WO2023010759A1 - Power distribution and sale competitive situation-based regional power distribution network gridding load interval prediction method - Google Patents

Power distribution and sale competitive situation-based regional power distribution network gridding load interval prediction method Download PDF

Info

Publication number
WO2023010759A1
WO2023010759A1 PCT/CN2021/138595 CN2021138595W WO2023010759A1 WO 2023010759 A1 WO2023010759 A1 WO 2023010759A1 CN 2021138595 W CN2021138595 W CN 2021138595W WO 2023010759 A1 WO2023010759 A1 WO 2023010759A1
Authority
WO
WIPO (PCT)
Prior art keywords
interval
power
load
charging
electricity
Prior art date
Application number
PCT/CN2021/138595
Other languages
French (fr)
Chinese (zh)
Inventor
钱仲豪
周亮
袁松
陆凌
杨俊�
张骏
徐晓轶
Original Assignee
国网江苏省电力有限公司南通供电分公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 国网江苏省电力有限公司南通供电分公司 filed Critical 国网江苏省电力有限公司南通供电分公司
Publication of WO2023010759A1 publication Critical patent/WO2023010759A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0206—Price or cost determination based on market factors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S50/00—Market activities related to the operation of systems integrating technologies related to power network operation or related to communication or information technologies
    • Y04S50/14—Marketing, i.e. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards

Definitions

  • the invention belongs to the technical field of distribution network interval prediction, and relates to a grid-based load interval prediction method for regional distribution networks under the competition situation of distribution and sale of electricity.
  • the purpose of the present invention is to provide a grid-based load interval prediction method for regional distribution networks under the situation of competition in distribution and sale of electricity. From the perspective of engineering investment and construction, this method is based on Monte Carlo and wavelet neural network modeling analysis, and proposes an optimization method for power supply and consumption intervals with PICP and DADI as evaluation indicators.
  • the technical solution of the present invention a grid-based load interval prediction method for a regional distribution network under the competition situation of distribution and sale of electricity, including the following steps:
  • Step 10 constructing a typical energy consumption model of distribution network end users based on local extremum clustering (CFDP);
  • CFDP local extremum clustering
  • Step 20 According to the power user feature set of the annual load curve and the power user feature set of the daily load curve, research the type of power supply and consumption based on the time scale;
  • Step 30 Research the power supply interval model of flexible loads under different power price mechanisms such as time-of-use power price and real-time power price;
  • Step 40 Carry out modeling research on the charging and discharging model of electric vehicles based on the Monte Carlo method, determine the charging demand of electric vehicles on each time section in combination with Monte Carlo sampling and interval numbers, and determine the charging and discharging of electric vehicles according to their expectations and standard deviations interval;
  • Step 50 Establish a dual-output single-hidden layer neural network model to obtain the upper and lower limits of the output of the photovoltaic power generation system, and use the particle swarm optimization algorithm to perform comprehensive optimization according to the evaluation indicators such as interval coverage rate ICP and interval average width IAW to determine the optimal network model output weight;
  • Step 60 Calculate the PICP and DADI power supply interval flexibility evaluation index according to the historical data and the upper and lower limits of the interval;
  • Step 70 Under the comprehensive consideration of the influence of photovoltaic system, flexible load and electric vehicle load output uncertainty, the social utility and flexibility indicators under the comprehensive power distribution and sales competition situation are the optimization goals, and the power supply and consumption interval value with high prediction accuracy is obtained.
  • CFDP local extremum clustering
  • the parameter d c is the cut-off distance
  • d ij is the distance between i and j
  • ⁇ i is the local density of node i
  • I s is the set of all nodes.
  • the type of power supply and consumption is studied based on the time scale.
  • l(t) is the load curve of month t
  • a is the load curve of the whole year
  • s(t) is the load rate of month t.
  • the modeling of power consumption behavior of power users' loads can be transformed into two sub-problems: the analysis of the total power consumption and the analysis of the time distribution of power consumption.
  • the flexibility of flexible loads and the power supply and consumption interval model are studied under different electricity price mechanisms such as time-of-use electricity price and real-time electricity price.
  • the amount of power variation during the peak period is:
  • ⁇ q 1 and ⁇ p 1 are the relative increments of electricity q and electricity price p during the peak period respectively;
  • ⁇ ij (i ⁇ j) is the cross elasticity coefficient, which means the change rate of electricity demand in period i and the change rate of electricity price in period j
  • ⁇ ii is the elasticity coefficient, which means the ratio between the rate of change of electricity demand and the rate of change of electricity price in period i.
  • the amount of power change during the valley period is:
  • ⁇ q 2 and ⁇ p 2 are the relative increments of electricity q and electricity price p during valley hours, respectively. After the implementation of the peak and valley electricity prices, the electricity consumption during the off-peak period will not be less than the electricity consumption before the implementation.
  • the charging and discharging model of the electric vehicle is modeled and studied based on the Monte Carlo method, and the charging demand of the electric vehicle on each time section is studied using Monte Carlo sampling, according to its expectation and standard deviation to determine the charging and discharging interval of electric vehicles, combined with the maximum likelihood estimation method:
  • the battery state of charge of an electric vehicle and its daily mileage d also approximately satisfy a linear relationship, then the charging time T C of an electric vehicle can be estimated as:
  • W 100 is the average power consumption per 100 kilometers of EV (unit: kW h/100km);
  • P C is the charging power of electric vehicles (unit: kW);
  • d is the daily mileage (unit: km).
  • electric vehicles generally adopt an orderly charging mode, so the charging power demand of a single electric vehicle at time t 0 can be expressed as:
  • P(t 0 ) is the power demand of a single electric vehicle on the t 0 time section
  • P C (t 0 ) is the charging power of a single electric vehicle on the t 0 time section
  • ⁇ C (t 0 ) is the time
  • ⁇ ( ⁇ ) is the probability density function of the initial charging time of the electric vehicle.
  • the number of charging intervals for electric vehicles is:
  • ⁇ is the radius adjustment parameter of the number of intervals
  • ⁇ EV and ⁇ EV are the corresponding expected values and standard deviations.
  • a dual-output single-hidden layer neural network model is established to obtain the upper and lower limits of the output of the photovoltaic power generation system, and the particle swarm optimization algorithm is used for comprehensive optimization according to the interval evaluation index to determine the network model.
  • Optimal output weights can be simplified as follows:
  • P PV is the actual output power of photovoltaics (kW); G STC and G ING are the actual solar radiation intensity (W/m 2 ) under standard conditions, respectively; P STC is the maximum power output of photovoltaic cells under standard test conditions (kW); k PV is the power temperature coefficient (%/°C); T c is the battery temperature; T r is the reference temperature.
  • the renewable patronage interval evaluation index is mainly divided into two aspects: interval coverage probability (ICP) and interval average width (IAW).
  • ICP is the probability that the measured value of the output of the photovoltaic power generation system falls into the interval model. The larger the probability value, the more accurate the interval model is; Accurate interval value, on this basis, construct the interval model comprehensive evaluation index (combinational coverage width-based criterion, CWC) as shown in the following formula:
  • CWC IAW[(1+ ⁇ )ICPe - ⁇ (ICP- ⁇ ) ] (11) where: ⁇ is the control coefficient of ICP, ⁇ is the confidence level, and ⁇ is the amplification factor of the difference between the confidence level and ICP.
  • the flexibility evaluation indexes of power supply and consumption intervals of PICP and DADI are calculated according to the historical data and the upper and lower limits of the interval.
  • PICP is the statistical probability that the actual value falls in the prediction interval, which can be written as:
  • U ij and L ij are given upper and lower boundary values respectively.
  • the PICP index can intuitively reflect the accuracy of the interval, and the higher the value, the greater the probability that the actual load value falls within the prediction interval, and the better the prediction result.
  • DADI can accurately and intuitively reflect the degree of deviation between the actual load value and the forecast interval:
  • d j is the deviation between the real value of load i at the jth time node and the prediction interval.
  • the optimization goal is to maximize the social utility under the competition situation of electricity distribution and sales, and the prediction accuracy is relatively high. High power supply interval value.
  • the cost objective function is:
