CN101777765A - On-line load simulation method of power system - Google Patents

Abstract

The invention relates to an on-line load simulation method of a power system, belonging to the technical field of power system measurement. The method utilizes user side power information collected by a power system marketing department load control system and transformer substation power information provided by an energy management system (EMS), combines with sampling manual survey statistics, utilizes a load node power supply area network chart provided by a dispatching department as well as related data, such as transformer number, volume, reactive compensation volume, line impedance and the like of a 220 kV transformer substation and subordinate 110 kV and 35 kV transformer substations and adopts a statistics composite method to realize the on-line load modelling of the load node so as to accurately model the load node. The invention improves the simulative calculation accuracy of a power grid and ensures that the power grid can safely, reliably and economically operate.

Description

A kind of on-line load simulation method of power system

Technical field

The invention belongs to the power system measuring technical field, be specifically related to a kind of on-line load simulation method of power system.

Background technology

The Mathematical Modeling of each element of electric power system is the basis that network analysis is calculated, the whether accurate quality that depends on model of analysis result.Detail mathematic model and modeling technique about generator, excitation system, governing system, transformer, transmission line in the power system analysis have obtained good development.But for a long time, as the load of one of electric power system critical elements, because factors such as its complexity, distributivity, time variation and randomness have determined the difficulty that its Mathematical Modeling is set up.

The researcher has proposed many load simulation methods.The basic thought of wherein adding up synthesis is the Mathematical Modeling that at first obtains various typical load elements (as: fluorescent lamp, home electronics, industrial electro motivation, air conditioner load etc.) by test and mathematical derivation, on some load point, add up certain composition of various loads constantly then, be every kind of percentage that typical load is shared, and the data of distribution line, transformer and reactive power compensation, last comprehensive these data draw the load model of this load point.The advantage of statistics synthesis is clear, the definite conception of physical model, is convenient to qualitative understanding part throttle characteristics.But also have certain shortcoming, comprising: (1) inquiry agency gets load capacity and actual load power and inconsistent because there is the problem of simultaneity factor, be not all devices all be to come into operation in 24 hours, therefore, need carry out investigation statistics at times; (2) As time goes on, actual load power, load structure and network configuration all may change, if load is carried out just wanting to put things right once and for all after the investigation statistics modeling work, are difficult to reach accuracy requirement.Certainly, the variation of real system structure and load structure also is progressively, slowly, therefore, needs the online various data of calling, and carries out the regular update of load model parameters, reaches the purpose of the long-acting dynamic management of load modeling; (3) investigation work needs thousands of users' of statistics load to form and parameter, and workload is huge, and is difficult to obtain statistics accurately.In China, the load model structure that the management and running department of each regional power grid of State Grid Corporation of China adopts always as Fig. 2 a) shown in: under the 220kV bus, hang the 220kV main transformer, the 110kV bus bar side that hangs over main transformer in parallel with static load again after induction-motor load and distribution equivalent impedance are together in series.This model structure is the same substantially with the transition model structure that WECC is adopted.China Electric Power Research Institutes in 2004 have proposed to consider integrated load model (the Synthesis Load Model of distribution network, SLM) structure, as Fig. 2 b) shown in: the 220kV main transformer under the 220kV bus, hung, connecting the distribution equivalent impedance in the 110kV of main transformer bus bar side, the little unit parallel connection of the reactive power compensation of induction-motor load, static load, distribution system and distribution system and is hung over below the distribution equivalent impedance.The characteristics of this model structure are as follows: 1. static load and induction-motor load can be considered the influence of distribution system impedance.Distribution system is adopted the impedance analogy method, guaranteed the relation of more realistic distribution system of model structure and power load.Can adopt suitable equivalence method, obtain distribution system equivalent impedance [16] more accurately.2. simulated the reactive power compensation of distribution system.Distribution network and power consumer have all disposed a large amount of reactive power compensators, and its dynamic characteristic has material impact to the stability of system, should simulate in detail, and SLM provides effective analogy method for the model of distribution system reactive power compensation.3. can consider the little unit of distribution system easily.In electric system simulation analysis (particularly upset test and accident analysis), need to consider to insert the little unit of distribution network sometimes, comprised little unit in the collective model structure, can make the simulation of little unit convenient.4. the static reactive load constant current and the firm power that can not occur bearing.The introducing of load power factor has guaranteed constant current and negative firm power that static load reactive component can not occur bearing, makes model more realistic.

The validity of this model structure has obtained checking by the big disturbance experiments in northeast of 2004 and 2005.The present invention will invent a kind of online SLM analogy method based on the SLM structure.

Chinese patent application 200810156278.2 has proposed a kind of high and low pressure side power load analogy method based on single side sampling, and step comprises: one, install and measure device in corresponding main transformer one side of loading electric substation; Two, gather corresponding electric parameters, and be converted to the desired data format of further analytical calculation; Three, online or processed offline data are carried out identification of Model Parameters, obtain the load model parameters of a side; Four, convert in conjunction with other operational factor of electric power system that is easier to obtain, obtain the load model and the parameter of an other side; Five, obtain the load model of both sides at last.Advantage: under the identical situation of model structure, only need to gather the electric parameters (noisy data) of main transformer one side, partial parameters is carried out the load model that suitable adjustment can obtain opposite side.The difficulty of this method maximum is to be difficult in system all transformer stations relative assembly all is installed, and secondly is to adopt total body examination to distinguish method identification load model parameters, and parameter has multi-solution and uncertainty, may not be inconsistent with actual conditions.The load model that the method that this paper proposes not only obtains has clear physics conception, be easy to the advantage understood by the Field Force, and all electrical networks all have EMS (EMS) and load control system have been installed at present, therefrom can obtain actual power information, need other device be installed in system again, overcome the shortcoming of the method for patent application 200810156278.2 propositions.In addition, this method has carried out simplifying to investigation statistics, has reduced the workload of investigation statistics.

Summary of the invention

The present invention is based on the user side power information of electric power system marketing department load control system collection and transformer station's power information that EMS (EMS) provides, in conjunction with sampling manual research statistics, and utilize the power supply area network topology data of the load bus that traffic department provides, adopt the statistics synthesis, realize the online load modeling of load bus, to reach is the purpose of load bus modeling exactly, improve the accuracy that grid simulation calculates, ensure power grid security, reliably, operation economically.

Therefore the present invention proposes a kind of on-line load simulation method of power system, it is characterized in that this method at first carries out the offline user sample investigation, analysis and arrangement obtain various typical 10kV or 6kV outlet the power load equipment ratio of confession; Utilize the user side power information of electric power system marketing department load control system collection and transformer station's power information that EMS EMS provides then, with every the 10kV of load bus that obtains to study or the online actual power of 6kV outlet; With the various typical 10kV of the power data of online collection and collected offline or 6kV outlet the power load equipment ratio of confession carry out COMPREHENSIVE CALCULATING, draw the composition of each various power load equipment of the moment of this load point, i.e. the percentage of every kind of shared total amount of power load equipment; Determine the part throttle characteristics parameter of each power load equipment again according to dynamic analog test or representative value, calculate static load constitute in the ratio of constant-impedance, constant current, permanent power, and adopt the polymerization of an induction electric group of planes to calculate equivalent dynamic load parameter; Utilize the power supply area network topology data of the load bus that traffic department provides at last: the data that comprise distribution line, transformer and reactive power compensation, adopt equivalent these data computation of algorithm synthesis of power distribution network to draw the load model of this load point, thereby realize the online load modeling of load bus.

