Power distribution system reliability index analysis and evaluation method considering DG and SOPs

Abstract

The invention relates to a power distribution system reliability index analysis and evaluation method considering DG and SOPs, and the method comprises the steps: building a DG island division and SOPs transfer model considering time sequence characteristics, precisely quantifying the influence of DG output fluctuation and SOPs dynamic transfer adjustment on a reliability index, and achieving the explicit expression of the reliability index of a power distribution network considering the access of DG and SOPs through a high-efficiency analysis method. And on the basis of time sequence element modeling accuracy, the model complexity is reduced, and the reliability evaluation efficiency is greatly improved. Besides, a reliability index analysis expression mode in a matrix form can clearly show the power failure influence of all equipment faults on load nodes, provides a visual tool for operation and maintenance personnel of a power distribution system, and can be used for subsequent sensitivity analysis and weak link identification of the system.

Description

Technical Field

[0001] The present invention belongs to the technical field of power systems, and in particular to a distribution system reliability index analysis and evaluation method taking DG and SOPs into account. Background Art

[0002] With the widespread integration of distributed generation (DG) and intelligent soft open points (SOPs), distribution networks are evolving from traditional radial, unidirectional power flow structures to multi-source coordinated, flexible, and controllable active networks. The islanding capability of DG and the dynamic power flow regulation of SOPs offer new paths to improving power supply reliability, but they also introduce challenges such as output volatility, random network topology, and complex protection and control strategies.

[0003] Existing distribution network reliability assessment methods usually assume that the operating status of the distribution network is constant within the statistical period, and calculate the reliability index through analytical methods (such as the minimum cut set method, the minimum path set method) or Monte Carlo simulation methods. This type of method is based on a radial network structure, with a fixed power source and load model as the core, and is suitable for simple distribution networks with unidirectional power flow. However, the access of DG breaks the boundary between power source and load. Its output is affected by the timing fluctuations of wind and solar resources, energy storage charging and discharging strategies, and dynamic changes in load demand. Traditional static models cannot accurately characterize the differences in multi-period operating scenarios. In addition, the island operation capability of DG depends on the dynamic coupling of fault location, output level and network topology reconstruction strategy. Traditional methods ignore the timing characteristics and dynamic island strategies, resulting in significant deviations in the calculation of reliability indicators.

[0004] In order to deal with the timing characteristics of DG and SOPs, existing research can be divided into two categories: analytical methods based on component probability models and simulation methods based on timing models. The analytical method characterizes the DG output fluctuations through Markov state transitions, multi-state models or non-parametric kernel density estimation, and calculates the reliability index by combining the probability of island formation. Its computational efficiency is high, but it is difficult to accurately describe the dynamic characteristics of timing and complex islanding strategies. The simulation method relies on the sequential Monte Carlo method to construct a wind, solar and storage timing model, and dynamically analyzes the islanding recovery range and network reconstruction strategy under fault scenarios. Although it can reflect the differences in operation in multiple time periods, it faces problems such as high computational cost and large sample requirements. Both methods have efficiency bottlenecks when dealing with large-scale distribution networks: the analytical method reduces accuracy due to simplified islanding strategies (such as the fixed islanding range assumption), while the simulation method is limited by the simulation complexity of high-dimensional timing scenarios and is difficult to meet real-time requirements. In summary, existing research is generally based on the assumption of a constant operating state, making it difficult to accurately quantify the impact of timing differences in DG output, load demand, and fault repair processes on reliability indicators. Furthermore, the ability to efficiently model the coordinated island operation of multiple DGs in large-scale networks is insufficient. Therefore, an evaluation method that takes into account both timing dynamic characteristics and computational efficiency is urgently needed.

[0005] Existing reliability assessment methods face many common challenges when dealing with high-penetration DG and SOPs access: First, the dynamic islanding strategy modeling is insufficient. Existing studies mostly rely on fixed island range assumptions or enumerated fault scenarios, and lack explicit expression of the coordinated mechanism of island division, network reconstruction and fault repair; Second, the timing-probability coupling characteristics are difficult to quantify. The timing randomness of DG output, load demand and component failures needs to be modeled through high-dimensional joint distribution. Traditional analytical methods and simulation methods both have the dimensionality curse problem when dealing with such high-dimensional uncertainties, and it is difficult to accurately describe the dynamic coupling relationship of multiple time scales; Finally, the contradiction between computational efficiency and accuracy is prominent. The analytical method suffers from accuracy loss due to model simplification, while the simulation method is difficult to apply to large-scale systems due to its high computational complexity. Both methods cannot take into account both evaluation efficiency and accuracy, which limits their practicality in high-penetration DG and SOPs scenarios. Summary of the Invention

[0006] The purpose of the present invention is to overcome the shortcomings of the existing technology and propose an analytical evaluation method for the reliability index of a distribution system taking into account DG and SOPs. By establishing a fault impact association matrix, the spatial correlation relationship between the consequences of a fault event and the faulty components is expressed, providing a model basis for subsequent reliability evaluation; secondly, for a complex distribution system taking into account DG and SOPs, a DG island partitioning and SOPs power transfer model taking into account time sequence characteristics is established, and the DG island recovery matrix and SOPs power transfer matrix after the fault are solved to characterize the relationship between the DG island operation and the SOPs flexible power transfer strategy and the power supply recovery of the load node; finally, based on matrix algebraic operations on the fault impact association matrix, the DG island recovery matrix, the SOPs power transfer matrix and the fault event parameter vector, an explicit analytical expression of the reliability index of the complex distribution system taking into account DG and SOPs is realized.

[0007] The present invention solves the technical problem by adopting the following technical solutions:

[0008] The analytical evaluation method for distribution system reliability indicators taking into account DGs and SOPs includes the following steps:

[0009] Step 1: Determine the impact of distribution network branch faults on load nodes and establish a fault impact correlation matrix;

[0010] Step 2: Establish the DG island division and SOPs transfer model, and solve the island recovery matrix and SOPs transfer matrix after the fault;

[0011] Step 3: Establish a reliability analysis and evaluation model based on the fault impact correlation matrix;

[0012] Step 4: Based on the fault impact correlation matrix established in step 1 and the evaluation model in step 3, perform explicit analytical calculation of the reliability index of the distribution system taking into account the access of DG and SOPs.

[0013] Furthermore, the specific implementation method of step 1 is: the impact of branch failure on load nodes is divided into four types:

[0014] Impact type a: A branch fault disconnects all power supply paths to the load, and power can only be restored after the fault is repaired.

[0015] Impact type b: A branch fault causes all power supply paths to the load to be disconnected. After the fault is isolated, the load is restored to the main power supply;

[0016] Impact type c: A branch fault causes all power supply paths to the load to be disconnected. After the fault is isolated, the load can be transferred to the backup power source through the tie line to restore power supply;

[0017] Impact type d: Branch failure has no impact on the load node;

[0018] According to the four types of impacts of branch faults on load nodes, the fault impact matrix FEIM is constructed, where FEIM includes FEIM-A, FEIM-B, FEIM-C and FEIM-D.

[0019] According to the fault location, the impact type of the fault on the load node and the load transfer path are analyzed in detail, and the power supply path matrix PSPM, segmented switch matrix SSM and connection matrix TSM are used to represent them:

[0020]

[0021] In FEIM, FEIM-A, FEIM-B, FEIM-C and FEIM-D are represented as follows:

[0022] FEIM-A=PSPM○SSM○(ONE-TSM)

[0023]

[0024] Among them, the matrix ONE represents an N-dimensional square matrix whose matrix elements are all 1.

[0025] Furthermore, the specific implementation method of step 2 is: constructing the DG power supply path matrix DGPSPM:

[0026]

[0027] Building an Island Recovery Matrix (IRM) DG , indicating the range of power supply restoration in isolated islands after branch failure at different locations, IRM DG It is an N-dimensional square matrix, the row number is the branch number, the column number is the load node number, when the matrix element When branch i fails, node j is powered by the island. Then, when branch i fails, node j cannot be restored to power by the island;

[0028] When DG has no capacity constraint, the island recovery matrix IRM DG,uncon The calculation method is:

[0029] IRM DG,uncon =FEIM-A-FEIM-A∩DGPSPM

[0030] The operator “∩” represents the bitwise AND operation of matrix elements, and also means that if the element values ​​at the same position in the two matrices are both “1”, then the element value at the corresponding position in the new matrix is ​​“1”, otherwise it is “0”;

[0031] Constructing SOPs transfer matrix PTM SOPs, indicates the range of power supply recovery of SOPs when different position branch faults, PTM SOPs is an N-dimensional square matrix, the row number is the branch number, and the column number is the load node number. When the matrix element , then node j is recovered by power supply by SOPs when branch i fails; when the matrix element , then node j cannot be recovered by power supply by SOPs when branch i fails.

