Power distribution network large-scale distributed photovoltaic transient equivalent modeling method, storage medium and electronic equipment
By dividing distributed photovoltaics into clusters and constructing equal value units, the problem of accurate characterization of large-scale distributed photovoltaics in the distribution network and the superior power grid is solved, and efficient equal value modeling is achieved, which is suitable for safety and stability analysis of large power grids , improves simulation accuracy and efficiency.
Patent Information
- Application Number
- CN202411011537.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-15
AI Technical Summary
The existing technology is difficult to accurately characterize the coupling interaction between large-scale distributed photovoltaics and superior power grids of the distribution network, and the distributed photovoltaic equivalent modeling efficiency is low and has great limitations, so it cannot be applied to safety and stability analysis of large power grids.
By dividing massive distributed photovoltaics into clusters, aggregating units according to the principle of power conservation, constructing an equal value network between the equal value unit and the original distribution network, eliminating unnecessary nodes, and establishing a REI equal value method and WARD equal value method using the same value method and the same value method of WARD The final equivalent model ensures that the injection power remains unchanged and improves simulation accuracy and efficiency.
Large-scale distributed photovoltaic equivalent modeling of distribution networks suitable for safety and stability analysis of large power grids is realized, which improves simulation accuracy and efficiency, and can accurately reflect the transient response between distributed photovoltaic and the power grid.
Smart Images

Figure CN118971131A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power grid transient simulation analysis, and in particular relates to a large-scale distributed photovoltaic transient equivalent modeling method for a distribution network, a storage medium and an electronic device. Background Art
[0002] Under the background of new power system construction, distributed photovoltaic power generation has become an important form of new energy development in the central and eastern power grid. In the first quarter of 2024, the national photovoltaic grid-connected capacity increased by 45.74 million kilowatts, of which distributed photovoltaic power generation was 23.8 million kilowatts; by the end of March 2024, the national photovoltaic power generation installed capacity reached 659 million kilowatts, of which distributed photovoltaic power generation was 280 million kilowatts. On the one hand, with the continuous advancement of large-scale distributed photovoltaic development, distributed photovoltaic power generation has shown a high-speed development trend of multi-point and wide-area, local high-density grid connection, and the distribution network has gradually changed from "passive" to "active", and the coupling interaction with the upper power grid has become increasingly stronger; on the other hand, considering the huge number of distributed photovoltaic power generation and its intricate grid structure with the distribution network, the establishment and calculation of detailed simulation models of distributed photovoltaic power generation are not feasible, so the existing simulation technology usually adopts the equivalent simulation model.
[0003] Publication No. CN114491886A discloses a general modeling method for active distribution network containing multiple types of distributed renewable energy, including: establishing a low-order general model structure of distributed renewable energy; connecting the obtained general model of distributed renewable energy in parallel with the traditional comprehensive load model as the general model structure of active distribution network; performing equivalent modeling on multiple nonlinear modules including low voltage active power management, low voltage reactive power management and current limiting of converter in multiple types of distributed renewable energy in active distribution network, and obtaining model parameters of each nonlinear module; using particle swarm optimization algorithm to perform parameter identification on linear modules in the general model of active distribution network; and calculating the final parameter value of the general model of active distribution network. The present invention can describe the general dynamics of multiple types of distributed renewable energy, and at the same time, it can also describe the equivalent characteristics of low voltage ride-through control and current limiting link of distributed renewable energy, and the model structure is simple and has high equivalent accuracy.
[0004] Publication No. CN116937665A discloses a distributed photovoltaic grid-connected management process optimization method, including constructing a Petri net model based on distributed photovoltaic grid-connected management data; constructing an association matrix based on the output matrix and input matrix of the Petri net model; optimizing the Petri net model based on the association matrix reorganization method; introducing time parameters into the Petri net model for simulation analysis and evaluating the simulation results. The present invention optimizes the distributed photovoltaic grid-connected service management process through Petri nets and association matrix reorganization methods. According to the simulation analysis results, the method of the present invention improves the efficiency of the 10kV distributed photovoltaic grid-connected management service process by 12.4%, greatly shortening the running time of the process. At the same time, it simplifies the cumbersome and redundant processes in the distributed grid-connected management service process, improves the work efficiency in the grid-connected management process, and further promotes the development of distributed photovoltaic management services. Summary of the invention
[0005] The technical problem to be solved by the present invention is to provide a large-scale distributed photovoltaic transient equivalent modeling method, storage medium and electronic equipment for distribution network in view of the deficiencies in the prior art.
