A power supply planning optimization method and system considering energy storage dynamic reserve contribution and comprehensive value evaluation

Abstract

The application discloses a power supply planning optimization method and system considering energy storage dynamic backup contribution and comprehensive value evaluation, and the method comprises the following steps: a flat distribution model is established based on the weekly net load peak value and the energy storage reliable capacity, the total maintenance capacity of the thermal power unit is optimally distributed to each week of the whole year to obtain an available capacity matrix; a time sequence operation simulation model is established, which contains the energy storage state of charge and the upward backup coupling constraint, and the model outputs the system operation cost, the load shedding sequence and the wind and light curtailment amount; the model is taken as the core, the dichotomy method is used to iteratively solve the reliable capacity of the energy storage, and the contribution of the energy storage to the system capacity adequacy is quantified; based on the operation simulation result and the reliable capacity, the energy value and the capacity value of the energy storage are brought into the annual total cost target function, and the optimal expansion scheme is searched under the reliability constraint. The application can effectively improve the economy and reliability of the power supply planning scheme, and is suitable for the power supply expansion optimization decision of the power system containing a high proportion of new energy power.

Description

Technical Field

[0001] This invention belongs to the field of power system planning and operation technology, and in particular relates to a power planning optimization method and system that takes into account the dynamic reserve contribution and comprehensive value assessment of energy storage. Background Technology

[0002] With the deepening implementation of the "dual carbon" goals, the installed capacity of new energy sources, mainly wind and solar power, continues to rise, profoundly changing the operating characteristics of the power system. The volatility and intermittency of new energy output, coupled with the temporal variations in load, significantly increases the system's demand for flexible adjustment resources. Energy storage systems, especially with their rapid charge and discharge response capabilities and energy transfer capabilities over time, have become one of the important means to support the absorption of new energy and improve system reliability. During the power planning phase, decision-makers need to assess the comprehensive value of energy storage systems to determine a reasonable scale of energy storage configuration and operating strategies. However, existing power planning methods still have several shortcomings in quantifying the value of energy storage.

[0003] First, the modeling of energy storage backup capacity is relatively simplified. Traditional planning methods typically focus primarily on the spinning reserve contribution of conventional units, paying insufficient attention to the ability of energy storage systems to rapidly provide backup power. While some studies have attempted to incorporate energy storage into the backup system, their modeling methods often assume the backup capacity of energy storage as a constant value independent of the state of charge. In reality, the upward backup capacity of energy storage is constrained by its current remaining energy: when the state of charge is low, even with sufficient power capacity, energy storage struggles to continuously provide positive backup power. Existing models rarely characterize this dynamic coupling relationship, leading to biases in the assessment of the actual reliability contribution of energy storage.

[0004] Second, the comprehensive value assessment of energy storage is not comprehensive enough. The value of energy storage in a power system typically encompasses multiple aspects, including energy time-shift value, capacity value, and environmental value. Existing planning methods mostly focus on quantifying energy time-shift value, and a unified and easily implemented assessment framework for the capacity value (reliable capacity) and environmental value (carbon emission reduction) of energy storage has not yet been established. As a result, the comprehensive benefits of energy storage are difficult to fully identify in planning decisions.

[0005] Third, there is a disconnect between planning and operational simulation. Traditional methods often use simplified simulations based on typical day or load duration curves, which fail to accurately reflect the temporal coupling between renewable energy output and load, the start-up, shutdown, and ramp-up constraints of conventional units, the continuous dynamics of energy storage state of charge, and the superimposed impact of random faults and renewable energy fluctuations in multiple scenarios. This may lead to insufficient reserves or load shedding events in actual operation of the planned scheme, resulting in distorted reliability assessments.

[0006] Fourth, existing methods do not adequately consider the impact of planned outages (maintenance) of thermal power units. Thermal power units require several weeks of maintenance annually. If maintenance periods are concentrated during peak load periods or low renewable energy output periods, it will significantly worsen reserve adequacy. Existing planning methods often ignore planned outages or only assume a uniform distribution of maintenance capacity throughout the year, failing to dynamically reflect the reduction in available capacity due to maintenance in time-series simulations, as well as the coordination between maintenance plans and load and renewable energy characteristics, leading to an overestimation of reliability.

