Parallel computing-based new energy voltage control method

Abstract

The invention provides a parallel computing-based new energy voltage control method, and belongs to the technical field of automatic voltage control of a power system. The method comprises the following steps: when a control cycle arrives, establishing a voltage-controlled quadratic programming model for each new energy gathering area; inputting the quadratic programming model of each new energy gathering area into a parallel computing platform and initiating a computing request, and solving by the parallel computing platform to obtain a reactive power regulation quantity of a new energy station equivalent machine in the new energy gathering area; and the parallel computing platform takes the reactive power regulation quantity as a control instruction and issues the control instruction to AVC substations of the new energy stations in the new energy gathering area, and the AVC substations complete reactive power regulation in the new energy stations. According to the method, on the basis of coordinating the secondary voltage control model, a parallel computing platform is combined to realize simultaneous solving of a plurality of new energy gathering areas, so that the real-time requirement of new energy voltage control is met.

Description

Technical Field

[0001] This invention belongs to the field of automatic voltage control technology for power systems, and specifically relates to a new energy voltage control method based on parallel computing. Background Technology

[0002] Automatic Voltage Control (AVC) systems are an important means of achieving safe (improving voltage stability margin), economical (reducing network losses), and high-quality (improving voltage qualification rate) operation of the power grid. Its basic principle is to achieve a reasonable distribution of reactive voltage in the power grid by coordinating and controlling the reactive power output of generators, transformer taps, and reactive power compensation equipment.

[0003] To date, there are three main modes of automatic voltage control worldwide:

[0004] The first approach, represented by the German power company RWE, involves two-level control without any regional control. The optimal power flow (OPF) calculation results are directly sent to the primary voltage controllers of each power plant for control. However, the OPF model involves a large amount of computation and takes a long time. When large disturbances, sudden increases or decreases in load occur in the system, if the system relies entirely on OPF, the response speed of the AVC (Automatic Voltage Controller) will be insufficient, and the dynamic quality of the control will be difficult to guarantee.

[0005] The second type is the three-level voltage control mode developed by EDF in France. Research and implementation of this mode began in the 1970s, and after more than 30 years of research, development, and application, it is currently recognized internationally as the most advanced voltage control system. While this control mode has been well-appointed, it still has drawbacks. This is because the regional two-level voltage control (SVC) is developed based on the locality of reactive voltage in the power system. However, reactive voltage is coupled between regions, so the quality of the control system fundamentally depends on the degree of coupling between reactive voltage controls in each region. However, with the development of the power system and the real-time changes in operating conditions, the regions that were initially considered relatively decoupled are not static. Furthermore, the control sensitivity, which exists in the form of fixed control parameters, changes in real-time with operating conditions. Therefore, this type of fixed-hardware regional controller is difficult to adapt to the continuous development of the power system and the significant changes in real-time operating conditions, making it difficult to consistently guarantee good control performance.

[0006] The third approach is a three-level voltage control mode based on "soft partitioning," proposed by the Dispatch Automation Laboratory of the Department of Electrical Engineering at Tsinghua University. This mode overcomes the shortcomings of hard partitioning in the EDF three-level voltage control system through soft partitioning. It has been widely applied in more than 20 regional and provincial power grids in China and has been successfully extended to the voltage control of the PJM power grid in North America. In this mode, the AVC application software in the dispatch center consists of a three-level control module and a two-level control module. The three-level control is the optimal power flow (OPF) for global reactive power optimization, providing a coordinated voltage optimization control objective for the entire network. The two-level control is the calculation of decoupled control strategies for each partition. Taking the optimized control objective of the central bus in each partition given by the three-level control as input, it considers reactive power regulation equipment such as power plants within the partition, calculates the control strategies for various reactive power resources within the partition, and distributes them to power plants and substations. The substation devices at the plant and substation ends complete the first-level control, receiving and executing the control strategies issued by the dispatch master station. Currently, automatic voltage control software developed using this mode is widely used in my country, installed and deployed in provincial and regional power grid dispatch control centers to automatically control voltage and reactive power at various voltage levels in the power grid.

[0007] Currently, the secondary control of the Automatic Voltage Control (AVC) system focuses on renewable energy aggregation zones, employing a coordinated secondary voltage control method to achieve hierarchical and zoned reactive power voltage control. In practical production applications, the calculation time for each renewable energy aggregation zone is approximately 0.1 seconds, resulting in a total calculation time of about 20 seconds due to the use of serial calculation. Considering the intermittent, variable, and random characteristics of wind and solar power generation, and the resulting voltage fluctuations, the renewable energy voltage control command cycle should be 1 minute. Therefore, the aforementioned serial calculation method can no longer meet the real-time requirements of renewable energy voltage control. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a new energy voltage control method based on parallel computing. This invention, based on the Coordinated Second-Level Voltage Control (CSVC) model and combined with a parallel computing platform, enables simultaneous solving of multiple new energy aggregation zones to meet the real-time requirements of new energy voltage control.

[0009] This invention proposes a new energy voltage control method based on parallel computing, comprising:

[0010] When the control cycle arrives, a quadratic programming model for voltage control is established for each new energy aggregation area.

[0011] The secondary planning model of each new energy gathering area is input into the parallel computing platform and a calculation request is initiated. The parallel computing platform solves for the reactive power regulation of the equivalent units of the new energy power stations in the new energy gathering area.

[0012] The parallel computing platform sends the reactive power adjustment amount as a control command to the AVC substations of each new energy power station in the new energy aggregation area, and the AVC substations complete the reactive power adjustment within the new energy power station.

[0013] In a specific embodiment of the present invention, the objective function expression of the quadratic programming model is as follows:

[0014]

[0015] Among them, V p This refers to the sampled value of the high-voltage bus voltage in the new energy aggregation area. The target voltage value for the high-voltage side busbar in the new energy aggregation area; As the objective, calculate the reactive power regulation ΔQ of the equivalent unit in the new energy power station. g ;Θ g To achieve the goal of balancing the reactive power output of various new energy power stations within the new energy aggregation area; W p W q These are the target weighting coefficients for voltage control and reactive power balance control, respectively. pg This is a matrix representing the reactive power regulation of equivalent generating units in new energy power plants within the new energy aggregation area to the reactive power voltage sensitivity matrix of the new energy aggregation area bus.

[0016] In one specific embodiment of the present invention, the constraints of the quadratic programming model include:

[0017] Voltage constraints on the high-voltage side bus of new energy power plants:

[0018]

[0019] Among them, V H , and These are the current voltage, upper voltage limit, and lower voltage limit of the high-voltage busbar at the new energy power station;

[0020] Voltage constraints on the busbars in the new energy aggregation area:

[0021]

[0022] in, and These are the upper and lower voltage limits for the busbar in the new energy aggregation area, respectively.

[0023] Reactive power constraints of power generation stations:

[0024]

[0025] Among them, Q g , These are the current reactive power, upper reactive power limit, and lower reactive power limit of the new energy equivalent generator, respectively.

[0026] The features and beneficial effects of this invention are as follows:

[0027] The new energy voltage control method proposed in this invention establishes a quadratic programming model (CSVC) for multiple new energy aggregation zones. Solving the CSVC is taken as a computational task, and a request is sent to the parallel computing platform. The parallel computing platform allocates computing resources to complete the solution, collects all the computational results, and then sends the instructions to each new energy power station for execution to meet the real-time requirements of new energy voltage control. Attached Figure Description

[0028] Figure 1 This is a flowchart of a new energy voltage control method based on parallel computing, according to an embodiment of the present invention.

[0029] Figure 2 This is a schematic diagram of a new energy collection area structure according to a specific embodiment of the present invention. Detailed Implementation

[0030] This invention proposes a new energy voltage control method based on parallel computing, which is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] This embodiment proposes a new energy voltage control method based on parallel computing, including:

[0032] When the control cycle arrives, a quadratic programming model for voltage control is established for each new energy aggregation area.

[0033] The secondary planning model of each new energy gathering area is input into the parallel computing platform and a calculation request is initiated. The parallel computing platform solves for the reactive power regulation of the equivalent units of the new energy power stations in the new energy gathering area.

[0034] The parallel computing platform sends the reactive power adjustment amount as a control command to the AVC substations of each new energy power station in the new energy aggregation area, and the AVC substations complete the reactive power adjustment within the new energy power station.

[0035] In one specific embodiment of the present invention, the overall process of the new energy voltage control method based on parallel computing is as follows: Figure 1 As shown, it includes the following steps:

[0036] 1) When the set control cycle arrives, establish a voltage control quadratic programming model (CSVC model) for each new energy collection area.

[0037] In this embodiment, the objective function expression of the quadratic programming model for any new energy agglomeration area is as follows:

[0038]

[0039] Among them, V p This is the sampled value of the high-voltage bus voltage in the new energy aggregation area. The target voltage value for the high-voltage busbar in the new energy aggregation area is generally provided by global reactive power optimization or set manually. As the objective, calculate the reactive power regulation ΔQ of the equivalent unit in the new energy power station. g ;Θ g To achieve the goal of balancing the reactive power output of various new energy power stations within the new energy aggregation area; W p W q These are the target weighting coefficients for voltage control and reactive power balance control, respectively. pg This is a matrix representing the reactive power regulation of equivalent generating units in new energy power plants within the new energy aggregation area to the reactive power voltage sensitivity matrix of the new energy aggregation area bus.

[0040] The calculation of the above parameters can be found in the paper "Research on Coordinated Second-Level Voltage Control" by Guo Qinglai, Sun Hongbin, Zhang Boming, et al. (Automation of Electric Power Systems, 2005, 29(23): 19-23), where the parameter Θ is calculated. g W p W q and sensitivity matrix C hg C pg The calculation methods are all introduced.

[0041] The constraints of the quadratic programming model include:

[0042] Voltage constraints on the high-voltage side bus of new energy power plants:

[0043]

[0044] Among them, V H , and These represent the current voltage, upper voltage limit, and lower voltage limit of the high-voltage busbar at the new energy power station. Equation (2) indicates the adjustment of ΔQ by the equivalent unit (e.g., wind power equivalent unit) at the new energy power station. g The voltage of the high-voltage busbar at the new energy power station should be within the upper and lower limits.

[0045] Voltage constraints on the busbars in the new energy aggregation area:

[0046]

[0047] in, and These are the upper and lower voltage limits for the busbar in the new energy aggregation area, respectively.

[0048] Equation (3) represents the new energy on-duty adjustment ΔQ g The bus voltage in the post-new energy aggregation area should be within the upper and lower limits.

[0049] Reactive power constraints of power generation stations:

[0050]

[0051] Among them, Q g , These are the current reactive power, upper reactive power limit, and lower reactive power limit of the new energy equivalent generator, respectively.

[0052] Equation (4) represents the adjustment ΔQ of the new energy equivalent machine. g The reactive power of the post-new energy equivalent generator should be within the upper and lower limits.

[0053] Furthermore, in this embodiment, the CSVC model of each new energy aggregation zone is output to a file, with the following file format:

[0054] Wp = W p

[0055] Vp = V p

[0056]

[0057] Cpg = [C pg1 C pg2 ,…,C pgn ]

[0058] Wq = W q

[0059]

[0060] Vh = [V H1 V H2 , ..., V Hn ]

[0061] Chg={[C h1g1 C h1g2 C h1gn ],[C h2g1 C h1g2 C h1gn ],…,[C hng1 C hng2 C hngn ]}

[0062] Q = [Q] g1 Q g2 Q gn ]

[0063]

[0064] In the file, the right side of the equals sign in each line represents the value of that parameter, where C hg Each of the equal-rate generators in the collection area is considered as a group for its reactive voltage sensitivity to one of the high-voltage side buses of a new energy power station. There are n groups in total, where n represents the number of high-voltage side buses of the new energy power station.

[0065] 2) Input the quadratic programming model of each new energy gathering area into the parallel computing platform and initiate a calculation request. The parallel computing platform will then complete the solution to obtain the reactive power regulation ΔQ of each equivalent generator. g .

[0066] In this embodiment, the parallel computing platform adopts the architecture proposed by Cheng Lun, Zhao Zigang, Gao Zeming, et al. in the literature "Research and Application of Online Parallel Computing Architecture for Anticipated Faults in Power Grids". The solution of the CSVC model adopts the quadratic programming problem solution method introduced by Guo Qinglai, Sun Hongbin, Zhang Boming, et al. in the literature "Research on Coordinated Secondary Voltage Control" (Automation of Electric Power Systems, 2005, 29(23): 19-23). ​​The quadratic programming model of each new energy gathering area is solved to obtain the reactive power regulation ΔQ of the equivalent generators (including equivalent generators and SVGs) in each new energy station. g .

[0067] 3) After the parallel computing platform completes the computation, it collects the results and expresses them as ΔQ. g Control commands are issued to the AVC substations of each new energy power station in the new energy aggregation area, and each AVC substation completes the reactive power regulation within the new energy power station.

[0068] The method described in this embodiment will be further explained in detail below with reference to specific embodiments:

[0069] In one specific embodiment of the present invention, there are two 220kV renewable energy collection areas, with the following structure: Figure 2 As shown. Among them, Figure 2 The left side is the collection area A, which includes: 1 220kV substation, 2 new energy power stations, and a total of 5 new energy equivalent units. Figure 2 The right side is collection area B, whose structure is the same as collection area A. The control cycle is set to 1 minute.

[0070] In this embodiment, the new energy voltage control method based on parallel computing includes the following steps:

[0071] 1) When the control cycle arrives, establish a voltage control quadratic programming model (CSVC model) for each new energy collection area.

[0072] In this embodiment, the objective function expression of the quadratic programming model for any new energy agglomeration area is as follows:

[0073]

[0074] In this embodiment, taking collection area A as an example, the collection area A is set as follows: Sampled value V of the high-voltage side bus voltage in collection area A p =231kV.

[0075] Among them, for the new energy gathering area A, W q =0.

[0076] The constraints of the quadratic programming model include:

[0077] Voltage constraints on the high-voltage side bus of new energy power plants:

[0078]

[0079] Voltage constraints on the busbars in the new energy aggregation area:

[0080]

[0081] Reactive power constraints of power generation stations:

[0082]

[0083] Furthermore, in this embodiment, the CSVC model of each new energy aggregation zone is output to a file.

[0084] Taking aggregation area A as an example, the file content is as follows:

[0085] Wp=1

[0086] Vp=231

[0087] VpSet=232

[0088] VpMax = 235

[0089] VpMin=225

[0090] Cpg=[0.066,0.081,0.074,0.054,0.067]

[0091] Wq=0

[0092] VhMax = [235, 235]

[0093] VhMin = [225, 225]

[0094] Vh = [228, 228]

[0095] Chg={[0.113,0.111,0.093,0.088,0.094],[0.073,0.061,0.063,0.078,0.054]}

[0096] Q=[0.12,0.45,-0.22,0.04,0.03]

[0097] QMax = [10, 10, 10, 10, 10]

[0098] QMin = [-10, -10, -10, -10, -10]

[0099] 2) Input the quadratic programming model of each new energy gathering area into the parallel computing platform and initiate a calculation request. The parallel computing platform will then complete the solution to obtain the reactive power regulation ΔQ of each equivalent generator. g .

[0100] 3) After the parallel computing platform completes the computation, it collects the results and expresses them as ΔQ. g Control commands are issued to the AVC substations of each new energy power station in the new energy aggregation area, and each AVC substation completes the reactive power regulation within the new energy power station.

[0101] The technical solutions provided by the embodiments of the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the embodiments of the present invention. The descriptions of the embodiments above are only for helping to understand the principles of the embodiments of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A new energy voltage control method based on parallel computing, characterized in that, include: When the control cycle arrives, a quadratic programming model for voltage control is established for each new energy aggregation area. The secondary planning model of each new energy gathering area is input into the parallel computing platform and a calculation request is initiated. The parallel computing platform solves for the reactive power regulation of the equivalent units of the new energy power stations in the new energy gathering area. The parallel computing platform sends the reactive power adjustment amount as a control command to the AVC substations of each new energy power station in the new energy aggregation area, and the AVC substations complete the reactive power adjustment within the new energy power station.

2. The method according to claim 1, characterized in that, The objective function expression of the quadratic programming model is as follows: Among them, V p This refers to the sampled value of the high-voltage bus voltage in the new energy aggregation area. The target voltage value for the high-voltage side busbar in the new energy aggregation area; As the objective, calculate the reactive power regulation ΔQ of the equivalent unit in the new energy power station. g ;Θ g To achieve the goal of balancing the reactive power output of various new energy power stations within the new energy aggregation area; W p W q These are the target weighting coefficients for voltage control and reactive power balance control, respectively. pg This is a matrix representing the reactive power regulation of equivalent generating units in new energy power plants within the new energy aggregation area to the reactive power voltage sensitivity matrix of the new energy aggregation area bus.

3. The method according to claim 2, characterized in that, The constraints of the quadratic programming model include: Voltage constraints on the high-voltage side bus of new energy power plants: Among them, V H , and These are the current voltage, upper voltage limit, and lower voltage limit of the high-voltage busbar at the new energy power station; Voltage constraints on the busbars in the new energy aggregation area: in, and These are the upper and lower voltage limits for the busbar in the new energy aggregation area, respectively. Reactive power constraints of power generation stations and other similar equipment: Among them, Q g , These are the current reactive power, upper reactive power limit, and lower reactive power limit of the new energy equivalent generator, respectively.

Images