[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0007] A method for transient equivalent modeling of large-scale distributed photovoltaic power generation in a distribution network is provided, comprising the following steps: S1, collecting information and data required for equivalent modeling; S2, building a detailed simulation model of distributed photovoltaic power generation in the distribution network, the detailed model is a standard for verifying whether the equivalent modeling is accurate or not, and the equivalent simulation models subsequently established need to compare and verify the simulation results with the detailed model; S3, building an equivalent simulation model of distributed photovoltaic power generation in the distribution network; S31, dividing a large amount of distributed photovoltaic power generation into a number of clusters according to certain principles, so that the distributed photovoltaic power generation characteristics within the clusters are similar, and the distributed photovoltaic power generation characteristics between the clusters are different; S32, on the basis of the completion of cluster division, performing unit aggregation, that is, according to the principle of power conservation, multiple distributed photovoltaic power generation units with similar characteristics within the same cluster are equivalent to one power source, so as to reduce the number of distributed photovoltaic power generation units; S33, according to the unit aggregation result, further constructing an equivalent network interconnected between the equivalent unit and the original distribution network, connecting the equivalent unit to the original distribution network through the equivalent network, and eliminating redundant nodes, so as to finally realize the equivalent modeling of distributed photovoltaic power generation; S4, verifying the effectiveness of the constructed equivalent simulation model according to the detailed simulation model.
[0008] Preferably, the step S31 further includes that, for a linear network containing any number of distributed power sources, when a short circuit fault occurs at a point k outside the system, the current response of each power source at the short circuit point k depends on the electromotive force of each power source and the impedance (i.e., transfer impedance) between the power source node and the short circuit point k. Considering the superposition principle of the linear network, the short circuit point current (AC component) can be expressed as:
[0009]
[0010] In the formula, represents the AC component value of the short-circuit current at point k outside the linear network, represents the electromotive force of the mth distributed power source in the linear network, Z mk represents the transfer impedance between the mth distributed power node and the short-circuit point k. From formula (1), we can know that the definition of transfer impedance is: The electromotive force of all distributed power sources outside is 0 (short circuit grounding). and The ratio is the transfer impedance between the power node m and the short-circuit point k.
[0011] Preferably, step S31 also includes, taking into account the uncertainty of the short-circuit fault location and the fact that the electrical distance between the PCC and the short-circuit point is negligible compared to the electrical distance between the distributed power source and the PCC, using the electrical distance between the distributed photovoltaic node and the PCC and the distributed photovoltaic model as cluster division indicators.
[0012] Preferably, the step S31 also includes, S311, determining the distribution network topology parameters and the model of each distributed photovoltaic manufacturer according to the preliminary fund collection results on site, and calculating the electrical distance between each distributed photovoltaic node and the PCC; S312, according to the model priority principle, performing initial grouping of each distributed photovoltaic according to the manufacturer model; S313, based on the electrical distance calculation results, using the k-means clustering algorithm to further divide the distributed photovoltaics of the same model after the initial grouping, to form a distributed photovoltaic cluster division result that comprehensively considers dynamic characteristics and electrical distance.
[0013] Preferably, step S32 also includes establishing an equivalent unit model according to the two principles of "model consistency" and "power conservation"; wherein the "model consistency" principle means that the model of the equivalent unit model needs to be the same as the original sub-cluster, so as to ensure that the dynamic characteristics of the distributed photovoltaic units before and after aggregation are relatively stable. The "power conservation" principle means that before and after the distributed photovoltaic units are aggregated, the power at the PCC where the distribution network is connected to the upper power grid needs to remain unchanged:
[0014]
[0015] Where Peqv,u represents the active power of the equivalent unit corresponding to sub-cluster u, P v,u represents the active power of the vth distributed photovoltaic unit in sub-cluster u, C u represents the set of all distributed photovoltaic units in subcluster u, Q eqv,u represents the reactive power of the equivalent unit corresponding to sub-cluster u, Q v,u Represents the reactive power of the vth distributed PV unit in subcluster u.
[0016] Preferably, the step S33 further includes: S331, constructing an equivalent network between the equivalent units / loads and the distributed photovoltaic access points of the original distribution network; S332, eliminating redundant nodes other than the equivalent units / loads and PCC.
[0017] Preferably, the step S331 further comprises: using the REI equivalent method to construct an equivalent network between the two, connecting the equivalent units / loads and the distribution network; wherein, for the original distributed photovoltaic access point, the injection current relationship is:
[0018]
[0019] In the formula, Represents the injected current of the original distributed photovoltaic access point, represents the injected power, Indicates the node voltage of the original distributed photovoltaic access point.