  • ⁇ b (t) and ⁇ s (t) are the purchase and sale prices of the distribution network from the main grid at time t; u s (t) and u b (t) are the purchase and sale prices of the distribution network from the main grid respectively Electricity sales; ⁇ t is the time interval, usually 1h; T is 24.
  • the present invention has the following advantages:
  • Fig. 1 is the flowchart of the embodiment of the present invention.
  • Fig. 2 is a result diagram of charging and discharging intervals of an electric vehicle according to an embodiment of the present invention
  • Fig. 3 is the result diagram of the power supply load interval of the distribution network according to the embodiment of the present invention.
  • the embodiment of the method of the present invention is based on the gridded load interval prediction method of the regional distribution network under the competition situation of distribution and sales, as shown in Figure 1, the method includes the following steps:
  • Step 10 constructing a typical energy consumption model of distribution network end users based on local extremum clustering (CFDP);
  • CFDP local extremum clustering
  • Step 20 According to the power user feature set of the annual load curve and the power user feature set of the daily load curve, research the type of power supply and consumption based on the time scale;
  • Step 30 Research the power supply interval model of flexible loads under different power price mechanisms such as time-of-use power price and real-time power price;
  • Step 40 Carry out modeling research on the charging and discharging model of electric vehicles based on the Monte Carlo method, determine the charging demand of electric vehicles on each time section in combination with Monte Carlo sampling and interval numbers, and determine the charging and discharging of electric vehicles according to their expectations and standard deviations interval;
  • Step 50 Establish a dual-output single-hidden layer neural network model to obtain the upper and lower limits of the output of the photovoltaic power generation system, and use the particle swarm optimization algorithm to perform comprehensive optimization according to the evaluation indicators such as interval coverage rate ICP and interval average width IAW to determine the optimal network model. output weight;
  • Step 60 Calculate the PICP and DADI power supply interval flexibility evaluation index according to the historical data and the upper and lower limits of the interval;
  • Step 70 Under the comprehensive consideration of the influence of photovoltaic system, flexible load and electric vehicle load output uncertainty, the social utility and flexibility indicators under the comprehensive power distribution and sales competition situation are the optimization goals, and the power supply and consumption interval value with high prediction accuracy is obtained.
  • CFDP local extremum clustering
  • the parameter d c is the cut-off distance
  • d ij is the distance between i and j
  • ⁇ i is the local density of node i
  • I s is the set of all nodes.
  • the type of power supply and consumption is studied based on the time scale.
  • l(t) is the load curve of month t
  • a is the load curve of the whole year
  • s(t) is the load rate of month t.
  • the modeling of power consumption behavior of power users' loads can be transformed into two sub-problems: the analysis of the total power consumption and the analysis of the time distribution of power consumption.
  • the flexibility of flexible loads and the power supply and consumption interval model are studied under different electricity price mechanisms such as time-of-use electricity price and real-time electricity price.
  • the amount of power variation during the peak period is:
  • ⁇ q 1 and ⁇ p 1 are the relative increments of electricity q and electricity price p during the peak period respectively;
  • ⁇ ij (i ⁇ j) is the cross elasticity coefficient, which means the change rate of electricity demand in period i and the change rate of electricity price in period j
  • ⁇ ii is the elasticity coefficient, which means the ratio between the rate of change of electricity demand and the rate of change of electricity price in period i.
  • the amount of power change during the valley period is:
  • ⁇ q 2 and ⁇ p 2 are the relative increments of electricity q and electricity price p during valley hours, respectively. After the implementation of the peak and valley electricity prices, the electricity consumption during the off-peak period will not be less than the electricity consumption before the implementation.
  • the charging and discharging model of the electric vehicle is modeled and studied based on the Monte Carlo method, and the charging demand of the electric vehicle on each time section is studied using Monte Carlo sampling, according to its expectation and standard deviation to determine the charging and discharging interval of electric vehicles, combined with the maximum likelihood estimation method:
  • the battery state of charge of an electric vehicle and its daily mileage d also approximately satisfy a linear relationship, then the charging time T C of an electric vehicle can be estimated as:
  • W 100 is the average power consumption per 100 kilometers of EV (unit: kW h/100km);
  • P C is the charging power of electric vehicles (unit: kW);
  • d is the daily mileage (unit: km).
  • electric vehicles generally adopt an orderly charging mode, so the charging power demand of a single electric vehicle at time t 0 can be expressed as:
  • P(t 0 ) is the power demand of a single electric vehicle on the t 0 time section
  • P C (t 0 ) is the charging power of a single electric vehicle on the t 0 time section
  • ⁇ C (t 0 ) is the time
  • ⁇ ( ⁇ ) is the probability density function of the initial charging time of the electric vehicle.
  • the number of charging intervals for electric vehicles is:
  • ⁇ is the radius adjustment parameter of the number of intervals
  • ⁇ EV and ⁇ EV are the corresponding expected values and standard deviations.
  • a dual-output single-hidden layer neural network model is established to obtain the upper and lower limits of the output of the photovoltaic power generation system, and the particle swarm optimization algorithm is used for comprehensive optimization according to the interval evaluation index to determine the network model.
  • Optimal output weights can be simplified as follows:
  • P PV is the actual output power of photovoltaics (kW); G STC and G ING are the actual solar radiation intensity (W/m2) under standard conditions, respectively; P STC is the maximum power output of photovoltaic cells under standard test conditions ( kW); k PV is the power temperature coefficient (%/°C); T c is the battery temperature; T r is the reference temperature.
  • the renewable patronage interval evaluation index is mainly divided into two aspects: interval coverage probability (ICP) and interval average width (IAW).
  • ICP is the probability that the measured value of the output of the photovoltaic power generation system falls into the interval model. The larger the probability value, the more accurate the interval model is; Accurate interval value, on this basis, construct the interval model comprehensive evaluation index (combinational coverage width-based criterion, CWC) as shown in the following formula:
  • CWC IAW[(1+ ⁇ )ICPe - ⁇ (ICP- ⁇ ) ] (26) where: ⁇ is the control coefficient of ICP, ⁇ is the confidence level, and ⁇ is the amplification factor of the difference between the confidence level and ICP.
  • the flexibility evaluation indexes of power supply and consumption intervals of PICP and DADI are calculated according to the historical data and the upper and lower limits of the interval.
  • PICP is the statistical probability that the actual value falls in the prediction interval, which can be written as:
  • U ij and L ij are given upper and lower boundary values respectively.
  • the PICP index can intuitively reflect the accuracy of the interval, and the higher the value, the greater the probability that the actual load value falls within the prediction interval, and the better the prediction result.
  • DADI can accurately and intuitively reflect the degree of deviation between the actual load value and the forecast interval:
  • d j is the deviation between the real value of load i at the jth time node and the prediction interval.
  • the optimization goal is to maximize the social utility under the competition situation of electricity distribution and sales, and the prediction accuracy is relatively high. High power supply interval value.
  • the cost objective function is:
  • ⁇ b (t) and ⁇ s (t) are the purchase and sale prices of the distribution network from the main grid at time t; u s (t) and u b (t) are the purchase and sale prices of the distribution network from the main grid respectively Electricity sales; ⁇ t is the time interval, usually 1h; T is 24.
  • the IEEE 33 node system is selected for specific analysis and description.
  • the power distribution system includes eight groups of photovoltaic power sources, and the rated capacity of the photovoltaic power generation system is set to 400kVA.
  • the rated capacity of the local photovoltaic equipment is 400kVA, and the photovoltaic power generation system is set on nodes 3, 4, 7, 12, 14, 18, 21, and 24.
  • the power distribution system includes the charging and discharging load of electric vehicles. Assume that the number of charging electric vehicles in one day is 50, the power of charging vehicles is 3kW, and the charging and discharging systems of electric vehicles are set on nodes 3, 4, 7, 12, 14, 18, 21, and 24. .
  • the electricity price t (unit: yuan/kWh) of electricity traded with the upper-level power grid is shown in the following formula:
  • T is the time period.
  • the specific electric vehicle load range, the interactive power range between the distribution network and the main grid, and the load range of nodes containing renewable energy are shown in Figures 2 to 5 below;

