CN104281919A - Method for evaluating control performance of grid system - Google Patents

Abstract

The invention discloses a method for evaluating the control performance of a grid system. The method for evaluating the control performance of the grid system comprises the following steps that the grid system is divided into multiple interconnected control areas, and the frequency deviation factors of all the control areas are obtained; whether the absolute value of the difference value betweenthe frequency deviation factors of any two control areas is larger than a preset threshold or not is judged, and all the control areas are evaluated according to a corrected CPS1 standard if the absolute value of the difference value between the frequency deviation factors of any two control areas is larger than the preset threshold, wherein the corrected CPS1 standard is obtained through an expression (please see the expression in the specifications); otherwise, all the control areas are evaluated according to a CPS1 standard, wherein the CPS1 standard is obtained through an expression (please see the expression in the specifications). According to the method for evaluating the control performance of the grid system, the problem that the situation is likely to happen that a frequency control target cannot be distributed fairly when control performance evaluation is carried out on control areas of different load scales can be solved well.

Description

A kind of network system control performance evaluation method

Technical field

The present invention relates to a kind of network system control performance evaluation method.

Background technology

In interconnected electric power system, the frequency control effect of each regional power grid is evaluated and specification by control performance assessment criteria, control performance assessment criteria demand fulfillment system operation reliability and the objectivity requirement evaluated regional power grid controlling behavior; At present, the CPS standard of what the control performance evaluation of each large regions electrical network of China adopted is North America electric reliability association (NERC), relative to the A standard previously adopted, CPS standard is the effect of behavior to interconnected electric power system in order to evaluate each control zone more objectively, guarantee that exchange power between control zone is within planned value scope, makes the frequency departure of interconnected electric power system within the limits prescribed; But, along with CPS standard economizes the enforcement of netting in part, find to there is particular law between CPS index and the scale of control area: the CPS index of larger control zone is generally better than the CPS index of the less control zone of scale, namely adopt CPS standard of the prior art to carry out easy frequency of occurrences control objectives ε when control performance is evaluated to interconnected electric power system ²can not fair allocat, and the adaptability problem of CPS standard evaluation scale great disparity Region control performance, as pointed out the regularity of distribution of CPS index between the region of different scales in existing document " CPS is for the applicability analysis of different load scale ", point out the average fluctuation due to district control deviation ACE and the non-linear proportional relation of frequency response coefficient simultaneously, cause the adaptability problem utilizing CPS standard evaluation scale great disparity Region control performance.

The formulation of CPS standard maintains system frequency by the district control deviation ACE of restriction interconnected electric power system to stablize, the specific requirement of described CPS standard in formula, Δ f is network system actual frequency f and scheduled frequency f ₀deviation, be the one-minute average value of Δ f, ε be in given 1 year period, root-mean-square value; Suppose that an interconnected electric power system contains M control zone, there is identical frequency individual features each control zone, namely each control zone is under same service condition, its free-running frequency characteristic coefficient β is identical, and under the pattern that controls in order wire frequency deviation of interconnected network system, because the β of control zone is identical, therefore there is identical frequency bias coefficient B any moment control zone.If the root mean square of each control zone frequency departure when independent operating is the same, and the controlling behavior of control zone remains unchanged, so control zone interconnected after have the responsibility of equality to go to maintain common interconnected network frequency departure, according to district control deviation ACE and between relation, formula (1) is converted, draws:

${\mathrm{RMS}}^2\left\{\stackrel{\‾}{\mathrm{\Δf}}\right\}=\mathrm{AVG}\left({\stackrel{\‾}{\mathrm{\Δf}}}^2\right)=\mathrm{AVG}\left\{\right[{\mathrm{\Σ}}_i\left(\frac{\stackrel{\‾}{{\mathrm{ACE}}_i}}{k}\right)]\×\stackrel{\‾}{\mathrm{\Δf}}\}=\mathrm{AVG}\left\{\right[...+\left(\frac{\stackrel{\‾}{{\mathrm{ACE}}_i}}{k}\right)+...]\×\stackrel{\‾}{\mathrm{\Δf}}\}$ $=...+\mathrm{AVG}\{\left(\frac{\stackrel{\‾}{{\mathrm{ACE}}_i}}{k}\right)\×\stackrel{\‾}{\mathrm{\Δf}}\}+...,$ ACE in formula _ibe the district control deviation of i-th control zone, district control deviation ACE _ione-minute average value, k be the frequency response coefficient of system; In order to meet frequency control target, get k=-10MB, B is the frequency bias coefficient of region i, then because each control zone has identical frequency response characteristic, the responsibility of each control zone to maintenance interconnected network frequency is identical, as long as therefore meet (2) responsibility, so maintaining system frequency is fair to each control zone; In summary it can be seen that said process is based on all identical situation of the frequency bias coefficient of control zone in interconnected network system, if when contain the different control zone of frequency bias coefficient in interacted system, set a large control zone and comprise L control area, namely this large control zone is made up of L standard control area linear superposition, its load scale is L times of standard control zone, therefore frequency bias coefficient B _c=LB, B _cfor the frequency bias coefficient of large control zone, if the ACE now during each control zone independent operating meets the requirement of formula (2), L control zone requires it is the same to the control performance of frequency, therefore when they integrally run, as long as this large control zone meets $\mathrm{AVG}\{\left(\frac{\stackrel{\‾}{{\mathrm{ACE}}_1+...+{\mathrm{ACE}}_L}}{-10B}\right)\×\stackrel{\‾}{\mathrm{\Δf}}\≤L\×{\mathrm{\ϵ}}^2\}---\left(3\right),$ This large control zone is to safeguarding that the frequency stabilization responsibility of interconnected network and standard control zone are to safeguarding that the responsibility of interconnected network frequency stabilization is identical, the district control deviation ACE of this large control zone _cbe the district control deviation sum of 1 to region, region L, substituted into formula (3) and obtain large control zone frequency responsibility born as a whole can find out thus, for the control zone that interconnected network load scale is different, as long as control this region to meet formula (4), it is fair that each control zone meets to interconnected network frequency the responsibility that frequency control target bears.

CPS1 standard focuses on the long-term assessment to system frequency quality, primary goal is the stability maintaining system frequency, namely require that the root-mean-square value of system frequency is less than system frequency control objectives value, because each control zone cannot be implemented to frequency requirement, by the relation of frequency and district control deviation, analysis conversion is carried out to formula (1), and then CPS1 index is formed to the ACE restriction of control zone; Lower mask body is divided into load scale identical and different two kinds of situations in each control zone to analyze:

When each control zone load scale is all identical, in order to the physical significance of every variable in formula is better described, first by formula the control zone number M of interconnected network is multiplied by both sides simultaneously, obtains m ε in formula (5) ²be defined as frequency adjustment responsibility, primary and foremost purpose frequency adjustment responsibility being assigned to M control zone is the allocation scheme confirming responsibility, frequency adjustment responsibility has two kinds of allocation scheme: distribute according to each control zone load scale, and distributes according to each control zone load fluctuation size; In employing load scale allocation scheme situation, the load scale of control zone is equal to the size of control zone to frequency regulation capability, and therefore load scale in control zone is larger, and the frequency regulating duty born is also larger; Under this allocation model, the frequency bias coefficient linear proportional example relation of the frequency regulating duty that control zone is born in interconnected network and this control zone.Under employing control zone load fluctuation size distribution condition, load fluctuation is larger, and the difficulty of control zone frequency adjustment is larger, to get frequency control responsibility larger; Under this allocation model, the frequency regulating duty that control zone is born in interconnected network should distribute according to the long-time statistical characteristic of this control zone load fluctuation.When carrying out assigned frequency control responsibility to formula (5) by load scale, because the load scale of each control zone is all identical, therefore the frequency control responsibility born of each control zone is also identical, therefore the frequency adjustment responsibility that each control zone is distributed is ε ²; When assigned frequency adjustment responsibility is come according to load fluctuation size to formula (5), because the service condition of control zone is all similar, can think the long-time statistical characteristic close of each control zone load fluctuation, the frequency adjustment responsibility that therefore control zone is born under this allocation model is all ε ²; Show that, when interconnected electric power system is when the load scale of each control zone is all identical, carry out assigned frequency regulating and controlling responsibility from load scale and load fluctuation two kinds of angles, formula (4) all can reach the requirement of fair allocat thus.

When in interconnected network containing load scale not identical control zone time, when assigned frequency adjustment responsibility is come according to load scale to formula (5), because the load scale of this large control zone is L times of standard control zone, therefore the frequency responsibility that this control zone is distributed is L ε ², again due to the frequency bias coefficient B of this control zone _c=LB, so under this allocation model, think that each control zone of interconnected network meets formula (4) and has general adaptability to meeting system to frequency quality requirement and control zone responsibility distributional equity, to formula (5) when carrying out assigned frequency according to load fluctuation size and regulating responsibility, along with the increase of load scale, load fluctuation is not increase according to the proportional linearity of load scale, because there is the smoothing effect of load fluctuation in interconnected network, namely along with load scale in control zone increases, because in region, the load fluctuation of diverse location is cancelled out each other, load overall undulatory property in control zone is weakened, such load scale is larger, load smoothing effect is further remarkable, and now the unit capacity of control zone is that the proportional linearity of same load scale increases, the control zone that therefore load scale is larger, it is more prone to the control of frequency, and formula (4) is not suitable for according to load fluctuation allocation model.

Summary of the invention

The present invention is directed to the proposition of above problem, and develop a kind of network system control performance evaluation method.