Wherein, this method specifically may further comprise the steps:

(1) determines the part throttle characteristics parameter of each power consumption equipment according to dynamic analog test or representative value;

(2) traffic department provides related datas such as the transformer platform number of power supply area network topological diagram, 220kV transformer station and the 110kV of subordinate of load bus and 35kV transformer station and capacity, reactive compensation capacity, line impedance;

(3) determine the load structure situation of 220kV transformer station and 110kV of subordinate and 35kV transformer station according to the user side power information data of load control system collection of marketing;

(4) according to the load structure situation of each load bus, determine each load bus the load type of confession;

(5) off-line carries out user's sample investigation, each load type transformer station is chosen some typical 10kV or 6kV outlet carry out the part throttle characteristics investigation, analysis and arrangement obtains each load type: comprise that the 10kV of industry, commerce, resident or agriculture load type or the power consumption equipment of 6kV outlet constitute situation: comprise power consumption equipment type i and each power consumption equipment proportion ρ _i

(6) relevant information of online harvest energy management system EMS is according to the voltage V of data acquisition 220kV/110kV/35kV every 10kV of main transformer side or 6kV circuit j among the EMS EMS _j, electric current I _j, power factor PF _j, calculate and obtain this circuit 10kV side outlet power P _j=V _jI _jPF _j

(7) constitute situation for j sample investigation 10kV outlet institute electricity supply and use equipment in the k type load that comprises industrial class, commercial, resident's class, agriculture, it is promoted the use of similar 10kV outlet, statistics draws the gross power of these all kinds of power consumption equipments of period, divided by the gross power of this whole load bus of period, can obtain the ratio k of whole 220kV of this period all kinds of power consumption equipment i of transformer station _i: If always total n bar 10kV of this 220kV transformer station or 6kV outlet;

(8), the load model of each power consumption equipment is decomposed into static load model and dynamic load model two parts according to the part throttle characteristics parameter of each power consumption equipment;

(9) utilize the static load equivalence method, calculate the static load model parameter of the integrated load model of considering distribution network;

(10) utilize the dynamic load equivalence method, calculate the dynamic load model parameter of the integrated load model of considering distribution network;

(11) utilize the distribution network impedance equivalence method, calculate the distribution network equivalent impedance of the integrated load model of considering distribution network;

(12) form 220kV transformer station power supply area network original system (detailed system, comprise 110kV, 35kV circuit, transformer, reactive power compensation, each node with all types of payload and model parameter etc.) trend stablize data, carry out trend and stability Calculation, form result of calculation file 1;

(13) trend of the valve systems such as integrated load model of the consideration distribution network of formation 220kV transformer station power supply area network is stablized data, carries out trend and stability Calculation, forms result of calculation file 2;

(14) validity of verifying with the actual distribution network dynamic response characteristics of valve system SLM modeling such as integrated load model of considering distribution network by comparing result file 1 and destination file 2.

Wherein, described load bus is a 220kV transformer station.

Wherein, described step (9) static load equivalence method comprises:

The static load model structure that IEEE Taskforce recommend to adopt is to be the multinomial load model (Polynomial Load Model) of polynomial equation form with the relationship description between load power and the voltage, the general type of this model suc as formula 1 and formula 2 shown in:

P=P _o[a * (V/V _o) ²+ b * (V/V _o)+c] (formula 1)

Q=Q _o[α * (V/V _o) ²+ β * (V/V _o)+γ] (formula 2)

Multinomial active power load model coefficient is a, b, c, reactive power load model coefficient is the power factor (PF) of α, β, γ and load, this load model is called as " ZIP " model, because it has comprised constant-impedance (Z), constant current (I) and permanent power (P), this model is used to describe specific load equipment or load cell, V _oThe rated voltage of expression load, P _oAnd Q _oThen be illustrated respectively in rated voltage V _oThe specified active power and the reactive power of following load, if when describing the synthetic load of bus with this model, V _o, P _oAnd Q _oBe commonly used to represent the numerical value under system's initial launch operating mode;

Equivalence to static load mainly is to FACTOR P _o, a, b, c and Q _o, α, β, γ equivalence, the equivalence of multinomial load model is based on the sensitivity of load power to load side voltage, promptly

${\frac{\&PartialD;P}{\&PartialD;V}|}_{V=\mathrm{Vo}}={\frac{\&PartialD;{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">P</figure-callout>}}_1}{\&PartialD;V}|}_{V=\mathrm{Vo}}+{\frac{\&PartialD;{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">P</figure-callout>}}_2}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+......+\frac{\&PartialD;P_n}{\&PartialD;V}|}_{V=\mathrm{Vo}}$ (formula 3)

${\frac{\&PartialD;Q}{\&PartialD;V}|}_{V=\mathrm{Vo}}={\frac{\&PartialD;{\mathrm{<figure-callout\; id="1"\; label="Q"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_1}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+\frac{\&PartialD;{\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">Q</figure-callout>}}_2}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+......+\frac{\&PartialD;Q_n}{\&PartialD;V}|}_{V=\mathrm{Vo}}$ (formula 4)

P ₁, P ₂P _nAnd Q ₁, Q ₂Q _nBe the active power and the reactive power of each static load, corresponding multinomial load model coefficient is respectively P _O1P _On, a ₁A _n, b ₁B _n, c ₁C _nAnd Q _O1Q _On, α ₁α _n, β ₁β _n, γ ₁γ _nWork as V=V _oIn time, have:

$\left.\begin{array}{c}{P}_o=P_{o1}+P_{o2}+...+P_{\mathrm{on}}\\ Q_o=Q_{o1}+Q_{o2}+...+Q_{\mathrm{on}}\end{array}\right\}$ (formula 5)

$\left.\begin{array}{c}a=\frac{P_{o1}{a}_1+P_{o2}{a}_2+......+P_{\mathrm{on}}{a}_n}{P_o}\\ b=\frac{P_{o1}{b}_1+P_{o2}{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">b</figure-callout>}}_2+......+P_{\mathrm{on}}{b}_n}{P_o}\\ c=1-a-b\end{array}\right\}$ (formula 6)

$\left.\begin{array}{c}\mathrm{\&alpha;}=\frac{Q_{o1}{\mathrm{\&alpha;}}_1+Q_{o2}{\mathrm{\&alpha;}}_2+......+Q_{\mathrm{on}}{\mathrm{\&alpha;}}_n}{Q_o}\\ \mathrm{\&beta;}=\frac{Q_{o1}{\mathrm{\&beta;}}_1+Q_{o2}{\mathrm{\&beta;}}_2+......+Q_{\mathrm{on}}{\mathrm{\&beta;}}_n}{Q_o}\\ \mathrm{\&gamma;}=1-\mathrm{\&alpha;}-\mathrm{\&beta;}\end{array}\right\}$ (formula 7)

Wherein, described step (10) dynamic load equivalence method comprises:

The rated slip Sn of the specified electromagnetic power Pemn of motor or nominal torque Temn, rotor and maximum electromagnetic torque Pem_max or breakdown torque multiple κ _mIt is the most important parameter that to represent mechanical property in the motor, the basic principle of this equivalence method is that new Equivalent Model must keep total specified active power ∑ Pn of original system absorption constant, total reactive power ∑ Qn or power factor (PF) Pf are constant, total electromagnetic power ∑ Pemn is constant, total rotor winding copper loss ∑ Pcu2 is constant, total constant and total kinetic energy ∑ Eenergy of maximum electromagnetic power ∑ Pem_max remains unchanged, can also obtain total stator winding copper loss ∑ Pcu1 according to this tittle, the rated slip Sn and the equivalent inertia time constant H of equivalent motor are calculated as follows:

∑ P _Cu1=∑ P _n-∑ P _Emn(formula 8)

S _n=∑ P _Cu2/ ∑ R _Emn(formula 9)

H=∑ E _Energy/ (∑ P _Emn-∑ P _Cu2) (formula 10)

∑ P wherein _Emn-∑ P _Cu2Be exactly the specified mechanical output of equivalent motor output, remain unchanged;

Calculate the electric parameter of equivalent motor model then according to these amounts of having tried to achieve, comprise stator resistance R _s, stator leakage reactance X _s, rotor resistance R _r, rotor leakage reactance X _rWith excitatory reactance X _m, establishing specified phase voltage is Un, the CALCULATION OF PARAMETERS flow process is as follows:

(1), calculates ∑ Pcu1, Sn and H according to formula 8 to formula 10 then, and make P according to known calculation of parameter ∑ Pn, ∑ Qn, ∑ Pemn, ∑ Pcu2, ∑ Pem_max and ∑ Eenergy _{Emt_max}=∑ P _{Em_max}

(2) establishing total stator phase current is

, then have

${\stackrel{\&CenterDot;}{I}}_n=\frac{\mathrm{\&Sigma;}{P}_n-\mathrm{j\&Sigma;}{Q}_n}{3U_n}$ (formula 11)

Then

$R_s=\frac{\mathrm{\&Sigma;}{P}_{\mathrm{cu}1}}{3{I_n}^2}$ (formula 12)

(3) according to ∑ Pn, ∑ Qn and Un ask the equivalent impedance Z of equivalent motor by following formula _Deq:

$Z_{\mathrm{deq}}=\frac{3{{\mathrm{<figure-callout\; id="2"\; label="U\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">U</figure-callout>}}_n}^2}{\mathrm{\&Sigma;}{P}_n-\mathrm{j\&Sigma;}{Q}_n}$

R _Deq=real (Z _Deq) (formula 13)

X _deq＝imag(Z _deq)

(4) formula of reduction by maximum electromagnetic power calculates X _sAnd X _r:

$X=\sqrt{{(\frac{3{{\mathrm{<figure-callout\; id="2"\; label="U\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">U</figure-callout>}}_n}^2}{2P_{\mathrm{emt}\_\max }}-R_s)}^2-{{\mathrm{<figure-callout\; id="2"\; label="R\; s"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">R</figure-callout>}}_s}^2}$

$X_s=\frac{X}{2}$ (formula 14)

X _r＝X _s

In this algorithm, always suppose X _r=X _s, and the X that calculates according to this formula _sAnd X _rInevitable less than normal, because the maximum electromagnetic power that calculates according to the maximum electromagnetic power formula of simplifying is bigger than the maximum electromagnetic power of reality, so need be by alternative manner to X _sAnd X _rRevise;

(5) according to the R that tries to achieve _s, X _s, X _rAnd equivalent impedance Z _Deq=R _Deq+ jX _DeqAsk R _rAnd X _m, order

K _r=R _Deq-R _s(formula 15)

K _x＝X _deq-X _s

$R_r=\frac{(K_r+{K_x}^2/K_r-\sqrt{{(K_r+{K_x}^2/K_r)}^2-4{X_s}^2})S_n}{2}$ (formula 16)

$X_m=\frac{K_r{X}_s+K_x\frac{R_r}{S_n}}{\frac{R_r}{S_n}-K_r}$

This calculating R _rAnd X _mMethod can guarantee P all the time _Em=∑ P _EmSet up;

(6) according to the R that tries to achieve _s, X _s, R _r, X _rAnd X _m, recomputate maximum electromagnetic power according to formula of reduction calculating:

$P_{\mathrm{emt}\_\max i}=\frac{3{{\mathrm{<figure-callout\; id="2"\; label="U\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">U</figure-callout>}}_n}^2}{2(R_s+\sqrt{{R_s}^2+X^2})}$ (formula 17)

(7) calculate maximum electromagnetic power actual under the new argument according to the Dai Weinan equivalent circuit:

The Dai Weinan equivalent impedance is:

$Z_{\mathrm{dp}}=j{X}_r+\frac{j{X}_m(R_s+j{X}_s)}{R_s+j(X_s+X_m)}$

R _Dp=real (Z _Dp) (formula 18)

X _dp＝imag(Z _dp)

The condition that produces maximum electromagnetic power is:

$R_{\mathrm{pm}}=\frac{R_r}{S_m}=\sqrt{{{\mathrm{<figure-callout\; id="2"\; label="R\; dp"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{dp}}}^2+{X_{\mathrm{dp}}}^2}$ (formula 19)

S _mBe critical slippage, the open circuit voltage of Dai Weinan equivalent circuit is:

${\stackrel{\&CenterDot;}{U}}_o=U_n\frac{j{X}_m}{R_s+j(X_s+X_m)}$ (formula 20)

Therefore, can recomputate the actual maximum electromagnetic torque of new argument correspondence according to following formula:

$P_{\mathrm{em}\_\max i}=\frac{3{U_o}^2{R}_{\mathrm{pm}}}{{(R_{\mathrm{dp}}+R_{\mathrm{pm}})}^2+{X_{\mathrm{dp}}}^2}$ (formula 21)

(8) calculate P _{Emt_maxi}With P _{Em_maxi}Ratio, revise P _{Emt_max}

$k_{\max i}=\frac{P_{\mathrm{emt}\_\max i}}{P_{\mathrm{em}\_\max i}}$

P _{Emt_max}=k _MaxiP _{Em_max}(formula 22)

(9) compare P _{Em_maxi}With P _{Em_max}

Err _{Pem_max}=| P _{Em_max}-P _{Em_maxi}| (formula 23)

If Err _{Pem_max}〉=1.0e ^-5, then returned for (4) step and recomputate, finish otherwise calculate.

Wherein, described step (11) is utilized the distribution network impedance equivalence method, and the distribution network equivalent impedance that calculates the integrated load model of considering distribution network comprises:

Equate with each transformer of power distribution network, each distribution line consumed power sum according to distribution network system impedance consumed power, can the computing system resistance value be

$Z_{\mathrm{ep}}=\left[\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{u}_j^2{(1/Z_j)}^{*}\right]/{\left(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{I}_i\right)}^2$ (formula 24)

In the formula: Z _EqThe impedance of expression distribution network system; u _jThe expression busbar voltage, Z _jIndication transformer and distribution line impedance; I _iThe expression load current.

The invention has the beneficial effects as follows:

1. transformer station's power information of providing of this method user side power information of gathering based on electric power system marketing department load control system and EMS (EMS), in conjunction with sampling manual research statistics, and utilize the power supply area network topology data of the load bus that traffic department provides, adopt the statistics synthesis, realize the online load modeling of load bus, to reach is the purpose of load bus modeling exactly, improve the accuracy that grid simulation calculates, ensure power grid security, reliably, operation economically.

2. this method is based on the real time data of actual electric network, having carried out online load modeling, changed the pattern of traditional dependence off-line manual research, is the further lifting to traditional approach, for load modeling work provides the effect of important guidance, help to improve the accuracy of load modeling.

3. this method has carried out simplifying to investigation statistics, has reduced the workload of investigation statistics.

Description of drawings

Fig. 1 is the design flow diagram of online load simulation method of the present invention;

Fig. 2 is the structure of existing load model a), Fig. 2 b) be the equivalent circuit of considering the integrated load model of distribution network;

Fig. 3 is an actual load regional distribution network network structure chart;

Fig. 4 is that the distribution network node injects the trend schematic diagram;

Fig. 5 is the power distribution network equivalent circuit;

Fig. 6 is the equivalent process of power distribution network.

Fig. 7 is the contrast of 110kV busbar voltage curve a), and wherein solid line "-" is for adopting the simulation result of actual distribution network analogue system, the simulation result of dotted line during for the equivalent SLM model parameter that adopts that the present invention draws;

Fig. 7 b) be the contrast of load active power curve, wherein solid line "-" is for adopting the simulation result of actual distribution network analogue system, the simulation result of dotted line during for the equivalent SLM model parameter that adopts that the present invention draws;

Fig. 7 c) be the contrast of reactive load power curve, wherein solid line "-" is for adopting the simulation result of actual distribution network analogue system, the simulation result of dotted line during for the equivalent SLM model parameter that adopts that the present invention draws.

Embodiment

The basic thought of online statistics synthesis is at first to obtain various typical load element (as: fluorescent lamps by test and mathematical derivation, home electronics, the industrial electro motivation, air conditioner load etc.) Mathematical Modeling, utilize the user side power information of electric power system marketing department load control system collection and the online power information of transformer station that EMS (EMS) provides then, in conjunction with sampling manual research statistics, the composition of certain various load of the moment of statistics on some load point, be every kind of percentage that typical load is shared, and distribution line, the data of transformer and reactive power compensation, last comprehensive these data draw the load model of this load point.

Technical scheme of the present invention mainly comprises four parts, the value calculating methods such as value calculating methods such as constituting value calculating method, static loads such as calculating, dynamic load with electric device and distribution network impedance of promptly loading.

1, load cell constitutes calculating section

Off-line carries out user's sample investigation, each load type transformer station is chosen some typical 10kV or 6kV outlet carry out the part throttle characteristics investigation, the power consumption equipment that analysis and arrangement obtains each load type (industry, commercial, resident or agriculture load type) 10kV or 6kV outlet constitutes situation: comprise power consumption equipment type i and each power consumption equipment proportion ρ _i

Obtain the voltage V of 220kV/110kV/35kV every 10kV of main transformer side or 6kV circuit j according to data among the EMS _j, electric current I _j, power factor PF _j, calculate and obtain this circuit 10kV side outlet power P _j=V _jI _jPF _jJ sample investigation 10kV outlet institute electricity supply and use equipment formation situation in k class (industrial class, commercial, resident's class, the agriculture) load promotes the use of similar 10kV outlet with it.Statistics draws all kinds of power consumption equipment gross powers of this period, divided by the gross power of this whole load bus of period, can obtain the ratio of whole 220kV of this period all kinds of power consumption equipment i of transformer station

(might as well establish always total n bar 10kV of this 220kV transformer station or 6kV outlet).

2, the equivalent calculating section of static load model

The static load model structure that IEEE Taskforce recommend to adopt is to be the multinomial load model (Polynomial Load Model) of polynomial equation form with the relationship description between load power and the voltage, the general type of this model suc as formula 1 and formula 2 shown in.