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Accounting & Taxation (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • Finance (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Health & Medical Sciences (AREA)
  • Educational Administration (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Data Mining & Analysis (AREA)
  • Public Health (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

Disclosed in the present invention is a power distribution and sale competitive situation-based regional power distribution network gridding load interval prediction method, comprising: carrying out modeling research on typical energy utilization of a terminal user of a power distribution network; carrying out research on a power supply and consumption classification method on the basis of a time scale; carrying out research on the flexibility and a power supply and consumption interval model of a flexible load; carrying out modeling research on a charging and discharging model of an electric vehicle, carrying out research on charging requirements of the electric vehicle on each time section, and determining a charging and discharging interval of the electric vehicle; using a particle swarm optimization algorithm to perform comprehensive optimization, and determining an optimal output weight of a network model; calculating PICP and DADI power supply and consumption interval flexibility evaluation indexes; and obtaining a power supply and consumption interval value having relatively high prediction accuracy. According to the present invention, a plurality of factors such as distributed photovoltaic, flexible load and electric vehicle power supply and consumption intervals are considered, and a power supply and consumption interval value having relatively high accuracy is determined for an optimization target in combination with social benefits and interval indexes in a power distribution and sale competitive situation.

Description

基于配售电竞争态势下区域配电网网格化负荷区间预测方法Gridded Load Interval Prediction Method of Regional Distribution Network Based on Distribution and Sales Competition Situation
本发明属于配电网区间预测技术领域，涉及一种基于配售电竞争态势下区域配电网网格化负荷区间预测方法。The invention belongs to the technical field of distribution network interval prediction, and relates to a grid-based load interval prediction method for regional distribution networks under the competition situation of distribution and sale of electricity.
背景技术Background technique
目前我国分布式电源发展迅猛，分布式电源装机容量不断增加，且分布式电源容量在全国总装机容量占比日趋变大。由于我国政府对开发利用新能源的大力支持和有关政策的推出，使得包括风力发电、光伏发电等形式的可再生能源发电在近些年得到了相当快速的发展，分布式电源尤其是分布式可再生能源实现了大规模的接入与应用。At present, my country's distributed power supply is developing rapidly, and the installed capacity of distributed power supply is increasing continuously, and the proportion of distributed power supply capacity in the total installed capacity of the country is increasing day by day. Due to the strong support of the Chinese government for the development and utilization of new energy and the introduction of relevant policies, renewable energy power generation including wind power generation and photovoltaic power generation has developed rapidly in recent years. Distributed power, especially distributed renewable energy Renewable energy has achieved large-scale access and application.
目前，国内外对可再生能源预测，柔性负荷供用区间模型有较为深入的研究，对含配售电竞争态势下供用电区间和区间评估指标的研究较少。根据已有的供用电区间，结合配售电竞争态势下以社会效益和PICP以及DADI区间评价指标优化供用电区间模型，建立更为合理的配电网供用电区间。At present, there are relatively in-depth studies on renewable energy forecasting and flexible load supply and use interval models at home and abroad, but there are few studies on power supply and use intervals and interval evaluation indicators under the competition situation including distribution and sales. According to the existing power supply and consumption interval, combined with the social benefits and PICP and DADI interval evaluation indicators to optimize the power supply and consumption interval model under the competition situation of power distribution and sales, a more reasonable power supply and consumption interval of the distribution network is established.
发明内容Contents of the invention
本发明的目的在于提供一种基于配售电竞争态势下区域配电网网格化负荷区间预测方法。该方法从工程投资建设的角度，基于蒙特卡洛和小波神经网络建模分析，并提出了以PICP和DADI为评价指标的供用电区间优化方法。The purpose of the present invention is to provide a grid-based load interval prediction method for regional distribution networks under the situation of competition in distribution and sale of electricity. From the perspective of engineering investment and construction, this method is based on Monte Carlo and wavelet neural network modeling analysis, and proposes an optimization method for power supply and consumption intervals with PICP and DADI as evaluation indicators.
本发明的技术解决方案：一种基于配售电竞争态势下区域配电网网格化负荷区间预测方法，包括以下步骤：The technical solution of the present invention: a grid-based load interval prediction method for a regional distribution network under the competition situation of distribution and sale of electricity, including the following steps:
步骤10)基于局部极值聚类(CFDP)构建配电网终端用户典型用能模型；Step 10) constructing a typical energy consumption model of distribution network end users based on local extremum clustering (CFDP);
步骤20)根据年负荷曲线的电力用户特征集和日负荷曲线电力用户特征集，基于时间尺度研究供用电类型；Step 20) According to the power user feature set of the annual load curve and the power user feature set of the daily load curve, research the type of power supply and consumption based on the time scale;
步骤30)基于分时电价和实时电价等不同电价机制下研究柔性负荷的供用电区间模型；Step 30) Research the power supply interval model of flexible loads under different power price mechanisms such as time-of-use power price and real-time power price;
步骤40)基于蒙特卡洛方法对电动汽车充放电模型进行建模研究，结合蒙特卡洛抽样和区间数确定每个时间断面上电动汽车的充电需求，按照其期望和标准差确定电动汽车充放电区间；Step 40) Carry out modeling research on the charging and discharging model of electric vehicles based on the Monte Carlo method, determine the charging demand of electric vehicles on each time section in combination with Monte Carlo sampling and interval numbers, and determine the charging and discharging of electric vehicles according to their expectations and standard deviations interval;
步骤50)建立双输出单隐层神经网络模型获取光伏发电***出力的上下限， 根据区间覆盖率ICP和区间平均宽度IAW等评价指标采用粒子群寻优算法进行综合优化，确定网络模型的最优输出权重；Step 50) Establish a dual-output single-hidden layer neural network model to obtain the upper and lower limits of the output of the photovoltaic power generation system, and use the particle swarm optimization algorithm to perform comprehensive optimization according to the evaluation indicators such as interval coverage rate ICP and interval average width IAW to determine the optimal network model output weight;
步骤60)根据历史数据和区间上下限计算PICP和DADI供用电区间灵活性评价指标；Step 60) Calculate the PICP and DADI power supply interval flexibility evaluation index according to the historical data and the upper and lower limits of the interval;
步骤70)综合考虑光伏***、柔性负荷和电动汽车负荷出力不确定性影响下，综合配售电竞争态势下社会效用和灵活性指标为优化目标，获得预测准确度较高的供用电区间值。Step 70) Under the comprehensive consideration of the influence of photovoltaic system, flexible load and electric vehicle load output uncertainty, the social utility and flexibility indicators under the comprehensive power distribution and sales competition situation are the optimization goals, and the power supply and consumption interval value with high prediction accuracy is obtained.
进一步的，本发明中，所述步骤10)中，基于局部极值聚类(CFDP)构建配电网终端用户典型用能模型，其局部密度为：Further, in the present invention, in the step 10), a typical energy consumption model of distribution network end users is constructed based on local extremum clustering (CFDP), and its local density is:
Figure PCTCN2021138595-appb-000001
Figure PCTCN2021138595-appb-000001
上式中：参数d c为截断距离，d ij为i与j之间的距离；ρ i为i节点局部密度；I s为所有节点集合。 In the above formula: the parameter d c is the cut-off distance, d ij is the distance between i and j; ρ i is the local density of node i; I s is the set of all nodes.
进一步的，本发明中，所述步骤20)中，根据年负荷曲线的电力用户特征集和日负荷曲线电力用户特征集，基于时间尺度研究供用电类型。对于一条给定的负荷曲线：Further, in the present invention, in the step 20), according to the power user feature set of the annual load curve and the power user feature set of the daily load curve, the type of power supply and consumption is studied based on the time scale. For a given load curve:
Figure PCTCN2021138595-appb-000002
Figure PCTCN2021138595-appb-000002
Figure PCTCN2021138595-appb-000003
Figure PCTCN2021138595-appb-000003
式中：l(t)为t月的负荷曲线；a为全年负荷曲线；s(t)为t月的负荷率。In the formula: l(t) is the load curve of month t; a is the load curve of the whole year; s(t) is the load rate of month t.
对于电力用户负荷用电行为的建模就可以转化为对其用电总量的分析和用电的时间分布分析两个子问题。The modeling of power consumption behavior of power users' loads can be transformed into two sub-problems: the analysis of the total power consumption and the analysis of the time distribution of power consumption.
进一步的，本发明中，所述步骤30)中基于分时电价和实时电价等不同电价机制下研究柔性负荷的灵活性和供用电区间模型。对于采用峰谷分时电价的用户来说，峰时段的电量变化量为：Further, in the present invention, in the step 30), the flexibility of flexible loads and the power supply and consumption interval model are studied under different electricity price mechanisms such as time-of-use electricity price and real-time electricity price. For users who adopt the peak-valley time-of-use electricity price, the amount of power variation during the peak period is:
Figure PCTCN2021138595-appb-000004
Figure PCTCN2021138595-appb-000004
式中：Δq 1和Δp 1分别是峰时段时电量q和电价p的相对增量；ε ij(i≠j)是交叉弹性系数，表示在i时段电量需求变化率与在j时段电价变动率之间的比值；ε ii为弹性系数，表示在i时段电量需求变化率与电价变动率之间的比值。 In the formula: Δq 1 and Δp 1 are the relative increments of electricity q and electricity price p during the peak period respectively; ε ij (i≠j) is the cross elasticity coefficient, which means the change rate of electricity demand in period i and the change rate of electricity price in period j The ratio between; ε ii is the elasticity coefficient, which means the ratio between the rate of change of electricity demand and the rate of change of electricity price in period i.
谷时段的电量变化量为：The amount of power change during the valley period is:
Figure PCTCN2021138595-appb-000005
Figure PCTCN2021138595-appb-000005
式中：Δq 2和Δp 2分别是谷时段电量q和电价p的相对增量。行峰谷电价后，谷时段的用电量不会小于执行前的用电量。 In the formula: Δq 2 and Δp 2 are the relative increments of electricity q and electricity price p during valley hours, respectively. After the implementation of the peak and valley electricity prices, the electricity consumption during the off-peak period will not be less than the electricity consumption before the implementation.
进一步的，本发明中，所述步骤40)中，基于蒙特卡洛方法对电动汽车充放电模型进行建模研究，利用蒙特卡洛抽样研究每个时间断面上电动汽车的充电需求，按照其期望和标准差确定电动汽车充放电区间，结合最大拟然估计法：Further, in the present invention, in the step 40), the charging and discharging model of the electric vehicle is modeled and studied based on the Monte Carlo method, and the charging demand of the electric vehicle on each time section is studied using Monte Carlo sampling, according to its expectation and standard deviation to determine the charging and discharging interval of electric vehicles, combined with the maximum likelihood estimation method:
Figure PCTCN2021138595-appb-000006
Figure PCTCN2021138595-appb-000006
式中：μ t＝3.2，σ t＝0.88，t为时间段。电动汽车的电池荷电状态与其日行驶里程d也近似满足线性关系，则电动汽车充电时长T C可估计为： In the formula: μ t =3.2, σ t =0.88, t is the time period. The battery state of charge of an electric vehicle and its daily mileage d also approximately satisfy a linear relationship, then the charging time T C of an electric vehicle can be estimated as:
Figure PCTCN2021138595-appb-000007
Figure PCTCN2021138595-appb-000007
式中，W 100为EV的百公里平均耗电量(单位：kW·h/100km)；P C为电动汽车的充电功率(单位：kW)；d为日行驶里程量(单位：km)。在优化后的峰谷电价时间段内，电动汽车一般采取有序充模式，则单辆电动汽车在t 0时刻的充电功率需求可表述为： In the formula, W 100 is the average power consumption per 100 kilometers of EV (unit: kW h/100km); P C is the charging power of electric vehicles (unit: kW); d is the daily mileage (unit: km). During the optimized peak-valley electricity price period, electric vehicles generally adopt an orderly charging mode, so the charging power demand of a single electric vehicle at time t 0 can be expressed as:
Figure PCTCN2021138595-appb-000008
Figure PCTCN2021138595-appb-000008
式中：P(t 0)为t 0时间断面上单辆电动汽车的功率需求；P C(t 0)为t 0时间断面上单辆电动汽车的充电功率；ζ C(t 0)为时间t 0断面上单辆电动汽车充电功率的概率，Ψ(·)则为电动汽车起始充电时刻的概率密度函数。电动汽车充电区间数为： In the formula: P(t 0 ) is the power demand of a single electric vehicle on the t 0 time section; P C (t 0 ) is the charging power of a single electric vehicle on the t 0 time section; ζ C (t 0 ) is the time The probability of the charging power of a single electric vehicle on the t 0 section, Ψ(·) is the probability density function of the initial charging time of the electric vehicle. The number of charging intervals for electric vehicles is:
Figure PCTCN2021138595-appb-000009
Figure PCTCN2021138595-appb-000009
式中，υ为区间数的半径调节参数，μ EV和σ EV为对应期望值和标准差。 In the formula, υ is the radius adjustment parameter of the number of intervals, and μ EV and σ EV are the corresponding expected values and standard deviations.
进一步的，本发明中，所述步骤50)中，建立双输出单隐层神经网络模型获取光伏发电***出力的上下限，根据区间评价指标采用粒子群寻优算法进行综合优化，确定网络模型的最优输出权重。光伏发电***的稳态功率输出可以用如 下简化模型：Further, in the present invention, in the step 50), a dual-output single-hidden layer neural network model is established to obtain the upper and lower limits of the output of the photovoltaic power generation system, and the particle swarm optimization algorithm is used for comprehensive optimization according to the interval evaluation index to determine the network model. Optimal output weights. The steady-state power output of the photovoltaic power generation system can be simplified as follows:
Figure PCTCN2021138595-appb-000010
Figure PCTCN2021138595-appb-000010
式中：P PV为光伏的实际输出功率(kW)；G STC和G ING分别为标准条件下和实际的太阳辐射强度(W/m 2)；P STC为标准测试条件下光伏电池最大功率输出(kW)；k PV为功率温度系数(％/℃)；T c为电池温度；T r为参考温度。 In the formula: P PV is the actual output power of photovoltaics (kW); G STC and G ING are the actual solar radiation intensity (W/m 2 ) under standard conditions, respectively; P STC is the maximum power output of photovoltaic cells under standard test conditions (kW); k PV is the power temperature coefficient (%/℃); T c is the battery temperature; T r is the reference temperature.
可再生光顾区间评价指标主要分为两个方面：区间覆盖概率(interval coverage probability，ICP)以及区间平均宽度(interval average width,IAW)。ICP即光伏发电***出力的实测值落入区间模型的概率，概率值越大，表明区间模型越准确；IAW则表示区间值的宽窄，理论上应使平均区间宽度尽可能小，以便获得更为精确的区间值，在此基础上构建如下式所示的区间模型综合评价指(combinational coverage width-based criterion,CWC)：The renewable patronage interval evaluation index is mainly divided into two aspects: interval coverage probability (ICP) and interval average width (IAW). ICP is the probability that the measured value of the output of the photovoltaic power generation system falls into the interval model. The larger the probability value, the more accurate the interval model is; Accurate interval value, on this basis, construct the interval model comprehensive evaluation index (combinational coverage width-based criterion, CWC) as shown in the following formula:
CWC＝IAW[(1+γ)ICPe -θ(ICP-μ)]           (11)式中：γ为ICP的控制系数，μ为置信水平，θ为置信水平与ICP的差异放大系数。 CWC＝IAW[(1+γ)ICPe -θ(ICP-μ) ] (11) where: γ is the control coefficient of ICP, μ is the confidence level, and θ is the amplification factor of the difference between the confidence level and ICP.
进一步的，本发明中，所述步骤60)中，根据历史数据和区间上下限计算PICP和DADI的供用电区间灵活性评价指标。PICP为实际值落在预测区间中的统计概率，可写为：Further, in the present invention, in the step 60), the flexibility evaluation indexes of power supply and consumption intervals of PICP and DADI are calculated according to the historical data and the upper and lower limits of the interval. PICP is the statistical probability that the actual value falls in the prediction interval, which can be written as:
Figure PCTCN2021138595-appb-000011
Figure PCTCN2021138595-appb-000011
上式中：m为每一组样本中的数据个数；a ij为预测结果判别指标，当给定值处于给定的上下边界之间时，其取值为1，否则取值为0： In the above formula: m is the number of data in each group of samples; a ij is the discriminant index of the prediction result. When the given value is between the given upper and lower boundaries, its value is 1, otherwise it is 0:
Figure PCTCN2021138595-appb-000012
Figure PCTCN2021138595-appb-000012
式中：U ij和L ij分别为给定上下边界值。PICP指标能够直观反映区间的准确性，其数值越高，表明负荷真实值落在预测区间中的概率越大，预测结果更好。DADI则可以准确直观地反映负荷实际值与预测区间之间的偏离程度： In the formula: U ij and L ij are given upper and lower boundary values respectively. The PICP index can intuitively reflect the accuracy of the interval, and the higher the value, the greater the probability that the actual load value falls within the prediction interval, and the better the prediction result. DADI can accurately and intuitively reflect the degree of deviation between the actual load value and the forecast interval:
Figure PCTCN2021138595-appb-000013
Figure PCTCN2021138595-appb-000013
上式中d j为第j个时间节点处负荷i真实值与预测区间的偏差。综合上述分析可知DADI指标数值越小，表明负荷真实值与预测区间的偏移程度越小，预测区间较为精确。 In the above formula, d j is the deviation between the real value of load i at the jth time node and the prediction interval. Based on the above analysis, it can be seen that the smaller the value of the DADI index, the smaller the deviation between the actual load value and the prediction interval, and the more accurate the prediction interval.
进一步的，本发明中，所述步骤70)中，综合考虑光伏***、柔性负荷和电动汽车负荷出力不确定性影响下，以配售电竞争态势下社会效用最大为优化目标，获得预测准确度较高的供用电区间值。对于配网与微网之间的能源交易分析可知，成本目标函数为：Further, in the present invention, in the step 70), under the comprehensive consideration of the influence of the photovoltaic system, flexible load and electric vehicle load output uncertainty, the optimization goal is to maximize the social utility under the competition situation of electricity distribution and sales, and the prediction accuracy is relatively high. High power supply interval value. For the analysis of energy transactions between the distribution network and the microgrid, the cost objective function is:
Figure PCTCN2021138595-appb-000014
Figure PCTCN2021138595-appb-000014
式中：μ b(t)和μ s(t)分别为t时刻配电网从主网的购售电价；u s(t)和u b(t)分别为配电网从主网的购售电量；Δt为时间间隔，通常取1h；T为24。 In the formula: μ b (t) and μ s (t) are the purchase and sale prices of the distribution network from the main grid at time t; u s (t) and u b (t) are the purchase and sale prices of the distribution network from the main grid respectively Electricity sales; Δt is the time interval, usually 1h; T is 24.
有益效果：与现有技术相比，本发明具有以下优点：Beneficial effect: compared with the prior art, the present invention has the following advantages:
已有研究仅从国内外的短期预测负荷分析，研究了不同类型负荷的耦合特性，针对负荷区间进行概念性的阐述归纳。对于考虑配售电竞争态势下配电网供用电区间模型，本发明基于蒙特卡洛抽样，综合考虑配电网整体社会效益和PICP以及DADI等供用电区间灵活性指标，确定基于配售电竞争态势下区域配电网网格化负荷区间预测方法。Existing studies only analyze the short-term load forecast at home and abroad, study the coupling characteristics of different types of loads, and conceptually summarize the load intervals. For the power supply and consumption interval model of the distribution network considering the situation of distribution and sales competition, the present invention is based on Monte Carlo sampling, and comprehensively considers the overall social benefits of the distribution network and the flexibility indicators of power supply and consumption intervals such as PICP and DADI, and determines the model based on the distribution and sales competition. Gridded load interval forecasting method for regional distribution network under situational conditions.
附图说明Description of drawings
图1为本发明实施例的流程图；Fig. 1 is the flowchart of the embodiment of the present invention;
图2为本发明实施例的电动汽车充放电区间结果图；Fig. 2 is a result diagram of charging and discharging intervals of an electric vehicle according to an embodiment of the present invention;
图3为本发明实施例的配电网供用电负荷区间结果图；Fig. 3 is the result diagram of the power supply load interval of the distribution network according to the embodiment of the present invention;
图4、5为本发明实施例的含可再生能源节点负荷区间结果图。4 and 5 are results diagrams of node load intervals containing renewable energy according to the embodiment of the present invention.
具体实施方式Detailed ways
下面结合附图，对本发明实施例的技术方案做进一步的说明。The technical solutions of the embodiments of the present invention will be further described below in conjunction with the accompanying drawings.
如图1所示，本发明方法的实施例，基于配售电竞争态势下区域配电网网格化负荷区间预测方法如图1所示，该方法包括以下步骤：As shown in Figure 1, the embodiment of the method of the present invention is based on the gridded load interval prediction method of the regional distribution network under the competition situation of distribution and sales, as shown in Figure 1, the method includes the following steps:
步骤10)基于局部极值聚类(CFDP)构建配电网终端用户典型用能模型；Step 10) constructing a typical energy consumption model of distribution network end users based on local extremum clustering (CFDP);
步骤20)根据年负荷曲线的电力用户特征集和日负荷曲线电力用户特征集，基于时间尺度研究供用电类型；Step 20) According to the power user feature set of the annual load curve and the power user feature set of the daily load curve, research the type of power supply and consumption based on the time scale;
步骤30)基于分时电价和实时电价等不同电价机制下研究柔性负荷的供用电区间模型；Step 30) Research the power supply interval model of flexible loads under different power price mechanisms such as time-of-use power price and real-time power price;
步骤40)基于蒙特卡洛方法对电动汽车充放电模型进行建模研究，结合蒙特卡洛抽样和区间数确定每个时间断面上电动汽车的充电需求，按照其期望和标准差确定电动汽车充放电区间；Step 40) Carry out modeling research on the charging and discharging model of electric vehicles based on the Monte Carlo method, determine the charging demand of electric vehicles on each time section in combination with Monte Carlo sampling and interval numbers, and determine the charging and discharging of electric vehicles according to their expectations and standard deviations interval;
步骤50)建立双输出单隐层神经网络模型获取光伏发电***出力的上下限，根据区间覆盖率ICP和区间平均宽度IAW等评价指标采用粒子群寻优算法进行综合优化，确定网络模型的最优输出权重；Step 50) Establish a dual-output single-hidden layer neural network model to obtain the upper and lower limits of the output of the photovoltaic power generation system, and use the particle swarm optimization algorithm to perform comprehensive optimization according to the evaluation indicators such as interval coverage rate ICP and interval average width IAW to determine the optimal network model. output weight;
步骤60)根据历史数据和区间上下限计算PICP和DADI供用电区间灵活性评价指标；Step 60) Calculate the PICP and DADI power supply interval flexibility evaluation index according to the historical data and the upper and lower limits of the interval;
步骤70)综合考虑光伏***、柔性负荷和电动汽车负荷出力不确定性影响下，综合配售电竞争态势下社会效用和灵活性指标为优化目标，获得预测准确度较高的供用电区间值。Step 70) Under the comprehensive consideration of the influence of photovoltaic system, flexible load and electric vehicle load output uncertainty, the social utility and flexibility indicators under the comprehensive power distribution and sales competition situation are the optimization goals, and the power supply and consumption interval value with high prediction accuracy is obtained.
进一步的，本发明中，所述步骤10)中，基于局部极值聚类(CFDP)构建配电网终端用户典型用能模型，其局部密度为：Further, in the present invention, in the step 10), a typical energy consumption model of distribution network end users is constructed based on local extremum clustering (CFDP), and its local density is:
Figure PCTCN2021138595-appb-000015
Figure PCTCN2021138595-appb-000015
上式中：参数d c为截断距离，d ij为i与j之间的距离；ρ i为i节点局部密度；I s为所有节点集合。 In the above formula: the parameter d c is the cut-off distance, d ij is the distance between i and j; ρ i is the local density of node i; I s is the set of all nodes.
进一步的，本发明中，所述步骤20)中，根据年负荷曲线的电力用户特征集和日负荷曲线电力用户特征集，基于时间尺度研究供用电类型。对于一条给定的负荷曲线：Further, in the present invention, in the step 20), according to the power user feature set of the annual load curve and the power user feature set of the daily load curve, the type of power supply and consumption is studied based on the time scale. For a given load curve:
Figure PCTCN2021138595-appb-000016
Figure PCTCN2021138595-appb-000016
Figure PCTCN2021138595-appb-000017
Figure PCTCN2021138595-appb-000017
式中：l(t)为t月的负荷曲线；a为全年负荷曲线；s(t)为t月的负荷率。In the formula: l(t) is the load curve of month t; a is the load curve of the whole year; s(t) is the load rate of month t.
对于电力用户负荷用电行为的建模就可以转化为对其用电总量的分析和用电的时间分布分析两个子问题。The modeling of power consumption behavior of power users' loads can be transformed into two sub-problems: the analysis of the total power consumption and the analysis of the time distribution of power consumption.
进一步的，本发明中，所述步骤30)中基于分时电价和实时电价等不同电价机制下研究柔性负荷的灵活性和供用电区间模型。对于采用峰谷分时电价的用户来说，峰时段的电量变化量为：Further, in the present invention, in the step 30), the flexibility of flexible loads and the power supply and consumption interval model are studied under different electricity price mechanisms such as time-of-use electricity price and real-time electricity price. For users who adopt the peak-valley time-of-use electricity price, the amount of power variation during the peak period is:
Figure PCTCN2021138595-appb-000018
Figure PCTCN2021138595-appb-000018
式中：Δq 1和Δp 1分别是峰时段时电量q和电价p的相对增量；ε ij(i≠j)是交叉弹性系数，表示在i时段电量需求变化率与在j时段电价变动率之间的比值；ε ii为弹性系数，表示在i时段电量需求变化率与电价变动率之间的比值。 In the formula: Δq 1 and Δp 1 are the relative increments of electricity q and electricity price p during the peak period respectively; ε ij (i≠j) is the cross elasticity coefficient, which means the change rate of electricity demand in period i and the change rate of electricity price in period j The ratio between; ε ii is the elasticity coefficient, which means the ratio between the rate of change of electricity demand and the rate of change of electricity price in period i.
谷时段的电量变化量为：The amount of power change during the valley period is:
Figure PCTCN2021138595-appb-000019
Figure PCTCN2021138595-appb-000019
式中：Δq 2和Δp 2分别是谷时段电量q和电价p的相对增量。行峰谷电价后，谷时段的用电量不会小于执行前的用电量。 In the formula: Δq 2 and Δp 2 are the relative increments of electricity q and electricity price p during valley hours, respectively. After the implementation of the peak and valley electricity prices, the electricity consumption during the off-peak period will not be less than the electricity consumption before the implementation.
进一步的，本发明中，所述步骤40)中，基于蒙特卡洛方法对电动汽车充放电模型进行建模研究，利用蒙特卡洛抽样研究每个时间断面上电动汽车的充电需求，按照其期望和标准差确定电动汽车充放电区间，结合最大拟然估计法：Further, in the present invention, in the step 40), the charging and discharging model of the electric vehicle is modeled and studied based on the Monte Carlo method, and the charging demand of the electric vehicle on each time section is studied using Monte Carlo sampling, according to its expectation and standard deviation to determine the charging and discharging interval of electric vehicles, combined with the maximum likelihood estimation method:
Figure PCTCN2021138595-appb-000020
Figure PCTCN2021138595-appb-000020
式中：μ t＝3.2，σ t＝0.88，t为时间段。电动汽车的电池荷电状态与其日行驶里程d也近似满足线性关系，则电动汽车充电时长T C可估计为： In the formula: μ t =3.2, σ t =0.88, t is the time period. The battery state of charge of an electric vehicle and its daily mileage d also approximately satisfy a linear relationship, then the charging time T C of an electric vehicle can be estimated as:
Figure PCTCN2021138595-appb-000021
Figure PCTCN2021138595-appb-000021
式中，W 100为EV的百公里平均耗电量(单位：kW·h/100km)；P C为电动汽车的充电功率(单位：kW)；d为日行驶里程量(单位：km)。在优化后的峰谷电价时间段内，电动汽车一般采取有序充模式，则单辆电动汽车在t 0时刻的充电功率需求可表述为： In the formula, W 100 is the average power consumption per 100 kilometers of EV (unit: kW h/100km); P C is the charging power of electric vehicles (unit: kW); d is the daily mileage (unit: km). During the optimized peak-valley electricity price period, electric vehicles generally adopt an orderly charging mode, so the charging power demand of a single electric vehicle at time t 0 can be expressed as:
Figure PCTCN2021138595-appb-000022
Figure PCTCN2021138595-appb-000022
式中：P(t 0)为t 0时间断面上单辆电动汽车的功率需求；P C(t 0)为t 0时间断面上单辆电动汽车的充电功率；ζ C(t 0)为时间t 0断面上单辆电动汽车充电功率的概率，Ψ(·)则为电动汽车起始充电时刻的概率密度函数。电动汽车充电区间数为： In the formula: P(t 0 ) is the power demand of a single electric vehicle on the t 0 time section; P C (t 0 ) is the charging power of a single electric vehicle on the t 0 time section; ζ C (t 0 ) is the time The probability of the charging power of a single electric vehicle on the t 0 section, Ψ(·) is the probability density function of the initial charging time of the electric vehicle. The number of charging intervals for electric vehicles is:
Figure PCTCN2021138595-appb-000023
Figure PCTCN2021138595-appb-000023
式中，υ为区间数的半径调节参数，μ EV和σ EV为对应期望值和标准差。 In the formula, υ is the radius adjustment parameter of the number of intervals, and μ EV and σ EV are the corresponding expected values and standard deviations.
进一步的，本发明中，所述步骤50)中，建立双输出单隐层神经网络模型获取光伏发电***出力的上下限，根据区间评价指标采用粒子群寻优算法进行综合优化，确定网络模型的最优输出权重。光伏发电***的稳态功率输出可以用如下简化模型：Further, in the present invention, in the step 50), a dual-output single-hidden layer neural network model is established to obtain the upper and lower limits of the output of the photovoltaic power generation system, and the particle swarm optimization algorithm is used for comprehensive optimization according to the interval evaluation index to determine the network model. Optimal output weights. The steady-state power output of the photovoltaic power generation system can be simplified as follows:
Figure PCTCN2021138595-appb-000024
Figure PCTCN2021138595-appb-000024
式中：P PV为光伏的实际输出功率(kW)；G STC和G ING分别为标准条件下和实际的太阳辐射强度(W/m2)；P STC为标准测试条件下光伏电池最大功率输出(kW)；k PV为功率温度系数(％/℃)；T c为电池温度；T r为参考温度。 In the formula: P PV is the actual output power of photovoltaics (kW); G STC and G ING are the actual solar radiation intensity (W/m2) under standard conditions, respectively; P STC is the maximum power output of photovoltaic cells under standard test conditions ( kW); k PV is the power temperature coefficient (%/°C); T c is the battery temperature; T r is the reference temperature.
可再生光顾区间评价指标主要分为两个方面：区间覆盖概率(interval coverage probability，ICP)以及区间平均宽度(interval average width,IAW)。ICP即光伏发电***出力的实测值落入区间模型的概率，概率值越大，表明区间模型越准确；IAW则表示区间值的宽窄，理论上应使平均区间宽度尽可能小，以便获得更为精确的区间值，在此基础上构建如下式所示的区间模型综合评价指(combinational coverage width-based criterion,CWC)：The renewable patronage interval evaluation index is mainly divided into two aspects: interval coverage probability (ICP) and interval average width (IAW). ICP is the probability that the measured value of the output of the photovoltaic power generation system falls into the interval model. The larger the probability value, the more accurate the interval model is; Accurate interval value, on this basis, construct the interval model comprehensive evaluation index (combinational coverage width-based criterion, CWC) as shown in the following formula:
CWC＝IAW[(1+γ)ICPe -θ(ICP-μ)]          (26)式中：γ为ICP的控制系数，μ为置信水平，θ为置信水平与ICP的差异放大系数。 CWC＝IAW[(1+γ)ICPe -θ(ICP-μ) ] (26) where: γ is the control coefficient of ICP, μ is the confidence level, and θ is the amplification factor of the difference between the confidence level and ICP.
进一步的，本发明中，所述步骤60)中，根据历史数据和区间上下限计算PICP和DADI的供用电区间灵活性评价指标。PICP为实际值落在预测区间中的统计概率，可写为：Further, in the present invention, in the step 60), the flexibility evaluation indexes of power supply and consumption intervals of PICP and DADI are calculated according to the historical data and the upper and lower limits of the interval. PICP is the statistical probability that the actual value falls in the prediction interval, which can be written as:
Figure PCTCN2021138595-appb-000025
Figure PCTCN2021138595-appb-000025
上式中：m为每一组样本中的数据个数；a ij为预测结果判别指标，当给定值处于给定的上下边界之间时，其取值为1，否则取值为0： In the above formula: m is the number of data in each group of samples; a ij is the discriminant index of the prediction result. When the given value is between the given upper and lower boundaries, its value is 1, otherwise it is 0:
Figure PCTCN2021138595-appb-000026
Figure PCTCN2021138595-appb-000026
式中：U ij和L ij分别为给定上下边界值。PICP指标能够直观反映区间的准确性，其数值越高，表明负荷真实值落在预测区间中的概率越大，预测结果更好。DADI则可以准确直观地反映负荷实际值与预测区间之间的偏离程度： In the formula: U ij and L ij are given upper and lower boundary values respectively. The PICP index can intuitively reflect the accuracy of the interval, and the higher the value, the greater the probability that the actual load value falls within the prediction interval, and the better the prediction result. DADI can accurately and intuitively reflect the degree of deviation between the actual load value and the forecast interval:
Figure PCTCN2021138595-appb-000027
Figure PCTCN2021138595-appb-000027
上式中d j为第j个时间节点处负荷i真实值与预测区间的偏差。综合上述分析可知DADI指标数值越小，表明负荷真实值与预测区间的偏移程度越小，预测区间较为精确。 In the above formula, d j is the deviation between the real value of load i at the jth time node and the prediction interval. Based on the above analysis, it can be seen that the smaller the value of the DADI index, the smaller the deviation between the actual load value and the prediction interval, and the more accurate the prediction interval.
进一步的，本发明中，所述步骤70)中，综合考虑光伏***、柔性负荷和电动汽车负荷出力不确定性影响下，以配售电竞争态势下社会效用最大为优化目标，获得预测准确度较高的供用电区间值。对于配网与微网之间的能源交易分析可知，成本目标函数为：Further, in the present invention, in the step 70), under the comprehensive consideration of the influence of the photovoltaic system, flexible load and electric vehicle load output uncertainty, the optimization goal is to maximize the social utility under the competition situation of electricity distribution and sales, and the prediction accuracy is relatively high. High power supply interval value. For the analysis of energy transactions between the distribution network and the microgrid, the cost objective function is:
Figure PCTCN2021138595-appb-000028
Figure PCTCN2021138595-appb-000028
式中：μ b(t)和μ s(t)分别为t时刻配电网从主网的购售电价；u s(t)和u b(t)分别为配电网从主网的购售电量；Δt为时间间隔，通常取1h；T为24。 In the formula: μ b (t) and μ s (t) are the purchase and sale prices of the distribution network from the main grid at time t; u s (t) and u b (t) are the purchase and sale prices of the distribution network from the main grid respectively Electricity sales; Δt is the time interval, usually 1h; T is 24.
下面列举一具体实施例。A specific embodiment is enumerated below.
选取IEEE 33节点***做具体分析说明，配电***包含八组光伏电源,光伏发电***的额定容量设为400kVA。当地的光伏设备额定容量为400kVA，光伏发电***设置在3，4，7，12，14，18，21，24节点上。配电***包含电动汽车充放电负荷，假定一天内电动汽车充电数目为50，充电汽车功率为3kW，电动汽车充放电***设置在3，4，7，12，14，18，21，24节点上。与上级电网之间交易电量电价t(单位：元/千瓦时)如下式所示：The IEEE 33 node system is selected for specific analysis and description. The power distribution system includes eight groups of photovoltaic power sources, and the rated capacity of the photovoltaic power generation system is set to 400kVA. The rated capacity of the local photovoltaic equipment is 400kVA, and the photovoltaic power generation system is set on nodes 3, 4, 7, 12, 14, 18, 21, and 24. The power distribution system includes the charging and discharging load of electric vehicles. Assume that the number of charging electric vehicles in one day is 50, the power of charging vehicles is 3kW, and the charging and discharging systems of electric vehicles are set on nodes 3, 4, 7, 12, 14, 18, 21, and 24. . The electricity price t (unit: yuan/kWh) of electricity traded with the upper-level power grid is shown in the following formula:
Figure PCTCN2021138595-appb-000029
Figure PCTCN2021138595-appb-000029
式中：T为时间段。具体电动汽车负荷区间，配电网与主网交互功率区间和含可再生能源节点负荷区间分别如下图2至5所示；In the formula: T is the time period. The specific electric vehicle load range, the interactive power range between the distribution network and the main grid, and the load range of nodes containing renewable energy are shown in Figures 2 to 5 below;
以上显示和描述了本发明的基本原理、主要特征和优点。本领域的技术人员应该了解，本发明不受上述具体实施例的限制，上述具体实施例和说明书中的描 述只是为了进一步说明本发明的原理，在不脱离本发明精神和范围的前提下，本发明还会有各种变化和改进，这些变化和改进都落入要求保护的本发明范围内。本发明要求保护的范围由权利要求书及其等效物界定。The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the art should understand that the present invention is not limited by the above-mentioned specific examples. The descriptions in the above-mentioned specific examples and the description are only to further illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention The invention also has various changes and improvements, and these changes and improvements all fall within the scope of the claimed invention. The protection scope of the present invention is defined by the claims and their equivalents.

Claims (8)

  1. 一种基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，该方法包括以下步骤：A grid-based load interval prediction method for a regional distribution network based on the competition situation of distribution and sale of electricity, characterized in that the method includes the following steps:
    步骤10)基于局部极值聚类(CFDP)构建配电网终端用户典型用能模型；Step 10) constructing a typical energy consumption model of distribution network end users based on local extremum clustering (CFDP);
    步骤20)根据年负荷曲线的电力用户特征集和日负荷曲线电力用户特征集，基于时间尺度研究供用电类型；Step 20) According to the power user feature set of the annual load curve and the power user feature set of the daily load curve, research the type of power supply and consumption based on the time scale;
    步骤30)基于分时电价和实时电价等不同电价机制下研究柔性负荷的供用电区间模型；Step 30) Research the power supply interval model of flexible loads under different power price mechanisms such as time-of-use power price and real-time power price;
    步骤40)基于蒙特卡洛方法对电动汽车充放电模型进行建模研究，结合蒙特卡洛抽样和区间数确定每个时间断面上电动汽车的充电需求，按照其期望和标准差确定电动汽车充放电区间；Step 40) Carry out modeling research on the charging and discharging model of electric vehicles based on the Monte Carlo method, determine the charging demand of electric vehicles on each time section in combination with Monte Carlo sampling and interval numbers, and determine the charging and discharging of electric vehicles according to their expectations and standard deviations interval;
    步骤50)建立双输出单隐层神经网络模型获取光伏发电***出力的上下限，根据区间覆盖率ICP和区间平均宽度IAW等评价指标采用粒子群寻优算法进行综合优化，确定网络模型的最优输出权重；Step 50) Establish a dual-output single-hidden layer neural network model to obtain the upper and lower limits of the output of the photovoltaic power generation system, and use the particle swarm optimization algorithm to perform comprehensive optimization according to the evaluation indicators such as interval coverage rate ICP and interval average width IAW to determine the optimal network model. output weight;
    步骤60)根据历史数据和区间上下限计算PICP和DADI供用电区间灵活性评价指标；Step 60) Calculate the PICP and DADI power supply interval flexibility evaluation index according to the historical data and the upper and lower limits of the interval;
    步骤70)综合考虑光伏***、柔性负荷和电动汽车负荷出力不确定性影响下，综合配售电竞争态势下社会效用和灵活性指标为优化目标，获得预测准确度较高的供用电区间值。Step 70) Under the comprehensive consideration of the influence of photovoltaic system, flexible load and electric vehicle load output uncertainty, the social utility and flexibility indicators under the comprehensive power distribution and sales competition situation are the optimization goals, and the power supply and consumption interval value with high prediction accuracy is obtained.
  2. 根据权利要求1所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤10)中，基于局部极值聚类构建配电网终端用户典型用能模型，其局部密度为：According to claim 1, the gridded load interval prediction method of regional distribution network based on the distribution and sales competition situation is characterized in that, in the step 10), the typical consumption of end users of the distribution network is constructed based on local extreme value clustering. energy model, its local density is:
    Figure PCTCN2021138595-appb-100001
    Figure PCTCN2021138595-appb-100001
    上式中：参数d c＞0为截断距离，d ij为i与j之间的距离；ρ i为i节点局部密度；I s为所有节点集合。 In the above formula: the parameter d c >0 is the cut-off distance, d ij is the distance between i and j; ρ i is the local density of node i; I s is the set of all nodes.
  3. 根据权利要求2所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤20)中，根据年负荷曲线的电力用户特征集和日负荷曲线电力用户特征集，基于时间尺度研究供用电类型；对于一条给定的负荷曲线：According to claim 2, the regional distribution network gridded load interval prediction method based on the distribution and sales competition situation, characterized in that, in the step 20), according to the electric power user characteristic set of the annual load curve and the daily load curve power User feature set, based on the time scale to study the type of electricity supply and consumption; for a given load curve:
    Figure PCTCN2021138595-appb-100002
    Figure PCTCN2021138595-appb-100002
    Figure PCTCN2021138595-appb-100003
    Figure PCTCN2021138595-appb-100003
    式中：l(t)为t月的负荷曲线；a为全年负荷曲线；s(t)为t月的负荷率；In the formula: l(t) is the load curve of month t; a is the load curve of the whole year; s(t) is the load rate of month t;
    对于电力用户负荷用电行为的建模就可以转化为对其用电总量的分析和用电的时间分布分析两个子问题。The modeling of power consumption behavior of power users' loads can be transformed into two sub-problems: the analysis of the total power consumption and the analysis of the time distribution of power consumption.
  4. 根据权利要求3所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤30)中基于分时电价和实时电价等不同电价机制下研究柔性负荷的灵活性和供用电区间模型；对于采用峰谷分时电价的用户来说，峰时段的电量变化量为：According to claim 3, the gridded load interval prediction method of regional distribution network based on the distribution and sales competition situation, characterized in that, in the step 30), the flexible load is studied under different electricity price mechanisms such as time-of-use electricity price and real-time electricity price The flexibility and power supply interval model; for users who adopt the peak-valley time-of-use electricity price, the amount of power variation during the peak period is:
    Figure PCTCN2021138595-appb-100004
    Figure PCTCN2021138595-appb-100004
    式中：Δq 1和Δp 1分别是峰时段时电量q和电价p的相对增量；ε ij(i≠j)是交叉弹性系数，表示在i时段电量需求变化率与在j时段电价变动率之间的比值；ε ii为弹性系数，表示在i时段电量需求变化率与电价变动率之间的比值； In the formula: Δq 1 and Δp 1 are the relative increments of electricity q and electricity price p during the peak period respectively; ε ij (i≠j) is the cross elasticity coefficient, which means the change rate of electricity demand in period i and the change rate of electricity price in period j The ratio between; ε ii is the elasticity coefficient, which means the ratio between the rate of change of electricity demand and the rate of change of electricity price in period i;
    谷时段的电量变化量为：The amount of power change during the valley period is:
    Figure PCTCN2021138595-appb-100005
    Figure PCTCN2021138595-appb-100005
    式中：Δq 2和Δp 2分别是谷时段电量q和电价p的相对增量；行峰谷电价后，谷时段的用电量不会小于执行前的用电量。 In the formula: Δq 2 and Δp 2 are the relative increments of electricity q and electricity price p during the off-peak period, respectively; after the peak-valley electricity price is applied, the electricity consumption during the off-peak period will not be less than the electricity consumption before the implementation.
  5. 根据权利要求4所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤40)中，基于蒙特卡洛方法对电动汽车充放电模型进行建模研究，利用蒙特卡洛抽样研究每个时间断面上电动汽车的充电需求，按照其期望和标准差确定电动汽车充放电区间，结合最大拟然估计法：According to claim 4, the regional distribution network gridded load interval prediction method based on the distribution and sales competition situation, characterized in that, in the step 40), the charging and discharging model of the electric vehicle is modeled based on the Monte Carlo method Research, use Monte Carlo sampling to study the charging demand of electric vehicles on each time section, determine the charging and discharging interval of electric vehicles according to their expectations and standard deviations, and combine the maximum likelihood estimation method:
    Figure PCTCN2021138595-appb-100006
    Figure PCTCN2021138595-appb-100006
    式中：μ t＝3.2，σ t＝0.88，t为时间段；电动汽车的电池荷电状态与其日行驶里程d也近似满足线性关系，则电动汽车充电时长T C可估计为： In the formula: μ t = 3.2, σ t = 0.88, t is the time period; the battery state of charge of the electric vehicle and its daily mileage d also approximately satisfy the linear relationship, then the charging time T C of the electric vehicle can be estimated as:
    Figure PCTCN2021138595-appb-100007
    Figure PCTCN2021138595-appb-100007
    式中，W 100为EV的百公里平均耗电量，单位：kW·h/100km；P C为电动汽车的充电功率，单位：kW；d为日行驶里程量，单位：km；在优化后的峰谷电价时间段内，电动汽车一般采取有序充模式，则单辆电动汽车在t 0时刻的充电功率需求可表述为： In the formula, W 100 is the average power consumption per 100 kilometers of EV, unit: kW h/100km; P C is the charging power of electric vehicles, unit: kW; d is the daily mileage, unit: km; after optimization During the period of peak and valley electricity prices, electric vehicles generally adopt the orderly charging mode, so the charging power demand of a single electric vehicle at time t 0 can be expressed as:
    Figure PCTCN2021138595-appb-100008
    Figure PCTCN2021138595-appb-100008
    式中：P(t 0)为t 0时间断面上单辆电动汽车的功率需求；P C(t 0)为t 0时间断面上单辆电动汽车的充电功率；ζ C(t 0)为时间t 0断面上单辆电动汽车充电功率的概率，Ψ(·)则为电动汽车起始充电时刻的概率密度函数；电动汽车充电区间数为： In the formula: P(t 0 ) is the power demand of a single electric vehicle on the t 0 time section; P C (t 0 ) is the charging power of a single electric vehicle on the t 0 time section; ζ C (t 0 ) is the time The probability of the charging power of a single electric vehicle on the t 0 section, Ψ( ) is the probability density function of the initial charging time of the electric vehicle; the number of electric vehicle charging intervals is:
    Figure PCTCN2021138595-appb-100009
    Figure PCTCN2021138595-appb-100009
    式中，υ为区间数的半径调节参数，μ EV和σ EV为对应期望值和标准差。 In the formula, υ is the radius adjustment parameter of the number of intervals, and μ EV and σ EV are the corresponding expected values and standard deviations.
  6. 根据权利要求5所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤50)中，建立双输出单隐层神经网络模型获取光伏发电***出力的上下限，根据区间评价指标采用粒子群寻优算法进行综合优化，确定网络模型的最优输出权重；光伏发电***的稳态功率输出可以用如下简化模型：According to claim 5, the gridded load interval prediction method of regional distribution network based on the distribution and sales competition situation, characterized in that, in the step 50), a dual-output single-hidden layer neural network model is established to obtain the output of the photovoltaic power generation system Based on the upper and lower limits of the interval evaluation index, the particle swarm optimization algorithm is used for comprehensive optimization to determine the optimal output weight of the network model; the steady-state power output of the photovoltaic power generation system can be simplified as follows:
    Figure PCTCN2021138595-appb-100010
    Figure PCTCN2021138595-appb-100010
    式中：P PV为光伏的实际输出功率kW；G STC和G ING分别为标准条件下和实际的太阳辐射强度(W/m2)；P STC为标准测试条件下光伏电池最大功率输出kW；k PV为功率温度系数％/℃；T c为电池温度；T r为参考温度； In the formula: P PV is the actual output power kW of photovoltaics; G STC and G ING are the actual solar radiation intensity (W/m2) under standard conditions and actual conditions respectively; P STC is the maximum power output kW of photovoltaic cells under standard test conditions; k PV is power temperature coefficient%/℃; T c is battery temperature; T r is reference temperature;
    可再生光顾区间评价指标主要分为两个方面：区间覆盖概率以及区间平均宽度；ICP即光伏发电***出力的实测值落入区间模型的概率，概率值越大，表明区间模型越准确；IAW则表示区间值的宽窄，理论上应使平均区间宽度尽可能小，以便获得更为精确的区间值，在此基础上构建如下式所示的区间模型综合评价指：The evaluation index of renewable patronage interval is mainly divided into two aspects: interval coverage probability and interval average width; ICP is the probability that the measured output value of the photovoltaic power generation system falls into the interval model. The larger the probability value, the more accurate the interval model; IAW is Indicates the width of the interval value. Theoretically, the average interval width should be made as small as possible in order to obtain a more accurate interval value. On this basis, the comprehensive evaluation index of the interval model shown in the following formula is constructed:
    Figure PCTCN2021138595-appb-100011
    Figure PCTCN2021138595-appb-100011
    式中：γ为ICP的控制系数，μ为置信水平，θ为置信水平与ICP的差异放大系数。In the formula: γ is the control coefficient of ICP, μ is the confidence level, and θ is the amplification factor of the difference between the confidence level and ICP.
  7. 根据权利要求6所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤60)中，根据历史数据和区间上下限计算PICP和DADI的供用电区间灵活性评价指标；PICP为实际值落在预测区间中的统计概率，可写为：According to claim 6, the gridded load interval prediction method of regional distribution network based on the distribution and sales competition situation, characterized in that, in the step 60), the supply and use of PICP and DADI are calculated according to the historical data and the upper and lower limits of the interval Electrical interval flexibility evaluation index; PICP is the statistical probability that the actual value falls in the prediction interval, which can be written as:
    Figure PCTCN2021138595-appb-100012
    Figure PCTCN2021138595-appb-100012
    上式中：m为每一组样本中的数据个数；a ij为预测结果判别指标，当给定值处于给定的上下边界之间时，其取值为1，否则取值为0： In the above formula: m is the number of data in each group of samples; a ij is the discriminant index of the prediction result. When the given value is between the given upper and lower boundaries, its value is 1, otherwise it is 0:
    Figure PCTCN2021138595-appb-100013
    Figure PCTCN2021138595-appb-100013
    式中：U ij和L ij分别为给定上下边界值；PICP指标能够直观反映区间的准确性，其数值越高，表明负荷真实值落在预测区间中的概率越大，预测结果更好；DADI则可以准确直观地反映负荷实际值与预测区间之间的偏离程度： In the formula: U ij and L ij are the given upper and lower boundary values respectively; the PICP index can intuitively reflect the accuracy of the interval, the higher the value, the greater the probability that the actual load value falls in the prediction interval, and the better the prediction result; DADI can accurately and intuitively reflect the degree of deviation between the actual load value and the forecast interval:
    Figure PCTCN2021138595-appb-100014
    Figure PCTCN2021138595-appb-100014
    上式中d j为第j个时间节点处负荷i真实值与预测区间的偏差；综合上述分析可知DADI指标数值越小，表明负荷真实值与预测区间的偏移程度越小，预测区间较为精确。 In the above formula, d j is the deviation between the actual value of load i and the prediction interval at the jth time node; based on the above analysis, it can be seen that the smaller the value of the DADI index, the smaller the deviation between the actual load value and the prediction interval, and the more accurate the prediction interval .
  8. 根据权利要求7所述的基于配售电竞争态势下区域配电网网格化负荷区间预测方法，其特征在于，所述步骤70)中，综合考虑光伏***、柔性负荷和电动汽车负荷出力不确定性影响下，以配售电竞争态势下社会效用最大为优化目标，获得预测准确度较高的供用电区间值；对于配网与微网之间的能源交易分析可知，成本目标函数为：According to claim 7, the gridded load interval prediction method of regional distribution network based on the competition situation of distribution and sale of electricity, is characterized in that, in the step 70), the uncertainty of the output of photovoltaic system, flexible load and electric vehicle load is comprehensively considered Under the influence of power distribution and sales competition, the optimization goal is to maximize the social utility under the competition situation of electricity distribution and sales, and obtain the interval value of power supply and consumption with high prediction accuracy; for the analysis of energy transactions between the distribution network and the microgrid, the cost objective function is:
    Figure PCTCN2021138595-appb-100015
    Figure PCTCN2021138595-appb-100015
    式中：μ b(t)和μ s(t)分别为t时刻配电网从主网的购售电价；u s(t)和u b(t)分别为配电网从主网的购售电量；Δt为时间间隔，通常取1h；T为24。 In the formula: μ b (t) and μ s (t) are the purchase and sale prices of the distribution network from the main grid at time t; u s (t) and u b (t) are the purchase and sale prices of the distribution network from the main grid respectively Electricity sales; Δt is the time interval, usually 1h; T is 24.
PCT/CN2021/138595 2021-08-04 2021-12-24 Power distribution and sale competitive situation-based regional power distribution network gridding load interval prediction method WO2023010759A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110889709.1 2021-08-04
CN202110889709.1A CN113642787A (en) 2021-08-04 2021-08-04 Regional power distribution network gridding load interval prediction method based on distribution and sale power competition situation

