CN109347108A - One kind preventing chain overload controlling method - Google Patents

One kind preventing chain overload controlling method Download PDF

Info

Publication number
CN109347108A
CN109347108A CN201811202701.8A CN201811202701A CN109347108A CN 109347108 A CN109347108 A CN 109347108A CN 201811202701 A CN201811202701 A CN 201811202701A CN 109347108 A CN109347108 A CN 109347108A
Authority
CN
China
Prior art keywords
branch
control node
overload
sensitivity
power
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201811202701.8A
Other languages
Chinese (zh)
Inventor
李本瑜
游昊
石恒初
陈剑平
方毅
杨远航
罗吉
张鑫
任文鹏
秦昊
蔡明芳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
WUHAN JIAHUA INNOVATION ELECTRICAL CO Ltd
Yunnan Power Grid Co Ltd
Original Assignee
WUHAN JIAHUA INNOVATION ELECTRICAL CO Ltd
Yunnan Power Grid Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by WUHAN JIAHUA INNOVATION ELECTRICAL CO Ltd, Yunnan Power Grid Co Ltd filed Critical WUHAN JIAHUA INNOVATION ELECTRICAL CO Ltd
Priority to CN201811202701.8A priority Critical patent/CN109347108A/en
Publication of CN109347108A publication Critical patent/CN109347108A/en
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02JCIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00Circuit arrangements for ac mains or ac distribution networks
    • H02J3/04Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
    • H02J3/06Controlling transfer of power between connected networks; Controlling sharing of load between connected networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Control Of Eletrric Generators (AREA)

Abstract

One kind preventing chain overload controlling method, crucial branch collection is first determined according to the operation information of each branch, pairing sensitivity of each control node to overload branch in broad sense power flow transfer region where computation overload branch again, and effective control node is filtered out to addition set omega, then in set of computations Ω each effective control node pair improvement overall sensitivity, and the maximum control node of overall sensitivity value is chosen to as epicycle optimum control unit, then calculate the adjustment amount Δ of control node pairP ij , further according to ΔP ij Control of decreasing load is carried out, refreshes Branch Power Flow using sensitivity relation and corrects node and contribute, while judging to overload whether phenomenon is eliminated.The calculation amount of emergent control can be not only greatly decreased in the design, but also enhances overload elimination ability, accelerates overload release rate.

