Distributed offshore wind turbine control method based on consensus algorithm

Abstract

The application provides a distributed offshore wind turbine control method based on a consistency algorithm, comprising: taking the distributed offshore wind turbine as a distributed intelligent agent, and obtaining total active power shortage of a plurality of intelligent agents; establishing an initial first adjacency matrix according to a communication relationship between adjacent intelligent agents, and adjusting the first adjacency matrix to obtain a second adjacency matrix; modifying an initial first Laplacian matrix according to the second adjacency matrix to obtain a second Laplacian matrix; establishing a target function taking maximization of a wind energy utilization coefficient of each distributed offshore wind turbine as an objective, and obtaining an output power increment allocated to each distributed offshore wind turbine according to the second Laplacian matrix, so that the distributed offshore wind turbine outputs corresponding output power; and the application can improve power quality and improve power control accuracy of the distributed offshore wind turbine.

Description

Technical Field

[0001] This invention relates to the field of offshore wind power generation technology, and more specifically to a distributed offshore wind turbine control method based on a consensus algorithm. Background Technology

[0002] Amidst the escalating energy and environmental crises, the installed capacity of distributed offshore wind turbines, represented by wind power generation, is continuously increasing. Distributed offshore wind turbines can be viewed as virtual flywheel energy storage systems, absorbing and releasing energy by changing their rotational speed, thereby altering their output power, regulating the power output of the power system, and adjusting the system frequency.

[0003] For distributed offshore wind turbines in offshore wind farms, the distribution of power output from the centralized control system to each turbine is subject to significant challenges. If the centralized controller fails, the entire control system collapses immediately, resulting in low communication reliability. Furthermore, processing large amounts of collected data simultaneously by a single centralized controller places a heavy computational burden on the microprocessor, increasing computational and communication latency and reducing control efficiency. Additionally, the common use of 0-1 adjacency matrices to represent the topological relationships between agents leads to low power quality and low information density between adjacent agents. Therefore, existing technologies for centralized control of distributed offshore wind turbine output power suffer from low communication reliability, high computational load, low power quality, and reliance on a centralized controller. Moreover, the real-time power imbalance within the power system itself results in low system robustness, leading to inaccurate power control of the distributed offshore wind turbines. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the prior art by proposing a distributed offshore wind turbine control method based on a consensus algorithm, which can improve power quality and increase the accuracy of power control for distributed offshore wind turbines.

[0005] The distributed offshore wind turbine control method based on consensus algorithm provided by this invention includes:

[0006] Distributed offshore wind turbines are treated as distributed intelligent agents, and the total active power deficit of several intelligent agents is obtained; wherein, the total active power is obtained based on grid frequency fluctuations and voltage fluctuations;

[0007] Based on the communication relationship between neighboring agents, an initial first adjacency matrix is ​​established, and then the first adjacency matrix is ​​adjusted sequentially according to the adjustable output power increment and the maximum output power increment of each agent to obtain a second adjacency matrix.

[0008] The first Laplace matrix in the consensus algorithm is corrected based on the second adjacency matrix to obtain the second Laplace matrix;

[0009] An objective function is established to maximize the wind energy utilization coefficient of each distributed offshore wind turbine, and the output power increment allocated to each distributed offshore wind turbine is obtained according to the second Laplace matrix; wherein the sum of the obtained output power increments is not greater than the total active power deficit.

[0010] The output power of each distributed offshore wind turbine is increased to enable the distributed offshore wind turbine to output the corresponding output power.

[0011] This invention employs distributed offshore wind turbines as distributed agents, which, compared to existing technologies, reduces reliance on centralized control, thereby enhancing system robustness. Establishing an adjacency matrix for agents interacting with each other avoids a large number of distributed agents transmitting information to the centralized controller simultaneously, preventing multiple agents from simultaneously occupying bandwidth and improving control efficiency. Adjusting the adjacency matrix based on output power increments and maximum output power increments increases the information content compared to existing adjacency matrices represented by 0s and 1s, resulting in higher accuracy in calculating the output power of each offshore wind turbine and improving power quality. Correcting the Laplace matrix in the consensus algorithm using the adjusted adjacency matrix increases the information density of the Laplace matrix, further improving data transmission quality and the accuracy of offshore wind turbine output power calculation, thus facilitating more precise control of the offshore wind turbines.

[0012] Further, the step of correcting the initial first Laplacian matrix in the consensus algorithm based on the second adjacency matrix to obtain the second Laplacian matrix includes:

[0013] Based on the second adjacency matrix, the communication delay between neighboring agents, and the input delay of the agent itself, the initial first Laplace matrix in the consensus algorithm is corrected to obtain the second Laplace matrix.

[0014] This invention modifies the initial Laplace matrix based on the adjusted adjacency matrix, communication delay, and input delay. This not only increases the data quality density of the Laplace matrix but also takes into account the delay between adjacent agents. As a result, the accuracy of the output power of offshore wind turbines can be improved based on the modified Laplace matrix, thereby enhancing the system's power regulation capability.

[0015] Preferably, the second Laplace matrix can be expressed as:

[0016]

[0017] Among them, I ij (s) is the element in the i-th row and j-th column of the second Laplace matrix, a ij Let N be the element in the i-th row and j-th column of the second adjacency matrix. i τ is a set of intelligent agents. ij >0 indicates the communication delay from agent i to agent j, τ ii >0 indicates the input delay of agent i itself.

[0018] Furthermore, the step of adjusting the first adjacency matrix sequentially based on the adjustable output power increment and the maximum output power increment of each agent to obtain the second adjacency matrix includes:

[0019] The communication topology between each agent is obtained, and a first adjacency matrix is ​​established based on the communication topology. The ratio of the adjustable output power increment to the maximum output power increment of each agent is obtained in sequence. Based on the obtained ratios, the corresponding elements in the first adjacency matrix are adjusted to obtain a second adjacency matrix.

[0020] Furthermore, the establishment of the objective function for maximizing the wind energy utilization coefficient of each distributed offshore wind turbine also includes:

[0021] During the iterative calculation of the consensus algorithm, the ratio of the output power increment corresponding to each distributed offshore wind turbine to the maximum output power increment is used as the consensus variable.

[0022] Furthermore, obtaining the output power increment allocated to each of the distributed offshore wind turbines includes:

[0023] Calculate the first output power increment of each distributed offshore wind turbine under the objective function, and according to the second adjacency matrix, if the obtained first output power increments all satisfy the constraint conditions and the consistency variables of each distributed offshore wind turbine are equal, then the system converges, and the output power increment allocated to each distributed offshore wind turbine is obtained.

[0024] Furthermore, the condition that all the obtained first output power increments satisfy the constraint conditions also includes:

[0025] If all the obtained first output power increments satisfy the constraints, then the second output power increment of the previous iteration is updated according to the first output power increment and the second Laplace matrix;

[0026] If the obtained first output power increments do not meet the constraints, then the maximum output power increment of each distributed offshore wind turbine shall be used as the first output power increment of the corresponding distributed offshore wind turbine.

[0027] Furthermore, if the consistency variables of each distributed offshore wind turbine are equal, then the system converges, which also includes:

[0028] If the system has not yet converged, then the next iteration of the second output power increment is performed based on the first output power increment, the second adjacency matrix, and the second Laplace matrix, so as to obtain the next second output power increment for each of the distributed offshore wind turbines.

[0029] Furthermore, prior to treating the distributed offshore wind turbines as distributed intelligent agents, the method also includes:

[0030] The control process of distributed offshore wind turbines is divided into three control layers, so that the output power increment of each distributed offshore wind turbine can be obtained in the secondary control layer; wherein, the three control layers include: a scheduling layer, the secondary control layer and a primary control layer.

[0031] Furthermore, the step of increasing the output power of each distributed offshore wind turbine to enable the distributed offshore wind turbine to output a corresponding output power includes:

[0032] Based on the output power increment of each distributed offshore wind turbine, each distributed offshore wind turbine is controlled for a corresponding pitch angle to enable wide-speed operation while adopting corresponding output power tracking control; wherein, if the wind speed of the distributed offshore wind turbine does not meet the power generation conditions, the adjustment of the pitch angle of the distributed offshore wind turbine is stopped. Attached Figure Description

[0033] Figure 1 This is a flowchart illustrating the distributed offshore wind turbine control method based on a consensus algorithm provided in an embodiment of the present invention.

[0034] Figure 2 This is a schematic diagram of the three-layer control framework provided in an embodiment of the present invention;

[0035] Figure 3 This is a schematic diagram of the communication topology of a distributed offshore wind turbine provided in an embodiment of the present invention;

[0036] Figure 4 This is a schematic diagram of the process for allocating the active power deficit of each offshore wind turbine based on a consensus algorithm, provided in a real-time example of the present invention.

[0037] Figure 5 This is a schematic diagram illustrating the control of a distributed offshore generator provided by the present invention;

[0038] Figure 6 This is a schematic diagram of the system provided in this embodiment of the invention, which is controlled by three control layers. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] It is worth noting that the technical concept of this invention is as follows: Distributed offshore wind turbines are treated as distributed intelligent agents; a communication topology is constructed; both the adjacency matrix and the Laplace matrix are adjusted; and the output power of each offshore wind turbine is obtained based on a consensus algorithm. This invention features strong system robustness and does not rely on a centralized controller, thereby enabling the use of the kinetic energy stored within the distributed offshore wind turbines to regulate system frequency and improve power quality.

[0041] Specifically, this invention uses distributed offshore wind turbines as distributed agents, which, compared to existing technologies, reduces reliance on centralized control and enhances system robustness. Establishing an adjacency matrix for agents interacting with each other avoids a large number of distributed agents transmitting information to the centralized controller simultaneously, thus preventing multiple agents from simultaneously occupying bandwidth and improving control efficiency. Adjusting the adjacency matrix based on output power increments and maximum output power increments increases the information content compared to existing adjacency matrices represented by 0s and 1s, resulting in higher accuracy in calculating the output power of each offshore wind turbine and improving power quality. Correcting the Laplace matrix in the consensus algorithm using the adjusted adjacency matrix increases the information density of the Laplace matrix, further improving data transmission quality and the accuracy of offshore wind turbine output power calculation, thus facilitating more precise control of the offshore wind turbines.

[0042] See Figure 1 This is a flowchart illustrating the distributed offshore wind turbine control method based on a consensus algorithm provided in this embodiment of the invention, including steps S11 to S15, specifically:

[0043] Step S11: Treat the distributed offshore wind turbines as distributed intelligent agents and obtain the total active power deficit of several intelligent agents; wherein, the total active power is obtained based on grid frequency fluctuations and voltage fluctuations.

[0044] It is worth noting that grid frequency fluctuations and voltage fluctuations are grid frequency offsets and voltage offsets. Therefore, to address the real-time power imbalance and frequency offset issues in the power system itself, the following steps are taken: First, the total output power increase is calculated based on the system's frequency offset. Then, each offshore wind turbine is treated as a distributed intelligent agent, and a consensus algorithm is used to allocate the total power increase, the output power increase that each offshore wind turbine should output, and finally, the output power of the offshore wind turbine is controlled based on the corresponding output power increase.

[0045] Before treating the distributed offshore wind turbines as distributed intelligent agents, the method further includes: dividing the control process of the distributed offshore wind turbines into three control layers, so that the output power increment of each distributed offshore wind turbine can be obtained in the secondary control layer; wherein the three control layers include: a scheduling layer, the secondary control layer and the primary control layer.

[0046] It is worth noting that this invention constructs a three-layer control framework. The first layer is the dispatch layer, which determines the total reactive power deficit and the total active power deficit based on grid frequency and voltage fluctuations. It also clarifies the allocation of the total active power deficit and the total reactive power deficit between two intelligent agents: one composed of distributed offshore wind turbines and the other composed of energy storage devices in offshore wind farms. In other words, at the dispatch layer, the power deficit is determined by the grid dispatch center based on frequency fluctuations. The power deficit includes both the total active power deficit and the total reactive power deficit; that is, the total active power deficit and the total reactive power deficit can be obtained at the dispatch layer.

[0047] Each adjacent agent can communicate with each other. The total active power deficit and total reactive power deficit that can be allocated among each agent are calculated at the scheduling layer. It is worth noting that this invention focuses on the specific allocation of the total active power deficit by the agents generated by the distributed offshore wind turbine, so as to control the pitch angle according to the corresponding power increment allocated to each agent.

[0048] It is worth noting that when using a permanent magnet synchronous motor as an offshore wind turbine, the offshore wind turbine adopts a wide speed control strategy under high wind speed conditions.

[0049] It is worth noting that each offshore wind turbine can form a virtual flywheel energy storage system through speed regulation, thereby achieving charge and discharge control and regulating the frequency of the power system.

[0050] It is worth noting that distributed offshore wind turbines can achieve wide-range operation of permanent magnet synchronous motors while using maximum power point tracking control by controlling the pitch angle.

[0051] The second layer is the secondary control layer, which calculates the distribution of the active power increment among each offshore wind turbine through a consensus algorithm; the third layer is the primary control layer, also known as the response layer, which controls the output active power reference value of each offshore wind turbine.

[0052] See Figure 2 This is a schematic diagram of the three-layer control framework provided in this embodiment of the invention, including grid voltage control and frequency control executed at the scheduling layer. The scheduling layer obtains the total active power deficit and total reactive power deficit, and transmits these deficits to the secondary control layer for calculating the output power increase. The secondary control layer includes a distributed offshore wind turbine generator group and an energy storage transposition group; therefore, it receives not only the total active power but also the total reactive power deficit. Specifically, the intelligent agent corresponding to the offshore wind turbine uses a consensus algorithm to control the output power increase of the distributed offshore wind turbine, obtaining a reference value for the output power. The intelligent agent corresponding to the energy storage transposition allocates the total reactive power deficit, which is not elaborated here. Therefore, the secondary control layer transmits the output power increment as a reference value and the total reactive power deficit to the primary control layer. The primary control layer includes several distributed offshore wind turbines and several distributed energy storage devices. The primary control layer allocates the total active power deficit and the total reactive power deficit according to the received reference value. Based on the corresponding output power increment, the pitch angle of each distributed offshore wind turbine is controlled to enable the distributed offshore wind turbines to operate at a wide speed range while adopting corresponding power tracking control. In addition, based on the corresponding reactive power, each energy storage device allocates the total reactive power deficit to achieve dynamic adjustment of the overall system power.

[0053] Step S12: Based on the communication relationship between adjacent agents, establish an initial first adjacency matrix, and adjust the first adjacency matrix in sequence according to the adjustable output power increment and maximum output power increment of each agent to obtain a second adjacency matrix.

[0054] Specifically, the first adjacency matrix is ​​adjusted sequentially based on the adjustable output power increment and the maximum output power increment of each agent to obtain the second adjacency matrix. This includes: obtaining the communication topology between each agent, establishing the first adjacency matrix based on the communication topology, and sequentially obtaining the ratio of the adjustable output power increment to the maximum output power increment of each agent. Based on the obtained ratios, the corresponding elements in the first adjacency matrix are adjusted to obtain the second adjacency matrix.

[0055] It is worth noting that the present invention adjusts the traditional adjacency matrix. Specifically, a first adjacency matrix is ​​first established based on the communication relationship between adjacent agents. The first adjacency matrix is ​​then adjusted based on the ratio of the adjustable output power increment of each agent to the maximum output power increment to obtain a second adjacency matrix.

[0056] Specifically, the ratio of the adjustable output power increment of each agent to the maximum output power increment is obtained. The second adjacency matrix can be obtained by directly replacing the corresponding element in the first adjacency matrix with the ratio, multiplying or dividing the corresponding element in the first adjacency matrix with the ratio, or multiplying the ratio matrix with the first adjacency matrix.

[0057] Preferably, the ratio is used to directly replace the elements in the corresponding first adjacency matrix to obtain the second adjacency matrix.

[0058] Preferably, the elements of the second adjacency matrix can be represented as:

[0059]

[0060] Among them, P j For the adjustable output power derating of agent j, P jmax The maximum adjustable output power increment for agent j.

[0061] It is worth noting that in traditional consensus algorithms, if there is communication between agent i and agent j, then a ij =1, thus obtaining the adjacency matrix. Therefore, the first adjacency matrix can be obtained by the same operation. However, the information density of the first adjacency matrix is ​​extremely low. After adjustment by the present invention, the density of data carried by the adjacency matrix can be increased. Moreover, the larger the value, the greater the information density carried and the greater the adjustability of the system power.

[0062] Step S13: Correct the initial first Laplace matrix in the consensus algorithm according to the second adjacency matrix to obtain the second Laplace matrix.

[0063] Specifically, based on the second adjacency matrix, the communication delay between neighboring agents, and the input delay of the agent itself, the initial first Laplace matrix in the consensus algorithm is modified to obtain the second Laplace matrix.

[0064] This invention modifies the initial Laplace matrix based on the adjusted adjacency matrix, communication delay, and input delay. This not only increases the data quality density of the Laplace matrix but also takes into account the delay between adjacent agents. As a result, the accuracy of the output power of offshore wind turbines can be improved based on the modified Laplace matrix, thereby enhancing the system's power regulation capability.

[0065] Preferably, the elements of the second Laplace matrix can be represented as:

[0066]

[0067] Among them, I ij (s) is the element in the i-th row and j-th column of the second Laplace matrix, a ij Let N be the element in the i-th row and j-th column of the second adjacency matrix. i τ is a set of intelligent agents. ij >0 indicates the communication delay from agent i to agent j, τ ii >0 indicates the input delay of agent i itself, and s is the Laplace operator.

[0068] It is worth noting that, To account for the values ​​of the Laplace elements that take into account communication delay and input delay, the time-domain function of the consensus algorithm with different delays can be expressed as:

[0069]

[0070] in, Let be the time-domain function at time t.

[0071] Applying a Laplace transform to the time-domain function yields:

[0072]

[0073] Among them, X i (s), X j (s) are the transformation functions of agents i and j with respect to the Laplace operator s, respectively. i (0) is the time-domain function of agent i at time 0.

[0074] Based on the transformed time-domain function and the Laplace time delay property, when j∈N i When, the second Laplace matrix can be taken as

[0075] Preferably, the second Laplace matrix can be expressed as:

[0076] L(s)={I ij (s)}.

[0077] It is worth noting that the input delay and communication delay of the agent are taken into account, and the Laplace matrix in the traditional consensus algorithm is modified according to the second adjacency matrix. This makes the modified second Laplace matrix contain complex data of input delay, communication delay and communication topology, thereby improving the data density of the second Laplace matrix and making the accuracy of solving the final output power through the second Laplace matrix higher.

[0078] It is worth noting that this invention controls distributed offshore wind turbines through a constant-speed variable pitch adjustment method. That is, by changing the pitch angle, the blades installed on the hub are adjusted. By changing the relationship between the wind speed and the rated wind speed, the pitch angle is changed to change the starting torque acting on the wind turbine.

[0079] See Figure 3 This is a schematic diagram of the communication topology of distributed offshore wind turbines provided in an embodiment of the present invention. In the diagram, distributed offshore wind turbines communicate with adjacent distributed offshore wind turbines. When two distributed offshore wind turbines communicate with each other, a corresponding second adjacency matrix and a second Laplace matrix can be established.

[0080] Step S14: Establish an objective function to maximize the wind energy utilization coefficient of each distributed offshore wind turbine, and obtain the output power increment allocated to each distributed offshore wind turbine according to the second Laplace matrix; wherein the sum of the obtained output power increments is not greater than the total active power deficit.

[0081] The objective function is to maximize the wind energy utilization coefficient of each distributed offshore wind turbine. The method further includes: during the iterative calculation of the consensus algorithm, the ratio of the output power increment corresponding to each distributed offshore wind turbine to the maximum output power increment is used as the consensus variable.

[0082] Preferably, the consistency variable can be represented as:

[0083]

[0084] Where, ΔP out,i For the adjustable output power adequacy of distributed offshore wind turbine i, ΔP out,max,i This is an increase for the corresponding maximum output power.

[0085] It is worth noting that the objective function is subject to real-time power balance constraints, which limit the maximum and minimum values ​​of the output power increment of each distributed offshore wind turbine, i.e., limit the maximum and minimum values ​​of the output power of each distributed offshore wind turbine.

[0086] It is worth noting that after the control layer obtains the total active power deficit of all agents, each agent in the secondary control layer calculates the corresponding output power that should be adjusted for the corresponding distributed offshore generator. The power deficit allocated from the total active power deficit is obtained as the corresponding power increase to be adjusted. In the primary control layer, each distributed offshore generator generates the corresponding output power according to the power deficit.

[0087] The output power increment allocated to each distributed offshore wind turbine is obtained according to the second Laplace matrix, including: calculating the first output power increment of each distributed offshore wind turbine under the objective function, and according to the second adjacency matrix, if the obtained first output power increments all satisfy the constraint conditions and the consistency variables of each distributed offshore wind turbine are equal, then the system converges and the output power increment allocated to each distributed offshore wind turbine is obtained.

[0088] If the obtained first output power increments all satisfy the constraints, the method further includes: if the obtained first output power increments all satisfy the constraints, then the second output power increment of the previous iteration is updated according to the first output power increment and the second Laplace matrix; if the obtained first output power increments do not satisfy the constraints, then the maximum output power increment of each distributed offshore wind turbine is used as the first output power increment of the corresponding distributed offshore wind turbine.

[0089] If the consistency variables of each distributed offshore wind turbine are equal, the system converges. The method further includes: if the system has not yet converged, then based on the first output power increment, the second adjacency matrix, and the second Laplace matrix, performing the next iteration calculation of the second output power increment to obtain the next second output power increment of each distributed offshore wind turbine.

[0090] See Figure 4This is a flowchart illustrating the process of allocating the active power deficit of each offshore wind turbine based on a consensus algorithm, as provided in a real-time example of this invention. In the diagram, the allocated output power increment for each distributed offshore wind turbine is calculated based on the total active power deficit and the second adjacency matrix. This yields the allocated output power for each distributed offshore wind turbine. The process then sequentially checks whether the output power increment meets the constraints of the distributed offshore wind turbine; that is, the output power increment must be less than and equal to the maximum output power increment that the corresponding distributed offshore wind turbine can output. For agents that do not meet the corresponding constraints, the maximum output power increment that the corresponding distributed offshore wind turbine can output is used as the output power increment for that distributed offshore wind turbine. If the corresponding constraints are met, the calculated output power increment value is used to update the output power increment of that distributed offshore wind turbine. Finally, the consensus variable for each distributed offshore wind turbine is calculated based on its output power increment. If the consistency variables of each wind turbine are equal, the system is considered to have converged; if the consistency variables of each distributed offshore wind turbine are not equal, the new output power of the distributed offshore wind turbine is calculated based on the new output power of the distributed offshore wind turbine and the second adjacency matrix.

[0091] Step S15: Increase the output power of each distributed offshore wind turbine to enable the distributed offshore wind turbine to output the corresponding output power.

[0092] Specifically, based on the output power increment of each distributed offshore wind turbine, each distributed offshore wind turbine is controlled for a corresponding pitch angle, so as to enable the distributed offshore wind turbine to operate at a wide speed range while adopting the corresponding output power tracking control; wherein, if the wind speed of the distributed offshore wind turbine does not meet the power generation conditions, the adjustment of the pitch angle of the distributed offshore wind turbine is stopped.

[0093] It is worth noting that the output power obtained by the distributed offshore wind turbines is controlled at the primary control layer.

[0094] See Figure 5 This is a schematic diagram illustrating the control of a distributed offshore wind turbine provided by the present invention. In the diagram, the turbine speed can be adjusted while maintaining a constant wind energy utilization rate by controlling the pitch angle, where β is the pitch angle. Specifically, if the wind speed of the distributed offshore wind turbine is less than a wind speed threshold, the corresponding reference value ω for the turbine's rotational speed is... ref 1p.u; otherwise, the reference value for rotational speed ω. ref It is 0.9 pu, and based on the actual rotational speed ω r Obtain the reference value ω of the rotational speed. ref With actual rotational speed ω rDifferential rotational speed Δω r Then, the pitch angle is adjusted by controlling the pitch controller through a proportional-integral-derivative (PID) controller. By inputting the differential speed, the adjusted pitch angle reference value β is output. ref By adjusting the pitch angle resolution, the final pitch angle β is obtained. By changing the pitch angle, the blades mounted on the hub are adjusted, enabling the distributed offshore wind turbine to operate at a wide speed range while using corresponding power tracking control.

[0095] See Figure 6 This is a schematic diagram of the system provided in this embodiment of the invention, which is controlled through a three-layer control layer, including steps S61 to S64, specifically:

[0096] Step S61: Determine the output power increase through the scheduling layer and transmit the output power increase to the secondary control layer; which includes the total active power deficit and the total reactive power deficit.

[0097] Step S62: After receiving the output power increment, the secondary control layer allocates the output power increment through a consensus algorithm. That is, each distributed offshore generator obtains the corresponding output power and transmits the output power as a reference value to the primary control layer.

[0098] Step S63: The distributed offshore generator adjusts the pitch angle according to the corresponding output power.

[0099] Step S64: If the system has not yet ended, proceed to step S61; otherwise, end. Specifically, if the distributed offshore generator stops operating, i.e., the wind speed does not meet the power generation conditions, the system ends the power regulation; otherwise, the system continues to regulate.

[0100] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention utilizes distributed offshore wind turbines to control the pitch angle to achieve wide-speed operation, forming a virtual flywheel energy storage system, and realizing the power balance of the power system, thereby regulating the system frequency. This invention proposes adjusting the adjacency matrix and Laplace matrix of the communication topology, and under the constraint of increasing the output power of each wind turbine, it also considers the influence of communication delay and input delay, greatly increasing the data density of the adjacency matrix and Laplace matrix, thereby improving power quality and ultimately improving the power control accuracy of distributed offshore wind turbines.

[0101] Those skilled in the art will understand that embodiments of this application may also include computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0103] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0104] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0105] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A distributed offshore wind turbine control method based on consensus algorithm, characterized in that, include: Distributed offshore wind turbines are treated as distributed intelligent agents, and the total active power deficit of several intelligent agents is obtained; wherein, the total active power deficit is obtained based on grid frequency fluctuations and voltage fluctuations. Based on the communication relationship between neighboring agents, an initial first adjacency matrix is ​​established, and then the first adjacency matrix is ​​adjusted sequentially according to the adjustable output power increment and the maximum output power increment of each agent to obtain a second adjacency matrix. The first Laplace matrix in the consensus algorithm is corrected based on the second adjacency matrix to obtain the second Laplace matrix; An objective function is established to maximize the wind energy utilization coefficient of each distributed offshore wind turbine, and the output power increment allocated to each distributed offshore wind turbine is obtained according to the second Laplace matrix; wherein the sum of the obtained output power increments is not greater than the total active power deficit. The output power of each distributed offshore wind turbine is increased to enable the distributed offshore wind turbine to output the corresponding output power. The step of correcting the initial first Laplacian matrix in the consensus algorithm based on the second adjacency matrix to obtain the second Laplacian matrix includes: Based on the second adjacency matrix, the communication delay between neighboring agents, and the input delay of the agent itself, the initial first Laplace matrix in the consensus algorithm is corrected to obtain the second Laplace matrix; The step of adjusting the first adjacency matrix sequentially based on the adjustable output power increment and the maximum output power increment of each agent to obtain the second adjacency matrix includes: The communication topology between each agent is obtained, and a first adjacency matrix is ​​established based on the communication topology. The ratio of the adjustable output power increment to the maximum output power increment of each agent is obtained in sequence. The corresponding elements in the first adjacency matrix are adjusted according to the obtained ratios to obtain a second adjacency matrix. The process of obtaining the output power increment allocated to each distributed offshore wind turbine includes: Calculate the first output power increment of each distributed offshore wind turbine under the objective function. If the obtained first output power increments all satisfy the constraints and the consistency variables of each distributed offshore wind turbine are equal, then the system converges and the output power increment allocated to each distributed offshore wind turbine is obtained. The step of increasing the output power of each distributed offshore wind turbine to enable the distributed offshore wind turbine to output a corresponding output power includes: Based on the output power increment of each distributed offshore wind turbine, each distributed offshore wind turbine is controlled with a corresponding pitch angle to enable the distributed offshore wind turbine to operate at a wide speed range while adopting the corresponding output power tracking control; wherein, if the wind speed of the distributed offshore wind turbine does not meet the power generation conditions, the adjustment of the pitch angle of the distributed offshore wind turbine is stopped.

2. The distributed offshore wind turbine control method as described in claim 1, characterized in that, The second Laplace matrix is ​​expressed as: ， Where s is the Laplace operator, For the second Laplace matrix, the first... Line 1 Column elements, The second adjacency matrix is ​​the first... Line 1 Column elements, A collection of intelligent agents, Representing neighboring agents Send information to the intelligent agent Communication latency, Represents intelligent agents Its own input latency.

3. The distributed offshore wind turbine control method as described in claim 1, characterized in that, The establishment of the objective function, which aims to maximize the wind energy utilization coefficient of each distributed offshore wind turbine, also includes: During the iterative calculation of the consensus algorithm, the ratio of the output power increment corresponding to each distributed offshore wind turbine to the maximum output power increment is used as the consensus variable.

4. The distributed offshore wind turbine control method as described in claim 1, characterized in that, If the obtained first output power derating all satisfy the constraint conditions, it also includes: If all the obtained first output power increments satisfy the constraints, then the output power increments of the previous iteration are updated based on the first output power increments and the second Laplace matrix. If the obtained first output power increments do not meet the constraints, then the maximum output power increment of each distributed offshore wind turbine shall be used as the first output power increment of the corresponding distributed offshore wind turbine.

5. The distributed offshore wind turbine control method as described in claim 1, characterized in that, If the consistency variables of each distributed offshore wind turbine are equal, then the system converges, which also includes: If the system has not yet converged, the next iteration of the output power increment is performed based on the first output power increment, the second adjacency matrix, and the second Laplace matrix to obtain the next output power increment of each distributed offshore wind turbine.

6. The distributed offshore wind turbine control method as described in claim 1, characterized in that, Before considering the distributed offshore wind turbines as distributed intelligent agents, the following is also included: The control process of distributed offshore wind turbines is divided into three control layers, so that the output power increment of each distributed offshore wind turbine can be obtained in the secondary control layer; wherein, the three control layers include: a scheduling layer, the secondary control layer and a primary control layer.

Images