Publications (1)

Publication Number Publication Date
WO2023010759A1 true WO2023010759A1 (en) 2023-02-09

Family

ID=78419567

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/138595 WO2023010759A1 (en) 2021-08-04 2021-12-24 Power distribution and sale competitive situation-based regional power distribution network gridding load interval prediction method

Country Status (2)

Country Link
CN (1) CN113642787A (en)
WO (1) WO2023010759A1 (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994627A (en) * 2023-03-22 2023-04-21 湖南大学 Residential building flexible load day-ahead optimal scheduling method, device, equipment and medium
CN116632842A (en) * 2023-07-26 2023-08-22 国网山东省电力公司信息通信公司 Clustering characteristic-based method and system for predicting distribution type photovoltaic load probability of platform
CN116901774A (en) * 2023-09-11 2023-10-20 南京安充智能科技有限公司 Flexible power distribution method, system and storage medium based on full-network charging pile
CN117056848A (en) * 2023-10-10 2023-11-14 湖北华中电力科技开发有限责任公司 Monitoring data quality analysis method for improving power system decision
CN117134361A (en) * 2023-10-27 2023-11-28 国网天津市电力公司宝坻供电分公司 High-proportion green electric cross-region cross-voltage class digestion method, device, equipment and medium
CN117332291A (en) * 2023-11-30 2024-01-02 中国电力科学研究院有限公司 Distributed photovoltaic-oriented regional resource monitoring layout method and system
CN117559566A (en) * 2024-01-11 2024-02-13 国网吉林省电力有限公司经济技术研究院 Optimization method for park system access to power distribution network based on load prediction uncertainty
CN117639036A (en) * 2023-11-21 2024-03-01 广东健电新能源科技有限公司 Charging and discharging planning method and system for charging pile
CN117634931A (en) * 2024-01-25 2024-03-01 华北电力大学 Electric automobile adjustment capability prediction method and system considering charging behavior
CN117895659A (en) * 2024-03-14 2024-04-16 山东理工大学 Automatic scheduling method and system for smart power grid

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113642787A (en) * 2021-08-04 2021-11-12 国网江苏省电力有限公司南通供电分公司 Regional power distribution network gridding load interval prediction method based on distribution and sale power competition situation
CN115549109A (en) * 2022-09-15 2022-12-30 清华大学 Mass flexible load rapid aggregation control method and device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140304025A1 (en) * 2011-10-28 2014-10-09 Viridity Energy, Inc. Managing energy assets associated with transport operations
CN107017619A (en) * 2017-03-29 2017-08-04 华北电力大学 The photovoltaic charge station network distribution type energy management method at non-cooperative game visual angle
CN109872228A (en) * 2019-01-17 2019-06-11 华北电力大学（保定） Meter and probabilistic flexible load polymerization quotient participate in electricity market Bidding system
CN113642787A (en) * 2021-08-04 2021-11-12 国网江苏省电力有限公司南通供电分公司 Regional power distribution network gridding load interval prediction method based on distribution and sale power competition situation

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107425520B (en) * 2017-06-12 2020-04-21 东南大学 Active power distribution network three-phase interval state estimation method containing node injection power uncertainty
US10825113B2 (en) * 2018-03-16 2020-11-03 Dalian University Of Technology Method for short-term generation scheduling of cascade hydropower plants coupling cluster analysis and decision tree
CN110956329B (en) * 2019-12-02 2023-04-25 国网浙江省电力有限公司绍兴供电公司 Load prediction method based on distributed photovoltaic and electric automobile space-time distribution
CN112131733B (en) * 2020-09-15 2022-03-11 燕山大学 Distributed power supply planning method considering influence of charging load of electric automobile
CN112217202B (en) * 2020-09-29 2022-06-14 东南大学 Distributed new energy, energy storage and power distribution network planning method considering flexibility investment

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140304025A1 (en) * 2011-10-28 2014-10-09 Viridity Energy, Inc. Managing energy assets associated with transport operations
CN107017619A (en) * 2017-03-29 2017-08-04 华北电力大学 The photovoltaic charge station network distribution type energy management method at non-cooperative game visual angle
CN109872228A (en) * 2019-01-17 2019-06-11 华北电力大学（保定） Meter and probabilistic flexible load polymerization quotient participate in electricity market Bidding system
CN113642787A (en) * 2021-08-04 2021-11-12 国网江苏省电力有限公司南通供电分公司 Regional power distribution network gridding load interval prediction method based on distribution and sale power competition situation

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994627A (en) * 2023-03-22 2023-04-21 湖南大学 Residential building flexible load day-ahead optimal scheduling method, device, equipment and medium
CN115994627B (en) * 2023-03-22 2023-07-14 湖南大学 Residential building flexible load day-ahead optimal scheduling method, device, equipment and medium
CN116632842A (en) * 2023-07-26 2023-08-22 国网山东省电力公司信息通信公司 Clustering characteristic-based method and system for predicting distribution type photovoltaic load probability of platform
CN116632842B (en) * 2023-07-26 2023-11-10 国网山东省电力公司信息通信公司 Clustering characteristic-based method and system for predicting distribution type photovoltaic load probability of platform
CN116901774A (en) * 2023-09-11 2023-10-20 南京安充智能科技有限公司 Flexible power distribution method, system and storage medium based on full-network charging pile
CN116901774B (en) * 2023-09-11 2023-11-14 南京安充智能科技有限公司 Flexible power distribution method, system and storage medium based on full-network charging pile
CN117056848B (en) * 2023-10-10 2023-12-29 湖北华中电力科技开发有限责任公司 Monitoring data quality analysis method for improving power system decision
CN117056848A (en) * 2023-10-10 2023-11-14 湖北华中电力科技开发有限责任公司 Monitoring data quality analysis method for improving power system decision
CN117134361A (en) * 2023-10-27 2023-11-28 国网天津市电力公司宝坻供电分公司 High-proportion green electric cross-region cross-voltage class digestion method, device, equipment and medium
CN117134361B (en) * 2023-10-27 2024-01-30 国网天津市电力公司宝坻供电分公司 High-proportion green electric cross-region cross-voltage class digestion method, device, equipment and medium
CN117639036A (en) * 2023-11-21 2024-03-01 广东健电新能源科技有限公司 Charging and discharging planning method and system for charging pile
CN117639036B (en) * 2023-11-21 2024-04-26 广东健电新能源科技有限公司 Charging and discharging planning method and system for charging pile
CN117332291A (en) * 2023-11-30 2024-01-02 中国电力科学研究院有限公司 Distributed photovoltaic-oriented regional resource monitoring layout method and system
CN117332291B (en) * 2023-11-30 2024-03-26 中国电力科学研究院有限公司 Distributed photovoltaic-oriented regional resource monitoring layout method and system
CN117559566A (en) * 2024-01-11 2024-02-13 国网吉林省电力有限公司经济技术研究院 Optimization method for park system access to power distribution network based on load prediction uncertainty
CN117634931A (en) * 2024-01-25 2024-03-01 华北电力大学 Electric automobile adjustment capability prediction method and system considering charging behavior
CN117634931B (en) * 2024-01-25 2024-04-16 华北电力大学 Electric automobile adjustment capability prediction method and system considering charging behavior
CN117895659A (en) * 2024-03-14 2024-04-16 山东理工大学 Automatic scheduling method and system for smart power grid
CN117895659B (en) * 2024-03-14 2024-05-31 山东理工大学 Automatic scheduling method and system for smart power grid

Also Published As

Publication number Publication date
CN113642787A (en) 2021-11-12

Similar Documents

Publication Publication Date Title
WO2023010759A1 (en) Power distribution and sale competitive situation-based regional power distribution network gridding load interval prediction method
Xuan et al. PV-load decoupling based demand response baseline load estimation approach for residential customer with distributed PV system
Sun A multi-objective optimization model for fast electric vehicle charging stations with wind, PV power and energy storage
CN109359389B (en) Urban electric vehicle charging decision method based on typical load dynamic game
Kang et al. How to better share energy towards a carbon-neutral city? A review on application strategies of battery energy storage system in city
Roberts et al. Characterisation of Australian apartment electricity demand and its implications for low-carbon cities
CN112488362A (en) Energy storage optimization configuration method for coordinating electric automobile to participate in auxiliary service market
Srithapon et al. Surrogate-Assisted multi-objective probabilistic optimal power flow for distribution network with photovoltaic generation and electric vehicles
Zhang et al. Grid-connected photovoltaic battery systems: A comprehensive review and perspectives
Mohammadi et al. Optimal operation management of microgrids using the point estimate method and firefly algorithm while considering uncertainty
Li et al. Optimal battery schedule for grid-connected photovoltaic-battery systems of office buildings based on a dynamic programming algorithm
Zhou A regression learner-based approach for battery cycling ageing prediction―advances in energy management strategy and techno-economic analysis
CN107480907A (en) The optimization method of provincial power network power purchase proportioning containing wind-powered electricity generation under a kind of time-of-use tariffs
Mutanen Improving electricity distribution system state estimation with AMR-based load profiles
Tong et al. Reusing electric vehicle battery for demand side management integrating dynamic pricing
CN105337574A (en) Robust-regression-based distributed photovoltaic generating electricity-stealing identification method
Assolami et al. Impact on voltage quality and transformer aging of residential prosumer ownership of plug-in electric vehicles: Assessment and solutions
CN109193643B (en) Method and system for calculating power distribution and distribution system network loss and reliability
Hashemipour et al. Uncertainty modeling for participation of electric vehicles in collaborative energy consumption
CN109636033B (en) Spontaneous self-use comprehensive electricity price prediction method for distributed photovoltaic project
CN112801343A (en) Energy storage system capacity planning method considering multi-meteorological-scene adaptive cost
CN102904252A (en) Method for solving uncertainty trend of power distribution network with distributed power supply
Telford et al. Dirichlet sampled capacity and loss estimation for LV distribution networks with partial observability
Han et al. Sensitivity analysis of acquisition granularity of photovoltaic output power to capacity configuration of energy storage systems
Naceur et al. Deep Learning Algorithm to Predict The Integration Efficiency Of Wind Turbine In Smart Grid

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21952622

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE