A substation energy storage system configuration method with high proportion of new energy access

Abstract

The application discloses a substation energy storage system configuration method with high proportion of new energy access, relates to the technical field of power system engineering design and energy storage system planning, and comprises the following steps: identifying main transformer overload samples, extracting an overload reference duration through probability modeling; performing power balance calculation under a typical operation mode according to the overload reference duration, determining energy storage system rated power and rated capacity; performing parameter selection in combination with discharge depth, and completing hardware configuration; after the configured electrochemical energy storage system is connected to a substation and outdoor arrangement is determined, performing multi-scheme evaluation and selection through a multi-dimensional evaluation model to determine an optimal configuration scheme. The application solves the problems of capacity mismatch caused by lack of overload sample probability law identification, insufficient hardware output caused by neglect of the constraint of discharge depth on effective energy storage capacity conversion, and difficulty in balancing treatment effect and safety caused by lack of multi-dimensional quantitative evaluation in the prior art.

Description

Technical Field

[0001] This invention relates to the field of power system engineering design and energy storage system planning technology, and more specifically, to a method for configuring a substation energy storage system with a high proportion of new energy access. Background Technology

[0002] After a high proportion of intermittent renewable energy is connected to the distribution network, the spatiotemporal imbalance between the randomness and volatility of power output and the rigid demand of electricity load leads to frequent local overload and reverse power flow problems in the main transformers of substations. This not only causes the insulation of the main transformers to age and shorten their lifespan, but also triggers the operation of protection devices, resulting in unplanned power outages. This has become the core bottleneck for the further large-scale consumption of renewable energy.

[0003] Currently, the mainstream solutions for substation overload problems fall into two categories: transformer capacity expansion and electrochemical energy storage configuration. Transformer capacity expansion can only passively improve the load-bearing capacity of the main transformer and cannot solve the intermittent problem of renewable energy generation. Moreover, the expansion space is significantly limited by the substation site conditions. On the other hand, energy storage systems, with their flexible charging and discharging characteristics, can simultaneously manage main transformer overload and absorb redundant photovoltaic power, making them the core solution for high-proportion renewable energy scenarios.

[0004] However, existing technologies for configuring energy storage systems in substation scenarios have significant drawbacks: First, capacity calculations rely heavily on empirical estimations, without probabilistic modeling of measured overload samples to extract overload baseline durations, and lack power balance calculations under typical operating conditions, which makes it easy for the determined rated power and capacity of energy storage to be redundant or insufficient. Second, the hardware selection for electrochemical energy storage is crude, failing to combine the depth of discharge for refined parameter selection, making it difficult to meet the requirements of long life and intrinsic safety of substations. Third, after the system is connected and deployed outdoors, there is a lack of evaluation models that cover multiple dimensions such as economy and safety to quantitatively compare and select candidate solutions, making it impossible to scientifically determine the optimal configuration. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a substation energy storage system configuration method with a high proportion of new energy access, in order to solve the problems of capacity mismatch caused by the lack of overload sample probability law identification, insufficient hardware output capability caused by ignoring the constraint of discharge depth on effective energy storage capacity conversion, and difficulty in balancing governance effectiveness and safety in real substation environments due to the lack of multi-dimensional quantitative evaluation.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The configuration method for energy storage systems in substations with a high proportion of renewable energy access includes the following steps: Operational data is collected and overload samples of the main transformer are identified. Overload baseline durations representing risk distribution are extracted through probabilistic modeling. Based on the overload baseline durations, power balance calculations are performed under typical operating conditions to determine the rated power and rated capacity of the energy storage system. Based on the rated power, rated capacity, and depth of discharge of the energy storage system, parameter selection of the electrochemical energy storage system is performed, and hardware configuration is completed. After the configured electrochemical energy storage system is connected to the substation and its outdoor layout is determined, multiple schemes are evaluated and compared through a multi-dimensional evaluation model to determine the optimal configuration scheme.

[0007] In a preferred embodiment, the process of identifying overload samples of the main transformer includes: setting a load rate threshold based on the insulation heat resistance level of the main transformer; traversing the operating data and extracting continuous time segments where the load rate exceeds the load rate threshold as valid overload samples; calculating the difference between the start time and the recovery time of each sample to construct an overload duration sample.

[0008] In a preferred embodiment, extracting the overload baseline duration characterizing the risk distribution through probabilistic modeling includes: fitting the overload duration sample with a Weibull distribution, wherein the shape and scale parameters of the Weibull distribution are solved using the maximum likelihood estimation method to construct a probability density function; and the validity of the probability density function is verified using distribution tests and goodness-of-fit indices.

[0009] In a preferred embodiment, the typical operating mode includes: substation summer peak load forward overload, spring and autumn photovoltaic peak reverse overload, and N-1 fault condition.

[0010] In a preferred embodiment, the calculation of the rated power of the energy storage system includes: calculating the forward overload power gap under the summer peak load operation mode, the reverse redundancy power under the spring and autumn photovoltaic peak operation mode, and the rated compensation power under the N-1 fault condition; comparing the forward overload power gap, the reverse redundancy power, and the rated compensation power, and selecting the maximum value among them as the original power reference value; multiplying the original power reference value by the preset engineering safety margin coefficient to obtain the rated power of the energy storage system.

[0011] In a preferred embodiment, the rated capacity of the energy storage system is obtained by multiplying the rated power of the energy storage system by the overload reference duration.

[0012] In a preferred embodiment, determining the rated power and rated capacity of the energy storage system further includes: scaling the rated power and rated capacity of the energy storage system proportionally using the product of the main transformer capacity scaling factor and the new energy penetration rate correction factor to obtain the rated parameters of the energy storage system after adaptive matching of the target substation; wherein, the main transformer capacity scaling factor is determined based on the ratio of the rated capacity of the main transformer of the target substation to that of the benchmark reference substation; and the new energy penetration rate correction factor is determined based on the product of the difference in new energy penetration rates between the target substation and the benchmark reference substation and a preset penetration rate sensitivity coefficient plus 1.

[0013] In a preferred embodiment, the hardware configuration includes: calculating the effective capacity based on the rated capacity and depth of discharge of the vanadium redox flow battery, and determining the vanadium ion concentration and tank volume in accordance with Faraday's law.

[0014] In a preferred embodiment, the multidimensional evaluation model includes safety indicators; wherein, the safety indicators include a battery safety factor, and the formula for the battery safety factor is as follows: , In the formula, This represents the probability of a fire being triggered. For battery safety factor, This represents the probability of fault propagation. To protect the system's fault tolerance, As the fire triggering probability weight, For the probability weight of fault propagation, To protect against errors, a fault tolerance weight is used.

[0015] In a preferred embodiment, the fire triggering probability formula is as follows: , In the formula, This represents the probability of a fire being triggered. This represents the probability of thermal runaway. This is a factor that amplifies the risk of heavy overload. This represents the failure factor of the temperature control system.

