Initial black start frequency and voltage distributed optimization control method and system in virtual power plant mode

Abstract

The invention relates to the technical field of electric power, and discloses an initial black-start frequency and voltage distributed optimization control method and system in a virtual power plant mode, and the method comprises the steps: building a virtual power plant structure relation, and reducing the state difference between adjacent agent nodes of a distributed multi-agent system until the state of each node is converged to a consistent value; establishing an initial black-start control model of the distributed new energy of the virtual power plant by taking the minimum transient frequency and the minimum voltage deviation value as optimization objectives; and overall optimization and power scheduling of initial black-start voltage and frequency control of the distributed power supply cluster are carried out through a consistency algorithm in combination with a hierarchical control mode. The method not only has better control stability, but also has stronger robustness and reliability for operation scenes such as new energy output fluctuation and unit random commissioning and quitting.

Description

Technical Field

[0001] The present invention relates to the field of power technology, and in particular to a method and system for distributed optimization control of initial black-start frequency and voltage under the virtual power plant mode. Background Art

[0002] Building a new power system with new energy as the main body is an important measure to promote the construction of a clean, low-carbon, safe and efficient modern energy system in China and accelerate the realization of the "dual-carbon" goal. Extreme scenarios resulting from the superposition of the strong volatility of new energy output and the inherent uncertainty of power grid operation are extremely likely to trigger large-scale systematic power outages such as the "9.28" in Australia and the "8.9" in the UK. Black start refers to the self-healing technology that, after a large-scale power outage of the system, without relying on other external assistance, uses black-start units with self-starting capabilities in the original network to restart the power-off system. With the large-scale grid connection of new energy, its advantages of low plant power consumption rate and fast startup speed make it have great potential as a black-start power source. Therefore, black-start research considering new energy is essential.

[0003] In order to achieve the effective aggregation of distributed energy in a wider area and participate in power system operation and power market transactions through the optimal scheduling of various controllable resources such as adjustable loads and energy storage, virtual power plant (VPP) technology has emerged.

[0004] Classified by function, virtual power plants are divided into commercial virtual power plants (CVPP) and technical virtual power plants (TVPP). The basic function of CVPP is to optimize the power generation plan based on load forecasting and power generation potential forecasting and participate in market bidding. It does not consider the impact of virtual power plants on the power grid and participates in the power market in the same way as traditional power plants. At present, the related research on VPP focuses on providing a framework and technical support for the access of distributed new energy power, promoting the improvement of China's power market system, and many beneficial results have been achieved. However, there are few reports on the use of VPP for black start. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the specification of this application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a method for distributed optimization control of initial black-start frequency and voltage under the virtual power plant mode to solve the above problems.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a distributed optimization control method for the initial black start frequency and voltage in a virtual power plant mode, including: establishing the structural relationship of the virtual power plant, reducing the state difference between adjacent agent nodes in the distributed multi-agent system until the states of each node converge to a consistent value;

[0009] Based on the structural relationship of the virtual power plant, defining the operating constraints of the virtual power plant, taking the minimum transient frequency and voltage deviation as the optimization goal, and establishing a distributed new energy initial black start control model for the virtual power plant;

[0010] Based on the initial black start control model, through the consensus algorithm and combined with the hierarchical control method, the overall optimization of the initial black start voltage and frequency control and power scheduling of the distributed power cluster are carried out.

[0011] As a preferred solution of the distributed optimization control method for the initial black start frequency and voltage in the virtual power plant mode of the present invention, wherein: establishing the structural relationship of the virtual power plant includes:

[0012] Establishing the relationship between nodes in the multi-agent of the virtual power plant through the adjacency matrix, and introducing the Laplace matrix to represent the influence of information feedback and state differences between agents on the control quantity;

[0013] Calculating the control variables of the agent to obtain the change of the agent state, and the agent control variables are calculated through the state difference between adjacent agents;

[0014] Controlling each agent node to continuously adjust according to the adjacent nodes through the first-order consensus algorithm, reducing the state difference between any adjacent agent nodes, that is, the control variables of each node converge consistently.

[0015] As a preferred solution of the distributed optimization control method for the initial black start frequency and voltage in the virtual power plant mode of the present invention, wherein: defining the operating constraints of the virtual power plant includes:

[0016] Obtaining the change amounts of the frequency and voltage at the power output end of the distributed power source, and defining the operating constraints of the virtual power plant to determine the frequency and voltage transient stability during the load connection process;

[0017] The change amounts of the frequency and voltage at the power output end of the distributed power source are inversely proportional to the change amounts of the active power and reactive power, and are respectively expressed as:

[0018]

[0019] wherein, m i and n iare the droop coefficients of frequency and voltage characteristics respectively, and ω, V, P, and Q are angular frequency, voltage, active power, and reactive power respectively;

[0020] The operating constraint conditions are expressed as:

[0021]

[0022] Among them, L represents the power exchange command value, P L and Q L represent the active and reactive power magnitudes of the connected load respectively, P i.max represents the rated active installed capacity of the distributed power source, Q i.max represents the maximum reactive power that the distributed power source can absorb, Q i.min represents the maximum reactive power that the distributed power source can send out, Δω i represents the carrier angular frequency of the inverters of each distributed power source, ΔV i represents the voltage fluctuation value of each power supply port.

[0023] As a preferred scheme of the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode of the present invention, among them: the initial black-start control model of the distributed new energy based on the virtual power plant is expressed as:

[0024]

[0025] Among them, λ represents the Lagrange multiplier corresponding to the equality constraint.