Description

One kind preventing chain overload controlling method
Technical field
The invention belongs to Operation of Electric Systems and control technology field, and in particular to one kind prevents chain overload controlling method.
Background technique
The rapid development of electric utility provides strong guarantee for social progress, productivity dynamics and national security, But the safe operation problem of its own also increasingly shows.It has been trend of the times that power grid interconnects on a large scale, system dynamic behaviour also because This becomes more sophisticated, and in the large scale electric network of grid structure even more complex, power flow transfer overload has become cascading failure The main propellant of development, and emergent control is still the important means of prevention large-scale blackout.
Branch overload emergent control is subordinate to strategy for security correction scope, keeps overload Branch Power Flow fast by adopting an effective measure Speed decline, is withdrawn into safe condition from the state of emergency for system, to thoroughly strangling cascading failure in the budding period.Due to sensitive Degree method has significant advantage in algorithm complexity and convergence rate level, therefore shows one's talent and obtain from all control algolithms Extensive utilization, still, existing Sensitivity Analysis Method, which mostly only obtains single control node as evaluation object, to be added, subtracts out Power nodal operation sequence, by participate in reversed pairing adjustment plus, subtract power output node and be considered as an entirety and carry out overall merit, nothing Method, which fully demonstrates, respectively to be added, subtracts power output node by cooperation coordinated with each other, and the integration capability of a plurality of branch overload in section is eliminated.
Summary of the invention
The purpose of the present invention is it is of the existing technology can not effectively eliminate in section a plurality of branch and overload aiming at the problem that, It provides and a kind of prevents chain overload controlling method to come eliminate overload using every wheel obtained optimum control node of adjustment.
In order to achieve the above object, technical scheme is as follows:
One kind preventing chain overload controlling method, successively the following steps are included:
Step A, the crucial branch being made of overload branch collection L and heavy duty branch collection M is determined according to the operation information of each branch Road collection K, wherein the operation information includes overload branch l and its overload quantity, heavy duty branch m and its amount of redundancy;
Step B, pairing spirit of each control node to overload branch in broad sense power flow transfer region where computation overload branch Sensitivity, and effective control node is filtered out to set omega is formed according to pairing sensitivity, if set omega is empty set, reduce effective Control node to filtering out effective control node to set omega is formed, until set omega is not empty set again after screening threshold value;
Step C, in set of computations Ω each effective control node pair improvement overall sensitivity, and choose improvement synthesis it is sensitive The maximum control node of angle value is to as optimum control node pair;
Step D, the adjustment amount Δ P of optimum control node pair is calculatedij
Step E, first according to Δ PijControl of decreasing load is carried out, sensitivity relation is recycled to refresh Branch Power Flow and correct node Power, while judging to overload whether phenomenon is eliminated, if overload is eliminated, terminate premature beats;If overload phenomenon still remains, more Step B to step E is repeated after new key branch collection K, until overload is eliminated.
In step B, the pairing sensitivity is calculated using equal and opposite quantities in pairs method:
ΔPl=Sl-ijΔPj
Sl-ij=Sl-j-Sl-i
In above formula, Δ PlFor the active power variable quantity for overloading branch l, Sl-ijTo add in overload branch l, subtracting power output node I, the pairing sensitivity of j, i ∈ B1∪B2, j ∈ B1, B1、B2Respectively there is generator node collection, the load bus collection of adjustment amount, ΔPjFor the active injection variable quantity of control node j, Sl-iSensitivity for control node i to overload branch l indicates control section Point i active power output, which increases when 1 unit power, corresponding reference mode such as balancing machine reduce by 1 unit power, overloads having for branch l Function variable quantity, B are the susceptance matrix of control node, and Δ θ is the voltage-phase variable quantity of control node, and Δ P is control node Active injection variable quantity, Δ PiFor the active injection variable quantity of control node i, Δ θlFor the phase difference for overloading the both ends branch l, xl For the reactance for overloading branch l.
In step C, the improvement overall sensitivity SK-ijAre as follows:
In above formula, λ is the ratio between the weight for overloading branch and heavily loaded branch, αl、αmRespectively overload branch l, heavily loaded branch m Weight, Sl-ijFor in overload branch l plus, the pairing sensitivity of wind farm control node i, j, Sm-ijTo add in heavily loaded branch m, The pairing sensitivity of wind farm control node i, j, Pl、Pl.limRespectively overload the power and its limit of branch l, Pm、Pm.limFor The power and its limit of heavily loaded branch m.
In step D, the adjustment amount Δ PijIt is calculated by following formula:
ΔPij=min (Δ PN,ΔPM,ΔPR)
ΔPN=min (APi.max, Δ Pj.max)
ΔPM=max { β (Pl-Pl.lim)/Sl-ij|l∈L,Sl-ij> 0 }
ΔPR=min { γ (Pz-Pz.lim)/Sz-ij|z∈F,Sz-ij< 0 }
In above formula, Δ PNIt can adjust power output, Δ P for maximum of the control node to (i, j)i.maxFor the maximum of control node i Power output, Δ P can be addedj.maxPower output, Δ P can be subtracted for the maximum of control node jMThe control that (i, j) is needed to adjust for control node Amount, β, γ are penalty coefficient, Pl、Pl.limRespectively overload the power and its limit of branch l, Sl-ijTo add in overload branch l, The pairing sensitivity of wind farm control node i, j, Δ PRFor the not out-of-limit control amount of normal branch, F is normal branch constraint collection, F=M ∪ B, B are the branch collection being affected by control node to (i, j), and z is the branch in F, Pz、Pz.limRespectively branch z Power and its limit, Sz-ijFor in branch z plus, the pairing sensitivity of wind farm control node i, j.
The branch collection B chooses to obtain using concentric relaxation principle, selection principle are as follows:
The branch that the 3 layers of concentric circles searched for centered on control node i or j respectively is included, then takes union.
In step B, the set omega is that can eliminate effective control node of its overload to collection, Ω=J (l1)∪J(l2) ∪…∪J(lN), { l1,l2,…,lN}=overload branch collection L, the screening technique of set omega are as follows:
For any overload branch l ∈ L, if control node is to (i, j) to the pairing sensitivity S of overload branch ll-ijMeet Following two condition, then control node to (i, j) be referred to as overload branch l effective control node to and be divided into set J (l):
Sl-ij> 0
Sl-ij≥ηmax{Sl-ij|i∈B1∪B2,j∈B1}
In above formula, η is effective control node to screening threshold value, η ∈ [0,1], B1、B2Respectively with the power generation of adjustment amount Machine node collection, load bus collection.
In step B, broad sense power flow transfer region is obtained by power flow transfer regional expansion, expanding method are as follows:
It will abut against generator that number of nodes is 1 or load bus be divided into power flow transfer region where its adjacent node;
If certain branch breaking makes system sectionalizing, and the equal non-power generator-bus-bar system of each subsystem after off-the-line, then by subsystem System divides different broad sense power flow transfer regions into;
In step B, for effective control node to (i, j), if control node i or j are cutpoint and are non-adjustable, selection with The maximum respective type node of controlled variable replaces the cutpoint in remaining connected broad sense power flow transfer region of the cutpoint.
Compared with prior art, the invention has the benefit that
1, the present invention is a kind of prevents that chain overload controlling method first determines crucial branch collection according to the operation information of each branch, then Pairing sensitivity of each control node to overload branch in broad sense power flow transfer region where computation overload branch, and filtered out Control node is imitated to set omega is added, then in set of computations Ω each effective control node pair improvement overall sensitivity, and select It takes the maximum control node of overall sensitivity value to as epicycle optimum control unit, then calculates the adjustment amount of control node pair ΔPij, further according to Δ PijControl of decreasing load is carried out, refreshes Branch Power Flow using sensitivity relation and corrects node and contribute, sentence simultaneously Whether disconnected overload phenomenon is eliminated, which is introduced broad sense power flow transfer region and saved control using improvement overall sensitivity method The search range of point is limited to a certain smaller area of system, is overloaded, can substantially be subtracted to branch is eliminated using optimum control node The calculation amount of few emergent control.Therefore, the calculation amount of emergent control can be greatly decreased in the present invention.
2, one kind of the present invention prevents that chain overload controlling method goes out by enabling balance nodes participate in emergent control but increasing unit The processing that power is adjusted can enhance overload elimination ability while avoiding balancing machine out-of-limit, accelerate overload release rate.Therefore, The present invention effectively enhances overload elimination ability, accelerates overload release rate.
3, one kind of the present invention prevents adjustment amount Δ P in chain overload controlling methodijCalculating on the one hand comprehensively considered control The not out-of-limit constraint of constraint and normal branch is eliminated in the controlled variable constraint of node pair, overload quantity, can effectively avoid carry out trend Verification, on the other hand, the error for introducing sensitivity under not only meter and DC flow model of penalty coefficient β, γ, moreover it is possible to ensure Each branch has certain nargin after Control of decreasing load, reduces branch since new out-of-limit probability being occurred by microvariations, improves tight The precision suddenly controlled, meanwhile, the normal branch constraint collection scale obtained using concentric relaxation principle is smaller and representative, right In reducing, emergent control calculation amount is of great advantage.Therefore, it invention further reduces the calculation amount of emergent control, improves tight The precision suddenly controlled.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Fig. 2 is the broad sense power flow transfer region division situation map of IEEE39 node system in the embodiment of the present invention 1.