[0016] The technical effects and advantages of the substation energy storage system configuration method with high proportion of new energy access in this invention are as follows: This invention collects substation operation data and identifies overload samples of main transformers. Based on probabilistic modeling, it extracts the overload baseline duration representing the risk distribution and performs power balance calculations under typical operating conditions. It then determines the rated power and rated capacity of the energy storage system and, combined with the depth of discharge, selects parameters and configures the hardware for the electrochemical energy storage system. This directly transforms the random overload distribution of the substation's main transformers into deterministic energy storage system power and capacity configuration parameters. This avoids insufficient or excessively redundant energy storage capacity configuration due to statistical biases in overload duration, improving the engineering adaptability of the solution. By connecting the electrochemical energy storage system to the substation and determining its outdoor layout, and then using a multi-dimensional evaluation model to evaluate and compare multiple solutions, it helps balance investment and safety within the limited physical space constraints of the substation. This effectively solves the problems of capacity mismatch caused by the lack of probabilistic identification of overload samples, insufficient hardware output capacity due to neglecting the constraint of the depth of discharge on the effective energy storage capacity conversion, and the difficulty in achieving both governance effectiveness and safety in real substation environments due to the lack of multi-dimensional quantitative evaluation in existing technologies. Attached Figure Description

[0017] Figure 1 A schematic diagram illustrating the configuration process for energy storage systems in substations with a high proportion of renewable energy access.

[0018] Figure 2 The figure shows the Weibull distribution fitting curve for the overload duration of the main transformer in the substation.

[0019] Figure 3 This is a typical daily photovoltaic / load output curve for a substation. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1 Please see Figure 1 This invention provides a method for configuring a substation energy storage system with a high proportion of new energy access, comprising the following steps: S1, collect operational data and identify overload samples of the main transformer, and extract the overload baseline duration representing the risk distribution through probability modeling; S2, Based on the overload reference duration, perform power balance calculations under typical operating conditions to determine the rated power and rated capacity of the energy storage system; S3, based on the rated power, rated capacity and depth of discharge of the energy storage system, selects parameters for the electrochemical energy storage system and completes the hardware configuration; S4. After the configured electrochemical energy storage system is connected to the substation and the outdoor layout is determined, a multi-dimensional evaluation model is used to evaluate and compare multiple schemes to determine the optimal configuration scheme.

[0022] This invention achieves a precise mapping of energy storage system scale from underlying statistical laws and operational constraints to engineering implementation by probabilistic modeling of overload duration samples and multi-condition power balance analysis. This is based on the deep coupling of overload reference duration, which characterizes the right-skewed overload distribution, and rated power of energy storage systems under multiple operating scenarios. Combined with hardware selection logic that decouples power and capacity and a multi-dimensional evaluation model, this invention eliminates capacity redundancy or insufficiency caused by normal distribution fitting. Through the synergistic evaluation of LCC and intrinsic safety based on the thermal runaway probability model, this invention helps to maximize asset absorption gains while ensuring the safety boundary of substation operation. It effectively solves the problems of capacity mismatch caused by the lack of overload sample probabilistic law identification, insufficient hardware output capability caused by ignoring the constraint of discharge depth on effective energy storage capacity conversion, and the difficulty in balancing governance effectiveness and safety in real substation environments due to the lack of multi-dimensional quantitative evaluation.

[0023] S1 collects operational data and identifies overload samples of the main transformer, and extracts the overload baseline duration representing the risk distribution through probability modeling.

[0024] With a high proportion of new energy access, the overload of the main transformer is driven by both meteorological changes and load fluctuations. The random fluctuations on the output side and the strong rigidity on the load side lead to the source-load imbalance in the spatiotemporal dimension. This makes the duration of the main transformer overload statistically exhibit a significant asymmetric "long-tail characteristic" (right-skewed distribution), which increases the evolution depth of extreme safety risks.

[0025] Traditional fixed-experience duration methods can easily lead to severe redundancy in energy storage capacity, while conventional normal distribution fitting can easily underestimate the probability of ultra-long overload under extreme weather conditions because it cannot effectively characterize long-tail characteristics, resulting in insufficient support capacity of the system under extreme operating conditions.

[0026] To overcome the aforementioned model mismatch defects, this embodiment introduces the Weibull distribution to probabilistically model the overload duration samples. Utilizing the distribution's strong shape adaptability, it accurately characterizes the long-tailed extreme failure model under the dual influence of weather and load, establishing an overload baseline duration that balances economy and safety.

[0027] S11, Collect main transformer operating data, filter main transformer overload samples from the main transformer operating data, and construct overload duration samples. This includes: Collect substation main transformer operation data, including the main transformer's historical operating power, load curve, and new energy output time sequence data; Set a load rate threshold (in this embodiment, it is based on the "Guidelines for the Safety and Stability of Power Systems"). And the equipment insulation heat resistance rating at Adjustments can be made within a certain range, preferably... ); The main transformer operating data is compared and traversed using a load rate threshold to filter out continuous time segments in which the main transformer load rate exceeds the load rate threshold, thus forming effective overload samples. Furthermore, the overload start time and recovery time of each valid overload sample are extracted, the duration of a single overload event is calculated, and all single duration data are summarized to construct an overload duration sample.

[0028] S12, Perform probabilistic statistical modeling on the overload duration sample to obtain the overload baseline duration. This includes: The Weibull distribution is used to fit the overload duration samples. Specifically, the shape and scale parameters of the distribution are solved by the maximum likelihood estimation (MLE) method to obtain the probability distribution model of the overload duration.

[0029] Preferably, the determination coefficients of the probability distribution model are calculated. Furthermore, the KS (Kolmogorov-Smirnov) test was used to double-validate the fit of the Weibull distribution, ensuring that the model can accurately and effectively represent the long-tail characteristics of overload duration.

[0030] Based on the validated probability distribution model, calculate the statistical quantiles of this distribution; select (General statistical standard) quantile duration corresponding to the confidence level, to preliminarily determine the overload baseline duration; Furthermore, considering the limited and discrete nature of the effective overload sample size in actual engineering, the t-distribution is used to calculate the confidence interval of the mean overload duration. Combined with the preset overload safety margin of the main transformer, the above-mentioned preliminarily determined overload reference duration is dynamically corrected. The confidence level is not a fixed constant and can be adjusted based on the safety importance level of the target substation (e.g., the difference in reliability requirements between a hub substation and a general terminal substation). It can be dynamically adjusted within a certain range.

[0031] like Figure 2 As shown, it displays the Weibull distribution fitting curve of the overload duration of the main transformer in the target substation. The horizontal axis represents the overload duration of the main transformer (e.g., in minutes), and the vertical axis represents the probability density of occurrence for that duration. Figure 2Intuitively, the fitted curve reaches its probability density peak at approximately 16 units of time on the horizontal axis, indicating that short-term overload events concentrated in this interval are the most frequent occurrences in actual operation. As the overload duration further increases, the curve extends to the right and the probability density slowly decreases, forming a significant asymmetric "long tail" characteristic. This distribution pattern accurately reveals the objective statistical law of substation overload conditions under high-proportion renewable energy access: short-term overloads dominate, while ultra-long-term overload events caused by extreme weather (such as sudden photovoltaic drops coupled with load surges), although having a lower probability of occurrence, objectively exist and their tail risks cannot be ignored. This embodiment uses the Weibull distribution, which has extremely strong shape adaptability, to fit the sample, effectively overcoming the defect of the traditional normal distribution, which seriously underestimates the probability of long-term extreme overloads due to the symmetry assumption. It can more realistically reflect the actual operating law of the power grid, providing solid statistical support for subsequent accurate calculation of the overload reference duration at a 95% confidence level and the scientific allocation of energy storage capacity.

[0032] S2, based on the overload reference duration, perform power balance calculations under typical operating conditions to determine the rated power and rated capacity of the energy storage system.

[0033] Based on the aforementioned extracted overload baseline duration, and combined with the main transformer capacity and renewable energy penetration rate, parameters are scaled proportionally and adaptively matched, specifically including: S21, Based on the overload reference duration, perform power balance calculations for different operating conditions to determine the rated power and rated capacity of the energy storage system; S211, extract typical scenarios covering different overload mechanisms; This embodiment first retrieves the annual main transformer operation data of the target substation, and uses a clustering algorithm combined with an overload rate threshold to extract typical scenarios covering different overload mechanisms, including but not limited to: (1) Pure load positive overload scenario: Typical operating mode such as summer night or winter night load peak, due to the lack of photovoltaic output to offset, the main transformer positive downward power transmission has an extremely high and steep peak. (2) Source-driven reverse feed scenario: The typical operating mode is the peak output of new energy in the middle of spring and autumn. At this time, the base load is extremely low, and the photovoltaic power generation is so large that the power cannot be consumed locally, resulting in serious power reverse feed exceeding the limit. The waveform usually shows a wide and flat characteristic.

[0034] S212 performs power balance calculations under different typical operating modes; To ensure that the substation can reduce the main transformer load rate to within the load rate threshold under any extreme operating conditions, the rated power of the energy storage system is determined using a multi-scenario envelope method. Specifically: S2121, positive overload gap calculation; Extract the real-time load power curve under the summer peak load operation mode, and calculate the difference between it and the rated load capacity corresponding to the safe operating load rate of the main transformer. The peak value of the difference is determined as the positive overload power gap. S2122, Calculation of reverse redundancy power; Extract the net load power curve under the peak photovoltaic operation mode in spring and autumn, calculate the power value of the new energy output exceeding the main transformer's allowable reverse power transmission limit, and determine its maximum value as the reverse redundant power; S2123, N-1 working condition compensation power calculation; The extreme condition of a single main transformer outage is simulated to obtain the total load demand of the substation and the maximum bearing limit of the remaining main transformer under short-term overload capacity. The power difference between the two is determined as the rated compensation power.

[0035] S2124, Rated power and rated output; By comparing the aforementioned positive overload power gap, reverse redundancy power, and rated compensation power, the maximum value is selected as the original power reference value. This value is then weighted and corrected using a preset engineering safety margin coefficient. The corrected value is ultimately determined as the rated power of the energy storage system. The formula for calculating the rated power of the energy storage system is as follows: , In the formula, P is the rated power of the energy storage system, in MW; and represents the power deficit in the summer positive overload scenario, in MW. Redundant power for reverse overload scenarios in spring and autumn, in MW; The rated compensation power is expressed in MW; γ is the engineering safety margin coefficient, which is taken as 1.1 in this embodiment.

[0036] S2125, Calculate the rated capacity of the energy storage system based on the rated power and overload reference duration of the energy storage system. The formula is as follows: , In the formula, P is the rated power of the energy storage system, and T is the overload reference duration.

[0037] S213 allows for the proportional adjustment of energy storage power and capacity parameters based on the rated capacity of the main transformer and the penetration rate of new energy sources, for substations of different voltage levels and sizes. Preferably, when the main transformer capacity is small or the scale of new energy access is low, the scale of energy storage configuration is reduced accordingly; when the main transformer capacity is large or the penetration rate of new energy is high, the scale of energy storage configuration is increased accordingly, so as to achieve adaptive matching of energy storage parameters under different operating scenarios. In one specific embodiment, an adaptive scaling model is constructed. Based on the rated power and rated capacity of the energy storage system obtained from the above steps, the rated power and rated capacity of the energy storage system of the target substation are corrected, thereby achieving adaptive matching of energy storage parameters under different operating scenarios. The implementation formula of the adaptive scaling model is as follows: , , In the formula, The rated power of the energy storage system after adaptive matching of the target substation; The rated capacity of the energy storage system after adaptive matching of the target substation; The rated power of the energy storage system calculated in the preceding steps; The rated capacity of the energy storage system calculated in the preceding steps; The main variable capacity scaling factor; This refers to the new energy penetration rate correction factor. The specific calculation methods for the main transformer capacity scaling factor and the new energy penetration rate correction factor are as follows: , , In the formula, The actual rated capacity of the main transformer in the target substation; The rated capacity of the main transformer in the reference substation; The actual renewable energy penetration rate of the target substation; The penetration rate of new energy in the benchmark substation; This is the preset permeability sensitivity coefficient; Wherein, the permeability sensitivity coefficient The coefficients are determined by least squares regression fitting based on multiple sets of historical configuration samples. The value range is set to 0.8 to 1.5; in a specific embodiment, the historical configuration sample is a database formed by collecting the operation data and design quotas of no less than 5 existing energy storage substations at the same voltage level. The reference substation is selected as a typical substation in the target area that meets the preset data collection integrity threshold and whose main transformer capacity is at the median level of its voltage level. The data collection integrity threshold is set to ensure that the probabilistic statistical model can accurately reproduce the "long tail characteristics" of overload. Specifically, it is set as follows: the annual time coverage rate is not less than 95%, the sampling frequency is not less than 96 points / day (i.e., 15 minutes / time), and the duration of a single continuous missing data does not exceed 4 hours.

[0038] It should be noted that the rated power and rated capacity used in subsequent steps have been adaptively matched with the target substation by default, and the variable names are consistent with those before the adaptive matching.

[0039] S3 selects parameters for the electrochemical energy storage system based on the rated power, rated capacity, and depth of discharge of the energy storage system, and completes the hardware configuration.

[0040] In this embodiment, considering the need for long-term overload mitigation in high-proportion renewable energy substations, vanadium redox flow batteries are preferentially selected as the energy storage medium. Based on the rated power and rated capacity determined through adaptive matching in the above steps, the core hardware is independently selected and modularly configured using the fully decoupled power and capacity characteristics of flow batteries. The specific configuration logic is as follows: (1) Capacity parameter mapping and electrolyte configuration: In response to the high capacity requirements caused by the long-tail overload characteristics, the effective energy storage capacity required by the system is first calculated based on the rated capacity after adaptive matching of the target substation and the preset depth of discharge (DOD); then, based on the law of conservation of energy and Faraday's law of electrolysis, the basic molar amount of active vanadium ions in the electrolyte is calculated; and taking into account the operating environment temperature and solubility limit, the vanadium ion concentration of the energy storage electrolyte and the total volume of the storage tank are determined, and the physical configuration of the energy storage electrolyte and the storage tank is completed accordingly.

[0041] (2) Power parameter mapping and stack and converter configuration: In response to the instantaneous power gap caused by sudden weather changes, the transient power output capability of the battery side is locked by adjusting the effective reaction area and the number of series and parallel connections of the stack unit based on the calculated rated power of the target substation energy storage system (i.e. the rated power to complete the adaptive matching of the target substation). When selecting a fuel cell stack, priority should be given to models with a cycle life greater than 15,000 cycles and a power regulation response time in the hundreds of milliseconds range to cope with high-frequency fluctuations in both the source and load. A bidirectional power converter (PCS) matching the rated power should be configured simultaneously to ensure the system meets the substation's grid connection voltage level and reactive power regulation requirements.

[0042] (3) Modular integration and auxiliary system configuration: Based on the capacity of the energy storage system and the layout space of the substation, the electrolyte storage tank, stack and PCS are designed in a containerized modular manner; and in combination with the environmental thermodynamic conditions of the substation, the auxiliary control unit including the circulating pumping system, pipeline valves and liquid / air cooling thermal management system is customized to ensure low loss and high reliability of the system fluid circulation.

[0043] Through the multi-dimensional selection and configuration based on power and capacity decoupling described above, this embodiment not only enables the energy storage system to accurately cover the complex overload management requirements calculated above, but also fully leverages the inherent safety and ultra-long lifespan physical advantages of vanadium redox flow batteries, realizing standardized and engineered hardware implementation for complex substation operating conditions.

[0044] S4. After the configured electrochemical energy storage system is connected to the substation and the outdoor layout is determined, a multi-dimensional evaluation model is used to evaluate and compare multiple schemes to determine the optimal configuration scheme.

[0045] Optimal energy storage configuration decisions are achieved through engineering physical access, spatial topology layout, and multi-dimensional quantitative evaluation. Specifically: S41, Connect the electrochemical energy storage system to the substation and determine its layout structure; S411, Engineering Physical Access; Comparing three connection methods for the energy storage system—connection to the low-voltage busbar of the substation, outgoing lines of important users, and emergency busbar—the selection of this method is based on the substation's voltage level and wiring configuration, prioritizing the reduction of electrical connection distance, line loss, and improved regulation response speed. This embodiment selects the 10kV low-voltage busbar connection method. This 10kV low-voltage busbar connection method can directly absorb redundant active power during peak photovoltaic periods and share the main transformer current during peak load periods. Calculations show that, due to the proximity of the connection point to the power balance center, the cable laying distance is shortened, and transmission loss is reduced by approximately [amount missing]. At the same time, it can share the existing 10kV power distribution operation and maintenance resources of the substation, reducing the initial construction and later operation and maintenance costs.

[0046] S412, Spatial topology arrangement; The location of the energy storage system is determined based on proximity to the low-voltage side of the main transformer, taking into account the layout of equipment within the station and the optimization of cable laying paths. The outdoor layout follows the substation electrical equipment installation specifications, meets the engineering requirements for fire prevention, seepage prevention, heat dissipation, and operation and maintenance access, and is adaptively adjusted according to the site conditions and electrical layout characteristics of substations of different voltage levels. In one specific embodiment, the energy storage system adopts a linear outdoor layout with independent functional zones to balance safety and maintenance space: The fuel cell stack unit area is arranged in a double-row symmetrical layout, with reserved space between each module. m is the maintenance and heat dissipation distance; Electrolyte storage tank area: Independently located on one side of the fuel cell stack unit, and equipped with a leak-proof dike. The logic behind this arrangement is to use physical isolation to prevent electrolyte leakage from corroding the fuel cell stack and electrical equipment; Combiner switch station: Located at the end of the system, it realizes DC combiner, inverter and fault protection, and shortens the outgoing line distance to the substation bus.

[0047] S42, through a multi-dimensional evaluation model, compare and analyze different preset schemes to determine the final configuration scheme; S421, Construct a multi-dimensional evaluation model, where the evaluation indicators include: S4211, Life Cycle Economic Indicator; The life-cycle economic indicators are calculated using the life-cycle cost (LCC) model, and the calculation formula is as follows: , In the formula, LCC is a life-cycle economic indicator. The LCC cost score is obtained after dimensionless processing (normalization / standardization). This is the initial investment, covering equipment purchase costs, installation costs, and construction costs. For operational costs, a 20-year operating cycle is preferred for calculation, including routine maintenance costs and energy loss compensation costs. Equipment replacement costs refer to the investment required to replace core components (such as battery clusters) when they reach the end of their lifespan during the operating cycle, as well as the resulting downtime losses. The cost of decommissioning and disposal includes the algebraic sum of residual value recovery revenue and environmental disposal expenditures.

[0048] S4212, operating indicators; The effectiveness of the configuration scheme is quantitatively evaluated by comparing the grid operation parameters before and after the integration of the energy storage system. The indicators include: Improvement in curtailment rate: The ratio of annual curtailed solar power to annual theoretical power generation, calculated after the integration of energy storage, due to the local absorption of redundant photovoltaic power by the energy storage system. Improvement in main transformer load rate: Quantified based on the percentage reduction in peak load rate of the main transformer.

[0049] Energy storage utilization rate: defined as the ratio of the total annual charge and discharge capacity to the theoretical maximum output capacity of the rated capacity, as shown in the following formula: , In the formula, For energy storage utilization rate; The effective charging and discharging capacity of the energy storage system during the statistical period (unit: MWh, i.e. the charging and discharging capacity that meets the needs of main transformer over-limit management and photovoltaic power consumption, excluding maintenance charging and discharging operations). Rated capacity of the energy storage system (unit: MWh); Therefore, the specific implementation steps to obtain the final running effect score are as follows: The system extracts the values ​​of the three indicators mentioned above from all preset candidate schemes. Since the improvement in curtailment rate, main transformer load rate, and energy storage utilization are all positive indicators (higher values ​​indicate better operational performance), the system iterates through all candidate schemes and extracts the maximum value for each individual indicator as a benchmark. Subsequently, the specific indicator values ​​of the target scheme are divided by the corresponding benchmark maximum values, thus unifying the three original indicators with different physical units into a single value. to Standard sub-item scores between; Obtain the system's preset sub-item weight set for the operational performance dimension. This weight set includes absorption weight, overload mitigation weight, and efficiency weight (the sum of the proportions of these three is always equal to a certain value). The scores of the three standard sub-items of the same target scheme are multiplied by their corresponding weight coefficients. Finally, the three product results under the target scheme are summed, and the resulting algebraic sum is the final performance score of the configuration scheme.

[0050] S4213, safety indicator; The security indicators include: System operational risk quantification value: using fire triggering probability The quantification and calculation formula is as follows: , Probability of thermal runaway under normal operating conditions (times / 10,000 clusters) Year); : Overload risk amplification factor; Failure coefficient of temperature control system.

[0051] Equipment safety level: A comprehensive evaluation of the fault tolerance rate and fault propagation prevention capability of the protection system is used to obtain the fault propagation capability and protection fault tolerance rate. Among them, the fault propagation capability quantifies the system's ability to contain a substantial fault (such as leakage) within a single physical area. The smaller the value of this indicator, the stronger the system's isolation and protection capability. Its specific acquisition process includes: At the physical boundaries of each independent functional module of the system, a quick-cut-off actuator (such as a pneumatic emergency shut-off valve installed at the inlet and outlet of the storage tank) is configured. When the system diagnoses an out-of-limit fault and issues an emergency isolation command, it records the mechanical action delay time from the issuance of the command to the complete closure of the rapid cut-off actuator. A safety risk spillover assessment is conducted to obtain the mechanical action delay time. This delay time is then multiplied by the system's pre-calibrated fault dynamic propagation rate (such as the maximum rated electrolyte flow rate of the main circulation pump) to calculate the estimated total fault spillover before the valve is closed. Subsequently, the estimated total fault spillover is divided by the physical isolation margin designed for the module (such as the effective remaining interception volume of the leak-proof dike), and the resulting ratio is directly determined as the fault propagation capability.

[0052] The specific process for obtaining the protection tolerance rate includes: The core safety nodes of the energy storage system adopt a dual-path heterogeneous sensing configuration, that is, two sensors with independent power supply, A and B, are deployed in parallel on the same node; The controller has a built-in arbitration program that compares the two signals in real time. When the main signal A experiences a communication interruption or an abnormal data jump, the controller triggers a fault-tolerant mechanism, switches control to the backup signal B, and records the actual time taken for this switchover. The actual time taken for the primary / backup link switchover is obtained, and then divided by the system's preset maximum allowable blind zone time threshold to obtain the time loss percentage. Next, the total value "1" is subtracted from the time loss percentage, and the percentage of the difference is determined as the final protection fault tolerance rate. According to this calculation logic, the shorter the actual time taken, the closer the obtained fault tolerance rate is to 100%.

[0053] Taking into account the probability of fire triggering, fault propagation capability, and protection fault tolerance rate, a weighted safety factor is calculated. The weights are then combined with the substation scenario to obtain the safety factor S. The formula for the battery safety factor is: , This represents the probability of fault propagation. To protect the system's fault tolerance, As the fire triggering probability weight, For the probability weight of fault propagation, To protect against errors, a fault tolerance weight is used.

[0054] S422, compare and analyze different preset schemes to determine the final configuration scheme; Specifically, in response to the overload management needs of the target substation, the system first generates a set of candidate solutions covering traditional power grid equipment transformation solutions (such as transformer expansion), the first type of electrochemical energy storage solutions (such as solid-state battery energy storage that requires periodic replacement of core components due to cycle life limitations), and the second type of electrochemical energy storage solutions (such as flow battery energy storage with power capacity decoupling and long cycle life). Subsequently, a quantitative comparison is carried out from three dimensions: full life cycle economic indicators, operation indicators, and safety indicators.

[0055] Preferably, after unifying the polarity and dimensionless processing of the LCC cost score, operational performance score, and safety factor S, the system calculates the comprehensive evaluation score of each candidate scheme by combining the actual scenario weight of the target substation. Finally, the candidate scheme with the optimal comprehensive evaluation score is output as the final configuration scheme.

[0056] Example 2 applies the above-mentioned general method to 110kV substations (Huaibei Substation, Kantuan Substation, Yuanzhuang Substation) in the Huaisu area, and provides a detailed implementation description of the energy storage system configuration method for substations with a high proportion of new energy access according to the present invention.

[0057] The basic parameters of the 110kV substation in this embodiment are as follows: the rated capacity of the main transformer is 50MVA, with 2 main transformers configured, model SFSZ11-50000 / 110, cooling method is ONAN / ONAF, the rated active power of a single main transformer is 45MW, the peak load rate is 111.1%, the N-1 condition load rate is 122.2%, the allowable reverse overload power is 54MW, the total installed capacity of photovoltaic power in the station is 154.75MW, the peak photovoltaic output during midday in spring and autumn reaches 108MW, and there are significant problems of main transformer forward overload and reverse power flow.

[0058] S1 collects operational data and identifies overload samples of the main transformer, and extracts the overload baseline duration representing the risk distribution through probability modeling.

[0059] Full-time operation data of three 110kV substations in the Huaisu area were collected. According to the standard of "main transformer load rate > 105%", a total of 127 main transformer overload samples were selected, including 43 samples from Huaibei substation, 41 samples from Kantuan substation, and 43 samples from Yuanzhuang substation. The main transformer overload samples are evenly distributed in all four seasons, under different weather conditions, and during key periods such as peak photovoltaic and peak load periods, and have good representativeness and unbiasedness.

[0060] Based on the effective overload samples, overload duration data are extracted to construct overload duration samples, and probabilistic modeling is performed using the Weibull distribution. A log-likelihood function is constructed using the maximum likelihood estimation method, and the derivatives with respect to the shape parameter x and scale parameter y are calculated to establish a system of maximum likelihood equations. The fixed-point iteration method is then used for numerical solution.

[0061] Specifically, the moment estimation method is used to perform preliminary fitting on the overload duration sample to obtain initial parameter values ​​( These are the initial shape parameters. (Initial scale parameters): Then, using this initial value as the starting point for iteration, the maximum likelihood estimation equations are solved using the Newton-Raphson iterative method. The convergence condition for the iteration is set as "the relative change in parameters between two adjacent iterations is less than..." The calculation was stopped "or the number of iterations reached 100". The final fitted parameters were x=2.37 and y=24.5 minutes. Since the scale parameter y of the Weibull distribution is the characteristic lifetime (i.e., the 63.2% quantile of the distribution), and considering the probability density distribution calculated based on the shape parameter x=2.37, the sample data exhibits a right-skewed distribution. The peak probability density interval of the overload duration is highly correlated with y=24.5 minutes, concentrated in the 20-25 minute range. The probability density function formula for the Weibull distribution is as follows: , Where t is the overload duration sample, x is the shape parameter, and y is the scale parameter.

[0062] After the fitting is completed, the fitting effect is verified using the coefficient of determination. The goodness of fit of the Weibull distribution model to the overload duration samples is evaluated using the following formula: , In the formula, For the first One sample of overload duration observations; Let be the fitted value of the Weibull distribution model for the i-th sample; η is the mean of the overload duration samples; and the sample size is n=127. The calculated r²=0.92 indicates that the Weibull distribution model can explain 92% of the overload duration data variation, demonstrating a high model fit. Based on this, the KS test is used to verify the statistical significance of the fitted distribution. The specific implementation steps are as follows: (1) Arrange the overload duration samples in ascending order to obtain ordered samples. ; (2) Calculate the empirical distribution function of the sample: ; (3) Based on the determined Weibull distribution parameters x=2.37 and y=24.5 minutes, calculate the theoretical distribution function value corresponding to the ordered sample. ; (4) Calculate the difference between the empirical distribution function and the theoretical distribution function, and take the maximum absolute value as the KS test statistic. (in ); (5) Based on the statistic D and the sample size n=127, the test P value is obtained by KS test distribution table or numerical calculation.

[0063] The KS test yielded a P=0.86, which is greater than the significance level of 0.05. Therefore, the null hypothesis that "the overload duration sample follows a Weibull distribution" is accepted. The fitting result is statistically significant and can effectively describe the distribution pattern of overload duration.

[0064] Based on the overload duration sample of the main transformer obtained by fitting the aforementioned Weibull distribution, and considering the safety level and reliability requirements of power system engineering design, a 95% confidence level was selected for statistical analysis. This confidence level is a commonly used high reliability confidence standard in the field of power equipment reliability design, is compatible with the safety design level of substation main transformer protection, and can cover most typical overload conditions at the statistical level, avoiding interference from extreme low-probability conditions on the benchmark value. The t-distribution was used to calculate the confidence interval of the mean of the overload duration sample, and the calculation formula is as follows: , In the formula, η is the sample mean of overload duration; For degrees of freedom n 1. Two-tailed quantiles of the t-distribution (corresponding to significance level α=0.05); σ is the sample standard deviation of overload duration; n is the sample size.

[0065] The calculated 95% confidence interval for the overload duration is [21.5, 26.2] minutes. Considering the safety margin requirements of the engineering design, the upper limit of the confidence interval is rounded up and a certain safety redundancy is reserved, ultimately determining the substation main transformer overload reference duration to be 30 minutes. This reference value can cover more than 95% of actual overload conditions, while also meeting the safety level requirements for main transformer overload protection in power system engineering design, providing a reliable time reference for subsequent energy storage system capacity configuration.

[0066] S2, based on the overload reference duration, perform power balance calculations under typical operating conditions to determine the rated power and rated capacity of the energy storage system.

[0067] This embodiment provides an adaptive power and capacity configuration method for energy storage systems in multi-scenario substation overload conditions, to achieve adaptive parameter matching based on load, source, and dynamic scaling. Specifically: The typical operating conditions of substations are broken down into two categories: peak nighttime load in summer and peak midday photovoltaic load in spring and autumn. Figure 3 As shown, power balance calculations are carried out based on typical operating conditions: refined power surplus and deficit calculations are performed for main transformers #1 and #2. The total power deficit during peak summer nighttime load periods is 8.5MW, and the reverse overload during peak midday photovoltaic periods in spring and autumn requires the absorption of 6MW of redundant power. Next, the N-1 operating condition verification is performed, specifically as follows: Based on the substation's main transformer configuration parameters, the total load power of the main transformer under operating condition N-1 is calculated to be 55MW. Combined with the rated active power of a single main transformer being 45MW, and considering safety margins, the rated compensation power required by the energy storage system is calculated to be 11MW. The formula for the rated power of the energy storage system is as follows: , In the formula, P is the rated power of the energy storage system, in MW; Power deficit for summer positive overload scenarios, in MW; Redundant power for reverse overload scenarios in spring and autumn, in MW; The rated compensation power is expressed in MW; γ is the engineering safety margin coefficient, which is taken as 1.1 in this embodiment.

[0068] Therefore, the rated power of the energy storage system is taken as 11MW, and the rated power and rated capacity of the energy storage system are determined as follows: Then, the rated capacity of the energy storage system is determined according to the overload reference duration. The rated capacity of the energy storage system is the product of the rated power of the energy storage system and the overload reference duration. The rated power of the energy storage system is 11MW, and the overload reference duration is 30 minutes (0.5 hours) determined in step S1. The rated capacity of the energy storage system is calculated to be 5.5MWh. S3 selects parameters for the electrochemical energy storage system based on the rated power, rated capacity, and depth of discharge of the energy storage system, and completes the hardware configuration.

[0069] In this embodiment, based on the rated power and rated capacity of the energy storage system determined in the above steps, a vanadium redox flow battery, whose power and capacity are physically completely decoupled, is used for engineering hardware implementation. The specific configuration steps are as follows: (1) Selection of core energy storage carrier and adaptation to operating conditions: For high-frequency fluctuation and long-term overload conditions, vanadium redox flow battery is selected as the electrochemical energy storage carrier; the selected core technical parameters meet the following requirements: cycle life Subsequent discharge depth (DoD) System energy efficiency Power regulation response time ms.

[0070] Meanwhile, the system has a wide operating temperature range ( It is fully adaptable to the extreme climate conditions of the target application area (such as Huaisu area in Anhui Province) to ensure operational reliability throughout the entire life cycle.

[0071] (2) Effective capacity calculation is based on the aforementioned calculations. Rated capacity of energy storage system and Based on the preset discharge depth, effective energy boundary calculations are performed; the actual effective energy storage capacity under this configuration is obtained as follows: Under the premise of meeting the rated full-load power output, this effective capacity can provide continuous power supply support for approximately 0.5225 hours, and its energy reserves are sufficient to cover the line switching and fault repair window under the extreme N-1 conditions of most substations.

[0072] (3) Parameter decoupling and standardized modular networking fully utilize the characteristics of flow batteries and adopt a design architecture of "unit-based splitting and modular networking" for hardware deployment: Power-side mapping: 11 optional Standardized fuel cell stack units are achieved through parallel networking. The rated active power output; One unit with a rated power of [missing information] will be provided simultaneously. A bidirectional power converter (PCS) is used to reserve sufficient power regulation margin.

[0073] Capacity-side mapping: A separate energy storage electrolyte tank with a total volume of 60 cubic meters is configured, and a closed-loop design is adopted to store positive and negative electrolytes separately and supply them in groups, providing abundant and stable active material circulation support for the large-scale stack network.

[0074] (4) Relying on the "intrinsic safety" physical characteristics of the aqueous electrolyte of the vanadium redox flow battery, the complex fire extinguishing and explosion-proof system of traditional lithium-ion batteries is eliminated, thereby greatly optimizing the initial investment cost of safety monitoring equipment (the total investment in safety equipment after optimization is about RMB140,000). The system is designed with three core protection subsystems to address its unique operating mechanism: electrolyte medium safety module (such as leak detector and emergency shut-off valve), fuel cell stack operation status monitoring module (such as high-precision temperature / current sensor), and electrical fault prevention module (such as centralized monitoring system).

[0075] Through the above-mentioned streamlined and efficient security architecture, the risk of thermal runaway is controlled and reduced from the underlying principle.

[0076] S4. After the configured electrochemical energy storage system is connected to the substation and the outdoor layout is determined, a multi-dimensional evaluation model is used to evaluate and compare multiple schemes to determine the optimal configuration scheme.

[0077] First, the grid connection scheme was selected. Comparing three connection methods—connecting the energy storage system to the low-voltage side bus of the substation, outgoing lines for important users, and emergency bus—and considering factors such as overload mitigation effectiveness and ease of operation and maintenance, the 10kV low-voltage side bus connection method was chosen. This 10kV low-voltage side bus connection method connects the energy storage system to the connection node between the 10kV low-voltage side bus and the photovoltaic grid connection point, as well as the local load. It can directly absorb peak redundant power from the photovoltaic system and share peak main transformer power, reducing the main transformer load rate from the source. Furthermore, it features short cable laying distances, low power transmission losses, and can share existing 10kV distribution and maintenance resources in the substation, reducing engineering and operation and maintenance costs. The outdoor layout of the energy storage units was then determined. The energy storage system adopts a linear outdoor layout, with the fuel cell stack units, electrolyte storage tanks, and combiner switch stations arranged independently according to functional zones; the fuel cell stack units are arranged in double rows, with reserved space for maintenance and heat dissipation; the electrolyte storage tanks are arranged separately on one side of the fuel cell stack units to avoid electrolyte leakage affecting core equipment; the combiner switch stations are located at the end of the system to achieve centralized power collection and fault protection; Finally, a multi-dimensional evaluation model was used to compare and analyze different options to determine the final configuration scheme, as follows: First, a three-dimensional multi-solution comparative analysis was conducted. Three substation overload solutions were selected: transformer capacity expansion, 12.5MW / 6.25MWh lithium battery energy storage, and 11MW / 5.5MWh vanadium redox flow battery energy storage. A quantitative comparison was carried out from three dimensions: full life-cycle economic indicators, operational indicators, and safety indicators. The core comparison results are as follows: Lifecycle economic indicators: The lifecycle economic indicators use total lifecycle cost as the evaluation indicator. The calculation method is that the lifecycle cost equals the sum of the initial investment cost, 20-year operation and maintenance cost, equipment replacement cost, and decommissioning cost. The initial investment cost includes equipment purchase cost, installation engineering cost, construction engineering cost, and other expenses. The operation and maintenance cost includes the cost of daily equipment inspection and loss compensation. The equipment replacement cost includes equipment purchase, dismantling and installation, and downtime losses. The decommissioning cost includes environmental treatment and residual value recovery costs.

[0078] The initial investment for transformer capacity expansion is 7.14 million yuan, including equipment costs of 6.11 million yuan and construction costs of 1.03 million yuan. This plan has no equipment replacement requirement during its 20-year operating cycle. Annual maintenance costs are calculated at 1% of the initial investment, approximately 70,000 yuan, totaling 1.4 million yuan. Decommissioning costs are negligible. The total life-cycle cost over 20 years is 8.54 million yuan. Lithium-ion battery energy storage is limited by the cycle life of lithium iron phosphate batteries. This solution requires two battery system replacements within a 20-year operating cycle, at the 7th and 14th years. Each replacement costs approximately 6.36 million yuan, including approximately 4.41 million yuan for battery pack purchase, approximately 1.2 million yuan for downtime losses, and approximately 750,000 yuan for the environmental disposal of old batteries during retirement. The annual operation and maintenance cost is calculated at 1% of the battery purchase cost, approximately 44,100 yuan, totaling 880,000 yuan over 20 years. The total life cycle cost over 20 years is 13.6 million yuan (including the costs of two equipment replacements, downtime, and waste disposal). The initial investment for vanadium redox flow battery energy storage is 10.9 million yuan, including approximately 9.2 million yuan for equipment purchase (including stack units, electrolyte storage tanks, bidirectional converters, and circulation systems), approximately 800,000 yuan for installation, approximately 500,000 yuan for construction, and approximately 400,000 yuan for other expenses. The vanadium redox flow battery stack has a cycle life of over 10,000 cycles, and the electrolyte can be recycled and reused for a long time. There is no need to replace core equipment during the 20-year operating cycle. During the decommissioning stage, the electrolyte can be recycled and reused, with a disposal cost of approximately 620,000 yuan. The annual operation and maintenance cost is approximately 90,000 yuan, totaling 1.8 million yuan over 20 years. The total life cycle cost over 20 years is 13.32 million yuan (excluding core equipment replacement costs). Operational indicators: Transformer capacity expansion can only solve the problem of main transformer overload, but cannot improve power quality and photovoltaic absorption. Both lithium battery energy storage and vanadium redox flow battery energy storage can achieve overload control, photovoltaic absorption, and voltage fluctuation smoothing. Among them, vanadium redox flow battery has stronger reactive power support capability (power factor -0.9 to +0.9) and can better help suppress harmonic pollution in the distribution network.

[0079] To verify the effectiveness of the 11MW / 5.5MWh vanadium redox flow battery energy storage system configured in this scheme, multi-dimensional quantitative verification was conducted with the core optimization indicators of reducing curtailment rate (improving curtailment rate), improving main transformer load rate, and increasing energy storage utilization. The definitions, calculation methods, and statistical methods of each indicator are as follows: (1) Reduce the light rejection rate The substation has a total photovoltaic (PV) installed capacity of 154.75MW, with an annual effective utilization of 1200 hours and a theoretical annual PV power generation of 185,700MWh. Before the energy storage system was installed, constrained by the main transformer's allowable reverse overload power limit of 54MW, 20.44MW of PV power was curtailed during the peak PV generation periods in spring and autumn, with a daily curtailment duration of 3 hours, resulting in 84 effective curtailment days per year and an annual curtailment volume of approximately 5170MWh, with a calculated curtailment rate of approximately 2.78%. The 11MW / 5.5MWh vanadium redox flow battery energy storage system can fully absorb the excess PV power beyond the main transformer's reverse transmission capacity during the peak PV generation periods at midday, absorbing 5.5MWh of excess PV power daily, adding 462MWh of PV power absorption annually, reducing the annual curtailment volume to 4708MWh, and lowering the curtailment rate to 2.54%. This solution can effectively eliminate the curtailment problem caused by main transformer overload constraints. Under the existing energy storage capacity configuration, it has achieved an effective improvement in the curtailment level and reduced the regional photovoltaic consumption loss.

[0080] (2) Improve the degree of improvement in the main transformer load rate After the energy storage was integrated, the load factor under N-1 operating condition decreased from 122.2% to 100%, meeting the requirements of the "Guidelines for the Safety and Stability of Power Systems". The constraints for the N-1 operating condition are met, and both forward and reverse overloads meet the constraints.

[0081] (3) Improve energy storage utilization rate Energy storage utilization rate is defined as the ratio of the total annual charge and discharge capacity to the theoretical maximum output capacity of the rated capacity, as shown in the following formula: , In the formula, For energy storage utilization rate; The effective charging and discharging capacity of the energy storage system during the statistical period is 3650MWh (unit: MWh, which is the charging and discharging capacity that meets the needs of main transformer over-limit management and photovoltaic power consumption, excluding maintenance charging and discharging operations; in this case, it is 3650MWh). The rated capacity of the energy storage system is (unit: MWh, 5.5MWh in this scheme); the calculated energy storage utilization rate is 7.6%. Considering additional scenarios such as emergency power supply and voltage regulation in actual operation, the total annual charging and discharging capacity is much higher than 3650MWh, which meets the constraint target of high utilization rate, with no idle resources, and the capacity configuration is highly compatible with the operating scenario.

[0082] Safety indicators: Transformer capacity expansion poses no fire risk, but there are potential hazards such as oil leakage and insulation failure due to equipment aging; the sole cause of lithium battery fires is thermal runaway, and the probability of fire triggering is... (times / ten thousand clusters) (The lower the annual value, the safer it is); the energy storage (electrolyte) and conversion (stacking) of vanadium redox flow batteries are completely separated. There are no solid active materials in the stack. The charge and discharge reaction is a reversible ionic reaction. There is no basis for an exothermic chain reaction. Industry tests and standards have clearly shown that there is no typical risk of thermal runaway and it is not easy to cause a fire.

[0083] In this embodiment, the fire triggering probability formula for the all-vanadium redox flow battery is expressed as: .

[0084] For lithium battery energy storage systems, the fire risk originates from thermal runaway and its propagation process. Based on the "Guidelines for Power Grid Safety Risk Assessment (DL / T1855-2018)," the following model is used for calculation: , The probability of thermal runaway under normal operating conditions (times / 10,000 clusters) (Year), in this embodiment, it is taken as 0.02; The overload risk amplification factor is set to 1.8 in this embodiment; The failure factor of the temperature control system is 1.2 in this embodiment.

[0085] In this embodiment, based on industry white papers, national standards, and regional operating environment data, the values ​​of key influencing parameters are analyzed as follows: According to the "White Paper on the Development of China's Energy Storage Industry (2024)" and the "Safety Technical Specification for Lithium-ion Batteries for Power Storage (GB / T36547-2018)," the measured thermal runaway probability of large-scale lithium iron phosphate energy storage power stations in China is 0.01~0.03 times / 10,000 clusters. Year, take the median value It is 0.02.

[0086] When the main transformer is overloaded, energy storage requires a high-rate charge / discharge of 1C. The internal temperature rise of the lithium battery is 25-30°C higher than that under the conventional 0.5C condition. According to the "Research on Thermal Runaway Characteristics of High-Rate Lithium-ion Batteries", the probability of thermal runaway increases by 1.5-2 times when the rate is increased to 1C. Taking the middle value... It is 1.8.

[0087] Anhui has a subtropical monsoon climate, with extreme summer temperatures reaching 38-40℃. Outdoor substation energy storage temperature control systems operate at full load, and the failure probability is 20% higher than in areas with normal temperatures. Referring to the "Anhui Power Grid Summer Equipment Operation White Paper (2023)," the failure probability of outdoor energy storage temperature control systems increases by 15%-25% in summer; taking the midpoint... It is 1.2.

[0088] The calculated probability of thermal runaway fire in lithium battery energy storage is 0.0432, indicating a fire risk and requiring the configuration of complex fire and explosion protection devices.