[0026] As a preferred scheme of the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode of the present invention, among them: the hierarchical control method includes:

[0027] The primary control layer, which controls active power - frequency and reactive power - voltage through the distributed power source droop control strategy to perform basic regulation of frequency and voltage;

[0028] The secondary control layer, which eliminates the deviation generated by the primary control layer and maintains the frequency and voltage within the rated value range;

[0029] The tertiary control layer, which is the dispatching layer and controls the output of each distributed power source and the power flow between the virtual power plant and the outside world.

[0030] As a preferred scheme of the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode of the present invention, among them: the primary control layer includes:

[0031] Through the distributed power source droop control strategy, automatic distribution of power among multiple power sources is performed, and the distributed power source voltage droop characteristic is expressed as:

[0032]

[0033] Among them, ω i and V i represent the carrier angular frequency and output terminal voltage of the i-th distributed power source, and V N represents the rated output voltage of the distributed power source, and ω N represents the rated angular frequency, m i and n i respectively represent the frequency of droop control and the voltage droop control coefficient. ΔP i and ΔQ i represent the changes in the active and reactive power outputs of the i-th distributed power source.

[0034] As a preferred solution of the initial black start frequency and voltage distributed optimization control method in the virtual power plant mode described in the present invention, where: the secondary control layer includes:

[0035] The secondary control of the distributed power source includes voltage control and power control. The voltage control is used to restore the bus voltage to the rated value, and the power control adjusts the power distribution among the distributed power sources;

[0036] The control target of the voltage control is that the bus voltages of all nodes reach the rated value;

[0037] For the power control, the droop control amount is corrected through a consensus algorithm. The voltages and frequencies of the distributed power source units are taken, and the change amounts based on the droop characteristics are used as the state variables of the intelligent agent nodes, and a secondary power control is designed to make the state variables tend to be consistent.

[0038] In a second aspect, the present invention provides an initial black start frequency and voltage distributed optimization control system in a virtual power plant mode, including:

[0039] The first construction module is used to establish the structural relationship of the virtual power plant, reduce the state difference between adjacent intelligent agent nodes of the distributed multi-intelligent agent system until the states of all nodes converge to a consistent value;

[0040] The second construction module is used to define the operation constraint conditions of the virtual power plant based on the structural relationship of the virtual power plant, and establish an initial black start control model for the distributed new energy of the virtual power plant with the minimum transient frequency and voltage deviation as the optimization target;

[0041] The control module is used to perform overall optimization and power scheduling of the initial black start voltage and frequency control of the distributed power source cluster based on the initial black start control model through a consensus algorithm and in combination with a hierarchical control method.

[0042] In a third aspect, the present invention provides an electronic device, comprising:

[0043] a memory and a processor;

[0044] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method for optimizing the control of the initial black-start frequency and voltage distribution in the virtual power plant mode are realized.

[0045] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for optimizing the control of the initial black-start frequency and voltage distribution in the virtual power plant mode.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention uses virtual power plant technology to unite geographically dispersed distributed new energy units to participate in black start. First, an initial black-start control model for a distributed new energy unit cluster is established with the minimum transient frequency and voltage deviation as the optimization goal. Secondly, the multi-agent consensus algorithm is used in combination with a hierarchical control method to realize the overall optimization of the initial black-start voltage and frequency control and power scheduling of the distributed power cluster. Compared with traditional centralized control, this method not only has better control stability, but also has stronger robustness and reliability for operation scenarios such as new energy output fluctuations and random connection and disconnection of units. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0048] Figure 1 It is a schematic diagram of the overall process of the method for optimizing the control of the initial black-start frequency and voltage distribution in the virtual power plant mode according to an embodiment of the present invention;

[0049] Figure 2 It is a schematic diagram of a virtual power plant of the method for optimizing the control of the initial black-start frequency and voltage distribution in the virtual power plant mode according to an embodiment of the present invention;

[0050] Figure 3 It is a schematic diagram of a control structure based on a consensus algorithm of the method for optimizing the control of the initial black-start frequency and voltage distribution in the virtual power plant mode according to an embodiment of the present invention;

[0051] Figure 4Schematic diagram of the initial black - start system for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0052] Figure 5 Schematic diagram of the initial black - start voltage waveform of the distributed power generation cluster for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0053] Figure 6 Schematic diagram of the initial black - start voltage waveform of a single photovoltaic - energy storage power station for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0054] Figure 7 Schematic diagram of the frequency fluctuation of the distributed power generation cluster with load during the initial black - start for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0055] Figure 8 Schematic diagram of the frequency fluctuation of a single photovoltaic - energy storage power station with load during the initial black - start for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0056] Figure 9 Schematic diagram of the convergence of the voltage control consistency variable in Scenario 1 for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0057] Figure 10 Schematic diagram of the convergence of the frequency control consistency variable in Scenario 1 for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0058] Figure 11 Schematic diagram of the initial black - start voltage waveform when the system power output fluctuates for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0059] Figure 12 Schematic diagram of the initial black - start frequency waveform when the system power output fluctuates for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0060] Figure 13 Schematic diagram of the convergence of the voltage control consistency variable in Scenario 2 for the initial black - start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0061] Figure 14Schematic diagram of the convergence of frequency control consistency variables in Scenario 2 of the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0062] Figure 15 Schematic diagram of the initial black-start voltage waveform when some power sources are taken out of operation in the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0063] Figure 16 Schematic diagram of the initial black-start frequency waveform when some power sources are taken out of operation in the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0064] Figure 17 Schematic diagram of the convergence of voltage control consistency variables in Scenario 3 of the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention;

[0065] Figure 18 Schematic diagram of the convergence of frequency control consistency variables in Scenario 3 of the initial black-start frequency and voltage distributed optimization control method under the virtual power plant mode according to an embodiment of the present invention. Detailed implementation manners

[0066] To make the above objects, features and advantages of the present invention more obvious and understandable, the following detailed description of the specific implementation manners of the present invention is made in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0067] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0068] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude other embodiments.

[0069] The present invention will be described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention, for the convenience of explanation, the cross-sectional views showing the device structure will be locally enlarged in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0070] Meanwhile, in the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper, lower, inner, and outer" is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0071] Unless otherwise clearly specified and defined in the present invention, the terms "installation, connection, and coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, and can also be indirectly connected through an intermediate medium, or can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0072] Embodiment 1

[0073] Referring to Figures 1 - 3 , an embodiment of the present invention provides an initial black-start frequency and voltage distributed optimization control method under a virtual power plant mode. As Figure 1 shown, it includes:

[0074] S101, establishing the structural relationship of the virtual power plant, reducing the state difference between adjacent agent nodes of the distributed multi-agent system until the states of each node converge to a consistent value;

[0075] S102, based on the structural relationship of the virtual power plant, defining the operation constraint conditions of the virtual power plant, and establishing an initial black-start control model for the distributed new energy of the virtual power plant with the minimum transient frequency and voltage deviation as the optimization goal;

[0076] S103, based on the initial black-start control model, through the consensus algorithm and combined with the hierarchical control method, performing the overall optimization and power scheduling of the initial black-start voltage and frequency control of the distributed power cluster.

[0077] It should be noted that this embodiment is based on the background of a large-scale power outage on the user side. A distributed optimization control method for transient voltage and frequency is proposed for black-starting important loads using a distributed new energy unit cluster in the VPP mode. First, with the minimum transient frequency and voltage deviation as the optimization goal, a distributed new energy initial black-start control model based on a virtual power plant is established. Second, the overall optimization and power scheduling of the initial black-start voltage and frequency control of the distributed power cluster are realized by using the consensus algorithm and combining hierarchical control. Finally, the effectiveness of the method proposed in the present invention is verified through a numerical example.

[0078] A virtual power plant is an intelligent energy system that aggregates one or more resources in different spaces and participates in the operation of the power system and power market transactions through autonomous coordinated optimization control. It can be equivalent to a power plant participating in various power markets such as capacity, electricity, and ancillary services, and obtaining corresponding economic benefits while ensuring the stable operation of the system. Adjustable (interruptible) loads, distributed power sources, and energy storage are the main basic resources for the development of virtual power plants. In order to effectively aggregate these resources and achieve a rapid response to grid dispatching instructions, from a control perspective, a virtual power plant presents as a multi-agent system (MAS), and the individual resources participating in the aggregation are all autonomous or semi-autonomous subsystems with functions of environmental perception, problem-solving, and communication and interaction, that is, agents. Based on network and communication technologies, agents are interconnected and exchange messages, and coordinate and optimize their behaviors through distributed algorithms, and interact with each other to achieve common control goals. A schematic diagram of a virtual power plant is as Figure 2 shown.

[0079] While a virtual power plant reliably and efficiently aggregates various resources, it should also have robustness and scalability to meet the needs of the dynamically changing aggregated resources. The distributed coordination and cooperation ability of agents in MAS is the key to achieving the above control goals of a virtual power plant, which is reflected in how agents transmit information and through mutual interaction and influence, make a certain state of multiple agents tend to be consistent, that is, maintain consistency. The communication mechanism and control protocol adopted by agents to achieve a consistent goal constitute the consensus algorithm.

[0080] In a preferred embodiment, establishing the structural relationship of a virtual power plant includes:

[0081] Establish the relationship between nodes in the multi-agent of the virtual power plant through an adjacency matrix, and introduce a Laplacian matrix to represent the influence of information feedback and state differences between agents on the control quantity;

[0082] Obtain the change of agent state by calculating the control variable of the agent, and the agent control variable is calculated by the state difference between adjacent agents;

[0083] Through the first-order consensus algorithm, each agent node is controlled to continuously adjust according to its adjacent nodes, reducing the state difference between any two adjacent agent nodes, that is, the control variables of each node converge uniformly.

[0084] In an alternative embodiment, the adjacency matrix A = (a ij ) n×n is used to describe the relationship between nodes in the multi-agent system, where n is the number of nodes. If there is an edge from the j-th node to the i-th node, then a ij = 1, i ≠ j, otherwise a ij = 0, and the diagonal element a ii = 0. On this basis, to characterize the influence of information feedback and state differences between agents on the control quantity, the Laplace matrix L = (l ij ) is further introduced, and the definition of l ij is as follows:

[0085]

[0086] For a distributed MAS with n agents, each agent communicates with its neighboring agents, and the neighboring agents feedback information. The change in the state of agent i depends on its control variable u i , which is determined by the state differences between it and all its adjacent agents, and is expressed as:

[0087]

[0088] where, x i is the state variable of agent i, which can represent physical quantities such as voltage, current, frequency, and power of the actual system. N i represents the set of neighboring nodes connected to agent i. The above formula can be written in the following matrix form, expressed as:

[0089] U = -LX

[0090] where, L is the Laplace matrix of the MAS, which is determined by the communication network topology structure of all agents that make up the MAS.