Specific embodiment
The present invention will be further described in detail with reference to the specific embodiments.
Referring to Fig. 1, one kind preventing chain overload controlling method, successively the following steps are included:
Step A, the crucial branch being made of overload branch collection L and heavy duty branch collection M is determined according to the operation information of each branch Road collection K, wherein the operation information includes overload branch l and its overload quantity, heavy duty branch m and its amount of redundancy;
Step B, pairing spirit of each control node to overload branch in broad sense power flow transfer region where computation overload branch Sensitivity, and effective control node is filtered out to set omega is formed according to pairing sensitivity, if set omega is empty set, reduce effective Control node to filtering out effective control node to set omega is formed, until set omega is not empty set again after screening threshold value;
Step C, in set of computations Ω each effective control node pair improvement overall sensitivity, and choose improvement synthesis it is sensitive The maximum control node of angle value is to as optimum control node pair;
Step D, the adjustment amount Δ P of optimum control node pair is calculatedij
Step E, first according to Δ PijControl of decreasing load is carried out, sensitivity relation is recycled to refresh Branch Power Flow and correct node Power, while judging to overload whether phenomenon is eliminated, if overload is eliminated, terminate premature beats;If overload phenomenon still remains, more Step B to step E is repeated after new key branch collection K, until overload is eliminated.
In step B, the pairing sensitivity is calculated using equal and opposite quantities in pairs method:
ΔPl=Sl-ijΔPj
Sl-ij=Sl-j-Sl-i
In above formula, Δ PlFor the active power variable quantity for overloading branch l, Sl-ijTo add in overload branch l, subtracting power output node I, the pairing sensitivity of j, i ∈ B1∪B2, j ∈ B1, B1、B2Respectively there is generator node collection, the load bus collection of adjustment amount, ΔPjFor the active injection variable quantity of control node j, Sl-iSensitivity for control node i to overload branch l indicates control section Point i active power output, which increases when 1 unit power, corresponding reference mode such as balancing machine reduce by 1 unit power, overloads having for branch l Function variable quantity, B are the susceptance matrix of control node, and Δ θ is the voltage-phase variable quantity of control node, and Δ P is control node Active injection variable quantity, Δ PiFor the active injection variable quantity of control node i, Δ θlFor the phase difference for overloading the both ends branch l, xl For the reactance for overloading branch l.
In step C, the improvement overall sensitivity SK-ijAre as follows:
In above formula, λ is the ratio between the weight for overloading branch and heavily loaded branch, αl、αmRespectively overload branch l, heavily loaded branch m Weight, Sl-ijFor in overload branch l plus, the pairing sensitivity of wind farm control node i, j, Sm-ijTo add in heavily loaded branch m, The pairing sensitivity of wind farm control node i, j, Pl、Pl.limRespectively overload the power and its limit of branch l, Pm、Pm.limFor The power and its limit of heavily loaded branch m.
In step D, the adjustment amount Δ PijIt is calculated by following formula:
ΔPij=min (Δ PN,ΔPM,ΔPR)
ΔPN=min (Δ P Δ Pi.max,ΔPj.max)
ΔPM=max { β (Pl-Pl.lim)/Sl-ij|l∈L,Sl-ij> 0 }
ΔPR=min { γ (Pz-Pz.lim)/Sz-ij|z∈F,Sz-ij< 0 }
In above formula, Δ PNIt can adjust power output, Δ P for maximum of the control node to (i, j)i.maxFor the maximum of control node i Power output, Δ P can be addedj.maxPower output, Δ P can be subtracted for the maximum of control node jMThe control that (i, j) is needed to adjust for control node Amount, β, γ are penalty coefficient, Pl、Pl.limRespectively overload the power and its limit of branch l, Sl-ijTo add in overload branch l, The pairing sensitivity of wind farm control node i, j, Δ PRFor the not out-of-limit control amount of normal branch, F is normal branch constraint collection, F=M ∪ B, B are the branch collection being affected by control node to (i, j), and z is the branch in F, Pz、Pz.limRespectively branch z Power and its limit, Sz-ijFor in branch z plus, the pairing sensitivity of wind farm control node i, j.
The branch collection B chooses to obtain using concentric relaxation principle, selection principle are as follows:
The branch that the 3 layers of concentric circles searched for centered on control node i or j respectively is included, then takes union.
In step B, the set omega is that can eliminate effective control node of its overload to collection, Ω=J (l1)∪J(l2) ∪…∪J(lN), { l1,l2,…,lN}=overload branch collection L, the screening technique of set omega are as follows:
For any overload branch l ∈ L, if control node is to (i, j) to the pairing sensitivity S of overload branch ll-ijMeet Following two condition, then control node to (i, j) be referred to as overload branch l effective control node to and be divided into set J (l):
Sl-ij> 0
Sl-ij≥ηmax{Sl-ij|i∈B1∪B2,j∈B1}
In above formula, η is effective control node to screening threshold value, η ∈ [0,1], B1、B2Respectively with the power generation of adjustment amount Machine node collection, load bus collection.
In step B, broad sense power flow transfer region is obtained by power flow transfer regional expansion, expanding method are as follows:
It will abut against generator that number of nodes is 1 or load bus be divided into power flow transfer region where its adjacent node;
If certain branch breaking makes system sectionalizing, and the equal non-power generator-bus-bar system of each subsystem after off-the-line, then by subsystem System divides different broad sense power flow transfer regions into;
In step B, for effective control node to (i, j), if control node i or j are cutpoint and are non-adjustable, selection with The maximum respective type node of controlled variable replaces the cutpoint in remaining connected broad sense power flow transfer region of the cutpoint.
The principle of the present invention is described as follows:
Match Calculation of Sensitivity:
Using equal and opposite quantities in pairs, two control nodes are adjusted every time, avoidable balancing machine is out-of-limit and improves tight Anxious control efficiency.If node i increases power output Δ Pi, node j reduction power output Δ Pj, and the absolute value of the two adjustment amount is equal, i.e. Δ Pi=-Δ Pj, wherein i ∈ B1∪B2, j ∈ B1, only it is above-mentioned plus, subtract power output node to adjustment under, branch l active power variable quantity ΔPlExpression formula is as follows:
ΔPl=Sl-ijΔPj
Sl-ij=Sl-j-Sl-i
If Sl-ij> 0, adjustment node can reduce overload branch l trend to (i, j), if Sl-ij< 0, adjust node to (i, j) then Overload branch l trend can be increased.
Overall sensitivity calculates:
Since overload branch overload quantity is bigger, then it is higher to eliminate out-of-limit priority;Heavily loaded branch amount of redundancy is smaller, then saves Overload may more occur when point power output adjustment, bigger weight ought to be assigned.The present invention is according to overload branch overload quantity size And heavily loaded branch amount of redundancy size determines that weight is as follows:
Adjustment amount Δ PijIt calculates:
In order to eliminate branch overload to greatest extent, while considering the calculating error of DC flow model, introduces penalty coefficient β, then control node needs the control amount Δ P adjusted to (i, j)MAre as follows:
ΔPM=max { β (Pl-Pl.lim)/Sl-ij|l∈L,Sl-ij> 0 }
In practical large scale system, two classes are broadly divided into (i, j) branch influenced vulnerable to adjustment node: amount of redundancy compared with Small heavily loaded branch M and node power output easily cause trend to fluctuate biggish branch B when adjusting.
Easily cause trend to fluctuate biggish branch B according to concentric relaxation principle when node is contributed and adjusted, can will adjust Node i or j are considered as disturbing source, and usual impacted biggish branch is in 3 layers expanded outwardly centered on adjusting node with one heart In circle.
In the method for the present invention, the screening of effective control node pair is that the acquisition of optimum control node pair is laid a good foundation, most Excellent control node is to being then embodiment to overload branch and the heavily loaded best off-load effect of branch.
Embodiment 1:
Referring to Fig. 1, one kind preventing chain overload controlling method, successively follows the steps below:
Step 1, using IEEE39 node system as object, each branch operation information is obtained by current flow state, determine by Overload the crucial branch collection K of branch collection L and heavy duty branch collection M composition, wherein the system includes 4 broad sense power flow transfer regions, That is D1, D2, D3 and D4, specific dividing condition is as shown in Fig. 2, enable 10% that each control node adjusts that nargin is Power generation limits, branch When (5-8,6-7) is cut-off on road, the out-of-limit 388.83MW of branch 2-1, Overflow RateHT 1.648, heavily loaded branch 1-39,5-4 load factor Respectively 0.885,0.890, specific Power Flow Information are as shown in table 1:
The Power Flow Information of the crucial branch collection K of table 1
Pairing spirit of each control node to overload branch in broad sense power flow transfer region where step 2, computation overload branch Sensitivity, wherein for effective control node to (i, j), if control node i or j are cutpoint and are non-adjustable, selection and the cutpoint The maximum respective type node of controlled variable replaces the cutpoint in remaining connected broad sense power flow transfer region;
Broad sense power flow transfer region is obtained by power flow transfer regional expansion, expanding method are as follows:
It will abut against generator that number of nodes is 1 or load bus be divided into power flow transfer region where its adjacent node;
If certain branch breaking makes system sectionalizing, and the equal non-power generator-bus-bar system of each subsystem after off-the-line, then by subsystem System divides different broad sense power flow transfer regions into, and due in the system, branch 2-1 will result directly in because overload continues to trip and be System off-the-line, therefore strategy for security correction control need to be carried out to branch 2-1, and overload branch 2-1 and be located at region D1, preferential use generates electricity Machine-generator control node pair;
In this system, sequence 3,4 effective control node centering contain cutpoint 16,26 respectively, and be subtract power output node, therefore The maximum power generator node that subtracts of controlled variable is searched for from other broad sense trend regions being attached thereto to replace;
The pairing sensitivity is calculated using equal and opposite quantities in pairs method:
ΔPl=Sl-ijΔPj
Sl-ij=Sl-j-Sl-i
In above formula, Δ PlFor the active power variable quantity for overloading branch l, Sl-ijTo add in overload branch l, subtracting power output node I, the pairing sensitivity of j, i ∈ B1∪B2, j ∈ B1, B1、B2Respectively there is generator node collection, the load bus collection of adjustment amount, ΔPjFor the active injection variable quantity of control node j, B is the susceptance matrix of control node, and Δ θ is the voltage-phase of control node Variable quantity, Δ P are the active injection variable quantity of control node, Δ PiFor the active injection variable quantity of control node i, Δ θlFor mistake Carry the phase difference at the both ends branch l, xlFor the reactance for overloading branch l, Sl-iSensitivity for control node i to overload branch l, table Show that control node i active power output increases overload when 1 unit power, corresponding reference mode such as balancing machine reduce by 1 unit power The active variable quantity of branch l;
Step 3, the effective control node for filtering out overload branch 2-1 form set omega to 6 (referring to tables 2), wherein The set omega is that can eliminate effective control node of its overload to collection, Ω=J (l1)∪J(l2)∪…∪J(lN), { l1, l2,…,lN}=overload branch collection L, the screening technique of set omega are as follows:
For any overload branch l ∈ L, if control node is to (i, j) to the pairing sensitivity S of overload branch ll-ijMeet Following two condition, then control node to (i, j) be referred to as overload branch l effective control node to and be divided into set J (l):
During ensuring adjustment, overload branch l trend declines steadily, need to meet Sl-ij> 0;
To control node to priority division is carried out, make to match sensitivity Sl-ijMeet:
Sl-ij≥ηmax{Sl-ij|i∈B1∪B2,j∈B1}
In above formula, η is that effective control node can reflect that control node offsets to threshold value, η ∈ [0,1], value size is screened Except the ability of branch overload, value is 0.2, B herein1、B2Respectively with generator node collection, the load bus collection of adjustment amount;
The improvement overall sensitivity (referring to table 2) of each effective control node pair in step 4, set of computations Ω, and choose and change Optimum control node pair is used as to (39,31) into overall sensitivity value maximum control node, wherein the improvement synthesis is sensitive Spend SK-ijAre as follows:
In above formula, λ is the ratio between the weight for overloading branch and heavily loaded branch, takes 1.5, α hereinl、αmRespectively overload branch l, The weight of heavily loaded branch m, Sl-ijFor in overload branch l plus, the pairing sensitivity of wind farm control node i, j, Sm-ijFor heavy duty branch Add in the m of road, the pairing sensitivity of wind farm control node i, j, Pl、Pl.limThe power and its limit of branch l are respectively overloaded, Pm、Pm.limFor the power and its limit of heavily loaded branch m;
The effective control node pair and its improvement overall sensitivity of 2 first round of table control
Step 5 calculates control node to the adjustment amount Δ P of (39,31) according to the following formulaijFor 408.27MW (i.e. unit 31 power outputs reduce 408.27MW, and the power output of unit 39 increases 408.27MW, and relevant information is referring to table 3):
ΔPij=min (Δ PN,ΔPM,ΔPR)
ΔPN=min (Δ Pi.max,ΔPj.max)
ΔPM=max { β (Pl-Pl.lim)/Sl-ij|l∈L,Sl-ij> 0 }
ΔPR=min { γ (Pz-Pz.lim)/Sz-ij|z∈F,Sz-ij< 0 }
In above formula, Δ PNIt can adjust power output, Δ P for maximum of the control node to (i, j)i.maxFor the maximum of control node i Power output, Δ P can be addedj.maxPower output, Δ P can be subtracted for the maximum of control node jMThe control that (i, j) is needed to adjust for control node Amount, β, γ are penalty coefficient, take β=1.05, γ=0.95, P hereinl、Pl.limRespectively overload power and its pole of branch l Limit, Sl-ijFor in overload branch l plus, the pairing sensitivity of wind farm control node i, j, Δ PRFor the not out-of-limit control of normal branch Amount processed, F are normal branch constraint collection, and F=M ∪ B, B are the branch collection being affected by control node to (i, j), using together Heart relaxation principle is chosen to obtain, selection principle are as follows: the 3 layers of concentric circles searched for centered on control node i or j respectively are included Branch, then takes union, and z is the branch in F, Pz、Pz.limThe respectively power and its limit of branch z, Sz-ijFor in branch z Add, the pairing sensitivity of wind farm control node i, j;
The relevant information of calculating matched group (39,31) adjustment amount of table 3
Step 6, elder generation are according to Δ PijControl of decreasing load is carried out, sensitivity relation is recycled to refresh Branch Power Flow and correct node Power.The power of branch 2-1 falls to 580.56MW, load factor 96.76%, the load factor point of branch 1-39 and 5-4 after adjustment Kong Zhi not be 47.72%, 45.34%, remaining branch does not occur overload phenomenon, it is seen that control node is to (39,31) by a wheel System is just withdrawn into safe operation state from the state of emergency by control.

Claims (8)

1. one kind prevents chain overload controlling method, it is characterised in that:
The control method successively the following steps are included:
Step A, the crucial branch collection being made of overload branch collection L and heavy duty branch collection M is determined according to the operation information of each branch K, wherein the operation information includes overload branch l and its overload quantity, heavy duty branch m and its amount of redundancy;
Step B, pairing sensitivity of each control node to overload branch in broad sense power flow transfer region where computation overload branch, And effective control node is filtered out to set omega is formed according to pairing sensitivity, if set omega is empty set, reduce effectively control Node to filtering out effective control node to set omega is formed, until set omega is not empty set again after screening threshold value;
Step C, in set of computations Ω each effective control node pair improvement overall sensitivity, and choose improve overall sensitivity value Maximum control node is to as optimum control node pair;
Step D, the adjustment amount Δ P of optimum control node pair is calculatedij
Step E, first according to Δ PijControl of decreasing load is carried out, sensitivity relation is recycled to refresh Branch Power Flow and corrects node power output, Judge to overload whether phenomenon is eliminated simultaneously, if overload is eliminated, terminates premature beats;If overload phenomenon still remains, update Step B to step E is repeated after crucial branch collection K, until overload is eliminated.
2. one kind according to claim 1 prevents chain overload controlling method, it is characterised in that:
In step B, the pairing sensitivity is calculated using equal and opposite quantities in pairs method:
ΔPl=Sl-ijΔPj
Sl-ij=Sl-j-Sl-i
In above formula, Δ PlFor the active power variable quantity for overloading branch l, Sl-ijFor in overload branch l plus, subtract power output node i, j Match sensitivity, i ∈ B1∪B2, j ∈ B1, B1、B2Respectively with generator node collection, the load bus collection of adjustment amount, Δ Pj For the active injection variable quantity of control node j, Sl-iSensitivity for control node i to overload branch l indicates that control node i has Function power output increases the active change that branch l is overloaded when 1 unit power, corresponding reference mode such as balancing machine reduce by 1 unit power Change amount, B are the susceptance matrix of control node, and Δ θ is the voltage-phase variable quantity of control node, and Δ P is the active of control node Inject variable quantity, Δ PiFor the active injection variable quantity of control node i, Δ θlFor the phase difference for overloading the both ends branch l, xlFor mistake Carry the reactance of branch l.
3. one kind according to claim 1 or 2 prevents chain overload controlling method, it is characterised in that:
In step C, the improvement overall sensitivity SK-ijAre as follows:
In above formula, λ is the ratio between the weight for overloading branch and heavily loaded branch, αl、αmRespectively overload the power of branch l, heavily loaded branch m Weight, Sl-ijFor in overload branch l plus, the pairing sensitivity of wind farm control node i, j, Sm-ijTo add in heavily loaded branch m, subtracting out The pairing sensitivity of power control node i, j, Pl、Pl.limRespectively overload the power and its limit of branch l, Pm、Pm.limFor heavy duty The power and its limit of branch m.
4. one kind according to claim 1 or 2 prevents chain overload controlling method, it is characterised in that:
In step D, the adjustment amount Δ PijIt is calculated by following formula:
ΔPij=min (Δ PN,ΔPM,ΔPR)
Δ PN=min ((Δ Pi.max,ΔPj.max)
ΔPM=max { β (Pl-Pl.lim)/Sl-ij|l∈L,Sl-ij> 0 }
ΔPR=min { γ (Pz-Pz.lim)/Sz-ij|z∈F,Sz-ij< 0 }
In above formula, Δ PNIt can adjust power output, Δ P for maximum of the control node to (i, j)i.maxIt can add for the maximum of control node i Power output, Δ Pj.maxPower output, Δ P can be subtracted for the maximum of control node jMThe control amount that adjusts needed to (i, j) for control node, β, γ is penalty coefficient, Pl、Pl.limRespectively overload the power and its limit of branch l, Sl-ijTo add in overload branch l, subtracting out The pairing sensitivity of power control node i, j, Δ PRFor the not out-of-limit control amount of normal branch, F is normal branch constraint collection, F=M ∪ B, B are the branch collection being affected by control node to (i, j), and z is the branch in F, Pz、Pz.limThe respectively function of branch z Rate and its limit, Sz-ijFor in branch z plus, the pairing sensitivity of wind farm control node i, j.
5. one kind according to claim 4 prevents chain overload controlling method, it is characterised in that:
The branch collection B chooses to obtain using concentric relaxation principle, selection principle are as follows:
The branch that the 3 layers of concentric circles searched for centered on control node i or j respectively is included, then takes union.
6. one kind according to claim 1 or 2 prevents chain overload controlling method, it is characterised in that:
In step B, the set omega is that can eliminate effective control node of its overload to collection, Ω=J (l1)∪J(l2)∪…∪J (lN), { l1,l2,…,lN}=overload branch collection L, the screening technique of set omega are as follows:
For any overload branch l ∈ L, if control node is to (i, j) to the pairing sensitivity S of overload branch ll-ijMeet following Two conditions, then control node to (i, j) be referred to as overload branch l effective control node to and be divided into set J (l):
Sl-ij> 0
Sl-ij≥ηmax{Sl-ij|i∈B1∪B2,j∈B1}
In above formula, η is effective control node to screening threshold value, η ∈ [0,1], B1、B2Respectively with the generator section of adjustment amount Point set, load bus collection.
7. one kind according to claim 1 or 2 prevents chain overload controlling method, it is characterised in that:
In step B, broad sense power flow transfer region is obtained by power flow transfer regional expansion, expanding method are as follows:
It will abut against generator that number of nodes is 1 or load bus be divided into power flow transfer region where its adjacent node;
If certain branch breaking makes system sectionalizing, and the equal non-power generator-bus-bar system of each subsystem after off-the-line, then subsystem is drawn For different broad sense power flow transfer regions.
8. one kind according to claim 1 or 2 prevents chain overload controlling method, it is characterised in that:
In step B, for effective control node to (i, j), if control node i or j are cutpoint and are non-adjustable, selection is cut with this The maximum respective type node of controlled variable replaces the cutpoint in remaining connected broad sense power flow transfer region of point.
CN201811202701.8A 2018-10-16 2018-10-16 One kind preventing chain overload controlling method Pending CN109347108A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811202701.8A CN109347108A (en) 2018-10-16 2018-10-16 One kind preventing chain overload controlling method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811202701.8A CN109347108A (en) 2018-10-16 2018-10-16 One kind preventing chain overload controlling method

Publications (1)

Publication Number Publication Date
CN109347108A true CN109347108A (en) 2019-02-15

Family

ID=65309501

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811202701.8A Pending CN109347108A (en) 2018-10-16 2018-10-16 One kind preventing chain overload controlling method

Country Status (1)

Country Link
CN (1) CN109347108A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113410834A (en) * 2021-05-26 2021-09-17 佳源科技股份有限公司 Network reconstruction method and device applied to distribution network terminal
CN115178369A (en) * 2022-06-30 2022-10-14 宁波爱乐吉电动工具股份有限公司 Control method and system of pulverizer

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101299537A (en) * 2007-11-08 2008-11-05 江苏省电力公司 Centralized decision-making real time emergency control method for large electric network overloading
CN105305389A (en) * 2015-09-07 2016-02-03 许继集团有限公司 Emergency control method for preventing interlocking overloading of power transmission section
CN105740610A (en) * 2016-01-27 2016-07-06 云南电网有限责任公司电力科学研究院 Relay protection minimum break point set assessment method
CN107526002A (en) * 2017-07-10 2017-12-29 郑州轻工业学院 A kind of HVDC circuit fault distance measurements based on concentric relaxation principle
CN108647830A (en) * 2018-05-16 2018-10-12 清华大学 The value generation methods such as the boundary of security constraint inside the province towards transregional electricity transaction transprovincially

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101299537A (en) * 2007-11-08 2008-11-05 江苏省电力公司 Centralized decision-making real time emergency control method for large electric network overloading
CN105305389A (en) * 2015-09-07 2016-02-03 许继集团有限公司 Emergency control method for preventing interlocking overloading of power transmission section
CN105740610A (en) * 2016-01-27 2016-07-06 云南电网有限责任公司电力科学研究院 Relay protection minimum break point set assessment method
CN107526002A (en) * 2017-07-10 2017-12-29 郑州轻工业学院 A kind of HVDC circuit fault distance measurements based on concentric relaxation principle
CN108647830A (en) * 2018-05-16 2018-10-12 清华大学 The value generation methods such as the boundary of security constraint inside the province towards transregional electricity transaction transprovincially

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
姚薇薇,等: "电网连锁性故障序列的模糊综合评价方法", 《电力系统自动化学报》 *
徐岩: "基于功率灵敏度的线路过载划区域紧急控制策略", 《电工技术学报》 *
杨文辉: "预防连锁跳闸的关键线路后备保护与紧急控制策略研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 *
邓佑满,等: "电力系统有功安全校正策略的反向等量配对调整法", 《电力系统自动化》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113410834A (en) * 2021-05-26 2021-09-17 佳源科技股份有限公司 Network reconstruction method and device applied to distribution network terminal
CN115178369A (en) * 2022-06-30 2022-10-14 宁波爱乐吉电动工具股份有限公司 Control method and system of pulverizer

Similar Documents

Publication Publication Date Title
CN108988322B (en) Microgrid operation strategy optimization method considering system time-varying property
CN110932281B (en) Multi-section cooperative correction method and system based on quasi-steady-state sensitivity of power grid
CN106099997B (en) Consider the interconnected network spare shared capacity allocation method of security constraint
CN110896231B (en) Distributed photovoltaic capacity receiving calculation method and system for power distribution network in poverty alleviation area
CN106374453B (en) Power system reconstruction method
CN104779607B (en) One of direct-current micro-grid distributed and coordinated control method and system
CN107359616A (en) One kind solves the not convergent method of Load flow calculation after large scale electric network mode adjusts
CN107069812B (en) The distributed collaboration control method of more energy-storage units in grid type micro-capacitance sensor
CN110380404B (en) Power transmission network excitation system adjustment coefficient optimization setting method considering high energy consumption point load
CN109687497B (en) Prevention control method and system for inhibiting extra-high voltage direct current continuous commutation failure
CN108767906B (en) Wind-solar primary frequency modulation reserve capacity determination method considering cluster electric automobile
CN109347108A (en) One kind preventing chain overload controlling method
CN107451743B (en) Active power distribution network island operation method considering source load operation characteristics
CN104821605A (en) Active safety correction method based on improved particle swarm optimization algorithm
CN105826946B (en) The power distribution network dynamic reactive power optimization method of large-scale photovoltaic access
CN107069757B (en) A kind of mixing multi-infeed HVDC system secondary voltage control method
CN117895461A (en) Power coordination control method for multiple energy storage systems
CN111049185B (en) Island division method for power distribution network comprising distributed power supply
KR101277317B1 (en) Microgrid system
CN110854853B (en) Generating method and system for starting scheme of generator set
Alhammad et al. Monotonic Controller for Battery Energy Storage System Using Neural Network
CN104767225B (en) The computational methods that generating set is exerted oneself in power system recovery
CN104915723B (en) Interruptible load participation rotation standby coordination optimization method considering frequency recovery effect
CN114744674A (en) Voltage and power self-adaptive control method for photovoltaic access power distribution network
CN113991686A (en) Fault recovery method, device and storage medium for multi-energy coupling power distribution system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20190215