P=P _o[a * (V/V _o) ²+ b * (V/V _o)+c] (formula 1)

Q=Q _o[α * (V/V _o) ²+ β * (V/V _o)+γ] (formula 2)

Multinomial active power load model coefficient is a, b, c, and reactive power load model coefficient is the power factor (PF) of α, β, γ and load.This load model is called as " ZIP " model sometimes, because it has comprised constant-impedance (Z), constant current (I) and permanent power (P).This model is used to describe specific load equipment or load cell, V _oThe rated voltage of expression load, P _oAnd Q _oThen be illustrated respectively in rated voltage V _oThe specified active power and the reactive power of following load.If but V when describing the synthetic load of bus with this model _o, P _oAnd Q _oBe commonly used to represent the numerical value under system's initial launch operating mode.

Equivalence to static load mainly is to FACTOR P _o, a, b, c and Q _o, α, β, γ equivalence.Equivalence to the multinomial load model is based on the sensitivity of load power to load side voltage, promptly

${\frac{\&PartialD;P}{\&PartialD;V}|}_{V=\mathrm{Vo}}={\frac{\&PartialD;{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">P</figure-callout>}}_1}{\&PartialD;V}|}_{V=\mathrm{Vo}}+{\frac{\&PartialD;{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">P</figure-callout>}}_2}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+......+\frac{\&PartialD;P_n}{\&PartialD;V}|}_{V=\mathrm{Vo}}$ (formula 3)

${\frac{\&PartialD;Q}{\&PartialD;V}|}_{V=\mathrm{Vo}}={\frac{\&PartialD;{\mathrm{<figure-callout\; id="1"\; label="Q"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_1}{\&PartialD;V}|}_{V=\mathrm{Vo}}+{\frac{\&PartialD;{\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">Q</figure-callout>}}_2}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+......+\frac{\&PartialD;Q_n}{\&PartialD;V}|}_{V=\mathrm{Vo}}$ (formula 4)

P ₁, P ₂P _nAnd Q ₁, Q ₂Q _nBe the active power and the reactive power of each static load, corresponding multinomial load model coefficient is respectively P _O1P _On, a ₁A _n, b ₁B _n, c ₁C _nAnd Q _O1Q _On, α ₁α _n, β ₁β _n, γ ₁γ _nWork as V=V _oIn time, have

$\left.\begin{array}{c}{P}_o={\mathrm{<figure-callout\; id="1"\; label="P\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">P</figure-callout>}}_o+{\mathrm{<figure-callout\; id="2"\; label="P\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">P</figure-callout>}}_o+...+P_{\mathrm{on}}\\ Q_o={\mathrm{<figure-callout\; id="1"\; label="Q\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_o+{\mathrm{<figure-callout\; id="2"\; label="Q\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">Q</figure-callout>}}_o+...+Q_{\mathrm{on}}\end{array}\right\}$ (formula 5)

$\left.\begin{array}{c}a=\frac{P_{o1}{a}_1+P_{o2}{a}_2+......+P_{\mathrm{on}}{a}_n}{P_o}\\ b=\frac{P_{o1}{\mathrm{<figure-callout\; id="1"\; label="b"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">b</figure-callout>}}_1+P_{o2}{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">b</figure-callout>}}_2+......+P_{\mathrm{on}}{b}_n}{P_o}\\ c=1-a-b\end{array}\right\}$ (formula 6)

$\left.\begin{array}{c}\mathrm{\&alpha;}=\frac{{\mathrm{<figure-callout\; id="1"\; label="Q\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_o{\mathrm{\&alpha;}}_1+{\mathrm{<figure-callout\; id="2"\; label="Q\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">Q</figure-callout>}}_o{\mathrm{\&alpha;}}_2+......+Q_{\mathrm{on}}{\mathrm{\&alpha;}}_n}{Q_o}\\ \mathrm{\&beta;}=\frac{{\mathrm{<figure-callout\; id="1"\; label="Q\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_o{\mathrm{\&beta;}}_1+{\mathrm{<figure-callout\; id="2"\; label="Q\; o"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">Q</figure-callout>}}_o{\mathrm{\&beta;}}_2+......+Q_{\mathrm{on}}{\mathrm{\&beta;}}_n}{Q_o}\\ \mathrm{\&gamma;}=1-\mathrm{\&alpha;}-\mathrm{\&beta;}\end{array}\right\}$ (formula 7)

3, the equivalent calculating section of dynamic load model

The rated slip Sn of the specified electromagnetic power Pemn of motor or nominal torque Temn, rotor and maximum electromagnetic torque Pem_max or breakdown torque multiple κ _mIt is the several most important parameter that to represent mechanical property in the motor.Based on this, the present invention adopts a kind of dynamic load model equivalence method with clear and definite Physical Mechanism.

The basic principle of equivalence method is that new Equivalent Model must keep total specified active power ∑ Pn of original system absorption constant, total reactive power ∑ Qn (or power factor (PF) Pf) is constant, total electromagnetic power ∑ Pemn (torque) is constant, total rotor winding copper loss ∑ Pcu2 is constant, and the constant and total kinetic energy ∑ Eenergy of total maximum electromagnetic power ∑ Pem_max (torque) remains unchanged.Can also obtain total stator winding copper loss ∑ Pcu1 according to this tittle, the rated slip Sn of equivalent motor and equivalent inertia time constant H are calculated as follows:

∑ P _Cu1=∑ P _n-∑ P _Emn(formula 8)

S _n=∑ P _Cu2/ ∑ R _Emn(formula 9)

H=∑ E _Energy/ (∑ P _Emn-∑ P _Cu2) (formula 10)

∑ P wherein _Emn-∑ P _Cu2Be exactly the specified mechanical output of equivalent motor output, also remain unchanged.

Calculate the electric parameter of equivalent motor model then according to these amounts of having tried to achieve, comprise stator resistance R _s, stator leakage reactance X _s, rotor resistance R _r, rotor leakage reactance X _rWith excitatory reactance X _mIf specified phase voltage is Un, the CALCULATION OF PARAMETERS flow process is as follows:

(1), calculates ∑ Pcu1, Sn and H according to formula 8 to formula 10 then, and make P according to known calculation of parameter ∑ Pn, ∑ Qn, ∑ Pemn, ∑ Pcu2, ∑ Pem_max and ∑ Eenergy _{Emt_max}=∑ P _{Em_max}

(2) establishing total stator phase current is

Then have

${\stackrel{\&CenterDot;}{I}}_n=\frac{\mathrm{\&Sigma;}{P}_n-\mathrm{j\&Sigma;}{Q}_n}{3U_n}$ (formula 11)

Then

$R_s=\frac{\mathrm{\&Sigma;}{\mathrm{<figure-callout\; id="1"\; label="P\; cu"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">P</figure-callout>}}_{\mathrm{cu}}}{3{I_n}^2}$ (formula 12)

(3) according to ∑ Pn, ∑ Qn and Un ask the equivalent impedance Z of equivalent motor by following formula _Deq:

$Z_{\mathrm{deq}}=\frac{3{{\mathrm{<figure-callout\; id="2"\; label="U\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">U</figure-callout>}}_n}^2}{\mathrm{\&Sigma;}{P}_n-\mathrm{j\&Sigma;}{Q}_n}$

R _Deq=real (Z _Deq) (formula 13)

X _deq＝imag(Z _deq)

(4) formula of reduction by maximum electromagnetic power calculates X _sAnd X _r:

$X=\sqrt{{(\frac{3{{\mathrm{<figure-callout\; id="2"\; label="U\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">U</figure-callout>}}_n}^2}{2P_{\mathrm{emt}\_\max }}-R_s)}^2-{{\mathrm{<figure-callout\; id="2"\; label="R\; s"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">R</figure-callout>}}_s}^2}$

$X_s=\frac{X}{2}$ (formula 14)

X _r＝X _s

In this algorithm, always suppose X _r=X _s, and the X that calculates according to this formula _sAnd X _rInevitable less than normal, because the maximum electromagnetic power that calculates according to the maximum electromagnetic power formula of simplifying is bigger than the maximum electromagnetic power of reality.So need be by alternative manner to X _sAnd X _rRevise.

(5) according to the R that tries to achieve _s, X _s, X _rAnd equivalent impedance Z _Deq=R _Deq+ jX _DeqAsk R _rAnd X _m, order

K _r＝R _deq-R _s

K _x=X _Deq-X _s(formula 15)

$R_r=\frac{(K_r+{K_x}^2/K_r-\sqrt{{(K_r+{K_x}^2/K_r)}^2-4{X_s}^2})S_n}{2}$ (formula 16)

$X_m=\frac{K_r{X}_s+K_x\frac{R_r}{S_n}}{\frac{R_r}{S_n}-K_r}$

This calculating R _rAnd X _mMethod can guarantee P all the time _Em=∑ P _EmSet up.

(6) according to the R that tries to achieve _s, X _s, R _r, X _rAnd X _m, recomputate maximum electromagnetic power according to formula of reduction calculating:

$P_{\mathrm{emt}\_\max i}=\frac{3{{\mathrm{<figure-callout\; id="2"\; label="U\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">U</figure-callout>}}_n}^2}{2(R_s+\sqrt{{R_s}^2+X^2})}$ (formula 17)

(7) calculate maximum electromagnetic power actual under the new argument according to the Dai Weinan equivalent circuit:

The Dai Weinan equivalent impedance is

$Z_{\mathrm{dp}}=j{X}_r+\frac{j{X}_m(R_s+j{X}_s)}{R_s+j(X_s+X_m)}$

R _Dp=real (Z _Dp) (formula 18)

X _dp＝imag(Z _dp)

The condition that produces maximum electromagnetic power is

$R_{\mathrm{pm}}=\frac{R_r}{S_m}=\sqrt{{{\mathrm{<figure-callout\; id="2"\; label="R\; dp"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{dp}}}^2+{X_{\mathrm{dp}}}^2}$ (formula 19)

S _mBe critical slippage, the open circuit voltage of Dai Weinan equivalent circuit is

${\stackrel{\&CenterDot;}{U}}_o=U_n\frac{j{X}_m}{R_s+j(X_s+X_m)}$ (formula 20)

Therefore, can recomputate the actual maximum electromagnetic torque of new argument correspondence according to following formula

$P_{\mathrm{em}\_\max i}=\frac{3{U_o}^2{R}_{\mathrm{pm}}}{{(R_{\mathrm{dp}}+R_{\mathrm{pm}})}^2+{X_{\mathrm{dp}}}^2}$ (formula 21)

(8) calculate P _{Emt_maxi}With P _{Em_maxi}Ratio, revise P _{Emt_max}

$k_{\max i}=\frac{P_{\mathrm{emt}\_\max i}}{P_{\mathrm{em}\_\max i}}$

P _{Emt max}=k _MaxiP _{Em_max}(formula 22)

(9) compare P _{Em_maxi}With P _{Em_max}

Err _{Pem_max}=| P _{Em_max}-P _{Em_maxi}| (formula 23)

If Err _{Pem_max}〉=1.0e ^-5, then returned for (4) step and recomputate, finish otherwise calculate.

4, consider the distribution network equivalent impedance calculating section of the integrated load model of distribution network

Any load can think to regulate automatically according to the variation of line voltage a kind of dynamic load of himself watt level, and difference only is that the complexity of the relation between various load powers and the change in voltage is different.Usually said static load model as constant-impedance, permanent power and constant current load model, has very simple and clear variation relation between the power of these models and the voltage, break away from the physical essence of load, and the power of load only is decided by the size of external voltage.And dynamic load, equipment as asynchronous motor load, thermostatic equipment and charged power electronic regulating unit, their power absorbed are not only relevant with the variation of line voltage, but also relate to time variable, has more complicated variation relation, especially asynchronous motor has complicated mechanical-electro-magnetic transient variation characteristic.

Therefore, any load is for realizing the active power balance of I/O, all the variation of with good grounds voltage and the intrinsic characteristic of regulating self-admittance automatically.So any load can be regarded the negative current source that is subjected to line voltage control as.

Load in the real system is not directly to be connected on the same bus simultaneously, but according to location distribution in electric power system, as shown in Figure 3: bus 1 is by a transmission line L ₁Connecting bus 2, bus 2 has 3 outlets, wherein transmission line L ₂And L ₄Directly be dynamic load M respectively ₁And M ₂Power supply, and transmission line L ₃Connecting transformer station of subordinate below, the low-pressure side bus of this transformer station has 3 outlets, wherein transmission line L again ₅Hang dynamic load M below ₃, transmission line L ₆Hang static load (comprising constant-impedance load, constant current load and permanent power load) below, transmission line L ₇Hang other load below.For load area, each load mainly is coupled by distribution network.So the node that in the load area network, has some directly not link to each other with load bus, thereby the injection current of these nodes always equals zero, this category node is called the contact node.

When carrying out electric calculating, static cell in the distribution network such as distribution transformer, distribution line, parallel capacitor etc. can be simulated with the equivalent circuit that R, L, C formed, or even some static loads also can replace with them or its combinational circuit.Therefore, the distribution network that has these static cells to be linked to be can be described with corresponding admittance matrix or impedance matrix, as shown in Figure 4.Load, especially induction-motor load as the main dynamic element of distribution network are coupled by these static cells.

Accompanying drawing 4 has been described the relation between actual distribution network (as shown in Figure 3) node injecting power, voltage, electric current and the network.Wherein the power of the injection of network represented in the positive sign of power and electric current among the figure, the power of supplying with to power distribution network as system and the compensation power of reactive-load compensation equipment etc.; Negative sign is then represented to load from the electrical network power absorbed or from the electric current of network flow to load.The linear network part that empty frame surrounds among the figure, the relation between its node current and the voltage can be described by modal equation:

I=YU (formula 24)

Wherein I is the node injection current, and U is a node voltage, and Y is an admittance matrix.Pass between node injecting power and the electric current is:

$I_i=\frac{P_i-j{Q}_i}{U_i^{*}}$ (formula 25)

P _iAnd Q _iBe respectively active power and reactive power that node i is injected to network, when node i is load bus, as network injecting power P _iAnd Q _iBefore should add negative sign; U _i ^*Conjugate for the node i voltage vector.

The purpose of load modeling is to provide load model more accurately for system stability analysis, and farthest reduce the complexity of load model, could in stability analysis, can guarantee certain accuracy like this, not increase extra computation burden again to stability analysis.Therefore, most realistic way is that all loads in low-voltage load zone are equivalent to same bus.But all be coupled between each load of actual load zone by power distribution network.Distribution network is to the interaction response characteristic important influence between load and the main system.So, in order to take into account the influence of distribution network, be an impedance with whole distribution equivalence, with its be connected in system and etc. between the duty value bus.

The modal equation of establishing electric network is

(formula 26)

$A=\left[\begin{array}{cc}{Y}_{11}& Y_{12}\\ Y_{21}& Y_{22}\end{array}\right]B=\left[\begin{array}{cccc}{Y}_{13}& Y_{14}& \&CenterDot;\&CenterDot;\&CenterDot;& Y_{1n}\\ Y_{23}& Y_{24}& \&CenterDot;\&CenterDot;\&CenterDot;& Y_{2n}\end{array}\right]$ (formula 27)

$C=\left[\begin{array}{ccc}{Y}_{31}& & Y_{32}\\ Y_{41}& & Y_{42}\\ & \&CenterDot;\&CenterDot;\&CenterDot;& \\ Y_{n1}& & Y_{n2}\end{array}\right]D=\left[\begin{array}{cccccc}{Y}_{33}& & Y_{34}& \&CenterDot;\&CenterDot;\&CenterDot;& & Y_{3n}\\ Y_{43}& & Y_{44}& \&CenterDot;\&CenterDot;\&CenterDot;& & Y_{4n}\\ & \&CenterDot;\&CenterDot;\&CenterDot;& & & \&CenterDot;\&CenterDot;\&CenterDot;& \\ Y_{n3}& & {\mathrm{<figure-callout\; id="4"\; label="Y\; n"\; filenames="HSA00000007349800022.png,HSA00000007349800031.png"\; state="\{\{state\}\}">Y</figure-callout>}}_n& \&CenterDot;\&CenterDot;\&CenterDot;& & Y_{\mathrm{nn}}\end{array}\right]$

The work that at first will do before the load in the electric power system is carried out modeling is exactly these contact nodes of cancellation, makes all loads all equivalently be connected on same the bus bus 2 as shown in Figure 3 simultaneously.

Can obtain according to formula 26 and formula 27

$\left[\begin{array}{c}{\stackrel{\&CenterDot;}{I}}_3\\ {\stackrel{\&CenterDot;}{I}}_4\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {\stackrel{\&CenterDot;}{I}}_n\end{array}\right]=C\left[\begin{array}{c}{\stackrel{\&CenterDot;}{V}}_1\\ {\stackrel{\&CenterDot;}{V}}_2\end{array}\right]+D\left[\begin{array}{c}{\stackrel{\&CenterDot;}{V}}_3\\ {\stackrel{\&CenterDot;}{V}}_4\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {\stackrel{\&CenterDot;}{V}}_n\end{array}\right]$ (formula 28)

$\left[\begin{array}{c}{\stackrel{\&CenterDot;}{V}}_3\\ {\stackrel{\&CenterDot;}{V}}_4\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {\stackrel{\&CenterDot;}{V}}_n\end{array}\right]=D^{-1}(\left[\begin{array}{c}{\stackrel{\&CenterDot;}{I}}_3\\ {\stackrel{\&CenterDot;}{I}}_4\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {\stackrel{\&CenterDot;}{I}}_n\end{array}\right]-C\left[\begin{array}{c}{\stackrel{\&CenterDot;}{V}}_1\\ {\stackrel{\&CenterDot;}{V}}_2\end{array}\right])$ (formula 29)

Wherein load current can calculate according to formula 30.