[0091] The first-order consensus algorithm is expressed as:

[0092] U(k + 1) = DU(k)

[0093] where, k is the number of iterations, and D is a random matrix based on the communication network topology structure of the MAS and with non-zero diagonal elements, so that the algorithm can adapt to the random dropout of nodes and meet the plug-and-play requirements of agents.

[0094] For agent i, the first-order consensus algorithm is specifically as follows:

[0095]

[0096] where d ij is the state transition coefficient that constitutes the state transition matrix, and max(n i , n j ) represents the maximum number of neighbors of an agent node and its neighbor nodes in the MAS.

[0097] The formula of the first-order consensus algorithm shows that each agent node in the distributed MAS continuously adjusts according to its adjacent nodes. As k gradually increases, the state difference between any two adjacent agent nodes will be small enough, that is, the consensus variables (control variables) of each node will converge uniformly. When any M agent nodes among them all satisfy the following formula, it represents that the entire system converges, which is expressed as:

[0098]

[0099] where the value of M is related to the convergence speed. The smaller the value of M, the faster the convergence speed. Generally, the value can be 1 - 5, and ε needs to be small enough, usually taking the order of magnitude of 10-2.

[0100] Under the action of the above consensus control law, the state difference between adjacent agent nodes in the distributed MAS can be reduced until the states of all nodes finally converge to a consistent value, that is: x 1∞ = x 2∞ =... = x n∞ .

[0101] It should be noted that the single-unit capacity of distributed new energy units distributed on the load side is small and the layout is relatively scattered. If we want to utilize their advantage of rapid startup to provide startup power support for power outage emergency restoration, the system dispatching must have the effective aggregation ability of the power generation capacity of distributed new energy units, that is: while suppressing the output fluctuations of new energy units and providing stable power output and power support for important loads, it can also effectively control the frequency and voltage fluctuations that may occur during the startup of the units until the load is restored. The VPP technology highly matches this requirement. Therefore, the present invention proposes an initial black start optimization control strategy based on the VPP multi-agent consensus algorithm.

[0102] As previously mentioned, using new energy units for black start usually includes main steps such as unit self-startup, charging no-load lines, and connecting to important loads.

[0103] (1) Briefly describe the voltage control process and requirements

[0104] Unit self-starting: A distribution transformer may be installed on the interface side of the distributed power inverter. A too rapid increase in the bus voltage amplitude may cause the distribution transformer to experience flux saturation, resulting in overvoltage and affecting the system recovery. Therefore, it is necessary to control the starting speed of the black start power source to ensure its zero-start voltage rise.

[0105] Charging an unloaded line: Distributed power sources are small-capacity power generation facilities with a capacity of 10 kW - 50 MW, distributed near the load with a transmission distance not exceeding 50 km. The voltage levels are mostly 220 V, 380 V, 10 kV, and 35 kV. The capacitance and inductance in the transmission line are energy storage elements. When performing a closing and charging operation on it, there will be a transient process, which can generate an operating overvoltage several times that of the power supply voltage. According to the regulations, for a system of 220 kV and below, the maximum allowable operating overvoltage is 3 times, and for a system below 35 kV, it should not be greater than 4 times.

[0106] Connecting a load: After the line charging is completed, due to the incomplete restoration of the system grid, the initial impact during the load connection may cause a relatively serious transient voltage drop in the weak system. Therefore, when connecting a load, the system needs to consider transient voltage safety, that is, after a fault, the system can ensure that the duration of the load node voltage being lower than a given value does not exceed a predetermined time period. Otherwise, the voltage is considered transiently unsafe. The State Grid of China stipulates that the time when the transient voltage is less than 0.75 p.u. should not exceed 1 s.

[0107] In the initial stage of new energy black start, due to the small capacity of distributed new energy power plants and the voltage levels sent out being below 220 kV, and the transmission line length being within 300 km (at this time, the capacitance rise problem has a relatively small impact on the operating overvoltage), the most likely factor causing voltage collapse is the process of load restoration.

[0108] (2) Frequency control requirements during the initial black start process

[0109] During the initial black start process, it is necessary to focus on ensuring the transient frequency stability when connecting a load. The initial impact during the load connection may cause a serious frequency drop in the system. The main factor affecting the frequency deviation is the proportion of the power deficit in the total system capacity. The black start safety regulations require that the frequency fluctuation cannot exceed ±0.5 Hz, that is, the frequency needs to be controlled within the range of 49.5 - 50.5 Hz.

[0110] In summary, during the entire initial black start process, the most important thing to pay attention to is the system transient stability during the load connection stage, including transient voltage stability and transient frequency stability.

[0111] When a large number of new energy power stations are connected to the power grid, they need to cooperate with each other to participate in frequency modulation and voltage regulation. At the same time, reactive power voltage regulation needs to follow the basic principles of hierarchical and zonal adjustment and reactive power local balance. Therefore, while each distributed power source achieves the consistent goal of frequency modulation and voltage regulation, it should also minimize the flow and loss of reactive power. The active and reactive powers of each distributed power source need to be scheduled and controlled through a virtual power plant.

[0112] The virtual power plant in this embodiment is composed of a wind farm and a photovoltaic power station. Since there are no problems such as generator in-phase in new energy units, and in addition, the degree of voltage amplitude drop during the black start process is closely related to the electrical distance between the load and the power source point, the steady-state power frequency overvoltage during the initial black start of new energy is small. Therefore, it is necessary to focus on studying the frequency and voltage transient stability during the load connection process. The following is the stable control model during the initial black start and load connection when a virtual power plant composed of multiple distributed power sources is used as the black start power source.