[0020] Similarly, the outflow current relationship of node G in sub-cluster u can be obtained:
[0021]
[0022] In the formula, represents the outflow current of node G in sub-cluster u, represents the outgoing power, It represents the voltage of node G in sub-cluster u. As mentioned above, the equivalent network has no power loss, so it can be known that the injected current of the original distributed photovoltaic access point remains unchanged before and after the equivalent value:
[0023]
[0024] In the formula, y v,u represents the branch admittance between the vth original distributed photovoltaic access point of sub-cluster u and node H in the REI equivalent network, Represents the voltage of node H in sub-cluster u. Similarly, the calculation formula of trunk admittance yu can be obtained:
[0025]
[0026] In the formula, yu represents the trunk admittance of subcluster u in the REI equivalent network. Considering formulas (5) and (6), The value of can be any. Substitute equations (3) and (4) into equations (5) and (6) respectively:
[0027]
[0028] Distributed photovoltaic grid-connected inverters usually operate at unity power factor, that is, they only send active power to the outside, so equation (7) can be further adjusted to:
[0029]
[0030] When performing load equalization, the construction process of the REI network is similar to the above content. The only difference is that the flow direction of the load power is opposite to that of the power supply. The positive and negative signs of the corresponding relationship results in equation (4-8) can be changed.
[0031] Preferably, the step S332 also includes using the WARD equivalent method to eliminate redundant nodes in the distribution network and establish a final equivalent model; wherein, first, it is necessary to determine the node sets to be retained and to be eliminated, the nodes to be retained are the equivalent power / load access points and PCC in the distribution network, and the remaining nodes in the distribution network are all nodes to be eliminated. Secondly, construct the node voltage equation of the distribution network. Define the equivalent power / load access points as the boundary node set (represented by symbol B), the PCC as the internal node set (represented by symbol I), and the nodes to be eliminated as the external node set (represented by symbol E), and write the node voltage equation of the distribution network in matrix form:
[0032]
[0033] Where Y represents the node admittance matrix, represents the node voltage matrix, Represents the node current matrix. Finally, the external node set is eliminated to obtain the WARD equivalent network containing only the boundary node set and the internal node set. After REI equivalent, the power / load of the original distributed photovoltaic / load access point of the distribution network has been equivalently migrated to the new equivalent power / load access point, and the original access point has been converted into a passive node. Combined with the previous node definition, it can be seen that the external nodes are all passive nodes, that is, The equivalent power / load access points are indirectly connected to the PCC through external nodes. There is no direct branch between the equivalent power / load access points and the PCC. BI =Y IB = 0. Substituting the above boundary conditions into equation (9), we can eliminate
[0034]
[0035] According to the simplified node admittance matrix, the equivalent network between equivalent power sources, load access points and PCCs can be obtained to form the final equivalent model.
[0036] A computer-readable storage medium is also provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any one of the methods for transient equivalent modeling of large-scale distributed photovoltaic power generation in a distribution network is implemented.
[0037] An electronic device is also provided, 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, any one of the methods for transient equivalent modeling of large-scale distributed photovoltaic power generation in a distribution network is implemented.
[0038] At present, the "decoupling property from the grid" of the distributed photovoltaic power electronic topology makes it quite different from the transient and steady-state response characteristics of the traditional power supply / load. The limitations of the traditional grid stability simulation analysis method that treats distributed photovoltaic as a "negative load" and directly offsets the load power are becoming more and more prominent, and it is difficult to accurately characterize the coupling interaction between the large-scale distributed photovoltaic of the distribution network and the upper grid. Publication No. CN114491886A discloses a general modeling method for an active distribution network containing multiple types of distributed new energy, including: establishing a low-order general model structure of distributed new energy; connecting the obtained general model of distributed new energy with the traditional comprehensive load model in parallel as the general model structure of the active distribution network. However, the application cannot model the equivalent of larger-scale distributed photovoltaics, and the modeling efficiency is low and the limitations are large. Publication No. CN116937665A discloses a distributed photovoltaic grid-connected management process optimization method, including constructing a Petri net model based on distributed photovoltaic grid-connected management data; and constructing an association matrix based on the output matrix and input matrix of the Petri net model. The application still cannot achieve a large-scale distributed photovoltaic equivalent modeling method for distribution networks suitable for large-scale power grid security and stability analysis.
[0039] Compared with the prior art, the present invention has the following beneficial effects:
[0040] The present invention adopts a large-scale distributed photovoltaic equivalent modeling method for distribution networks that is suitable for safety and stability analysis of large power grids. It is of great significance to study the coupling interaction between massive distributed photovoltaics and power grids, and is of great significance to building a new type of power system that is clean, low-carbon, safe, abundant, economical and efficient, coordinated in supply and demand, flexible and intelligent.
[0041] By classifying a large number of distributed photovoltaic units one by one according to certain standards or indicators, the operating characteristics of the same type of units are made similar or the same. On this basis, an equivalent model of the same type of units is constructed to ensure the simulation accuracy before and after the equivalence.
[0042] By constructing the REI equivalent network, the power injected into the distributed photovoltaic / load access point of the original distribution network remains unchanged, that is, the equivalent network ensures no power loss, which greatly improves the simulation accuracy; by using the WARD equivalent method to eliminate redundant nodes in the distribution network and establish the final equivalent model, the simulation efficiency and effect are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative labor.
