Electrochemical energy storage capacity optimization method considering wide-load operation of thermal power generating unit

Abstract

The invention discloses an electrochemical energy storage capacity optimization method considering wide-load operation of a thermal power generating unit. The method comprises the following steps: aiming at the optimization requirements of a thermal power generating unit and an electrochemical energy storage cooperative system, constructing a main target and sub-target coupled hierarchical target function; the main target is to maximize the benefit of the power grid side and preferentially guarantee the safe and economical operation of the power grid, and the benefit of the power grid side comprises the frequency modulation service benefit, the network loss cost and the reserve capacity cost considering the wide-load operation of the thermal power generating unit; the sub-targets are power plant side income maximization and energy storage side life loss rate minimization, and the participation enthusiasm of the power plant and the energy storage configuration economy are ensured; cooperative system constraint setting; a multi-objective improved vulture optimization algorithm is adopted as a core solving tool of a hierarchical objective function. According to the method, the limitation of a single target is broken through based on a hierarchical target function, and multi-target collaborative optimization is realized; the dynamic characteristics of the target function are matched, and local optimum caused by multi-peak characteristics of the target function is avoided.

Description

Technical Field

[0001] This invention belongs to the field of electrochemical energy storage technology, specifically a method for optimizing electrochemical energy storage capacity considering the wide-load operation of thermal power units. Background Technology

[0002] With the large-scale integration of new energy sources into the power grid, the uncertainty and volatility of power grid operation have significantly increased. As a flexible adjustment resource, the capacity configuration of electrochemical energy storage is of great significance to the safe, stable, and economical operation of the power grid. Current electrochemical energy storage capacity configuration optimization technologies have significant shortcomings: On the one hand, existing solutions mostly design objective functions from the power plant's perspective, focusing on economic indicators such as payback period and internal rate of return, neglecting the core benefits for system operators. They fail to incorporate system stability and source-grid synergy into core optimization objectives and lack a multi-benefit synergy framework, easily leading to configuration schemes that cannot meet the needs of power grid security and absorption. On the other hand, existing methods use static weights for multi-objective optimization, making it difficult to adapt to dynamic operating environments such as power grid load fluctuations, changes in new energy output, and unit operating condition switching.

[0003] Therefore, there is an urgent need for an electrochemical energy storage capacity configuration optimization method that comprehensively considers system security, power source-grid coordination, and peak-valley difference regulation, has dynamic scenario adaptability (matching load and new energy fluctuations), and can achieve multi-objective coordination among the power grid, power plant, and energy storage. This method aims to solve the problems of perspective bias, static rigidity, and insufficient practicality of existing technologies, and improve the support effect of energy storage configuration on power grid operation. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides an electrochemical energy storage capacity configuration optimization method considering the wide-load operation of thermal power units. It constructs a hierarchical objective function with "main objective + sub-objective coupling", setting the grid-side revenue as the main objective and maximizing the power plant-side revenue and minimizing the energy storage-side lifetime loss rate as sub-objectives, thus breaking through the limitations of single objectives and achieving multi-objective collaborative optimization; and matching the dynamic characteristics of the objective function to avoid getting trapped in local optima due to the multi-peak characteristics of the objective function.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an electrochemical energy storage capacity optimization method considering wide-load operation of thermal power units, comprising:

[0006] Hierarchical Objective Function Construction: To address the optimization needs of the collaborative system of thermal power units and electrochemical energy storage, a hierarchical objective function with coupled main and sub-objectives is constructed. The main objective is to maximize grid-side revenue, prioritizing the safe and economical operation of the grid. Grid-side revenue includes frequency regulation service revenue considering the wide-load operation of thermal power units, grid loss costs, and reserve capacity costs. The sub-objectives are to maximize power plant-side revenue and minimize the lifespan loss rate of energy storage, ensuring the incentive for power plant participation and the economic efficiency of energy storage configuration.

[0007] Cooperative system constraint setting: Two types of cooperative constraints are set: thermal power unit operation constraints and electrochemical energy storage operation constraints;

[0008] Optimization solution: The multi-objective improved vulture optimization algorithm is used as the core solution tool for the hierarchical objective function to obtain the optimal electrochemical energy storage capacity configuration scheme.

[0009] Furthermore, the electrochemical energy storage capacity optimization method also includes: data acquisition and preprocessing: acquiring technical parameters, operating status quantities and economic indicators of thermal power units, rated capacity, rated voltage, charge and discharge efficiency, life decay characteristics and full life cycle cost of electrochemical energy storage equipment, and power grid operation data.

[0010] Furthermore, the formula for calculating the grid-side revenue is as follows:

[0011] F main =C p +C grid,spare -C loss

[0012] In the formula, F main For grid-side revenue, C p For the revenue from power grid frequency regulation services, C grid,spare For annual standby capacity costs, C loss Cost of network loss.

[0013] Furthermore, when the thermal power units are operating under normal load conditions, the revenue from the first frequency regulation service of the power grid is calculated as follows:

[0014]

[0015] When thermal power units operate under wide load conditions, the revenue from the second frequency regulation service of the power grid is calculated as follows:

[0016]

[0017] Among them, c by and c tk λ and ε represent the unit prices of the spinning reserve and regulation capacity of thermal power units, respectively, in yuan / (MW·h); λ and ε represent the contributions of energy storage frequency regulation to the spinning reserve and regulation capacity of thermal power units, respectively; c pf and c pm The compensation unit prices for reducing the maximum frequency deviation of the power system and improving the frequency recovery speed of the power system are respectively, c pf The unit is yuan / (MW·h·Hz), c pm The unit is yuan / (MW·h·(Hz / s)); Δf mi and Δν siThese represent the maximum decrease in frequency deviation and the increase in frequency recovery speed within the i-th cycle after energy storage-assisted frequency regulation, respectively; P Gn (t) represents the output of the thermal power unit under normal load, in MW; This indicates the lower limit of the output of a thermal power unit under normal load operation; P represents the maximum allowable output of a thermal power unit. Gw (t) represents the output of a thermal power unit under wide load conditions, in MW; This indicates the lower limit of output capacity for wide-load operation;

[0018] Power Grid Frequency Regulation Service Revenue C p For: C p =C p1 +C p2 .

[0019] Furthermore, the network loss cost is calculated as follows:

[0020] C loss =C e,t ·∫ΔP loss (t)dt

[0021] Among them, C loss Cost of network loss, yuan; C e,t The real-time grid connection price is yuan / kWh; ΔP loss (t) represents the line loss power, in kW;

[0022] The formula for line loss power is: ΔP loss (t)=I line (t) 2 ·R line ·L line ·10 -6

[0023] Among them, I line (t) represents the line current at time t, in A; R line Resistance per unit length of the line, in ohms / km; L line The length of the line is in km.

[0024] Furthermore, the formula for calculating the annual standby capacity cost is as follows:

[0025] C grid,spare =(S spare0 -S spareopt )·T year ·C spare ·10 -4

[0026] Among them, C grid,spare For annual standby capacity costs, S spare0 This represents the reserve capacity when no energy storage is available; S spareoptThis refers to the grid's reserve capacity when energy storage is available; T year =Annual duration; C spare Cost per unit of standby capacity.

[0027] Furthermore, the formula for maximizing the power plant's revenue is as follows:

[0028] F sub1 =max(R) G -C h -C inv )

[0029] In the formula, F sub1 To maximize power plant-side revenue, R G To increase the power generation revenue of thermal power units, C h For the coal consumption cost of thermal power units, C inv For energy storage investment costs;

[0030] Energy storage investment costs are divided into operation and maintenance costs C1 and initial investment costs C2.

[0031] C inv =C1+C2

[0032] C1 = C E E ess +C P P BN

[0033] C2=k E E ess +k P P BN

[0034] In the formula, C E C represents the unit price per unit capacity of the energy storage system. P E represents the unit price per unit power of the energy storage system. ess P is the rated capacity of the energy storage system. BN K represents the rated power of the energy storage system. E K represents the unit capacity operation and maintenance cost of an energy storage system. P This refers to the operation and maintenance cost per unit power of the energy storage system.

[0035] Furthermore, the formula for minimizing the lifetime loss rate of the energy storage side is as follows:

[0036]

[0037] Among them, F sub2 To minimize the energy storage side lifetime loss rate, η ess,loss,annual Δη is the annual energy storage lifespan loss rate. ess,loss T represents the energy storage lifetime loss per unit time step; essThe rated cycle life of the energy storage is given by Δt, which represents the time step interval.

[0038] Furthermore, the operating constraints of the thermal power units include constraints on the output range of the thermal power units, the ramp rate, the coal consumption rate, and the number of start-stop cycles; the operating constraints of the electrochemical energy storage include constraints on capacity, charge and discharge power, state of charge, and the SOC level of the energy storage system.

[0039] Furthermore, when the deviation between the main objective and sub-objectives in the objective function is large, the global exploration probability of the algorithm is increased, and the search range is expanded through a spiral flight strategy; when the objective function value tends to be stable, the algorithm switches to local development mode, focusing on refining the area around the optimal solution, and avoiding getting trapped in local optima due to the multi-peak characteristics of the objective function.

[0040] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0041] This invention abandons the one-sidedness of traditional single-objective optimization (such as pursuing only grid revenue or power plant revenue), and constructs a hierarchical architecture of "main objective (grid-side revenue) + sub-objectives (power plant-side revenue, energy storage lifespan loss)". The main objective prioritizes ensuring grid safety and economy, while the sub-objectives quantify power plant revenue and energy storage loss respectively, achieving a multi-objective balance of "grid safety - power plant profitability - equipment durability", avoiding the hidden dangers caused by single-objective optimization.

[0042] This invention addresses the multi-objective balance requirement of the objective function "grid-power plant-energy storage" by upgrading the traditional single search strategy of IVOA (Inertial Weighted Optimization Algorithm) to a dual-mode "global exploration + local development". When the deviation between the main objective and sub-objectives in the objective function is large, the probability of global exploration of the algorithm is increased, and the search range is expanded by using a spiral flight strategy. When the objective function value tends to be stable, the local development mode is switched to focus on refining the periphery of the optimal solution, so as to avoid getting trapped in local optima due to the multi-peak characteristics of the objective function. Attached Figure Description

[0043] Figure 1 This is a schematic flowchart of an electrochemical energy storage capacity optimization method for thermal power units under wide load operation, according to the present invention. Detailed Implementation

[0044] The present invention will now be described in further detail with reference to the embodiments and accompanying drawings.

[0045] This invention provides a method for optimizing electrochemical energy storage capacity considering wide-load operation of thermal power units, such as... Figure 1 As shown, it includes the following steps:

[0046] Data acquisition and preprocessing: Acquire technical parameters, operating status parameters and economic indicators of thermal power units, rated capacity, rated voltage, charge and discharge efficiency, life decay characteristics and total life cycle cost of electrochemical energy storage equipment, and power grid operation data;

[0047] Hierarchical Objective Function Construction: To address the optimization requirements of synergistic operation between thermal power units and electrochemical energy storage, a hierarchical objective function with coupled main and sub-objectives is constructed. The main objective is to maximize grid-side revenue, prioritizing the safe and economical operation of the grid. Grid-side revenue includes frequency regulation service revenue considering the wide-load operation of thermal power units, grid loss costs, and reserve capacity costs. The sub-objectives are to maximize power plant-side revenue and minimize the lifespan loss rate of energy storage, ensuring the enthusiasm of power plants to participate and the economic efficiency of energy storage configuration.

[0048] Coordinated system constraint settings: Set operating constraints for thermal power units and electrochemical energy storage;

[0049] Optimization solution: The multi-objective improved vulture optimization algorithm is used as the core solution tool for the hierarchical objective function to obtain the optimal electrochemical energy storage capacity configuration scheme.

[0050] The following is a detailed explanation of each step in the above method.

[0051] 1. Data Acquisition and Preprocessing

[0052] 1.1 Data Acquisition

[0053] Data is collected on the technical parameters, operating status, and economic indicators of thermal power units; rated capacity, rated voltage, charge / discharge efficiency, lifespan degradation characteristics, and total lifecycle cost of electrochemical energy storage devices; and power grid operation data. Data acquisition employs a multi-source fusion approach, obtaining real-time operating data at the second to minute level through SCADA systems and intelligent measurement terminals, extracting historical data from the dispatch database and equipment ledger system for the past year, and collecting market data from the power trading platform.

[0054] 1.2 Data Preprocessing

[0055] Data preprocessing eliminates defects in raw data and transforms it into standardized data that meets subsequent requirements, ensuring data integrity and reliability.

[0056] (1) Outlier Removal: The "3σ criterion + physical constraints" are used for screening. For data such as load and unit output that conform to a normal distribution, the mean μ and standard deviation σ are calculated, and data outside the range of μ-3σ to μ+3σ are removed to avoid interference from single-point anomalies on scene clustering and objective function calculation;

[0057] (2) Missing value imputation: Short-term missing values ​​are imputed using linear interpolation, with the following formula:

[0058]

[0059] Where: x t x represents the missing value at time t. t-1 x t+1 This represents valid data for adjacent time points.

[0060] (3) Feature normalization

[0061] Min-Max normalization is performed on data of different scales, such as load fluctuation rate and unit output level, and mapped to the [0,1] interval.

[0062]

[0063] Where: x is the original data, x min x max These are the minimum and maximum values ​​for this type of data, respectively.

[0064] 2. Hierarchical objective function construction (coupling of main objective and sub-objectives)

[0065] To address the optimization needs of the "thermal power unit + electrochemical energy storage" collaborative system, a hierarchical architecture is constructed, consisting of a "main objective (maximizing grid-side revenue)" and sub-objectives (maximizing power plant-side revenue and minimizing energy storage lifetime loss). The core design logic is as follows: the main objective prioritizes ensuring the safe and economical operation of the power grid; the sub-objectives quantify power plant-side revenue and energy storage lifetime loss respectively, ensuring the enthusiasm of power plants to participate and the economic efficiency of energy storage configuration, ultimately achieving multi-objective collaborative optimization of "guaranteed grid safety, improved power plant revenue, and controllable energy storage loss".

[0066] 2.1 Definition of the main objective: Maximizing grid-side revenue

[0067] The benefits on the grid side encompass two major dimensions: "improved economic benefits + reduced safety costs," specifically including three items: frequency regulation service revenue, grid loss costs, and savings in reserve capacity costs.

[0068] (1) Frequency modulation service revenue

[0069] In the composition of grid-side revenue, frequency regulation service revenue is the most critical and directly reflects the value of system operation, as it reflects the degree to which generating units contribute to frequency regulation in different load ranges.

[0070] When the generating units are operating under normal load conditions, the revenue from the first frequency regulation service of the power grid can be calculated as follows:

[0071]

[0072] When the generating units operate under wide load conditions, the revenue from the second frequency regulation service of the power grid can be calculated as follows:

[0073]

[0074] Among them, c by and c tk λ and ε represent the unit prices of the spinning reserve and regulation capacity of thermal power units, respectively, in yuan / (MW·h); λ and ε represent the contributions of energy storage frequency regulation to the spinning reserve and regulation capacity of thermal power units, respectively; c pf and c pm The compensation unit prices for reducing the maximum frequency deviation of the power system and improving the frequency recovery speed of the power system are respectively, c pf The unit is yuan / (MW·h·Hz), c pm The unit is yuan / (MW·h·(Hz / s)); Δf mi and Δν si These represent the maximum decrease in frequency deviation and the increase in frequency recovery speed within the i-th cycle after energy storage-assisted frequency regulation, respectively; P Gn (t) represents the output of the thermal power unit under normal load, in MW; This indicates the lower limit of the output of a thermal power unit under normal load operation; P represents the maximum allowable output of a thermal power unit. Gw (t) represents the output of a thermal power unit under wide load conditions, in MW; This indicates the lower limit of output capacity for wide-load operation;

[0075] Therefore, the revenue from frequency regulation services of the power grid is: C p =C p1 +C p2 .

[0076] (2) Network loss cost

[0077] In addition to the revenue from frequency regulation services, the unavoidable energy losses during grid operation also have a significant impact on overall economic benefits. Grid losses are essentially the electricity generated that is consumed in the grid before it is delivered to users.

[0078] The formula for line loss power is: ΔP loss (t)=I line (t) 2 ·R line ·L line ·10 -6

[0079] Among them, I line (t) represents the line current at time t, in A; R line Resistance per unit length of the line, in ohms / km; L line The length of the line is in km.

[0080] Therefore, the network loss cost can be calculated as follows:

[0081] C loss=C e,t ·∫ΔP loss (t)dt

[0082] Among them, C loss Cost of network loss, yuan; C e,t The real-time grid connection price is yuan / kWh; ΔP loss (t) represents the line loss power, in kW.

[0083] (3) Cost savings in standby capacity

[0084] Energy storage can replace part of the standby capacity of thermal power units, reducing the costs of unit start-up, shutdown, and no-load operation required by the power grid to maintain standby.

[0085] Annual standby capacity cost savings are: C grid,spare =(S spare0 -S spareopt )·T year ·C spare ·10 -4

[0086] Among them, S spare0 This represents the reserve capacity when no energy storage is available; S spareopt This refers to the grid's reserve capacity when energy storage is available; T year = 8760h (annual duration); C spare Cost per unit of standby capacity.

[0087] (4) Comprehensive calculation of main objective: grid-side revenue

[0088] F main =C p +C grid,spare -C loss

[0089] In the formula, F main For grid-side revenue, C p For the revenue from power grid frequency regulation services, C grid,spare For annual standby capacity costs, C loss Cost of network loss.

[0090] 2.2 Sub-objective 1: Maximize revenue on the power plant side

[0091] (1) Energy storage investment cost

[0092] Energy storage investment costs can be divided into: operation and maintenance costs C1 and initial investment costs C2.

[0093] C inv =C1+C2

[0094] C1 = C E E ess +C P PBN

[0095] C2=k E E ess +k P P BN

[0096] In the formula: C E C represents the unit price per unit capacity of the energy storage system. P E represents the unit price per unit power of the energy storage system. ess P is the rated capacity of the energy storage system. BN K represents the rated power of the energy storage system. E K represents the unit capacity operation and maintenance cost of an energy storage system. P This refers to the operation and maintenance cost per unit power of the energy storage system.

[0097] (2) Coal consumption cost

[0098] The coal consumption cost of a thermal power unit over a period of time can be calculated by examining the unit's coal consumption under different loads. By integrating the product of the unit's power output and its coal consumption rate at that power output over a period of time, the unit's coal consumption during that period can be obtained. The coal consumption cost then equals the coal consumption multiplied by the coal price.

[0099] C h =C coal ∫(R ate,cc P G (t))dt

[0100] R ate =a r P G (t) 2 +b r P G (t)+c r

[0101] Among them, C h Coal consumption cost of thermal power units; R ate Coal consumption rate of thermal power units; C coal The unit price of coal for the current season; a r b r c r This is the coal consumption cost coefficient for thermal power units.

[0102] (3) Increased power generation revenue of the unit

[0103] With the addition of energy storage systems, the extra reserve capacity of conventional generating units can be used for electricity market transactions, thereby increasing the generating revenue of the units.

[0104] R G =(ρ i -c i )·PBN ·T

[0105] In the formula: ρ i It's the grid electricity price, c i It is the power generation cost, T is the frequency regulation calculation period, and P is the power generation cost. BN It is the rated power of energy storage.

[0106] The formula for maximizing revenue on the power plant side is:

[0107] F sub1 =max(R) G -C h -C inv )

[0108] In the formula, F sub1 To maximize power plant-side revenue, R G To increase the power generation revenue of thermal power units, C h For the coal consumption cost of thermal power units, C inv This refers to the investment cost of energy storage.

[0109] 2.3 Sub-objective 2: Minimize the lifetime loss rate of energy storage side

[0110] To reflect the impact of energy storage performance degradation on system benefits during long-term operation, its lifetime attrition rate can be defined and calculated as follows:

[0111]

[0112] Among them, F sub2 To minimize the energy storage side lifetime loss rate, η ess,loss,annual Δη is the annual energy storage lifespan loss rate. ess,loss T represents the energy storage lifetime loss per unit time step; ess The rated cycle life of the energy storage is represented by Δt, which indicates the time step interval.

[0113] 3. Setting Constraints for Collaborative Systems

[0114] To ensure that the optimization results meet the requirements of equipment operating characteristics and power grid safety, two types of collaborative constraints are set: thermal power unit operating constraints and electrochemical energy storage operating constraints, as detailed below:

[0115] 3.1 Constraints of Thermal Power Units

[0116] Output range constraints: The output of thermal power units must be between the minimum stable output and the rated capacity to avoid unstable combustion at low loads or operation exceeding the rated power. The formula is as follows:

[0117] P unit,min ≤P unit (t)≤P unit,max

[0118] Among them, P unit,min For the minimum stable output of the thermal power unit, P unit,max This is to ensure the maximum stable output of the thermal power unit.

[0119] Rate of ramping constraint: The rate of change of output of thermal power units must be less than the maximum rate of ramping to prevent excessive boiler thermal stress or turbine damage. The formula is as follows:

[0120]

[0121] Where, r unit,up The maximum load increase rate; r unit,down This represents the maximum load reduction rate.

[0122] Coal consumption rate constraint: The coal consumption rate of thermal power units varies with power output in a quadratic function relationship and must meet the coal consumption characteristic curve provided by the manufacturer. The formula is as follows:

[0123] R ate =a r P G (t) 2 +b r P G (t)+c r

[0124] Among them, a r b r c r This is the coal consumption characteristic coefficient.

[0125] Start-up / shutdown frequency constraint: Frequent start-ups and shutdowns of thermal power units will shorten equipment lifespan and increase maintenance costs. Therefore, the annual number of start-ups and shutdowns needs to be controlled, as shown in the following formula:

[0126] 0≤N unit,start (t)≤N unit,start,max

[0127] Where, N unit,start,max This represents the maximum number of start-ups and shutdowns allowed per year for thermal power units.

[0128] 3.2 Operational Constraints of Electrochemical Energy Storage

[0129] Capacity constraint: The real-time remaining capacity of the energy storage must be between 0 and the rated capacity to avoid overcharging or over-discharging and damaging the battery. The formula is as follows:

[0130] 0≤E ess (t)≤E ess

[0131] Among them, E ess (t) represents the remaining energy storage capacity at time t; E ess Configure the rated capacity for energy storage.