[0113] When the new energy power station starts up by itself, it should make full use of the reactive power capacity and voltage regulation ability of the inverter. In order to make all distributed power sources (DGs) in an equal position, communicate independently, and can make decisions on operating states and power exchanges autonomously to achieve distributed consistency optimization, the peer-to-peer mode is used to control each distributed power source. In this mode, all DGs use droop control.

[0114] In a preferred implementation manner, it is defined that the operating constraint conditions of the virtual power plant include:

[0115] Obtain the change amounts of the frequency and voltage at the power output end of the distributed power source, and define the operating constraint conditions of the virtual power plant to determine the frequency and voltage transient stability during the load connection process;

[0116] The change amounts of the frequency and voltage at the power output end of the distributed power source are inversely proportional to the change amounts of the active power and reactive power, which is expressed as:

[0117]

[0118] where, m i and n i are the droop coefficients of the frequency and voltage characteristics respectively, and ω, V, P, and Q are the angular frequency, voltage, active power, and reactive power respectively;

[0119] The operating constraint conditions are expressed as:

[0120]

[0121] Among them, L represents the power exchange instruction value, that is, the power imbalance of the power grid. Due to the small transformer capacity and short transmission distance, the reactive power loss in this part is small, and the active power loss is ignored. Therefore, at this time, L is approximately equal to the power of the connected load, that is, P L and Q L are the active and reactive power of the connected load respectively; P i.max represents the rated active installed capacity of the distributed power source. Since there is no energy storage device, all distributed power sources cannot absorb active power, that is, P i.min = 0; Q i.max represents the maximum reactive power that the distributed power source can absorb, which is equal to the rated reactive power of the power source inverter, and Q i.min represents the maximum reactive power that the distributed power source can send out, which is equal to the negative value of the rated reactive power of the power source inverter; Δω i represents the carrier angular frequency of each distributed power source inverter. According to the black start requirement, the frequency fluctuation shall not exceed ±0.5 Hz. Therefore, Δω i.min = -π (rad / s), Δω i.max = π (rad / s); ΔV i represents the voltage fluctuation value of each power source port. Considering the transient fluctuation when connecting the load and the limitation of the duration at the same time, the threshold of the voltage fluctuation is strictly considered, and ΔV i.min = -0.2U N , ΔV i.max = 0.2U N .

[0122] In a preferred embodiment, the control objective is to minimize the fluctuation. Therefore, the absolute value needs to be taken here. That is, the optimization objective is expressed as:

[0123]

[0124] Applying the Lagrange multiplier method, the optimization problem is transformed. The initial black start control model of distributed new energy based on the virtual power plant is expressed as:

[0125]

[0126] Among them, λ represents the Lagrange multiplier corresponding to the equality constraint.

[0127] It should be noted that since the initial black start system is weak and lacks inertia, has weak anti-disturbance ability, and the control links are complex and diverse, therefore, the virtual power plant can use the hierarchical control method to more effectively control the initial black start of the distributed power source cluster as a whole. Therefore, it is mainly divided into three control layers.

[0128] In a preferred embodiment, the hierarchical control method includes:

[0129] The primary control layer controls active power - frequency and reactive power - voltage through the droop control strategy of distributed power sources, performing basic regulation of frequency and voltage.

[0130] The secondary control layer eliminates the deviation generated by the primary control layer and maintains the frequency and voltage within the rated value range.

[0131] The tertiary control layer is the dispatching layer, which controls the power output of each distributed power source and the power flow between the virtual power plant and the outside world.

[0132] It should be noted that since the distributed control algorithm can achieve the purpose of global control with only a small number of communication links and information exchanges, applying the consensus algorithm to hierarchical control can better meet the requirements of initial black - start control. When the system power is unbalanced (such as when connecting loads), the primary droop control of the inverter itself maintains the basic stability of the system frequency and voltage. On this basis, at the secondary control layer, based on the consensus algorithm, the transient frequency and voltage fluctuations are further stabilized, the error of the droop control is reduced, and the reasonable power distribution is completed, further realizing the power dispatching of the third layer.

[0133] Adopt Figure 3 The shown control structure to optimize the frequency and voltage control during the initial black - start process. Among them, the blue area is the primary control, which adopts the primary droop control to maintain the basic supply - demand balance and stability of the system; the red area is the secondary control, which uses the distributed control based on the consensus algorithm to correct the reference frequency and voltage values of the primary control.

[0134] In a preferred embodiment, the primary control layer includes:

[0135] Through the droop control strategy of distributed power sources, automatically allocate the power among multiple power sources. The voltage droop characteristic of the distributed power source is expressed as:

[0136]

[0137] where, ω i and V i represent the carrier angular frequency and the output - terminal voltage of the i - th distributed power source, V N represents the rated output voltage of the distributed power source, ω N represents the rated angular frequency, and the rated angular frequency corresponding to 50Hz is 100π (rad / s); m i and n i respectively represent the frequency and voltage droop control coefficients of the droop control, ΔP i and ΔQ i represent the changes in the active and reactive power outputs of the i - th distributed power source.

[0138] It should be noted that droop control is the basic control strategy for coordinating distributed power sources. Its principle is similar to the droop characteristics of traditional generator primary frequency regulation, aiming to achieve automatic power distribution among multiple power sources, and its working mode is simple and reliable.

