CN105591388A - Electric power system rectangular coordinate PQ decomposition method tidal data storage method based on sparse technology - Google Patents
Electric power system rectangular coordinate PQ decomposition method tidal data storage method based on sparse technology Download PDFInfo
- Publication number
- CN105591388A CN105591388A CN201610130999.0A CN201610130999A CN105591388A CN 105591388 A CN105591388 A CN 105591388A CN 201610130999 A CN201610130999 A CN 201610130999A CN 105591388 A CN105591388 A CN 105591388A
- Authority
- CN
- China
- Prior art keywords
- array
- node
- data
- row
- battle array
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 89
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 31
- 238000005516 engineering process Methods 0.000 title claims abstract description 8
- 238000013500 data storage Methods 0.000 title abstract description 6
- 238000003860 storage Methods 0.000 claims abstract description 52
- 238000003491 array Methods 0.000 claims abstract description 9
- 230000007935 neutral effect Effects 0.000 claims description 6
- 239000003990 capacitor Substances 0.000 claims description 2
- 230000001186 cumulative effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000007796 conventional method Methods 0.000 description 10
- 239000011159 matrix material Substances 0.000 description 10
- 230000003068 static effect Effects 0.000 description 3
- 238000010205 computational analysis Methods 0.000 description 2
- 238000002940 Newton-Raphson method Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
- H02J3/04—Circuit arrangements for AC mains or AC distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Complex Calculations (AREA)
Abstract
The invention provides an electric power system rectangular coordinate PQ decomposition method tidal data storage method based on sparse technology. According to the method of the invention, non-zero elements at upper and lower corners of three arrays of Y(n,2n), B'(n-1,n-1) and B''(n-1,n-1) in a traditional method are stored in an A(n,d) array, thereby greatly reducing number of storage units, improving data file reading and writing speed, and improving calculation speed of Ipi, Iqi or deltaPi and deltaQi in a PQ decomposition method trend. Furthermore the number of storage units is furthermore reduced by means a fact that the positions of non-zero elements in the arrays of the three arrays of Y(n,2n), B'(n-1,n-1) and B''(n-1,n-1). Compared with methods such as linked list method, the number of storage units is greatly reduced. Compared with the traditional method such as an IEEE-118 node system, the number of maximal storage units and the number of actual storage units of the electric power system rectangular coordinate PQ decomposition method tidal data storage method are respectively 10.00% and 4.48% of the number of maximal storage units and the number of actual storage units of the IEEE-118 node system, and data writing time and data reading time of the actual storage units of the electric power system rectangular coordinate PQ decomposition method tidal data storage method are respectively 11.45% and 6.20% of data writing time and data reading time of the IEEE-118 node system. Along with node number increase, the advantages of the electric power system rectangular coordinate PQ decomposition method tidal data storage method become more remarkable.
Description
Technical field
The invention belongs to Power System Analysis and calculate field.
Background technology
Newton-Raphson method (Newton method) is the most frequently used method during electric power system tide calculates, can be by by computational processIt is divided into rectangular co-ordinate Newton method and polar coordinates Newton method, can derive respectively polar coordinates PQ according to both characteristics simplifiedsDecomposition method and rectangular co-ordinate PQ decomposition method. In general rectangular co-ordinate Newton method is set forth in principle than polar coordinates Newton methodBetter understand with calculation and programming aspect, but computational process, polar coordinates Newton method is repaiied than rectangular co-ordinate Newton methodJust equational equation number will lack, and the kind of Jacobian matrix element and quantity are also few, and trend iterations is also few sometimes1 time or identical, therefore polar coordinates Newton method seems more extensive than the application of rectangular co-ordinate Newton method. By the same token,Polar coordinates PQ decomposition method has become the synonym of PQ decomposition method, almost without Introduction of Literatures rectangular co-ordinate PQ decomposition method.In fact,, owing to not relating to trigonometric function in the calculating of rectangular co-ordinate trend, this is especially sparse in consideration to many systemsIn the situation of property and rectangular co-ordinate trend design factor matrix feature, its memory cell number, trend iteration speed etc. are not sub-In polar index.