[0132] Charging and discharging power constraints: The charging and discharging power of energy storage must be within the rated power range, with charging being positive and discharging being negative, as shown in the following formula:

[0133] -P ess,r ≤P ess (t)≤P ess,r

[0134] Among them, P ess,r This is the rated power of the energy storage, and P ess,r =k·E ess ;P ess (t) represents the energy storage charging and discharging power at time t.

[0135] State of charge constraints:

[0136] Overcharging and over-discharging can lead to aging and failure of energy storage batteries. Therefore, it is necessary to constrain the State of Charge (SOC) of energy storage within a limited range. The SOC of energy storage is calculated as the ratio of remaining capacity to rated capacity. The SOC of energy storage at time t is shown in the following formula:

[0137]

[0138] Energy storage system SOC level constraints:

[0139] SOC min ≤SOC(t)≤SOC max

[0140] In the formula, and Here, E represents the discharge and charging power of the energy storage at time i, in MW; E0 represents the initial capacity of the energy storage, in MWh; and typically, the State of Charge (SOC) is used for charging and discharging. min Set to 0.1, SOC max Take 0.9.

[0141] 4. Optimize the solution

[0142] To achieve coordinated optimization of the capacity and operation strategy of thermal power units and electrochemical energy storage, the Multi-Objective Improved Vulture Optimization Algorithm (MO-IVOA) is adopted as the core solution tool. Based on the Vulture Optimization Algorithm (VOA) proposed in 2021, MO-IVOA simulates the predation behavior of vultures—"scouting prey-encircling prey-attacking prey"—and handles multi-objective optimization problems through non-dominated sorting and crowding distance calculation. Its core advantages lie in its strong global search capability, fewer parameters, and fast convergence speed.

[0143] 4.1 Algorithm Improvement Design

[0144] (1) Population initialization: Abandoning traditional random initialization, a feasible initial population is generated based on historical operating data. Unit output and load data are extracted from one year of historical data of the power plant, and "unit output-load" combinations that meet the constraints are screened. Then, based on the energy storage power P ess (t)=P load (t)-P unit (t), P load (t) represents the grid load power at time t. Calculate the energy storage power demand corresponding to each combination. Based on the energy storage power demand and the continuous frequency regulation duration, determine the initial range E of the energy storage capacity. ess,range =[E ess,min E ess,max ],in, T freq,min Minimum continuous frequency modulation duration; in E ess,range N initial individuals are randomly generated to form the initial population.

[0145] (2) Multi-objective fitness calculation

[0146] The fitness of an individual is determined by non-dominated ranking and crowding distance calculation, and the steps are as follows:

[0147] a) Perform a non-dominated ranking of all individuals in the population: if individual A is not inferior to individual B in all objectives and is superior to B in at least one objective, then A dominates B; divide the population into different dominance layers, the first layer is the non-dominated solution (Pareto optimal solution candidate), the second layer is the solution dominated by the first layer, and so on.

[0148] b) Calculate the crowding distance for individuals within each dominance layer: Where CD(i) is the crowding distance of individual i, f m (i) represents the function value of individual i on the m-th objective (main objective, sub-objective 1, sub-objective 2). These are the maximum and minimum values ​​of the m-th target, respectively;

[0149] c) The fitness ranking is based on both the priority of the dominant layer and the descending order of the crowding distance. Individuals with the first layer and the largest crowding distance are given priority to enter the next generation to ensure the diversity of solutions.

[0150] 4.2 Algorithm Flow

[0151] The execution flow of the MO-IVOA optimization solution module consists of 7 core steps, as follows:

[0152] (1) Parameter initialization: Input algorithm parameters: population size N, maximum number of iterations T, probability p of the reconnaissance stage, step size coefficient α in the attack stage; input constraint parameters: generator output range, SOC range, frequency deviation threshold, etc.; input objective function parameters: coal price, electricity price, frequency regulation unit price, etc.

[0153] (2) Initial population generation: Generate an initial population based on historical data to obtain N initial individuals X = {X1, X2,..., X N}}, where each individual corresponds to a set of rated capacities E of energy storage configurations ess and energy storage power P ess (t); then calculate the main objective, sub-objective 1, and sub-objective 2 of each individual.

[0154] (3) Non-dominated sorting and crowding degree calculation: First, perform non-dominated sorting on the initial population to divide the dominance levels; then calculate the crowding distance of individuals within each dominance level to construct the initial Pareto optimal solution set P.

[0155] (4) Iterative optimization (t = 1 to T)

[0156] a) Stage 1: Reconnaissance for prey (global optimization)

[0157] Generate a random number r1 ∈ [0, 1]. If r1 < p, where p represents the probability of the reconnaissance stage, then the vulture conducts reconnaissance flight and updates the individual position: X(t + 1) = X rand (t) - r2 × |X rand (t) - 2r3 × X(t)|, where X rand (t) is an individual randomly selected from the population, and r2 and r3 ∈ [0, 1] are random numbers; if r1 > p, then enter the stage of surrounding the prey.

[0158] b) Stage 2: Surrounding the prey (local search)

[0159] When r1 > p in Stage 1, enter the local search in Stage 2. First, calculate the fitness of all individuals in the population, and select the individual with the best fitness as the "prey position" X best (t). Then generate a random number r4 ∈ [0, 1]. If r4 < 0.5, then contract the enclosure: X(t + 1) = X best (t) - r5 × |X best (t) - X(t)|, where r5 ∈ [0, 2] is a random number that linearly decreases from 2 to 0 with the number of iterations; if r4 > 0.5, then perform spiral flight to surround: X(t + 1) = |X best (t) - X(t)| × e αr6 × cos(2πr6) + X best (t), where r6 ∈ [-1, 1] is a random number and α is the spiral shape coefficient.

[0160] c) Phase 3: Attacking the Prey (Partial Development)

[0161] Generate a random number r7 ∈ [0,1]. If r7 < 0.5, then launch a frontal attack: X(t+1) = X best (t)-r5×|X best (t)-X mean (t)|, where X mean (t) represents the average position of the current population; if r7 > 0.5, then a flank attack is necessary: ​​X(t+1) = X rand (t)-r5×|X rand (t)-X mean (t)|.

[0162] d) Phase 4: New Population Assessment and Update

[0163] First, calculate the objective function value of the newly generated individuals, merge the current population with the new population, and perform non-dominated sorting and crowding calculation. Then, retain the N best individuals to form the next generation population X(t+1), update the Pareto optimal solution set P. If the new individual is a non-dominated solution and its crowding distance is greater than the minimum crowding distance solution in the existing solution set, then replace the minimum distance solution (to ensure solution set diversity). If the new individual is dominated by any solution in the solution set, then discard the new individual. Finally, record the optimal objective function value of the current iteration (main objective, sub-objective 1, sub-objective 2) for subsequent convergence determination.

[0164] (5) Convergence criterion

[0165] The "dual-index convergence criterion method" is adopted, which simultaneously satisfies the convergence of the objective function and the convergence of the Pareto solution set. The specific criterion logic is as follows:

[0166] Indicator 1: Objective function convergence:

[0167] Calculate the mean of the principal objective over the last 5 iterations. If the current primary objective value F main,current The deviation from the mean of the main objective satisfies (ε1=10 -3 If the relative convergence threshold is reached, then the main objective converges.

[0168] Indicator 2: Pareto solution set convergence

[0169] Calculate the "similarity distance" D between the current Pareto solution set and the solution set from the previous iteration. sim

[0170]

[0171] Among them, P tLet P be the solution set of the current iteration. t-1 For the solution set of the previous iteration, |P t | represents the current size of the solution set; if D sim <ε2(ε2=5×10 -3 If the similarity threshold is met, then the solution set converges.

[0172] If both indicators converge simultaneously, or the number of iterations reaches the maximum number of iterations T, then stop iterating; otherwise, return to step 4 to continue iterating.

[0173] (6) Optimal solution selection and output

[0174] From the final Pareto optimal solution set P final In this process, the optimal energy storage capacity configuration scheme is selected based on the actual needs of the project.

[0175] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for optimizing electrochemical energy storage capacity considering wide-load operation of thermal power units, characterized in that, include: Hierarchical Objective Function Construction: To address the optimization requirements of the thermal power unit and electrochemical energy storage collaborative system, a hierarchical objective function with coupled main and sub-objectives is constructed. The main objective is to maximize grid-side revenue, prioritizing the safe and economical operation of the grid. Grid-side revenue includes frequency regulation service revenue considering the wide-load operation of thermal power units, grid loss costs, and reserve capacity costs. The sub-objectives are to maximize power plant-side revenue and minimize energy storage-side lifetime loss rate, ensuring the incentive for power plant participation and the economic efficiency of energy storage configuration. Coordinated system constraint settings: Set operating constraints for thermal power units and electrochemical energy storage; Optimization solution: The multi-objective improved vulture optimization algorithm is used as the core solution tool for the hierarchical objective function to obtain the optimal electrochemical energy storage capacity configuration scheme.

2. The electrochemical energy storage capacity optimization method according to claim 1, characterized in that, Also includes: Data acquisition and preprocessing: Collect technical parameters, operating status parameters and economic indicators of thermal power units, rated capacity, rated voltage, charge and discharge efficiency, life decay characteristics and total life cycle cost of electrochemical energy storage equipment, and power grid operation data.

3. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, The formula for calculating grid-side revenue is: F main =C p +C grid,spare -C loss In the formula, F main For grid-side revenue, C p For the revenue from power grid frequency regulation services, C grid,spare For annual standby capacity costs, C loss Cost of network loss.

4. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, When thermal power units are operating under normal load conditions, the revenue from the first frequency regulation service of the power grid is calculated as follows: When thermal power units operate under wide load conditions, the revenue from the second frequency regulation service of the power grid is calculated as follows: Among them, c by and c tk λ and ε represent the unit prices of the spinning reserve and regulation capacity of thermal power units, respectively, in yuan / (MW·h); λ and ε represent the contributions of energy storage frequency regulation to the spinning reserve and regulation capacity of thermal power units, respectively; c pf and c pm The compensation unit prices for reducing the maximum frequency deviation of the power system and improving the frequency recovery speed of the power system are respectively, c pf The unit is yuan / (MW·h·Hz), c pm The unit is yuan / (MW·h·(Hz / s)); Δf mi and Δν si These represent the maximum decrease in frequency deviation and the increase in frequency recovery speed within the i-th cycle after energy storage-assisted frequency regulation, respectively; P Gn (t) represents the output of the thermal power unit under normal load, in MW; This indicates the lower limit of the output of a thermal power unit under normal load operation; P represents the maximum allowable output of a thermal power unit. Gw (t) represents the output of a thermal power unit under wide load conditions, in MW; This indicates the lower limit of output capacity for wide-load operation; Power Grid Frequency Regulation Service Revenue C p For: C p =C p1 +C p2 .

5. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, The network loss cost is calculated as follows: C loss =C e,t ·∫ΔP loss (t)dt Among them, C loss Cost of network loss, yuan; C e,t The real-time grid connection price is yuan / kWh; ΔP loss (t) represents the line loss power, in kW; The formula for line loss power is: ΔP loss (t)=I line (t) 2 ·R line ·L line ·10 -6 Among them, I line (t) represents the line current at time t, in A; R line Resistance per unit length of the line, in ohms / km; L line The length of the line is in km.

6. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, The formula for calculating the annual standby capacity cost is as follows: C grid,spare =(S spare0 -S spareopt )·T year ·C spare ·10 -4 Among them, C grid,spare For annual standby capacity costs, S spare0 This represents the reserve capacity when no energy storage is available; S spareopt This refers to the grid's reserve capacity when energy storage is available; T year =Annual duration; C spare Cost per unit of standby capacity.

7. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, The formula for maximizing the power plant's revenue is as follows: F sub1 =max(R G -C h -C inv ) In the formula, F sub1 To maximize power plant-side revenue, R G To increase the power generation revenue of thermal power units, C h For the coal consumption cost of thermal power units, C inv For energy storage investment costs; Energy storage investment costs are divided into operation and maintenance costs C1 and initial investment costs C2. C inv =C1+C2 C1=C E E ess +C P P BN C2=k E E ess +k P P BN In the formula, C E C represents the unit price per unit capacity of the energy storage system. P E represents the unit price per unit power of the energy storage system. ess P is the rated capacity of the energy storage system. BN K represents the rated power of the energy storage system. E K represents the unit capacity operation and maintenance cost of an energy storage system. P This refers to the operation and maintenance cost per unit power of the energy storage system.

8. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, The formula for minimizing the lifespan loss rate of the energy storage side is as follows: Among them, F sub2 To minimize the energy storage side lifetime loss rate, η ess,loss,annual Δη is the annual energy storage lifespan loss rate. ess,loss T represents the energy storage lifetime loss per unit time step; ess The rated cycle life of the energy storage is represented by Δt, which indicates the time step interval.

9. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, The aforementioned operating constraints for thermal power units include constraints on the output range of thermal power units, ramp-up rate constraints, coal consumption rate constraints, and start-up / shutdown frequency constraints; the aforementioned operating constraints for electrochemical energy storage include constraints on capacity, charge / discharge power, state of charge constraints, and SOC level constraints of the energy storage system.

10. The method for optimizing electrochemical energy storage capacity according to claim 1 or 2, characterized in that, When the deviation between the main objective and sub-objectives in the objective function is large, the global exploration probability of the algorithm is increased, and the search range is expanded by using a spiral flight strategy; when the objective function value tends to be stable, the local development mode is switched to focus on refining the area around the optimal solution, so as to avoid getting trapped in local optima due to the multi-peak characteristics of the objective function.

Images