A heterogeneous battery energy storage system cooperative control method based on a distributed preset time observer

Abstract

This invention discloses a cooperative control method for heterogeneous battery energy storage systems based on a distributed preset time observer, comprising the following steps: 1) establishing a heterogeneous battery energy storage system model; 2) establishing a preset time observer based on a time base generator and a heterogeneous battery energy storage system cooperative controller; 3) using the preset time observer to estimate the average battery cell state and average expected power, and transmitting this estimate to the heterogeneous battery energy storage system cooperative controller; 4) the heterogeneous battery energy storage system cooperative controller processes the average battery cell state and average expected power to obtain the microgrid expected power that meets the SoC equalization requirements of the battery cells. Under the premise that only some battery cells can receive the expected power, this invention achieves SoC equalization and power point tracking in heterogeneous battery energy storage systems under detailed balanced directed and undirected graph topologies, reducing communication costs and avoiding single points of failure.

Description

Technical Field

[0001] This invention relates to the field of collaborative control of heterogeneous battery energy storage systems in DC microgrids, specifically a collaborative control method for heterogeneous battery energy storage systems based on a distributed preset time observer. Background Technology

[0002] In recent years, the power system has shown a trend of gradually increasing electricity load and transmission capacity. The high operating costs, operational difficulties, and weak regulation capabilities of large-scale interconnected power grids with centralized generation and long-distance high-voltage transmission have become increasingly prominent, making it difficult to meet users' growing demands for safer, more reliable, diverse, and flexible power supply. To mitigate the impact of large-scale distributed generation on the main power grid, compensate for the insufficient capacity of the power system to support the widespread penetration of distributed generation, and fully leverage the advantages of distributed generation technology, the concept of microgrids has emerged.

[0003] Microgrid loads vary with time and seasons, while power plants generate electricity almost around the clock. If the generated electricity is not used, it results in significant energy waste. On the other hand, during peak electricity demand periods, insufficient power supply can cause power shortages or even blackouts in some areas, leading to severe losses for industrial production. Energy storage technology can effectively solve these problems by storing electrical energy during off-peak hours and then inverting it to the grid during peak periods.

[0004] In the energy sector, driven by growing global market demand and increasingly prominent environmental issues, renewable energy's share in the world's energy mix reached a new high of 13.7% in 2021. Integrating renewable energy into DC microgrids is a crucial strategy for energy conservation, emission reduction, and achieving dual-carbon goals at this stage. However, renewable energy sources (such as wind and solar power) are susceptible to the effects of climate change (e.g., wind speed, solar radiation, temperature), leading to intermittent and unpredictable power output that challenges the normal operation of DC microgrids. Adding energy storage between renewable energy sources and the microgrid ensures a smooth and stable output of new energy power to the grid, enabling large-scale, safe, and reliable grid connection for wind and solar power. This ensures power quality and reliability while reducing energy loss, making energy storage the preferred solution to this problem.

[0005] Energy storage methods mainly include pumped hydro storage, compressed air storage, flywheel energy storage, superconducting magnetic energy storage, and battery energy storage. Superconducting magnetic energy storage faces unresolved technical challenges and is unlikely to be widely applied in the short term. Flywheel energy storage also has many unresolved technical issues, low conversion efficiency, and high-power flywheel energy storage is difficult to implement. Compressed air storage has high safety requirements, and large-capacity compressed air storage typically uses enclosed caverns, requiring specific geographical conditions. Currently, the main energy storage methods used are pumped hydro storage and battery energy storage.

[0006] Pumped hydro storage technology is relatively mature, with large storage capacity and long operating life, making it suitable for large-capacity energy storage in power systems. However, it is greatly affected by water resources and geographical conditions. Compared with other energy storage systems (such as flywheel energy storage, supercapacitor energy storage, and superconducting magnetic energy storage), battery energy storage systems have an irreplaceable position in DC microgrid systems that require flexible configuration and plug-and-play operation due to their advantages of fast response speed, high energy density, and high storage efficiency.