$i_x=-\frac{{\mathrm{PV}}_x+{\mathrm{QV}}_y}{{V_x}^2+{{\mathrm{<figure-callout\; id="2"\; label="V\; y"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">V</figure-callout>}}_y}^2}=-\frac{{\mathrm{PV}}_x+{\mathrm{QV}}_y}{V^2}$ (formula 30)

$i_y=-\frac{{\mathrm{PV}}_y-{\mathrm{QV}}_x}{{V_x}^2+{{\mathrm{<figure-callout\; id="2"\; label="V\; y"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">V</figure-callout>}}_y}^2}=-\frac{{\mathrm{PV}}_y-{\mathrm{<figure-callout\; id="2"\; label="QV\; x\; V"\; filenames="HSA00000007349800022.png,HSA00000007349800032.png"\; state="\{\{state\}\}">QV</figure-callout>}}_x}{V^{}}$

V wherein _x, V _y, i _x, i _yReal part and the imaginary part of representing load bus voltage and electric current respectively, P+jQ is a load power.Calculate according to formula 26 and formula 29

$\left[\begin{array}{c}{\stackrel{\&CenterDot;}{I}}_1\\ {\stackrel{\&CenterDot;}{I}}_2\end{array}\right]-{\mathrm{BD}}^{-1}\left[\begin{array}{c}{\stackrel{\&CenterDot;}{I}}_3\\ {\stackrel{\&CenterDot;}{I}}_4\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {\stackrel{\&CenterDot;}{I}}_n\end{array}\right]=(A-{\mathrm{BD}}^{-1}C)\left[\begin{array}{c}{\stackrel{\&CenterDot;}{V}}_1\\ {\stackrel{\&CenterDot;}{V}}_2\end{array}\right]$ (formula 31)

The electric current that wherein is increased on node 1 and 2 is

$\left[\begin{array}{c}\mathrm{\&Delta;}{\stackrel{\&CenterDot;}{I}}_1\\ \mathrm{\&Delta;}{\stackrel{\&CenterDot;}{I}}_2\end{array}\right]=-{\mathrm{BD}}^{-1}{\left[\begin{array}{cccc}{\stackrel{\&CenterDot;}{I}}_3& {\stackrel{\&CenterDot;}{I}}_4& \&CenterDot;\&CenterDot;\&CenterDot;& {\stackrel{\&CenterDot;}{I}}_n\end{array}\right]}^T$ (formula 32)

And the matrix A-BD in the formula 31 ^-1C just represents the equivalent admittance matrix between connected node 1 and the node 2, can calculate new impedance between node 1 and the node 2, i.e. distribution equivalent impedance according to its main diagonal element and non-main diagonal element.

Calculate as can be known according to real network, if node 2 directly is connected with node 1, except that node 1, all injection currents in the network have all been transferred on the node 2, promptly

As shown in Figure 5.Z wherein ₁₂' be according to A-BD ^-1The distribution equivalent impedance that C obtains.

According to Kirchhoff's current law (KCL), the size of equivalent front and back network injection current still is zero, because

With

Constant, so Therefore, if node 2 directly is connected with node 1, except that node 1, the load form on each load bus all can be returned on the node 2.But because distribution network is downward from node 1, each node busbar voltage can progressively reduce.Load bus voltage after the node 2 all to be lower than basically node 2 voltage (this be since going up certainly of all having basically of distribution network-high voltage bus downwards-the radial network configuration decision of low-voltage bus bar).So finally calculate

Total load current than reality is smaller.So in order to reduce error, some adjustment are done in position to node 2 in the formula 3, promptly in distribution network, select a node (not necessarily will directly be connected) with node 1, it is equivalent to directly linking to each other with node 1, that is to say that the distribution equivalent impedance that obtains at last is connected node 1 and this node.Generally can select to be positioned at the node of loading the area of concentration or directly linking to each other with big load, bigger because this node goes out the proportion of injection current, the final error of calculating can be smaller.

Suppose and select node i, therefore, modal equation is demarcated as formula 26, replace the position of node 2 with node i for equivalent back links to each other with node 1.According to the division of formula 3 matrix-blocks and variable, with different, can obtain similar four blocked admittance matrixes with respect to the A in the formula 4, B, C and D, they are respectively A ', B ', C ' and D '.The same relation that just can obtain the injection current and the node voltage of node 1 and node i.

What situation about directly linking to each other with node 2 and node 1 was different is that the load current in the distribution network has been transferred to respectively on node 1 and the i, promptly

With

All non-vanishing.Shown in accompanying drawing 6 (a), the increment current on the node 1 shown in the figure (a)

Be not 0.

For all loads in the distribution network are all equivalent to the bus of node i, so also need further equivalence, the load of the increment in the economize on electricity 1 is all equivalent to economize on electricity i, and the result is shown in accompanying drawing 6 (b).

Z in the accompanying drawing 6 (a) _1i' be that basis is suc as formula the 8 equivalent admittance matrix A '-B ' D ' that equally calculate ^-1C ', and the Z in the accompanying drawing 6 (b) _1i" be calculated as follows

${Z_{1i}}^{\&prime;\&prime;}={Z_{1i}}^{\&prime;}+\frac{\mathrm{\&Delta;}{\stackrel{\&CenterDot;}{I}}_1}{{\stackrel{\&CenterDot;}{I}}_i+\mathrm{\&Delta;}{\stackrel{\&CenterDot;}{I}}_i}$ (formula 33)

Z in the formula _1i" be the equivalent impedance of distribution network just.

The load current that keeps on the bus 1 is

${\stackrel{\&CenterDot;}{I}}_1=\frac{P-\mathrm{jQ}}{{{\stackrel{\&CenterDot;}{V}}_1}^{*}}$ (formula 34)

The power of all loads in the establishing electric network under the node 1 is ∑ P _L+ j ∑ Q _L, then Dui Ying load current is

${\stackrel{\&CenterDot;}{I}}_L=\frac{\mathrm{\&Sigma;}{P}_L-\mathrm{j\&Sigma;}{Q}_L}{{{\stackrel{\&CenterDot;}{V}}_i}^{*}}$ (formula 35)

Then can calculate the equivalent impedance of distribution network according to the loss of distribution network

$Z_{\mathrm{eq}}=\frac{(P-\mathrm{\&Sigma;}{P}_L)+j(Q-\mathrm{\&Sigma;}{Q}_L-\mathrm{\&Sigma;}{Q}_c)}{{I_1}^2}$ (formula 36)

∑ Q wherein _cBe all the reactive power compensation power in the distribution network, then equivalent reactive power compensation to economize on electricity i is

$Q_{\mathrm{ceq}}=\mathrm{imag}\left[{\stackrel{\&CenterDot;}{V}}_i{({\stackrel{\&CenterDot;}{I}}_1-{\stackrel{\&CenterDot;}{I}}_L)}^{*}\right]$ (formula 37)

According to the equivalent principle of distribution network as can be seen, the distribution equivalent impedance Z in the formula 33 _1i" and the Z in the formula 13 _EqBe consistent.

5, the design cycle of online statistics synthetic load analogy method

The input, load that the entire flow of this method comprises data with electric device constitute to calculate, dynamic load equivalently calculates, static load is equivalent calculates and part such as the equivalent calculating of distribution network impedance, as shown in Figure 1.

The basic step of this method is as follows:

(1) determines the part throttle characteristics parameter of each power consumption equipment according to dynamic analog test or representative value;

(2) traffic department provides related datas such as the transformer platform number of power supply area network topological diagram, 220kV transformer station and the 110kV of subordinate of load bus and 35kV transformer station and capacity, reactive compensation capacity, line impedance;

(3) determine the load structure situation of 220kV transformer station and 110kV of subordinate and 35kV transformer station according to the user side power information data of load control system collection of marketing;

(4) according to the load structure situation of each load bus, determine each load bus the load type of confession;

(5) off-line carries out user's sample investigation, each load type transformer station is chosen some typical 10kV or 6kV outlet carry out the part throttle characteristics investigation, the power consumption equipment that analysis and arrangement obtains each bar 10kV or 6kV outlet constitutes situation: comprise power consumption equipment type and each power consumption equipment proportion;

(6), it is promoted the use of other 10kV outlet with load type for j sample investigation 10kV outlet institute electricity supply and use equipment formation situation in k class (industrial class, commercial, resident's class, the agriculture) load.

(7) relevant information of online collection EMS according to voltage, electric current, the power factor of data acquisition 220kV/110kV/35kV every 10kV of main transformer side or 6kV circuit among the EMS, is calculated this circuit 10kV side outlet power;

(8) power data of online collection and the sample investigation data of collected offline are carried out comprehensively, statistics draws all kinds of power consumption equipment gross powers of this period, divided by the gross power of this whole load bus of period, obtains the ratio of these all kinds of power consumption equipments of period.

(9), the load model of each power consumption equipment is decomposed into static load model and dynamic load model two parts according to the part throttle characteristics parameter of each power consumption equipment;

(10) utilize the static load equivalence method, calculate the static load model parameter of the integrated load model of considering distribution network;

(11) utilize the dynamic load equivalence method, calculate the dynamic load model parameter of the integrated load model of considering distribution network;

(12) utilize the distribution network impedance equivalence method, calculate the distribution network equivalent impedance of the integrated load model of considering distribution network;

(13) form 220kV transformer station power supply area network original system (detailed system, comprise 110kV, 35kV circuit, transformer, reactive power compensation, each node with all types of payload and model parameter etc.) trend stablize data, carry out trend and stability Calculation, form result of calculation file 1;

(14) trend of the valve systems such as integrated load model of the consideration distribution network of formation 220kV transformer station power supply area network is stablized data, carries out trend and stability Calculation, forms result of calculation file 2;

(15) validity of verifying with the actual distribution network dynamic response characteristics of valve system SLM modeling such as integrated load model of considering distribution network by comparing result file 1 and destination file 2.

Technical scheme of the present invention is applied in the load model of Beijing Electric Power Corp. in-depth research and Adaptability Analysis, and effect is better, has good society and economic benefit.

Invention has been described according to specific exemplary embodiment herein.It will be conspicuous carrying out suitable replacement to one skilled in the art or revise under not departing from the scope of the present invention.Exemplary embodiment only is illustrative, rather than to the restriction of scope of the present invention, scope of the present invention is by appended claim definition

Claims (6)

1. on-line load simulation method of power system is characterized in that this method at first carries out the offline user sample investigation, analysis and arrangement obtain various typical 10kV or 6kV outlet the power load equipment ratio of confession; Utilize the user side power information of electric power system marketing department load control system collection and transformer station's power information that EMS EMS provides then, with every the 10kV of load bus that obtains to study or the online actual power of 6kV outlet; With the various typical 10kV of the power data of online collection and collected offline or 6kV outlet the power load equipment ratio of confession carry out COMPREHENSIVE CALCULATING, draw the composition of each various power load equipment of the moment of this load point, i.e. the percentage of every kind of shared total amount of power load equipment; Determine the part throttle characteristics parameter of each power load equipment again according to dynamic analog test or representative value, calculate static load constitute in the ratio of constant-impedance, constant current, permanent power, and adopt the polymerization of an induction electric group of planes to calculate equivalent dynamic load parameter; Utilize the power supply area network topology data of the load bus that traffic department provides at last: the data that comprise distribution line, transformer and reactive power compensation, adopt equivalent these data computation of algorithm synthesis of power distribution network to draw the load model of this load point, thereby realize the online load modeling of load bus.

2. online load simulation method as claimed in claim 1 is characterized in that specifically may further comprise the steps:

(1) determines the part throttle characteristics parameter of each power consumption equipment according to dynamic analog test or representative value;

(2) traffic department provides related datas such as the transformer platform number of power supply area network topological diagram, 220kV transformer station and the 110kV of subordinate of load bus and 35kV transformer station and capacity, reactive compensation capacity, line impedance;

(3) determine the load structure situation of 220kV transformer station and 110kV of subordinate and 35kV transformer station according to the user side power information data of load control system collection of marketing;

(4) according to the load structure situation of each load bus, determine each load bus the load type of confession;

(5) off-line carries out user's sample investigation, each load type transformer station is chosen some typical 10kV or 6kV outlet carry out the part throttle characteristics investigation, analysis and arrangement obtains each load type: comprise that the 10kV of industry, commerce, resident or agriculture load type or the power consumption equipment of 6kV outlet constitute situation: comprise power consumption equipment type i and each power consumption equipment proportion ρ _i

(6) relevant information of online harvest energy management system EMS is according to the voltage V of data acquisition 220kV/110kV/35kV every 10kV of main transformer side or 6kV circuit j among the EMS EMS _j, electric current I _j, power factor PF _j, calculate and obtain this circuit 10kV side outlet power P _j=V _jI _jPF _j

(7) constitute situation for j sample investigation 10kV outlet institute electricity supply and use equipment in the k type load that comprises industrial class, commercial, resident's class, agriculture, it is promoted the use of similar 10kV outlet, statistics draws the gross power of these all kinds of power consumption equipments of period, divided by the gross power of this whole load bus of period, can obtain the ratio k of whole 220kV of this period all kinds of power consumption equipment i of transformer station _i:

If always total n bar 10kV of this 220kV transformer station or 6kV outlet;

(8), the load model of each power consumption equipment is decomposed into static load model and dynamic load model two parts according to the part throttle characteristics parameter of each power consumption equipment;

(9) utilize the static load equivalence method, calculate the static load model parameter of the integrated load model of considering distribution network;

(10) utilize the dynamic load equivalence method, calculate the dynamic load model parameter of the integrated load model of considering distribution network;

(11) utilize the distribution network impedance equivalence method, calculate the distribution network equivalent impedance of the integrated load model of considering distribution network;

(12) form 220kV transformer station power supply area network original system (detailed system, comprise 110kV, 35kV circuit, transformer, reactive power compensation, each node with all types of payload and model parameter etc.) trend stablize data, carry out trend and stability Calculation, form result of calculation file 1;

(13) trend of the valve systems such as integrated load model of the consideration distribution network of formation 220kV transformer station power supply area network is stablized data, carries out trend and stability Calculation, forms result of calculation file 2;

(14) validity of verifying with the actual distribution network dynamic response characteristics of valve system SLM modeling such as integrated load model of considering distribution network by comparing result file 1 and destination file 2.

3. as the described online load simulation method of claim 1-2, it is characterized in that described load bus is a 220kV transformer station.

4. as the arbitrary described online load simulation method of claim 1-3, it is characterized in that described step (9) static load equivalence method comprises:

The static load model structure that IEEE Taskforce recommend to adopt is to be the multinomial load model (Polynomial Load Model) of polynomial equation form with the relationship description between load power and the voltage, the general type of this model suc as formula 1 and formula 2 shown in:

P=P _o[a * (V/V _o) ²+ b * (V/V _o)+c] (formula 1)

Q=Q _o[α * (V/V _o) ²+ β * (V/V _o)+γ] (formula 2)

Multinomial active power load model coefficient is a, b, c, reactive power load model coefficient is the power factor (PF) of α, β, γ and load, this load model is called as " ZIP " model, because it has comprised constant-impedance (Z), constant current (I) and permanent power (P), this model is used to describe specific load equipment or load cell, V _oThe rated voltage of expression load, P _oAnd Q _oThen be illustrated respectively in rated voltage V _oThe specified active power and the reactive power of following load, if when describing the synthetic load of bus with this model, V _o, P _oAnd Q _oBe commonly used to represent the numerical value under system's initial launch operating mode;

Equivalence to static load mainly is to FACTOR P _o, a, b, c and Q _o, α, β, γ equivalence, the equivalence of multinomial load model is based on the sensitivity of load power to load side voltage, promptly

${\frac{\&PartialD;P}{\&PartialD;V}|}_{V=\mathrm{Vo}}{=\frac{\&PartialD;P_1}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+\frac{\&PartialD;P_2}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+......+\frac{\&PartialD;P_n}{\&PartialD;V}|}_{V=\mathrm{Vo}}$ (formula 3)

${\frac{\&PartialD;Q}{\&PartialD;V}|}_{V=\mathrm{Vo}}{=\frac{\&PartialD;Q_1}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+\frac{\&PartialD;Q_2}{\&PartialD;V}|}_{V=\mathrm{Vo}}{+......+\frac{\&PartialD;Q_n}{\&PartialD;V}|}_{V=\mathrm{Vo}}$ (formula 4)