[0044] Figure 1 : Schematic diagram of the distribution network structure of the large-scale distributed photovoltaic transient equivalent modeling method of the distribution network of the present invention;
[0045] Figure 2 : A specific flow chart of the REI equivalent method of the method for modeling the transient equivalent of large-scale distributed photovoltaic power distribution networks of the present invention;
[0046] Figure 3 : A specific flow chart of the WARD equivalent method of the transient equivalent modeling method of large-scale distributed photovoltaic power distribution network of the present invention;
[0047] Figure 4 : Schematic diagram of transient response curves of each model under 98% fault in the transient equivalent modeling method of large-scale distributed photovoltaic power distribution network of the present invention; DETAILED DESCRIPTION
[0048] In order to better understand the present invention, the content of the present invention is further clearly set forth in conjunction with the embodiments below, but the protection content of the present invention is not limited to the following embodiments. In the following description, a large number of specific details are provided in order to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details.
[0049] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.
[0050] Embodiment 1: This application provides a method for modeling transient equivalent of large-scale distributed photovoltaic power distribution network, specifically as follows Figure 1-Figure 4 As shown, the following steps are included:
[0051] S1, collect the information and data required for equivalent modeling;
[0052] The actual distribution network usually spans different voltage levels and has a complex grid structure. The capacity and model parameters of distributed photovoltaics are unknown, and a large number of distributed photovoltaics and load units are scattered on the low-voltage feeders of the distribution network, which further increases the complexity of distributed photovoltaic modeling in the actual distribution network. Therefore, it is necessary to fully collect key information such as the grid topology, line impedance, and source-load parameters of the actual distribution network to provide a solid foundation for the subsequent establishment of detailed models and equivalent models.
[0053] S2, build a detailed simulation model of distributed photovoltaic power distribution network. The detailed model is the standard for verifying the accuracy of equivalent modeling. The equivalent simulation models established subsequently need to compare and verify the simulation results with the detailed model;
[0054] The distributed photovoltaic power supply model considered in this embodiment is composed of a photovoltaic array and a grid-connected inverter. When the grid operates normally without faults, the grid-connected inverter adopts the maximum power point tracking (MPPT) control mode to convert the solar energy captured by the photovoltaic array into electrical energy as much as possible. When a fault occurs in the grid, the power supply makes a judgment by measuring the terminal voltage. If the voltage drop exceeds the set threshold, the inverter enters the fault ride-through control mode and actively generates reactive power to support the recovery of the grid voltage. Figure 1 The typical structure of a distribution network with large-scale distributed photovoltaics is shown. The root node of the distribution network system under study is a 110kV regional substation, which is connected to the upper power grid through transformers and lines, and is connected to multiple regional substations of the same voltage level through lines; the regional substation is then connected to the distribution network feeder of a lower voltage level through a step-down transformer; the distributed photovoltaic power sources and loads are connected to the distribution network system from the distribution network feeder of the low voltage level. In general, the grid structure of the distribution system is radial, and the distributed photovoltaics and loads are dispersed on the terminal low-voltage feeder of the distribution system.
[0055] S3, build a distributed photovoltaic equivalent simulation model for the distribution network;
[0056] S31, according to certain principles, the massive distributed photovoltaics are divided into several clusters, so that the distributed photovoltaic characteristics within the clusters are similar, and the distributed photovoltaic characteristics between the clusters are different;
[0057] S32, after the cluster division is completed, unit aggregation is performed, that is, according to the principle of power conservation, multiple distributed photovoltaic units with similar characteristics in the same cluster are regarded as one power source, so as to reduce the number of distributed photovoltaic units;
[0058] S33, according to the aggregation results of the units, further construct an equivalent network that connects the equivalent units with the original distribution network, connect the equivalent units to the original distribution network through the equivalent network, and eliminate redundant nodes, finally realizing the equivalent modeling of distributed photovoltaics.
[0059] S4, verify the validity of the constructed equivalent simulation model based on the detailed simulation model.
[0060] Preferably, the step S31 further includes that, for a linear network containing any number of distributed power sources, when a short circuit fault occurs at a point k outside the system, the current response of each power source at the short circuit point k depends on the electromotive force of each power source and the impedance (i.e., transfer impedance) between the power source node and the short circuit point k. Considering the superposition principle of the linear network, the short circuit point current (AC component) can be expressed as:
[0061]
[0062] In the formula, represents the AC component value of the short-circuit current at point k outside the linear network, represents the electromotive force of the mth distributed power source in the linear network, Z mk represents the transfer impedance between the mth distributed power node and the short-circuit point k. From formula (1), we can know that the definition of transfer impedance is: The electromotive force of all distributed power sources outside is 0 (short circuit grounding). and The ratio is the transfer impedance between the power node m and the short-circuit point k.
