A hybrid energy storage double-layer optimization configuration method for reducing frequency modulation loss of a thermal power unit

Abstract

The application discloses a hybrid energy storage double-layer optimization configuration method for reducing frequency modulation loss of a thermal power unit, belongs to the technical field of hybrid energy storage of thermal power units, and designs a hybrid energy storage configuration double-layer model, wherein a bottom-layer model is composed of a hybrid energy storage control strategy considering energy storage state of charge recovery and a thermal power-hybrid energy storage frequency modulation response model, and provides each frequency modulation resource output and energy storage SOC state for an upper-layer model; the upper-layer model is a multi-objective optimization configuration model with the minimum difference between cost and benefit in a full life cycle and the maximum AGC comprehensive evaluation index Kp, and provides energy storage power and energy constraints for the control strategy of the bottom-layer model. A distance method multi-objective particle swarm algorithm is introduced to solve the double-layer model, and the contingency of a traditional multi-objective particle swarm algorithm in determining a global optimum in a Pareto solution set is avoided. The application can improve the frequency modulation performance of the unit, reduce the frequency modulation loss of the thermal power unit, and realize the double-layer optimization configuration method of the hybrid energy storage system for reducing the full cycle life cost of the energy storage.

Description

Technical Field

[0001] This invention belongs to the field of hybrid energy storage technology for thermal power units, and in particular relates to a method for optimizing the configuration of a hybrid energy storage dual-layer system to reduce frequency regulation losses in thermal power units. Background Technology

[0002] In new power systems, the proportion of renewable energy generation is increasing. However, the intermittency and volatility of renewable energy sources impact the frequency stability of the power system, making it increasingly difficult for traditional thermal power units to meet frequency regulation requirements. Compared to thermal power units, electrochemical energy storage offers faster, more precise, and adjustable capacity. On average, the frequency regulation effect of electrochemical energy storage is more than 20 times that of thermal power units. Assisting thermal power units in frequency regulation can effectively improve unit frequency regulation performance, increase generator operating efficiency, and thus enhance the absorption of renewable energy. However, determining a reasonable configuration that balances the excellent frequency regulation characteristics and high cost of electrochemical energy storage is crucial for its large-scale application. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a hybrid energy storage dual-layer optimization configuration method for reducing frequency regulation losses of thermal power units, overcoming the shortcomings of the prior art, and providing a scientific, reasonable, and highly applicable hybrid energy storage system dual-layer optimization configuration method that can improve the frequency regulation performance of the unit, reduce the frequency regulation losses of thermal power units, and ultimately reduce the life cycle cost of energy storage.

[0004] A method for optimizing the configuration of a hybrid energy storage dual-layer system to reduce frequency regulation losses in thermal power units includes the following steps, which are performed sequentially.

[0005] Step 1: Construct a two-layer model for optimal energy storage configuration

[0006] The two-layer model consists of a bottom layer model and an upper layer model. The bottom layer model includes a hybrid energy storage control strategy that takes into account the recovery of the energy storage state of charge and a thermal power-hybrid energy storage frequency regulation response model, which provides the upper layer model with the output of each frequency regulation resource and the energy storage SOC state. The upper layer model is an optimization configuration model with the goal of minimizing the cost-benefit difference and maximizing the AGC comprehensive evaluation index Kp over the entire life cycle, which provides the energy storage power and energy constraints for the control strategy of the bottom layer model.

[0007] Step 2: Constructing the constraints of a two-layer model for optimal energy storage configuration.

[0008] The constraints of the underlying model include battery charge and discharge power constraints, power balance constraints, energy storage SOC constraints, and energy storage power constraints; the constraints of the upper-level model are the rated power and rated capacity constraints of the lithium-ion battery.

[0009] The method for establishing the objective function of the underlying model in step one is as follows: first, the plant-level AGC instructions are decomposed and reconstructed into n thermal power-hybrid energy storage instruction allocation schemes through EMD; second, each allocation scheme is input into the underlying model to solve for the output of each frequency regulation resource and the SOC curve of the hybrid energy storage.

[0010] The method for establishing the objective function of the upper-level model in step one is to set two objective functions: the difference between cost and benefit over the entire life cycle of hybrid energy storage and the Kp index.

[0011] Costs include hybrid life-cycle costs, costs associated with thermal power-hybrid energy storage AGC regulation, and other costs. Benefits include direct benefits (f1) from AGC frequency regulation mileage and indirect benefits (f2) from reduced system coal costs and power generation emissions costs.

[0012]

[0013] f2=Kp (2)

[0014] In the formula C inv For hybrid energy storage investment costs; C OM For hybrid energy storage operating costs; C scr C is the cost of disposing of lithium batteries at the end of their service life; ram and C pun These represent the frequency regulation loss cost and the regulation offset penalty cost of thermal power units, respectively; Bkp represents the AGC frequency regulation mileage revenue; B coal To reduce the indirect benefits of lowering the system's coal-fired costs; B env This is an indirect benefit to reduce the system's power generation and pollution discharge costs.

[0015] Solving the objective function of the upper-level model using a multi-objective particle swarm optimization algorithm includes the following steps: Step 1, converting all indices into extremely large indices and standardizing them.

[0016] In the upper-level model, objective function 1 is the difference between frequency regulation cost and frequency regulation revenue over the entire life cycle, which is a very small indicator; objective function 2 is the Kp indicator, which is a very large indicator. Since the objective function types are inconsistent, they need to be transformed into very large indicators and normalized.

[0017] Objective function 1:

[0018]

[0019] Objective function 2:

[0020]

[0021] Step 2: Standardize the normalized matrix formed in Step 1. The normalized matrix is ​​as follows:

[0022]

[0023] The standardized formula is as follows:

[0024]

[0025] The maximum value is defined as the distance between the indicator and its maximum and minimum values, and the indicator's fit.

[0026]

[0027] Define the minimum value:

[0028]

[0029] Distance between evaluation metric and maximum value:

[0030]

[0031] Distance between evaluation index and minimum value:

[0032]

[0033] The degree of fit between the individual particles is as follows:

[0034]

[0035] Calculate the fit degree Si of each particle, and take the particle with the largest Si as the global optimal value.

[0036] The battery charging and discharging power constraints of the underlying model in step two are reflected by a double Logistic function, which uses a double Logistic function to reflect the output constraints of the battery energy storage under different SOCs. The function expression is as follows:

[0037]

[0038] When SOC < 0.1 or SOC ≥ 0.9, K(SOC) = 0; where P0 is 0.01; K max Take 1; n, which describes the rate of change of the curve, is taken as 15; the energy storage SOC is defined as follows:

[0039]

[0040] In the formula, E rated P represents the rated capacity of the energy storage. b This refers to the energy storage and discharge power of the lithium battery; the SOC range is between 0 and 1.

[0041] The power balance constraint is P A =P g +P b +P fw (14)

[0042] In the formula PA For AGC power commands, P g For the thermal power unit to output power, P fw It provides energy storage and power to the flywheel;

[0043] Energy storage SOC constraint is

[0044]

[0045] In the formula, SOC L and SOC H These are the maximum and minimum values ​​of SOC, respectively, taking values ​​of 0.1 and 0.9; SOC b (t) and SOC fw (t) represents the SOC of battery energy storage and flywheel energy storage at time t, respectively;

[0046] Energy storage power constraints include lithium battery power constraints and flywheel energy storage power constraints. Lithium battery power constraints...

[0047] -P B.rated ×K(SOC b ) < P b <P B.rated ×K(SOC b (16)

[0048] Flywheel energy storage power constraints

[0049] -P FW.rated <P fw <P FW.rated (17)

[0050] In the formula, P B.rated and P FW.rated These are the rated power of battery energy storage and flywheel energy storage, respectively.

[0051] The upper-level model constraints in step two are: rated power and rated capacity constraints for lithium-ion batteries.

