A Multi-Level Autonomous Operation and Control Method for Medium and Low Voltage Power Distribution Systems

Abstract

This invention discloses a multi-level autonomous operation and control method for medium- and low-voltage distribution systems. In the day-ahead phase, each distribution substation establishes a substation adjustable potential assessment model based on the operational characteristics of its adjustable resources, calculates the interactive power range for the next day, and reports it to the distribution network dispatch center. The distribution network then establishes a corresponding day-ahead safe operation range planning model for each substation based on the interactive power range, calculates the maximum allowable fluctuation range of interactive power for each substation, and transmits this value to the substation to provide interactive power boundaries for its autonomous operation. In the intraday phase, based on the safe operation range, the substations conduct autonomous rolling dispatch with the goal of optimal substation economics, and transmit the interactive power curves obtained from dispatching at each time period to the distribution network. Then, the distribution network formulates corresponding dispatch strategies based on the interactive power information of each substation, with the goal of optimal economics, to achieve efficient autonomous operation of the entire medium- and low-voltage distribution system.

Description

Technical Field

[0001] This invention relates to the field of power distribution system optimization scheduling, and in particular to a multi-level autonomous operation and control method for medium and low voltage power distribution systems. Background Technology

[0002] With the continuous grid connection and consumption of large-scale distributed renewable energy on the distribution side, the current distribution network has evolved from the traditional "passive" distribution network into a new form of "active" distribution network. However, due to the intermittency and volatility of renewable energy power generation, the source-load mismatch phenomenon of active distribution networks is becoming increasingly prominent, and its safe and stable operation faces severe challenges.

[0003] The intermittent renewable energy sources connected in a distributed manner exacerbate the uncertainty of active distribution network operation from the source side. Simultaneously, the increasing demand for flexible loads, such as electric vehicles and air conditioners, adds to the uncertainty on the load side. To address the source-load uncertainty and efficiently absorb widely dispersed renewable energy, the best approach is to consider inter-regional complementarity and mutual support, achieving localized regional autonomous operation of sources and loads through the flexible allocation of distributed resources. The widespread integration of multi-source-storage-load devices into the distribution network, coupled with increasing user autonomy and participation on the demand side, complicates and diversifies the regulation characteristics of massive controllable resources. Distribution networks are gradually exhibiting new characteristics such as active operation, intelligent source and load management, and network power electronics, significantly increasing the complexity and control difficulty of the distribution system and posing a significant challenge to its operation and regulation.

[0004] To address the pressures of massive data on the operation of traditional centralized distribution networks, including latency, data loss, and security risks, edge computing, as a novel computing model, deploys edge servers at the edge of the distribution network. This allows data to be directly transmitted to the vicinity of the data source for analysis and processing, effectively alleviating the burden on network communication and computing resources in distribution network operation and control, and significantly improving the independent operation and scheduling capabilities of distribution substations. Under these circumstances, the distributed and flattened characteristics of my country's active distribution network operation and management are gradually emerging, forming a network structure of regional autonomy and cloud-edge collaboration. Faced with the aforementioned challenges and changes in the form of active distribution networks, how to build and develop a highly reliable, flexible, and user-friendly active distribution network management and control system has become an urgent problem to be solved. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a multi-level autonomous operation and control method for medium and low voltage power distribution systems. This method aims to fully leverage the autonomous optimization capabilities of distribution transformers in medium and low voltage power distribution systems, achieve multi-level flat scheduling of medium and low voltage power distribution systems, promote the full exploitation of adjustable resources in distribution transformers and power distribution networks, and facilitate the efficient local consumption of distributed energy, thereby contributing to the safe and efficient operation of medium and low voltage power distribution systems.

[0006] Technical solution: The present invention provides a multi-level autonomous operation and control method for medium and low voltage power distribution systems, comprising the following steps:

[0007] (1) Based on the operational characteristics of adjustable resources in each transformer area, establish a current transformer area regulation potential assessment model and solve for the interaction power range of each transformer area for the next day; the transformer area regulation potential assessment model is a multi-objective programming model for interaction power and energy upper and lower limits at the grid connection point of the transformer area.

[0008] (2) Based on the interactive power range of each transformer area for the next day, establish an optimization model for the safe operation range of the transformer area before the day, and solve for the maximum safe operation range of the transformer area interactive power to maintain the safe and stable operation of the distribution network. The optimization model for the safe operation range of the transformer area before the day is a three-layer robust optimization model. The first layer is used to obtain the upper and lower boundary values ​​of the maximum safe operation range of the transformer area power. The second layer obtains the worst transformer area interactive power trajectory for the safe operation of the distribution network from the safe operation range of the power. The third layer verifies whether the distribution network can achieve safe and stable operation under the interactive power trajectory.

