physical characteristics and historical data of the energy storage resource scheduling optimization platform to establish a digital twin model that describes the operating status of the energy storage equipment. Figure 1 As shown in the figure, the energy storage resource scheduling optimization platform based on digital twin mainly includes three parts: measurement system, modeling and simulation system and operation optimization system.

[0110] The measurement system is responsible for collecting, storing and communicating data from various sources in the energy storage system, including historical, real-time and simulated data. The measurement system is used to provide the basic data required for energy storage system modeling and optimization.

[0111] The digital twin model of the distributed power grid, energy storage system (ESS) and power grid is implemented in the modeling and simulation system. A virtual image is established, and the real-time interaction, monitoring and control of the operating status of the system between the virtual space and the physical entity are realized through physical mechanisms, data-driven and the integration of the two. The modeling and simulation system is used to simulate the dynamic operation of the system to provide support for the site selection and capacity determination of the energy storage system.

[0112] The operation optimization system uses simulation results and real-time data to optimize the scheduling of energy storage systems. In this system, the time-series power flow simulation, location and capacity determination, and autonomous regulation of energy storage systems are completed. The operation optimization system provides technical support for the control and planning of the power grid to ensure the economy and stability of system operation.

[0113] Step 2: Based on the digital twin model of the scheduling optimization platform constructed in step 1, wind power generation, photovoltaic system and energy storage system are jointly modeled to establish an equivalent model of the joint power generation system and simplify the power flow calculation of the system. The following describes the modeling and simulation system in the optimization platform in step 1. In this joint power generation system, wind power generation, photovoltaic system and energy storage system are modeled and equivalent to different nodes in the power grid for power flow analysis.

[0114] First, for wind turbines, the PQ node modeling method is adopted, that is, the active power and reactive power output by the wind turbine are used as known variables. A mathematical model is established by describing the relationship between the power output of the wind turbine and system voltage, leakage reactance and other parameters through formulas. The power in the wind turbine can be calculated as:

[0115]

[0116] Among them, Q W is the reactive power output by the wind turbine; U and P W are the voltage and active power of the wind turbine generator respectively. 1 and x 2 Respectively represent the stator leakage reactance and rotor leakage reactance of the wind turbine. m is the magnetizing reactance of the wind turbine.

[0117] Next, the photovoltaic system is mathematically modeled. The photovoltaic system adjusts the output voltage through the voltage source control inverter, so it can be modeled as a node that outputs reactive power. In the power flow, the reactive power output of the photovoltaic system can be calculated as:

[0118]

[0119] Among them, Q P is the reactive power output by the photovoltaic system; P P is the active power output by the photovoltaic system, I is the effective value of the output current of the photovoltaic system, e is the output voltage factor of the photovoltaic system inverter control module, and f is the reactive power adjustment factor of the photovoltaic system, which is used to reflect the adjustment characteristics of the reactive power of the photovoltaic system.

[0120] Based on the above analysis, wind power generation and photovoltaic systems constitute a wind-solar power generation system. The overall active power and reactive power of the wind-solar power generation system can be expressed as:

[0121] P IDG =P W +P P

[0122] Q IDG =Q W +Q P

[0123] Among them, P IDG and Q IDG are the active power and reactive power of the wind and solar power generation systems respectively; P W and Q W are the active power and reactive power of the wind turbine generator respectively; P P and Q P They are the active power and reactive power output by the photovoltaic system respectively.

[0124] When charging and discharging, the energy storage system behaves as a load model, absorbing or releasing active power and reactive power.

[0125] The injection power formula of the energy storage system describes the charging P GC and discharge P GD The power flow calculation under the state. It includes the active power P e and reactive power Q e The two components are important indicators that describe the impact of energy storage systems on the power grid. Assuming i represents an imaginary unit, the injected power of the energy storage system can be expressed as:

[0126]

[0127] Among them, P GC and P GD are the power flows of the energy storage system in charging and discharging states respectively; P e and Q e They are the active power and reactive power output by the energy storage system respectively.

[0128] Step 3: Based on the equivalent model of the combined power generation system established in step 2, the energy storage system resource scheduling optimization problem is constructed. The load demand and distributed energy output data predicted by the digital twin model are used to establish the physical and operational constraints of the multi-objective optimization function and the scheduling algorithm; by optimizing the site selection and sizing of the energy storage system, the operating cost can be reduced and the utilization efficiency of distributed energy can be improved.

[0129] Specifically, the objective function of optimizing the configuration of the energy storage system is set. The optimization goal is to minimize the total operating cost during the entire life cycle of the energy storage system, which includes power generation cost, investment and construction cost, operation and maintenance cost, power loss reduction cost and peak-shaving and valley-filling benefits. And achieve higher economic benefits and system stability through optimization.

[0130] In a preferred but non-limiting embodiment of the present invention, the optimization objectives mainly include: power generation cost, investment and construction cost of energy storage system, operation and maintenance cost, power loss reduction cost and peak load shifting benefit. The detailed calculation process is as follows:

[0131] The cost of electricity generation is calculated using the following formula:

[0132]

[0133] in, For the cost of electricity generation; refers to the power generation of the i-th generator. The generator here refers to the traditional generator used for auxiliary power generation, such as thermal power generation, gas turbine, etc.; a i ,b i ,c i Respectively represent the power generation cost coefficient; S gen is a collection of generators.

[0134] The investment and construction cost of the energy storage system is proportional to its capacity, and the calculation formula is:

[0135]

[0136] in, is the investment and construction cost of the energy storage system; K E Refers to the investment cost per unit capacity; S E Indicates the rated capacity of the energy storage system; T E It is the life cycle of the energy storage system.

[0137] The operation and maintenance cost is related to the actual output power and capacity of the energy storage system, and the calculation formula is:

[0138]

[0139] in, K is the operation and maintenance cost; F ,K V Respectively represent the operating cost per unit capacity and per unit charge and discharge; T C is the charging and discharging time of the energy storage system; P C Indicates the actual charging and discharging power of the energy storage system.

[0140] Optimizing the configuration can reduce the power loss of the system. The power loss reduction cost calculation formula is:

[0141]

[0142] Among them, f LOSS The cost of reducing power loss; P LOSS Represents the power loss before the energy storage system is connected; P LOSS,E Indicates the power loss after the energy storage system is connected; E L Indicates the unit electricity price.

[0143] The energy storage system achieves peak-shaving and valley-filling through charging and discharging, improving the absorption capacity of distributed power generation. The peak-shaving and valley-filling benefit calculation formula is:

[0144] f LS =|E L (ΔP S,E -ΔP)|, (7)

[0145] Among them, f LS is the absorption capacity of distributed generation; ΔP represents the peak-to-valley difference before the energy storage system is connected; ΔP S,E Indicates the peak-to-valley difference after the energy storage system is connected.

[0146] In summary, the multi-objective optimization function can be expressed as

[0147]

[0148] Among them, f ESS is a multi-objective optimization function, For the cost of electricity generation; The investment and construction costs of the energy storage system; is the operation and maintenance cost; LOSS Cost of power loss reduction; f LS The capacity to absorb distributed power generation.

[0149] Furthermore, in order to ensure the safe and stable operation of the energy storage system in the actual power grid, the scheduling optimization problem needs to meet the following constraints: power balance constraint, voltage constraint, thermal stability constraint, state of charge (SOC) constraint, output power constraint and energy storage device quantity constraint.

[0150] After the energy storage system is connected, the power balance constraint means that the active and reactive power of the system nodes must satisfy the power balance constraint relationship:

[0151]

[0152] Among them, P Iinj and Q Iinj Respectively represent the active and reactive injection power of the node; P IDG and Q IDG are the active power and reactive power of the wind and solar power generation systems respectively; P GC and P GD are the power flows of the energy storage system in the charging and discharging states respectively; P ILOAD and Q ILOAD Respectively represent the active power and reactive power energy consumption of the load;

[0153] Active and reactive power P of system nodes Iinj and Q Iinj It can be expressed as:

[0154]

[0155] Among them, V k and V j are the voltage amplitudes of two different nodes k and j respectively; G kj is the conductance between nodes; θ kj is the phase difference between nodes; B kj is the inter-node susceptance.

[0156] Voltage constraint means that the system node voltage must be kept within the allowable range, that is,

[0157] V min ≤V k,t ≤V max , where V k,t Represents the working voltage of node k at time t. V min ,V max Represents the minimum and maximum voltage limits respectively.

[0158] Thermal stability constraint means that the current of the grid line should not exceed the maximum allowable value, that is, I L,T ≤I R , where I L,T ,I R Represent the actual current and maximum current of L line respectively.

[0159] The state of charge constraint refers to the need to limit the system's state of charge to avoid overcharging. The SOC state is related to the previous time and can be expressed as

[0160]

[0161] Among them, B soc,i (t) is the SOC value of the energy storage system; E B,i Indicates rated capacity; B soc,i (t) and B soc,i (t-1) represents the SOC value of the energy storage system at different times; P i represents the charge and discharge power; Δt is the time interval.

[0162] The SOC of the energy storage system needs to be kept within a safe range, that is,

[0163]

[0164] Among them, B soc,i (t) is the SOC value of the energy storage system; SOC min and SOC max They respectively represent the upper and lower limits of the SOC of the energy storage system.

[0165] In addition, the initial and final values ​​of SOC must be equal to ensure the daily scheduling closed loop of the system. This constraint can be expressed as

[0166] B soc,i (t=1)=B soc,i (t=24). (13)

[0167] Among them, B soc,i (t = 1) and B soc,i (t=24) represents the SOC value of the energy storage system at the start and end time respectively.

[0168] The output power constraint means that the energy storage system can only be in the charging and discharging state, and the output power cannot exceed the rated power. The constraint equation can be expressed as:

[0169] -P E,out ≤P e ≤P E,in (14)

[0170] Among them, P e Represents the active power output of the energy storage system; P E,out and P E,in They are respectively the limit values ​​of the active power output by the energy storage system.

[0171] The energy storage equipment quantity constraint means that the number of newly built energy storage systems cannot exceed the maximum allowed value, that is,

[0172] N E ≤N E,max (15)

[0173] Among them, N E and NE,max are the actual number and permitted number of new energy storage systems, respectively.

[0174] In summary, the energy storage resource scheduling optimization based on digital twins can be expressed as:

[0175]

[0176] Step 4, problem solving and dynamic adjustment are as follows: Differential Evolution (DE) algorithm is used to solve the above resource optimization problem. The following is a detailed introduction to the algorithm.

[0177] First, encode and initialize the population, and encode the configuration variables of the energy storage system with real numbers. Set N initial solutions, each solution x i is a D-dimensional vector, where D represents the dimension of the optimization variable. Among them, are the lower and upper bounds of the j-th variable, respectively.

[0178] Next, we define the fitness function to measure the quality of each individual. Assume that w 1 ,w 2 ,...is the weight of each optimization objective, which is used to achieve weighted solution of multiple objectives. The calculation formula is:

[0179]

[0180] Perform differential mutation to generate new candidate solutions V i :

[0181] V i =X r1 +F·(X r2 -X r3 ) (18)

[0182] Among them, X r1 ,X r2 ,X r3 are three different individuals randomly selected from the population; F is a scaling factor, usually F∈[0.5,1], which is used to adjust the mutation intensity.

[0183] Solve U by cross generation test i :

[0184]

[0185] Where CR is the crossover probability, j rand are randomly selected indices that guarantee that at least one dimension changes.

[0186] Solve the test i and the current solution x i Compare and retain individuals with higher fitness:

[0187]

[0188] Determine the convergence conditions. Check whether one of the following convergence conditions is met: the optimal value of the fitness function remains unchanged in multiple iterations; the maximum number of iterations k is reached max If satisfied, stop iteration and output the optimal solution; otherwise return and continue iteration. Finally, output the optimal solution, that is, the optimal location, capacity and scheduling strategy of the energy storage system.

[0189] Based on the above statements, the overall approach of smart grid energy storage fault monitoring method based on digital twin is given below.

[0190] Table 1 Energy storage resource scheduling optimization method based on digital twin

[0191]

[0192] The embodiment of the present invention further provides a resource scheduling optimization system based on the energy storage resource scheduling optimization method of the digital twin, including:

[0193] Platform construction module, equivalent model building module, optimization problem construction module and problem solving and dynamic adjustment module;

[0194] Among them, the platform construction module is used to build a digital twin-based energy storage resource scheduling optimization platform;

[0195] The equivalent model building module is used to model wind power generation, photovoltaic systems and energy storage systems, and equate them to different nodes in the power grid;

[0196] The optimization problem building module is used to set the objective function for optimizing the configuration of the energy storage system;

[0197] The problem solving and dynamic adjustment module is used to solve optimization problems using differential evolution algorithm.

[0198] The present invention realizes real-time dynamic capture of the energy storage system status and power grid environment by introducing digital twin technology. Digital twin combines physical models, real-time measurement data and historical information to build a virtual image of the power grid, allowing the platform to monitor the charging and discharging behavior of the energy storage system and the fluctuating output of distributed energy, and dynamically optimize based on real-time data. At the same time, the synergy of the measurement system and high-performance simulation technology provides technical support for the control and planning of power grid operation, enabling the energy storage system to effectively respond to environmental changes and load demands, and improve the stability and economy of system operation.

[0199] This paper proposes an effective strategy to deal with the volatility of wind and photovoltaic power generation by combining digital twin technology and optimized configuration of energy storage system. Digital twin technology dynamically captures the impact of wind speed and light changes on distributed energy output. The energy storage system achieves charge and discharge balance by peak shaving and valley filling, smoothing power fluctuations and improving energy utilization.