[0007] In recent years, battery energy storage system control technology has developed rapidly, with the relevant theoretical and technical foundations basically mature, and corresponding research and demonstrations are being carried out extensively. However, many technical problems still exist. For example, how to achieve faster and more accurate speed in equalizing the state of charge (SoC) and meeting the charging and discharging power requirements of the battery energy storage system to meet the microgrid requirements; how to control battery energy storage systems with different battery cell capacities; and how to ensure SoC equalization and charging and discharging power tracking of microgrid needs when battery cells have different sensing ranges, i.e., they may not be able to exchange state information. Therefore, researching control strategies for heterogeneous battery energy storage systems has certain theoretical and engineering value. Summary of the Invention

[0008] The purpose of this invention is to provide a cooperative control method for heterogeneous battery energy storage systems based on distributed preset time observers, comprising the following steps:

[0009] 1) Establish a model for a heterogeneous battery energy storage system;

[0010] 2) Establish a pre-set time observer based on a time base generator and a collaborative controller for heterogeneous battery energy storage systems;

[0011] 3) The average battery cell state and average expected power are estimated using a preset time observer and transmitted to the heterogeneous battery energy storage system co-controller.

[0012] 4) The heterogeneous battery energy storage system collaborative controller processes the average battery cell state and average expected power to obtain the microgrid expected power that meets the SoC equalization requirements of the battery cells.

[0013] Furthermore, the steps for establishing a heterogeneous battery energy storage system model include:

[0014] a) Establish the dynamic equations of battery cell i in the heterogeneous battery energy storage system, namely:

[0015]

[0016] In the formula, s i (t), s i (0), C i i i (t), p i (t) and V i (t) represents the SoC value, initial SoC value, battery capacity, output current, output power, and output voltage of battery cell i, respectively; N is the number of battery cells; τ is the integration variable;

[0017] b) Differentiating both sides of the dynamic equation of battery cell i, we get:

[0018]

[0019] c) Establish the state expression for the battery cell, i.e.:

[0020]

[0021] In the formula, x i (t) represents the state of the battery cell at time t.

[0022] Furthermore, the heterogeneous battery energy storage system collaborative controller is shown below:

[0023]

[0024] In the formula, the average cell state p a (t) represents the average expected power; p * p(t) represents the desired power of the microgrid; p(t) represents the output power.

[0025] Furthermore, the preset time observer based on the time base generator includes an average expected power preset time observer and an average battery state preset time observer.

[0026] Furthermore, when the battery cell topology is a detailed balanced directed graph, the average expected power preset time observer is as follows:

[0027]

[0028] In the formula, For battery cells i and j, pair p a The estimated value of (t); a ijThese are the corresponding elements of the adjacency matrix; vector B = dig{b1, b2, ..., b} N When battery cell i can receive information about the desired power, element b i =1, otherwise b i =0; Vector M = L + B is used to represent the detailed balanced directed graph information of the battery cell topology; L is the Laplace matrix; Vector K = dig{k1, k2, ..., k N}; kmax is the largest element in vector K; the elements in vector K satisfy k j a ij =k i a ji ;λ min (MK) is the smallest eigenvalue of matrix MK; constants ψ, σ1, δ1 > 0; v i (t) and u1 are intermediate parameters; p a (t) represents the average expected power;

[0029] The time base generator ε1(t) is shown below:

[0030]

[0031] In the formula, t f t represents the preset time; t represents the actual time.

[0032] Furthermore, when the battery cell topology is a detailed balanced directed graph, the average battery state preset time observer is as follows:

[0033]

[0034] In the formula, For battery cells i and j, the average battery cell state x a Estimates of (t); constant φ>0; τ i (t) is an intermediate parameter.

[0035] Furthermore, when the battery cell topology is an undirected graph, the average expected power preset time observer is as follows:

[0036]

[0037] In the formula, vector H = L + B; L is the Laplace matrix; λ max (H -1 ) is H -1 The largest eigenvalue; u2 is an intermediate parameter; constants α, σ2, δ2>0;

[0038] The time base generator ε2(t) is shown below:

[0039]

[0040] In the formula, t f This is the preset time.

[0041] Furthermore, when the battery cell topology is an undirected graph, the average battery state preset time observer is as follows:

[0042]

[0043] In the formula, λ2(L T L) is a matrix L T The second smallest eigenvalue of L; L is the Laplace matrix; q i q j , u3 is an intermediate parameter. The constant β > 0.