[0063] Preferably, step S31 also includes, taking into account the uncertainty of the short-circuit fault location and the fact that the electrical distance between the PCC and the short-circuit point is negligible compared to the electrical distance between the distributed power source and the PCC, using the electrical distance between the distributed photovoltaic node and the PCC and the distributed photovoltaic model as cluster division indicators.
[0064] Preferably, the step S31 further includes:
[0065] S311, based on the preliminary funding results at the site, determine the topological parameters of the distribution grid and the models of each distributed photovoltaic manufacturer, and calculate the electrical distance between each distributed photovoltaic node and the PCC;
[0066] S312, according to the principle of model priority, each distributed photovoltaic is initially grouped according to the manufacturer model;
[0067] S313, based on the electrical distance calculation result, the k-means clustering algorithm is used to further divide the distributed photovoltaics of the same model after the initial grouping, so as to form a distributed photovoltaic cluster division result that comprehensively considers the dynamic characteristics and electrical distance.
[0068] Preferably, step S32 also includes establishing an equivalent unit model according to the two principles of "model consistency" and "power conservation"; wherein the "model consistency" principle means that the model of the equivalent unit model needs to be the same as the original sub-cluster, so as to ensure that the dynamic characteristics of the distributed photovoltaic units before and after aggregation are relatively stable. The "power conservation" principle means that before and after the distributed photovoltaic units are aggregated, the power at the PCC where the distribution network is connected to the upper power grid needs to remain unchanged:
[0069]
[0070] Where P eqv,u represents the active power of the equivalent unit corresponding to sub-cluster u, P v,u represents the active power of the vth distributed photovoltaic unit in sub-cluster u, C u represents the set of all distributed photovoltaic units in subcluster u, Q eqv,u represents the reactive power of the equivalent unit corresponding to sub-cluster u, Q v,u Represents the reactive power of the vth distributed PV unit in subcluster u.
[0071] Preferably, the step S33 further includes:
[0072] S331, constructing an equivalent network between equivalent units / loads and distributed photovoltaic access points of the original distribution network;
[0073] S332, eliminate redundant nodes other than equivalent units / loads and PCC.
[0074] Preferably, the step S331 further comprises: using the REI equivalent method to construct an equivalent network between the two, connecting the equivalent units / loads and the distribution network; wherein, for the original distributed photovoltaic access point, the injection current relationship is:
[0075]
[0076] In the formula, Represents the injected current of the original distributed photovoltaic access point, represents the injected power, Indicates the node voltage of the original distributed photovoltaic access point.
[0077] Similarly, the outflow current relationship of node G in sub-cluster u can be obtained:
[0078]
[0079] In the formula, represents the outflow current of node G in sub-cluster u, represents the outgoing power, It represents the voltage of node G in sub-cluster u. As mentioned above, the equivalent network has no power loss, so it can be known that the injected current of the original distributed photovoltaic access point remains unchanged before and after the equivalent value:
[0080]
[0081] In the formula, y v,u represents the branch admittance between the vth original distributed photovoltaic access point of sub-cluster u and node H in the REI equivalent network, Represents the voltage of node H in sub-cluster u. Similarly, the calculation formula of trunk admittance yu can be obtained:
[0082]
[0083] In the formula, yu represents the trunk admittance of subcluster u in the REI equivalent network. Considering formulas (5) and (6), The value of can be any. Substitute equations (3) and (4) into equations (5) and (6) respectively:
[0084]
[0085] Distributed photovoltaic grid-connected inverters usually operate at unity power factor, that is, they only send active power to the outside, so equation (7) can be further adjusted to:
[0086]
[0087] When performing load equalization, the construction process of the REI network is similar to the above content. The only difference is that the flow direction of the load power is opposite to that of the power supply. The positive and negative signs of the corresponding relationship results in equation (4-8) can be changed.
[0088] Preferably, the step S332 also includes using the WARD equivalent method to eliminate redundant nodes in the distribution network and establish a final equivalent model; wherein, first, it is necessary to determine the node sets to be retained and to be eliminated, the nodes to be retained are the equivalent power / load access points and PCC in the distribution network, and the remaining nodes in the distribution network are all nodes to be eliminated. Secondly, construct the node voltage equation of the distribution network. Define the equivalent power / load access points as the boundary node set (represented by symbol B), the PCC as the internal node set (represented by symbol I), and the nodes to be eliminated as the external node set (represented by symbol E), and write the node voltage equation of the distribution network in matrix form:
[0089]
[0090] Where Y represents the node admittance matrix, represents the node voltage matrix, Represents the node current matrix. Finally, the external node set is eliminated to obtain the WARD equivalent network containing only the boundary node set and the internal node set. After REI equivalent, the power / load of the original distributed photovoltaic / load access point of the distribution network has been equivalently migrated to the new equivalent power / load access point, and the original access point has been converted into a passive node. Combined with the previous node definition, it can be seen that the external nodes are all passive nodes, that is, The equivalent power / load access points are indirectly connected to the PCC through external nodes. There is no direct branch between the equivalent power / load access points and the PCC. BI =Y IB = 0. Substituting the above boundary conditions into equation (9), we can eliminate
[0091]
[0092] According to the simplified node admittance matrix, the equivalent network between equivalent power sources, load access points and PCCs can be obtained to form the final equivalent model.
[0093] Preferably, step S4 also includes applying the same disturbance to the constructed detailed simulation model and the equivalent simulation model, drawing and comparing the transient response curves of the key physical quantities of each simulation model, so as to judge the degree of fit between the equivalent simulation model and the detailed simulation model. If the degree of fit between the transient response curves of the models is high, it means that the equivalent effect is good; otherwise, it means that there is a certain deviation in the equivalent modeling, and it is necessary to find out the cause of the simulation error, improve the existing equivalent modeling method, and re-perform equivalent modeling and model verification.
[0094] The method proposed in this embodiment establishes an equivalent simulation model, including 6 equivalent power sources and 1 equivalent load, as well as an equivalent grid between the equivalent power sources, load access points and PCC. This embodiment also establishes other simulation models for comparative analysis, including a single-machine equivalent model (the power of the distributed power sources and loads included in the feeder system of each 10kV substation in the distribution network is summed up respectively, and directly connected to the 10kV substation node as an equivalent model), an equivalent model that only considers the electrical distance grouping basis (referred to as equivalent model A), and an equivalent model that only considers the dynamic characteristics grouping basis (referred to as equivalent model B).
[0095] Fill in the fault card in the transient calculation module of the comprehensive stability program, set a short-circuit fault disturbance on a line in the outer area of the distribution network system under study, and a permanent three-phase short-circuit fault occurs somewhere on the line at 1s. At 1.12s, the line protection action removes the fault, and observes the dynamic change process of the electrical quantity at the PCC connected to the upper power grid under different simulation models under fault disturbance. The severity of the disturbance is determined by the location of the fault. The closer the fault location is to the distribution network system, the more serious the fault is, and the greater the disturbance to the distribution network system under study. According to the rules for filling in the fault card of the comprehensive stability program, the line fault location is defined as the percentage of the total length of the line from the node on the I side of the line, which is referred to as "x% fault" in this embodiment. Based on this, this embodiment selects three typical fault locations to fill in the fault card, namely 98% fault, 70% fault and 10% fault. According to the previous definition, it is easy to know that the 98% fault location is far away from the distribution network system under study, the 10% fault location is close to the distribution network system under study, and the 70% fault location is between the above two.
[0096] The transient response curves of each model under 98% fault are analyzed. Figure 4 (a)-(b) show the transient voltage and current response curves of each model. When t = 0 ~ 1s, the short circuit fault does not occur and the regional distribution network is in a steady-state operation process. Figure 4 (a)-(b) show that there are obvious voltage and current simulation deviations between the single-machine equivalent model, equivalent model A and equivalent model B and the detailed model; compared with several other equivalent models, the steady-state simulation results of PCC voltage and current of the proposed equivalent model are more consistent with the detailed model, and can accurately reflect the steady-state operation of voltage and current at the PCC of the regional distribution network and the upper-level power grid.
[0097] When t=1s, a permanent fault occurs on a line in the nearby area outside the distribution network, and the distribution network is disturbed by the fault, with voltage drop and current oscillation at the PCC. Figure 4 (a)-(b) It can be found that although the single-machine equivalent model has the advantages of simple calculation and convenient operation, it ignores the differences in electrical distance and dynamic characteristics between each distributed photovoltaic in the regional distribution network and the PCC. Compared with the detailed model simulation results, the amplitude and waveform trend of the voltage and current transient response curves of the model are significantly different. Equivalent models A and B respectively considered the electrical distance factor and dynamic characteristic factor of the distributed power source from the PCC when modeling. Figure 4 As shown in (a), the PCC voltage simulation results of equivalent models A and B have similar waveform trends to the detailed model, but there is a deviation in the amplitude. Figure 4In (b), the current dynamic response curve of equivalent model A is almost always higher than that of the detailed model after the fault occurs because the dynamic characteristics of different distributed photovoltaics are ignored; the equivalent model B takes into account the dynamic characteristics of the units. After the fault occurs and in the early stage of the fault recovery process, it has a higher simulation accuracy, but in the later stage of the fault recovery, the deviation of the current response curve of the model from the detailed model gradually increases. This is because the difference in the electrical distance between different distributed photovoltaics and PCC is not taken into account in the modeling process of the equivalent model B, which makes it difficult for the equivalent model B to accurately reflect the contribution of different distributed power sources in the detailed model to the current at the PCC. Therefore, when the distribution network system is in a steady state and in the transition process of recovering to a steady state, the current response at the PCC deviates from the detailed model.
[0098] Compared with the single-machine equivalent model, equivalent models A and B mentioned above, the proposed equivalent model comprehensively considers the electrical distance between distributed photovoltaic and PCC and the dynamic characteristics of the unit, and has higher simulation accuracy under fault disturbance. Figure 4 As shown in (a), after the fault occurs, the PCC voltage simulation curve of the proposed equivalent model almost completely coincides with the detailed model; Figure 4 In (b), the current response curve of the proposed equivalent model is basically consistent with that of the detailed model, with a relatively smaller amplitude deviation, and has higher transient simulation accuracy, which can accurately reflect the transient response of the distribution network with large-scale distributed photovoltaic access.
[0099] Embodiment 2: The present invention also provides an electronic device for a method for transient equivalent modeling of large-scale distributed photovoltaic power distribution networks, the electronic device comprising a memory, at least one processor, a computer program stored in the memory and executable on the at least one processor, and at least one communication bus. The memory can be used to store the computer program, and the processor implements a method for transient equivalent modeling of large-scale distributed photovoltaic power distribution networks in Embodiment 1 by running or executing the computer program stored in the memory and calling the data stored in the memory.
[0100] Preferably, the electronic device is a server, and the system needs to build a suitable server in the cloud and deploy corresponding applications. The server needs to support functions such as multi-user access, data storage and real-time control.
[0101] Data processing and control: Run data processing and control programs on the server, analyze the data collected by the sensor through data analysis and algorithm processing, and issue control instructions based on the analysis results to achieve remote control.
[0102] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.
[0103] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0104] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0105] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0106] In addition, those skilled in the art should understand that, although there are many problems in the prior art, each embodiment or technical solution of the present invention can be improved in only one or several aspects, without having to solve all the technical problems listed in the prior art or background technology at the same time. Those skilled in the art should understand that the content not mentioned in a claim should not be used as a limitation on the claim.
[0107] Finally, it should be noted that the above embodiments are only used to illustrate the preferred embodiments of the present invention and are not limiting. Other modifications or equivalent substitutions made to the technical solutions of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solutions of the present invention.
Claims
1. A method for transient equivalent modeling of large-scale distributed photovoltaic power distribution networks, characterized by: The following steps are included: S1, collect the information and data required for equivalent modeling; S2, build a detailed simulation model of distributed photovoltaic power distribution network. The detailed model is the standard for verifying the accuracy of equivalent modeling. The equivalent simulation models established subsequently need to compare and verify the simulation results with the detailed model; S3, build a distributed photovoltaic equivalent simulation model for the distribution network; S31, according to certain principles, the massive distributed photovoltaics are divided into several clusters, so that the distributed photovoltaic characteristics within the clusters are similar, and the distributed photovoltaic characteristics between the clusters are different; S32, after the cluster division is completed, unit aggregation is performed, that is, according to the principle of power conservation, multiple distributed photovoltaic units with similar characteristics in the same cluster are regarded as one power source, so as to reduce the number of distributed photovoltaic units; S33, according to the aggregation results of the units, further construct an equivalent network that connects the equivalent units with the original distribution network, connect the equivalent units to the original distribution network through the equivalent network, and eliminate redundant nodes, finally realizing the equivalent modeling of distributed photovoltaics; S4, verify the validity of the constructed equivalent simulation model based on the detailed simulation model.
2. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 1, characterized in that: The step S31 also includes that, for a linear network containing any number of distributed power sources, when a short circuit fault occurs at a point k outside the system, the current response of each power source at the short circuit point k depends on the electromotive force of each power source and the impedance (i.e., transfer impedance) between the power source node and the short circuit point k; considering the superposition principle of the linear network, the short circuit point current (AC component) can be expressed as: In the formula, represents the AC component value of the short-circuit current at point k outside the linear network, represents the electromotive force of the mth distributed power source in the linear network, Z mk represents the transfer impedance between the mth distributed power node and the short-circuit point k; from formula (4-1), we can know the definition of transfer impedance: when The electromotive force of all distributed power sources outside is 0 (short circuit grounding). and The ratio is the transfer impedance between the power node m and the short-circuit point k.
3. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 1, characterized in that: The step S31 also includes, considering the uncertainty of the short-circuit fault location and that the electrical distance between the PCC and the short-circuit point is negligible compared to the electrical distance between the distributed power source and the PCC, using the electrical distance between the distributed photovoltaic node and the PCC and the distributed photovoltaic model as cluster division indicators.
4. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 1, characterized in that: The step S31 further includes: S311, based on the preliminary funding results at the site, determine the topological parameters of the distribution grid and the models of each distributed photovoltaic manufacturer, and calculate the electrical distance between each distributed photovoltaic node and the PCC; S312, according to the principle of model priority, each distributed photovoltaic is initially grouped according to the manufacturer model; S313, based on the electrical distance calculation result, the k-means clustering algorithm is used to further divide the distributed photovoltaics of the same model after the initial grouping, so as to form a distributed photovoltaic cluster division result that comprehensively considers the dynamic characteristics and electrical distance.
5. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 1, characterized in that: The step S32 further includes: Establish equivalent unit model according to the two principles of "model consistency" and "power conservation"; Among them, the "model consistency" principle means that the model of the equivalent unit model needs to be the same as the original sub-cluster, so as to ensure that the dynamic characteristics of the distributed photovoltaic units are relatively stable before and after aggregation; the "power conservation" principle means that the power at the PCC where the distribution network is connected to the upper power grid needs to remain unchanged before and after the distributed photovoltaic units are aggregated: Where P eqv,u represents the active power of the equivalent unit corresponding to sub-cluster u, P v,u represents the active power of the vth distributed photovoltaic unit in sub-cluster u, C u represents the set of all distributed photovoltaic units in subcluster u, Q eqv,u represents the reactive power of the equivalent unit corresponding to sub-cluster u, Q v,u Represents the reactive power of the vth distributed PV unit in subcluster u.
6. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 1, characterized in that: The step S33 further includes: S331, constructing an equivalent network between equivalent units / loads and distributed photovoltaic access points of the original distribution network; S332, eliminate redundant nodes other than equivalent units / loads and PCC.
7. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 6, characterized in that: The step S331 also includes using the REI equivalent method to construct an equivalent network between the two, connecting the equivalent units / loads and the distribution network grid; Among them, for the original distributed photovoltaic access point, the injection current relationship is: In the formula, Represents the injected current of the original distributed photovoltaic access point, represents the injected power, Indicates the node voltage of the original distributed photovoltaic access point; Similarly, the outflow current relationship of node G in sub-cluster u can be obtained: In the formula, represents the outflow current of node G in sub-cluster u, represents the outgoing power, represents the voltage of node G of sub-cluster u; As mentioned above, there is no power loss in the equivalent network, so the injected current of the original distributed photovoltaic access point remains unchanged before and after the equivalent value: In the formula, y v,u represents the branch admittance between the vth original distributed photovoltaic access point of sub-cluster u and node H in the REI equivalent network, represents the voltage of node H of sub-cluster u; Similarly, the calculation formula of trunk admittance yu can be obtained: Where yu represents the trunk admittance of subcluster u in the REI equivalent network; Considering equations (5) and (6), The value of can be any. Substitute equations (3) and (4) into equations (5) and (6) respectively: Distributed photovoltaic grid-connected inverters usually operate at unity power factor, that is, they only send active power to the outside, so equation (7) can be further adjusted to: When performing load equalization, the construction process of the REI network is similar to the above content. The only difference is that the flow direction of the load power is opposite to that of the power supply. The positive and negative signs of the corresponding relationship results in equation (8) can be changed.
8. The method for transient equivalent modeling of large-scale distributed photovoltaic power distribution network according to claim 6, characterized in that: The step S332 also includes using the WARD equivalent method to eliminate redundant nodes in the power distribution network and establish a final equivalent model; Among them, first, it is necessary to determine the node sets to be retained and to be eliminated. The nodes to be retained are the equivalent power / load access points and PCC in the distribution network, and the remaining nodes in the distribution network are all nodes to be eliminated; secondly, the node voltage equation of the distribution network is constructed; the equivalent power / load access points are defined as the boundary node set (symbol B), the PCC is defined as the internal node set (symbol I), and the nodes to be eliminated are the external node set (symbol E), and the node voltage equation of the distribution network is written in matrix form: Where Y represents the node admittance matrix, represents the node voltage matrix, Represents the node current matrix; finally, the external node set is eliminated to obtain the WARD equivalent network containing only the boundary node set and the internal node set; after REI equivalent, the power / load of the original distributed photovoltaic / load access point of the distribution network has been equivalently migrated to the new equivalent power / load access point, and the original access point has been converted into a passive node. Combined with the previous node definition, it can be seen that the external nodes are all passive nodes, that is, The equivalent power / load access points are indirectly connected to the PCC through external nodes. There is no direct branch between the equivalent power / load access points and the PCC. BI =Y IB = 0; Substituting the above boundary conditions into equation (9), we can eliminate According to the simplified node admittance matrix, the equivalent network between equivalent power sources, load access points and PCCs can be obtained to form the final equivalent model.
9. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method for transient equivalent modeling of large-scale distributed photovoltaic power generation in a distribution network as claimed in any one of claims 1 to 8.
10. An electronic 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 method for transient equivalent modeling of large-scale distributed photovoltaic power generation in a distribution network as described in any one of claims 1 to 8 is implemented.