[0032] The calculation method of PTM SOPs,uncon under the condition of no capacity constraint of SOPs is:

[0033] PTM SOPs,uncon = FEIM-C

[0034] Moreover, the specific implementation method of step 3 is that there are N k DGs in the power distribution system, and there are N l SOPs ports, so the island recovery matrix formed by the kth DG is IRM DG,k , and the transfer matrix of the lth SOPs port is PTM SOPs,l , a DG island division and SOPs transfer optimization model at a moment t is constructed, and the objective function of the model is:

[0035]

[0036] , wherein represents the element in the ith row and jth column of the kth modified island recovery matrix, represents the element in the ith row and jth column of the lth modified SOPs transfer matrix, ω j represents the importance of load node j, represents the active power demand of load node j at the current moment t.

[0037] Moreover, the transfer optimization model further includes constraint conditions, and the constraint conditions include DG capacity constraints, SOPs capacity constraints, power supply range constraints, network connectivity constraints, load weight constraints, system power flow constraints and system operation constraints.

[0038] The advantages and positive effects of the present application are:

[0039] This paper establishes a DG islanding and SOPs transfer model that takes into account timing characteristics, accurately quantifying the impact of DG output fluctuations and SOPs dynamic transfer adjustments on reliability indicators. Furthermore, through an efficient analytical method, it explicitly expresses the reliability indicators of the distribution network that takes into account DG and SOPs access. This reduces model complexity based on the accuracy of sequential element modeling, significantly improving the efficiency of reliability assessment. Furthermore, the matrix-based analytical expression of reliability indicators clearly demonstrates the impact of all equipment failures on load node outages, providing distribution system operators with an intuitive tool for subsequent sensitivity analysis and identifying system weaknesses. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of the explicit analytical evaluation method for power supply reliability indicators of a complex power distribution system according to the present invention;

[0041] Figure 2 Schematic diagram of the distribution network node branch containing DG and SOPs of the present invention;

[0042] Figure 3 IRM before and after the present invention is modified DG Schematic diagram of;

[0043] Figure 4 PTM before and after the present invention is modified SOPs Schematic diagram of;

[0044] Figure 5 This is a structural diagram of a 137-node example of the present invention;

[0045] Figure 6 Schematic diagram of the number of power outages at each load node in various scenarios according to an embodiment of the present invention;

[0046] Figure 7 Schematic diagram of power outage time of each load node in various scenarios of the present invention. DETAILED DESCRIPTION

[0047] The present invention is further described below in conjunction with the accompanying drawings.

[0048] Analytical evaluation method of distribution system reliability index considering DG and SOPs, such as Figure 1 As shown, the following steps are included:

[0049] Step 1: Determine the impact of distribution network branch faults on load nodes and establish a fault impact correlation matrix.

[0050] This step spatially categorizes the impact of branch faults on load nodes in traditional passive distribution networks, constructing a fault impact correlation matrix that characterizes the location and degree of impact of fault events and affected loads. This provides a matrix model foundation for subsequent analytical evaluation of distribution network reliability indicators.

[0051] by Figure 2 Taking the dual-feeder distribution system with DGs and SOPs as an example, spatial correlation matrix modeling for reliability assessment is performed. It is assumed that a circuit breaker is installed at the busbar outlet and each branch is equipped with a sectionalizer. The sectionalizer is normally closed and the tie switch is normally open. The substation busbar, circuit breaker, and switchgear are highly reliable, so only a single branch failure scenario is considered.

[0052] The present invention classifies the impact of branch faults on load nodes into four types, and the characteristics of each fault impact type are as follows:

[0053] Impact type a: A branch fault causes all power supply paths to the load to be disconnected, and power supply can only be restored after the fault is repaired; Impact type b: A branch fault causes all power supply paths to the load to be disconnected. After the fault is isolated, the load can be restored to the main power supply; Impact type c: A branch fault causes all power supply paths to the load to be disconnected. After the fault is isolated, the load can be transferred to the backup power supply through the interconnection line to restore power supply; Impact type d: The branch fault has no impact on the load node.

[0054] In order to correspond to the four types of fault impact, the present invention constructs four types of fault effect association matrices (FEIM), namely FEIM-A, FEIM-B, FEIM-C and FEIM-D. The row number in FEIM corresponds to the branch number, and the column number corresponds to the load node number. The elements in FEIM are "0" or "1". When the matrix element is "1", it means that the branch fault will cause the load to lose power. When the matrix element is "0", it means that the branch fault with the corresponding number has no effect on the load. Taking FEIM-A as an example, if the element FEIM-A in the matrix is ij =1, indicating that the impact of branch i fault on load node j is type a, otherwise, FEIM-A ij =0.

[0055] By analyzing the impact of branch faults on load nodes, we found that the type of fault impact is primarily determined by the following factors: whether the faulty branch is located in the load node's power supply path, the configuration of each branch's sectionalizing switches, and the presence of a valid tie line downstream of the faulty branch. Therefore, the present invention determines the fault impact type based on the upstream and downstream relationship between the faulty branch and the load node, as well as the location of the sectionalizing switches and tie lines.

[0056] When the fault branch i is upstream of the load node j, that is, the fault branch i is on the power supply path of the load node j, then the fault branch i and the load node j belong to the same feeder, then the fault impact type is a or c. If there is no interconnecting line that plays a power transfer role downstream of the fault branch i, then the fault impact type is a; if there is a interconnecting line that plays a power transfer role downstream of the fault branch i, then the fault impact type is c.

[0057] When faulty branch i is not on the power supply path to load node j, meaning that the faulty branch i and load node j may belong to the same feeder or different feeders, the fault impact type is b or d. If a sectionalizing switch exists between the faulty branch i and load node j, the impact type is b. If there is no sectionalizing switch between the faulty branch i and load node j, the faulty branch i and load node j belong to different feeders, and the impact type is d.

[0058] Based on the fault location, we will analyze in detail the impact type of the fault on the load node and the load transfer path. To facilitate subsequent matrix operations, three matrices are used to analytically represent the four fault impact type judgment dimensions mentioned above: the power supply path matrix (PSPM), the sectionalizing switch matrix (SSM), and the tie switch matrix (TSM). The elements in each matrix are defined as follows:

[0059]

[0060] Based on the above judgment of the fault impact type and the definitions of PSPM, SSM, and TSM, the analytical expressions of FEIM-A, FEIM-B, FEIM-C, and FEIM-D are as follows.

[0061]

[0062]

[0063] Wherein, the matrix ONE represents an N-dimensional square matrix whose elements are all 1. Based on the derived four types of FEIM, a matrix model foundation is provided for the subsequent analytical evaluation of distribution network reliability indicators.

[0064] Step 2: Establish the DG island division and SOPs transfer model, and solve the island recovery matrix and SOPs transfer matrix after the fault.

[0065] This step targets complex distribution systems connected to DGs and SOPs. At the temporal level, a DG islanding and SOPs power transfer model is established that takes into account time series characteristics. The post-fault DG islanding recovery matrix and SOPs power transfer matrix are solved to characterize the spatial correlation between DG islanding operation, SOPs flexible power transfer strategy, and load node power supply recovery.

[0066] Analysis of the four types of load impacts of branch faults after DG integration reveals that, after isolation of the sectionalizer for Type B and Type C faults, power is restored by the main power source and tie lines. The transfer paths remain unchanged before and after DG integration. However, Type D faults have no impact on load nodes, so only Type A faults require island power restoration. Therefore, DG island power restoration strategies targeting Type A faults require research. Similarly, comparing the four types of load impacts of branch faults before and after SOPs integration, Type A and Type B faults are both restored by the main power source, while Type D faults have no impact on load nodes. Therefore, SOPs-based power restoration strategies require research targeting Type C faults.

[0067] According to the above analysis, we can get Figure 2 FEIM-A and FEIM-C of the distribution network structure. The present invention defines a distributed generation power supply path matrix (DGPSPM).

[0068]

[0069] In order to express the relationship between load nodes and island recovery range, the present invention defines the island recovery matrix (IRM) of DGs. DG ). IRM DG It is an N-dimensional square matrix, which represents the range of power supply restoration of the island after branch failure at different locations. Among them, the row number is the branch number, the column number is the load node number, and when the matrix element When branch i fails, node j is powered by the island. Then, when branch i fails, node j cannot be restored to power by the island.

[0070] When DG has no capacity constraint, the island recovery matrix IRM DG,uncon Calculate according to the following formula:

[0071] IRM DG,uncon =FEIM-A-FEIM-A∩DGPSPM (9)

[0072] The operator "∩" represents a bitwise AND operation on matrix elements. That is, if the values ​​of the elements at the same position in both matrices are simultaneously "1," the value of the element at the corresponding position in the new matrix is ​​"1," otherwise it is "0." Equation (9) implies that, without considering the capacity constraints of the DG, a power outage will only occur if a branch fault disconnects the power supply path from the load node to both the main power source and the DG. Otherwise, the load node will be restored to power by the DG.

[0073] Similarly, in order to express the relationship between load nodes and SOPs transfer range, the present invention defines the SOPs transfer matrix (power transfer matrix of SOPs, PTM SOPs ),PTM SOPs It is an N-dimensional matrix, which indicates the scope of power supply restoration by SOPs when branches at different locations fail. The row number is the branch number, and the column number is the load node number. When the matrix element When branch i fails, node j is restored to power supply by SOPs; when the matrix element Then, when branch i fails, node j cannot be restored to power by SOPs.

[0074] Under the condition of no capacity constraint in SOPs, PTM SOPs,uncon Calculate according to the following formula:

[0075] PTM SOPs,uncon =FEIM-C (10)

[0076] Since SOPs is a new type of distribution component that replaces traditional tie switches, PTM SOPs,uncon It is FEIM-C. Since the capacity of DG and SOPs actually connected to the distribution network is limited, the impact of capacity constraints on DG island power supply and SOPs transfer strategy must be considered. Therefore, it is necessary to reduce some loads within the DG island range and SOPs transfer range. The corresponding matrix operation is to IRM DG,uncon Some of the "1" elements in the _ are set to "0" to form a modified IRM DG , for PTM SOPs,uncon Some of the "1" elements in are set to "0" to form a modified PTM SOPs .

[0077] Step 3: Establish a reliability analysis and evaluation model based on the fault impact correlation matrix.

[0078] This step is based on the fault impact correlation matrix, DG island recovery matrix, SOPs transfer matrix, fault event parameter vector and reliability parameters to achieve a one-time explicit analytical expression of the reliability index of the complex distribution system taking into account DGs and SOPs.

[0079] In order to maximize the role of DG and SOPs in improving the reliability of system power supply and restore load power supply as much as possible, the present invention establishes an optimization model for DG island division and SOPs transfer range under fault conditions. k There are N SOPs ports in total. l The island recovery matrix formed by the kth DG is recorded as IRM DG,k , the transfer matrix of the lth SOPs port is recorded as PTM SOPs,l , then the DG island division and SOPs transfer optimization model at a certain time t can be constructed.

[0080] The objective function of the model is to recover as many loads of higher importance as possible, which is expressed as follows:

[0081]

[0082] Where, represents the element in row i and column j of the kth island recovery matrix after correction, represents the element in row i and column j of the modified l-th SOPs port transfer matrix, ω j represents the importance of load node j, It represents the active power demand of load node j at the current time t.

[0083] The constraints are as follows.

[0084] 1) DG capacity constraints

[0085]

[0086] Where, They represent the active and reactive outputs of the DG in the kth island at the current time t, Represents the reactive power demand of load node j at the current time t.

[0087] 2) SOPs capacity constraints

[0088]

[0089] Where, and is the active power and reactive power injected into the node by the lth SOPs port at time t, is the capacity of the lth SOPs port.

[0090] 3) Power supply range constraints

[0091]

[0092] Where, It represents the power supply restoration range of the kth island before correction, The lth SOPs port forwarding range before correction. This constraint states that the power supply range restored to the kth island after correction cannot exceed the kth island's power supply range before correction, and the lth SOPs port forwarding range after correction cannot exceed the lth SOPs port forwarding range before correction. Power supply ranges cannot overlap. Each load node can only be restored by at most one island per fault. Each load node can only be restored by at most one SOPs forwarding range per fault.

[0093] 4) Network connectivity constraints

[0094]

[0095] Where R j M represents the node set on the power supply path from DG to load node j. j represents the set of nodes on the power supply path from SOPs to load node j. If node j can recover power when branch i fails, all nodes on the power supply path from DG and SOPs to node j can also recover power.

[0096] 5) Load weight constraint

[0097]

[0098] The sum of the weighted importance of all loads in the entire network is 1.

[0099] 6) System power flow constraints

[0100]

[0101] Where r ij is the resistance of branch ij, x ij is the reactance of branch ij, P ij is the active power at the head end of branch i, j; Q ij is the reactive power of the branch i, j; v(j) is the set of end nodes of the branch with j as the first node; u(j) is the set of first node of the branch with j as the last node.

[0102] 7) System operation constraints

[0103]

[0104] Where U imin is the minimum voltage amplitude allowed at node i, U imax is the maximum voltage allowed at node i, I imax It is expressed as the maximum current carrying capacity allowed to pass through the i-th branch. ois the minimum per-unit value of the fault side node voltage, which is generally 1.0. is the node voltage of the kth DG connection at time t, It is the node voltage connected to the lth SOPs port at time t.

[0105] After obtaining the kth island recovery matrix IRM DG,k After that, the island recovery matrix IRM of the entire system at a certain time t can be obtained DG :

[0106]

[0107] After obtaining the matrix PTM of the lth SOPs port SOPs,l After that, the SOPs transfer matrix PTM of the whole system at a certain time t can be obtained SOPs :

[0108]

[0109] The constraints (16) and (27) are convexly relaxed, that is:

[0110]

[0111] Through conversion, the model solution is transformed into a standard second-order cone programming problem, which can be efficiently solved using mathematical optimization tools.

[0112] by Figure 2 Taking the distribution network as an example, the IRM after model modification is explained. DG When branch ⑥ fails, the section switch isolates the fault and load nodes 6-8 are affected by the fault and lose power. Considering the capacity limitation of the DG connected to the distribution network, it is impossible to restore the power supply of nodes 6-8. The island range must be optimized. The application model is used to optimize the IRM DG,uncon For example, considering the load amount and importance, load node 8 is cut, and load nodes 6 and 7 are restored by the island. The IRM before and after the correction is DG like Figure 3 shown.

[0113] Similarly, Figure 2 Taking the distribution network as an example, the PTM after model correction is explained. SOPs When branch ③ fails, the sectionalizer isolates the fault and load nodes 3-8 are affected by the fault and lose power. Considering the capacity limitation of the SOP port, it is impossible to restore power to all nodes 3-8. The optimization model is applied to correct the PTM. SOPs,uncon For example, considering the load and importance, load nodes 5 and 8 are reduced, and load nodes 3, 4, 6, and 7 are restored by SOP. SOPs like Figure 4 shown.

[0114] Step 4: Based on the fault impact correlation matrix established in step 1 and the evaluation model in step 3, perform explicit analytical calculation of the reliability index of the distribution system taking into account the access of DG and SOPs.

[0115] This step conducts a reliability assessment on a 137-node example system with the characteristics of a new distribution system. Through algebraic and logical operations between the above-constructed matrices and vectors, the impact of DGs and SOPs on the power supply reliability of the distribution network is quantified, aiming to verify the effectiveness of the model and method proposed in this invention.

[0116] The three fault impact correlation matrices for the entire network, FEIM-A, FEIM-B, and FEIM-C, are constructed by combining the FEIM-A, FEIM-B, and FEIM-C of each feeder using a diagonal block matrix. Therefore, the FEIM-C for the entire network can be expressed as:

[0117] FEIM-C =diag(FEIM-C 1 ,FEIM-C 2 ,···,FEIM-C n ) (36)

[0118] Where, FEIM-C i is the impact correlation matrix of the type C fault on the i-th feeder. For the impact of type C fault, the de-energized load is restored to power by the backup power source through the tie line. Based on the different feeder transfer methods, it can be divided into the following four cases.

[0119] If there are p feeders in the distribution network with multiple types of connections, that is, the power can be restored by the backup power source through SOPs and tie switches, then the transfer matrix of this type is defined as FEIM-C Ⅰ , then in the matrix, only the diagonal block matrices corresponding to the p feeders are non-zero matrices, and the diagonal block matrices corresponding to the remaining np feeders are zero matrices.

[0120] If the distribution network has q feeders that can only be restored through SOPs, then the transfer matrix is ​​defined as FEIM-C Ⅱ In this matrix, only the diagonal block matrices corresponding to the q feeders are non-zero matrices, and the diagonal block matrices corresponding to the remaining nq feeders are zero matrices.

[0121] If the distribution network has d feeders that can only be restored through the tie switch, then the transfer matrix is ​​defined as FEIM-C Ⅲ In this matrix, only the diagonal block matrices corresponding to the d feeders are non-zero matrices, and the diagonal block matrices corresponding to the remaining nd feeders are zero matrices.

[0122] If there are e feeders in the distribution network that are single radial lines and cannot be restored by power transfer, the power transfer matrix is ​​FEIM-C Ⅳ , obviously, this matrix is ​​an N-dimensional zero square matrix.

[0123] Based on the above analysis, n=p+q+d+e, and FEIM-C can be represented by the superposition of the above four types of matrices.

[0124] FEIM-C=FEIM-C Ⅰ +FEIM-C Ⅱ +FEIM-C Ⅲ +FEIM-C Ⅳ (37)

[0125] For PTM SOPs If the feeder connected to the SOPs can only be restored through the SOPs, then the transfer matrix is ​​defined as If the feeders connected to the SOPs can be restored through both the tie switch and the SOPs, the transfer matrix is ​​defined as Obviously, SOPs transfer matrix PTM SOPs , which can be expressed as:

[0126]

[0127] FEIM-A, FEIM-B, FEIM-C, IRM constructed according to the present invention DG and PTM SOPs , the reliability index of the complex distribution system after the DG and SOPs are connected can be explicitly calculated. The calculation method of the reliability index at the load level is shown in the following formula:

[0128] λ LP =λ×(FEIM-A+FEIM-B+FEIM-C) (39)

[0129]

[0130] Where, t island The switching time used to form an island, t sop In the absence of DG, part of the load would have to wait until the fault was repaired before the main power supply could resume power. After DG was connected, this part of the load could be restored to power by the DG island first, so the power outage time was shortened from the original fault repair time to t island Considering the limitation of DG capacity, the load that has been cut needs to wait until the fault is repaired before the main power supply can resume power supply. Included in the plan.

[0131] After connecting to SOPs, the power-off load can be transferred through SOPs, and the load outage time is shortened from the original contact switch operation time to t sop Considering the limitation of SOPs port capacity, for feeders that can only be restored through SOPs, part of the power-lost load on the feeder can be restored by SOPs, and the remaining loads that have been cut need to wait until the fault is repaired before they can be restored by the main power supply. Included in the plan.

[0132] For feeders with multiple types of interconnections, some of the power-lost loads on the feeder can be restored by SOPs first, and the remaining loads that have been cut need to wait until the interconnection switch is closed to restore power. The impact of the fault is Included in the plan.

[0133] The reliability index of the entire system can be obtained by summing up the reliability index of all load nodes. The analytical calculation formula for the system reliability index is as follows.

[0134]

[0135] EENS=μ LP ×L T 44)

[0136] Where N T The column vector representing the number of users at each load node arranged in ascending order, N total Indicates the total number of users in the distribution network.

[0137] According to the above-mentioned analytical evaluation method of distribution system reliability index taking into account DG and SOPs, the reliability evaluation is carried out by taking a 137-node example. The structure of the 137-node example is as follows: Figure 5 As shown in the figure, the calculation example includes two substation buses numbered 201 and 202, and 135 load nodes. A circuit breaker is installed at the outlet of the substation bus, and a section switch is installed at both ends of each branch. The section switch action time t is set. sw is 0.5 hours, the contact switch action time t op The switching operation time t used to form an island is 0.5 hours. island is 0.5 hours, SOPs action time t sop 0.05 hours.

[0138] The branch failure rate and fault repair time, the load demand of each node and the number of users are shown in Tables 1 and 2.

[0139] Table 1137 Node system component failure rate and failure repair time

[0140]

[0141]

[0142]

[0143] Table 2137 Node load requirements and number of users for each node in the node calculation example

[0144]

[0145]

[0146] This paper sets up three comparison scenarios. Scenario 1: No tie feeders exist, the entire network is radial, and no DGs or SOPs are connected; Scenario 2: Tie lines exist between different feeders, and no DGs or SOPs are connected; Scenario 3: Multiple types of ties exist between different feeders, and DGs and SOPs are considered. The tie switch, DG, and SOP configurations for each scenario are shown in Table 3, where an "○" indicates that the component is not considered for connection.

[0147] Table 3 Configuration of tie switches, DGs and SOPs in each scenario

[0148]

[0149]

[0150] The system reliability indicators under each scenario are shown in Table 4, and the number of power outages and power outages of each load node are shown in Table 4. Figure 6 and Figure 7 shown.

[0151] Table 4137 Node system reliability indicators

[0152]

[0153] Depend on Figure 6 It can be seen that the number of power outages at the load nodes on each feeder is exactly the same in the three scenarios. According to formula (39), the number of power outages at a load node is only related to the sum of the product of the branch failure rate and the number of faults caused by that branch. When any branch on a feeder fails, the circuit breaker at the feeder outlet instantly disconnects, causing power outages at all load nodes on that feeder. Therefore, the number of power outages at the load nodes on each feeder is exactly the same. In addition, if the positions of the circuit breakers and section switches remain unchanged, whether the distribution network is connected to a tie switch, DG, and SOP only affect the power outage time and power shortage of the load nodes and do not reduce the number of faults. Therefore, the number of power outages at the load nodes in each scenario is exactly the same.

[0154] From Table 4 and Figure 7It can be seen that in Scenario 2, by configuring the tie switch, the power outage time of some load nodes is shifted from fault repair time to tie switch operation time, significantly reducing the power outage time. Compared with Scenario 1, SAIDI is reduced by about 32.6% and EENS is reduced by about 33.9%.

[0155] After the DG is connected, some nodes that were previously affected by Class A faults and needed to wait for fault repair can now have their power restored by islanding. The power outage time shifts from the fault repair time to the islanding time. Considering that the scope of islanding recovery is closely related to the size of the DG and the load demand of each node, the degree of reduction in power outage time varies from node to node. When the SOP replaces the traditional tie switch, some nodes that were previously affected by Class C faults and needed to wait for the tie switch to close before power can now have their power restored by SOP. Therefore, the power outage time shifts from tie switch operation time to SOP operation time, significantly reducing the power outage time. However, for load nodes (105-117), when an upstream branch fault occurs, power cannot be restored by SOP due to SOP capacity limitations. The power outage time increases to the fault repair time, resulting in a power outage time between scenarios 1 and 2. In summary, by configuring DG and SOP in Scenario 3, the average power outage time and power shortage of the system are further reduced compared to Scenario 2. The SAIDI is reduced by about 19.7%, and the EENS is reduced by about 12.5%, significantly improving the reliability of the distribution network.

[0156] It should be emphasized that the embodiments described in the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific embodiments. Any other embodiments derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.

Claims

1. A distribution system reliability index analytical evaluation method taking into account DG and SOPs is characterized by: The following steps are involved: Step 1: Determine the impact of distribution network branch faults on load nodes and establish a fault impact correlation matrix; Step 2: Establish the DG island division and SOPs transfer model, and solve the island recovery matrix and SOPs transfer matrix after the fault; Step 3: Establish a reliability analysis and evaluation model based on the fault impact correlation matrix; Step 4: Based on the fault impact correlation matrix established in step 1 and the evaluation model in step 3, perform explicit analytical calculation of the reliability index of the distribution system taking into account the access of DG and SOPs.

2. The distribution system reliability index analytical evaluation method taking into account DG and SOPs according to claim 1 is characterized by: The specific implementation method of step 1 is: the impact of branch failure on load nodes is divided into four types: Impact type a: A branch fault disconnects all power supply paths to the load, and power can only be restored after the fault is repaired. Impact type b: A branch fault causes all power supply paths to the load to be disconnected. After the fault is isolated, the load is restored to the main power supply; Impact type c: A branch fault causes all power supply paths to the load to be disconnected. After the fault is isolated, the load can be transferred to the backup power source through the tie line to restore power supply; Impact type d: Branch failure has no impact on the load node; According to the four types of impacts of branch faults on load nodes, the fault impact matrix FEIM is constructed, where FEIM includes FEIM-A, FEIM-B, FEIM-C and FEIM-D. According to the fault location, the impact type of the fault on the load node and the load transfer path are analyzed in detail, and the power supply path matrix PSPM, segmented switch matrix SSM and connection matrix TSM are used to represent them: In FEIM, FEIM-A, FEIM-B, FEIM-C and FEIM-D are represented as follows: Among them, the matrix ONE represents an N-dimensional square matrix whose matrix elements are all 1.

3. The distribution system reliability index analytical evaluation method taking into account DG and SOPs according to claim 1 is characterized by: The specific implementation method of step 2 is: constructing the DG power supply path matrix DGPSPM: Building an Island Recovery Matrix (IRM) DG , indicating the range of power supply restoration in isolated islands after branch failure at different locations, IRM DG It is an N-dimensional square matrix, the row number is the branch number, the column number is the load node number, when the matrix element When branch i fails, node j is powered by the island. Then, when branch i fails, node j cannot be restored to power by the island; When DG has no capacity constraint, the island recovery matrix IRM DG,uncon The calculation method is: MRI DG,uncon =FEIM-A-FEIM-A∩DGPSPM The operator "∩" represents the bitwise AND operation of matrix elements, and also means that if the element values ​​at the same position in the two matrices are both "1", then the element value at the corresponding position in the new matrix is ​​"1", otherwise it is "0". Constructing SOPs transfer matrix PTM SOPs , indicating the scope of power supply restoration by SOPs when branch failure occurs at different locations, PTM SOPs It is an N-dimensional square matrix, the row number is the branch number, the column number is the load node number, when the matrix element When branch i fails, node j is restored to power supply by SOPs; when the matrix element Then, when branch i fails, node j cannot be restored to power supply by SOPs; Under the condition of no capacity constraint of SOPs, PTM SOPs,uncon The calculation method is: PTM SOPs,uncon =FEIM-C 4. The distribution system reliability index analytical evaluation method taking into account DG and SOPs according to claim 1 is characterized by: The specific implementation method of step 3 is: there are N DGs in the power distribution system. k There are N SOPs ports in total. l Then the island recovery matrix formed by the kth DG is IRM DG,k , then the transfer matrix of the lth SOPs port is PTM SOPs,l , construct the DG island division and SOPs transfer optimization model at a certain time t, and the objective function of the model is: in, represents the element in row i and column j of the kth island recovery matrix after correction, represents the element in row i and column j of the modified l-th SOPs port transfer matrix, ω j represents the importance of load node j, It represents the active power demand of load node j at the current time t.

5. The distribution system reliability index analytical evaluation method taking into account DG and SOPs according to claim 1 is characterized by: The power transfer optimization model also includes constraints, which include DG capacity constraints, SOPs capacity constraints, power supply range constraints, network connectivity constraints, load weight constraints, system flow constraints and system operation constraints.

Images