CN-122026534-A - Power distribution network optimal scheduling method and system based on centralized and distributed architecture

CN122026534ACN 122026534 ACN122026534 ACN 122026534ACN-122026534-A

Abstract

The invention discloses a power distribution network optimal scheduling method and system based on a centralized and distributed architecture, and belongs to the technical field of power control. The method comprises the steps of constructing a day-ahead stage centralized scheduling objective function and a day-in stage distributed scheduling objective function of a power distribution network, forming a centralized distributed stage optimal scheduling model of the power distribution network, constructing day-ahead stage centralized scheduling constraint according to a next day load demand value and a renewable energy output long-term predicted value of the power distribution network, solving to obtain a day-ahead stage centralized optimal scheduling strategy, constructing day-in stage distributed scheduling constraint according to next day load actual measurement data and a renewable energy output short-term predicted value, solving to obtain a day-in stage distributed optimal scheduling strategy, and optimizing power distribution network scheduling according to the day-ahead stage centralized optimal scheduling strategy and the day-in stage distributed optimal scheduling strategy, so that efficiency and accuracy of power distribution network optimal scheduling are improved.

Inventors

WU XUGUANG
YE KEJIAN
LIN XIAODONG
HU YUNHAO
ZHENG YANG
ZHOU JINHUI
ZHOU TAIBIN
NING XINSHEN
TANG YAOJING
CAO WANGSHU
YE YIJUN
MA JU

Assignees

国网浙江省电力有限公司温州供电公司

Dates

Publication Date: 20260512
Application Date: 20260415

Claims (10)

1. The power distribution network optimal scheduling method based on the centralized and distributed architecture is characterized by comprising the following steps of: Acquiring operation condition data of a power distribution network, and constructing a day-ahead stage centralized scheduling objective function and a day-in stage distributed scheduling objective function of the power distribution network based on the operation condition data to form a centralized distributed stage optimized scheduling model of the power distribution network; acquiring a next-day load demand value and a renewable energy source output long-term predicted value of the power distribution network, and constructing a day-ahead stage centralized scheduling constraint by combining the operation condition data; based on the centralized scheduling constraint of the day-ahead stage, solving the centralized and distributed stage optimization scheduling model, and determining a structural optimization strategy and an on-load voltage regulating transformer operation strategy of the power distribution network; Acquiring actual measurement data of the next day load of the power distribution network and a short-term predicted value of the output of renewable energy sources, and constructing a daily stage distributed scheduling constraint by combining the operation condition data; Solving the centralized distributed phase optimization scheduling model based on the intra-day phase distributed scheduling constraint, determining a photovoltaic output scheduling strategy and an energy storage output scheduling strategy of the power distribution network, and further constructing an intra-day phase distributed optimal scheduling strategy; and carrying out optimal scheduling on the power distribution network based on the day-ahead stage centralized optimal scheduling strategy and the day-in stage distributed optimal scheduling strategy.
2. The method for optimizing and scheduling a power distribution network based on a centralized and distributed architecture according to claim 1, wherein the obtaining operation condition data of the power distribution network, and constructing a centralized and distributed scheduling objective function of a day-ahead stage and a distributed scheduling objective function of a day-in stage of the power distribution network based on the operation condition data, to form a centralized and distributed stage optimizing and scheduling model of the power distribution network, comprises: Acquiring operation condition data of a power distribution network, wherein the operation condition data comprise branch operation data, reconstruction sectional switch state data, distributed photovoltaic output power data and on-load voltage regulating transformer gear data; Constructing a centralized scheduling objective function of a day-ahead stage of the power distribution network based on the branch running data, the reconstructed sectional switch state data, the distributed photovoltaic output power data and the on-load voltage regulating transformer gear data; Performing cluster division on the power distribution network to obtain a plurality of clusters of the power distribution network, and constructing a daily phase distributed scheduling objective function of the power distribution network by combining the branch operation data and the distributed photovoltaic output power data; And forming a centralized and distributed stage optimization scheduling model of the power distribution network based on the day-ahead stage centralized scheduling objective function and the day-in stage distributed scheduling objective function of the power distribution network.
3. The optimized scheduling method for a power distribution network based on a centralized and distributed architecture according to claim 2, wherein the constructing a centralized scheduling objective function for a day-ahead stage of the power distribution network based on the branch operation data, the reconstructed segmented switch state data, the distributed photovoltaic output power data and the on-load step-down transformer gear data comprises: the branch operation data comprises branch current data and branch resistance data; Constructing network loss cost of the power distribution network based on the product of the branch current data and the branch resistance data; Determining reconstruction segment switch state transformation data based on the reconstruction segment switch state data, and constructing the reconstruction cost of the power distribution network based on the reconstruction segment switch state transformation data; determining a distributed photovoltaic output power prediction deviation value based on the distributed photovoltaic output power data, and constructing a abandoned renewable energy punishment cost based on the distributed photovoltaic output power prediction deviation value; Determining on-load tap changer gear change data based on the on-load tap changer gear data, and constructing on-load tap changer operation cost based on the on-load tap changer gear change data; based on a preset day-ahead cost weight coefficient set, carrying out weighted summation on the power distribution network loss cost, the power distribution network reconstruction cost, the abandoned renewable energy punishment cost and the on-load voltage regulating transformer operation cost to obtain the total operation cost of the power distribution network; and constructing a centralized scheduling objective function of the power distribution network in a day-ahead stage by taking the minimum total running cost of the power distribution network as a target.
4. The method for optimizing and scheduling the power distribution network based on the centralized and distributed architecture according to claim 3, wherein the step of performing cluster division on the power distribution network to obtain a plurality of clusters of the power distribution network and combining the branch operation data and the distributed photovoltaic output power data to construct a daily stage distributed scheduling objective function of the power distribution network comprises the following steps: performing cluster division on the power distribution network to obtain a plurality of clusters of the power distribution network; Based on the branch current data, the branch resistance data and the distributed photovoltaic output power data, cluster branch current data, cluster branch resistance data and cluster distributed photovoltaic output power data of each cluster are respectively obtained; Calculating cluster network loss cost of each cluster based on the product of cluster branch current data and cluster branch resistance data of each cluster; Determining a cluster distributed photovoltaic output power prediction bias value for each of the clusters based on cluster distributed photovoltaic output power data for each of the clusters; calculating cluster abandoned renewable energy punishment cost of each cluster based on cluster distributed photovoltaic output power prediction deviation values of each cluster; Based on a preset daily cost weight coefficient set, weighting and summing the cluster abandoned renewable energy punishment cost and the cluster network loss cost of each cluster to obtain the total cluster running cost of the power distribution network; and constructing a daily phase distributed scheduling objective function of the power distribution network by taking the minimum total running cost of the clusters of the power distribution network as a target.
5. The method for optimizing and scheduling a power distribution network based on a centralized and distributed architecture according to claim 2, wherein the step of obtaining the next-day load demand value and the long-term prediction value of renewable energy output of the power distribution network, and the step of constructing a centralized scheduling constraint in a day-ahead stage by combining the operation condition data, comprises the following steps: The operation condition data also comprises node voltage data, on-load voltage regulating transformer operation data, static reactive compensator operation data and energy storage system operation data; The method comprises the steps of obtaining a next-day load demand value and a renewable energy source output long-term predicted value of the power distribution network, wherein the next-day load demand value comprises a next-day load active demand and a next-day load reactive demand; Constructing a daily front branch operation constraint set of the power distribution network based on the branch operation data; determining a node voltage upper limit value and a node voltage lower limit value based on the node voltage data, and constructing a day-ahead node voltage operation constraint of the power distribution network based on the node voltage upper limit value and the node voltage lower limit value; Constructing on-load voltage regulating transformer operation constraints of the power distribution network based on the on-load voltage regulating transformer operation data; constructing a day-ahead distributed photovoltaic output constraint of the power distribution network based on the distributed photovoltaic output power data and the distributed photovoltaic active power long-term predicted value; Based on the static var compensator operation data, constructing a daily static var compensator operation constraint of the power distribution network; Based on the energy storage system operation data, constructing a day-ahead energy storage device operation constraint of the power distribution network; Inputting the branch operation data, the static var compensator operation data, the energy storage system operation data, the next-day load active demand and the next-day load reactive demand into a preset power flow calculation model, and constructing a day-ahead node power balance constraint of the power distribution network; And constructing a centralized scheduling constraint in a day-ahead stage of the power distribution network based on the day-ahead branch operation constraint set, the day-ahead node voltage operation constraint, the on-load voltage regulating transformer operation constraint, the day-ahead distributed photovoltaic output constraint, the day-ahead static reactive power compensator operation constraint, the day-ahead energy storage equipment operation constraint and the day-ahead node power balance constraint of the power distribution network.
6. The method for optimizing scheduling of a power distribution network based on a centralized and distributed architecture according to claim 5, wherein the constructing a set of daily front branch operation constraints of the power distribution network based on the branch operation data comprises: The branch operation data comprise branch opening and closing data, branch power data, branch current data and branch resistance data; Constructing a power distribution network reconstruction radiation communication composite constraint of the power distribution network based on the branch opening and closing data; Determining the maximum branch transmission capacity, the active power and the reactive power of the branch based on the branch power data, and constructing the daily branch transmission power constraint of the power distribution network based on the maximum branch transmission capacity, the active power and the reactive power of the branch; determining a branch current upper limit value and a branch current lower limit value based on the branch current data, and constructing a daily front branch transmission current constraint of the power distribution network based on the branch current upper limit value and the branch current lower limit value; Determining a branch resistance value and a branch impedance value based on the branch resistance data, and constructing a daily forward branch ohm law constraint of the power distribution network based on the branch resistance value and the branch impedance value; Constructing a daily second order cone relaxation constraint of the power distribution network based on the branch current data and the node voltage data; And constructing a daily branch operation constraint set of the power distribution network based on the power distribution network reconstruction radiation communication composite constraint, the daily branch transmission power constraint, the daily branch transmission current constraint, the daily branch ohm law constraint and the daily second order cone relaxation constraint of the power distribution network.
7. The method for optimizing scheduling of a power distribution network based on a centralized and distributed architecture according to claim 5, wherein the steps of obtaining the measured data of the next day load of the power distribution network and the short-term predicted value of the output of renewable energy sources, and constructing a daily stage distributed scheduling constraint by combining the operation condition data include: Acquiring actual measurement data of the next day load and short-term predicted values of the output of the renewable energy source, wherein the actual measurement data of the next day load comprises actual measurement data of active load of the next day and actual measurement data of reactive load of the next day; acquiring a daily scheduling period of the power distribution network; based on the node voltage upper limit value and the node voltage lower limit value, combining the intra-day scheduling period to construct intra-day node voltage operation constraint of the power distribution network; Constructing cluster coupling branch constraint of the power distribution network according to a preset alternate direction multiplier method and the intra-day scheduling period; Based on the branch operation data, constructing a daily branch operation constraint set of the power distribution network by combining the daily scheduling period; based on the short-term predicted value of the distributed photovoltaic active power and the distributed photovoltaic output power data, combining the intra-day scheduling period to construct intra-day distributed photovoltaic output constraint of the power distribution network; based on the operation data of the static var compensator, combining the daily scheduling period to construct the daily static var compensator operation constraint of the power distribution network; Based on the energy storage system operation data, constructing an intra-day energy storage device operation constraint of the power distribution network in combination with the intra-day scheduling period; Inputting the branch operation data, the static var compensator operation data, the energy storage system operation data, the next day load active actual measurement data and the next day load reactive actual measurement data into a preset power flow calculation model, and constructing a daily node power balance constraint of the power distribution network; And constructing the intra-day stage distributed scheduling constraint of the power distribution network based on the intra-day node voltage operation constraint, the cluster coupling branch constraint, the intra-day branch operation constraint set, the intra-day distributed photovoltaic output constraint, the intra-day static reactive power compensator operation constraint, the intra-day energy storage equipment operation constraint and the intra-day node power balance constraint of the power distribution network.
8. The power distribution network optimal scheduling system based on the centralized and distributed architecture is characterized by comprising a centralized and distributed stage optimal scheduling model construction module, a day-ahead stage centralized scheduling constraint construction module, a day-ahead stage centralized optimal scheduling strategy acquisition module, a day-in stage distributed scheduling constraint construction module, a day-in stage distributed optimal scheduling strategy acquisition module and a power distribution network optimal scheduling module; The centralized and distributed stage optimization scheduling model construction module is used for acquiring operation condition data of a power distribution network, and constructing a day-ahead stage centralized scheduling objective function and a day-in stage distributed scheduling objective function of the power distribution network based on the operation condition data so as to form a centralized and distributed stage optimization scheduling model of the power distribution network; The day-ahead stage centralized scheduling constraint construction module is used for acquiring a next day load demand value and a renewable energy source output long-term predicted value of the power distribution network and constructing day-ahead stage centralized scheduling constraint by combining the operation condition data; The day-ahead stage centralized optimal scheduling strategy acquisition module is used for solving the centralized distributed stage optimal scheduling model based on the day-ahead stage centralized scheduling constraint, and determining a structure optimization strategy and an on-load voltage regulating transformer operation strategy of the power distribution network; The daily stage distributed scheduling constraint construction module is used for acquiring the measured data of the next day load of the power distribution network and the short-term predicted value of the output of the renewable energy source, and constructing daily stage distributed scheduling constraint by combining the operation condition data; the intra-day stage distributed optimal scheduling strategy acquisition module is used for solving the centralized distributed stage optimal scheduling model based on the intra-day stage distributed scheduling constraint, determining a photovoltaic output scheduling strategy and an energy storage output scheduling strategy of the power distribution network, and further constructing the intra-day stage distributed optimal scheduling strategy; the power distribution network optimal scheduling module is used for optimally scheduling the power distribution network based on the day-ahead stage centralized optimal scheduling strategy and the day-in stage distributed optimal scheduling strategy.
9. The power distribution network optimal scheduling system based on a centralized and distributed architecture according to claim 8, wherein the centralized and distributed stage optimal scheduling model building module comprises a centralized and distributed stage optimal scheduling model building unit; The centralized and distributed stage optimization scheduling model construction unit is used for acquiring operation condition data of the power distribution network, wherein the operation condition data comprise branch operation data, reconstruction sectional switch state data, distributed photovoltaic output power data and on-load voltage regulating transformer gear data; Constructing a centralized scheduling objective function of a day-ahead stage of the power distribution network based on the branch running data, the reconstructed sectional switch state data, the distributed photovoltaic output power data and the on-load voltage regulating transformer gear data; Performing cluster division on the power distribution network to obtain a plurality of clusters of the power distribution network, and constructing a daily phase distributed scheduling objective function of the power distribution network by combining the branch operation data and the distributed photovoltaic output power data; And forming a centralized and distributed stage optimization scheduling model of the power distribution network based on the day-ahead stage centralized scheduling objective function and the day-in stage distributed scheduling objective function of the power distribution network.
10. The power distribution network optimized dispatching system based on the centralized and distributed architecture as set forth in claim 9, wherein the centralized and distributed stage optimized dispatching model constructing unit comprises a day-ahead stage centralized dispatching objective function constructing subunit; in the day-ahead stage centralized scheduling objective function construction subunit, the branch operation data comprises branch current data and branch resistance data; the day-ahead stage centralized scheduling objective function construction subunit is used for constructing network loss cost of the power distribution network based on the product of the branch current data and the branch resistance data; Determining reconstruction segment switch state transformation data based on the reconstruction segment switch state data, and constructing the reconstruction cost of the power distribution network based on the reconstruction segment switch state transformation data; determining a distributed photovoltaic output power prediction deviation value based on the distributed photovoltaic output power data, and constructing a abandoned renewable energy punishment cost based on the distributed photovoltaic output power prediction deviation value; Determining on-load tap changer gear change data based on the on-load tap changer gear data, and constructing on-load tap changer operation cost based on the on-load tap changer gear change data; based on a preset day-ahead cost weight coefficient set, carrying out weighted summation on the power distribution network loss cost, the power distribution network reconstruction cost, the abandoned renewable energy punishment cost and the on-load voltage regulating transformer operation cost to obtain the total operation cost of the power distribution network; and constructing a centralized scheduling objective function of the power distribution network in a day-ahead stage by taking the minimum total running cost of the power distribution network as a target.

Description

Power distribution network optimal scheduling method and system based on centralized and distributed architecture Technical Field The invention belongs to the technical field of power control, and particularly relates to a power distribution network optimal scheduling method and system based on a centralized and distributed architecture. Background With the continuous development of new energy technology at present, the occupancy rate of distributed resources such as renewable energy sources, energy storage equipment, electric automobiles and the like in a power distribution network is continuously improved, and the power distribution network gradually evolves from a traditional single power supply network to a complex system of multi-source collaborative and bidirectional interaction. Although the wide access of the distributed resources can provide a new driving force for the efficient development of the power distribution network, the current power distribution network optimization scheduling is increasingly complex due to the obvious fluctuation and intermittence of the renewable energy output, and new influences are brought to the safe and stable operation and the power balance control of the power distribution network. In the current power distribution network optimal scheduling scheme, a complete centralized architecture is adopted in power distribution network scheduling, namely, operation data of distributed resources, loads and equipment of the whole network are collected through a scheduling center, then decision making is carried out through a unified optimization model, and control instructions are issued. In addition, the current optimal scheduling of the power distribution network is usually realized based on single scheduling of daily scheduling or daily scheduling, and the single daily scheduling can cause scheduling errors, so that real-time power unbalance occurs in the optimal scheduling of the power distribution network, and the single daily scheduling can cause frequent scheduling of the optimal scheduling of the power distribution network, so that the scheduling loss of the power distribution network is caused. Therefore, a method and a system for optimizing and scheduling a power distribution network based on a centralized and distributed architecture are needed to solve the defects in the prior art. Disclosure of Invention The invention aims to provide a power distribution network optimal scheduling method and system based on a centralized and distributed architecture, which are used for solving the technical problems, and the efficiency and the accuracy of power distribution network optimal scheduling are improved by constructing a centralized and distributed stage optimal scheduling model and adopting a day-before stage centralized scheduling constraint and a day-in stage distributed scheduling constraint. In order to solve the technical problems, the embodiment of the invention provides a power distribution network optimal scheduling method based on a centralized distributed architecture, which comprises the steps of obtaining operation condition data of a power distribution network, constructing a day-ahead stage centralized scheduling objective function and a day-in stage distributed scheduling objective function of the power distribution network based on the operation condition data to form a centralized distributed stage optimal scheduling model of the power distribution network, obtaining a day-ahead stage centralized scheduling constraint by combining a day-ahead stage load demand value and a renewable energy output long-term predicted value of the power distribution network and the operation condition data, solving the centralized distributed stage optimal scheduling model based on the day-ahead stage centralized scheduling constraint, determining a structure optimization strategy of the power distribution network and an on-load voltage regulating transformer operation strategy, further constructing a day-ahead stage centralized optimal scheduling strategy, obtaining day-in stage distributed scheduling constraint by combining the operation condition data, solving the centralized distributed stage optimal scheduling constraint, determining the day-ahead stage optimal scheduling strategy and the optimal energy storage output strategy, and further constructing the day-ahead stage optimal scheduling strategy based on the day-ahead stage optimal scheduling strategy. It can be understood that the invention constructs the centralized scheduling objective function of the day-ahead stage and the distributed scheduling objective function of the day-in stage of the power distribution network, thereby forming a centralized and distributed stage optimizing scheduling model of the power distribution network; the centralized distributed phase optimization scheduling model can be integrated with characteristics of a centralized architecture and a distributed architecture, characteristics