[0089] Taking into account the probability of fire triggering, fault propagation capability, and protection fault tolerance rate, a weighted safety factor is calculated. The weights are then combined with the substation scenario to obtain the safety factor S. The formula for the battery safety factor is: , In the formula, For battery safety factor, This represents the probability of fault propagation. To protect the system's fault tolerance, As the fire triggering probability weight, For the probability weight of fault propagation, To protect against errors, a fault tolerance weight is used.

[0090] The key influencing parameters are analyzed as follows: The above weight allocation follows the risk factor weight setting principles of the "Guidelines for Power Grid Safety Risk Assessment (DL / T1855-2018)": Fire trigger probability weight Substation fires can directly cause main transformer tripping and photovoltaic grid connection interruption, affecting the entire new district power grid. This is classified as a Level 1 safety risk, with the highest weight. Fault propagation probability weight The spread of the fault will expand the scope of the accident, but it can be mitigated through zone isolation, which is classified as a level 2 safety risk; Protection tolerance weight The failure of the protection system is a "passive risk," which can be reduced through regular maintenance. It belongs to the third level of security risk and has the lowest weight.

[0091] The substation's energy storage system features a compact layout, with a fire-prevention distance of only 3 meters between clusters, lower than the 5-meter requirement of the national standard GB / T36547-2018. According to the "Experimental Study on Thermal Runaway Propagation Characteristics of Energy Storage Power Stations," the probability of thermal runaway propagation is 70% when the cluster spacing is 3 meters. Take 0.7.

[0092] Substation energy storage commonly employs a high-precision BMS (Body Management System) + heptafluoropropane gas fire suppression system, which is a mainstream protection scheme for power energy storage. According to the "Safety Regulations for Power Energy Storage Power Stations (DL / T2505-2024)," the average fault tolerance rate of this protection scheme is 70%~80%, taking the midpoint. The value is 0.75. The calculated safety factor for lithium iron phosphate batteries is:

[0093] Vanadium redox flow batteries require only leak-proof dikes and corrosion-resistant piping, without complex electronic protection systems. Failure points are limited to pipe joints or the pump body, and the annual failure probability of the vanadium redox flow battery protection system is <1%, with a fault tolerance rate >98%. To meet project requirements, the safety factor of the vanadium redox flow battery is calculated to be 0.99.

[0094] After comprehensive quantitative comparison, the vanadium redox flow battery energy storage system was determined to be the optimal solution for the 110kV substation in the Huaisu area under the scenario of high proportion of new energy access.

[0095] Finally, a comprehensive evaluation of the configuration effect was conducted. The 11MW / 5.5MWh vanadium redox flow battery energy storage system configured in this invention, after being applied to a 110kV substation in the Huaisu area, achieved the following engineering effects: Effective management of main transformer overload: Under N-1 operating conditions, the main transformer load rate is reduced from 122.2% to below the safe range. Both forward and reverse overloads are controlled within the safe operating range, effectively preventing main transformer insulation aging and malfunction of protection devices. Reduced curtailment rate: The energy storage system can absorb 5MWh / day of redundant photovoltaic power during peak photovoltaic periods, reducing the annual curtailment of photovoltaic power to 4708MWh, effectively reducing the curtailment rate of substations; Power quality improvement: By absorbing redundant photovoltaic power and smoothing out photovoltaic output fluctuations, the voltage fluctuation of the 110kV bus is controlled within ±2%, with no additional harmonics generated, thus helping to improve the overall power quality of the distribution network. Energy storage utilization rate meets the target: The energy storage system can achieve cyclical operation of summer peak load discharge and spring and autumn photovoltaic peak charging, with a total annual charge and discharge capacity of 3650MWh and an energy storage utilization rate of 7.6%. Considering additional scenarios such as emergency power supply and voltage regulation in actual operation, the total annual charge and discharge capacity is much higher than 3650MWh, meeting the constraint target of high utilization rate, with no idle resources, and the capacity configuration is highly adapted to the operating scenario.

[0096] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0097] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0098] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0100] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0101] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for configuring a substation energy storage system with a high proportion of new energy access, characterized in that, include: Collect operational data and identify overload samples of the main transformer, and extract the overload baseline duration that represents the risk distribution through probabilistic modeling; Based on the overload reference duration, power balance calculations are performed under typical operating conditions to determine the rated power and rated capacity of the energy storage system. Based on the rated power, rated capacity and depth of discharge of the energy storage system, the parameters of the electrochemical energy storage system are selected and the hardware configuration is completed. After the configured electrochemical energy storage system is connected to the substation and its outdoor layout is determined, a multi-dimensional evaluation model is used to evaluate and compare multiple schemes to determine the optimal configuration scheme.

2. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The process of identifying overload samples of the main transformer includes: The load rate threshold is set based on the insulation heat resistance level of the main transformer. Traverse the running data and extract continuous time segments where the load rate exceeds the load rate threshold as valid overload samples; Calculate the difference between the start time and the recovery time for each sample to construct an overload duration sample.

3. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 2, characterized in that, The overload baseline duration representing the risk distribution is extracted through probabilistic modeling, including: The overload duration sample is fitted with a Weibull distribution, wherein the shape and scale parameters of the Weibull distribution are solved by the maximum likelihood estimation method to construct the probability density function. The validity of the probability density function was verified using distribution tests and goodness-of-fit indices.

4. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The typical operating modes include: substation summer peak load forward overload, spring and autumn photovoltaic peak reverse overload, and N-1 fault condition.

5. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The calculation of the rated power of the energy storage system includes: Calculate the forward overload power gap under the summer peak load operation mode, the reverse redundancy power under the spring and autumn photovoltaic peak operation mode, and the rated compensation power under the N-1 fault condition. By comparing the positive overload power gap, the reverse redundant power, and the rated compensation power, the maximum value among them is selected as the original power reference value. The rated power of the energy storage system is obtained by multiplying the original power reference value by the preset engineering safety margin coefficient.

6. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The rated capacity of the energy storage system is obtained by multiplying the rated power of the energy storage system by the overload reference duration.

7. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The determination of the rated power and rated capacity of the energy storage system also includes: By using the product of the main transformer capacity scaling factor and the new energy penetration rate correction factor, the rated power and rated capacity of the energy storage system are scaled proportionally to obtain the rated parameters of the energy storage system after adaptive matching of the target substation. The main transformer capacity scaling factor is determined based on the ratio of the rated capacity of the main transformers of the target substation to that of the benchmark substation. The new energy penetration rate correction factor is determined by multiplying the difference in new energy penetration rate between the target substation and the benchmark reference substation by 1 and adding 1 to the product of the preset penetration rate sensitivity coefficient.

8. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The hardware configuration includes: The effective capacity is calculated based on the rated capacity and depth of discharge of the vanadium redox flow battery, and the vanadium ion concentration and tank volume are determined by combining Faraday's law.

9. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 1, characterized in that, The multidimensional evaluation model includes security indicators; The safety indicators include the battery safety factor, which is calculated using the following formula: ， In the formula, This represents the probability of a fire being triggered. For battery safety factor, This represents the probability of fault propagation. To protect the system's fault tolerance, As the fire triggering probability weight, For the probability weight of fault propagation, To protect against errors, a fault tolerance weight is used.

10. The method for configuring a substation energy storage system with a high proportion of new energy access according to claim 9, characterized in that, The formula for the fire triggering probability is as follows: ， In the formula, This represents the probability of a fire being triggered. This represents the probability of thermal runaway. This is a factor that amplifies the risk of heavy overload. This represents the failure factor of the temperature control system.

Images