[0009] (3) Based on the safe operating range of the power distribution area, with the goal of minimizing the comprehensive operating cost of the distribution area during the day, solve the optimal economic dispatch strategy for each time period during the day, and transmit the interactive power value of each time period to the distribution network.

[0010] (4) Based on the interactive power values ​​of each time period, the distribution network calculates the optimal intraday autonomous dispatch strategy with the goal of minimizing the overall cost of the distribution network.

[0011] Furthermore, the operational characteristics of the adjustable resources mentioned in step (1) include energy storage operational characteristics and transferable load operational characteristics;

[0012] Energy storage operating characteristics include:

[0013]

[0014] E es,0 =E es,24

[0015] Where Δt is the unit time interval. E represents the maximum charge / discharge power per unit time of energy storage. es,t and and These are the capacity and charging / discharging power of the energy storage system at time t, respectively. and These refer to the charging efficiency and discharging efficiency of the energy storage system, respectively. and These are 0-1 variables representing the charging and discharging state of energy storage at time t, respectively. When the energy storage is in the charging state...

[0016] Transferable load operating characteristics include:

[0017]

[0018]

[0019] in, and These represent the increase or decrease in load in the distribution area at time t due to the scheduling needs of the converged terminal. and These represent the maximum load that the transformer area can increase or decrease at time t.

[0020] Furthermore, the method for establishing the pre-existing regional regulation potential assessment model described in step (1) is as follows:

[0021] (a) Establish an approximate descriptive model of the regulation potential of the transformer area;

[0022]

[0023] Among them, P DSS It is a vector composed of the interactive power at the grid connection point of the transformer area at different time periods; The interactive power value at the grid connection point of the transformer area at time t; Let be the energy value of the transformer area at time t; and as well as and These represent the allowable interactive power and the upper and lower limits of energy at time t at the grid connection point of the transformer substation;

[0024] (b) Establish the aforementioned daily front-area regulation potential assessment model;

[0025]

[0026] The current local area regulation potential assessment model is a multi-objective planning model, and its constraints include the regulation potential approximate description model Ω2, the operation constraints of each adjustable resource in the local area, and the power balance constraints of the local area.

[0027] Furthermore, the daily front-area safe operation zone optimization model mentioned in step (2) is as follows:

[0028]

[0029] Among them, Ω n A set of distribution network nodes. and The upper and lower limits of the safe operating range of the power distribution area at node i, respectively. and as well as and These are the slack variables for active and reactive power flow in the distribution network;

[0030] The constraints of the current safe operation range optimization model for distribution substations include the regulation potential constraints uploaded by each substation, the operation constraints of each equipment in the distribution network, and the power balance constraints of the distribution network.

[0031] Further, in step (3), based on the safe operating range of the transformer area's power, and with the objective of minimizing the transformer area's overall daily operating cost, the optimal economic dispatch strategy for each time period during the day is calculated, including:

[0032] During the intraday phase, an intraday economic self-regulation scheduling model for the distribution area is established with the goal of minimizing the overall intraday operating cost of the distribution area:

[0033]

[0034] in, C represents the daily integrated operating cost of the transformer area located at node i. ess C represents the operating cost of energy storage equipment during the day in the distribution area. DR For the scheduling cost of transferable load, C grid The cost of purchasing and selling electricity for the distribution area and the power grid;

[0035] Solving the intraday economic autonomous scheduling model of the aforementioned transformer area yields the optimal economic scheduling strategy for each time period within the day.

[0036] Furthermore, in step (4), the method for calculating the optimal intraday autonomous dispatch strategy with the goal of minimizing the overall cost of the distribution network is as follows:

[0037] To minimize the overall cost of the distribution network, an intraday economic autonomous dispatch model for the distribution network is established:

[0038]

[0039] Among them, Ω dis Ω is the collection of distribution network branches. i C is the set of distribution network nodes. loss,ij For the intraday network loss of the ij branch of the distribution network, C ess,i This represents the intraday operating cost of the medium-voltage energy storage located at node i, C. grid,dis This represents the power purchase and sale value between the distribution network and the upstream power grid during the daily period.

[0040] Solving the intraday economic autonomous dispatch model of the distribution network yields the optimal intraday autonomous dispatch strategy for the distribution network.

[0041] Further, in step (1), the boundary node shrinkage algorithm is used to solve the current front-area regulation potential assessment model.

[0042] Further, in step (2), an iterative algorithm based on variable range interval constraints is used to solve the daily front-area safe operation interval optimization model; the iterative algorithm based on variable range interval constraints includes:

[0043] By utilizing decision variables, the robust optimization problem with a variable range interval is transformed into a robust optimization problem with a fixed range interval. Then, an iterative method based on the C&CG concept is used to solve the robust optimization problem with a fixed range interval.

[0044] Furthermore, the transformation of the robust optimization problem with a variable range interval into a robust optimization problem with a fixed range interval using decision variables includes:

[0045] For those within the transformer area range Possible interaction power values ​​within By introducing the decision variable ξ i,t The following constraints ensure the boundary is fixed:

[0046]

[0047] Through the above constraints, the variable range constraints in the second layer of the original robust optimization problem are resolved. Convert to a fixed range constraint ξ i,t ∈[0,1]; where, ξ i,t It is the ratio of the worst-case interactive power value of the transformer area to its maximum safe operating range, that is, the worst-case interactive power trajectory of the transformer area.

[0048] Furthermore, the iterative method based on the C&CG concept for solving the robust optimization problem within a given range includes:

[0049] (a) The robust optimization problem within a given range is split into a main problem and a subproblem:

[0050] The main question is:

[0051]

[0052] In the formula, x is the first-level decision variable. and A set; Let ξ be the decision variable i,t The set of values ​​after the l-th iteration is used to characterize the worst-case interactive power output trajectory of the substation area within the operating range determined in the first layer; y is the power flow relaxation variable. The set of power distribution network operation decision variables, including energy storage charging and discharging power, and c, b, A, B, E, F, H, J, L are parameter matrices;

[0053] The subproblems are:

[0054]

[0055] In the formula, b is the parameter matrix of the second-level objective function that constitutes the robust optimization problem with a variable range; this subproblem is a max-min two-level optimization problem, which is transformed into a single-level max problem for solution through the cone duality transformation strategy;

[0056] (b) Based on the transformed main problem and subproblems, the following iterative strategy is constructed for solving the problem:

[0057] 1) Given a set of initial uncertain variable ξ values ​​as the initial scenario, set the number of iterations k = 1, the threshold judgment condition P = +∞ and the algorithm threshold condition δ = 0;

[0058] 2) Calculate the main problem based on the initial scenario ξ to obtain the optimal solution.

[0059] 3) The optimal solution Substituting these values ​​into the subproblem, we obtain the objective function of the subproblem. And the corresponding worst-case scenario's interactive power trajectory of the transformer substation. Update threshold judgment conditions

[0060] 4) Determine if P≤δ holds true. If it does, it means that even under the worst-case operating scenario for the distribution network, the safe and stable operation of the distribution network can still be guaranteed. Within the defined operating range, there are no power flow exceedances; the iteration terminates, and the maximum power safe operating range value for the transformer area is returned. If the conditions are not met, add variables to the main problem. y k+1 and the following constraints:

[0061]

[0062] Let k = k + 1 and jump to step b) until the algorithm converges.

[0063] Beneficial effects: Compared with the prior art, the advantages of the present invention are: (1) The present invention proposes a multi-level autonomous operation and control method for medium and low voltage power distribution systems. In this strategy, a regulation potential assessment model for each distribution area is constructed on the basis of the distribution area. Since the model does not need to consider the interaction with other distribution areas, the distribution network does not have a synchronization requirement for uploading the potential assessment model of each distribution area, thereby greatly alleviating the communication pressure during distribution network scheduling; (2) The present invention introduces the optimization of the safe operation range of the distribution area during the scheduling process. Since this range is obtained based on the adjustable potential characteristics reported by the distribution area and comprehensively considering the safety operation constraints of the distribution network, from the perspective of the distribution network, this operation range reserves a certain adjustment margin for the interaction power between the distribution area and the distribution network scheduling; from the perspective of the distribution area, by complying with this interaction power range, the distribution area does not need to consider the impact of the uncertainty of its interaction with the distribution network power on the safety operation of the distribution network, thereby realizing its own economic autonomous operation. It effectively promotes the safe and efficient operation of medium and low voltage power distribution systems. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of the multi-level autonomous operation and control method for medium and low voltage power distribution systems according to the present invention;

[0065] Figure 2 This is a topology diagram of a medium- and low-voltage power distribution system according to an embodiment of the present invention;

[0066] Figure 3 This is a diagram showing the safe operation range of the four transformer substations obtained from the previous stage optimization in this embodiment of the invention. Detailed Implementation

[0067] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0068] like Figure 1 As shown, the multi-level autonomous operation and control method for medium and low voltage power distribution systems includes the following steps.

[0069] (1) Establish a model for assessing the adjustment potential of each transformer area based on the operational characteristics of the adjustable resources within each transformer area, and solve for the interactive power range of each transformer area for the next day.

[0070] The operational characteristics of adjustable resources within the distribution area mainly include the operational characteristics of energy storage and the operational characteristics of transferable loads, as detailed below:

[0071] (a) Energy storage operation characteristics

[0072]

[0073] E es,0 =E es,24

[0074] Where Δt is the unit time interval. E represents the maximum charge / discharge power per unit time of energy storage. es,t and and These are the capacity and charging / discharging power of the energy storage system at time t, respectively. and These refer to the charging efficiency and discharging efficiency of the energy storage system, respectively. and These are 0-1 variables representing the charging and discharging state of energy storage at time t, respectively. When the energy storage is in the charging state...

[0075] (b) Transferable load operation characteristics

[0076]

[0077] in, and These represent the increase or decrease in load in the distribution area at time t due to the scheduling needs of the converged terminal. and These represent the maximum load that the transformer area can increase or decrease at time t.

[0078] The method for establishing a local area adjustment potential assessment model is as follows:

[0079] (a) Establishing an approximate descriptive model of the regulation potential of the transformer area

[0080] The regulation potential Ω2 of a transformer substation can be viewed as a mapping of the power operating domain of each adjustable resource within the substation at each moment to the grid connection point, based on satisfying the substation network constraints and adjustable resource operation constraints. Furthermore, considering that the operating characteristics of the adjustable resources within the substation can be represented by power and energy constraints, a storage-like model can be established to approximate the regulation potential of the substation at the grid connection point at each moment:

[0081]

[0082] Among them, P DSS It is a vector composed of the interactive power at the grid connection point of the transformer area at different time periods; The interactive power value at the grid connection point of the transformer area at time t; Let be the energy value of the transformer area at time t; and as well as and These represent the upper and lower limits of the allowed interactive power and energy at time t at the grid connection point of the transformer substation.

[0083] (b) Establish the aforementioned assessment model for the regulatory potential of the front area.

[0084]

[0085] The aforementioned local transformer area regulation potential assessment model is a multi-objective planning model. Its constraints, in addition to the regulation potential approximation model Ω2, should also include operational constraints on each adjustable resource within the transformer area and power balance constraints within the transformer area. The objective function is... and Together, these constitute the boundary of the transformer area's regulation potential considering time coupling. By solving this model, the maximum interactive power and energy range that the transformer area may reach at each time point can be obtained, thereby enabling a description of the transformer area's adjustable potential.

[0086] (2) Based on the collected interaction power range of each distribution area, and based on the safety operation constraints of the distribution network, establish an optimization model for the safe operation range of the current distribution area, and solve for the maximum safe operation range of the distribution area that can maintain the safe and stable operation of the distribution network.

[0087] The current optimization model for the safe operation zone of the front area is as follows:

[0088]

[0089] The daily front-area safe operation zone optimization model is a three-stage robust optimization model, where Ω n A set of distribution network nodes. and These are the upper and lower limits of the safe operating range of the transformer substation power at node i, obtained from the distribution network optimization. and as well as and These are the slack variables for active and reactive power flow in the distribution network.

[0090] The constraints mainly include the regulation potential constraints of each distribution area, the operation constraints of energy storage, photovoltaic inverters and other equipment in the distribution network, and the power balance constraints of the distribution network.

[0091] In this three-layer robust optimization model, the minimization problem in the first layer aims to obtain the safe operating range of the maximum possible transformer area interaction power by solving a mixed-integer linear programming problem, which determines the uncertainty set of transformer area interaction power fluctuations; the maximization problem in the second layer aims to obtain the transformer area interaction power value that will cause the worst operating condition of the distribution network from the interaction power uncertainty set obtained in the first layer; and the last layer of the model aims to solve an optimal power flow problem to verify whether the distribution network can operate safely and stably under the worst transformer area interaction power scenario.

[0092] The obtained safe operating range of the distribution area will be provided to each distribution area as a reference for its daily operation. As long as the power of its grid connection point is kept within the range, the distribution area does not need to consider the impact of its operation and scheduling on the safety of the distribution network, thereby effectively improving the efficiency and economy of the distribution area's autonomy.

[0093] (3) The safe operation range of the distribution area transmitted by the base distribution network, each distribution area constructs a corresponding intraday economic autonomous dispatch model with the goal of optimal economy, solves the optimal economic dispatch strategy for each time period of the day, and transmits the interactive power value of each time period to the distribution network.

[0094] During the intraday phase, each distribution area will operate autonomously based on the safe operating range issued by the distribution network prior to the day, with the goal of optimal economic efficiency. The model is as follows:

[0095]

[0096] in, C represents the daily integrated operating cost of the transformer area located at node i. ess C represents the operating cost of energy storage equipment during the day in the distribution area. DR For the scheduling cost of transferable load, C grid This represents the cost of purchasing and selling electricity in the distribution area and network. The specific calculation formulas for each cost are shown below:

[0097]

[0098] Where τ is the scheduling time interval, T is the total time step of the intraday scheduling phase, and ρ ess , ρ FL , ρ buy and ρ sell These represent the unit power cost of energy storage, transferable load, and electricity purchase and sale during the intraday dispatch phase of the distribution area. and These represent the charging and discharging power values ​​during the daily energy storage phase. and These represent the increases and decreases in the load of the transformer area during time period t, respectively. and These represent the power purchased and sold by the transformer substation to the distribution network. In addition to the cost calculation constraints mentioned above, this optimization model should also include interactive power limit constraints at the grid connection point, operational constraints of adjustable resources such as energy storage, and power balance constraints for the transformer substation.

[0099] By solving this model, the autonomous operation strategy of the distribution area during the day can be obtained. Simultaneously, the optimized interactive power values ​​at the grid connection point for each time period need to be transmitted to the distribution network in real time to enable intraday autonomous dispatching, thereby achieving safe and economical operation of the entire system.

[0100] (4) Based on the collected intraday interactive power curves of each distribution area, the distribution network will formulate corresponding intraday autonomous dispatch strategies with the goal of optimal economic efficiency, thereby improving the economic efficiency of the entire medium and low voltage distribution system while ensuring the safe operation of the distribution network.

[0101] The intraday economic autonomous dispatch model for the distribution network is as follows:

[0102]

[0103] Among them, Ω dis Ω is the collection of distribution network branches. i C is the set of distribution network nodes. loss,ij For the intraday network loss of the ij branch of the distribution network, C ess,i This represents the intraday operating cost of the medium-voltage energy storage located at node i, C. grid,dis This represents the power purchase and sale values ​​between the distribution network and the upstream power grid during the daily period. By solving this model, the economic dispatch strategy with the optimal overall cost of the distribution network can be obtained under the condition of deterministic power exchange between distribution areas, thereby effectively improving the economic efficiency of the entire medium and low voltage distribution system.

[0104] In step (1), the boundary node shrinkage algorithm is used to solve the daily front-area regulation potential assessment model. In step (2), the iterative solution algorithm based on variable range interval constraints is used to solve the daily front-area safe operation interval optimization model. The models in steps (3) and (4) are mixed integer linear programming problems, which can be solved directly by calling commercial solvers.

[0105] The iterative solution algorithm based on variable range interval constraints in step (2) mainly includes the following two steps:

[0106] (A) Introduce new variables to fix the variable range interval, thereby transforming the original variable range interval robust optimization problem into a fixed range interval robust optimization problem;

[0107] (B) Referencing the C&CG concept, construct a corresponding iterative method to solve the transformed fixed-range interval robust optimization problem.

[0108] Furthermore, the method for fixing the variable range interval in step (1) of the above iterative solution algorithm based on variable range interval constraints is as follows:

[0109] For those within the transformer area range Possible interaction power values ​​within By introducing a new decision variable ξ i,t The following constraints are used to fix the boundary.

[0110]

[0111] Through the above constraints, the variable range constraints in the second layer of the original robust optimization problem are resolved. It can be converted into a fixed-range constraint ξ i,t ∈[0,1]. Where, ξ i,t It can be understood as the ratio of the worst-case interactive power value of the transformer area to its maximum safe operating range, that is, the worst possible interactive power trajectory of the transformer area.

[0112] Furthermore, the iterative solution method in step (B) of the above iterative solution algorithm based on variable range interval constraints is as follows:

[0113] (a) Decompose the transformed fixed-range robust optimization problem into a main problem and a subproblem.

[0114] The main question is:

[0115]

[0116] In the formula, x is the first-level decision variable. and A set; The decision variable ξ introduced in the above variable range constraint transformation step after the lth iteration i,t The set of values ​​is used to characterize the worst interactive power output trajectory of the substation area within the operating range determined in the first layer; y is the power flow relaxation variable. The set of power distribution network operation decision variables, including energy storage charging and discharging power, and c, b, A, B, E, F, H, J, L are parameter matrices.

[0117] It is worth noting that, unlike conventional robust optimization algorithms, in the main problem of this algorithm, the power flow relaxation variables in the model are directly set to 0. This ensures that, under the operating trajectory propagated by the subproblems, the obtained maximum interactive power operating range can maintain the safe and stable operation of the distribution network.

[0118] The subproblems are:

[0119]

[0120] In the formula, b is the parameter matrix that constitutes the second-level objective function of the original robust optimization problem. This subproblem is a typical max-min bilevel optimization problem, which can be transformed into a single-level max problem for solution using strategies such as cone duality transformation.

[0121] (b) Based on the transformed main problem and subproblems, the following iterative strategy is constructed for solving the problem:

[0122] 1) Given a set of initial uncertain variable ξ values ​​as the initial scenario, set the number of iterations k = 1, the threshold judgment condition P = +∞ and the algorithm threshold condition δ = 0;

[0123] 2) Calculate the above main problem based on the initial scenario ξ to obtain the optimal solution.

[0124] 3) The optimal solution Substituting these values ​​into the subproblem above, we can obtain the objective function of the subproblem. And the corresponding worst-case scenario's interactive power trajectory of the transformer substation. Update threshold judgment conditions

[0125] 4) Determine if P≤δ holds true. If it does, it means that even under the worst-case operating scenario for the distribution network, the safe and stable operation of the distribution network can still be guaranteed. Within the defined operating range, there are no power flow exceedances. The iteration terminates, returning the maximum power safe operating range value for the transformer area. If the conditions are not met, add variables to the main problem. y k+1 and the following constraints

[0126]

[0127] Let k = k + 1 and jump to step 2) until the algorithm converges.

[0128] Based on the above calculations, the maximum power safe operating range of each distribution transformer area that the distribution network can accept while ensuring its own safe operation can be obtained. This range will be transmitted to each distribution transformer area. In the subsequent dispatching phase, each distribution transformer area only needs to comply with the constraints of this power range to ensure the safe and stable operation of the distribution network, thereby greatly reducing the complexity of distribution transformer area operation decisions and effectively promoting the efficient and autonomous operation of the distribution transformer areas.

[0129] The method of the present invention will be illustrated below with specific implementation examples.

[0130] To verify the effectiveness of the method described herein, this embodiment employs the following... Figure 2 The medium- and low-voltage power distribution system shown was tested. The network has 33 nodes, 37 branches, and four transformer substations.

[0131] Figure 3The optimized power safety operating ranges for the four distribution substations under this embodiment are shown. Before 9:00 AM, due to the relatively low overall power generation and high load output of distributed generation within the distribution network, the power safety operating range of each substation tends to approach its lower limit to ensure the safe and stable operation of the distribution network, in order to avoid the problem of node voltage exceeding the lower limit caused by the increase in load demand at distribution network nodes. During the midday period, as the photovoltaic power in the distribution network reaches its peak value, the safe operating range of the substations will gradually approach its upper limit to cope with the overvoltage problem caused by the excess active power in the distribution network. Furthermore, after 5:00 PM, as the load in each distribution network and substation increases, the distribution network will again limit the power safety operating range of the substations to approach its lower limit to avoid the problem of node voltage exceeding the lower limit.

[0132] Furthermore, to verify the reliability of the proposed method, based on intraday forecast data, this invention randomly generated 1000 distributed power supply output and load scenarios between 21:00 and 22:00 using Monte Carlo sampling, with error ranges of ±5% and ±10%, respectively, to simulate real-time uncertainty.

[0133] To verify the performance of this invention in operation scheduling, the following two methods are compared. Regarding the setting of intraday source-load output scenarios, this invention, based on the predicted source-load output during the day-ahead period, randomly generates 500 sets of source-load output data for distribution substations and the distribution network using Monte Carlo sampling within a ±10% error range. This serves as the intraday operation scenario set for the distribution network and distribution substations. Then, 10% of the photovoltaic output scenarios are selected within each distribution network and distribution substation, and their output value is set to 50% of the original value between 12:00 and 13:00 to simulate photovoltaic ramp-up events such as cloud cover.

[0134] Method 1: Centralized economic dispatch strategy. This method aims to minimize the total operating cost of the medium and low voltage power distribution system and does not consider the impact of the regulation potential of the distribution area.

[0135] Method 2: Optimization scheduling strategy for medium and low voltage power distribution systems based on the distribution area resource aggregation model, wherein the interactive power curve of each low voltage distribution area is directly specified by the medium voltage distribution network based on the resource aggregation model reported by each distribution area and its own economic needs.

[0136] Table 1 shows the distribution network operation and scheduling results under the three methods.

[0137] Table 1. Simulation Operation and Scheduling Results of Medium Voltage Power Distribution System

[0138]

[0139] As shown in Table 1, by constructing a distribution area regulation potential assessment model, the distribution network dispatch center can effectively tap into the flexible and adjustable resources in the distribution areas without significantly increasing communication pressure, thereby improving the economic efficiency of medium and low voltage distribution system operation. Furthermore, comparing Method 2 and the method of this invention, it can be seen that Method 2 cannot effectively guarantee the safe and stable operation of the medium and low voltage system in some intraday testing scenarios. This is because the distribution area interaction power in Method 2 is directly specified by the distribution network, which greatly reduces the distribution area's ability to cope with its own intraday source-load fluctuations. In contrast, the method of this invention provides the distribution network with a safe operating range for the distribution areas. On the one hand, the distribution network can provide a certain interaction power regulation margin for the distribution areas to help them cope with their own source-load fluctuations and enhance the distribution network's power support capability for the distribution areas. On the other hand, by maintaining their own interaction power within the specified operating range, the distribution areas greatly reduce the possibility of distribution network power flow exceeding limits due to intraday interaction power, thereby effectively enhancing the safety of the entire medium and low voltage distribution system operation. This effectively improves the autonomous operation capability of the distribution network and distribution areas, as well as the efficiency of the entire medium and low voltage distribution system operation.

Claims

1. A multi-level autonomous operation and control method for medium and low voltage power distribution systems, characterized in that, Includes the following steps: (1) Based on the operational characteristics of adjustable resources in each transformer area, establish a current transformer area regulation potential assessment model and solve for the interaction power range of each transformer area for the next day; the transformer area regulation potential assessment model is a multi-objective programming model for interaction power and energy upper and lower limits at the grid connection point of the transformer area. (2) Based on the interaction power range of each transformer area for the next day, establish an optimization model for the safe operation range of the transformer area before the day, and solve for the maximum safe operation range of the transformer area interaction power to maintain the safe and stable operation of the distribution network. The current transformer substation safe operation range optimization model is a three-layer robust optimization model. The first layer is used to obtain the upper and lower boundary values ​​of the maximum transformer substation power safe operation range. The second layer obtains the worst transformer substation interactive power trajectory for the safe operation of the distribution network from the power safe operation range. The third layer verifies whether the distribution network can achieve safe and stable operation under the interactive power trajectory. (3) Based on the safe operating range of the power distribution area, with the goal of minimizing the comprehensive operating cost of the distribution area during the day, solve the optimal economic dispatch strategy for each time period during the day, and transmit the interactive power value of each time period to the distribution network. (4) Based on the interactive power values ​​of each time period, the distribution network calculates the optimal intraday autonomous dispatch strategy with the goal of minimizing the overall cost of the distribution network; The method for establishing the pre-existing regional adjustment potential assessment model mentioned in step (1) is as follows: (a) Establish an approximate descriptive model of the regulation potential of the transformer area; Among them, P DSS It is a vector composed of the interactive power at the grid connection point of the transformer area at different time periods; The interactive power value at the grid connection point of the transformer area at time t; Let be the energy value of the transformer area at time t; and as well as and These represent the allowable interactive power and the upper and lower limits of energy at time t at the grid connection point of the transformer substation; (b) Establish the aforementioned daily front-area regulation potential assessment model; The current local area regulation potential assessment model is a multi-objective planning model, and its constraints include the regulation potential approximate description model Ω2, the operation constraints of each adjustable resource in the local area, and the power balance constraints of the local area. The daily front-area safe operation zone optimization model mentioned in step (2) is as follows: Among them, Ω n A set of distribution network nodes. and The upper and lower limits of the safe operating range of the power distribution area at node i, respectively. and as well as and These are the slack variables for active and reactive power flow in the distribution network; The constraints of the current safe operation range optimization model for distribution areas include the regulation potential constraints uploaded by each distribution area, the operation constraints of each equipment in the distribution network, and the power balance constraints of the distribution network. In step (2), an iterative algorithm based on variable range interval constraints is used to solve the daily front-area safe operation interval optimization model; the iterative algorithm based on variable range interval constraints includes: By utilizing decision variables, the robust optimization problem with a variable range interval is transformed into a robust optimization problem with a fixed range interval. The iterative method based on the C&CG concept is then used to solve the robust optimization problem with a fixed range interval. The method of transforming a robust optimization problem with a variable range interval into a robust optimization problem with a fixed range interval using decision variables includes: For those within the transformer area range Possible interaction power values ​​within By introducing the decision variable ξ i,t The following constraints ensure the boundary is fixed: Through the above constraints, the variable range constraints in the second layer of the original robust optimization problem are... Convert to a fixed range constraint ξ i,t ∈[0,1]; where, ξ i,t It is the ratio of the worst-case interactive power value of the transformer area to its maximum safe operating range, that is, the worst-case interactive power trajectory of the transformer area.

2. The multi-level autonomous operation and control method for medium and low voltage power distribution systems according to claim 1, characterized in that, The operational characteristics of the adjustable resources mentioned in step (1) include energy storage operational characteristics and transferable load operational characteristics; Energy storage operating characteristics include: AND es,0 =And es,24 Where Δt is the unit time interval. E represents the maximum charge / discharge power per unit time of energy storage. es,t and and These are the capacity and charging / discharging power of the energy storage system at time t, respectively. and These refer to the charging efficiency and discharging efficiency of the energy storage system, respectively. and These are 0-1 variables representing the charging and discharging state of energy storage at time t, respectively. When the energy storage is in the charging state... Transferable load operating characteristics include: in, and These represent the increase or decrease in load in the distribution area at time t due to the scheduling needs of the converged terminal. and These represent the maximum load that the transformer area can increase or decrease at time t.

3. The multi-level autonomous operation and control method for medium and low voltage power distribution systems according to claim 1, characterized in that, In step (3), based on the safe operating range of the transformer area's power, and with the goal of minimizing the transformer area's overall daily operating cost, the optimal economic dispatch strategy for each time period during the day is calculated, including: During the intraday phase, an intraday economic self-regulation scheduling model for the distribution area is established based on the goal of minimizing the overall intraday operating cost of the distribution area: minf i DSS =C ess +C DR +C grid Among them, f i DSS C represents the daily integrated operating cost of the transformer area located at node i. ess C represents the operating cost of energy storage equipment during the day in the distribution area. DR For the scheduling cost of transferable load, C grid The cost of purchasing and selling electricity for the distribution area and the power grid; Solving the intraday economic autonomous scheduling model of the aforementioned transformer area yields the optimal economic scheduling strategy for each time period within the day.

4. The multi-level autonomous operation and control method for medium and low voltage power distribution systems according to claim 1, characterized in that, In step (4), the method for calculating the optimal intraday autonomous dispatch strategy with the goal of minimizing the overall cost of the distribution network is as follows: To minimize the overall cost of the distribution network, an intraday economic autonomous dispatch model for the distribution network is established: Among them, f i DSS Ω represents the intraday comprehensive operating cost of the transformer area located at node i. dis For the collection of distribution network branches, Ω i C is the set of distribution network nodes. loss,ij For the intraday network loss of the ij branch of the distribution network, C ess,i This represents the intraday operating cost of the medium-voltage energy storage located at node i, C. grid,dis This represents the power purchase and sale value between the distribution network and the upstream power grid during the daily period. Solving the intraday economic autonomous dispatch model of the distribution network yields the optimal intraday autonomous dispatch strategy for the distribution network.

5. The multi-level autonomous operation and control method for medium and low voltage power distribution systems according to claim 1, characterized in that, In step (1), the boundary node shrinkage algorithm is used to solve the current front-area regulation potential assessment model.

6. The multi-level autonomous operation and control method for medium and low voltage power distribution systems according to claim 1, characterized in that, The iterative method based on C&CG principles for solving robust optimization problems within a given range includes: (a) The robust optimization problem within a given range is split into a main problem and a subproblem: The main question is: In the formula, x is the first-level decision variable. and A set; Let ξ be the decision variable. i,t The set of values ​​after the l-th iteration is used to characterize the worst-case interactive power output trajectory of the substation area within the operating range determined in the first layer; y is the power flow relaxation variable. The set of power distribution network operation decision variables, including energy storage charging and discharging power, and c, b, A, B, E, F, H, J, L are parameter matrices; The subproblems are: In the formula, b is the parameter matrix of the second-level objective function that constitutes the robust optimization problem with a variable range; this subproblem is a max-min two-level optimization problem, which is transformed into a single-level max problem for solution through the cone duality transformation strategy; (b) Based on the transformed main problem and subproblems, the following iterative strategy is constructed for solving the problem: 1) Given a set of initial uncertain variable ξ values ​​as the initial scenario, set the number of iterations k = 1, the threshold judgment condition P = +∞ and the algorithm threshold condition δ = 0; 2) Calculate the main problem based on the initial scenario ξ to obtain the optimal solution. 3) The optimal solution Substituting these values ​​into the subproblem, we obtain the objective function of the subproblem. And the corresponding worst-case scenario's interactive power trajectory of the transformer substation. Update threshold judgment conditions 4) Determine if P≤δ holds true. If it does, it means that even under the worst-case operating scenario for the distribution network, the safe and stable operation of the distribution network can still be guaranteed. Within the defined operating range, there are no power flow exceedances; the iteration terminates, and the maximum power safe operating range value for the transformer area is returned. If the conditions are not met, add variables to the main problem. y k+1 and the following constraints: Let k = k + 1 and jump to step b) until the algorithm converges.

Images