[0052] P B.rated ≥0 (18)

[0053]

[0054] The power and capacity of a flywheel energy storage system have a fixed relationship based on the charging and discharging time.

[0055] Through the above design scheme, this invention can bring the following beneficial effects: a hybrid energy storage two-layer optimization configuration method to reduce frequency regulation losses of thermal power units. First, the automatic generation control command is decomposed and reconstructed into a command allocation scheme for thermal power units and hybrid energy storage systems through empirical mode decomposition, so as to reduce the frequency regulation mileage of thermal power units. The key design is a hybrid energy storage configuration two-layer model. The bottom layer model consists of a hybrid energy storage control strategy considering the recovery of the energy storage state of charge and a thermal power-hybrid energy storage frequency regulation response model, providing the upper layer model with the output of each frequency regulation resource and the energy storage SOC state. The upper layer model is a multi-objective optimization configuration model that minimizes the cost-benefit difference over the entire life cycle and maximizes the AGC comprehensive evaluation index Kp, providing the energy storage power and energy constraints for the control strategy of the bottom layer model. Then, a multi-objective particle swarm optimization algorithm based on the superior-inferior solution distance method is introduced to solve this two-layer model, avoiding the randomness of traditional multi-objective particle swarm optimization algorithms in determining the global optimum in the Pareto solution set. This invention can improve the frequency regulation performance of the unit, reduce the frequency regulation loss of the thermal power unit, and finally realize a two-layer optimized configuration method for hybrid energy storage system that reduces the life cycle cost of energy storage. Attached Figure Description

[0056] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0057] Figure 1 This is a schematic diagram of a hybrid energy storage dual-layer optimization configuration method to reduce frequency regulation losses in thermal power units.

[0058] Figure 2 This is a schematic diagram of a hybrid energy storage dual-layer optimized configuration method for reducing frequency regulation losses in thermal power units, representing a thermal power AGC regulation structure. Detailed Implementation

[0059] The following description, using the accompanying drawings and embodiments, further illustrates the dual-layer optimized configuration method for the hybrid energy storage system of the present invention.

[0060] Reference Figure 1 and Figure 2 The present invention proposes a two-layer optimized configuration method for a hybrid energy storage system to reduce frequency regulation losses in thermal power units, comprising the following:

[0061] 1) Construct a two-layer model for optimal energy storage configuration

[0062] ① The underlying model consists of a hybrid energy storage control strategy that takes into account the recovery of the energy storage state of charge (SOC) and a thermal power-hybrid energy storage frequency regulation response model, which provides the upper-level model with the output of each frequency regulation resource and the SOC state of the energy storage.

[0063] ②The upper-level model is a multi-objective optimization configuration model that minimizes the cost-benefit difference over the entire life cycle and maximizes the AGC comprehensive evaluation index Kp, providing energy storage power and energy constraints for the control strategy of the lower-level model.

[0064] 2) Optimize the objective function of the model

[0065] (a) Objective function of the underlying model

[0066] The underlying model consists of a hybrid energy storage control strategy that takes into account the recovery of the energy storage's State of Charge (SOC) and a thermal power-hybrid energy storage frequency regulation response model. First, the plant-level AGC commands are decomposed and reconstructed into n thermal power-hybrid energy storage command allocation schemes using EMD. Then, each allocation scheme is input into the underlying model to solve for the output of each frequency regulation resource and the hybrid energy storage SOC curves.

[0067] (b) Objective function of the upper-level model

[0068] The upper-level model for hybrid energy storage optimization configuration sets two objective functions: the cost-benefit difference over the entire life cycle of hybrid energy storage and the Kp index. The costs include the total life cycle cost of hybrid energy storage, and other costs associated with thermal power-hybrid energy storage AGC regulation. The benefits include the direct benefits of AGC frequency regulation mileage and the indirect benefits of reducing system coal-fired power costs and power generation pollution costs.

[0069]

[0070] f2=Kp (2)

[0071] In the formula C inv For hybrid energy storage investment costs; C OM For hybrid energy storage operating costs; C scr C is the cost of disposing of lithium batteries at the end of their service life; ram and C pun These are the frequency regulation loss cost and the regulation offset power penalty cost of thermal power units, respectively; B kp For AGC frequency modulation mileage gain; B coal To reduce the indirect benefits of lowering the system's coal-fired costs; B env This is an indirect benefit to reduce the system's power generation and pollution discharge costs.

[0072] 3) Optimization model solution method

[0073] The multi-objective particle swarm optimization algorithm is used to solve the problem, and the best-inferior solution distance method (TOPSIS method) is used to find the global optimal solution in the Pareto solution set.

[0074] ① Convert all indicators into extremely large indicators and standardize them.

[0075] In the upper-level model, objective function 1 is the difference between frequency regulation costs and revenues over the entire life cycle; a smaller value is better, making it a very small indicator. Objective function 2 is the Kp indicator; a larger value is better, making it a very large indicator. Because the objective functions are of different types, they need to be converted into very large indicators and normalized.

[0076] Objective function 1:

[0077]

[0078] Objective function 2:

[0079]

[0080] ② Standardize the positive matrix formed in the first step.

[0081] The normalized matrix is:

[0082]

[0083] The standardized formula is as follows:

[0084]

[0085] Calculate the distance between the indicator and its maximum and minimum values, and the indicator's fit.

[0086] Define the maximum value:

[0087]

[0088] Define the minimum value:

[0089]

[0090] Distance between evaluation metric and maximum value:

[0091]

[0092] Distance between evaluation index and minimum value:

[0093]

[0094] The degree of fit between the individual particles is as follows:

[0095]

[0096] Calculate the fit degree Si of each particle. The larger the value, the better. Select the particle with the largest Si as the global optimal value.

[0097] 4) Constraints

[0098] (a) Constraints of the underlying model

[0099] ① Battery charging and discharging power constraints

[0100] To avoid battery life loss due to overcharging and over-discharging, a dual logistic function is used to reflect the output constraint of the battery under different SOCs. The function expression is as follows:

[0101]

[0102] When SOC < 0.1 or SOC ≥ 0.9, K(SOC) = 0. In the formula, P0 is taken as 0.01; K max Take 1; n, which describes the rate of change of the curve, is taken as 15; the energy storage SOC is defined as follows:

[0103]

[0104] In the formula, E rated P represents the rated capacity of the energy storage. b This refers to the energy storage and discharge power of the lithium battery; the SOC range is between 0 and 1.

[0105] ② Power balance constraints

[0106] P A =P g +P b +P fw (14)

[0107] In the formula P A For AGC power commands, P g For the thermal power unit to output power, P fw Powering flywheel energy storage

[0108] ③ Energy storage SOC constraint

[0109]

[0110] In the formula, SOC L and SOC H These are the maximum and minimum values ​​of SOC, respectively, taking values ​​of 0.1 and 0.9; SOC b (t) and SOC fw (t) represents the SOC of battery energy storage and flywheel energy storage at time t, respectively.

[0111] ④ Energy storage power constraints

[0112] Lithium battery power constraints

[0113] -P B.rated ×K(SOC b ) < P b <P B.rated ×K(SOC b (16)

[0114] Flywheel energy storage power constraints

[0115] -P FW.rated <P fw <P FW.rated (17)

[0116] In the formula, P B.rated and P FW.rated These are the rated power of battery energy storage and flywheel energy storage, respectively.

[0117] (b) Constraints of the upper-level model

[0118] The discharge timescale of lithium-ion batteries is from minutes to hours; therefore, the constraints on capacity configuration are as follows:

[0119] Lithium-ion battery rated power and rated capacity constraints:

[0120] P B.rated ≥0 (18)

[0121]

[0122] Depending on the specific product's charging and discharging time, the power and capacity of a flywheel energy storage system have a fixed ratio.

[0123] Simulations were conducted using typical daily plant-level AGC command data from a 300MW unit at a power plant in North China, with hybrid energy storage installed on the side of the thermal power unit.

[0124] The AGC signal was decomposed and reconstructed into nine schemes for hybrid energy storage capacity configuration verification and economic analysis.

[0125] 1. The hybrid energy storage optimization configuration model proposed in this paper is used for configuration, and the results are shown in Table 1. Then, the TOPSIS method is used to select Scheme 3 as the best scheme.

[0126] 2. The same method was used to optimize the configuration of a single energy storage system. The results are shown in Table 2 and compared with those without energy storage configuration and without SOC recovery strategy.

[0127] Table 1 Configuration Results of Each Scheme

[0128]

[0129] Table 2 Results of configurations with single flywheel, single battery, hybrid energy storage, and no energy storage

[0130]

[0131] The results of each energy storage configuration are compared with those of the configuration without energy storage, as shown in Table 3 below.

[0132] Table 3 Comparison of various economic indicators for single flywheel, single battery, hybrid energy storage, and no energy storage configurations.

[0133]

[0134] Table 4 shows the unit investment cost of hybrid energy storage when the net income is equal to that of no energy storage configuration.

[0135]

[0136]

[0137] As can be seen from the results in Table 3:

[0138] 1) Due to the high cost of hybrid energy storage systems, the net profit of the configuration without energy storage is still higher than that of the other configurations, even though the Kp frequency regulation benefit is lower than that of the first four energy storage configurations. This indicates that the high cost of energy storage remains the main obstacle to its development. Under the same hybrid energy storage configuration, when the cost of lithium-ion batteries and flywheel energy storage is reduced to the levels shown in Table 4, the net profit is the same as that of the configuration without energy storage.

[0139] 2) Compared to strategies that only restrict energy storage output, the total cost of frequency regulation over its entire lifecycle decreased by RMB 282 million. Although the total revenue from frequency regulation decreased by RMB 4.36 million, the net profit over its entire lifecycle increased from RMB -78 million to RMB 200 million. The main reason for this is that the strategy of only restricting energy storage output requires more energy storage while improving the frequency regulation Kp index, and the excessively high investment cost of hybrid energy storage makes it unprofitable over its entire lifecycle.

[0140] 3) In a single-battery configuration, battery replacement costs account for 65% of the total cost of frequency regulation over its entire lifespan. Compared to a single-battery configuration, the hybrid energy storage configuration increases the lifespan of lithium batteries by approximately 5 times, effectively reducing battery replacement costs. In a single-flywheel configuration, initial investment costs account for 78% of the total cost of frequency regulation over its entire lifespan. The lower investment cost of lithium batteries in hybrid energy storage reduces the total cost of the hybrid energy storage configuration by 45 million yuan compared to the single-flywheel configuration.

[0141] 4) According to Table 3, since the energy storage system undertakes the high-frequency frequency regulation demand, the output of thermal power units is slow. The frequency regulation loss cost of thermal power units without energy storage configuration is 17 times that of mixed energy storage frequency regulation cost and about 12 times that of single energy storage configuration.

[0142] The hybrid energy storage capacity configuration method proposed in this paper, based on the above analysis, greatly reduces the frequency regulation loss of thermal power units and improves the economic efficiency of the thermal power-hybrid energy storage joint frequency regulation system; it also improves battery life and enhances economic efficiency while improving the Kp index.

[0143] At current energy storage costs, the net benefit of no energy storage configuration is higher than that of a thermal power unit frequency regulation system with energy storage. As energy storage costs continue to decrease, the economics of energy storage-assisted thermal power unit frequency regulation will improve in the future. The calculation conditions, figures, tables, etc., in the embodiments of this invention are only for further explanation of the invention and are not exhaustive, nor do they constitute a limitation on the scope of the claims. Those skilled in the art, based on the inspiration gained from the embodiments of this invention, can conceive of other substantially equivalent alternatives without inventive effort, all of which are within the scope of protection of this invention.

Claims

1. A method for optimizing the configuration of a hybrid energy storage dual-layer system to reduce frequency regulation losses in thermal power units, characterized by: The steps are as follows, and the steps are performed in sequence. Step 1: Construct a two-layer model for optimal energy storage configuration The two-layer model consists of a bottom-layer model and an upper-layer model. The bottom-layer model includes a hybrid energy storage control strategy that takes into account the recovery of the energy storage state of charge (SOC) and a thermal power-hybrid energy storage frequency regulation response model, providing the upper-layer model with the output of each frequency regulation resource and the SOC state of the energy storage. The upper-layer model is an optimization configuration model that aims to minimize the cost-benefit difference over the entire life cycle and maximize the AGC comprehensive evaluation index Kp, providing the energy storage power and energy constraints for the control strategy of the bottom-layer model. Step 2: Constructing the constraints of a two-layer model for optimal energy storage configuration. The constraints of the underlying model include battery charging and discharging power constraints, power balance constraints, energy storage SOC constraints, and energy storage power constraints; the constraints of the upper-level model are the rated power and rated capacity constraints of the lithium battery. The battery charging and discharging power constraints of the underlying model in step two are reflected by a double Logistic function, which uses a double Logistic function to reflect the output constraints of the battery energy storage under different SOCs. The function expression is as follows: （12） when SOC <0.1 or SOC When ≥0.9, K(SOC) =0; where, P 0 Take 0.01; K max Take 1; n describes the rate of change of the curve, take 15; energy storage SOC The definition is as follows: （13） In the formula, E B.rated This refers to the rated energy storage capacity of the lithium battery. P b This refers to the discharge power of lithium battery energy storage. SOC The range is between 0 and 1, where t represents time. Power balance constraint is (14) In the formula P A For AGC power commands, P g To provide power to thermal power units, P fw It provides energy storage and power to the flywheel; Energy storage SOC constraint is （15） In the formula, SOC L and SOC H These are the maximum and minimum values ​​of SOC, respectively, and are set to 0.1 and 0.

9. SOC b (t) and SOC fw (t) These are the SOC of lithium battery energy storage and flywheel energy storage at time t, respectively; Energy storage power constraints include lithium battery power constraints and flywheel energy storage power constraints. Lithium battery power constraints （16） Flywheel energy storage power constraints （17） In the formula, P B.rated and P FW.rated These are the rated power of lithium battery energy storage and flywheel energy storage, respectively.

2. The method for optimizing the configuration of a hybrid energy storage dual-layer system to reduce frequency regulation losses in thermal power units according to claim 1, characterized in that: The method for establishing the objective function of the underlying model in step one is as follows: first, the plant-level AGC instructions are decomposed and reconstructed into n thermal power-hybrid energy storage instruction allocation schemes through EMD; second, each allocation scheme is input into the underlying model to solve for the output of each frequency regulation resource and the SOC curve of the hybrid energy storage.

3. The method for optimizing the configuration of a hybrid energy storage dual-layer system to reduce frequency regulation losses in thermal power units according to claim 1, characterized in that: The method for establishing the objective function of the upper-level model in step one is to set two objective functions. The objective function is the difference between frequency regulation costs and benefits over the entire life cycle, representing a very small indicator. Kp is the comprehensive evaluation index for AGC, and it is an extremely large index. （1） （2） In the formula C inv Investment costs for hybrid energy storage; C OM For hybrid energy storage operating costs; C scr C is the cost of disposing of lithium batteries at the end of their service life; ram and C pun These are the frequency regulation loss cost and the regulation offset power penalty cost of thermal power units, respectively. B kp For AGC frequency modulation mileage revenue; B coal To reduce the indirect benefits of lowering the system's coal-fired costs; B env This is an indirect benefit to reduce the system's power generation and pollution discharge costs.

4. The method for optimizing the configuration of a hybrid energy storage dual-layer system to reduce frequency regulation losses in thermal power units according to claim 1, characterized in that: The constraints of the upper-level model in step two include the rated power and rated capacity constraints of the lithium battery: (18) (19) In the formula, P B.rated This refers to the rated power of lithium battery energy storage.

Images