Power system regulation method, system and related equipment

Abstract

The present invention relates to the field of smart grid technology, and more specifically to a power system regulation method, system, and related equipment, comprising: clustering all power generation nodes in a power grid to obtain node clusters; obtaining the initial trigger angles of all power generation nodes in the power grid, generating new trigger angles for each power generation node, and iteratively optimizing these using a particle swarm optimization algorithm; calculating the node cluster regulation weights of each node cluster and the node cluster associations between the node clusters during the iteration process; using the combined result of the node cluster regulation weights and node cluster associations as the fitness of the particle swarm optimization algorithm; outputting an optimal trigger angle control vector when the iteration terminates; and using the optimal trigger angle control vector to control and regulate the thyristors within the static VAR compensator of each power generation node in the power grid. While ensuring node voltage balance, the present invention reduces interference between different power generation nodes and improves the flexible regulation performance of the power system.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a power system regulation method, system and related equipment. Background Art

[0002] Currently, a wide range of clean energy generation nodes have been connected to the power grid on a large scale, with the number growing year by year, forming a complex power network system. Furthermore, the increasing number of interactive facilities, such as distributed energy resources, has transformed the power grid system from a passive network to an active network, and the power system flow has also shifted from a one-way flow to a two-way interaction, placing higher demands on the power system's optimal configuration capabilities. By introducing a variety of power electronic conversion, compensation, and control technologies, smart grids can effectively achieve flexible regulation of the power system, supporting the diverse operating modes and frequent power flow shifts brought about by the large-scale centralized connection of clean energy generation nodes such as wind and solar power.

[0003] Flexible AC transmission system technologies are advanced transmission and distribution technologies commonly used in the power industry. They utilize power electronics to combine communications, power electronics, and intelligent control. These technologies can effectively address power system flow control, reactive power compensation, and power quality issues, enabling flexible regulation of the power system. Static VAR compensators (SVs) use internal variable transmission susceptance to adjust the system's equivalent impedance, thereby adjusting the system voltage and improving system stability. They are currently widely used flexible AC transmission controllers. Currently, when smart grids control the node voltage of clean energy generation nodes through flexible AC transmission systems, they only independently consider the voltage regulation of clean energy generation nodes by a single SVAR compensator, ignoring the interactions between multiple SVAR compensators in cross-regional power systems. This increases interference between different generation nodes, reduces the stability of the power system, and, in turn, reduces the flexible regulation performance of the power system. Summary of the Invention

[0004] The present invention provides a power system regulation method, system and related equipment to solve the existing problems.

[0005] The present invention provides a power system regulation method, system and related equipment using the following technical solutions:

[0006] An embodiment of the present invention provides a method for regulating a power system, the method comprising the following steps:

[0007] Obtain several node voltage sequences of each power generation node in the power grid and the distances between the power generation nodes;

[0008] Using the node difference value obtained by the distance between different power generation nodes, the overall difference of node voltage, and the correlation between node voltage series, all power generation nodes in the power grid are clustered to obtain several node clusters;

[0009] The initial trigger angles of all power generation nodes in the power grid are obtained, new trigger angles for each power generation node are randomly generated, and all new trigger angles are optimized and iterated using a particle swarm optimization algorithm. During the iteration process, the node cluster adjustment weight of each node cluster and the node cluster association between the node clusters are calculated. The node cluster adjustment weight is determined based on the difference between the node voltage and the rated voltage and the trigger angle change, and the node cluster association is determined based on the node difference value and the new trigger angle. The combined result of the node cluster adjustment weight and the node cluster association is used as the fitness in the particle swarm optimization algorithm. When the iteration terminates, the optimal trigger angle control vector is output.

[0010] The optimal firing angle control vector is used to control and adjust the thyristors in the static VAR compensator of each power generation node in the power grid.

[0011] Preferably, the specific method for obtaining the node cluster is:

[0012] The node difference value of any two power generation nodes is calculated based on the distance between them, the overall difference of node voltages, and the correlation between the voltage smoothing series corresponding to the two power generation nodes after fitting all node voltage series.

[0013] The node difference value between different power generation nodes in the power grid is used as the distance measurement method of the DBSCAN algorithm, and the DBSCAN algorithm is used to cluster all power generation nodes in the smart grid to obtain several clusters, and any cluster is recorded as a node cluster.

[0014] Preferably, the specific method for obtaining the voltage smoothing sequence is:

[0015] The node voltage in each node voltage sequence of all power generation nodes is normalized by the maximum and minimum value normalization method to obtain a voltage normalized sequence; the voltage normalized sequence is used as the input of the polynomial fitting algorithm, and the fitting curve of the voltage normalized sequence is output. The fitting values ​​of the voltage normalized sequence on the fitting curve are combined into a new sequence, which is recorded as a voltage smoothing sequence.

[0016] Preferably, the method of randomly generating a new trigger angle for each power generation node and iteratively optimizing all new trigger angles using a particle swarm optimization algorithm includes the following specific methods:

[0017] In the interval The random values ​​in the form a vector with the same length as the initial trigger angle control vector, which is recorded as the new trigger angle control vector of the power grid. The elements in the new trigger angle control vector are recorded as the new trigger angle, and the obtained The new trigger angle control vector is used as the initial population, and the particle swarm optimization algorithm is used to optimize the initial population iteratively. Each optimization iteration outputs a new trigger angle control vector under the corresponding iteration order, where 、 and They are the preset first parameter, second parameter and third parameter respectively.

[0018] Preferably, the specific method for obtaining the node cluster adjustment weight is:

[0019] During the optimization iteration process of the particle swarm optimization algorithm on the initial population, the numerical level difference between the node voltage sequence of all power generation nodes in any node cluster and the rated voltage of the power grid, as well as the angular difference between the initial trigger angle control vector and the new trigger angle control vector under the corresponding iterative order are analyzed to obtain the node cluster adjustment weight of the node cluster after any number of optimization iterations. The numerical level difference and the angular difference are both positively correlated with the node cluster adjustment weight.

[0020] Preferably, the specific method for obtaining the node cluster association degree is:

[0021] During the optimization iteration process of the particle swarm optimization algorithm on the initial population, the node cluster association between different node clusters after any number of optimization iterations is calculated based on the node difference value between the power generation nodes in different node clusters and the new trigger angle output by the optimization iteration. The node difference value is negatively correlated with the node cluster association degree, and the new trigger angle output by the optimization iteration is positively correlated with the node cluster association degree.

[0022] Preferably, the specific method for obtaining the fitness is:

[0023] During the optimization iteration process of the particle swarm optimization algorithm on the initial population, all node cluster adjustment weights are weighted by the node cluster association to obtain the fitness of the new trigger angle control vector after any number of optimization iterations. The node cluster adjustment weights and the node cluster association are both positively correlated with the fitness.

[0024] Preferably, the method of controlling and adjusting the thyristors in the static VAR compensator of each power generation node in the power grid by using the optimal trigger angle control vector includes the following specific methods:

[0025] Each element in the optimal firing angle control vector is obtained as the optimal firing angle of the corresponding power generation node. A static VAR compensator is installed at each power generation node. The thyristor in the static VAR compensator is controlled by the optimal firing angle of the power generation node, thereby changing the equivalent impedance of the thyristor-controlled reactor.

[0026] A power system regulation system adopts any one of the power system regulation methods described above, the system comprising the following modules:

[0027] Data acquisition module: used to obtain several node voltage sequences of each power generation node in the power grid and the distance between the power generation nodes;

[0028] Node clustering module: It is used to cluster all power generation nodes in the power grid into several node clusters by using the node difference values ​​obtained by the distance between different power generation nodes, the overall difference of node voltages, and the correlation between node voltage sequences;

[0029] Optimization and iteration module: used to obtain the initial trigger angles of all power generation nodes in the power grid, randomly generate new trigger angles for each power generation node, and use the particle swarm optimization algorithm to optimize and iterate all new trigger angles. During the iteration process, the node cluster adjustment weight of each node cluster and the node cluster association between the node clusters are calculated. The node cluster adjustment weight is determined based on the difference between the node voltage and the rated voltage and the trigger angle change. The node cluster association is determined based on the node difference value and the new trigger angle. The combined result of the node cluster adjustment weight and the node cluster association is used as the fitness in the particle swarm optimization algorithm. When the iteration terminates, the optimal trigger angle control vector is output.

[0030] Power regulation module: used to control and regulate the thyristors in the static VAR compensator of each power generation node in the power grid using the optimal trigger angle control vector.

[0031] A power system regulation device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of a power system regulation method when executing the computer program.

[0032] The beneficial effects of the technical solution of the present invention are: through node voltage sequence correlation analysis and distance clustering, a node cluster that can reflect the topological characteristics of the power grid is formed. Compared with the traditional global unified regulation strategy, the partition optimization based on the node cluster difference value can accurately identify the voltage weak link, so that the trigger angle adjustment of the power generation node is more targeted; at the same time, the fitness function obtained based on the particle swarm algorithm combined with the dual dimensions of node cluster regulation weight and correlation degree realizes the collaborative optimization of trigger angle control. The node cluster regulation weight avoids excessive regulation of a single node by quantifying the voltage deviation and the trigger angle change cost. The node cluster correlation degree ensures the coordination between adjacent clusters, reduces the interaction between multiple static VAR compensators in the cross-regional power system, and reduces the interference between different power generation nodes while ensuring the node voltage balance, thereby improving the flexible regulation performance of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 A flowchart of the steps of a power system regulation method of the present invention;

[0035] Figure 2 This is a structural block diagram of a power system regulation system of the present invention. DETAILED DESCRIPTION

[0036] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a power system regulation method, system, and related equipment proposed in accordance with the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0038] The following describes in detail a specific scheme of a power system regulation method, system and related equipment provided by the present invention with reference to the accompanying drawings.

[0039] See also Figure 1 , which shows a flowchart of a power system regulation method provided by an embodiment of the present invention, the method comprising the following steps:

[0040] Step S001: collecting node voltage sequences through voltage sensors and obtaining distances between different clean energy generation nodes.