P ₁, P ₂P _nAnd Q ₁, Q ₂Q _nBe the active power and the reactive power of each static load, corresponding multinomial load model coefficient is respectively P _O1P _On, a ₁A _n, b ₁B _n, c ₁C _nAnd Q _O1Q _On, α ₁α _n, β ₁β _n, γ ₁γ _nWork as V=V _oIn time, have:

$\left.\begin{array}{c}{P}_o=P_{o1}+P_{o2}+...+P_{\mathrm{on}}\\ Q_o=Q_{o1}+Q_{o2}+...+Q_{\mathrm{on}}\end{array}\right\}$ (formula 5)

$\left.\begin{array}{c}a=\frac{P_{o1}{a}_1+P_{o2}{a}_2+......+P_{\mathrm{on}}{a}_n}{P_o}\\ b=\frac{P_{o1}{b}_1+P_{o2}{b}_2+......+P_{\mathrm{on}}{b}_n}{P_o}\\ c=1-a-b\end{array}\right\}$ (formula 6)

$\left.\begin{array}{c}\mathrm{\&alpha;}=\frac{Q_{o1}{\mathrm{\&alpha;}}_1+Q_{o2}{\mathrm{\&alpha;}}_2+......+Q_{\mathrm{on}}{\mathrm{\&alpha;}}_n}{Q_o}\\ \mathrm{\&beta;}=\frac{Q_{o1}{\mathrm{\&beta;}}_1+Q_{o2}{\mathrm{\&beta;}}_2+......+Q_{\mathrm{on}}{\mathrm{\&beta;}}_n}{Q_o}\\ \mathrm{\&gamma;}=1-\mathrm{\&alpha;}-\mathrm{\&beta;}\end{array}\right\}$ (formula 7)

5. as the arbitrary described online load simulation method of claim 1-3, it is characterized in that described step (10) dynamic load equivalence method comprises:

The rated slip Sn of the specified electromagnetic power Pemn of motor or nominal torque Temn, rotor and maximum electromagnetic torque Pem_max or breakdown torque multiple κ _mIt is the most important parameter that to represent mechanical property in the motor, the basic principle of this equivalence method is that new Equivalent Model must keep total specified active power ∑ Pn of original system absorption constant, total reactive power ∑ Qn or power factor (PF) Pf are constant, total electromagnetic power ∑ Pemn is constant, total rotor winding copper loss ∑ Pcu2 is constant, total constant and total kinetic energy ∑ Eenergy of maximum electromagnetic power ∑ Pem_max remains unchanged, can also obtain total stator winding copper loss ∑ Pcu1 according to this tittle, the rated slip Sn and the equivalent inertia time constant H of equivalent motor are calculated as follows:

∑ P _Cu1=∑ P _n-∑ P _Emn(formula 8)

S _n=∑ P _Cu2/ ∑ P _Emn(formula 9)

H=∑ E _Energy/ (∑ P _Emn-∑ P _Cu2) (formula 10)

∑ P wherein _Emn-∑ P _Cu2Be exactly the specified mechanical output of equivalent motor output, remain unchanged;

Calculate the electric parameter of equivalent motor model then according to these amounts of having tried to achieve, comprise stator resistance R _s, stator leakage reactance X _s, rotor resistance R _r, rotor leakage reactance X _rWith excitatory reactance X _m, establishing specified phase voltage is Un, the CALCULATION OF PARAMETERS flow process is as follows:

(1), calculates ∑ Pcu1, Sn and H according to formula 8 to formula 10 then, and make P according to known calculation of parameter ∑ Pn, ∑ Qn, ∑ Pemn, ∑ Pcu2, ∑ Pem_max and ∑ Eenergy _{Emt_max}=∑ P _{Em_max}

(2) establishing total stator phase current is

, then have

$i_n=\frac{\mathrm{\&Sigma;}{P}_n-j{\mathrm{\&Sigma;Q}}_n}{{3U}_n}$ (formula 11)

Then

$R_s=\frac{{\mathrm{\&Sigma;P}}_{\mathrm{cu}1}}{{{3I}_n}^2}$ (formula 12)

(3) according to ∑ Pn, ∑ Qn and Un ask the equivalent impedance Z of equivalent motor by following formula _Deq:

$Z_{\mathrm{deq}}=\frac{{{3U}_n}^2}{{\mathrm{\&Sigma;P}}_n-j{\mathrm{\&Sigma;Q}}_n}$

R _Deq=real (Z _Deq) (formula 13)

X _deq＝imag(Z _deq)

(4) formula of reduction by maximum electromagnetic power calculates X _sAnd X _r:

$X=\sqrt{{(\frac{{{3U}_n}^2}{{2P}_{\mathrm{emt}\_\max }}-R_s)}^2-{R_s}^2}$

$X_s=\frac{X}{2}$ (formula 14)

X _r＝X _s

In this algorithm, always suppose X _r=X _s, and the X that calculates according to this formula _sAnd X _rInevitable less than normal, because the maximum electromagnetic power that calculates according to the maximum electromagnetic power formula of simplifying is bigger than the maximum electromagnetic power of reality, so need be by alternative manner to X _sAnd X _rRevise;

(5) according to the R that tries to achieve _s, X _s, X _rAnd equivalent impedance Z _Deq=R _Deq+ jX _DeqAsk R _rAnd X _m, order

K _r＝R _deq-R _s

(formula 15)

K _x＝X _deq-X _s

$R_r=\frac{(K_r+{K_x}^2/K_r-\sqrt{{(K_r+{K_x}^2/K_r)}^2-4{X_s}^2})S_n}{2}$ (formula 16)

$X_m=\frac{K_r{X}_s+K_x\frac{R_r}{S_n}}{\frac{R_r}{S_n}-K_r}$

This calculating R _rAnd X _mMethod can guarantee P all the time _Em=∑ P _EmSet up;

(6) according to the R that tries to achieve _s, X _s, R _r, X _rAnd X _m, recomputate maximum electromagnetic power according to formula of reduction calculating:

$P_{\mathrm{emt}\_\max i}=\frac{{{3U}_n}^2}{2(R_s+\sqrt{{R_s}^2+X^2})}$ (formula 17)

(7) calculate maximum electromagnetic power actual under the new argument according to the Dai Weinan equivalent circuit:

The Dai Weinan equivalent impedance is:

$Z_{\mathrm{dp}}=j{X}_r+\frac{j{X}_m(R_s+j{X}_s)}{R_s+j(X_s+X_m)}$

R _Dp=real (Z _Dp) (formula 18)

X _dp＝imag(Z _dp)

The condition that produces maximum electromagnetic power is:

$R_{\mathrm{pm}}=\frac{R_r}{S_m}=\sqrt{{R_{\mathrm{dp}}}^2+{X_{\mathrm{dp}}}^2}$ (formula 19)

S _mBe critical slippage, the open circuit voltage of Dai Weinan equivalent circuit is:

${\stackrel{\&CenterDot;}{U}}_o=U_n\frac{j{X}_m}{R_s+j(X_s+X_m)}$ (formula 20)

Therefore, can recomputate the actual maximum electromagnetic torque of new argument correspondence according to following formula:

$P_{\mathrm{em}\_\max i}=\frac{{{3U}_o}^2{R}_{\mathrm{pm}}}{{(R_{\mathrm{dp}}+R_{\mathrm{pm}})}^2+{X_{\mathrm{dp}}}^2}$ (formula 21)

(8) calculate P _{Emt_maxi}With P _{Em_maxi}Ratio, revise P _{Emt_max}

$k_{\max i}=\frac{P_{\mathrm{emt}\_\max i}}{P_{\mathrm{em}\_\max i}}$

P _{Emt_max}=k _MaxiP _{Em_max}(formula 22)

(9) compare P _{Em_maxi}With P _{Em_max}

Err _{Pem_max}=| P _{Em_max}-P _{Em_maxi}| (formula 23)

If Err _{Pem_max}〉=1.0e ^-5, then returned for (4) step and recomputate, finish otherwise calculate.

6. as the arbitrary described online load simulation method of claim 1-3, it is characterized in that described step (11) utilizes the distribution network impedance equivalence method, the distribution network equivalent impedance that calculates the integrated load model of considering distribution network comprises:

Equate with each transformer of power distribution network, each distribution line consumed power sum according to distribution network system impedance consumed power, can the computing system resistance value be

$Z_{\mathrm{eq}}=\left[\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{u}_j^2{(1/Z_j)}^{*}\right]/{\left(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{I}_i\right)}^2$ (formula 24)

In the formula: Zeq represents the distribution network system impedance; Uj represents busbar voltage, Zj indication transformer and distribution line impedance; Ii represents load current.

Images