[0007] In conclusion, it is necessary to develop a technical method that can accurately simulate the dynamic operating characteristics of energy storage, fully quantify its multiple values, and closely couple planning and operation simulation. Summary of the Invention

[0008] To address the problems mentioned in the background section, this invention discloses a power planning optimization method and system that considers dynamic backup capacity and comprehensive value assessment of energy storage, providing a decision-making basis for power expansion schemes. To achieve the above objectives, the technical solution adopted by this invention is as follows:

[0009] A power planning optimization method that considers the dynamic reserve contribution and comprehensive value assessment of energy storage, characterized by the following steps:

[0010] S1. Based on the weekly net load peak and reliable energy storage capacity, a leveling allocation model is established to optimize the allocation of the total planned maintenance capacity of thermal power units to each week of the year, obtain the weekly planned outage capacity, and then spread it to each hour of that week to obtain the final available capacity matrix.

[0011] S2, establish a time-series operation simulation model with constraints on energy storage state of charge and upward reserve coupling, input the given power configuration, and output the system operating cost, load shedding sequence and wind and solar curtailment amount;

[0012] S3, using the aforementioned time-series simulation model as the core of the calculation, employs a bisection method to iteratively solve for the reliable capacity of energy storage, thereby quantifying the contribution of energy storage to the system's capacity adequacy.

[0013] S4. Based on the operating results output by the time-series simulation model and the reliable capacity, the energy value and capacity value of energy storage are incorporated into the annual total cost objective function, and the optimal expansion scheme is searched under the reliability constraint.

[0014] Furthermore, the weekly net load peak in S1 is calculated by subtracting the theoretical maximum output of wind power from the hourly load within that week, then subtracting the theoretical maximum output of photovoltaic power, and taking the maximum value. The calculation formula is as follows:

[0015]

[0016] in, This represents the weekly net load peak. For load; To maximize the theoretical output of wind power; To maximize the contribution of photovoltaic theory;

[0017] The formula for calculating the weekly reliable energy storage capacity is:

[0018]

[0019] in, For reliable energy storage capacity per week; This refers to the rated discharge duration of the energy storage. The duration of weekly peak net load is defined as the maximum number of consecutive hours during which net load power exceeds 95% of its peak value within that week. This is the rated power of the energy storage.

[0020] Furthermore, the decision variable in the levelized allocation model in S1 is the weekly planned shutdown capacity; the formula for the optimization model is:

[0021]

[0022] in, This refers to the total installed capacity of thermal power plants. This is the thermal reserve factor. The target is the average value; the optimization model can be solved quickly using the CPLEX solver.

[0023] Furthermore, the thermal power unit in S2 adopts a piecewise linear cost model, dividing it into K output segments. The formula for calculating the total output of the thermal power unit is as follows:

[0024]

[0025] Each segment's output meets the requirements The total output of thermal power plants does not exceed the available capacity for the current period and meets the ramping constraint. The formula for the ramping constraint is:

[0026]

[0027] in, , These are the per-unit values ​​for the downward and upward ramp rates, respectively.

[0028] Furthermore, the objective function of the time-series simulation model in S2 is to minimize the sum of the segmented fuel cost of thermal power, load shedding penalty cost, wind and solar curtailment penalty cost, energy storage discharge marginal cost, and carbon emission cost.

[0029] 6. The power planning optimization method considering the dynamic reserve contribution and comprehensive value assessment of energy storage according to claim 1, characterized in that the reliable capacity of energy storage in S3 is quantified using effective load carrying capacity, and the calculation process includes:

[0030] S31, calculate the baseline annual energy deficit without energy storage, using the following formula:

[0031]

[0032] in, For the number of scenes, For the scene Time period The amount of load shedding;

[0033] S32, a constant load increment is superimposed on the original load. The timing simulation model was called to obtain A binary search method is used to satisfy... The load increment value is used as the reliable capacity of energy storage, and the convergence condition is that the search interval width is less than the preset tolerance.

[0034] S33, the capacity credit ratio is obtained by dividing the reliable capacity by the rated power of the energy storage.

[0035] Furthermore, the annual total cost objective function in S4 includes the sum of the annualized investment cost and the system's average annual operating cost; the annualized investment cost uses an equal-annual value coefficient, and the calculation formula is as follows:

[0036]

[0037] Converting one-time investments to annual averages, of which The discount rate is... The economic lifespan of the equipment is defined as follows: the annualized investment cost includes new investment in wind power, new investment in photovoltaic power, new investment in energy storage power, and new investment in energy storage energy; the average annual operating cost is obtained by averaging multiple random scenarios using the time-series operation simulation model; the reliability constraint is that the expected load loss does not exceed 2.4 hours / year.

[0038] Furthermore, in S4, the optimal solution for thermal power capacity expansion is searched by enumerating all possible combinations of units in the given capacity set. The total new capacity does not exceed the given capacity limit. For each combination, the timing simulation model is called to calculate the operating cost and reliability. Solutions that do not meet the reliability constraint that the load loss expectation does not exceed 2.4 hours / year are eliminated. Finally, the combination with the minimum annual total cost is selected.

[0039] Furthermore, in S4, for the wind-solar-storage expansion scheme, Bayesian optimization is used to simultaneously search for the wind power expansion multiple, photovoltaic expansion multiple, new energy storage power, and new energy storage duration. After initially sampling several points, iterative optimization is performed. For each sampling point, the time-series simulation model is called to evaluate the operating cost and reliability, with the goal of minimizing the total annual cost.

[0040] This invention also discloses a power planning optimization system that considers the dynamic reserve contribution and comprehensive value assessment of energy storage, used to implement the above-mentioned power planning optimization method that considers the dynamic reserve contribution and comprehensive value assessment of energy storage, characterized in that the system includes:

[0041] The planned shutdown allocation module is used to execute step S1 above;

[0042] Run the simulation module to perform step S2 above;

[0043] The capacity assessment module is used to perform step S3 described above;

[0044] The comprehensive value quantification and planning decision-making module is used to execute step S4 above.

[0045] The present invention has the following beneficial effects:

[0046] (1) In the operation simulation, the present invention explicitly establishes the coupling constraint between the energy storage upside reserve capacity and the remaining energy, so that the reserve capacity is dynamically adjusted with the state of charge. Compared with the prior art, which usually sets the energy storage reserve capacity as a constant value that is independent of the state of charge, and still assumes that it can provide full reserve when the energy storage energy is low, resulting in an overestimation of the system reliability, the present invention more realistically reflects the physical limitations of energy storage.

[0047] (2) The present invention adopts a piecewise linear aggregation model, which divides the thermal power units into a finite number of output segments after sorting them according to their marginal costs. Compared with existing planning methods that often use a single equivalent unit or introduce binary start-stop variables, it is difficult to balance scheduling accuracy and computational efficiency. The present invention retains the priority characteristics of economic scheduling and maintains linear programming, which can quickly solve and output the system marginal electricity price.

[0048] (3) The present invention proposes a leveling allocation model to optimize the allocation of the total maintenance capacity to each week. The constraints include the contribution of reliable energy storage capacity to the reserve. Compared with most planning methods that ignore the planned maintenance of thermal power units or assume that the maintenance capacity is evenly distributed throughout the year, the present invention enables the maintenance plan to naturally avoid the peak net load week, while minimizing the fluctuation of the maintenance capacity in each week.

[0049] (4) This invention constructs a three-level quantitative framework that includes energy value (operating cost + carbon cost), capacity value (ELCC) and environmental value (carbon emission reduction), and provides two optimization methods: combined enumeration and Bayesian optimization. Compared with existing technologies that usually only evaluate the energy time-shift value of energy storage, capacity value and environmental value are difficult to be uniformly included in the optimization. This invention achieves the decision of minimizing the total annual cost under reliability constraints. Attached Figure Description

[0050] Figure 1 This is a flowchart of a power planning optimization method that takes into account the dynamic reserve contribution and comprehensive value assessment of energy storage, as described in a specific embodiment.

[0051] Figure 2 This is a flowchart illustrating the planned shutdown leveling allocation in a specific embodiment.

[0052] Figure 3 This is a schematic diagram of a timing simulation model containing SOC-backup coupling constraints, as described in a specific embodiment.

[0053] Figure 4 This is a schematic diagram illustrating the cost change trend of coal-fired power plant expansion as described in a specific embodiment.

[0054] Figure 5 This is a schematic diagram illustrating the optimal total annual cost of wind and solar capacity expansion as described in a specific embodiment.

[0055] Figure 6 This is a schematic diagram comparing the economic efficiency of the extended solutions described in a specific embodiment. Detailed Implementation

[0056] To facilitate understanding by those skilled in the art, the present invention will be further described below in conjunction with embodiments and accompanying drawings.

[0057] Before optimizing the power planning, a forced shutdown scenario for conventional units is first generated as the basic input for all subsequent operational simulations. A sequential Monte Carlo simulation method is used to generate the time-series state sequence for each unit: the operating state of each thermal power unit is described by a two-state Markov process, with alternating states of "normal operation" and "fault shutdown". When a unit is in normal operation, its time to failure (TTF) is sampled according to an exponential distribution with a mean equal to the unit's mean time to failure (MTTF); after a unit experiences a fault and enters a shutdown state, its time to repair (TTR) is sampled according to an exponential distribution with a mean equal to the unit's mean time to repair (MTTR). The forced shutdown rate of the units satisfies:

[0058]

[0059] Each unit is independently sampled for its state sequence. Then, the available capacity of all units at the same moment is summed to obtain the hourly available capacity matrix of the system under each scenario. ,in Number the scene. For a given time period. A total of [number] were generated. Each independent scenario, in this embodiment This serves as input for subsequent planned shutdown allocation and timing simulation.

[0060]

Example 1

[0061] This embodiment provides a power planning optimization method that considers the dynamic reserve capacity and comprehensive value assessment of energy storage. The overall approach is as follows: using hourly time-series operational simulation as the core tool, a coupling relationship between the state of charge (SOC) of energy storage and its uplink reserve capacity is constructed in the model. Based on this, the operational simulation model is used to quantify the operational economics, capacity reliability (ELCC), and carbon emission reduction benefits of energy storage. Finally, economic and reliability indicators are incorporated into the planning objective function that minimizes the total annual cost, providing a decision-making basis for power expansion schemes. (See also...) Figure 1 The method includes the following steps:

[0062] S1, see reference Figure 2 Based on the weekly net load peak and reliable energy storage capacity, the weekly maintenance capacity of thermal power units is allocated in a levelized manner; specifically including:

[0063] S11, calculate the weekly net load peak and reliable energy storage capacity; assume a total annual load of [missing information]. Weekly peak net load The result is obtained by subtracting the theoretical maximum output of wind and solar power from the load:

[0064]

[0065] Average weekly unplanned available capacity The average available capacity during forced outage scenarios is obtained by averaging over a week. This represents the reliable capacity that energy storage can provide each week. Calculated in the following way:

[0066] Statistics on the duration of weekly peak net load For example, the load exceeds The number of consecutive hours of peak value;

[0067] Reliability capacity factor ,in This refers to the rated discharge duration of the energy storage.

[0068] , This is the rated power of the energy storage.

[0069] S12, Levelized Allocation Model, where the decision variable is the weekly planned outage capacity. (MW). The optimization model is:

[0070]

[0071] in This refers to the total installed capacity of thermal power plants. This is the thermal reserve factor. The target is the average value. This model is a convex quadratic programming problem, which can be solved quickly using solvers such as CPLEX. After solving, the weekly planned shutdown capacity is calculated. Tile this up to each hour of the week, subtract it from the forced shutdown matrix, and you get the final available capacity matrix. This is used for all subsequent simulation scenarios.

[0072] S2, see reference Figure 3 An economic dispatch model is established, incorporating SOC-reserve coupling constraints and segmented thermal power costs. Given a power supply configuration, the model outputs system operating costs, load shedding sequences, and wind and solar curtailment rates. Specifically, this includes:

[0073] S21, this model aggregates multiple thermal power units, uses piecewise linear cost to describe their economic dispatch characteristics, and describes the hourly operation process of the system in a single scenario in the form of linear programming. The scenario includes random outages of conventional units and fluctuations in renewable energy output. The system parameter settings are shown in Table 1:

[0074]

[0075] Table 1

[0076] Known parameters include:

[0077] Thermal power: Total installed capacity Piecewise capacity vector Piecewise marginal cost vector Per-unit value of uphill / downhill ramp rate Available capacity after random outages .

