CN-121984030-A - Power distribution network planning operation coordination method considering voltage safety and operation economy

CN121984030ACN 121984030 ACN121984030 ACN 121984030ACN-121984030-A

Abstract

The invention discloses a power distribution network planning operation coordination method considering voltage safety and operation economy, which comprises the steps of firstly obtaining power distribution network topology and source load data to generate a typical time sequence operation scene, secondly constructing a capacitor bank site selection model, optimizing by utilizing a discrete Kepler algorithm to generate a capacitor bank configuration initial solution, inputting the initial solution into a full-time domain simulation kernel, carrying out closed loop iterative correction based on voltage qualification rate and comprehensive cost to output an optimal installation position of the capacitor bank, finally executing coordination operation based on the optimal installation position, optimizing on-load voltage regulating transformer and energy storage action by utilizing the Kepler algorithm based on a voltage potential energy field by a slow layer to establish a voltage reference, controlling capacitor bank switching by utilizing an event-driven discrete differential evolution to eliminate steady-state deviation by utilizing a sagging control strategy by a fast layer, and compensating instantaneous shortage by utilizing a sagging control strategy. The invention can effectively eliminate voltage out-of-limit and realize the optimal balance of voltage safety and operation economy.

Inventors

ZHANG DENGJI
YANG LIAN
LI YIBO
WU QITING
CHEN YICHUAN

Assignees

广东工业大学

Dates

Publication Date: 20260505
Application Date: 20260210

Claims (4)

1. A power distribution network planning operation coordination method taking voltage safety and operation economy into consideration is characterized by comprising the following steps: S1, acquiring topological structure and line impedance parameters of a power distribution network, reading source-load time sequence data, and preprocessing the data to generate a power distribution network typical time sequence operation scene covering multiple typical working conditions; S2, constructing a capacitor bank site selection planning model, determining candidate installation nodes of the capacitor bank based on voltage sensitivity indexes, constructing the site selection planning model aiming at minimizing comprehensive operation cost, and solving the model by utilizing a discrete Kepler algorithm to generate a capacitor bank configuration initial solution; S3, constructing a full-time domain dynamic simulation kernel containing slow, medium and fast multi-level control logic, inputting the initial solution of the capacitor bank configuration into the full-time domain dynamic simulation kernel for deduction verification, and performing closed-loop iterative correction based on the voltage qualification rate and the comprehensive operation cost index until the safety constraint is met, and outputting an optimal installation position; S4, starting multi-time scale collaborative operation based on the optimal installation position, wherein the collaborative operation comprises the steps of performing collaborative scheduling on an on-load voltage regulating transformer and stored energy by using a Kepler optimization algorithm based on a voltage potential field at a slow layer time scale, establishing a voltage reference, constructing an event-driven gating mechanism based on a physical saturation feedback closed loop at a middle layer time scale, controlling switching of a capacitor bank to eliminate steady-state deviation, compensating instantaneous reactive power shortage by using a photovoltaic inverter and a static reactive generator at a fast layer time scale and adopting a collaborative droop control strategy based on a priority order, and feeding back capacity saturation to the middle layer in real time.
2. The collaborative method for planning operation of a power distribution network for both voltage safety and operational economy according to claim 1, wherein step S2 comprises: s2-1, calculating network loss sensitivity factors of all nodes of the power distribution network by adopting perturbation method : In the formula (1), the components are as follows, Is the active loss of the whole network in the reference state, To at the first Individual node injection micro-increment The active loss of the whole network after that, each node is divided into a plurality of nodes according to the sensitivity factor of the network loss The sizes of the (B) are arranged in descending order before selection The individual nodes are used as candidate node sets for capacitor bank installation; s2-2, constructing an addressing planning model containing operation network loss cost and voltage out-of-limit penalty terms by taking the minimum comprehensive operation cost of the capacitor bank as an objective function, wherein the mathematical expression is as follows: In the formula (2), the amino acid sequence of the compound, For the overall objective function of the planning model, For the full-time operation of the network loss cost, For the voltage out-of-limit penalty term, Is a set of time slices for a typical scene, Is that The total active power loss of the system at the moment, In order to simulate the step of time, Is the electricity price per unit of electricity, As a voltage out-of-limit penalty factor, Is that Time node Is used to determine the actual voltage amplitude of (a), And The lower limit and the upper limit of the node voltage safety range are respectively, As a maximum function for calculating only when the voltage exceeds Out-of-limit magnitude at range; s2-3, solving individual positions in the defined population by adopting an integer coding mode based on a discrete Kepler optimization algorithm, wherein each individual represents a numbered combination of a group of candidate node sets, simulating an celestial body gravity mechanism, and executing discrete updating operation, wherein the expression is as follows: In the formula (3), the amino acid sequence of the compound, For updated individual locations, i.e. newly generated capacitor bank candidate configurations, In order to update the current individual's location prior to the update, For the optimal individual in the current population, For gravitation adsorption operation, the optimal individual is carried out with a certain probability The locus gene in (c) is directly replaced to the current individual, For orbital perturbation operations, new nodes are selected from the candidate node set with random probability to replace the current gene, And the random mixed execution logic representing the two operations repeatedly executes the discrete updating operation until the preset population iteration times are reached, and outputs the initial solution of the capacitor bank configuration.
3. The collaborative method for planning operation of a power distribution network for both voltage safety and operational economy according to claim 1, wherein step S3 comprises: S3-1, full-time domain dynamic simulation deduction, mapping the capacitor bank configuration generated in the step S2-3 into a simulation kernel, simulating source load fluctuation and control response for 24 hours all day, and recording Time node Is the actual voltage amplitude of (a) ; S3-2, constructing a closed loop correction index and defining a voltage out-of-limit severity index When any node voltage is detected to be out of the safe range When the correction mechanism is triggered: In the formula (4), the amino acid sequence of the compound, Is an index of the severity of the full network voltage out-of-limit, Is the total number of time periods for the full time domain simulation, The total number of nodes of the power distribution network; S3-3, executing weight self-adaptive iteration if Identifying the node with the most serious voltage threshold violation And feedback the weight to the sensitivity of the node in step S2-1 And (4) performing adaptive updating: in the formula (5), the amino acid sequence of the compound, Is the first Node at time of iteration Is used for the sensitivity feedback weight of the (c), Is the first Node at time of iteration Is used for the sensitivity feedback weight of the (c), Correcting step length coefficients for the weights; s3-4, feeding the updated weight of the step S3-3 back to the step S2-2, and re-performing site selection optimization until the iteration termination condition is met And the comprehensive operation cost is not reduced any more, the iteration is stopped, and the optimal installation position of the capacitor bank is output.
4. The collaborative method for planning operations of a power distribution network for both voltage safety and operational economy according to claim 1, wherein step S4 comprises: S4-1, executing global reference correction of a slow layer based on pressure feedback, and acquiring source load prediction data by taking an hour as a time step; s4-1-1, at the starting time of the dispatching, receiving the accumulated action pressure signal uploaded by the middle layer Calculating voltage potential punishment weight capable of reflecting middle layer equipment pressure : In the formula (6), the amino acid sequence of the compound, For the total number of actual actions of the capacitor bank in the previous cycle, As the reference threshold value for the number of actions, As a basis weight for the weight of the base, Is the feedback gain; s4-1-2, based on the voltage potential energy penalty weight calculated in the step S4-1-1 Constructing a slow layer comprehensive objective function containing network loss cost and voltage potential energy penalty term : In the formula (7), the amino acid sequence of the compound, The total number of scheduling periods optimized for the slow layer, Is a voltage safety limit; s4-1-3, executing a Kepler optimization algorithm (PI-KOA) based on a voltage potential field to solve, wherein the algorithm introduces a voltage feedback gravitation mechanism based on the traditional Kepler planetary operation rule, namely, for each individual in the population, calculating a slow layer comprehensive objective function in the step S4-1-2 Negative gradient gravitation vector received in constructed voltage safety potential energy field And superimpose it on the kepler orbit update equation using the negative gradient gravity vector Driving the population to evolve towards a voltage safety low potential energy region, wherein the negative gradient attraction vector The calculation formula of (2) is as follows: in the formula (8), the amino acid sequence of the compound, For the purpose of the gradient operator, Is that Time node The voltage pair control variable of the on-load voltage regulating transformer OLTC gear and the sensitivity gradient vector of the energy storage ESS output; S4-2, executing middle-layer event-driven control, and monitoring the system state in real time at a minute-level frequency above a reference determined by slow-layer optimization; S4-2-1, based on an event-driven gating mechanism, activating an algorithm to avoid frequent invalid actions of the equipment if and only if the node voltage is detected to exceed an early warning range or the interval between the current moment and the last action moment exceeds a cooling threshold; S4-2-2 constructing an fitness function with minimum switch action cost as a core : In the formula (9), the amino acid sequence of the compound, For real-time active loss of the power distribution network at the current moment, For the cost factor of the switching action of the capacitor bank, As the total number of groups of the capacitor groups, Is the first The shift change amount of the group capacitor at the current time and the last sampling time, A real-time voltage out-of-limit penalty term; s4-2-3 the fitness function calculated if and only if step S4-2-2 When the preset threshold value or the voltage exceeds the limit, a discrete differential evolution algorithm (D-MODE) is activated to solve the model, the integer rounding mapping operation is carried out on the continuous real number vector generated by the mutation and crossover operation, so that the continuous real number vector is forcedly placed in a legal discrete gear set, and an executable capacitor switching instruction is output The calculation formula is as follows: in the formula (10), the amino acid sequence of the compound, In order to obtain the variation vector, As a result of the fact that the target vector, For the cross-over probability to be the case, As a rounding function, for mapping the continuous variable into discrete capacitor files; S4-3, executing fast layer on-site response control, and constructing an on-site feedback loop based on voltage droop characteristics by using millisecond-level response characteristics of the photovoltaic inverter PV and the static var generator SVG; S4-3-1, establishing voltage dead zone discrimination logic, and calculating total reactive power compensation demand of nodes ; S4-3-2, according to equipment capacity constraint, executing SVG priority and PV auxiliary residual allocation calculation of a photovoltaic inverter, and solving SVG instruction of the SVG by utilizing capacity limiting logic With photovoltaic inverter PV commands ; S4-3-3 SVG instruction of static var generator With photovoltaic inverter PV commands Issuing to corresponding equipment for execution, and calculating the saturation of the total capacity of the fast layer in real time Uploading to the middle layer: In the formula (11), the amino acid sequence of the compound, And The actual output reactive power of the static var generator SVG and the photovoltaic inverter PV respectively, And Rated capacities of SVG and PV of photovoltaic inverter respectively, when the total capacity saturation of fast layer When the preset threshold value is exceeded, the middle layer control is forcedly locked to prevent oscillation, so that closed loop feedback with multiple time scales is formed.

Description

Power distribution network planning operation coordination method considering voltage safety and operation economy Technical Field The invention relates to the technical field of power system planning and operation control, in particular to a power distribution network planning operation coordination method considering voltage safety and operation economy. Background The voltage qualification rate and the network loss are key indexes for measuring the running quality of the power distribution network. With the improvement of the permeability of the distributed new energy, the tidal current characteristics of the power distribution network are radically changed, and the uncertainty of the source load aggravates the contradiction between the voltage out-of-limit risk and the operation economy. In the prior art, planning site selection and operation control fracturing are often considered, dynamic voltage fluctuation of minute level is difficult to deal with based on static configuration of extreme scenes, and when the traditional optimization algorithm is used for solving the problem of mixed integer nonlinear programming involving discrete and continuous variables, local optimization is easy to fall into, and physical safety boundary and economic cost are difficult to accurately balance. Therefore, a synergistic approach that enables overall planning and site selection and multi-time scale operation control is needed to fully exploit equipment regulatory potential. Therefore, the invention provides a power distribution network planning operation coordination method which gives consideration to voltage safety and operation economy, and realizes global optimal allocation and efficient utilization of power distribution network resources through discrete optimization of a planning layer and multi-level closed-loop control of an operation layer. Disclosure of Invention In order to achieve the above purpose, the technical scheme provided by the invention is as follows: S1, acquiring topological structure and line impedance parameters of a power distribution network, reading source-load time sequence data, and preprocessing the data to generate a power distribution network typical time sequence operation scene covering multiple typical working conditions; S2, constructing a capacitor bank site selection planning model, determining candidate installation nodes of the capacitor bank based on voltage sensitivity indexes, constructing the site selection planning model aiming at minimizing comprehensive operation cost, and solving the model by utilizing a discrete Kepler algorithm to generate a capacitor bank configuration initial solution; S3, constructing a full-time domain dynamic simulation kernel containing slow, medium and fast multi-level control logic, inputting the initial solution of the capacitor bank configuration into the full-time domain dynamic simulation kernel for deduction verification, and performing closed-loop iterative correction based on the voltage qualification rate and the comprehensive operation cost index until the safety constraint is met, and outputting an optimal installation position; S4, starting multi-time scale collaborative operation based on the optimal installation position, wherein the collaborative operation comprises the steps of performing collaborative scheduling on an on-load voltage regulating transformer and stored energy by using a Kepler optimization algorithm based on a voltage potential field at a slow layer time scale, establishing a voltage reference, constructing an event-driven gating mechanism based on a physical saturation feedback closed loop at a middle layer time scale, controlling switching of a capacitor bank to eliminate steady-state deviation, compensating instantaneous reactive power shortage by using a photovoltaic inverter and a static reactive generator at a fast layer time scale and adopting a collaborative droop control strategy based on a priority order, and feeding back capacity saturation to the middle layer in real time. The step S2 specifically includes the following steps: s2-1, calculating network loss sensitivity factors of all nodes of the power distribution network by adopting perturbation method : In the formula (1), the components are as follows,Is the active loss of the whole network in the reference state,To at the firstIndividual node injection micro-incrementThe active loss of the whole network after that, each node is divided into a plurality of nodes according to the sensitivity factor of the network lossThe sizes of the (B) are arranged in descending order before selectionThe individual nodes are used as candidate node sets for capacitor bank installation; s2-2, constructing an addressing planning model containing operation network loss cost and voltage out-of-limit penalty terms by taking the minimum comprehensive operation cost of the capacitor bank as an objective function, wherein the mathematical expression is as follows: In the