[0044] It is worth noting that the principle of this invention is as follows: each battery cell is considered an intelligent agent, and the heterogeneous battery energy storage system is considered a multi-agent system. Each battery cell can monitor and control its own charging and discharging power, and communicate with adjacent battery cells to exchange its status in real time. Only some battery cells can receive the desired power from the DC microgrid. Through information exchange between battery cells, the active power output of the inverter in the distributed battery energy storage system is adjusted, achieving SoC equalization and power point tracking in the distributed heterogeneous battery energy storage system.

[0045] The technical effects of this invention are undeniable, and its beneficial effects are as follows:

[0046] 1) Under the premise that only some battery cells can receive the desired power, the present invention realizes SoC equalization and power tracking of heterogeneous battery energy storage system under detailed balanced directed graph and undirected graph topology, reducing communication costs and avoiding single point of failure.

[0047] 2) The observer proposed in this invention can estimate the average expected power and average battery state within a preset time. The preset time can be set in advance according to the task, so it does not depend on the initial state and control parameters of the battery energy storage system. This preset time observer does not require excessive gain to achieve, which is more practical. Attached Figure Description

[0048] Figure 1 This is a diagram of a DC microgrid structure with distributed battery energy storage systems and renewable energy sources.

[0049] Figure 2 The topology of a distributed heterogeneous battery energy storage system (detailed balanced directed graph).

[0050] Figure 3The waveform diagram of the battery cell state (discharge state) under different preset time.

[0051] Figure 4 The power waveform (discharge state) of the battery energy storage system is shown at different preset times.

[0052] Figure 5 The waveform diagram of the battery cell state (charging state) under different preset time periods.

[0053] Figure 6 The power waveform (state of charge) of the battery energy storage system is shown at different preset times.

[0054] Figure 7 This is the topology (undirected graph) of a distributed heterogeneous battery energy storage system.

[0055] Figure 8 Battery cell state waveforms under different control strategies (Example 1, discharge state).

[0056] Figure 9 The power waveform diagram of the battery energy storage system under different control strategies (Example 1, discharge state).

[0057] Figure 10 Battery cell state waveforms under different control strategies (Example 1, charging state).

[0058] Figure 11 The power waveform diagram of the battery energy storage system under different control strategies (Example 1, state of charging).

[0059] Figure 12 Power waveforms of battery energy storage systems under different control strategies (Example 2, discharge state).

[0060] Figure 13 Power waveforms of battery energy storage systems under different control strategies (Example 2, state of charging). Detailed Implementation

[0061] The present invention will be further described below with reference to embodiments, but it should not be construed that the scope of the present invention is limited to the following embodiments. Various substitutions and modifications made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention should be included within the scope of protection of the present invention.

[0062] Example 1:

[0063] See Figures 1 to 13 A collaborative control method for heterogeneous battery energy storage systems based on distributed preset time observers includes the following steps:

[0064] 1) Establish a model for a heterogeneous battery energy storage system;

[0065] 2) Establish a pre-set time observer based on a time base generator and a collaborative controller for heterogeneous battery energy storage systems;

[0066] 3) The average battery cell state and average expected power are estimated using a preset time observer and transmitted to the heterogeneous battery energy storage system co-controller.

[0067] 4) The heterogeneous battery energy storage system collaborative controller processes the average battery cell state and average expected power to obtain the microgrid expected power that meets the SoC equalization requirements of the battery cells.

[0068] The steps for establishing a model of a heterogeneous battery energy storage system include:

[0069] a) Establish the dynamic equations of battery cell i in the heterogeneous battery energy storage system, namely:

[0070]

[0071] In the formula, s i (t), s i (0), C i i i (t), p i (t) and V i (t) represents the SoC value, initial SoC value, battery capacity, output current, output power, and output voltage of battery cell i, respectively; N is the number of battery cells; τ is the integration variable;

[0072] b) Differentiating both sides of the dynamic equation of battery cell i, we get:

[0073]

[0074] c) Establish the state expression for the battery cell, i.e.:

[0075]

[0076] In the formula, x i (t) represents the state of the battery cell at time t.

[0077] The heterogeneous battery energy storage system collaborative controller is shown below:

[0078]

[0079] In the formula, the average cell state p a (t) represents the average expected power; p * p(t) represents the desired power of the microgrid; p(t) represents the output power.

[0080] The preset time observer based on the time base generator includes an average expected power preset time observer and an average battery state preset time observer.