[0200] The energy storage system charges when there is excess power generation and discharges when there is insufficient power generation, smoothing out the fluctuations in distributed energy output. The dynamic scheduling of the energy storage system is achieved through time-series power flow simulation and optimization.

[0201] The energy storage system reduces load demand during peak periods by shaving peak loads and filling valleys, balancing power generation and consumption and improving energy utilization.

[0202] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0203] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0204] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0205] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for optimizing energy storage resource scheduling based on digital twins, characterized in that: include: Step 1: Build a digital twin-based energy storage resource scheduling optimization platform; Combine the physical characteristics and historical data of the energy storage resource scheduling optimization platform to establish a digital twin model that describes the operating status of energy storage equipment; Step 2: Based on the digital twin model of the scheduling optimization platform constructed in step 1, jointly model the wind power generation, photovoltaic system and energy storage system to establish an equivalent model of the joint power generation system; Step 3: Based on the equivalent model of the combined power generation system established in step 2, physical and operational constraints of the multi-objective optimization function and the scheduling algorithm are established to construct the energy storage system resource scheduling optimization problem; Step 4: Based on the scheduling optimization problem established in step 3, the differential evolution algorithm is used to solve the problem and dynamically adjust it to achieve the optimal energy storage resource scheduling.

2. The energy storage resource scheduling optimization method based on digital twin according to claim 1 is characterized in that: In step 1, the energy storage resource scheduling optimization platform based on digital twins includes: measurement system, modeling and simulation system, and operation optimization system; The measurement system is used to provide the basic data required for energy storage system modeling and optimization; The modeling and simulation system is used to simulate the dynamic operation of the system to provide support for the site selection and sizing of the energy storage system; The operation optimization system is used to optimize the scheduling of the energy storage system using simulation results and real-time data.

3. The energy storage resource scheduling optimization method based on digital twin according to claim 1 is characterized in that: In step 2, wind power generation is modeled using the PQ node, and the reactive power output by the wind turbine is: Among them, Q W is the reactive power output by the wind turbine; U and P W are the voltage and active power of the wind turbine generator respectively; x1 and x2 represent the stator leakage reactance and rotor leakage reactance of the wind turbine generator respectively; x m is the excitation reactance of the wind turbine; The photovoltaic system adjusts the output voltage through the voltage source control inverter. The reactive power output by the photovoltaic system is: Among them, Q P is the reactive power output by the photovoltaic system; P P is the active power output by the photovoltaic system, I is the effective value of the output current of the photovoltaic system, e is the output voltage factor of the inverter control module, and f is the reactive power adjustment factor of the photovoltaic system; The energy storage system describes its impact on the grid by injecting power. The injection power formula is: P GC =-P e -i*Q e P GD =P e +i*Q e Among them, P GC and P GD are the power flows of the energy storage system in charging and discharging states respectively; P e and Q e They are the active power and reactive power output by the energy storage system respectively.

4. The energy storage resource scheduling optimization method based on digital twin according to claim 3 is characterized in that: In step 2, the wind power generation and photovoltaic system form a wind-solar power generation system. The active power and reactive power of the wind-solar power generation system can be expressed as: P IDG =P W +P P Q IDG =Q W +Q P Among them, P IDG and Q IDG are the active power and reactive power of the wind and solar power generation systems respectively; P W and Q W are the active power and reactive power of the wind turbine generator respectively; P P and Q P They are the active power and reactive power output by the photovoltaic system respectively.

5. The energy storage resource scheduling optimization method based on digital twin according to claim 4 is characterized in that: In step 3, the multi-objective optimization function is: Among them, f ESS is a multi-objective optimization function, For the cost of electricity generation; The investment and construction costs of the energy storage system; is the operation and maintenance cost; LOSS The cost of reducing power loss; f LS The capacity to absorb distributed generation; The cost of electricity generation is: Among them, P Gi refers to the power generated by the i-th generator; a i ,b i ,c i Respectively represent the power generation cost coefficient; S gen is a collection of generators; The investment and construction cost of the energy storage system is: Among them, K E Refers to the investment cost per unit capacity; S E Indicates the rated capacity of the energy storage system; T E is the life cycle of the energy storage system; The operation and maintenance costs are: Among them, K F ,K V Represent the operating cost per unit capacity and per unit charge and discharge respectively; S E Indicates the rated capacity of the energy storage system; T E is the life cycle of the energy storage system; T C is the charging and discharging time of the energy storage system; P C Indicates the actual charging and discharging power; The cost of the reduced power loss is: Among them, P LOSS Represents the power loss before the energy storage system is connected; P LOSS,E Indicates the power loss after the energy storage system is connected; E L Indicates the unit electricity price; The benefits of peak shaving and valley filling are: f LS =|E L (ΔP S,E -ΔP)| Among them, ΔP represents the peak-to-valley difference before the energy storage system is connected; ΔP S,E Represents the peak-to-valley difference after the energy storage system is connected; E L Indicates the unit electricity price.

6. The energy storage resource scheduling optimization method based on digital twin according to claim 5 is characterized in that: In step 3, the operating constraints include power balance constraints, voltage constraints, thermal stability constraints, state of charge constraints, output power constraints, and energy storage device quantity constraints; The power balance constraint is: Among them, P Iinj and Q Iinj Respectively represent the active and reactive injection power of the node; P IDG and Q IDG are the active power and reactive power of the wind and solar power generation systems respectively; P GC and P GD are the power flows of the energy storage system in the charging and discharging states respectively; P ILOAD and Q ILOAD Respectively represent the active power and reactive power energy consumption of the load; The voltage constraint is: In min ≤V k,t ≤V max Among them, V k,t represents the working voltage of node k at time t; V min ,V max Respectively represent the minimum and maximum voltage limit values; The thermal stability constraint is: I L,T ≤I R Among them, I L,T ,I R Respectively represent the actual current and maximum current of L line; The charge state constraints are: Among them, B soc,i (t) is the SOC value of the energy storage system; SOC min and SOC max They represent the upper and lower limits of the SOC of the energy storage system respectively; B soc,i (t=1)=B soc,i (t=24) Among them, B soc,i (t = 1) and B soc,i (t=24) represents the SOC value at the start and end time of the energy storage system respectively; The output power constraint is: -P E,out ≤P e ≤P E,in Among them, P e Represents the active power output of the energy storage system; P E,out and P E,in They are respectively the limit values ​​of the active power output by the energy storage system; The number of energy storage devices is constrained as follows: N E ≤N E,max Among them, N E and N E,max are the actual number and permitted number of new energy storage systems, respectively.

7. The energy storage resource scheduling optimization method based on digital twin according to claim 6 is characterized in that: Active and reactive power P of the system node Iinj and Q Iinj for: Among them, V k and V j are the voltage amplitudes of two different nodes k and j respectively; G kj is the conductance between nodes; θ kj is the phase difference between nodes; B kj is the inter-node susceptance.

8. The energy storage resource scheduling optimization method based on digital twin according to claim 6 is characterized in that: The calculation formula of the SOC value of the energy storage system is: Among them, E B,i Indicates rated capacity; B soc,i (t) and B soc,i (t-1) represents the SOC value of the energy storage system at different times; P i represents the charge and discharge power; Δt is the time interval.

9. The energy storage resource scheduling optimization method based on digital twin according to claim 5 is characterized in that: In step 4, the differential evolution algorithm includes: encoding and population initialization, setting N initial solutions, each solution x i is a D-dimensional vector, where D represents the dimension of the optimization variable; Define the fitness function to measure the quality of each individual. The calculation formula is: Among them, w1, w2, ... are the weights of each optimization objective; Perform differential mutation to generate new candidate solutions: V i =X r1 +F·(X r2 -X r3 ) Among them, V i is a candidate solution, X r1 ,X r2 ,X r3 are three different individuals randomly selected from the population; F is the scaling factor, which takes the value F∈[0.5,1]; Generate trial solutions via crossover: Among them, U i is the experimental solution; CR is the crossover probability, j rand is a randomly selected index, ensuring that at least one dimension changes; Solve the test i and the current solution x i Compare and retain individuals with higher fitness: Determine the convergence conditions, which include: the optimal value of the fitness function remains unchanged in multiple iterations; the maximum number of iterations k is reached max ; If one of the above conditions is met, the iteration stops and the optimal solution is output; otherwise, the iteration continues. The optimal solution output is the optimal location, sizing and scheduling strategy for the energy storage system.

10. A resource scheduling optimization system, using the energy storage resource scheduling optimization method based on digital twins according to any one of claims 1 to 9, characterized in that: include: Platform construction module, equivalent model building module, optimization problem construction module and problem solving and dynamic adjustment module; Among them, the platform construction module is used to build a digital twin-based energy storage resource scheduling optimization platform; The equivalent model building module is used to model wind power generation, photovoltaic systems and energy storage systems, and equate them to different nodes in the power grid; The optimization problem building module is used to set the objective function for optimizing the configuration of the energy storage system; The problem solving and dynamic adjustment module is used to solve optimization problems using differential evolution algorithm.

Images