[0078] Wind power and photovoltaic power: installed capacity per-unit value sequence Punishment for abandoning wind and light (RMB / MWh).

[0079] load: (MW).

[0080] Energy storage: Rated power Rated duration Charging efficiency Discharge efficiency Marginal discharge cost (RMB / MWh), minimum SOC ratio .

[0081] Reliability: Load shedding penalty (RMB / MWh), thermal reserve coefficient .

[0082] Carbon emissions: carbon price (Yuan / ton CO2), carbon emission intensity of thermal power plants (tons CO2 / MWh), where γ is the weighted average carbon emission intensity of the thermal power unit, determined according to the fuel type and power generation ratio of each unit; in this embodiment, γ = 0.85 tons CO2 / MWh.

[0083] S22, the constraints include:

[0084] S221, the output constraint of thermal power plants in sections, the formula is:

[0085]

[0086]

[0087] S222, the ramp-up constraint for total thermal power output, the formula is:

[0088]

[0089] S223, the constraint on new energy output, the formula is:

[0090]

[0091] S224, Energy Storage Energy and Power Constraints, the formula is:

[0092]

[0093]

[0094]

[0095]

[0096] S225, Daily Cycle Constraint: Energy returns to its initial value at the end of each 24-hour period. , That is, the daily starting SOC is set to the minimum allowable value. In this embodiment, =10%.

[0097] S22,6, Power balance constraint, the formula is:

[0098]

[0099] S227, System Upward Reserve Constraint: The total upward reserve capacity of the system must meet a certain proportion of the load, as shown in the formula:

[0100]

[0101] Among them, the upward backup capacity provided by energy storage Subject to the following two restrictions:

[0102] Power Limitation:

[0103] Energy Limitation:

[0104] at the same time .

[0105] The physical meaning of this set of constraints is as follows: when the energy storage SOC is low, the remaining energy is insufficient, and even if the power capacity is sufficient, it cannot provide effective uplink backup; when the SOC is high, the backup capacity is mainly limited by the power upper limit. This modeling approach enables the operational simulation during the planning phase to reflect the actual constraints of energy storage SOC on backup capacity.

[0106] S23, Objective function: Minimize operating costs in a single scenario, including fuel costs, load shedding penalties, wind and solar curtailment penalties, marginal costs of energy storage discharge, and carbon emission costs.

[0107]

[0108] This model is a linear programming problem and can be solved efficiently using commercial solvers.

[0109] S3, using the operational simulation model of S2 as its core, employs a bisection method to iteratively solve for the reliable capacity ELCC of energy storage, quantifying its contribution to the system's capacity adequacy; specifically including:

[0110] S31, Effective Load Carrying Capability (ELCC) of energy storage refers to the additional load power that the system can generate by connecting energy storage, while maintaining the original reliability level of the system (e.g., using the annual energy deficit energy EENS as an indicator). This embodiment of the invention uses the annual energy deficit energy EENS as the reliability indicator and solves the ELCC iteratively using a bisection method. The operational simulation model of Module 2 is embedded in the iterative process, enabling the ELCC evaluation to take into account the SOC-standby coupling constraints of the energy storage.

[0111] S32, Calculation steps, specifically including:

[0112] S321, calculate the energy shortage without energy storage baseline. (Settings / settings) That is, without energy storage, the operation simulation model of module two is invoked for all Simulations were performed using pre-generated random scenarios to obtain the total annual load shedding for each scenario. These random scenarios included conventional unit failures and fluctuations in renewable energy output. The average load was then calculated. :

[0113]

[0114] S322, binary search for ELCC. In the original load sequence... Superimposed constant load increment , forming new load For a given Then, the simulation model from Module 2 is invoked again, this time with the energy storage parameters to be evaluated. Calculations yielded ELCC means that the condition is met. of value.

[0115] Solve using the bisection method iteratively:

[0116] Initialize lower bound upper limit , The annual average load;

[0117] Set tolerance and maximum number of iterations In this embodiment, the tolerance is set to 1MW, and the maximum number of iterations is [not specified]. Take 20 times;

[0118] Repeat the following steps until Or reach the maximum number of iterations:

[0119] ;

[0120] Parallel computing of all In each scenario And take the average;

[0121] like ,but ;otherwise ;

[0122] final .

[0123] S323, Calculate the capacity-to-credit ratio using the following formula:

[0124]

[0125] S4, based on the operational results of S2 and combined with the ELCC results of S3, incorporates the energy value and capacity value of energy storage into the annual total cost objective function, with environmental value as an independent output indicator. It then searches for the optimal expansion scheme using enumeration or Bayesian optimization methods. Specifically, it includes:

[0126] S41, the objective function for the total annual cost. In power expansion planning, it is necessary to compare the economics of different expansion schemes. This invention uses minimizing the total annual cost as the planning objective. The total annual cost consists of two parts: the annualized investment cost of the new equipment and the average annual operating cost of the system.

[0127]

[0128] Annualized investment cost Using the equivalent annual value coefficient Converting one-time investments to annual averages, of which The discount rate is... For equipment lifespan.

[0129]

[0130] in Investment per unit for wind power and photovoltaic power (RMB / MW) Investment in energy storage capacity and energy unit (yuan / MW, yuan / MWh). To add capacity.

[0131] Average annual operating cost : Based on the running simulation model in S2 The average of the scenarios is obtained as follows:

[0132]

[0133] S42, Reliability Constraints: Each candidate solution must meet a preset reliability threshold, for example:

[0134]

[0135] in (Expected load shedding) is obtained from the statistics of load shedding events in all scenarios. A reliability of 2.4 hours per year is acceptable. Solutions that do not meet reliability constraints will be eliminated.

[0136] S43, Planning and Optimization Solution Framework. This invention provides two optimization methods, including:

[0137] Coal-fired power capacity expansion optimization: Enumerate all possible combinations of units in a given capacity set, with the total new capacity not exceeding a given capacity limit. This limit is pre-set by planners based on factors such as system scale, land use conditions, and fuel supply; in this embodiment, it is set to 3000 MW. For each combination, S1 is called to calculate operating costs and reliability, and the combination that meets the requirements is selected. The combination with the lowest hourly / yearly and lowest total annual cost.

[0138] Optimization of wind-solar-storage capacity expansion: Bayesian optimization is used to simultaneously search for wind-solar expansion multiples. and new energy storage capacity Duration The objective function is the total annual cost. During the optimization process, module two needs to be called at each sampling point to perform a simulation to evaluate the operating cost and reliability.

[0139] Ultimately, the output is a power expansion scheme that meets reliability constraints and minimizes total annual cost.

[0140]

Example 2

[0141] The following describes the specific implementation of the present invention in detail, taking the actual load and wind and solar power output data of a certain power grid as an example, in conjunction with the accompanying drawings.

[0142] Basic data preparation is shown in Table 2.

[0143]

[0144] Table 2

[0145] The baseline scenario assessment includes the following: The total annual operating cost of the energy storage system is RMB 10.037 billion, of which fuel costs are RMB 7.524 billion (approximately 75.0%), representing the main cost item for thermal power units. Carbon emission costs are RMB 2.509 billion (approximately 25.0%), calculated based on a carbon price of RMB 80 / ton and a carbon emission intensity of 0.85 tons / MWh, which is highly correlated with fuel costs. Load shedding penalties are RMB 3.5508 million (approximately 0.035%), corresponding to EENS = 355.08 MWh, with a penalty unit price of RMB 10,000 / MWh. Energy storage costs are RMB 158,900 (a very small percentage), with a marginal cost of RMB 100 / MWh for energy storage discharge, indicating that the actual discharge volume of energy storage is very small. Curtailment penalties are RMB 0.00, meaning that current new energy installations (800MW wind power and 1000MW photovoltaic) can be fully absorbed by the system, and wind and solar curtailment penalties have not been triggered.

[0146] Without energy storage, the Loop Length Expiration (LOLE) reaches 21.83 hours / year, far exceeding common reliability standards (such as 2.4 hours / year). After configuring energy storage, the LOLE drops to 1.81 hours / year, meeting reliability requirements. EENS decreases by a significant 91.9%, and the load shedding penalty is substantially reduced (from approximately 43.64 million yuan to 3.55 million yuan). The main contribution of energy storage lies in providing upward reserve capacity (SOC-reserve coupling constraint), discharging during peak load periods or conventional unit failures, thus avoiding numerous load shedding events. System performance comparisons are shown in Table 3.