[0041] It should be noted that there are a large number of clean energy power generation nodes in the power grid system. However, when using smart grid technology to control the node voltage of the clean energy power generation node through a flexible AC transmission system, only the voltage regulation of the clean energy power generation node by a single static VAR compensator is considered independently, and the interaction between multiple static VAR compensators in the cross-regional power system is ignored, resulting in increased interference between different power generation nodes, reduced stability of the power system, and further reduced flexible regulation performance of the power system. Therefore, in order to solve this technical problem, an embodiment of the present invention proposes a power system regulation method.

[0042] Specifically, in order to implement the power system regulation method proposed in this embodiment, it is first necessary to collect the node voltage sequence of all clean energy power generation nodes in the power grid system and the distances between the clean energy power generation nodes. The specific process is as follows:

[0043] First, the clean energy power generation nodes in the power grid system are referred to as power generation nodes. By installing voltage sensors at each power generation node, the node voltage data of each clean energy power generation node is collected in real time, and the distance between all different clean energy power generation nodes is obtained at the same time.

[0044] Then, the node voltage data collected by each clean energy power generation node is segmented to obtain the node voltage sequence measured by the power generation node in different time periods before, where the length of the node voltage sequence is ,in is the preset length parameter.

[0045] It should be noted that, in the embodiment of the present invention, the data acquisition frequency of the node voltage is preset to 100 Hz based on experience, and the length parameter The value is 50, which can be adjusted according to actual conditions in other embodiments and is not specifically limited in the embodiment of the present invention.

[0046] So far, the node voltage sequence of each power generation node and the distances between different power generation nodes are obtained through the above method.

[0047] Step S002: clustering all power generation nodes in the power grid using node difference values ​​obtained from the distances between different power generation nodes, the overall differences in node voltages, and the correlations between node voltage sequences to obtain a number of node clusters.

[0048] It's important to note that with the large-scale integration of clean energy into smart grids, multiple static VAR compensators (SVARs) are involved in controlling the flexible AC transmission system (FACTS) across a cross-regional power system comprised of clean energy generation nodes in diverse locations. This leads to strong interactions between clean energy generation nodes with strong electrical coupling and low short-circuit capacity, causing electromechanical oscillations and transient disturbances in the power system, degrading its flexible regulation performance.

[0049] It should be further explained that the change in the power generation output of clean energy is highly correlated with the region in which it is located. For example, similar changes in light intensity and wind intensity in the same region make the solar and wind power generation nodes in the region have similar power generation output characteristics. When the static VAR compensator adjusts the voltage of the power generation node, the nodes in the adjacent area are electrically coupled, and the control interaction between the two is relatively large; at the same time, the triggering angle changes of their static VAR compensators are similar, which affects the equivalent impedance of the power generation node, and thus affects the short-circuit capacity and flexible regulation performance of the power system in the region. In order to distinguish the strength of the interaction between power generation nodes in different regions and facilitate the subsequent optimization of the flexible regulation of the static VAR compensator in the smart grid, the embodiment of the present invention clusters the power generation nodes according to the regional change characteristics of the power generation output of the clean energy power generation nodes to obtain different node clusters.

[0050] Specifically, as a preferred embodiment, the specific method for obtaining the node cluster is:

[0051] First, the node difference value of any two power generation nodes is calculated based on the distance between them, the overall difference of node voltages, and the correlation between the voltage smoothing series corresponding to the two power generation nodes after fitting all node voltage series.

[0052] As an optional embodiment, the specific method for obtaining the voltage smoothing sequence is: using the maximum and minimum value normalization method to normalize the node voltages in each node voltage sequence of all power generation nodes to obtain a voltage normalized sequence; and using the voltage normalized sequence as the input of a polynomial fitting algorithm, outputting a fitting curve of the voltage normalized sequence, and forming a new sequence with the fitting values ​​of the voltage normalized sequence on the fitting curve, which is recorded as a voltage smoothing sequence.

[0053] It should be noted that, considering that the power generation nodes in the same area have similar output changes, but the degree of power generation output varies, for example, under the influence of clouds, the output changes of solar power generation nodes in the same area are similar, but the output degree of the power generation nodes in the center of the cloud cover is smaller, while the power generation nodes at the edge of the cloud cover are subject to stronger light intensity and have a larger output degree. At the same time, the same area may experience short-term and drastic changes in natural factors within a smaller location range, such as an instantaneous increase in wind force, which will cause the node voltages of different power generation nodes to fluctuate instantaneously. Therefore, the embodiment of the present invention eliminates the influence of the output degree of different power generation nodes on the subsequent judgment of the similarity of output changes by normalizing the node voltage sequence, and further obtains a voltage smoothing sequence by curve fitting the normalized sequence, effectively reducing the influence of the instantaneous fluctuation of node voltage on the similarity of output changes of power generation nodes.

[0054] It should also be noted that since the fluctuations in clean energy power generation output in a short period of time are random, and the overall temperature, sunlight and wind power in the adjacent area are relatively stable over a longer period of time, the voltage smoothing sequence of several node voltage sequences before the current measurement time of each clean energy power generation node is obtained according to the above method, so as to analyze the changes in power generation output of different power generation nodes over a longer period of time.

