A Two-Layer AGC Frequency Modulation Control Method Considering Operating Economic Cost and Consistency of Energy Storage SOC

Abstract

The present invention discloses a two - layer AGC frequency modulation control method considering operating economic cost and the consistency of energy storage SOC, including: (1) constructing a frequency modulation operating cost function of a thermal power unit; (2) constructing a frequency modulation operating cost function of energy storage; (3) formulating a secondary frequency modulation control method at the regional control level; (4) formulating a control method for the consistency of the SOC of energy storage units at the energy storage station level; (5) formulating an SOC management scheme for energy storage batteries; (6) constructing a frequency modulation performance evaluation index; (7) building a dynamic model of unit frequency modulation with energy storage batteries. The control method proposed by the present invention considers the influence of different disturbance conditions on the frequency modulation responsibility allocation method of thermal - energy storage AGC, gives full play to the frequency modulation characteristics of energy storage with fast and flexible response, while optimizing the frequency modulation operating cost, improving the frequency modulation performance, increasing the frequency modulation compensation income, effectively improving the service life of energy storage and the comprehensive operating efficiency of the system.

Description

Technical Field

[0001] The present invention relates to an energy storage frequency modulation control method, and particularly to a two-layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency. Background Art

[0002] With the increasing scale of new energy power generation, the problems of its power generation intermittency, volatility and even anti-regulation are becoming increasingly prominent. There is an urgent need for effective technical solutions to solve the power grid frequency modulation problem brought about by the large-scale grid connection of new energy. Conventional thermal power units are generally large in scale and slow in frequency modulation response speed. Frequent start-stop of the units will cause energy waste and great damage to the units. The multiplicity of tasks also makes it difficult for thermal power units to fully play the frequency modulation function. Equipping a certain proportion of energy storage systems in conventional units and giving full play to the fast and flexible regulation characteristics of the energy storage system can greatly improve the frequency modulation performance of the units, reduce unit losses and increase the power grid frequency modulation capacity. At present, the research on the combined heat and energy storage frequency modulation control method has become a hot topic in this field.

[0003] In the research on the combined frequency regulation control strategy of thermal energy storage, scholars such as X. Xie introduced the full-power compensation control strategy adopted in the Shijingshan energy storage project in Beijing, that is, the energy storage system automatically compensates for the difference between the actual output of the unit and the AGC command. However, this full-power compensation strategy lacks effective management of the energy storage battery power, which is likely to cause the energy storage system to be unable to participate in frequency regulation continuously for a long time, and is not conducive to the improvement of the frequency regulation effect. At the same time, it will shorten the replacement cycle of the energy storage system and increase the cost of the power plant. Scholars such as Chen Lijuan proposed a control method aiming to improve the AGC regulation accuracy and shorten the AGC response time, and designed a strategy for the energy storage battery state of charge (SOC) to return to the limit. However, this SOC management method belongs to after-the-fact management and does not adjust the energy storage output according to the SOC during the normal frequency regulation of the energy storage, and the regulation effect on the energy storage SOC is limited. Scholars such as Hu Zechun proposed to decompose the area control error (ACE) signal into low-frequency and high-frequency components by filtering, and allocate the low-frequency component to the unit and the high-frequency component to the energy storage system. Although this method takes into account the characteristics of fast and flexible regulation of the energy storage, using a filter to decompose the signal will cause inaccurate allocation of the frequency regulation command due to the influence of signal amplitude attenuation, phase shift, distortion, etc. after filtering, which is not conducive to the improvement of the frequency regulation effect. Scholars such as MANOJ DATTA proposed a control method for proportionally allocating the ACE signal among electric vehicles, photovoltaic power generation systems and energy storage systems. Although this static proportional allocation method is simple and easy to operate, it does not fully consider the differences in the frequency regulation output characteristics of different power sources, ignores the influence of different load disturbance conditions on the allocation of frequency regulation commands, and lacks effective management of the energy storage SOC. In addition, existing research mostly regards the energy storage system as a single entity to participate in frequency regulation, and there is less research on the coordinated control of each energy storage unit in the energy storage power station. The differences (such as SOC differences) among energy storage units during frequency regulation will reduce the frequency regulation efficiency and is not conducive to the safe and stable operation of the system. Summary of the Invention

[0004] Object of the Invention: In view of the current situation and deficiencies in the research on the combined frequency regulation control of energy storage assisting units, considering the influence of different disturbance conditions on the frequency regulation responsibility allocation method of thermal energy storage AGC, based on the purposes of giving full play to the fast and flexible response frequency regulation characteristics of energy storage, optimizing the frequency regulation operation cost, improving the frequency regulation performance, effectively suppressing the fluctuation of energy storage SOC, increasing the service life of energy storage batteries and the comprehensive operation efficiency of the system, etc., a two-layer AGC frequency regulation control method considering the operating economic cost and the consistency of energy storage SOC is proposed.

[0005] Technical Solution: A two-layer AGC frequency regulation control method considering the operating economic cost and the consistency of energy storage SOC includes the following steps:

[0006] (1) Construct the frequency regulation operation cost function of thermal power units;

[0007] (2) Construct the frequency regulation operation cost function of energy storage;

[0008] (3) Based on steps (1) and (2), construct the combined frequency regulation operation cost function of energy storage units. Taking the minimization of the combined frequency regulation operation cost function of energy storage units as the optimization goal, fully consider the multi-technical characteristics of different frequency regulation power sources, and formulate the secondary frequency regulation control method at the regional control level;

[0009] (4) The upper-layer regional control center dynamically and optimally allocates the frequency regulation responsibility in real time between thermal power units and the energy storage system according to the AGC frequency regulation demand and the corresponding control strategy. After receiving the AGC frequency regulation instruction allocated based on step (3), the lower-layer energy storage station formulates the SOC consistency control method of the energy storage units at the energy storage station level;

[0010] (5) Formulate the SOC management scheme of energy storage batteries;

[0011] (6) Construct the frequency regulation performance evaluation index;

[0012] (7) Build the frequency regulation dynamic model of the unit with energy storage batteries.

[0013] As a further solution of the present invention, the frequency regulation operation cost of the thermal power unit in step (1) includes coal consumption cost, environmental cost and wear cost.

[0014] As a further solution of the present invention, the frequency regulation operation cost of the energy storage in step (2) includes operation and maintenance cost and aging cost.

[0015] As a further solution of the present invention, step (3) specifically includes the following steps:

[0016] (31) Establish a dynamic wear coefficient based on the power change rate;

[0017] (32) Establish the AGC frequency regulation control objective function;

[0018] (33) Establish the AGC frequency regulation control constraint conditions.

[0019] As a further solution of the present invention, the AGC frequency regulation control constraint conditions include two dimensions of frequency regulation demand and frequency regulation.

[0020] As a further solution of the present invention, step (4) specifically includes the following steps:

[0021] (41) Establish the dynamic model of the energy storage battery unit;

[0022] (42) Formulate the cooperative control algorithm of the energy storage system based on leader-follower multi-agent consistency.

[0023] As a further solution of the present invention, the energy storage battery SOC management solution in step (5) includes: performing fine-tuning management on the energy storage battery SOC under the condition that the load disturbance changes slowly and the energy storage battery is in the idle state of frequency modulation. By reasonably charging and discharging the energy storage battery, its SOC is gradually restored to the reference value, so that it can be put into the next frequency modulation operation in a better state in dynamic frequency modulation services.

[0024] As a further solution of the present invention, the specific implementation steps of the SOC management are as follows:

[0025] First, it is judged whether coarse-tuning management of the energy storage SOC is required, that is, whether the energy storage battery SOC is within the upper and lower limit thresholds of the deep charge / deep discharge range. The upper and lower limit thresholds of the energy storage battery SOC are 80% and 20% respectively. If it is in the range exceeding the upper and lower limit thresholds, the energy storage battery suspends the frequency modulation service and gives priority to the coarse-tuning management of the energy storage SOC, and performs constant-power charging / discharging on the energy storage battery; until the energy storage SOC returns to the normal range, the range value is set to 40% - 60%;

[0026] If the energy storage SOC is within the normal range, it is judged whether fine-tuning management of the energy storage battery SOC can be carried out, that is, whether the conditions that the change rate and acceleration of the frequency modulation command are small and the output of the thermal power unit has basically reached the command requirements have been met, that is, the energy storage battery is in the idle state of frequency modulation. If it is satisfied, it enters the fine-tuning management stage of the energy storage SOC. In the fine-tuning management stage of the energy storage SOC, if the energy storage SOC is not within the set range, the set range is 49% - 51%, then the energy storage battery is subjected to constant-power charging / discharging until the energy storage SOC returns to the reference range, and the set range is 49.9% - 50.1%; the fine-tuning management priority of the energy storage SOC is the lowest, that is, as long as the frequency modulation command and the output of the thermal power unit do not meet the requirements for entering the SOC fine-tuning management, or the energy storage needs to participate in a new frequency modulation command, the fine-tuning management of the energy storage SOC is suspended and the frequency modulation service is given priority.

[0027] As a further solution of the present invention, the evaluation indicators in step (6) include performance evaluation indicators under step load disturbance, performance evaluation indicators under continuous load disturbance, and economic indicators of frequency modulation compensation benefits.

[0028] As a further solution of the present invention, step (7) specifically includes:

[0029] (71) Construct an energy storage battery simulation model;

[0030] (72) Construct a dynamic model of frequency modulation for the regional power grid equipped with energy storage

[0031] Beneficial effects: Compared with the prior art, the present invention takes into account the influence of different disturbance conditions on the frequency modulation responsibility allocation method of the thermal energy storage AGC, introduces a dynamic wear coefficient based on the power change rate into the unit frequency modulation operation cost, and can give full play to the frequency modulation characteristics of the energy storage with fast and flexible response;

[0032] During the energy storage frequency modulation process, the energy storage SOC is dynamically controlled, which can effectively suppress the SOC fluctuation, and make full use of the idle state of the energy storage for SOC fine-tuning, so that it can adaptively recover to the reference value;

[0033] While optimizing the frequency modulation operation cost and improving the frequency modulation performance, the SOC difference between energy storage units is adjusted in real time, effectively improving the service life of the energy storage battery and the comprehensive operation efficiency of the system, and having great application value and prospects. Description of the Drawings

[0034] Figure 1 It is a flow schematic diagram of the present invention;

[0035] Figure 2 It is a control flow chart of the energy storage battery SOC;

[0036] Figure 3 It is a model diagram of the energy storage battery unit considering the state of charge;

[0037] Figure 4 It is a regional power grid frequency modulation dynamic model based on the ARR signal;

[0038] Figure 5 It is a frequency deviation response curve;

[0039] Figure 6 It is a change curve of the energy storage battery SOC;

[0040] Figure 7 It is an active power output change curve;

[0041] Figure 8 It is a change curve of the active power output of the energy storage battery pack;

[0042] Figure 9 It is a dynamic load disturbance curve;

[0043] Figure 10 It is a frequency deviation response curve;

[0044] Figure 11 It is a change curve of the energy storage battery SOC;

[0045] Figure 12 It is an active power output change curve;

[0046] Figure 13 It is a change curve of the frequency modulation operation cost;

[0047] Figure 14 It is the SOC change curve when there is no charge state management. Detailed implementation manners

[0048] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0049] As Figure 1 shown, a two-layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency includes the following steps:

[0050] (1) Construct a thermal power unit frequency modulation operation cost function;

[0051] (2) Construct an energy storage frequency modulation operation cost function;

[0052] (3) Based on steps (1) and (2), construct a combined frequency modulation operation cost function of the energy storage unit, taking the minimization of the combined frequency modulation operation cost function of the energy storage unit as the optimization goal, fully considering the multi-technical characteristics of different frequency modulation power sources, and formulating a secondary frequency modulation control method at the regional control level;

[0053] (4) The upper-layer regional control center dynamically and optimally allocates the frequency modulation responsibility in real time between the thermal power unit and the energy storage system according to the AGC frequency modulation demand and the corresponding control strategy. After receiving the AGC frequency modulation instruction allocated based on step (3), the lower-layer energy storage station, in order to meet the requirement that the output of each energy storage unit in the energy storage station quickly realizes the tracking of the target power, and at the same time ensure the consistency of the SOC of each energy storage unit, effectively improve the service life of the energy storage battery and the comprehensive operation efficiency of the system, formulates an energy storage unit SOC consistency control method at the energy storage station level;

[0054] (5) In order to enable the energy storage battery to participate in the AGC frequency modulation service sustainably, suppress the deep charge and discharge of each battery unit, extend the service life of the energy storage battery, and improve the frequency modulation reliability, formulate an energy storage battery SOC management scheme;

[0055] (6) In order to quantitatively evaluate the quality of the frequency modulation performance, and be able to more intuitively reflect the frequency modulation effect, the energy storage SOC management situation, and the quality of the frequency modulation compensation income, construct a frequency modulation performance evaluation index;

[0056] (7) Build a unit frequency modulation dynamic model including an energy storage battery.

[0057] Specifically, the step (1) includes:

[0058] (11) Coal consumption cost. The coal consumption cost generated by the thermal power unit due to frequency modulation output is as follows:

[0059]

[0060] In the formula, CU 1,k represents the frequency regulation coal consumption cost of the thermal power unit at time k; a 1,i , b 1,i respectively represent the consumption coefficient of the i-th thermal power unit; P U,i,k represents the frequency regulation output value of the i-th thermal power unit at time k.

[0061] (12) Environmental cost. When the thermal power unit participates in frequency regulation, it will generate greenhouse gas emissions, thus bringing the environmental cost of treating power generation sewage. The calculation formula is as follows:

[0062]

[0063] In the formula, C U 2,k represents the frequency regulation environmental cost of the thermal power unit at time k; c NOx represents the unit sewage discharge cost of nitrogen oxides; c SO2 represents the unit sewage discharge cost of sulfur dioxide; c CO2 represents the unit sewage discharge cost of carbon dioxide; Δt represents the sampling interval time.

[0064] (13) Wear cost. When the thermal power unit participates in frequency regulation, the unit will be worn due to climbing and changing output, etc. The calculation formula for the resulting wear cost is established as follows:

[0065]

[0066] In the formula, C U 3,k represents the frequency regulation wear cost of the thermal power unit at time k; m i,k represents the frequency regulation wear coefficient of the i-th thermal power unit at time k.

[0067] The said step (2) includes:

[0068] (21) Operation and maintenance cost. The operation and maintenance cost of the energy storage system mainly refers to the cost caused by the system equipment during production and operation, including employee salaries, equipment maintenance costs, and operation costs of related equipment, etc. The calculation formula for the operation and maintenance cost of the energy storage system is as follows:

[0069]

[0070] In the formula, C B 1,k represents the frequency regulation operation and maintenance cost of the energy storage system at time k; c m,j represents the unit capacity operation and maintenance cost of the j-th energy storage battery; P B,j,k represents the frequency regulation output value of the j-th energy storage unit at time k.

[0071] (22) Aging cost. The deep charge and discharge of the energy storage battery will accelerate the wear and aging of the battery. The calculation formula of the aging cost generated during its participation in frequency regulation is established as follows:

[0072]

[0073] In the formula, C B 2,k represents the frequency regulation aging cost of the energy storage system at time k; c n,j represents the aging coefficient of the jth energy storage battery; SOC j,k represents the state of charge of the jth energy storage battery at time k; SOC ref represents the reference state of charge of the energy storage system.

[0074] Furthermore, the specific content of step (3) is as follows:

[0075] (31) Establish a dynamic wear coefficient based on the power change rate. Since the frequency regulation wear cost of thermal power units mainly stems from the output change of the units during ramping, the faster the output changes, the greater the wear cost. Therefore, in order to effectively describe the frequency regulation wear cost of thermal power units, the present invention establishes a dynamic wear coefficient based on the power change rate of the frequency regulation output of thermal power units as follows:

[0076] m i,k = m i,0 +α·v k (6)

[0077] In the formula, m i,0 is the reference value of the wear coefficient; v k is the frequency regulation power change rate of the thermal power unit at time k; α is the influence factor coefficient.

[0078] Among them, the calculation expression of the frequency regulation power change rate of the thermal power unit at time k is as follows:

[0079]

[0080] In the formula, P U,i,k is the frequency regulation output value of the ith thermal power unit at time k.

[0081] (32) Establish an AGC frequency regulation control objective function. The essence of the AGC frequency regulation control of the energy storage assisting the thermal power unit is to reasonably allocate the frequency regulation responsibility of the AGC instruction issued by the dispatching center between the thermal power unit and the energy storage system. At the regional control level, in order to ensure the economic optimality of the frequency regulation power distribution at each adjustment moment, the present invention takes the minimization of the combined frequency regulation operation cost function of the energy storage - unit as the optimization goal, fully considers the multi - technical characteristics of different frequency regulation power sources, and uses an improved particle swarm algorithm to perform real - time solution under multiple constraint conditions to realize the dynamic optimization allocation of the frequency regulation responsibility between the unit and the energy storage.

[0082] The objective function of the hierarchical frequency modulation control in the k-th moment area is established as follows:

[0083]

[0084] In the formula, C U 1,k , C U 2,k , C U 3,k respectively represent the frequency modulation coal consumption cost, the frequency modulation environmental cost and the frequency modulation wear cost of the thermal power unit at the k-th moment; C B 1,k , C B 2,k respectively represent the frequency modulation operation and maintenance cost and the frequency modulation aging cost of the energy storage system at the k-th moment.

[0085] (33) Establish the AGC frequency modulation control constraint conditions. The present invention establishes the AGC frequency modulation control constraint conditions from two aspects of "frequency modulation demand" and "frequency modulation capacity".

[0086] For the "frequency modulation demand", in order to better exert the frequency modulation characteristics of the thermal power and energy storage, the present invention sets that the frequency modulation commands of the thermal power unit and the energy storage system are within the range of their respective active power output capabilities, so that their respective outputs can meet the command requirements in real time. Therefore, the sum of the frequency modulation active power commands borne by the thermal power unit and the energy storage should be equal to the total AGC command at this moment, that is:

[0087] P agc,k =∑P U,i,k +∑P B,j,k (9)

[0088] In the formula, P agc,k is the AGC frequency modulation command issued by the dispatching center at the k-th moment.

[0089] For the "frequency modulation capacity", it mainly includes the ramp rate and load reserve capacity of the thermal power unit, the charge and discharge power limit of the energy storage and the variable range of the state of charge, that is:

[0090]

[0091] In the formula, v i is the ramp rate of the i-th thermal power unit; P i,max , P i,min are the upper and lower limits of the frequency modulation output of the i-th thermal power unit; P u,i,k is the actual frequency modulation output of the i-th unit at the k-th moment; P b,j,k is the actual frequency modulation output of the j-th energy storage unit at the k-th moment; SOC j,k is the state of charge of the j-th energy storage unit at the k-th moment; Pj,max , P j,min is the upper and lower limits of the output of the j-th energy storage unit; SOC j,max , SOC j,min are the upper and lower limits of the state of charge of the j-th energy storage unit.

[0092] Furthermore, in the step (4):

[0093] (41) Establish a dynamic model of the energy storage battery unit. Since the power regulation inertia time constant of each group of energy storage battery units is small (in milliseconds), while the regulation period of the energy storage system in response to AGC frequency modulation is usually long (from several seconds to several minutes), therefore, in the dynamic frequency modulation model of the large-scale energy storage system, the inertia time constant of each battery energy storage unit is considered negligible. Thus, the relationship between the SOC of the energy storage battery unit and its charge and discharge power can be obtained as shown in the following formula:

[0094]

[0095] In the formula, E B,j represents the rated capacity of the j-th energy storage unit.

[0096] For simplified calculation and to better apply the multi-agent cooperative control algorithm, a proportionality coefficient λ is introduced here to satisfy:

[0097]

[0098] In the formula, S B,j,k is the value of the state of charge of the j-th energy storage unit at time k after proportional transformation.

[0099] In a large-scale energy storage system, the second-order dynamic characteristics of each group of energy storage units can be expressed by the following formula:

[0100]

[0101] In the formula, u B,j,k is the power step factor of the j-th energy storage unit at time k.

[0102] (42) Develop a cooperative control algorithm for the energy storage system based on leader-follower multi-agent consensus. The large-scale energy storage system consists of multiple energy storage battery units with second-order dynamic characteristics and is a multi-agent system. Therefore, the present invention constructs a multi-agent system with a second-order leader-follower structure and uses the consensus cooperative control algorithm to enable each group of energy storage battery units to track the target value output and at the same time ensure that the state of charge of each group of energy storage units remains consistent.

[0103] In the theory of multi-agent systems with a second-order leader-follower structure, the knowledge of graph theory and matrix theory is mainly used. The network structure of the system can be represented by a graph G=(V, E), where V={0, 1, 2, …, n} represents the set of n follower nodes and one leader node (node 0) in the network. represents the set of edges. Denote G f =(V f , E f ) to represent the topological structure among followers. Use the adjacency matrix A=(a ij ) to represent the relationship between nodes and edges.

[11] :

[0104]

[0105] When the graph G f is an undirected graph, A is a symmetric matrix. In an undirected graph, use D = diag{d ii} to represent the degree matrix, where Use the Laplacian matrix L=(l ij ) to represent another relationship between nodes and edges:

[0106]

[0107] That is, L = D - A.

[0108] For a multi-agent system with a second-order leader-follower structure composed of the leader agent in the EMS layer of the energy storage station and the follower agents of each group of energy storage battery units, the dynamic characteristics of the leader can be expressed as:

[0109]

[0110] In the formula, S B,0,k represents the value after proportional transformation of the state of charge that the energy storage unit needs to follow at time k; P B,0,k represents the frequency regulation output value that the energy storage unit needs to follow at time k; P ref B,k represents the total AGC frequency regulation command value assigned to the energy storage system at time k; n represents the number of energy storage battery packs.

[0111] To achieve S B,j,k = S B,0,k and P B,j,k = P B,0,k , the present invention uses the following consensus protocol:

[0112]

[0113] In the formula, N i is the set of neighbors of node i. Denote the relationship between the leader and follower edges as γ 0 , γ 1 ∈R.

[0114] For the closed-loop multi-agent system, its consensus protocol can be expressed in the following matrix form:

[0115]

[0116] where d = [d 1 , d 2 ,..., d n T , D d = diag(d 1 , d 2 ,..., d n ), L is the Laplacian matrix of graph G f .

[0117] For the multi-agent energy storage system to achieve collaborative consensus, the requirement for the communication topology is that there is at least 1 path for the information transmission of the energy storage station EMS layer to reach any group of energy storage battery units, that is, the communication topology is required to be connected. In addition, the multi-agent cooperative control algorithm must also meet other convergence conditions, such as the value requirements of γ 0 , γ 1 , which will not be elaborated here.

[0118] The specific steps of step (5) include:

[0119] (51) To enable the energy storage battery to participate in the AGC frequency modulation service sustainably, the present invention performs real-time management of the state of charge (SOC) of each battery pack in the energy storage station, inhibits deep charging and discharging of each battery unit, controls its SOC to be maintained within a range as small as possible near the reference value (set by the present invention to be 50%), prolongs the service life of the energy storage battery, and improves the reliability of frequency modulation.

[0120] To minimize the impact of the energy storage battery SOC management on the performance of the thermal-storage combined AGC frequency modulation, the present invention selects to perform fine-tuning management of the energy storage battery SOC under the condition of slow load disturbance change and when the energy storage battery is in a frequency modulation idle state. By reasonably charging and discharging the energy storage battery, its SOC is gradually restored to the reference value, enabling it to be put into the next frequency modulation operation in a better state in the dynamic frequency modulation service.

[0121] ​The specific implementation strategy for the SOC management of the energy storage battery is as follows: First, it is judged whether coarse adjustment management of the energy storage SOC is required, that is, whether the SOC of the energy storage battery is in the deep charge / discharge range (the upper and lower limit thresholds of the SOC of the energy storage battery are set at 80% and 20% respectively in the present invention). If it is in the range exceeding the upper and lower limit thresholds, the energy storage battery suspends the frequency regulation service and preferentially conducts the coarse adjustment management of the energy storage SOC, and the energy storage battery is charged / discharged at a constant power (set its value to 1 / 2*P BN , P BN represents the rated power of the energy storage battery), until the energy storage SOC returns to the normal range (set at 40% - 60% in the present invention). If the energy storage SOC is within the normal range, it is judged whether fine adjustment management of the energy storage battery SOC can be carried out, that is, whether the change rate and acceleration of the frequency regulation command are small enough (the working condition with slow load disturbance change), and the output of the thermal power unit has basically reached the command requirement (the idle state of the energy storage battery for frequency regulation). If the conditions are met, it enters the fine adjustment management stage of the energy storage SOC. In the fine adjustment management stage of the energy storage SOC, if the energy storage SOC is not within the better range (set at 49% - 51% in the present invention), the energy storage battery is charged / discharged at a constant power (set its value to 1 / 15*P BN ), until the energy storage SOC returns to the reference range (set at 49.9% - 50.1% in the present invention). The fine adjustment management of the energy storage SOC has the lowest priority, that is, as long as the frequency regulation command and the output of the thermal power unit do not meet the requirements for entering the SOC fine adjustment management, or the energy storage needs to participate in a new frequency regulation command, the fine adjustment management of the energy storage SOC is suspended and the frequency regulation service is preferentially responded to.

[0122] In addition, in order to avoid unnecessary output fluctuations caused by the change of the SOC of the energy storage battery and have an adverse impact on the frequency regulation performance, after the fine adjustment management of the energy storage battery SOC is implemented in the present invention, for the working condition with a large load disturbance amplitude and slow change, a blocking control is adopted, that is, in the frequency regulation optimization control, a small offset is introduced on the basis of the original SOC, so that the frequency regulation responsibility distribution method remains stable, preventing unnecessary output fluctuations and ensuring the stability of the system frequency regulation performance.

[0123] The real-time management process of the energy storage battery SOC is as Figure 2 shown.

[0124] The specific steps of step (6) include:

[0125] (61) Construct a performance evaluation index under a step load disturbance. Referring to the frequency regulation performance indexes in the "Implementation Rules for the Operation Management of Grid-connected Power Plants" and the "Implementation Rules for the Auxiliary Service Management of Grid-connected Power Plants", and combining with the performance evaluation method of the frequency regulation process, the present invention proposes a frequency regulation performance evaluation index under a step load disturbance as shown in Table 1,

[0126] Table 1 Frequency Regulation Performance Indexes under Step Disturbance

[0127]

[0128] v is the AGC regulation rate of the unit; P s and P e are the output powers of the unit at the start and end of regulation; T e and T s are the start and end times of the ramp section during AGC regulation; P bias is the average regulation deviation; P A is the AGC command power; P(t) is the output power of the unit during the oscillation period; T oc is the duration of the oscillation period; t is the response time; t b is the start time of regulation. v r is the frequency recovery rate; d m and t m are the maximum value of the absolute value of the frequency deviation and the corresponding time; d s and t s are the steady-state frequency deviation value and the corresponding time; σ 1 represents the overall standard deviation of the frequency; n s is the sampling point number when reaching the steady-state frequency; f i represents the system frequency corresponding to the i-th sampling point; f N represents the reference frequency; t r represents the frequency recovery duration; v N and P N,bias and t N and t N,r are the standard values of the corresponding parameters respectively; d N,m and d N,s and v N,m and v N,r and σ N are the adjustment multiples used to increase the discrimination of the corresponding indicators respectively.

[0129] According to the refined indicators in Table 1, define the comprehensive regulation performance index K p1 as shown in the following formula. The larger K p1 , the better the performance.

[0130]

[0131] In the formula, a 1 and b 1 and c 1 and a 2 and b 2 and c 2 and d 2 and e 2 and f 2 and a, b are weight coefficients.

[0132] (62) Construct performance evaluation indicators under continuous load disturbances. To more intuitively reflect the frequency regulation effect and the management of the energy storage SOC, considering the frequency regulation characteristics under continuous load disturbances, the present invention introduces two refined indicators as shown in Table 2, where σ 2 , σ 3 represent the overall standard deviations of frequency and SOC respectively; f i , SOC i represent the system frequency and the energy storage SOC corresponding to the i-th sampling point respectively; n z is the number of the last sampling point; f N , SOC ref represent the reference values of frequency and SOC respectively.

[0133] Table 2 Frequency Regulation Performance Indicators under Continuous Disturbances

[0134]

[0135] According to the refined indicators in Table 2, define the comprehensive regulation performance indicator K p2 as shown in the following formula. The larger K p2 , the better the performance.

[0136]

[0137] In the formula, γ and λ are weight coefficients.

[0138] (63) Construct economic indicators for frequency regulation compensation benefits. To better reflect the advantages and disadvantages of frequency regulation compensation benefits, referring to the "Jiangsu Electric Power Auxiliary Service (Frequency Regulation) Market Trading Rules", the present invention proposes the following economic indicators for frequency regulation compensation benefits:

[0139] F J = K agc × Min(K p , 2) × P agc (21)

[0140] In the formula, F J is the economic indicator for frequency regulation compensation benefits; K agc is the basic compensation standard, 2 yuan / MW; K p is the comprehensive regulation performance indicator; P agc is the AGC adjustable capacity of the frequency regulation power source, taken as the difference between the upper and lower limits of AGC regulation.

[0141] The specific steps of step (7) include:

[0142] (71) Construct a simulation model of the energy storage battery. For each group of energy storage battery units, establish a simulation model that includes the energy storage SOC and can be used to study energy storage assisted frequency regulation, as Figure 3 shown, KT represents the integration power calculation time constant; E B represents the rated capacity of the energy storage battery unit; S SOC,in represents the initial value of the state of charge of the energy storage battery unit; P B,ref represents the active power output command of the energy storage battery unit; P B represents the actual active power output of the energy storage battery unit; S SOC is the actual SOC of the energy storage battery unit.

[0143] (72)Construct a regional power grid frequency modulation dynamic model equipped with energy storage. Based on the signal allocation mode of the Area Regulation Requirement (ARR), construct a regional power grid frequency modulation dynamic model for combined frequency modulation of thermal power units equipped with energy storage, as Figure 4 shown, Δf is the system frequency deviation; ΔP line is the power exchange deviation of the interconnected power grid tie line; K I is the integral coefficient of the PI regulator; K k is the proportional coefficient of the PI regulator; B is the system frequency deviation coefficient; ACE represents the regional control deviation; P AGC represents the AGC frequency modulation output command; P ref Gi represents the AGC command of the i-th conventional thermal power unit; P ref Bj represents the AGC command of the j-th energy storage system; P Gi1 represents the primary frequency modulation output of the i-th traditional thermal power unit; P Gi represents the actual active power output of the i-th traditional thermal power unit; P Bj is the actual active power output of the j-th energy storage system; P Ld is the net load disturbance of the system; T g 、T t 、T r respectively represent the time constants of the governor, generator and reheater; R represents the unit speed regulation coefficient; K r is the reheater coefficient; K p is the system gain; T p is the system time constant.

[0144] The calculation formula for the Area Control Error (ACE) is:

[0145] ACE = ΔP line + B·Δf (22)

[0146] In the regional power grid frequency modulation dynamic model of the present invention, a constant frequency adjustment method is selected, that is, the power exchange deviation of the interconnected power grid tie line is not considered, and the regional control deviation is ACE = B·Δf.

[0147] Embodiment

[0148] Model parameters:

[0149] Considering the technical characteristics, output characteristics, frequency regulation characteristics, economy, etc. of thermal power units and energy storage, the parameters of the frequency regulation control method and model are selected as follows:

[0150] Table 3 Frequency regulation control method and simulation model parameters

[0151]

[0152] Use the Matlab / Simulink platform to build a simulation model, and establish a two-layer control method for combined thermal energy storage AGC frequency regulation based on improved PSO in the Matlab Function module. Assume that the installed capacity of the regional power grid is 1000 MW, and the reference power is selected as 1000 MW. The initial state of charge of six groups of energy storage battery units is 55%, 50%, 52%, 45%, 48% and 53% respectively. The variable range of energy storage SOC is controlled within 10% - 90%, the rated power of each group of energy storage units is ±5 MW, the rated capacity of each group of energy storage units is 2.5 MW·h, the optimal state of charge of energy storage is 50%, the standby capacity of the thermal power unit is 40 MW, and the ramp rate is 3% / min of the rated power. Simulation results of two-layer AGC frequency regulation control considering operating economic cost and energy storage SOC consistency

[0153] Step load disturbance condition

[0154] Select the disturbance condition as: a step load disturbance of 0.02 p.u. is added to the system at 500 s. The relevant parameter settings of the indicators described in Table 1 are as follows:

[0155] Table 4 Frequency regulation performance index parameters under step disturbance

[0156]

[0157] Based on the frequency regulation dynamic model of the regional power grid, the method of the present invention is simulated and compared with two common methods in engineering and existing research. Among them, Method 1 is the differential compensation method, and Method 2 is the static proportional distribution method (the coefficient ratio of the unit and the energy storage is 7:3). Since the above two methods do not involve the control of energy storage SOC consistency, when simulating them, each energy storage unit is jointly regarded as an overall energy storage system, and the SOC regulation of each internal unit is not considered. The initial SOC value of the energy storage system is set to 50%. The simulation comparison results are as Figures 5 - 8 and shown in Table 5:

[0158] Table 5 Evaluation index values under step load disturbance condition

[0159]

[0160] As Figure 5 shown, after a step disturbance occurs at 500 s, the frequency deviation when using the method of the present invention and Method 1 is significantly smaller than that of Method 2. It should be noted that, as Figure 7 、 Figure 8 shown, in the initial stage (0 - 230 s), the frequency deviation that occurs when using the method of the present invention is generated during the process of adjusting the SOC of each energy storage unit to be consistent. If the initial SOC of each energy storage unit is consistent, the above-mentioned frequency deviation will not occur.

[0161] Figure 6 It reflects the change of the SOC of each energy storage unit under the method of the present invention and the comparison results with Method 1 and Method 2. It can be seen that after about 230 s, the SOC of each energy storage unit under the method of the present invention tends to be consistent and remains the same as the leading battery pack. After a step disturbance occurs at 500 s, as Figure 6 and Figure 7 shown, since the method of the present invention undertakes the most frequency modulation responsibility in the initial stage of the disturbance and the energy storage output exits more slowly, its energy storage SOC drops the most severely during the period from 500 s to 600 s. When its energy storage SOC drops below 49%, the method of the present invention immediately enters the fine-tuning management of the energy storage SOC, and the energy storage battery charges at a constant power, and gradually adjusts the SOC during the period from 600 s to 1400 s until it recovers to 50.06%. For the other two methods, due to the lack of effective energy storage SOC management, the SOC cannot recover to near the reference value by itself.

[0162] Figure 8 It reflects the change process of the active power output of the energy storage battery pack. It can be seen that in the initial stage of adjustment, each energy storage unit charges / discharges to the greatest extent possible to make its SOC tend to be consistent with the leading battery pack. When the SOC of each energy storage unit remains consistent, its output follows the leading battery pack to remain the same, realizing the tracking of the target power.

[0163] As shown in Table 5, under the step load disturbance condition, the vast majority of the refined index values of the method of the present invention are the highest, and its comprehensive regulation performance is the best.

[0164] Continuous load disturbance condition

[0165] According to data statistics, about 80% of the AGC command values in actual projects are within 3% of the total installed capacity. Therefore, the continuous load disturbance mode selected by the present invention is: the net load fluctuates within the range of ±30 MW for about 5000 s, including various typical working conditions such as continuous low frequency and continuous high frequency. The dynamic load disturbance curve is as Figure 9 shown. The relevant parameter settings of the said index are as follows:

[0166] Table 6 Frequency modulation performance index parameters under continuous load disturbance

[0167]

[0168] Under the above disturbance conditions, the method of the present invention is simulated and compared with two common methods in engineering and existing research. Among them, Method 1 is the differential compensation method, and Method 2 is the static proportional distribution method (the coefficient ratio of the unit to the energy storage is 7:3). The simulation comparison results are as Figures 10 - 14 shown in Table 7.

[0169] Table 7 Evaluation index values under continuous load disturbance conditions

[0170]

[0171] Figure 10 It reflects the change of the frequency deviation response of the three methods under continuous load disturbance conditions. It can be seen that whether it is under low-frequency or high-frequency continuous load disturbance conditions, the method of the present invention can control the frequency deviation within a small range. Especially under high-frequency load disturbance, the frequency deviation is significantly smaller than the other two methods, and the frequency deviation optimization effect is remarkable.

[0172] As Figure 11 shown, under the method of the present invention, the SOC of each energy storage unit tends to be consistent after about 230 seconds and the tracking of the target output command is achieved. During 3200 - 3700 s and 5850 - 6300 s, the method of the present invention performs fine-tuning management of the energy storage SOC, so that the SOC of each energy storage unit gradually returns to the reference value. Compared with the other two methods, when the method of the present invention is adopted, the SOC fluctuation of the energy storage system is significantly smaller, and it can self-recover to the reference value, with a larger adjustable margin, which can effectively improve the sustainability of the energy storage participating in frequency modulation auxiliary services.

[0173] As Figure 12 shown, in the high-frequency load disturbance stage, the energy storage undertakes the main frequency modulation responsibility under the method of the present invention, and the unit undertakes the main AGC frequency modulation command under low-frequency and large-amplitude load disturbances. Compared with the other two methods, the frequency modulation output of the thermal power unit is smoother under the method of the present invention, which can reduce the wear of the unit and improve the reliability of system operation.

[0174] As Figure 13 shown, the frequency modulation operation cost is the smallest and the economy is better when the method of the present invention is adopted.

[0175] Figure 14It reflects the SOC changes of each energy storage unit when Method 2 (without state of charge management) is adopted. It can be seen that under continuous load disturbance conditions, due to continuous discharge, the SOC of each energy storage unit seriously deviates from the reference value (50%). At about 6000 seconds, the SOC of Battery Pack 6 has dropped below 20%, being in a state of deep charge and deep discharge, which is not conducive to the service life of the battery. Moreover, if continuous discharge continues, there will be a risk that the battery stops discharging and exits frequency modulation, resulting in a secondary frequency drop, seriously threatening the sustainability and reliability of energy storage frequency modulation.

[0176] As shown in Table 7, the two refined index values of the method of the present invention are both the highest, and the comprehensive regulation performance advantage is significant.

[0177] Frequency modulation compensation revenue comparison

[0178] Under the above two disturbance conditions, the economic index value F of the frequency modulation compensation revenue of the 3 frequency modulation control methods (the same as above) J is as follows:

[0179] Table 8 Economic index values of frequency modulation compensation revenue

[0180]

[0181] As shown in the above table, under step and continuous load disturbance conditions, the economic index value of the frequency modulation compensation revenue of the method of the present invention is the highest, reflecting the economic superiority of the method of the present invention in terms of frequency modulation compensation revenue.

[0182] In summary, in view of the power grid frequency regulation problem brought about by the large-scale grid connection of new energy, considering the influence of different disturbance conditions on the AGC frequency regulation responsibility allocation method of thermal energy storage, a two-layer AGC frequency regulation control method considering operating economic cost and the consistency of energy storage state of charge is proposed: 1) An optimization model for the combined AGC frequency regulation control of thermal energy storage is established. With the goal of minimizing the frequency regulation operating cost at the regional control level, the improved particle swarm optimization algorithm is used to solve the optimization problem, realizing the economically optimized dynamic frequency regulation responsibility allocation of thermal energy storage. A dynamic wear coefficient based on the power change rate is introduced into the unit frequency regulation operating cost, giving full play to the frequency regulation characteristics of energy storage with fast and flexible response. 2) At the energy storage station level, the consistency cooperative control algorithm is used to achieve the output of each energy storage battery unit tracking the target value, while ensuring that the SOC of each energy storage unit remains consistent. During the energy storage frequency regulation process, the energy storage SOC is dynamically managed and controlled, effectively suppressing the fluctuation of the energy storage SOC, and making full use of the idle state of the energy storage for SOC fine-tuning, enabling it to adaptively recover to the reference value, effectively improving the service life of the energy storage and the comprehensive operating efficiency of the system. 3) A dynamic model for the combined regional power grid frequency regulation of thermal energy storage is built, and simulation comparison experiments are carried out using the Matlab / Simulink platform. The results show that the control strategy proposed in the present invention can effectively improve the frequency regulation performance, optimize the economic cost of frequency regulation operation, and increase the frequency regulation compensation income. At the same time, the energy storage SOC fluctuates less, the frequency regulation output of the unit is smoother, which can effectively reduce the unit loss and improve the operating reliability of the system.

Claims

1. A two - layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency, characterized in that, it includes the following steps: (1) Construct the frequency modulation operating cost function of thermal power units; (2) Construct the frequency modulation operating cost function of energy storage; (3) Based on steps (1) and (2), construct the combined frequency modulation operating cost function of energy storage units. Taking the minimization of the combined frequency modulation operating cost function of energy storage units as the optimization goal, fully considering the multi - technical characteristics of different frequency modulation power sources, formulate the secondary frequency modulation control method at the regional control level; (4) The upper - layer regional control center dynamically and optimally distributes the frequency modulation responsibility in real - time between thermal power units and the energy storage system according to the AGC frequency modulation demand and corresponding control strategies. After the lower - layer energy storage station receives the AGC frequency modulation command allocated based on step (3), formulate the SOC consistency control method of energy storage units at the energy storage station level; (5) Formulate the SOC management plan of energy storage batteries; (6) Construct the frequency modulation performance evaluation index; (7) Build the frequency modulation dynamic model of the unit with energy storage batteries; In step (1), the frequency modulation operating cost of thermal power units includes coal consumption cost, environmental cost, and wear cost; The calculation formula for the coal consumption cost generated by thermal power units due to frequency modulation output is as follows: （1） In the formula, C U 1,k represents the frequency regulation coal consumption cost of the thermal power unit at time k; and respectively represent the consumption coefficients of the i th thermal power unit; P U,i,k represents the frequency regulation output value of the i th thermal power unit at k time; Thermal power units will emit greenhouse gases during frequency modulation. The calculation formula for the environmental cost of treating power generation pollution is as follows: （2） In the formula, C U 2,k represents the frequency regulation environmental cost of the thermal power unit k at the moment; c NOx represents the unit sewage discharge cost of nitrogen oxides; c SO2 represents the unit sewage discharge cost of sulfur dioxide; c CO2 represents the unit sewage discharge cost of carbon dioxide; represents the sampling interval time; When thermal power units perform frequency modulation, the wear caused by the change in output due to ramping results in wear cost. The calculation formula is as follows: （3） In the formula, C U 3,k represents the frequency regulation wear cost of the thermal power unit k at the moment; m i,k represents the i th k frequency regulation wear coefficient of the thermal power unit at the moment; In step (2), the frequency modulation operating cost of energy storage includes operation and maintenance cost and aging cost; The operation and maintenance cost of the energy storage system refers to the cost caused by the system equipment during production and operation, including employee salaries, equipment maintenance costs, and the operating costs of related equipment. The calculation formula is as follows: （4） Wherein, C B 1,k represents the frequency regulation operation and maintenance cost of the energy storage system k at a certain moment; c m,j represents the operation and maintenance cost per unit capacity of the j th energy storage battery; P B,j,k represents the j th energy storage unit k at a certain moment for the frequency regulation output value; The deep charge and discharge of energy storage batteries will accelerate the wear and aging of the batteries. The calculation formula for the aging cost generated when they participate in frequency modulation is as follows: （5） In the formula, C B 2,k represents the frequency regulation aging cost of the energy storage system k at time c n,j represents the aging coefficient of the j nth energy storage battery; SOC j,k represents the j nth energy storage battery k state of charge at time SOC ref represents the reference state of charge of the energy storage system; Step (3) specifically includes the following steps; (31) Establish a dynamic wear coefficient based on the power change rate; The calculation formula for the dynamic wear coefficient is as follows: （6） In the formula, m i,0 is the reference value of the wear coefficient; v k is k the change rate of the frequency regulation power of the thermal power unit at the moment; α is the influence factor coefficient; Among them, k The calculation formula for the change rate of the frequency modulation power of the thermal power unit at each moment is as follows: （7） In the formula, P U,i,k is the i th thermal power unit k 's frequency regulation output value at the moment; (32) Establish the AGC frequency modulation control objective function; The objective function of frequency modulation control at the regional control level at time k is as follows: （8） Wherein, C U 1,k , C U 2,k , C U 3,k respectively represent the frequency regulation coal consumption cost, frequency regulation environmental cost and frequency regulation wear cost of the thermal power unit at the k moment; C B 1,k , C B 2,k respectively represent the frequency regulation operation and maintenance cost and frequency regulation aging cost of the energy storage system at the k moment. (33) Establish the AGC frequency modulation control constraint conditions; Establish the AGC frequency modulation control constraint conditions from two aspects: frequency modulation demand and frequency modulation capacity, as follows: For frequency modulation demand, in order to better utilize the frequency modulation characteristics of thermal power and energy storage, by setting the frequency modulation commands of thermal power units and the energy storage system within their respective active power output capabilities, so that their respective outputs can meet the command requirements in real - time. Therefore, the sum of the frequency modulation active power commands borne by thermal power units and energy storage is equal to the total AGC command at this moment, that is: （9） In the formula, P agc,k is k the AGC frequency modulation command issued by the dispatching center at a certain moment; For frequency modulation capacity, it includes the ramp rate and load reserve capacity of thermal power units, the charge - discharge power limit of energy storage, and the variable range of state of charge, that is: （10） Wherein, v i is the i th ramp rate of the thermal power unit; P i,max , P i,min are the upper and lower limits of the frequency regulation output of the i th thermal power unit; P u,i,k is the actual frequency regulation output of the i th unit at the k moment; P b,j,k is the actual frequency regulation output of the j th energy storage unit at the k moment; SOC j,k is the state of charge of the j th energy storage unit at the k moment; P j,max , P j,min are the upper and lower limits of the output of the j th energy storage unit; SOC j,max , SOC j,min are the upper and lower limits of the state of charge of the j th energy storage unit.

2. The two - layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency according to claim 1, characterized in that, step (4) specifically includes the following steps: (41) Establish the dynamic model of energy storage battery units; (42)Develop a collaborative control algorithm for energy storage systems based on leader-follower multi-agent consensus.

3. The two-layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency according to claim 1, characterized in that the energy storage battery SOC management scheme in step (5) includes: performing fine-tuning management of the energy storage battery SOC under the condition of slow load disturbance change and the idle state of the energy storage battery for frequency modulation. By reasonably charging and discharging the energy storage battery, its SOC gradually returns to the reference value, and it can be put into the next frequency modulation operation in a better state in dynamic frequency modulation services.

4. The two-layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency according to claim 3, characterized in that the specific implementation steps of the SOC management are as follows: First, judge whether coarse-tuning management of the energy storage SOC is required, that is, whether the energy storage battery SOC is in the deep charge / discharge range. The upper and lower limit thresholds of the energy storage battery SOC are 80% and 20% respectively. If it is in the range exceeding the upper and lower limit thresholds, the energy storage battery suspends frequency modulation services and gives priority to coarse-tuning management of the energy storage SOC, and performs constant-power charging / discharging on the energy storage battery; until the energy storage SOC returns to the normal range, the range value is set to 40% - 60%; If the energy storage SOC is within the normal range, judge whether fine-tuning management of the energy storage battery SOC can be performed, that is, whether the change rate and acceleration of the frequency modulation command are small enough, and the output of the thermal power unit has basically reached the command requirement, that is, the idle state of the energy storage battery for frequency modulation. If it is satisfied, enter the fine-tuning management stage of the energy storage SOC. In the fine-tuning management stage of the energy storage SOC, if the energy storage SOC is not within the set range, the set range is 49% - 51%, then perform constant-power charging / discharging on the energy storage battery until the energy storage SOC returns to the reference range, the set range is 49.9% - 50.1%; The fine-tuning management priority of the energy storage SOC is the lowest, that is, as long as the frequency modulation command and the output of the thermal power unit do not meet the requirements for entering the SOC fine-tuning management, or the energy storage needs to participate in a new frequency modulation command, then suspend the fine-tuning management of the energy storage SOC and give priority to responding to frequency modulation services.

5. The two-layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency according to claim 1, characterized in that the evaluation indexes in step (6) include performance evaluation indexes under step load disturbance, performance evaluation indexes under continuous load disturbance, and economic indexes of frequency modulation compensation benefits.

6. The two-layer AGC frequency modulation control method considering operating economic cost and energy storage SOC consistency according to claim 1, characterized in that step (7) specifically includes: (71) Construct an energy storage battery simulation model; (72) Construct a dynamic model of frequency modulation for a regional power grid equipped with energy storage.

Images