[0147]

[0148] Table 3

[0149] The reliable capacity (ELCC) calculation for energy storage specifically includes: using the EENS (4363.97 MWh / year) of a system without energy storage as a baseline, a bisection method is used to search for the ELCC of energy storage. A tolerance of 1 MW is set, with a maximum of 20 iterations, and the average EENS at each test point is evaluated in parallel. The final result is... MW, capacity-to-credit ratio This result can be used to assess the capacity value of energy storage during the planning phase.

[0150] Comparison of planning schemes, specifically including: setting a target annual load growth to MW, reliability target Hours / Year, Curtailment Rate .

[0151] Option 1 (Coal-fired Power Plant Expansion): See details Figure 4 The algorithm enumerates all combinations of 300 MW, 600MW, and 1000MW units, and calculates the operating cost and reliability for each combination using Module 2 (fixed existing wind, solar, and energy storage). Among the combinations meeting reliability requirements, the minimum annual total cost occurs when an additional 1300MW (one 300MW unit and one 1000MW unit) is added, with an annual total cost of RMB 12.318 billion (including investment and operation), a Loose-Leaf (LOLE) of 1.57 hours / year, and a curtailment rate of 0.0%.

[0152] Option 2 (Wind, Solar and Storage Expansion): See details Figure 5 Bayesian optimization was employed, with the wind-solar expansion factor as the search variable. New energy storage capacity MW, new energy storage duration Hours. Initially sampled at 5 points, iterated 12 times, with the objective function being the total annual cost (including new investment in wind, solar, and energy storage). The optimal solution is... (Wind power 4800MW, photovoltaic 6000MW), new energy storage MW, h, with a total annual cost of RMB 11.318 billion, LOLE = 2.05 hours / year, and a curtailment rate of 2.0%.

[0153] See Figure 6 Option 2 (wind, solar and energy storage expansion) has a total annual cost that is about 8.1% lower than Option 1 and has a higher carbon emission reduction benefit. Therefore, Option 2 is recommended as the target year power expansion option.

[0154] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of this invention.

Claims

1. A power planning optimization method that considers the dynamic reserve contribution and comprehensive value assessment of energy storage, characterized in that, The method includes the following steps: S1. Based on the weekly net load peak and reliable energy storage capacity, a leveling allocation model is established to optimize the allocation of the total planned maintenance capacity of thermal power units to each week of the year, obtain the weekly planned outage capacity, and then spread it to each hour of that week to obtain the final available capacity matrix. S2, establish a time-series operation simulation model with constraints on energy storage state of charge and upward reserve coupling, input the given power configuration, and output the system operating cost, load shedding sequence and wind and solar curtailment amount; S3, using the aforementioned time-series simulation model as the core of the calculation, employs a bisection method to iteratively solve for the reliable capacity of energy storage, thereby quantifying the contribution of energy storage to the system's capacity adequacy. S4. Based on the operating results output by the time-series simulation model and the reliable capacity, the energy value and capacity value of energy storage are incorporated into the annual total cost objective function, and the optimal expansion scheme is searched under the reliability constraint.

2. The power planning optimization method considering the dynamic reserve contribution and comprehensive value assessment of energy storage as described in claim 1, characterized in that, The weekly net load peak in S1 is calculated by subtracting the theoretical maximum output of wind power from the hourly load within that week, then subtracting the theoretical maximum output of photovoltaic power, and taking the maximum value. The formula is as follows: ； in, This represents the weekly net load peak. For load; To maximize the theoretical output of wind power; To maximize the contribution of photovoltaic theory; The formula for calculating the weekly reliable energy storage capacity is: ； in, For reliable energy storage capacity per week; This refers to the rated discharge duration of the energy storage. This refers to the duration of the weekly peak net load. This is the rated power of the energy storage.

3. The power planning optimization method considering the dynamic reserve contribution and comprehensive value assessment of energy storage as described in claim 1, characterized in that, The decision variable in the levelized allocation model in S1 is the weekly planned downtime capacity; the formula for the optimization model is: ； in, This refers to the total installed capacity of thermal power plants. This is the thermal reserve factor. The target is the average value; the optimization model can be solved quickly using the CPLEX solver.

4. The power planning optimization method considering the dynamic reserve contribution and comprehensive value assessment of energy storage as described in claim 1, characterized in that, The thermal power unit in S2 adopts a piecewise linear cost model, dividing it into K output segments. The formula for calculating the total output of the thermal power unit is: ； Each segment's output meets the requirements The total output of thermal power plants does not exceed the available capacity for the current period and meets the ramping constraint. The formula for the ramping constraint is: ； in, , These are the per-unit values ​​for the downward and upward ramp rates, respectively.

5. The power planning optimization method according to claim 1, which takes into account the dynamic reserve contribution and comprehensive value assessment of energy storage, is characterized in that... The objective function of the time-series operation simulation model in S2 is to minimize the sum of the segmented fuel cost of thermal power, load shedding penalty cost, wind and solar curtailment penalty cost, energy storage discharge marginal cost, and carbon emission cost.

6. The power planning optimization method according to claim 1, which takes into account the dynamic reserve contribution and comprehensive value assessment of energy storage, is characterized in that... The reliable capacity of energy storage in S3 is quantified using effective load carrying capacity, and the calculation process includes: S31, calculate the baseline annual energy deficit without energy storage, using the following formula: ； in, For the number of scenes, For the scene Time period The amount of load shedding; S32, a constant load increment is superimposed on the original load. The timing simulation model was called to obtain A binary search method is used to satisfy... The load increment value is used as the reliable capacity of energy storage, and the convergence condition is that the search interval width is less than the preset tolerance. S33, the capacity credit ratio is obtained by dividing the reliable capacity by the rated power of the energy storage.

7. The power planning optimization method according to claim 1, which takes into account the dynamic reserve contribution and comprehensive value assessment of energy storage, is characterized in that... The annual total cost objective function in S4 includes the sum of the annualized investment cost and the system's average annual operating cost; the annualized investment cost uses an equal-annual value coefficient, and the calculation formula is as follows: ； Converting one-time investments to annual averages, of which The discount rate is... The economic lifespan of the equipment is defined as follows: the annualized investment cost includes new investment in wind power, new investment in photovoltaic power, new investment in energy storage power, and new investment in energy storage energy; the average annual operating cost is obtained by averaging multiple random scenarios using the time-series operation simulation model; the reliability constraint is that the expected load loss does not exceed 2.4 hours / year.

8. The power planning optimization method according to claim 1, which takes into account the dynamic reserve contribution and comprehensive value assessment of energy storage, is characterized in that... In S4, the optimal solution for thermal power capacity expansion is searched by enumerating all possible combinations of units in the given capacity set. The total new capacity does not exceed the given capacity limit. For each combination, the timing simulation model is called to calculate the operating cost and reliability. Solutions that do not meet the reliability constraint that the load loss expectation does not exceed 2.4 hours / year are eliminated. Finally, the combination with the minimum annual total cost is selected.

9. The power planning optimization method considering the dynamic reserve contribution and comprehensive value assessment of energy storage as described in claim 1, characterized in that, In S4, for the wind-solar-storage expansion scheme, Bayesian optimization is used to simultaneously search for the wind power expansion multiple, photovoltaic expansion multiple, new energy storage power, and new energy storage duration. After initial sampling of several points, iterative optimization is performed. For each sampling point, the time-series simulation model is called to evaluate the operating cost and reliability, with the goal of minimizing the total annual cost.

10. A power planning optimization system that considers the dynamic reserve contribution and comprehensive value assessment of energy storage, used to implement the power planning optimization method that considers the dynamic reserve contribution and comprehensive value assessment of energy storage as described in any one of claims 1 to 9, characterized in that, The system includes: A planned shutdown allocation module is used to execute step S1 as described in claim 1; Run the simulation module to perform step S2 as described in claim 1; A capacity assessment module is used to perform step S3 as described in claim 1; The comprehensive value quantification and planning decision-making module is used to perform step S4 as described in claim 1.

Images