[0139] In a preferred embodiment, the secondary control layer includes:

[0140] Distributed power secondary control includes voltage control and power control. Voltage control is used to restore the bus voltage to the rated value, and power control adjusts the power distribution among distributed power sources;

[0141] The control objective of voltage control is that the bus voltages of all nodes reach the rated value;

[0142] For power control, the droop control amount is corrected through a consensus algorithm. The voltage and frequency of the distributed power unit are taken, and the change amount based on the droop characteristic is used as the state variable of the agent node, and secondary power control is designed to make the state variables tend to be consistent.

[0143] In an alternative embodiment,

[0144] For voltage control, the control objective is that the bus voltages of all nodes reach the rated value, which is expressed as:

[0145]

[0146] where Δω ωi and ΔV Vi are the reference value correction amounts obtained from secondary frequency control and voltage control respectively, and K P and K I are the proportional coefficient and integral coefficient of the PI controller respectively.

[0147] For power control, the droop control amount is corrected based on the consensus algorithm. The change amounts of the voltage and frequency of the DG unit based on the droop characteristic are used as the state variables of the agent node, which is expressed as:

[0148]

[0149] To make the state variable x i tend to be consistent, the following secondary power control is designed, which is expressed as:

[0150]

[0151] where K I is the integral coefficient, Δω Pi and ΔV Qi are the frequency and voltage correction amounts obtained from the secondary power control link respectively;

[0152] After considering the secondary control, the inner-loop angular frequency and voltage command value of the system become the following equations:

[0153]

[0154] It should be noted that new energy units require small starting power and can start quickly, which plays an important supporting role in enhancing the power supply guarantee ability of the new power system to cope with extreme power outage scenarios. With the large-scale access of small-capacity distributed new energy units on the load side, the emergency recovery potential hidden in them is urgently to be explored. In this embodiment, virtual power plant technology is used to combine geographically dispersed distributed new energy units to participate in black start. First, with the minimum transient frequency and voltage deviation as the optimization goal, an initial black start control model for a distributed new energy unit cluster is established. Secondly, the multi-agent consensus algorithm and the hierarchical control method are used to realize the overall optimization of the initial black start voltage and frequency control and power scheduling of the distributed power source cluster. The simulation results show that: compared with the traditional centralized control, the proposed method not only has better control stability, but also has stronger robustness and reliability for operation scenarios such as new energy output fluctuations and random connection and disconnection of units.

[0155] The above is a schematic solution of an initial black start frequency and voltage distributed optimization control method in a virtual power plant mode of this embodiment. It should be noted that the technical solution of the initial black start frequency and voltage distributed optimization control system in this virtual power plant mode belongs to the same concept as the technical solution of the above initial black start frequency and voltage distributed optimization control method in the virtual power plant mode. For the details not described in detail in the technical solution of the initial black start frequency and voltage distributed optimization control system in this embodiment, reference can be made to the description of the technical solution of the initial black start frequency and voltage distributed optimization control method in the virtual power plant mode above.

[0156] The initial black start frequency and voltage distributed optimization control system in the virtual power plant mode of this embodiment includes:

[0157] The first construction module is used to establish the virtual power plant structure relationship, reduce the state difference between adjacent agent nodes of the distributed multi-agent system until the states of all nodes converge to the same value;

[0158] The second construction module is used to define the operation constraint conditions of the virtual power plant based on the virtual power plant structure relationship, and establish an initial black start control model for the distributed new energy of the virtual power plant with the minimum transient frequency and voltage deviation as the optimization goal;

[0159] The control module is used to perform the overall optimization of the initial black start voltage and frequency control and power scheduling of the distributed power source cluster based on the initial black start control model through the consensus algorithm and in combination with the hierarchical control method.

[0160] This embodiment also provides an electronic device, which is applicable to the situation of distributed optimization control of the initial black start frequency and voltage in the virtual power plant mode, and includes:

[0161] A memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for realizing distributed optimization control of the initial black start frequency and voltage in the virtual power plant mode as proposed in the above embodiment.

[0162] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for realizing distributed optimization control of the initial black start frequency and voltage in the virtual power plant mode as proposed in the above embodiment.

[0163] The storage medium proposed in this embodiment and the method for realizing distributed optimization control of the initial black start frequency and voltage in the virtual power plant mode proposed in the above embodiment belong to the same inventive concept. For technical details not described in detail in this embodiment, reference can be made to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0164] From the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk, or optical disc of a computer, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.

[0165] Embodiment 2

[0166] Refer to Figures 4 - 18 And Tables 1 - 2. As an embodiment of the present invention, a method for distributed optimization control of the initial black start frequency and voltage in the virtual power plant mode is provided, and scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0167] To verify the effectiveness of the strategy of the present invention for the frequency and voltage control of the initial black start of the distributed power generation cluster, an initial black start scenario of a 10-machine 1-load 11-node system is constructed as Figure 4As shown in the figure, the 10 distributed power sources include small-scale distributed photovoltaic power stations and wind farms in different regions. There are a total of five regions. Among them, G1, G2, and G3 in Region 1 are three distributed wind farms, and the remaining distributed power sources are distributed photovoltaics. Each distributed new energy power station and its inverter act as an agent, which can transmit information with its adjacent agents and cooperate to restore the auxiliary load of the thermal power plant under the control of the virtual power plant.

[0168] The rated output of each distributed power source and the droop coefficient of its inverter are shown in Table 1.

[0169] Table 1: Distributed Power Source Parameters

[0170] Power Supply Name Power Supply Power (kW + kVar) Droop Coefficient G1 1350+1283 12.7 G2 2140+917 10 G3 1080+577 6.3 G4 1270+1019 11.1 G5 2320+1283 14 G6 2400+1237 13.5 G7 3540+990 10.8 G8 2340+1054 11.5 G9 1740+843 9.2 G10 1820+797 8.7

[0171] The total rated output of the distributed power sources is 20MW + 10Mvar. The outlet voltage of each distributed power source is 10kV, and it is stepped up to 35kV by a substation 5km away from its respective area. The transmission line length from each substation to the thermal power plant is 30km, and the auxiliary capacity of the thermal power plant is 1MW + 0.62Mvar. The light intensity of the photovoltaic power station is set to 1000W / m2, and the wind speed during the operation of the wind turbine is 15m / s.

[0172] Next, a simulation model will be built for the system shown in Figure 4 based on the MATLAB / Simulink simulation platform, and the following three scenarios will be used to verify the stability, robustness, and reliability of the proposed control method.

[0173] (1) Scenario 1 - Stability Verification

[0174] To verify the control stability of the method proposed in this embodiment, a photovoltaic and energy storage power station (the ratio of photovoltaic to energy storage capacity is 10:1) with the same total capacity as the small-scale distributed power source cluster described above is set as a comparison scenario. The charging no-load line during the initial black start process and the transient frequency and voltage fluctuations when carrying a load are simulated and compared. The total simulation duration is set to 2s, and a load of 1MW + 0.62Mvar is connected at 1s. The simulation results are as Figures 5 - 8 shown.

[0175] By observing the following two groups of waveforms, the following two groups of simulation results can be obtained:

[0176] Table 2: Simulation Results of Scenario 1

[0177]

[0178] Comparing the simulation results, it can be seen that the distributed power source cluster under distributed optimal control meets the safety specifications in terms of various indicators during power self-starting and load carrying, and shows better control effects compared to a single photovoltaic and energy storage power station.

[0179] Observe the convergence of the consensus variables when the load is applied as Figure 9 and Figure 10 shown. The consensus variables can represent the voltage and frequency fluctuations in the secondary control link. Within 10 ms, the voltage consensus variable has basically converged to about 11.5 V. For the 10 kV outlet voltage level of small-scale distributed power sources, the voltage control effect is relatively good, which explains the good transient voltage control effect of the distributed power source cluster. At the same time, the frequency consensus variable has converged to about 0.077 Hz, which also plays a certain role in optimizing the frequency control.

[0180] The above results show that the stability of the small-scale distributed power source cluster using the strategy of this embodiment during the initial black start is guaranteed.

[0181] (2) Scenario 2 - Robustness verification

[0182] The simulation process is the same as that of the distributed power source cluster in Scenario 1. Set the wind speed in Region 1 to decrease to 11 m / s at 1.4 s, the light intensity in Region 2 to increase to 1200 W / m2, and the light intensity of the photovoltaic power stations in the remaining regions to decrease to 800 W / m2. Without the assistance of energy storage devices, as Figures 11 - 14 , simulate and observe the transient voltage and frequency fluctuations of the system at this time.

[0183] It can be seen from the simulation results that when the output of the distributed power sources fluctuates to a large extent, the peak value of the transient voltage fluctuation is close to 1.2 times, and the peak value of the frequency fluctuation is about 0.24 Hz. Then both quickly return to the steady state. It can also be observed that after the corresponding changes occur in each distributed power source, the voltage consensus variable quickly converges to about 13 V and the frequency consensus variable converges to about 0.087 Hz, without having a great impact on the system stability. Thus, it can be known that the system has a certain adaptability to the output fluctuations of the distributed power sources and has a certain robustness.

[0184] (3) Scenario 3 - Reliability verification

[0185] To verify the reliability of the proposed method, an adaptive analysis is carried out on the possible random connection and disconnection of the power sources in the system. Set to observe the frequency and voltage fluctuations of the system when some distributed power sources suddenly withdraw from operation.

[0186] The simulation process is the same as that of the distributed power source cluster in Scenario 1. Disconnect G2 and G5 at 0.5 s. At this time, the communication between G4 and G5, G2 and G1, and G3 is disconnected, and the matrix D changes from Equation (16) to Equation (17) (the changed matrix elements are marked in red):

[0187]

[0188]

[0189] The voltage and frequency conditions of the simulation observation system are as follows Figures 15 - 16 . By comparing and observing the simulation waveform this time with the initial black start waveform of the distributed power generation cluster in Scenario 1, it can be found that when some distributed power generations are disconnected, the system voltage only fluctuates slightly by about 0.05 pu, and at the same time, a frequency fluctuation of 0.09 Hz occurs. Then it quickly returns to the steady state without affecting the safe operation of the system.

[0190] As Figures 17 - 18 . By observing the convergence of the consensus variables of the remaining units after some distributed power generations are taken out of operation, it can be seen that after the D matrix changes, the consensus convergence is not affected, and the consensus variables of voltage and frequency both converge well to about 11.5 V and 0.077 Hz. This shows that the distributed optimization based on the consensus algorithm enables the system to have a certain ability to cope with power loss situations. Even if some units in the distributed power generation cluster experience off-grid faults and the communication network between the power sources changes, it will not have a great impact on the convergence of the consensus variables. It can still maintain a good control effect on voltage and frequency stability and will not seriously affect the initial black start safety level, thus enabling the system to have a certain reliability in coping with power loss situations.

[0191] In this embodiment, starting from distributed optimal scheduling, the potential of small-scale distributed power generation clusters in emergency recovery is explored. The distributed power generations in different regions are managed and controlled by means of a virtual power plant, and the technology of multi-agent consensus is used for distributed optimization. The case simulation analysis shows that using the above strategy can reduce the transient voltage and frequency fluctuation degrees when the distributed power generation cluster connects to the load compared with a single centralized large-scale distributed power generation, and the voltage recovery speed is also improved. Moreover, the optimal scheduling using the multi-agent consensus algorithm can also enable the system to have good adaptability to the random connection and disconnection of power sources and the output fluctuations of new energy. The initial black start system has a certain stability, robustness, and reliability.

[0192] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode, characterized in that: include: Establish a virtual power plant structure relationship to reduce the state difference between adjacent agent nodes in the distributed multi-agent system until the states of each node converge to a consistent value; Based on the virtual power plant structure relationship, the virtual power plant operation constraints are defined, and the initial black start control model of the distributed renewable energy of the virtual power plant is established with the minimum transient frequency and voltage deviation as the optimization goal; Based on the initial black start control model, the overall optimization of the initial black start voltage and frequency control and power scheduling of the distributed power supply cluster are performed through a consistency algorithm combined with a hierarchical control method.

2. The method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode according to claim 1, characterized in that: Establishing the virtual power plant structural relationship includes: The relationship between nodes in the multi-agent virtual power plant is established through the adjacency matrix, and the Laplace matrix is ​​introduced to represent the information feedback between agents and the impact of state differences on the control quantity; By calculating the control variables of the intelligent agents, the state change of the intelligent agents is obtained, wherein the control variables of the intelligent agents are calculated by the state difference between adjacent intelligent agents; Through the first-order consensus algorithm, each intelligent agent node is controlled to continuously adjust according to the adjacent nodes to reduce the state difference between any adjacent intelligent agent nodes, that is, the control variables of each node converge consistently.

3. The method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode according to claim 1 or 2, characterized in that: The operating constraints of the virtual power plant are defined as follows: Obtain the frequency and voltage changes at the power output of the distributed power source, and define the operating constraints of the virtual power plant to determine the transient stability of the frequency and voltage during the load connection process; The changes in frequency and voltage at the power output terminal of the distributed power source are inversely proportional to the changes in active power and reactive power, and are expressed as: Among them, m i and n i are the frequency and voltage characteristic droop coefficients, ω, V, P and Q are the angular frequency, voltage, active power and reactive power respectively; The operation constraints are expressed as: Where L represents the power exchange command value, P L and Q L Respectively represent the active and reactive power of the load, P i.max Indicates the rated active installed capacity of distributed generation, Q i.max Indicates the maximum reactive power that the distributed generation can absorb, Q i.min Indicates the maximum reactive power that the distributed generation can deliver, Δω i Indicates the carrier angular frequency of each distributed power inverter, ΔV i Indicates the voltage fluctuation value of each power port.

4. The method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode according to claim 3, characterized in that: The initial black start control model of distributed renewable energy based on virtual power plant is expressed as: where λ represents the Lagrange multiplier corresponding to the equality constraint.

5. The method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode according to claim 1, characterized in that: The hierarchical control method includes: A primary control layer, which controls active power-frequency and reactive power-voltage through a distributed power droop control strategy, and performs basic frequency and voltage regulation; A secondary control layer, which eliminates the deviation generated by the primary control layer and maintains the frequency and voltage within the rated value range; The third control layer is the dispatching layer, which controls the output of each distributed power source and the power flow between the virtual power plant and the outside world.

6. The method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode according to claim 5, characterized in that: The primary control layer includes: The distributed power droop control strategy is used to automatically distribute power among multiple power sources. The voltage droop characteristics of the distributed power sources are expressed as follows: Among them, ω i and V i represents the carrier angular frequency and output voltage of the ith distributed power source, V N Represents the rated output voltage of the distributed power supply, ω N Indicates the rated angular frequency, m i and n i They represent the frequency and voltage droop control coefficients of droop control, ΔP i and ΔQ i Represents the change in active and reactive power output of the i-th distributed generation.

7. The method for distributed optimization control of initial black start frequency and voltage in a virtual power plant mode according to claim 6, characterized in that: The secondary control layer comprises: The secondary control of the distributed power source includes voltage control and power control, wherein the voltage control is used to restore the bus voltage to the rated value, and the power control regulates the power distribution between the distributed power sources; The voltage control target is that the bus voltage of all nodes reaches the rated value; The power control modifies the droop control amount through a consistency algorithm, takes the voltage and frequency of the distributed power supply unit, takes the change amount based on the droop characteristic as the state variable of the intelligent node, and designs secondary power control to make the state variables tend to be consistent.

8. A distributed optimization control system for initial black start frequency and voltage in a virtual power plant mode, characterized in that: include: The first building module is used to establish a virtual power plant structure relationship and reduce the state difference between adjacent agent nodes of the distributed multi-agent system until the states of each node converge to a consistent value; The second building module is used to define the operating constraints of the virtual power plant based on the virtual power plant structure relationship, take the minimization of transient frequency and voltage deviation as the optimization goal, and establish the initial black start control model of the distributed renewable energy of the virtual power plant; The control module is used to perform overall optimization and power scheduling of the initial black start voltage and frequency control of the distributed power supply cluster based on the initial black start control model through a consistency algorithm combined with a hierarchical control method.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the initial black start frequency and voltage distributed optimization control method under the virtual power plant mode according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, implement the steps of the initial black start frequency and voltage distributed optimization control method in a virtual power plant mode as described in any one of claims 1 to 7.

Images