[0055] As an optional embodiment, the specific calculation method of the node difference value is:

[0056]

[0057] in, Indicates the and The distance between power generation nodes; Indicates the number of voltage smoothing sequences involved in calculating node difference values; and Respectively represent and The first power generation node The average value of the node voltage in the node voltage sequence; and Respectively represent and The first power generation node A voltage smoothing sequence; represents the cosine function; is the preset first constant value; Indicates the smart grid and Node difference value of power generation nodes; represents the absolute value function.

[0058] It should be noted that in the embodiment of the present invention, The value is 500, and the first constant value is preset to 2, in order to map the cosine similarity to a positive value. and The specific values ​​of can be adjusted according to actual conditions and are not specifically limited in the embodiments of the present invention.

[0059] It should be noted that the node difference value between generation nodes reflects the degree of output difference between the two generation nodes during the power generation process. On the one hand, the smaller the distance between the generation nodes, the closer they are to the impact of natural factors, and the smaller the difference in node power generation output. At the same time, the smaller the impedance of the transmission line between the two nodes, the greater the degree of electrical coupling between the two nodes, the closer the node voltage changes, and the smaller the calculated node difference value. On the other hand, the voltage smoothing series reflects the changing trend of the power generation output of different nodes. The greater the cosine similarity between the two nodes, the smaller the difference in power generation output. Moreover, in the same area, the smaller the difference in the node voltage average values, the closer the power generation output is and the smaller the node difference value.

[0060] Then, the node difference value between different power generation nodes in the power grid is used as the distance measurement method of the DBSCAN algorithm, and the DBSCAN algorithm is used to cluster all power generation nodes in the smart grid to obtain several clusters, and any cluster is recorded as a node cluster.

[0061] It should be noted that, in the embodiment of the present invention, the neighborhood radius in the DBSCAN algorithm is taken as the average value of the node difference values ​​of all different power generation nodes, and the minimum number of clustering points of the core object is preset to 5, which can be adjusted according to actual conditions.

[0062] It should be further explained that clustered outliers are treated as node clusters containing only one power generation node. By analyzing the regional variation characteristics of clean energy power generation output and calculating a voltage smoothing sequence, the influence of the output levels of different power generation nodes and instantaneous voltage fluctuations on the judgment of the similarity of subsequent output changes is eliminated. Node difference values ​​are further calculated, and the DBSCAN algorithm is used to cluster nodes with the same power generation output variation characteristics into node clusters. This facilitates the subsequent flexible regulation optimization process, while ensuring voltage balance among nodes within the node cluster, while improving the short-circuit capacity between node clusters with a high degree of electrical coupling, thereby improving the stability of the power system.

[0063] So far, several node clusters have been obtained through the above method.

[0064] Step S003: Obtain the initial trigger angles of all power generation nodes in the power grid, randomly generate new trigger angles for each power generation node, and use the particle swarm optimization algorithm to optimize and iterate all the new trigger angles. During the iteration process, the node cluster adjustment weight of each node cluster and the node cluster association between the node clusters are calculated. The node cluster adjustment weight is determined based on the difference between the node voltage and the rated voltage and the trigger angle change, and the node cluster association is determined based on the node difference value and the new trigger angle; the comprehensive result of the node cluster adjustment weight and the node cluster association is used as the fitness in the particle swarm optimization algorithm; when the iteration terminates, the optimal trigger angle control vector is output.

[0065] It should be noted that the initial trigger angle obtained by the flexible AC transmission system through PI control only independently considers a single power generation node in the power system. However, there are interactions between multiple static VAR compensators in a cross-regional power system. In severe cases, this may deteriorate the voltage control performance of the static VAR compensator and even cause power system instability. Therefore, the embodiment of the present invention takes into account the interactions between power generation nodes within a node cluster in the smart grid, as well as the interactions between different node clusters, and adopts a particle swarm optimization algorithm to optimize the trigger angles of different power generation nodes in the smart grid, thereby improving the stability of the power system and its flexible regulation performance.

[0066] Specifically, as a preferred embodiment, the method for obtaining the optimal firing angle control vector includes:

[0067] Firstly, the initial trigger angle of each power generation node in the power grid is obtained, and the vector formed by the initial trigger angles of all power generation nodes is recorded as the initial trigger angle control vector. Then, several new trigger angle control vectors are obtained by random generation, and all the new trigger angle control vectors are optimized iteratively using the particle swarm optimization algorithm.

[0068] As an optional embodiment, the method of obtaining the initial trigger angle of each power generation node in the power grid and recording the vector formed by the initial trigger angles of all power generation nodes as the initial trigger angle control vector includes:

[0069] The PI parameters of the static VAR compensator of the power generation node are adjusted by the Ziegler-Nichols method. A PI control algorithm is adopted, and the difference between the node voltage obtained by real-time measurement and the rated voltage is used as input to output the initial trigger angle of the power generation node.

[0070] It should be noted that the flexible AC transmission system adopts the PI control method to obtain the trigger angle of a single static VAR compensator thyristor and adjust its equivalent impedance to achieve the purpose of stabilizing the voltage of a single node. Therefore, the specific process of obtaining the initial trigger angle of the power generation node is well known to those skilled in the art and will not be repeated here.

[0071] As an optional embodiment, the method of obtaining a plurality of new firing angle control vectors by random generation and iteratively optimizing all the new firing angle control vectors by using a particle swarm optimization algorithm includes:

[0072] In the interval The random values ​​in the form a vector with the same length as the initial trigger angle control vector, which is recorded as the new trigger angle control vector of the power grid. The elements in the new trigger angle control vector are recorded as the new trigger angle, and so on. The new trigger angle control vector is used as the initial population, and the particle swarm optimization algorithm is used to optimize the initial population iteratively. Each optimization iteration outputs a new trigger angle control vector under the corresponding iteration order, where 、 and They are the preset first parameter, second parameter and third parameter respectively.

[0073] It should be noted that the trigger angle is usually within the range Therefore, in the embodiment of the present invention, the preset interval for , that is, preset the first parameter for , the second parameter for In addition, in the embodiment of the present invention, the third parameter is preset based on experience is 30, the first parameter , the second parameter and the third parameter It can be adjusted according to actual conditions and is not specifically limited in the embodiment of the present invention.

[0074] It should be further explained that the purpose of the same length between the initial trigger angle control vector and the new trigger angle control vector in the embodiment of the present invention is that, since the elements in the two vectors represent the trigger angle, the embodiment of the invention ensures that the trigger angle and the power generation node always maintain a corresponding relationship by making the number of elements in the vector the same, so that each element representing the optimal trigger angle in the optimal trigger angle control vector obtained through continuous optimization and iteration will still have a corresponding relationship with each power generation node, thereby achieving effective control of each power generation node in the power grid.

[0075] Then, during the optimization iteration process of the particle swarm optimization algorithm on the initial population, the numerical level differences between the node voltage sequences of all power generation nodes in any node cluster and the rated voltage of the power grid, as well as the angular differences between the initial trigger angle control vector and the new trigger angle control vector under the corresponding iterative order are analyzed to obtain the node cluster adjustment weight of the node cluster after any number of optimization iterations. Both the numerical level differences and the angular differences are positively correlated with the node cluster adjustment weight.

[0076] It should be noted that when optimizing and iterating the trigger angles of power generation nodes within a node cluster, the degree to which clean energy power generation nodes in different node clusters are affected by natural factors varies, and the difference between their node voltages and the rated voltage is different. Therefore, in order to ensure that the node voltages of different power generation nodes remain stable, the embodiment of the present invention obtains the node cluster adjustment weights of the node cluster to adjust the optimization iterative process of the trigger angle, thereby ensuring the voltage stability of different power generation nodes.

[0077] As an optional embodiment, a specific method for calculating the node cluster adjustment weight of the node cluster is:

[0078]

[0079] in, Indicates the The number of power generation nodes in a node cluster; Indicates the The average value of the node voltage in the node voltage sequence currently measured by the power generation node; Indicates the rated voltage of the power grid; and The distribution represents the The initial trigger angle of the power generation node and the The firing angle output after the optimization iteration; represents a logarithmic function with a natural constant as its base; Indicates the The node cluster is in Node cluster adjustment weights during the optimization iteration; Indicates the absolute value symbol.

[0080] It should be noted that the node cluster adjustment weight reflects the degree of impact of the static VAR compensator (SVC) firing angle adjustment within the corresponding node cluster on the power system. In the specific calculation method, a logarithmic function is used to control the size of the node cluster adjustment weight. The smaller the difference between the average node voltage of a clean energy generation node and the rated voltage, the smaller the weight of the node cluster in the subsequent fitness calculation, and the smaller the calculated node cluster adjustment weight. This allows the trigger angle to be adjusted over a wider range, ensuring voltage stability at the corresponding generation node while improving the short-circuit capacity of the power system in the node cluster region. To balance the node voltage of a single generation node, the greater the difference between the firing angle output by the optimization iteration and the initial firing angle, the smaller the adjustment range, the larger the calculated node cluster adjustment weight, and the greater the fitness obtained in the subsequent corresponding firing angle calculation, thus ensuring node voltage stability. Furthermore, the larger the firing angle of the static VAR compensator thyristor, the greater its equivalent impedance, the smaller the short-circuit capacity of the power system, and the greater the degree of interaction between different generation nodes. Therefore, when the absolute value of the difference between the trigger angle and the initial trigger angle is equal, the trigger angle greater than the initial trigger angle has a greater impact on the stability of the power system. A logarithmic function is used to represent this impact feature, so that the weight set at this time is larger to reduce the adjustment range of the subsequent corresponding trigger angle.

[0081] Secondly, during the optimization iteration process of the particle swarm optimization algorithm on the initial population, the node cluster association between different node clusters after any number of optimization iterations is calculated based on the node difference value between the power generation nodes in different node clusters and the new trigger angle output by the optimization iteration. The node difference value is negatively correlated with the node cluster association degree, and the new trigger angle output by the optimization iteration is positively correlated with the node cluster association degree.

[0082] It should be noted that the power generation nodes within different node clusters are located in different regions. Due to the influence of transmission line impedance between different regions and natural factors on power generation output, the degree of electrical coupling of cross-regional power systems varies. At the same time, the trigger angle of the thyristor of the static VAR compensator of the power generation node is negatively correlated with the short-circuit capacity of the power system. In addition, the interaction between different power generation nodes in a strongly coupled cross-regional power system is weaker at higher short-circuit capacities. Therefore, the embodiments of the present invention analyze the degree of electrical coupling between power generation nodes in different node clusters and adjust their trigger angles to reduce the interaction in the flexible AC transmission system.

[0083] As an optional embodiment, the specific calculation method of the node cluster association degree is:

[0084]

[0085] in, and Respectively represent and The number of power generation nodes in a node cluster; and Respectively represent The output after the optimization iteration The node in the cluster The new trigger angle of the power generation node and the The node in the cluster New trigger angles for each power generation node; Indicates the The node in the cluster The power generation node and The node in the cluster Node difference value of power generation nodes; is the preset second constant value; Indicates the node clusters and the The node cluster is in The node cluster correlation degree under the trigger angle control of the sub-optimization iteration output.

[0086] It should be noted that the node cluster correlation reflects the short-circuit capacity of the power system in the area where the two node clusters are located and their electrical coupling. and When they are equal, it indicates the degree of interaction between the electrical coupling between different power generation nodes in the same node cluster on the regional power system where the node cluster is located. and When equal, The value is 0.

[0087] It should be noted that the second constant value is preset based on experience The value of is 1, which is used to prevent the denominator from being zero and to control the size of the node cluster association.

[0088] It should be further explained that the node difference value reflects the degree of difference in the output of two clean energy generation nodes. The smaller the node difference value between the generation nodes, the stronger the electrical coupling and the greater the correlation between the two node clusters. The greater the calculated node cluster correlation, the greater the degree of adjustment required for the generation node trigger angle in the corresponding cross-regional power system to adjust its short-circuit capacity and reduce the interaction between its internal nodes. Therefore, the greater the fitness obtained in the subsequent calculation. In addition, the larger the trigger angle of the static VAR compensator of the generation node in different node clusters, the smaller the short-circuit capacity of the cross-regional power system and the stronger the interaction effect. The greater the calculated node cluster correlation, the greater the fitness of the trigger angle in the subsequent calculation, and the worse the performance of the corresponding trigger angle in the flexible regulation of the power system.

[0089] Furthermore, during the optimization iteration process of the particle swarm optimization algorithm on the initial population, all node cluster adjustment weights are weighted by the node cluster association to obtain the fitness of the new trigger angle control vector after any number of optimization iterations. The node cluster adjustment weights and the node cluster association are both positively correlated with the fitness.

[0090] As an optional embodiment, a specific method for calculating the fitness of the trigger angle control vector is:

[0091]

[0092] in, represents the number of node clusters in the smart grid; and Respectively represent node clusters and the The node cluster is in Node cluster adjustment weights during the optimization iteration; Indicates the node clusters and the The node cluster is in The node cluster correlation degree under the trigger angle control output by the sub-optimization iteration; Indicates the firing angle control vector in The fitness of the optimization iteration.

[0093] It should be noted that the fitness of the firing angle control vector during the iterative optimization process reflects the degree of its performance in the flexible regulation of the smart grid's interregional power system. On the one hand, the node cluster regulation weight reflects the degree of influence of the static VAR compensator firing angle adjustment within the node cluster on the power system. A larger node cluster regulation weight indicates a greater impact of the current firing angle of the generation nodes within the node cluster on the power system. A higher calculated fitness indicates poorer flexible regulation performance. On the other hand, the node cluster association reflects the short-circuit capacity and electrical coupling of the power systems in the regions where the two node clusters are located. A higher node cluster association indicates a greater electrical coupling of the interregional power systems in which the two node clusters are located, a lower short-circuit capacity, and greater interaction between different generation nodes in the power system, resulting in decreased power system stability. Consequently, the calculated fitness of the firing angle control vector increases.

[0094] Finally, the fitness is used as the fitness in the particle swarm optimization algorithm, and after the number of iterations of the particle swarm optimization algorithm reaches the maximum number of iterations, the trigger angle control vector obtained in the last iteration is used as the optimal trigger angle control vector, where the maximum number of iterations of the particle swarm optimization algorithm is ,in It is the preset fourth parameter.

[0095] It should be noted that in the embodiment of the present invention, the fourth parameter is preset based on experience. is 30. It should be emphasized that the third parameter and the fourth parameter The values ​​of are not required to be equal; the fourth parameter The value of can be adjusted according to actual conditions and is not specifically limited in the embodiment of the present invention.

[0096] At this point, the optimal trigger angle control vector of the power grid is obtained.

[0097] Step S004: using the optimal trigger angle control vector to control and adjust the thyristors in the static VAR compensator of each power generation node in the power grid.

[0098] Specifically, as a specific embodiment, the method of controlling and adjusting the thyristors in the static VAR compensator of each power generation node in the power grid by using the optimal trigger angle control vector includes:

[0099] Each element in the optimal firing angle control vector is obtained as the optimal firing angle of the corresponding power generation node. A static VAR compensator is installed at each power generation node. The thyristor in the static VAR compensator is controlled by the optimal firing angle of the power generation node, changing the equivalent impedance of the thyristor-controlled reactor to achieve power compensation for the power grid, balance the node voltage at the power generation node, and thus realize the regulation of the power grid power system.

[0100] It should be noted that, in the embodiment of the present invention, by utilizing the regional characteristics of the power generation output of the clean energy power generation nodes, all power generation nodes in the cross-regional power system are divided into different node clusters. By analyzing the short-circuit capacity of each power generation node in the node cluster and the degree of coupling between different node clusters, the particle swarm algorithm is used to optimize the triggering angle of the static VAR compensator of each clean energy power generation node, thereby reducing the interaction between multiple static VAR compensators in the cross-regional power system. While ensuring the node voltage balance, the interference between different power generation nodes is reduced, thereby improving the flexible regulation performance of the power system.

[0101] Through the above steps, the flexible regulation performance of the power system is completed.

[0102] See also Figure 2 , which shows a structural block diagram of a power system regulation system provided by an embodiment of the present invention, the system includes the following modules:

[0103] Data acquisition module: used to obtain several node voltage sequences of each power generation node in the power grid and the distance between the power generation nodes;

[0104] Node clustering module: It is used to cluster all power generation nodes in the power grid into several node clusters by using the node difference values ​​obtained by the distance between different power generation nodes, the overall difference of node voltages, and the correlation between node voltage sequences;

[0105] Optimization and iteration module: used to obtain the initial trigger angles of all power generation nodes in the power grid, randomly generate new trigger angles for each power generation node, and use the particle swarm optimization algorithm to optimize and iterate all new trigger angles. During the iteration process, the node cluster adjustment weight of each node cluster and the node cluster association between the node clusters are calculated. The node cluster adjustment weight is determined based on the difference between the node voltage and the rated voltage and the trigger angle change. The node cluster association is determined based on the node difference value and the new trigger angle. The combined result of the node cluster adjustment weight and the node cluster association is used as the fitness in the particle swarm optimization algorithm. When the iteration terminates, the optimal trigger angle control vector is output.

[0106] Power regulation module: used to control and regulate the thyristors in the static VAR compensator of each power generation node in the power grid using the optimal trigger angle control vector.

[0107] In another embodiment of the present invention, a power system regulation device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, steps S001 to S004 in the power system regulation method are implemented.

[0108] Furthermore, in an optional embodiment, the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. The memory may also include a non-volatile random access memory. For example, the memory may also store device type information.

[0109] The memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory, wherein the non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available. For example, static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct RAM bus random access memory (DRRAM).

[0110] In an optional embodiment, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that the processor may be a processor that supports the Advanced RISC Machines (ARM) architecture.

[0111] The present invention may implement all or part of the processes in the above-described method embodiments by means of a computer program that instructs related hardware. The computer program may be stored in a computer-readable storage medium, and when executed by a processor, the computer program may implement the steps of each of the above-described method embodiments. The computer program includes computer program code, which may be in source code form, object code form, an executable file, or some intermediate form.

[0112] The embodiment of the present invention forms a node cluster that can reflect the topological characteristics of the power grid through node voltage sequence correlation analysis and distance clustering. Compared with the traditional global unified regulation strategy, the partition optimization based on the node cluster difference value can accurately identify the voltage weak link, so that the trigger angle adjustment of the power generation node is more targeted; at the same time, the fitness function obtained based on the particle swarm algorithm combined with the dual dimensions of node cluster regulation weight and correlation degree realizes the collaborative optimization of trigger angle control. The node cluster regulation weight avoids excessive regulation of a single node by quantifying the voltage deviation and the trigger angle change cost. The node cluster correlation degree ensures the coordination between adjacent clusters, reduces the interaction between multiple static VAR compensators in the cross-regional power system, and reduces the interference between different power generation nodes while ensuring the node voltage balance, thereby improving the flexible regulation performance of the power system.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A power system regulation method, characterized in that: The method comprises the following steps: Obtain several node voltage sequences of each power generation node in the power grid and the distances between the power generation nodes; Using the node difference value obtained by the distance between different power generation nodes, the overall difference of node voltage, and the correlation between node voltage series, all power generation nodes in the power grid are clustered to obtain several node clusters; The initial trigger angles of all power generation nodes in the power grid are obtained, new trigger angles for each power generation node are randomly generated, and all new trigger angles are optimized and iterated using a particle swarm optimization algorithm. During the iteration process, the node cluster adjustment weight of each node cluster and the node cluster association between the node clusters are calculated. The node cluster adjustment weight is determined based on the difference between the node voltage and the rated voltage and the trigger angle change, and the node cluster association is determined based on the node difference value and the new trigger angle. The combined result of the node cluster adjustment weight and the node cluster association is used as the fitness in the particle swarm optimization algorithm. When the iteration terminates, the optimal trigger angle control vector is output. The optimal firing angle control vector is used to control and adjust the thyristors in the static VAR compensator of each power generation node in the power grid; The specific method for obtaining the node cluster adjustment weight is: During the optimization iteration process of the particle swarm optimization algorithm on the initial population, the numerical level difference between the node voltage sequence of all power generation nodes in any node cluster and the rated voltage of the power grid, as well as the angular difference between the initial trigger angle control vector and the new trigger angle control vector under the corresponding iterative order are analyzed to obtain the node cluster adjustment weight of the node cluster after any number of optimization iterations. The numerical level difference and the angular difference are both positively correlated with the node cluster adjustment weight; The specific method for obtaining the node cluster association degree is: During the optimization iteration process of the particle swarm optimization algorithm on the initial population, the node cluster association between different node clusters after any number of optimization iterations is calculated based on the node difference value between the power generation nodes in different node clusters and the new trigger angle output by the optimization iteration. The node difference value is negatively correlated with the node cluster association degree, and the new trigger angle output by the optimization iteration is positively correlated with the node cluster association degree. The specific method for obtaining the fitness is: During the optimization iteration process of the particle swarm optimization algorithm on the initial population, all node cluster adjustment weights are weighted by the node cluster association to obtain the fitness of the new trigger angle control vector after any number of optimization iterations. The node cluster adjustment weights and the node cluster association are both positively correlated with the fitness.

2. A power system regulation method according to claim 1, characterized in that: The specific method for obtaining the node cluster is: The node difference value of any two power generation nodes is calculated based on the distance between them, the overall difference of node voltages, and the correlation between the voltage smoothing series corresponding to the two power generation nodes after fitting all node voltage series. The node difference value between different power generation nodes in the power grid is used as the distance measurement method of the DBSCAN algorithm, and the DBSCAN algorithm is used to cluster all power generation nodes in the smart grid to obtain several clusters, and any cluster is recorded as a node cluster.

3. A power system regulation method according to claim 2, characterized in that: The specific method for obtaining the voltage smoothing sequence is: The node voltage in each node voltage sequence of all power generation nodes is normalized by the maximum and minimum value normalization method to obtain a voltage normalized sequence; the voltage normalized sequence is used as the input of the polynomial fitting algorithm, and the fitting curve of the voltage normalized sequence is output. The fitting values ​​of the voltage normalized sequence on the fitting curve are combined into a new sequence, which is recorded as a voltage smoothing sequence.

4. The power system regulation method according to claim 1, characterized in that: The method of randomly generating a new trigger angle for each power generation node and iteratively optimizing all new trigger angles using a particle swarm optimization algorithm includes the following specific methods: In the interval The random values ​​in the form a vector with the same length as the initial trigger angle control vector, which is recorded as the new trigger angle control vector of the power grid. The elements in the new trigger angle control vector are recorded as the new trigger angle, and the obtained The new trigger angle control vector is used as the initial population, and the particle swarm optimization algorithm is used to optimize the initial population iteratively. Each optimization iteration outputs a new trigger angle control vector under the corresponding iteration order, where 、 and They are the preset first parameter, second parameter and third parameter respectively.

5. The power system regulation method according to claim 1, characterized in that: The specific method of controlling and adjusting the thyristors in the static VAR compensator of each power generation node in the power grid by using the optimal trigger angle control vector is as follows: Each element in the optimal firing angle control vector is obtained as the optimal firing angle of the corresponding power generation node. A static VAR compensator is installed at each power generation node. The thyristor in the static VAR compensator is controlled by the optimal firing angle of the power generation node, thereby changing the equivalent impedance of the thyristor-controlled reactor.

6. A power system regulation system, adopting a power system regulation method according to any one of claims 1 to 5, characterized in that: The system includes the following modules: Data acquisition module: used to obtain several node voltage sequences of each power generation node in the power grid and the distance between the power generation nodes; Node clustering module: It is used to cluster all power generation nodes in the power grid into several node clusters by using the node difference values ​​obtained by the distance between different power generation nodes, the overall difference of node voltages, and the correlation between node voltage sequences; Optimization and iteration module: used to obtain the initial trigger angles of all power generation nodes in the power grid, randomly generate new trigger angles for each power generation node, and use the particle swarm optimization algorithm to optimize and iterate all new trigger angles. During the iteration process, the node cluster adjustment weight of each node cluster and the node cluster association between the node clusters are calculated. The node cluster adjustment weight is determined based on the difference between the node voltage and the rated voltage and the trigger angle change. The node cluster association is determined based on the node difference value and the new trigger angle. The combined result of the node cluster adjustment weight and the node cluster association is used as the fitness in the particle swarm optimization algorithm. When the iteration terminates, the optimal trigger angle control vector is output. Power regulation module: used to control and regulate the thyristors in the static VAR compensator of each power generation node in the power grid using the optimal trigger angle control vector.

7. A power system regulation device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the power system regulation method according to any one of claims 1 to 5 are implemented.

Images