In power system, PQ decomposition method trend relates to coefficient matrix application multiple symmetries, extremely sparse in calculating.If do not consider the sparse property of these coefficient matrixes, the data of a large amount of neutral elements storage, read and computational process in can leadCause memory space significant wastage, file and computational process overlong time read and write data. Therefore, in conjunction with rectangular co-ordinate PQThe feature of decomposition method coefficient matrix also considers that sparse technology not only can save memory cell in a large number to the storage mode of element,Also can greatly reduce data file storage, read and time of computational process.
Rectangular co-ordinate PQ decomposition method trend needs to use 3 arrays in calculating, and wherein admittance matrix Y is for computing nodeElectric current (Ipi、Iqi) and node power (Δ Pi、ΔQi), coefficient matrix B ' is for calculating voltage imaginary part increment Delta fi, coefficientMatrix B is " for calculating voltage real part increment Delta ei. Storage to these 3 coefficient matrix data in conventional P Q decomposition method,Read, application etc. has the following disadvantages:
(1) " obtain manner of array element element is unreasonable for B ', B.
Due to Y, B ', B, " formation of array element element is larger on trend convergence or convergence rate impact, generally its element structureOne-tenth should be different. " array element element obtain manner is the imaginary part of directly taking from Y array element element, so only for the simplest B ', BStore 1 Y array element element. If " array element element is taken the imaginary part of Y array element element completely, if or only for but B ', B" array element element is identical and be different from the imaginary part of Y array element element, all may affect the iteration time of PQ decomposition method trend for B ', BNumber or convergence.
(2) " storage mode of array element element is unreasonable for Y, B ', B.
If the nodes that n is system, the PQ nodes that m is system is corresponding in traditional rectangular co-ordinate PQ decomposition methodArray be respectively Y (n, 2n), B ' (n-1, n-1), B " (n-1, n-1), corresponding respectively deposit extremely sparse Y, B ', B "Array element element. This storage mode can cause the significant wastage of memory space long with read and write data file and computing time,And with Y (n, 2n) array calculating Ipi、IqiOr Δ Pi、ΔQiEfficiency is very low. If with coordinate storage, sequential storing,The methods such as chained list storage, although can save many memory cells, make because diagonal element and nondiagonal element separate storageData to read process loaded down with trivial details, be unfavorable for its calculating and processing, calculate Ipi、IqiOr Δ Pi、ΔQiTime efficiency not high yet.
(3) Y, B ', B " battle array data file storage number more, access time is longer.
" three data files of battle array, at trend meter to store respectively Y, B ', B if do not considered sparse property by conventional methodIn calculation program, to open respectively three data files reading out data; As considered, sparse property is right respectively by chained list memory method" battle array is stored, and the number of storing file will reach 9 for Y, B ', B. The data file number of storage is more,The time of writing data file is longer, causes equally needing in PQ decomposition method flow calculation program the data file of openingThe number time more and read data file is longer, is unfavorable for the real-time calculating of PQ decomposition method.
(4) Y, B ', B " underuse by the relation between battle array nonzero element position.
Y, B ', B " array corresponding to battle array is respectively Y (n, 2n), B ' (n-1, n-1), B " (n-1, n-1). The exponent number of Y battle array isN, and B ', B " order of matrix number is (n-1). 1～(n-1) in order range, form in the process of Y, B ' battle array exponent number phasePosition and the number of its static nonzero element are just the same simultaneously; And in the forward steps of Y, B ' battle array, exponent number phasePosition and the number of its dynamic nonzero element are also just the same simultaneously. In 1st～m line range, B ', B " the each row of battle arrayThe position of nonzero element is identical in Static and dynamic situation with number, and at (m+1)～(n-1) row, B " battle arrayIn except diagonal element be-2, all the other elements are zero, with in the row of B ' battle array (m+1)～(n-1) in Static and dynamic situationThe position of nonzero element is different with number. In conventional method owing to not utilizing this two characteristics, thereby cause memory cellWaste and factor table forming process in some unnecessary calculating.
Summary of the invention
In order to overcome above-mentioned the deficiencies in the prior art, a kind of power system rectangular co-ordinate PQ based on sparse technology is proposedDecomposition method flow data memory method.
Three the array Y (n, 2n) that separately deposit, B ' (n-1, n-1) during the present invention calculates conventional P Q decomposition method trend," (n-1, n-1) uses respectively three virtual array Y (n, 3d to B1)、B′(n-1,d2)、B″(m,d2) corresponding substituting, merging is deposited inIn array A (n, d). In A (n, d), contain line number, each host node and the nonzero element child node sum of host node, Y (n, 3d1)Corresponding storage Y battle array information, for calculating Ipi、IqiOr Δ Pi、ΔQi, containing the row number of host node and nonzero element child nodeAnd parameter; B ' (n-1, d2)、B″(m,d2) corresponding storage B ', B " battle array information, calculates for completing PQ decomposition method trend,Wherein B ' (n-1, d2) in containing the parameter of host node and nonzero element child node, and B " (m, d2) in only containing the capable master of 1st～mThe parameter of node and nonzero element child node.
The present invention is achieved by the following technical solutions.
A kind of power system rectangular co-ordinate PQ decomposition method flow data storage side based on sparse technology of the present inventionMethod, comprises the following steps:
Step 1: definition data file array A (n, d);
(1) node corresponding with line number i is father node, and coupled nonzero element node is child node, neutral elementChild node do not occur;
(2) " three virtual arrays of array element element are Y (n, 3d to deposit Y, B ', B1)、B′(n-1,d2)、B″(m,d2)。Y(n,3d1) deposit row j and corresponding self-admittance and the transadmittance of all father nodes and nonzero element child node in arrayReal part and imaginary values, because the maximum non-zero entry prime number of each row father node and nonzero element child node sum isS1max＝d1, therefore its maximum number of column is 3 × d1。B′(n-1,d2)、B″(m,d2) all only deposit in array all father nodes andThe imaginary values of the corresponding self-admittance of nonzero element child node and transadmittance, and B " (m, d2) in only deposit PQ node correspondenceParameter, now the maximum non-zero entry prime number of each row father node and nonzero element child node sum is S2max＝d2, therefore itsMaximum number of column is d2, and the row j of its nonzero element can be by Y (n, 3d1) in obtain;
(3) by Y (n, 3d1)、B′(n-1,d2)、B″(m,d2) data of three arrays coexist in A (n, d) array, itsMiddle d=3d1+2d2+ 3. "+3 " are respectively line number row and S1max、S2maxThe count column at place, i.e. 1st～3 in table 1Row;
(4) the maximum storage unit number U of this methodmax.newCalculate by non-zero entry prime number maximum in each row:
Umax.new＝U1max+U2max+U3max＝Y(n,3d1)+B′(n-1,d2)+B″(m,d2)
＝n×(d1+2)+(n-1)×(d2+1)+m×d2。
Note: "+2 " correspond to line number row and Y battle array count column; " the count column of battle array that "+1 " corresponds to B ' battle array and B;
(5) the actual memory cell of this method is counted Uact.newBy each row SiThe non-zero entry prime number sum of actual count is calculated,Uact.new＝Y(n,3d′1)+B′(n-1,d′2)+B″(m,d′2). Computational analysis shows, the actual memory cell number of new method accounts for50% left and right of new method maximum storage unit number, and percentage reduces along with the increase of nodes;
(6) A (n, d) array is divided into " line number group ", " node array ", " Y battle array group ", " B ' battle array group ", " B " battle array group ",Store form is as follows:
Line number group i: deposit the line number corresponding with father node, be positioned at the 1st row;
Node array Si1、Si2: deposit Y, B ' (each row father node and nonzero element son node number sum in B ") battle array, positionIn the 2nd, 3 row, Si1、Si2Value is automatically cumulative to ensure to read quickly and efficiently corresponding father node and non-zero by programThe parameter of element child node, thus make actual memory cell number of the present invention much smaller than its maximum storage unit number;
Y battle array group: deposit (n, 3d with Y1) Y battle array data corresponding to array, be positioned at the 4th～(3d1+ 3) row, press row number and increase progressivelyThe row j of sequential storing father node and nonzero element child node and corresponding self-admittance, the real part of transadmittance, imaginary values;
B ' battle array group: deposit (n-1, d with B '2) B ' battle array data corresponding to array, be positioned at (3d1+4)～(3d1+d2+ 3) row, pressThe imaginary values of row incremental order storage father node and the corresponding self-admittance of nonzero element child node, transadmittance, its non-zeroThe row j of element can be by Y (n, 3d1) in obtain;
B " battle array group: deposit and B " (m, d2) " battle array data, are positioned at (3d to B corresponding to array1+d2+4)～(3d1+2d2+3)Row, the imaginary values of pressing row incremental order storage father node and the corresponding self-admittance of nonzero element child node, transadmittance,The row j of its nonzero element can be by Y (n, 3d1) in obtain.
The storage mode of A (n, d) array element is as shown in table 1.
The storage mode of table 1A (n, d) array element
Note: in table 1, except containing the row that maximum nonzero element node is corresponding, be not that all memory cells have data,And use Si1、Si2Can ensure that actual memory cell number is far smaller than maximum storage unit number, further improves the read-write of dataEfficiency.
Step 2: read in all branches data from data file;
" all elements of battle array is also stored in data in A (n, d) array step 3: calculate Y, B ', B;
Y array element element generally includes real part and the imaginary part of the Y array element element that all branch road parameters form, only for follow-up PQI in decomposition method Load Flow Programpi、IqiOr Δ Pi、ΔQiCalculating; " array element element solves Δ f for down-stream for B ', Bi、Δei; Generally in the process of calculating B ' array element element, remove line mutual-ground capacitor c and calculating B " in the process of array element elementRemove line resistance r;
Step 4: the data data writing file of A (n, d) array is used in order to down-stream.
Consider the modularization of program, the program that forms A (n, d) array leaves it at that, and A (n, d) array data fileCall and carried out by PQ decomposition method flow calculation program.
In PQ decomposition method flow calculation program, open the data file of A (n, d) array, by Y battle array data read in Y (n,3d1) array to be to calculate fast Ipi、IqiOr Δ Pi、ΔQi; By B ', B " battle array data read in respectively B ' (n-1, n-1) and B " (n-1,N-1) array is to solve Δ fi、Δei. Read in the data of A (n, d) data file than reading in respectively Y (n, 2n), B ' (n-1, n-1)" the data required time of (n-1, n-1) data file is wanted much less to and B.
The present invention is with the traditional Y that does not consider the sparse property of element, B ', B " compared with the storage mode of array element element, by multipleThe storage of data file is simplified to the storage of 1 data file, will the storage of all data be simplified to only to non-zero entry" " battle array non-zero entry that the almost identical relation in battle array nonzero element position, saves B ', B that the storage of element, utilizes Y, B ', BThe lower target storage of element row and to B " storage of battle array PV node corresponding data, greatly reduced data memory cell andTo the access time of data file and calculating Ipi、IqiOr Δ Pi、ΔQiTime. The present invention with press coordinate store, by suitableOrder is stored, is compared by methods such as chained list storages, still can more reduce the memory cell of data and the storage of data fileQuantity, has improved the read or write speed to data file, and storage mode is more simple, the computing of follow-up data etc.More convenient, quick.
Brief description of the drawings
Fig. 1 is the flow chart that the present invention forms rectangular co-ordinate PQ decomposition method flow data file.
Fig. 2 is the flow chart that conventional method forms rectangular co-ordinate PQ decomposition method flow data file.
Detailed description of the invention
The present invention will be described further by following examples.
Embodiment. The chained list memory method of more traditional memory method of not considering sparse property and the sparse property of consideration respectively" the plain required data file of array element that and the inventive method is stored IEEE-30 ,-57 ,-118 node system Y, B ', BThe average time of number, memory cell number, data file read-write process. Table 2 has provided the comparative result of the whole bag of tricks.
(1) comparison of data file number.
1) traditional memory method respectively by Y, B ', B " array element element leaves respectively Y (n, 2n), B ' (n-1, n-1) and B in " (n-1,N-1), in 3 arrays, need 3 data files;
2) chained list memory method stores diagonal element, off-diagonal element, line number (chained list) respectively by 3 arrays,Therefore " array element element needs 9 arrays, needs 9 data files to store respectively corresponding Y, B ', B;
3) the present invention is by Y, B ', B, and " array element element leaves in A (n, d) array jointly, only needs 1 data file;
(2) storage Y, B ', the B " comparison of array element element unit number.
If the nodes that n is system, the PQ nodes that m is system; Average each node is connected with 4 branch roads;Owing to being complex matrix, Y array element element × 2. To IEEE-118 node system, n=118, m=64,3d1＝3×S1max＝30，d2＝S2max＝9。
1) Traditional Method memory cell number: Uc＝n×2×n+2(n-1)×(n-1)＝55226；
2) chained list method memory cell number: Ut＝(2n+2N+N)＝(2n+3N)＝(14×n)＝6580；
Calculate for simplifying, be similar to herein and think that " exponent number of battle array is (n-1) for B ', B. Due to the calculating public affairs of above-mentioned chained list methodFormula is applicable to real number matrix, does not consider Y array element element × 2, and therefore depositing of chained list method of calculating are calculated in this simplification not moreStorage unit number.
3) maximum storage unit number of the present invention: calculate by non-zero entry prime number maximum in each row:
Umax.new＝U1max+U2max+U3max＝n×(3d1+2)+(n-1)×(d2+1)+m×d2＝5522。
Due to d=3d1+2d2+ 3, therefore to IEEE-118 node system, use A (118,51) array can store all information.
4) the actual memory cell number of the present invention: by S in each rowiThe non-zero entry prime number sum of actual count is calculated:
Uact.new＝Y(n,3d′1)+B′(n-1,d′2)+B″(m,d′2)＝2474。
Above-mentionedly relatively can find out, maximum storage unit number of the present invention accounts for 10.00% of conventional method memory cell number,And reduce along with the increase of nodes; Actual memory cell number of the present invention accounts for new method maximum storage unit number44.80%, also reduce along with the increase of nodes; And actual memory cell number of the present invention accounts for conventional method storage list4.48% of unit's number, the same significantly minimizing along with the increase of nodes; Actual memory cell number of the present invention accounts for chained list37.60% of method memory cell number, reduces along with the increase of nodes equally. Therefore nodes is larger, the present invention's jointThe memory cell of economizing is more, and to Ipi、IqiOr Δ Pi、ΔQiComputational efficiency high.
The comparison of table 2 the whole bag of tricks to IEEE system data file read-write time and memory cell number
tw.c、tr.c、Uc: conventional method is to Y, B ', B, and " battle array data file is on average write read time, required memory cell number;
tw.new、tr.new、Umax.new、Uact.new: the present invention is to Y, B ', B, and " battle array data file is on average write read time, instituteThe maximum, the actual memory cell number that need;
tw.new/tw.c、tr.new/tr.c、Umax.new/Uc、Uact.new/Uc: the present invention and conventional method on average write read time,The percentage of large memory cell number, the percentage of actual required memory cell number.
Ut、Uact.new/Ut: the required memory cell number of chained list method, actual memory cell number of the present invention and chained list method are requiredThe percentage of memory cell number.
Above-mentioned computational analysis shows:
1. the speed of memory cell number of the present invention and write-read data file is better than conventional method greatly.
The present invention by 3 required PQ decomposition method data file coexistences in 1 array, with conventional method by 3Data file is separated storage mode and is compared, and has greatly reduced number, memory cell and the data file of desired data fileWrite read time. As to IEEE-118 node system, the time that the present invention writes data file is only for conventional method11.45%, the time of read data file is only 6.20% of conventional method, and maximum storage unit number is only conventional method10.00%, and physical memory location is only 4.48% of conventional method.
2. memory cell number of the present invention is better than considering the chained list storage mode of sparse property equally.
The present invention only stores 1 data file, chained list method need be stored 9 data files; The maximum of the inventive method is depositedStorage unit number is about 60%～80% of chained list method, but the actual memory cell number of the inventive method is about chained list method40％。
3. power system nodes is more, and data-storing of the present invention is larger with the advantage that reads process.
The present invention is along with the increase memory cell of system node number can't significantly increase, and the time of write-read data file alsoCan significantly not increase.
4. the factor table of " therefore the nonzero element of the upper and lower triangle of battle array is forming B ', B " owing to storing Y, B ', B simultaneouslyBattle array, calculating Ii、ΔPi、ΔQiTime will be very convenient, and reading and writing data time or memory cell number and the upper triangle of storage onlyThe mode of nonzero element is compared does not almost have difference.
5. the present invention is according to Y, B ', B, and " the almost identical relation in battle array nonzero element position, to B ', B, " battle array is only stored non-Neutral element parameter, and the row subscript of nonzero element is directly used Y (n, 3d1) in the row subscript of storing.
6. consider B " particularity of PV node data in battle array, B of the present invention " (m, d2) in only store PQ node data,And directly PV node data is composed again call the data file of A (n, d) array at PQ decomposition method Load Flow Program afterBe worth, further reduce the storage capability of element.
7. for read-write and the application of the multiple asymmetric data files of various engineering fields, the present invention has same advantage.
The present invention can adopt any programming language and programmed environment to realize, and adopts C++ programming language here, opensHair ring border is visual c++, and operation platform is Intel (R) Corei7-4790CPU3.60GHZ, internal memory 8.00GB.
Claims (1)
1. the power system rectangular co-ordinate PQ decomposition method flow data memory method based on sparse technology, its featureComprise the following steps:
Step 1: definition data file array A (n, d);
(1) the definition node corresponding with line number i is father node, and coupled nonzero element node is child node, zeroThe child node of element and parameter do not appear in memory cell;
(2) definition is deposited Y, B ', B " three virtual arrays of array element element is Y (n, 3d1)、B′(n-1,d2)、B″(m,d2)；Y(n,3d1) deposit row j and the corresponding parameter of all father nodes and nonzero element child node, corresponding maximum in arrayNon-zero entry prime number is S1max; And B ' (n-1, d2)、B″(m,d2) all only deposit all father nodes and non-zero entry sub-prime in arrayThe corresponding parameter of node, the row j of its nonzero element all can be by Y (n, 3d1) in obtain, corresponding maximum nonzero elementNumber is S2max, and B " (m, d2) in only deposit the parameter that PQ node is corresponding;
(3) by Y (n, 3d1)、B′(n-1,d2)、B″(m,d2) data of three arrays coexist in A (n, d) array;
(4) be defined as maximum storage unit number U by non-zero entry prime number maximum in each rowmax.new；
(5) by each row SiThe non-zero entry prime number of actual count is defined as actual memory cell and counts Uact.new；
(6) A (n, d) array is divided into " line number group ", " node array ", " Y battle array group ", " B ' battle array group ", " B " battle array group ",Store form is as follows:
Line number group i: be the use of check data, deposit the line number corresponding with father node;
Node array Si1、Si2: be the use efficiently reading and writing data, deposit Y, B ' (each row father node and non-zero in B ") battle arrayElement son node number sum, its Si1、Si2Value is automatically cumulative by program;
Y battle array group: deposit (n, 3d with Y1) Y battle array data corresponding to array, press row incremental order storage father node and non-The row j of neutral element child node and corresponding self-admittance, the real part of transadmittance, imaginary values;
B ' battle array group: deposit (n-1, d with B '2) B ' battle array data corresponding to array, press row incremental order storage father node and non-The imaginary values of the corresponding self-admittance of neutral element child node, transadmittance, its nonzero element row j is by Y (n, 3d1) in obtain;
B " battle array group: deposit and B " (m, d2) " battle array data are pressed row incremental order storage father node and non-to B corresponding to arrayThe imaginary values of the corresponding self-admittance of neutral element child node, transadmittance, its nonzero element row j is by Y (n, 3d1) in obtain;
Step 2: read in all branches data from data file;
" all elements of battle array is also stored in data in A (n, d) array step 3: calculate Y, B ', B;
Y array element element generally includes real part and the imaginary part element of the Y battle array that all branch road parameters form, only for follow-up PQI in decomposition method Load Flow Programpi、IqiOr Δ Pi、ΔQiCalculating; " array element element solves Δ f for down-stream for B ', Bi、Δei, in the process of calculating B ' array element element, remove line mutual-ground capacitor c and " in the process of array element element, remove calculating BLine resistance r;
Step 4: the data data writing file of A (n, d) array is used in order to down-stream.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610130999.0A CN105591388B (en) | 2016-03-08 | 2016-03-08 | A kind of electric system rectangular co-ordinate PQ decomposition method flow data memory methods based on Sparse technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610130999.0A CN105591388B (en) | 2016-03-08 | 2016-03-08 | A kind of electric system rectangular co-ordinate PQ decomposition method flow data memory methods based on Sparse technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105591388A true CN105591388A (en) | 2016-05-18 |
| CN105591388B CN105591388B (en) | 2018-08-03 |
Family
ID=55930727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610130999.0A Expired - Fee Related CN105591388B (en) | 2016-03-08 | 2016-03-08 | A kind of electric system rectangular co-ordinate PQ decomposition method flow data memory methods based on Sparse technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105591388B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107957974A (en) * | 2017-11-29 | 2018-04-24 | 广西大学 | A kind of method for extracting polar coordinates power flow equation Jacobian matrix |
| CN109446551A (en) * | 2018-09-13 | 2019-03-08 | 南昌大学 | The formation and application of random nonzero element factor table in electric system PQ method trend |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317553A (en) * | 2014-10-13 | 2015-01-28 | 南昌大学 | Method for fast forming, reading and writing power system node admittance matrix data based on sparse matrix technology |
| CN104933528A (en) * | 2015-06-24 | 2015-09-23 | 南昌大学 | Method for quickly forming jacobian matrix in electric system load flow calculation based on sparse matrix technology |
| CN105045767A (en) * | 2015-06-24 | 2015-11-11 | 南昌大学 | Method for fast storing and reading sparse matrix data of electric power system |
| CN105786984A (en) * | 2016-02-15 | 2016-07-20 | 南昌大学 | Rapid-reading-writing electric-power-system PQ-decomposition-method flow data storage method based on sparse technology |
-
2016
- 2016-03-08 CN CN201610130999.0A patent/CN105591388B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317553A (en) * | 2014-10-13 | 2015-01-28 | 南昌大学 | Method for fast forming, reading and writing power system node admittance matrix data based on sparse matrix technology |
| CN104933528A (en) * | 2015-06-24 | 2015-09-23 | 南昌大学 | Method for quickly forming jacobian matrix in electric system load flow calculation based on sparse matrix technology |
| CN105045767A (en) * | 2015-06-24 | 2015-11-11 | 南昌大学 | Method for fast storing and reading sparse matrix data of electric power system |
| CN105786984A (en) * | 2016-02-15 | 2016-07-20 | 南昌大学 | Rapid-reading-writing electric-power-system PQ-decomposition-method flow data storage method based on sparse technology |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107957974A (en) * | 2017-11-29 | 2018-04-24 | 广西大学 | A kind of method for extracting polar coordinates power flow equation Jacobian matrix |
| CN107957974B (en) * | 2017-11-29 | 2020-12-18 | 广西大学 | Method for extracting polar coordinate tidal current equation Jacobi matrix |
| CN109446551A (en) * | 2018-09-13 | 2019-03-08 | 南昌大学 | The formation and application of random nonzero element factor table in electric system PQ method trend |
| CN109446551B (en) * | 2018-09-13 | 2023-03-14 | 南昌大学 | Formation and application of random non-zero element factor table in power system PQ method power flow |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105591388B (en) | 2018-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104317553B (en) | Method for fast forming, reading and writing power system node admittance matrix data based on sparse matrix technology | |
| Van der Sluis | Condition numbers and equilibration of matrices | |
| WO2010039325A1 (en) | Method for solving reservoir simulation matrix equation using parallel multi-level incomplete factorizations | |
| CN109359269A (en) | It carries out the system of matrix-vector multiplication operation and carries out neural network computing method | |
| CN104933528B (en) | A kind of method that Jacobian matrix during electric power system tide calculates quickly is formed based on sparse matrix technology | |
| JP7461081B2 (en) | Method and apparatus for deconvolution processing of feature data using convolution hardware | |
| CN105591388A (en) | Electric power system rectangular coordinate PQ decomposition method tidal data storage method based on sparse technology | |
| CN113628329B (en) | Zero-sample sketch three-dimensional point cloud retrieval method | |
| CN102832616B (en) | Node voltage calculation method and device for power system | |
| CN105354422B (en) | A method of polar coordinates Newton-Raphson approach trend is quickly sought based on symmetrical and sparse technology | |
| CN103034621A (en) | Address mapping method and system of radix-2*K parallel FFT (fast Fourier transform) architecture | |
| CN105045767B (en) | A kind of method of immediate access and reading power system sparse matrix data | |
| CN107563080B (en) | GPU-based two-phase medium random model parallel generation method and electronic equipment | |
| CN105786984B (en) | It is a kind of based on Sparse technology can fast reading and writing electric system PQ decomposition method flow data memory method | |
| CN110826186B (en) | LR triangular decomposition method based on symmetric sparse matrix technology and non-zero element random storage | |
| CN105786769B (en) | Application of method based on rapid data reading and symmetric sparse factor table in polar coordinate PQ decomposition method trend | |
| CN109447839B (en) | Method for randomly generating non-zero elements and forming Jacobian matrix of power system | |
| Riker et al. | The mixed-cell-complex partition-of-unity method | |
| CN105703359A (en) | Application of sparse symmetric factor table method in PQ decomposition method-based load flow calculation in rectangular coordinate system | |
| Estabrook et al. | The geometric approach to sets of ordinary differential equations and Hamiltonian dynamics | |
| CN105375468B (en) | A kind of method that rectangular co-ordinate Newton-Laphson method trend is quickly asked for based on symmetrical and sparse technology | |
| D’Adda et al. | Two-Dimensional Gauge Theories of the Symmetric Group S n in the Large-n Limit | |
| Bullett et al. | A gallery of iterated correspondences | |
| EP2867791A1 (en) | Asynchronous distributed computing based system | |
| CN103995812A (en) | Similar commodity searching method and system and mobile terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180803 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |