CN-121984108-A - Power distribution network collaborative optimization control method integrating cable planning and charging load

CN121984108ACN 121984108 ACN121984108 ACN 121984108ACN-121984108-A

Abstract

The invention relates to the technical field of power distribution network optimization control, in particular to a power distribution network collaborative optimization control method integrating cable planning and charging load, which is characterized in that a planning scheme and an operation strategy can be adapted to source load fluctuation in advance by constructing a photovoltaic and charging load probability model, and a collaborative scheduling strategy of a transformer, a reactive compensation device and the charging load is optimized and formulated in advance according to different scenes, so that network-source-load active collaboration is realized, network loss is reduced, equipment utilization efficiency is improved, simultaneously, the investment cost of a cable and long-term operation cost are optimized uniformly, a scheme with the lowest comprehensive cost of a full life cycle is automatically screened, long-term economic loss caused by traditional planning and operation cutting is avoided, full-period verification and extreme working condition test are carried out on the optimal collaborative scheme through digital twin simulation, safety, economy and performance indexes are evaluated quantitatively, an optimization-verification decision loop is formed, the feasibility of the scheme is ensured, and the implementation risk is reduced.

Inventors

LI BIN
SUN XUEBIN
LI YANG
WANG JIAFENG
LI BAICHENG
JIN HONGYU

Assignees

国网辽宁省电力有限公司鞍山供电公司

Dates

Publication Date: 20260505
Application Date: 20251216

Claims (8)

1. The power distribution network collaborative optimization control method integrating cable planning and charging load is characterized by comprising the following steps of: acquiring and processing multi-dimensional basic data of a target area, and preprocessing and extracting features of the multi-dimensional basic data to obtain a basic feature data set; the method comprises the steps of generating an uncertainty probability model and a scene, namely constructing a photovoltaic output probability model and a charging load probability model based on a basic characteristic data set, and generating a plurality of initial scenes covering different level combinations of the photovoltaic output and the charging load by sampling based on the output of the photovoltaic output probability model and the charging load probability model; based on the initial scene, adopting a clustering algorithm to reduce and obtain a plurality of typical operation scenes with known probability; step four, generating a cable planning scheme and making a collaborative optimization scheme decision, namely acquiring planning candidate schemes of a cable line in a target area, constructing a total cost objective function based on the cable investment cost and the expected running total cost of each planning candidate scheme, and carrying out optimization solution to output an optimal collaborative scheme; And fifthly, collaborative scheme simulation and verification, namely carrying out full-period operation simulation and extreme working condition verification on the optimal collaborative scheme in a digital twin environment, collecting and evaluating safe, economic and feasible core operation indexes in the simulation, and judging whether the optimal collaborative scheme passes the verification or not based on collaborative simulation scores.
2. The collaborative optimization control method for a power distribution network integrating cable planning and charging loads according to claim 1, wherein the uncertainty probability model and scene generation analysis process is as follows: establishing an uncertainty probability model based on the basic characteristic data set, wherein the uncertainty probability model comprises a photovoltaic output probability model and a charging load probability model; sampling the number of N groups (N=1000) of initial scenes aiming at the core dimension according to the data sample size and the calculation requirement; dividing the probability distribution interval of the photovoltaic output rate and the charging load rate output by the uncertainty probability model into N mutually non-overlapping subintervals respectively; Randomly extracting 1 sample point in each subinterval, and combining to form 1 group of photovoltaic-load coupling scenes, wherein the scenes comprise photovoltaic output rate, charging load rate values and corresponding occurrence probability; And reversely calculating an actual power value based on the sampled photovoltaic output rate and the sampled charging load rate, wherein the actual photovoltaic output rate is equal to the actual photovoltaic output rate multiplied by the installed capacity, the actual charging load rate is equal to the actual charging load rate multiplied by the maximum bearing load of the charging station, and simultaneously, a time tag (such as different time periods of a typical day) is given to each group of scenes to form a complete initial scene data set.
3. The collaborative optimization control method for the power distribution network, which fuses cable planning and charging loads according to claim 2, is characterized in that scene screening is specifically that an existing K-means clustering algorithm is adopted to cut down an initial scene data set, and 10 typical operation scene sets S with known probability and with uncertainty characteristics of the charging loads reserved are obtained.
4. The collaborative optimization control method for a power distribution network integrating cable planning and charging loads according to claim 1, wherein the cable planning scheme generation and collaborative optimization scheme decision analysis process is as follows: T1, acquiring a planning candidate scheme of a cable line of a target area, wherein the planning candidate scheme comprises a cable type, a path and topological connection; T2, calculating expected total operation cost including network loss, electricity purchasing cost and control equipment action cost under the typical operation scene set S for each planning candidate scheme; t3, acquiring the cable investment cost of each cable planning candidate scheme, and constructing an objective function based on the cable investment cost and the expected total running cost, wherein the objective function is represented by the ratio of the cable investment cost to the expected total running cost; and T4, carrying out convergence processing after preset iteration on the objective function, and finally outputting a cable planning candidate scheme and setting the cable planning candidate scheme as an optimal cooperative scheme.
5. The collaborative optimization control method for a power distribution network integrating cable planning and charging loads according to claim 1, wherein the collaborative scheme simulation and verification analysis process is as follows: disassembling the output optimal collaborative scheme into planning parameters and operation control parameters, respectively importing a digital twin model and completing simulation configuration; and sequentially selecting all typical operation scenes, executing full-period operation simulation of the power distribution network, simulating the action process of a transformer tap and a capacitor bank within 24 hours according to a day-ahead switching plan and a charging reference curve, and calculating the electricity purchasing cost and the network loss cost of each period.
6. And an extreme working condition scene (such as photovoltaic full-power and charging load peak superposition and temporary line fault removal) is additionally added, and the safety margin of the scheme under the extreme condition is verified.
7. The collaborative optimization control method for a power distribution network integrating cable planning and charging loads according to claim 5, wherein core operation indexes in the simulation process are collected in real time, wherein the core operation indexes comprise safety indexes, economic indexes and performance feasible indexes; Based on the collected core operation indexes, acquiring each positive operation index and each negative operation index in the safety index, the economic index and the performance feasibility index; normalizing positive running indexes and negative running indexes in the safety indexes, the economic indexes and the performance feasible indexes, and mapping all indexes to a [0,1] interval; Forward operation index normalization processing, namely GX= (X-Xmin)/(Xmax-Xmin); and (3) carrying out negative operation index normalization processing, namely GX= (Xmax-X)/(Xmax-Xmin), wherein X is a positive operation index or a negative operation index actual value.
8. The collaborative optimization control method for a power distribution network integrating cable planning and charging loads according to claim 6, wherein corresponding weight coefficients Lk, k=1, 2,3 of a safety class index, an economic class index and a performance feasible class index are obtained; setting a index in a safety layer index after normalization processing as GXg, setting a index in an economic layer index as GXm and setting a index in a feasible layer index as GXb, wherein g, m and b are natural numbers larger than zero; Calculating to obtain a collaborative simulation score based on L1× (GXg ×preset weight coefficient g) +L2× (GXm ×preset weight coefficient m) +L3× (GXb ×preset weight coefficient b); and judging the co-simulation score, and outputting an optimal co-scheme or outputting a simulation failing list.

Description

Power distribution network collaborative optimization control method integrating cable planning and charging load Technical Field The invention relates to the technical field of power distribution network optimization control, in particular to a power distribution network collaborative optimization control method integrating cable planning and charging loads. Background With deep propulsion and acceleration of energy transformation of a 'double-carbon' strategy, the form and the operation mode of a power distribution network are deeply changed, on one hand, renewable energy sources such as distributed Photovoltaics (PV) are connected into the power distribution network with high permeability, the intermittence and randomness of output bring serious challenges to the power balance and voltage stability of a system, on the other hand, the large-scale popularization of Electric Vehicles (EV) rapidly increases the charging load, the high uncertainty of space-time distribution further aggravates the operation complexity of the power distribution network, and the high uncertainty of the two sides of the 'source-charge' causes the traditional power distribution network to face huge pressure in the planning and operation aspect. In the operation stage, a dispatcher passively handles actual fluctuation of photovoltaic output and charging load under a fixed power grid structure, the system safety is maintained by means of regulating a transformer tap, switching a capacitor and the like, and economic optimization is difficult to realize from the global angle, and the operation is not considered in the planning, so that the operation is limited by a planned splitting mode, and the two main problems are that firstly, the planning stage is to meet all possible extreme scenes, excessive investment is often caused, a line is increased, the operation cost (such as network loss and electricity purchasing cost) is high, secondly, the adaptability is insufficient, the certainty load growth prediction based on the planning is difficult to accurately describe the random characteristics of photovoltaic and electric automobile charging, the safety problems of voltage overrun, line overload and the like are easy to be caused, and the system flexibility and toughness are insufficient. In view of the above technical drawbacks, a solution is now proposed. Disclosure of Invention The invention aims to provide a power distribution network collaborative optimization control method integrating cable planning and charging load, which solves the technical defects. The invention aims at realizing the technical scheme that the power distribution network collaborative optimization control method integrating cable planning and charging load comprises the following steps: acquiring and processing multi-dimensional basic data of a target area, and preprocessing and extracting features of the multi-dimensional basic data to obtain a basic feature data set; the method comprises the steps of generating an uncertainty probability model and a scene, namely constructing a photovoltaic output probability model and a charging load probability model based on a basic characteristic data set, and generating a plurality of initial scenes covering different level combinations of the photovoltaic output and the charging load by sampling based on the output of the photovoltaic output probability model and the charging load probability model; based on the initial scene, adopting a clustering algorithm to reduce and obtain a plurality of typical operation scenes with known probability; step four, generating a cable planning scheme and making a collaborative optimization scheme decision, namely acquiring planning candidate schemes of a cable line in a target area, constructing a total cost objective function based on the cable investment cost and the expected running total cost of each planning candidate scheme, and carrying out optimization solution to output an optimal collaborative scheme; And fifthly, collaborative scheme simulation and verification, namely carrying out full-period operation simulation and extreme working condition verification on the optimal collaborative scheme in a digital twin environment, collecting and evaluating safe, economic and feasible core operation indexes in the simulation, and judging whether the optimal collaborative scheme passes the verification or not based on collaborative simulation scores. Preferably, the uncertainty probability model and scene generation analysis process is as follows: establishing an uncertainty probability model based on the basic characteristic data set, wherein the uncertainty probability model comprises a photovoltaic output probability model and a charging load probability model; sampling the number of N groups (N=1000) of initial scenes aiming at the core dimension according to the data sample size and the calculation requirement; dividing the probability distribution interval of the photovoltaic