Technological means of the present invention is as follows:

A kind of network system control performance evaluation method, comprises the steps:

Step 1: described network system is divided into several intarconnected cotrol districts, draws the frequency bias coefficient of each control zone;

Step 2: judge whether the absolute difference of the frequency bias coefficient of any two control zones is greater than predetermined threshold value, is perform step 3, otherwise perform step 4;

Step 3: evaluate each control zone respectively according to revised CPS1 standard, described revised CPS1 standard is $\mathrm{AVG}\{\left(\frac{\stackrel{\‾}{{\mathrm{VCE}}_i}}{-10B_s}\right)\×\stackrel{\‾}{\mathrm{\Δf}}\}\≤\frac{\sqrt{{-10B}_i}}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\sqrt{-10B_i}}\×{\mathrm{\ϵ}}^2,$ In formula for the control deviation ACE of control zone i _ione-minute average value, B _sfor the frequency bias coefficient that network system is total, for the one-minute average value of each control zone frequency deviation f, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

Step 4: evaluate each control zone respectively according to CPS1 standard, described CPS1 standard is in formula for the control deviation ACE of control zone i _ione-minute average value, B _sfor the frequency bias coefficient that network system is total, for the one-minute average value of each control zone frequency deviation f, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

Further, described step 3 comprises the steps:

1. Real-time Collection network system actual frequency f, obtains network system actual frequency f and scheduled frequency f ₀difference DELTA f=f-f ₀, then the frequency departure of each control zone is Δ f, and calculates the one-minute average value of each control zone frequency deviation f wherein f is interconnected network system frequency, the f of Real-time Collection ₀for network system scheduled frequency;

2. the Tie line Power P of Real-time Collection connection control district i _t, obtain the Tie line Power P of control zone i _twith scheduled net interchange P ₀difference DELTA P _t=P _t-P ₀, according to ACE _i=-10B _iΔ f+ Δ P _tobtain the control deviation ACE of control zone i _i, wherein ACE _ifor the control deviation of control zone i, Δ f is the frequency departure of control zone i, Δ P _tfor Tie line Power P _twith scheduled net interchange P ₀difference, B _ifor the frequency bias coefficient of control zone i, i=1,2 ..., n;

3. according to formula calculate in the revised CPS1 standard of control zone i and meet factor CF ₁, wherein for the control deviation ACE of control zone i _ione-minute average value, for the one-minute average value of control zone i frequency deviation f, B _sfor the frequency bias coefficient that network system is total, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

What 4. 3. calculate when step meets factor CF ₁when≤0, automatic generation amount control mode is the control deviation ACE of retentive control district i _iconstant;

What 3. calculate when step meets the factor 0 < CF ₁when≤1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ithe direction change of trend zero;

What 3. calculate when step meets factor CF ₁during > 1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ibe less than wherein b _ifor the frequency bias coefficient of control zone i, B _sfor the frequency bias coefficient that network system is total, ε ₁₀for the root-mean-square value of ten minutes mean value of network system actual frequency and scheduled frequency deviation in given 1 year period;

Further, described step 4 comprises the steps:

I, Real-time Collection network system actual frequency f, obtain network system actual frequency f and scheduled frequency f ₀difference DELTA f=f-f ₀, then the frequency departure of each control zone is Δ f, and calculates the one-minute average value of each control zone frequency deviation f wherein f is interconnected network system frequency, the f of Real-time Collection ₀for network system scheduled frequency;

The Tie line Power P of II, Real-time Collection connection control district i _t, obtain the Tie line Power P of control zone i _twith scheduled net interchange P ₀difference DELTA P _t=P _t-P ₀, according to ACE _i=-10B _iΔ f+ Δ P _tobtain the control deviation ACE of control zone i _i, wherein ACE _ifor the control deviation of control zone i, Δ f is the frequency departure of control zone i, Δ P _tfor Tie line Power P _twith scheduled net interchange P ₀difference, B _ifor the frequency bias coefficient of control zone i, i=1,2 ..., n;

III, according to formula calculate in the iCPS1 standard of control zone and meet factor CF ₁, wherein for the control deviation ACE of control zone i _ione-minute average value, for the one-minute average value of control zone i frequency deviation f, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

IV, calculate when step III meet factor CF ₁when≤0, automatic generation amount control mode is the control deviation ACE of retentive control district i _iconstant;

What calculate when step III meets the factor 0 < CF ₁when≤1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ithe direction change of trend zero;

What calculate when step III meets factor CF ₁during > 1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ibe less than wherein b _ifor frequency bias coefficient, the B of control zone i _sfor frequency bias coefficient, ε that network system is total ₁₀for the root-mean-square value of ten minutes mean value of network system actual frequency and scheduled frequency deviation in given 1 year period;

Further, described revised CPS1 standard draws as follows:

A, detect corresponding time point t _kcontrol zone i load fluctuation value Δ L _i(t _k), described time point forms a set T ⁿ={ t _k, k=1,2 ..., K, have time interval W between adjacent time point, the setting of W makes K meet central limit theorem;

B, calculated load undulating quantity Δ L _i(t _k) expectation value and standard deviation ;

C, structure control zone i load fluctuation model t in formula _kfor measuring the time point of load fluctuation, Δ L _i(t _k) equal t for time point _ktime load fluctuation value;

D, calculated load volatility model expectation value and standard deviation wherein $\mathrm{\&sigma;}{\stackrel{\&OverBar;}{\mathrm{\&Delta;L}}}^n=\frac{1}{\sqrt{K}}{\mathrm{\&sigma;}}_{\mathrm{\&Delta;}{L}_I},$ Draw load fluctuation model submit to normal distribution;

E, calculate the expected value and standard deviation of the large control zone be made up of M control zone, the expectation value of described large control zone is the standard deviation of described large control zone is draw the increase along with load scale, the expectation value of load fluctuation increases according to increasing the consistent multiple of multiple with load scale, and the standard deviation of load fluctuation increases according to increasing the consistent multiple of multiple with the load scale after evolution;

F, by the frequency control target ε of interconnected electric power system ²increase multiple according to the load scale after evolution to distribute, draw revised CPS1 standard $\mathrm{AVG}\{\left(\frac{\stackrel{\&OverBar;}{{\mathrm{VCE}}_i}}{-10B_s}\right)\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}\&le;\frac{\sqrt{{-10B}_i}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\sqrt{-10B_i}}\&times;{\mathrm{\&epsiv;}}^2.$

Owing to have employed technique scheme, a kind of network system control performance evaluation method provided by the invention, predetermined threshold value whether is greater than as decision condition using the absolute difference of the frequency bias coefficient of any two control zones, and then each control zone with different load scale otherness can be distinguished, thus adopt CPS1 standard or revised CPS1 standard to evaluate to each control zone correspondence with different load scale otherness, solve the adaptability problem adopting CPS standard evaluation scale of the prior art great disparity Region control performance, the CPS1 index of control zone load scale being differed greatly by the method for the invention is similar in the allocation proportion in each interval.Formula medium frequency after adjustment controls responsibility allocation scheme and considers the impact of load smoothing effect on the long-time statistical characteristic of load fluctuation, easy frequency of occurrences control objectives to be carried out when control performance is evaluated in the control zone that can solve preferably between different load scale can not the problem of fair allocat, the revised CPS1 standard of the present invention considers the smoothing effect of load fluctuation to the impact of the fluctuation mean value of district control deviation, the Region control performance of different load scale can be evaluated more objectively, and the frequency control target of interconnected electric power system can be met.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the method for the invention.

Embodiment

A kind of network system control performance evaluation method as shown in Figure 1, comprises the steps:

A kind of network system control performance evaluation method, comprises the steps:

Step 1: described network system is divided into several intarconnected cotrol districts, draws the frequency bias coefficient of each control zone;

Step 2: judge whether the absolute difference of the frequency bias coefficient of any two control zones is greater than predetermined threshold value, is perform step 3, otherwise perform step 4;

Step 3: evaluate each control zone respectively according to revised CPS1 standard, described revised CPS1 standard is $\mathrm{AVG}\{\left(\frac{\stackrel{\&OverBar;}{{\mathrm{VCE}}_i}}{-10B_s}\right)\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}\&le;\frac{\sqrt{{-10B}_i}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\sqrt{-10B_i}}\&times;{\mathrm{\&epsiv;}}^2,$ In formula for the control deviation ACE of control zone i _ione-minute average value, B _sfor the frequency bias coefficient that network system is total, for the one-minute average value of each control zone frequency deviation f, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n, n be the number of the control zone that network system marks off;

Step 4: evaluate each control zone respectively according to CPS1 standard, described CPS1 standard is in formula for the control deviation ACE of control zone i _ione-minute average value, B _sfor the frequency bias coefficient that network system is total, for the one-minute average value of each control zone frequency deviation f, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

Further, described step 3 comprises the steps:

1. Real-time Collection network system actual frequency f, obtains network system actual frequency f and scheduled frequency f ₀difference DELTA f=f-f ₀, then the frequency departure of each control zone is Δ f, and calculates the one-minute average value of each control zone frequency deviation f wherein f is interconnected network system frequency, the f of Real-time Collection ₀for network system scheduled frequency;

2. the Tie line Power P of Real-time Collection connection control district i _t, obtain the Tie line Power P of control zone i _twith scheduled net interchange P ₀difference DELTA P _t=P _t-P ₀, according to ACE _i=-10B _iΔ f+ Δ P _tobtain the control deviation ACE of control zone i _i, wherein ACE _ifor the control deviation of control zone i, Δ f is the frequency departure of control zone i, Δ P _tfor Tie line Power P _twith scheduled net interchange P ₀difference, B _ifor the frequency bias coefficient of control zone i, i=1,2 ..., n;

3. according to formula calculate in the revised CPS1 standard of control zone i and meet factor CF ₁, wherein for the control deviation ACE of control zone i _ione-minute average value, for the one-minute average value of control zone i frequency deviation f, B _sfor the frequency bias coefficient that network system is total, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

What 4. 3. calculate when step meets factor CF ₁when≤0, automatic generation amount control mode is the control deviation ACE of retentive control district i _iconstant;

What 3. calculate when step meets the factor 0 < CF ₁when≤1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ithe direction change of trend zero;

What 3. calculate when step meets factor CF ₁during > 1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ibe less than wherein b _ifor the frequency bias coefficient of control zone i, B _sfor the frequency bias coefficient that network system is total, ε ₁₀for the root-mean-square value of ten minutes mean value of network system actual frequency and scheduled frequency deviation in given 1 year period;

Further, described step 4 comprises the steps:

I, Real-time Collection network system actual frequency f, obtain network system actual frequency f and scheduled frequency f ₀difference DELTA f=f-f ₀, then the frequency departure of each control zone is Δ f, and calculates the one-minute average value of each control zone frequency deviation f wherein f is interconnected network system frequency, the f of Real-time Collection ₀for network system scheduled frequency;

The Tie line Power P of II, Real-time Collection connection control district i _t, obtain the Tie line Power P of control zone i _twith scheduled net interchange P ₀difference DELTA P _t=P _t-P ₀, according to ACE _i=-10B _iΔ f+ Δ P _tobtain the control deviation ACE of control zone i _i, wherein ACE _ifor the control deviation of control zone i, Δ f is the frequency departure of control zone i, Δ P _tfor Tie line Power P _twith scheduled net interchange P ₀difference, B _ifor the frequency bias coefficient of control zone i, i=1,2 ..., n;

III, according to formula calculate in the iCPS1 standard of control zone and meet factor CF ₁, wherein for the control deviation ACE of control zone i _ione-minute average value, for the one-minute average value of control zone i frequency deviation f, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

IV, calculate when step III meet factor CF ₁when≤0, automatic generation amount control mode is the control deviation ACE of retentive control district i _iconstant;

What calculate when step III meets the factor 0 < CF ₁when≤1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ithe direction change of trend zero;

What calculate when step III meets factor CF ₁during > 1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ibe less than wherein b _ifor frequency bias coefficient, the B of control zone i _sfor frequency bias coefficient, ε that network system is total ₁₀for the root-mean-square value of ten minutes mean value of network system actual frequency and scheduled frequency deviation in given 1 year period;

Further, described revised CPS1 standard draws as follows:

A, detect corresponding time point t _kcontrol zone i load fluctuation value Δ L _i(t _k), described time point forms a set T ⁿ={ t _k, k=1,2 ..., K, have time interval W between adjacent time point, the setting of W makes K meet central limit theorem;

B, calculated load undulating quantity Δ L _i(t _k) expectation value and standard deviation ; Be be converted to another steady state (SS) from current steady state (SS) under network system load disturbance at any time, load fluctuation be not independent of each other in the same time as stochastic variable, set { the Δ L of load fluctuation _i(t _k): k=1,2 ..., K} is one group of independently random distribution data;

C, structure control zone i load fluctuation model t in formula _kfor measuring the time point of load fluctuation, Δ L _i(t _k) for time point be t _ktime load fluctuation value;

D, calculated load volatility model expectation value and standard deviation wherein $\mathrm{\&sigma;}{\stackrel{\&OverBar;}{\mathrm{\&Delta;L}}}^n=\frac{1}{\sqrt{K}}{\mathrm{\&sigma;}}_{\mathrm{\&Delta;}{L}_I},$ Draw load fluctuation model submit to normal distribution;

E, calculate the expected value and standard deviation of the large control zone be made up of M control zone, the expectation value of described large control zone is the standard deviation of described large control zone is draw the increase along with load scale, the expectation value of load fluctuation increases according to increasing the consistent multiple of multiple with load scale, and the standard deviation of load fluctuation increases according to increasing the consistent multiple of multiple with the load scale after evolution; A large control zone is made up of several control zones linear superposition, and then the load fluctuation of this large control zone is also be made up of the load linear superposition of each control zone, and the load fluctuation of each control zone is separate, according to Probability Theory and Math Statistics relevant knowledge, the load fluctuation of this large control zone is also Normal Distribution;

F, by the frequency control target ε of interconnected electric power system ²increase multiple according to the load scale after evolution to distribute, draw revised CPS1 standard $\mathrm{AVG}\{\left(\frac{\stackrel{\&OverBar;}{{\mathrm{VCE}}_i}}{-10B_s}\right)\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}\&le;\frac{\sqrt{{-10B}_i}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\sqrt{-10B_i}}\&times;{\mathrm{\&epsiv;}}^2.$

The present invention adopts the standard deviation of load fluctuation to characterize load fluctuation characteristic, standard deviation has better intuitive, and proximity test value itself more, the fluctuation size of a sample can be weighed, the dispersion degree of reflected sample exactly, and then characterize load fluctuation characteristic fully and effectively.

The present invention gives the allocation scheme of two kinds of frequency adjustment responsibilities according to the difference of each control zone load scale otherness, lower mask body passes judgment on the responsibility distributional equity sex chromosome mosaicism of each control zone under two kinds of allocation scheme, namely under these two kinds of allocation models, whether the Region control effect of each control zone is the same, the index of evaluation region control effects is the statistical property from seeing the frequency departure of this control zone within longer time period, represents with the root-mean-square value of frequency departure; Under the pattern of distributing according to load scale, due to be with load scale for distribute foundation, Region control performance has nothing to do with the load fluctuation size of control zone, therefore when each control zone independent operating, keep and interconnected operation time consistent, to the requirement of the root-mean-square value of frequency departure all the same (frequency control target is the same) during each control zone independent operating, be then fair when control zone is interconnected to the distribution of frequency adjustment responsibility; Under the pattern of distributing according to load fluctuation, due to be with load fluctuation be distribute foundation, therefore Region control performance is relevant with load fluctuation, the service condition when service condition when each control area is in isolated operation and their interconnected operations is consistent, and the root-mean-square value of the frequency departure of each control zone independent operating time all the same, each control zone just arrives fair principle to the distribution of frequency adjustment responsibility.

When control zone c independent operating (each control zone interconnected after frequency departure be consistent), according to control zone control deviation ACE _ccomputing formula, can obtain the frequency deviation f during c independent operating of control zone _cone-minute average value, therefore can change into being the one-minute average value of network system frequency departure, is the coefficient result in each control zone, frequency departure during reflection control zone c independent operating and the relation between the frequency departure of interconnected network; And then $\mathrm{AVG}\{\stackrel{\&OverBar;}{\left(\frac{{\mathrm{ACE}}_c}{-10B_c}\right)}\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}\&le;{\mathrm{\&epsiv;}}^2$ Be converted into $\mathrm{AVG}\{{\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}}_c\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}{\mathrm{\&epsiv;}}^2---\left(6\right),$ From formula (6), after the frequency adjustment responsibility distribution of interconnected network, frequency departure when limiting control zone independent operating and the relation between the frequency departure of system, frequency departure due to system is identical, therefore after distributing according to load scale pattern, each control zone is when service condition is constant, and Region control effect during independent operating all needs to be not more than identical frequency control target ε ₀, namely under this allocation model, meet the principle of frequency fair allocat, therefore CPS1 standard is applicable to load scale as distributing the pattern of foundation, according under load fluctuation allocation model, according to CPS1 standard, each control zone all demand fulfillment formula (7) during independent operating after frequency adjustment responsibility is distributed, because the existence of load fluctuation smoothing effect in interconnected network, make the control zone that load scale is larger, load fluctuation is relatively more level and smooth, AGC (control of automatic generation amount) performance is better, and then frequency departure during the independent operating of control zone is less, if or according to load scale, frequency control general objective is distributed, then each control zone demand fulfillment formula (7) requirement, but the control zone that scale is larger, the one-minute average value of the frequency departure of independent operating is less, control zone also less, therefore the requirement of formula (7) more easily meets, and this is obviously unfair to the control zone of small scale.

Existing CPS1 standard does not consider the smoothing effect of load, thinks the long-time statistical characteristic of load fluctuation and the frequency bias coefficient linear proportional relation of this control zone, thus distributes the overhead control target ε of interconnected electric power system according to frequency bias coefficient size ², a kind of network system control performance evaluation method provided by the invention, predetermined threshold value whether is greater than as decision condition using the absolute difference of the frequency bias coefficient of any two control zones, and then each control zone with different load scale otherness can be distinguished, thus adopt CPS1 standard or revised CPS1 standard to evaluate to each control zone correspondence with different load scale otherness, solve the adaptability problem adopting CPS standard evaluation scale of the prior art great disparity Region control performance, the CPS1 index of control zone load scale being differed greatly by the method for the invention is similar in the allocation proportion in each interval.Formula medium frequency after adjustment controls responsibility allocation scheme and considers the impact of load smoothing effect on the long-time statistical characteristic of load fluctuation, easy frequency of occurrences control objectives to be carried out when control performance is evaluated in the control zone that can solve preferably between different load scale can not the problem of fair allocat, the revised CPS1 standard of the present invention considers the smoothing effect of load fluctuation to the impact of the fluctuation mean value of district control deviation, the Region control performance of different load scale can be evaluated more objectively, and the frequency control target of interconnected electric power system can be met.

MATLAB/Simulink is utilized to carry out simulation analysis to the network system being divided into two intarconnected cotrol districts below, two intarconnected cotrol districts are control zone A and B, wherein the frequency bias coefficient of control zone A is 42.66, the frequency bias coefficient of control zone B is 21.33, and namely the frequency bias coefficient of control zone A is 2 times of control zone B; The load fluctuation model of control zone B submits to normal distribution, and the load fluctuation model of control zone A is made up of the load fluctuation model linear superposition of 2 control zone B.Table 1 is each interval statistics value (the evaluation period is 1min) of CPS1 standard, as known from Table 1, the CPS1 index entirety of control zone A is better than the index of control zone B, and the ratio being less than 100% interval that is distributed in of control zone A is starkly lower than the ratio that control zone B distributes in this interval; And the ratio that control zone A is distributed in 100% ~ 200% interval is significantly higher than the ratio that control zone B distributes in this interval, control zone A is being greater than the ratio in 200% interval a little less than the ratio of control zone B in this interval.

Table 1. liang intarconnected cotrol district CPS1 index distribution.

Between CPS1 Index areas	＜100％	100％～200％	＞200％
				Control zone A	0.0945	0.7120	0.1934
Control zone B	0.1689	0.6121	0.2189

Table 2 carries out revised simulation result to CPS1 standard, now the ratio that CPS1 index is being less than 100% interval of control zone A is increased to 11.34% by 9.45%, distribution in 100% ~ 200% interval drops to 69.32% by 71.2%, does not become in the distribution being greater than 200% interval; Be distributed in the ratio being less than 100% interval in the B index set of control zone and drop to 11.26% by 16.89%, be increased to 66.84% at the interval distribution proportion of 100% ~ 200% by 61.21%, do not change in the interval being greater than 200%; After A and B each interval Distribution Indexes ratio in control zone changes, control zone A is substantially identical with the distribution proportion that control zone B is being less than 100% interval, and the distribution proportion in 100% ~ 200% interval is also close; Therefore revised CPS1 standard can solve the malapportioned problem in control zone between different load scale preferably.

The revised CPS1 index distribution in table 2. liang intarconnected cotrol district.

Between CPS1 Index areas	＜100％	100％～200％	＞200％
				Control zone A	0.1134	0.6932	0.1934
Control zone B	0.1126	0.6684	0.2189

The above; be only the present invention's preferably embodiment; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses; be equal to according to technical scheme of the present invention and inventive concept thereof and replace or change, all should be encompassed within protection scope of the present invention.

Claims (4)

1. a network system control performance evaluation method, is characterized in that comprising the steps:

Step 1: described network system is divided into several intarconnected cotrol districts, draws the frequency bias coefficient of each control zone;

Step 2: judge whether the absolute difference of the frequency bias coefficient of any two control zones is greater than predetermined threshold value, is perform step 3, otherwise perform step 4;

Step 3: evaluate each control zone respectively according to revised CPS1 standard, described revised CPS1 standard is $\mathrm{AVG}\{\left(\frac{\stackrel{\&OverBar;}{{\mathrm{VCE}}_i}}{-10B_s}\right)\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}\&le;\frac{\sqrt{{-10B}_i}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\sqrt{-10B_i}}\&times;{\mathrm{\&epsiv;}}^2,$ In formula for the control deviation ACE of control zone i _ione-minute average value, B _sfor the frequency bias coefficient that network system is total, for the one-minute average value of each control zone frequency deviation f, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

Step 4: evaluate each control zone respectively according to CPS1 standard, described CPS1 standard is in formula for the control deviation ACE of control zone i _ione-minute average value, B _sfor the frequency bias coefficient that network system is total, for the one-minute average value of each control zone frequency deviation f, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n.

2. a kind of power grid control method of evaluating performance according to claim 1, is characterized in that described step 3 comprises the steps:

1. Real-time Collection network system actual frequency f, obtains network system actual frequency f and scheduled frequency f ₀difference DELTA f=f-f ₀, then the frequency departure of each control zone is Δ f, and calculates the one-minute average value of each control zone frequency deviation f wherein f is interconnected network system frequency, the f of Real-time Collection ₀for network system scheduled frequency;

2. the Tie line Power P of Real-time Collection connection control district i _t, obtain the Tie line Power P of control zone i _twith scheduled net interchange P ₀difference DELTA P _t=P _t-P ₀, according to ACE _i=-10B _iΔ f+ Δ P _tobtain the control deviation ACE of control zone i _i, wherein ACE _ifor the control deviation of control zone i, Δ f is the frequency departure of control zone i, Δ P _tfor Tie line Power P _twith scheduled net interchange P ₀difference, B _ifor the frequency bias coefficient of control zone i, i=1,2 ..., n;

3. according to formula calculate in the revised CPS1 standard of control zone i and meet factor CF ₁, wherein for the control deviation ACE of control zone i _ione-minute average value, for the one-minute average value of control zone i frequency deviation f, B _sfor the frequency bias coefficient that network system is total, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

What 4. 3. calculate when step meets factor CF ₁when≤0, automatic generation amount control mode is the control deviation ACE of retentive control district i _iconstant;

What 3. calculate when step meets the factor 0 < CF ₁when≤1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ithe direction change of trend zero;

What 3. calculate when step meets factor CF ₁during > 1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ibe less than wherein b _ifor the frequency bias coefficient of control zone i, B _sfor the frequency bias coefficient that network system is total, ε ₁₀for the root-mean-square value of ten minutes mean value of network system actual frequency and scheduled frequency deviation in given 1 year period.

3. a kind of power grid control method of evaluating performance according to claim 1, is characterized in that described step 4 comprises the steps:

I, Real-time Collection network system actual frequency f, obtain network system actual frequency f and scheduled frequency f ₀difference DELTA f=f-f ₀, then the frequency departure of each control zone is Δ f, and calculates the one-minute average value of each control zone frequency deviation f wherein f is interconnected network system frequency, the f of Real-time Collection ₀for network system scheduled frequency;

The Tie line Power P of II, Real-time Collection connection control district i _t, obtain the Tie line Power P of control zone i _twith scheduled net interchange P ₀difference DELTA P _t=P _t-P ₀, according to ACE _i=-10B _iΔ f+ Δ P _tobtain the control deviation ACE of control zone i _i, wherein ACE _ifor the control deviation of control zone i, Δ f is the frequency departure of control zone i, Δ P _tfor Tie line Power P _twith scheduled net interchange P ₀difference, B _ifor the frequency bias coefficient of control zone i, i=1,2 ..., n;

III, according to formula calculate in the iCPS1 standard of control zone and meet factor CF ₁, wherein for the control deviation ACE of control zone i _ione-minute average value, for the one-minute average value of control zone i frequency deviation f, B _ifor the frequency bias coefficient of control zone i, ε is the root-mean-square value of the one-minute average value of Δ f in given 1 year period, i=1,2 ..., n;

IV, calculate when step III meet factor CF ₁when≤0, automatic generation amount control mode is the control deviation ACE of retentive control district i _iconstant;

What calculate when step III meets the factor 0 < CF ₁when≤1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ithe direction change of trend zero;

What calculate when step III meets factor CF ₁during > 1, automatic generation amount control mode is the active power adjusting corresponding automatic-generation-control unit, makes the control deviation ACE of control zone i _ibe less than wherein b _ifor frequency bias coefficient, the B of control zone i _sfor frequency bias coefficient, ε that network system is total ₁₀for the root-mean-square value of ten minutes mean value of network system actual frequency and scheduled frequency deviation in given 1 year period.

4. a kind of power grid control method of evaluating performance according to claim 1, is characterized in that described revised CPS1 standard draws as follows:

A, detect corresponding time point t _kcontrol zone i load fluctuation value Δ L _i(t _k), described time point forms a set T ⁿ={ t _k, k=1,2 ..., K, have time interval W between adjacent time point, the setting of W makes K meet central limit theorem;

B, calculated load undulating quantity Δ L _i(t _k) expectation value and standard deviation ;

C, structure control zone i load fluctuation model t in formula _kfor measuring the time point of load fluctuation, Δ L _i(t _k) equal t for time point _ktime load fluctuation value;

D, calculated load volatility model expectation value and standard deviation wherein $\mathrm{\&sigma;}{\stackrel{\&OverBar;}{\mathrm{\&Delta;L}}}^n=\frac{1}{\sqrt{K}}{\mathrm{\&sigma;}}_{\mathrm{\&Delta;}{L}_I},$ Draw load fluctuation model submit to normal distribution;

E, calculate the expected value and standard deviation of the large control zone be made up of M control zone, the expectation value of described large control zone is the standard deviation of described large control zone is draw the increase along with load scale, the expectation value of load fluctuation increases according to increasing the consistent multiple of multiple with load scale, and the standard deviation of load fluctuation increases according to increasing the consistent multiple of multiple with the load scale after evolution;

F, by the frequency control target ε of interconnected electric power system ²increase multiple according to the load scale after evolution to distribute, draw revised CPS1 standard $\mathrm{AVG}\{\left(\frac{\stackrel{\&OverBar;}{{\mathrm{VCE}}_i}}{-10B_s}\right)\&times;\stackrel{\&OverBar;}{\mathrm{\&Delta;f}}\}\&le;\frac{\sqrt{{-10B}_i}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\sqrt{-10B_i}}\&times;{\mathrm{\&epsiv;}}^2.$