[0081] When the battery cell topology is a detailed balanced directed graph, the average expected power preset time observer is as follows:

[0082]

[0083] In the formula, For battery cells i and j, pair p a The estimated value of (t); a ij These are the corresponding elements of the adjacency matrix; vector B = dig{b1, b2, ..., b} N When battery cell i can receive information about the desired power, element b i =1, otherwise b i =0; Vector M = L + B is used to represent the detailed balanced directed graph information of the battery cell topology; L is the Laplace matrix; Vector K = dig{k1, k2, ..., k N}; kmax is the largest element in vector K; the elements in vector K satisfy k j a ij =k i a ji ;λ min (MK) is the smallest eigenvalue of matrix MK; constants ψ, σ1, δ1 > 0; v i (t) and u1 are intermediate parameters; p a (t) represents the average expected power;

[0084] The time base generator ε1(t) is shown below:

[0085]

[0086] In the formula, t f t represents the preset time; t represents the actual time.

[0087] When the battery cell topology is a detailed balanced directed graph, the average battery state preset time observer is as follows:

[0088]

[0089] In the formula, For battery cells i and j, the average battery cell state x a Estimates of (t); constant φ>0; τ i (t) is an intermediate parameter.

[0090] When the battery cell topology is an undirected graph, the average expected power preset time observer is as follows:

[0091]

[0092] In the formula, vector H = L + B; L is the Laplace matrix; λ max (H -1 ) is H -1 The largest eigenvalue; u2 is an intermediate parameter; constants α, σ2, δ2>0;

[0093] The time base generator ε2(t) is shown below:

[0094]

[0095] In the formula, t f This is the preset time.

[0096] When the battery cell topology is an undirected graph, the average battery state preset time observer is as follows:

[0097]

[0098] In the formula, λ2(L T L) is a matrix L T The second smallest eigenvalue of L; L is the Laplace matrix; q i q j , u3 is an intermediate parameter. The constant β > 0.

[0099] Example 2:

[0100] A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer includes the following steps:

[0101] S1: Heterogeneous Battery Energy Storage System Model

[0102] According to the amperometric method, the dynamics of battery cell i in the heterogeneous battery energy storage system are as follows:

[0103]

[0104] Where s i (t), s i (0), C i i i (t), p i (t) and V i (t) represents the SoC value, initial SoC value, battery capacity, output current, output power, and output voltage of battery cell i, respectively.

[0105] Differentiate both sides of (1)

[0106]

[0107] Define battery cell state

[0108]

[0109] S2: Collaborative Control Strategy for Heterogeneous Battery Energy Storage Systems

[0110] The heterogeneous battery energy storage system co-controller is

[0111]

[0112] in, p * (t) represents the expected power of the microgrid.

[0113] S3: Preset time observer based on time base generator

[0114] Because x a (t), p a (t) represents global information, which is not available for all battery cells. Therefore, an observer needs to be designed to estimate the global information required by the controller.

[0115] The following is a preset time observer for the average expected power of a detail-balanced directed graph:

[0116]

[0117] in, For the i-th battery cell pair p a The estimated value of (t), a ij These are the corresponding elements of the adjacency matrix. B = dig{b1, b2, ..., b} N When battery cell i can receive information about the desired power, b i =1, otherwise b i = 0. M = L + B, where L is the Laplace matrix. There exists K = dig{k1, k2, ..., k N} makes k j a ij =k i a ji If so, the diagram is called a detail-balanced diagram. min (MK) is the smallest eigenvalue of matrix MK.

[0118] It is a time base generator, where ψ, σ1, and δ1 are the positive parameters to be designed. 0 < δ1 < < 1. ρ is The upper limit of k; min Let K be the smallest element in vector K.

[0119] The following is a preset time observer for the average battery state suitable for detail-balanced directed graphs:

[0120]

[0121] in, For the i-th battery cell pair x a The estimated value of (t), where φ is the positive value to be designed, and These are the upper and lower limits of power.

[0122] The following are preset time observers for the average expected power of undirected graphs:

[0123]

[0124] Where H = L + B, λ max (H -1 ) is H -1 The largest eigenvalue, ε2(t), is a time base generator that satisfies the following structure:

[0125]

[0126] Where α, σ², and δ² are the positive parameters to be designed. 0 < δ2 < < 1.

[0127] The following is a preset time observer for the average battery state of an undirected graph:

[0128]

[0129] Wherein, λ2(L T L) is a matrix L T The second smallest eigenvalue of L, λ2(L) is the second smallest eigenvalue of matrix L.

[0130] Using the above control strategy, the power required for SoC equalization and power point tracking can be achieved when the topology details of a heterogeneous battery energy storage system are balanced in a directed / undirected graph.

[0131] Example 3:

[0132] A verification experiment of a cooperative control method for a heterogeneous battery energy storage system based on a distributed preset time observer is conducted, including:

[0133] In this specific embodiment, the structure diagram of a DC microgrid with a distributed battery energy storage system and renewable energy is as follows: Figure 1 As shown, the distributed battery energy storage system supplies power to the common load or local load through its respective inverters connected in parallel.

[0134] S1: Heterogeneous Battery Energy Storage System Model

[0135] The dynamics of the i-th battery cell in the heterogeneous battery energy storage system are as follows:

[0136]

[0137] Where s i (t), s i (0), C i i i (t), p i (t) and V i (t) represents the SoC value, initial SoC value, battery capacity, output current, output power, and output voltage of the i-th battery cell, respectively.

[0138] Differentiate both sides of (1)

[0139]

[0140] Define battery cell state

[0141]

[0142] S2: Collaborative Control Strategy for Heterogeneous Battery Energy Storage Systems

[0143] The heterogeneous battery energy storage system co-controller is

[0144]

[0145] in, p * (t) represents the expected power of the microgrid.

[0146] S3: Preset time observer based on time base generator

[0147] Because x a (t), p a (t) represents global information, which is not available for all battery cells. Therefore, an observer needs to be designed to estimate the global information in the controller.

[0148] The following is a preset time observer for the average expected power of a detail-balanced directed graph:

[0149]

[0150] in, For the i-th battery cell pair p a The estimated value of (t), a ij These are the corresponding elements of the adjacency matrix. B = dig{b1, b2, ..., b} NWhen the i-th battery cell can receive the desired power information, b i =1, otherwise b i = 0. M = L + B, where L is the Laplace matrix. K = dig{k1, k2, ..., k N} contains information related to the topology, λ min (MK) is the smallest eigenvalue of matrix MK. ε1(t) is a time-base generator that satisfies the following relationship: t f This is the preset time. ψ, σ1, and δ1 are the positive values ​​to be designed, where... 0 < δ1 < < 1.

[0151] The following is a preset time observer for the average battery state suitable for detail-balanced directed graphs:

[0152]

[0153] in, For the i-th battery cell pair x a The estimated value of (t), where φ is the positive value to be designed, and

[0154] Figure 2 Considering the topology (balanced details) of the distributed battery energy storage system in this embodiment, the heterogeneous battery energy storage system has 6 battery cells with parameters V = 20V, cell capacity C of (190, 215, 230, 210, 205, 235) Ah, and initial SoC of (0.95, 0.86, 0.83, 0.93, 0.97, 0.88). Only battery cell number 1 can receive the desired power information. Furthermore, assume the required power is as follows: p * (t)=±(4200sin(t)+4200)W.

[0155] The parameters of the preset time observers (5) and (6) are selected as σ1=2, ψ=1400, φ=2800, K=dig{1,2,2.5,1.25,2,1.6}. The initial state of the observers (5) and (6) is selected as v i (0) = 0,

[0156] Figures 3-6 The waveforms of battery cell states and power tracking of the battery energy storage system under different preset times are selected when the control strategy is applied to the heterogeneous battery energy storage system in two states: charging and discharging. The preset times are 0.5h and 3h.

[0157] Compared to existing control strategies, this control strategy is suitable for battery energy storage systems with a detailed balanced directed graph topology. Furthermore, the smaller the preset time, the faster the convergence speed, making it more practically valuable.

[0158] The following are preset time observers for the average expected power of undirected graphs:

[0159]

[0160] Where H = L + B, λ max (H -1 ) is H -1 The largest eigenvalue, ε2(t), is the time base generator. α, σ², and δ² are the positive values ​​that need to be designed, where 0 < δ2 < < 1.

[0161] The following are preset time observers for the average expected power of undirected graphs:

[0162]

[0163] Where λ2(L T L) is a matrix L T The second smallest eigenvalue of L,

[0164] Figure 7 Considering the topology (undirected graph) of the distributed battery energy storage system in the embodiment, the parameters of the preset time observers (7) and (8) are selected as σ² = 10, α = 1000, β = 3430, and the preset time is selected as t. f =0.5h. For observers (7) and (8), the initial state is chosen as v. i (0) = 0,

[0165] In specific implementations, different control strategies are employed to compare and verify the effectiveness and superiority of the invention. Control methods based on finite-time observers and fixed-time observers are selected for comparison with the invention.

[0166] Example 1: Required power p * (t)=±(4200sin(t)+4200)W.

[0167] Figures 8-11 The waveforms of the battery cell state and the power of the battery energy storage system are compared under different control strategies in the charging and discharging states.

[0168] Example 2: Required power:

[0169]

[0170] Figures 12-13 The waveforms of the power of the battery energy storage system are compared under different control strategies in both charging and discharging states.

[0171] Solid lines represent simulation results under the control method of this invention, dashed lines represent simulation results based on the finite-time observer control strategy, dashed lines represent simulation results based on the fixed-time observer control strategy, and thickened solid lines represent the desired power p. * (t). It can be seen that the control method proposed in this invention is superior to control methods based on finite-time observers and fixed-time observers.

[0172] Example 4:

[0173] A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer includes the following steps:

[0174] 1) Establish a model for a heterogeneous battery energy storage system;

[0175] 2) Establish a pre-set time observer based on a time base generator and a collaborative controller for heterogeneous battery energy storage systems;

[0176] 3) The average battery cell state and average expected power are estimated using a preset time observer and transmitted to the heterogeneous battery energy storage system co-controller.

[0177] 4) The heterogeneous battery energy storage system collaborative controller processes the average battery cell state and average expected power to obtain the microgrid expected power that meets the SoC equalization requirements of the battery cells.

[0178] Example 5:

[0179] A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer is described in Example 4. The steps for establishing a model of the heterogeneous battery energy storage system include:

[0180] 1) Establish the dynamic equations of battery cell i in the heterogeneous battery energy storage system, namely:

[0181]

[0182] In the formula, s i (t), s i (0), C i i i (t), p i (t) and V i(t) represents the SoC value, initial SoC value, battery capacity, output current, output power, and output voltage of battery cell i, respectively; N is the number of battery cells; τ is the integration variable;

[0183] 2) Differentiating both sides of the kinetic equation for battery cell i, we get:

[0184]

[0185] 3) Establish the state expression of the battery cell, that is:

[0186]

[0187] In the formula, x i (t) represents the state of the battery cell at time t.

[0188] Example 6:

[0189] A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer is described in Example 4. The collaborative controller for the heterogeneous battery energy storage system is shown below:

[0190]

[0191] In the formula, the average cell state p a (t) represents the average expected power; p * p(t) represents the desired power of the microgrid; p(t) represents the output power.

[0192] Example 7:

[0193] A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer is described in Example 4. The preset time observer based on the time base generator includes an average expected power preset time observer and an average battery state preset time observer.

[0194] Example 8:

[0195] A collaborative control method for heterogeneous battery energy storage systems based on distributed preset time observers is described in Example 4. When the battery cell topology is a detailed balanced directed graph, the average expected power preset time observer is as follows:

[0196]

[0197] In the formula, For battery cells i and j, pair p a The estimated value of (t); a ij These are the corresponding elements of the adjacency matrix; vector B = dig{b1, b2, ..., b}N When battery cell i can receive information about the desired power, element b i =1, otherwise b i =0; Vector M = L + B is used to represent the detailed balanced directed graph information of the battery cell topology; L is the Laplace matrix; Vector K = dig{k1, k2, ..., k N}; kmax is the largest element in vector K; the elements in vector K satisfy k j a ij =k i a ji ;λ min (MK) is the smallest eigenvalue of matrix MK; constants ψ, σ1, δ1 > 0; v i (t) and u1 are intermediate parameters; p a (t) represents the average expected power;

[0198] The time base generator ε1(t) is shown below:

[0199]

[0200] In the formula, t f t represents the preset time; t represents the actual time.

[0201] Example 9:

[0202] A collaborative control method for heterogeneous battery energy storage systems based on a distributed preset time observer is described in Example 4. When the battery cell topology is a detailed balanced directed graph, the average battery state preset time observer is as follows:

[0203]

[0204] In the formula, For battery cells i and j, the average battery cell state x a Estimates of (t); constant φ>0; τ i (t) is an intermediate parameter.

[0205] Example 10:

[0206] A collaborative control method for heterogeneous battery energy storage systems based on a distributed preset time observer is described in Example 4. When the battery cell topology is an undirected graph, the average expected power preset time observer is as follows:

[0207]

[0208] In the formula, vector H = L + B; L is the Laplace matrix; λ max (H -1 ) is H-1 The largest eigenvalue; u2 is an intermediate parameter; constants α, σ2, δ2>0;

[0209] The time base generator ε2(t) is shown below:

[0210]

[0211] In the formula, t f This is the preset time.

[0212] Example 11:

[0213] A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer is described in Example 4. When the battery cell topology is an undirected graph, the average battery state preset time observer is as follows:

[0214]

[0215] In the formula, λ2(L T L) is a matrix L T The second smallest eigenvalue of L; L is the Laplace matrix; q i q j , u3 is an intermediate parameter. The constant β > 0.

Claims

1. A collaborative control method for a heterogeneous battery energy storage system based on a distributed preset time observer, characterized in that, Includes the following steps: Step 1. Establish the model of the heterogeneous battery energy storage system; Step 2. Establish a pre-set time observer based on a time base generator and a collaborative controller for heterogeneous battery energy storage systems; Step 3. Use a preset time observer to estimate the average battery cell state and average expected power, and transmit the results to the heterogeneous battery energy storage system co-controller; Step 4. The heterogeneous battery energy storage system co-controller processes the average battery cell state and average expected power to obtain the microgrid expected power that meets the SoC equalization requirements of the battery cells; The preset time observer based on the time base generator includes an average expected power preset time observer and an average battery state preset time observer; When the battery cell topology is a detailed balanced directed graph, the average expected power preset time observer is as follows: （5） In the formula, , For battery cells i and j The estimated value; These are the corresponding elements of the adjacency matrix; vectors When battery cell i can receive information about the desired power, element ,otherwise ;vector Used to characterize the detailed balance directed graph information of the battery cell topology; It is a Laplace matrix; a vector kmax is the largest element in vector K; the elements in vector K satisfy... ; It is a matrix The smallest eigenvalue; a constant >0; v i (t) and u1 are intermediate parameters; p a (t) represents the average expected power; Among them, time base generator As shown below: （6） In the formula, t f t represents the preset time; t is the actual time. When the battery cell topology is an undirected graph, the average expected power preset time observer is as follows: （8） In the formula, vector ; It is a Laplace matrix; yes The largest eigenvalue; u2 is an intermediate parameter; a constant. >0; Among them, time base generator As shown below: （9） In the formula, t f This is the preset time.

2. The method for coordinated control of a heterogeneous battery energy storage system based on a distributed preset time observer according to claim 1, characterized in that, The steps for establishing a model of a heterogeneous battery energy storage system include: Step 1. Establish the dynamic equations of battery cell i in the heterogeneous battery energy storage system, namely: （1） In the formula, , , , , and These are the SoC value, initial SoC value, battery capacity, output current, output power, and output voltage of battery cell i, respectively; N is the number of battery cells. It is an integral variable; Step 2. Differentiate both sides of the kinetic equation for battery cell i to obtain: （2） Step 3. Establish the state expression of the battery cell, that is: （3） In the formula, x i (t) represents the state of the battery cell at time t.

3. The method for coordinated control of a heterogeneous battery energy storage system based on a distributed preset time observer as described in claim 1, characterized in that, The heterogeneous battery energy storage system collaborative controller is shown below: （4） In the formula, the average cell state ;p a (t) represents the average expected power; p(t) represents the expected power of the microgrid; p(t) represents the output power.

4. The method for coordinated control of a heterogeneous battery energy storage system based on a distributed preset time observer according to claim 1, characterized in that, When the battery cell topology is a detailed balanced directed graph, the average battery state preset time observer is as follows: （7） In the formula, , For battery cells i and j, the average battery cell state Estimates; constants >0; This is an intermediate parameter.

5. The method for coordinated control of a heterogeneous battery energy storage system based on a distributed preset time observer according to claim 1, characterized in that, When the battery cell topology is an undirected graph, the average battery state preset time observer is as follows: （10） In the formula, It is a matrix The second smallest eigenvalue; It is the Laplace matrix; q i q j , u3 is an intermediate parameter; a constant. >0.

Images