CN112036553B - Genetic algorithm-based non-signal injection type household phase topological relation identification method - Google Patents
Genetic algorithm-based non-signal injection type household phase topological relation identification method Download PDFInfo
- Publication number
- CN112036553B CN112036553B CN202011126494.XA CN202011126494A CN112036553B CN 112036553 B CN112036553 B CN 112036553B CN 202011126494 A CN202011126494 A CN 202011126494A CN 112036553 B CN112036553 B CN 112036553B
- Authority
- CN
- China
- Prior art keywords
- data
- phase
- num
- pop
- gene
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 41
- 230000002068 genetic effect Effects 0.000 title claims abstract description 32
- 238000002347 injection Methods 0.000 title claims abstract description 12
- 239000007924 injection Substances 0.000 title claims abstract description 12
- 210000000349 chromosome Anatomy 0.000 claims abstract description 99
- 239000011159 matrix material Substances 0.000 claims abstract description 42
- 230000008569 process Effects 0.000 claims abstract description 19
- 238000004140 cleaning Methods 0.000 claims abstract description 10
- 238000013507 mapping Methods 0.000 claims abstract description 6
- 108090000623 proteins and genes Proteins 0.000 claims description 66
- 238000009826 distribution Methods 0.000 claims description 56
- 230000008859 change Effects 0.000 claims description 35
- 238000005070 sampling Methods 0.000 claims description 17
- XOFYZVNMUHMLCC-ZPOLXVRWSA-N prednisone Chemical compound O=C1C=C[C@]2(C)[C@H]3C(=O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 XOFYZVNMUHMLCC-ZPOLXVRWSA-N 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000035772 mutation Effects 0.000 claims description 12
- 238000005192 partition Methods 0.000 claims description 11
- 238000012216 screening Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 230000005070 ripening Effects 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000003491 array Methods 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000005065 mining Methods 0.000 abstract description 5
- 238000013473 artificial intelligence Methods 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 239000003446 ligand Substances 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 238000000547 structure data Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 208000035126 Facies Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000004043 dyeing Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010429 evolutionary process Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a non-signal injection type household phase topological relation identification method based on a genetic algorithm, which comprises the following steps: obtaining effective data after data cleaning according to the electrical quantity data information; randomly generating an initialized chromosome population data matrix pop_data according to the effective data; entering a circulation process: calling a fitness function; calling a selection function, a cross function and a variation function of a genetic algorithm to obtain updated newPopdata; entering a reset function call of a genetic algorithm; and obtaining a final NewPopdata, and obtaining a discrimination result best_Pop of each user table corresponding to each row through a mapping relation. The invention can solve the problem of topology identification of the low-voltage transformer area by utilizing the existing data of the mining system, namely, other equipment is not required to be installed.
Description
Technical Field
The invention relates to the technical field of power distribution network topology identification, in particular to a non-signal injection type household phase topology relation identification method based on a genetic algorithm.
Background
At present, the identification of the household-phase topological relation of the low-voltage station area is mainly a line checking method: for the power supply of the users in the platform area, each user meter box is opened, the specific wiring mode of each user meter is checked, the specific use is determined, and then the record marking is carried out, and the method is characterized in that: the workload is great, the efficiency is low, and once the user changes the line or increases the capacity, the user can re-mark the line.
The existing low-voltage distribution topology of the transformer area is basically blank or needs to be manually participated, the work is complicated, the time is long, automatic updating and identification cannot be achieved, and an adaptive topology identification method of the low-voltage distribution topology of the transformer area is urgently needed to solve the problems.
Meanwhile, the system for collecting electric energy of the low-voltage side outgoing line multifunctional meter and district household meter of the distribution transformer area is widely applied in the whole country, and the system collects the electric energy data of the low-voltage side outgoing line multifunctional meter and district household meter of the distribution transformer area, and comprises the following steps: voltage, current, active, reactive, and electrical metrics, etc.; although the data acquisition mode, the communication mode, the management architecture and the like are slightly different, the daily accumulation and the monthly accumulation of the massive data clearly provide the possibility for intelligent analysis of big data.
Disclosure of Invention
The invention aims to: aiming at the defects of the prior art, the invention provides a non-signal injection type household phase topological relation identification method based on a genetic algorithm, which is based on a large amount of data accumulated by a sampling system, adopts an artificial intelligence Genetic Algorithm (GA) to automatically infer a low-voltage topological connection relation of a platform area, automatically generates a household-phase topological relation, and can achieve an accurate identification rate of not less than 97 percent according to two months of data accumulated by the existing user acquisition system, thereby basically meeting the commercial demand.
The technical scheme is as follows: the low-voltage distribution network topology identification method based on the cooperation of the unilateral optimization algorithm and the genetic algorithm, disclosed by the invention, comprises the following steps of:
(1) Obtaining effective data after data cleaning according to the electrical quantity data information;
(2) Randomly generating an initialized chromosome population data matrix pop_data according to the effective data;
(3) Entering a circulation process:
(4) Calling an Fitness function, returning the proximity delta E between the distribution transformer side and the corresponding user table in each chromosome, and converting the delta E into Fitness for measuring the proximity degree in each chromosome;
(5) Calling a selection function, a cross function and a variation function of a genetic algorithm to obtain updated newPopdata, wherein the updated newPopdata contains chromosomes with Pop_num selection results;
(6) Setting an early ripening condition, entering a reset function call of a genetic algorithm, if the early ripening condition is met, entering a step (2), otherwise, continuing a circulation process;
(7) And obtaining a final NewPopdata, and obtaining a discrimination result best_Pop of each user table corresponding to each row through a mapping relation.
Further, the electrical quantity data is any one of current, voltage, reactive power, electric energy and active power, and the data cleaning method comprises the steps of eliminating all 0 family table sets, eliminating weak value family table sets and eliminating invalid time sample space sets in the effective family table sets; screening the cleaned data comprises obtaining P in the electrical quantity data of the outgoing line side of the transformer in the transformer area a ，P b ,P c ,P Total-variation The data obtained from the user side ammeter sample has P a ，P b ,P c ,P General-household Obtaining P of outgoing line of distribution transformer Total-variation And accumulating and calculating an error delta P corresponding to the user table P at the same sampling moment, setting a parameter epsilon, and when delta P is less than or equal to epsilon, reserving the distribution transformer at the moment and the sampling data of the user electric meter, adding the distribution transformer and the sampling data of the user electric meter into an effective data sample set, otherwise, giving up epsilon and the quantity K of the effective moments satisfy the following conditions:
ε≤10％,K≥300；
ε≤15％,K≥400；
ε≤20％,K≥600；
ε≤25％,K≥1200；
ε≤35％,K≥2000。
further, in the step (1), according to the electric quantity data information, effective data after data cleaning is obtained, and a storage list ValidDataSet is returned, wherein each qualified sample comprises a combined set sample set of electric quantity of a station distribution transformer and three phases and a single phase of users, m_num in the storage list ValidDataSet represents the total number of user tables, validcount_num represents the qualified number of data samples, and a distribution transformer side total electric quantity matrix E is obtained according to the data of the storage list ValidDataDet Total of distribution transformer [j]Distribution side A, B, C single-phase electric quantity matrix E 1-a [j]、E 1-b [j]、E 1-c [j]Single-phase electric quantity matrix for sum household meter level
The step (2) randomly generates M×N initialized chromosome population data matrix pop_data, M is the number of chromosome initial population, N=3×m_num,3 represents the discrimination result of the household table with only A, B, C phases, and randomly generates chromosome population according to m_num and pop_num
Further, the fitness function construction process is as follows:
calculating the change rate delta E of the A/B/C three phases of the distribution transformer 1-a [i]、ΔE 1-b [i]、ΔE 1-c [i]Calculating the change rate delta E of the user meter P 1-a1 、ΔE 1-b1 、ΔE 1-c1 ；
Obtaining Pop_data_A [ i, j ], pop_data_B [ i, j ], pop_data_C [ i, j ] matrix;
obtaining E 1-a1 Rate of change matrix, E 1-b1 Rate of change matrix, E 1-c1 A rate of change matrix;
calculating the cumulative sum delta E of the distribution change rate and the absolute value of the corresponding household table change rate a ，ΔE b ，ΔE c ；
Proximity delta E between distribution transformer and corresponding user table in each chromosome 1 ＝ΔE a +ΔE b +ΔE c ；
Calculating Fit [ p ]]＝ΔE 1 [p],p＝1,2,…Pop_num-1，Fit_max＝max(Fit[p])，
Fitness[P]＝Fit_max-Fit[P],P＝1,2,…,Pop_num-1。
Further, the selection function construction method comprises
Fitness normalization:
P_fitness[j]＝Fitness[j]/Sumfit_ness，j＝0,1,…,Pop_num-1；
the 5 chromosomes corresponding to the P_fitness [ j ] from big to small are placed in the first five chromosomes to be ranked to store a Fit_max5[ w ] array;
randomly selecting ms and ms epsilon [0,1] values, and meeting the conditions: ms.ltoreq.max (P_fitness [ j ]), j=0, 1,2, …, pop_num-1.
Further, the cross function is: selecting an operation random number c_rand epsilon (0, 1) and crossover probability factors pc and pc epsilon (0.8,1), and if c_rand is more than or equal to pc, not performing crossover operation on the two groups of chromosomes, wherein the chromosome groups keep the original genes unchanged; if c_rand < pc is satisfied, the chromosome set performs a crossover operation.
Further, the variation function is:
1) Judging whether mutation operation is needed or not: when the random number count_rand is more than or equal to pm, the count_rand epsilon (0, 1) is not operated at this time, and when the count_rand is less than pm, the next operation is continued;
2) Selecting a random integer t 1 The_rand is less than or equal to Poplenge-1, androunding downwards, when t 1 1=0 represents the gene of phase a, when t 1 1=1 represents the gene of phase B, when t 1 1=2 represents the gene of the C phase, t is taken as 1 The value of the_rand gene is put in the variable t_sample_1;
3) The same steps as above, the second random number t is taken 2 _rand≤Poplenge-1，t 2 _rand≠t 1 And (d) rand, and rounding downwards, when t 2 1=0 represents the gene of phase a, when t 2 1=1 represents the gene of phase B, when t 2 1=2 represents the gene of the C phase, t is taken as 2 The value of the_rand gene is put in the variable t_sample_2;
4) The discriminating process comprises the following steps:
4.1 If t_pattern_1+t_pattern_2=0, the gene bit to be exchanged belongs to an invalid bit, the mutation operation is not performed, and the process returns to 3) to continue random selection;
4.2 If t_temp_1+t_temp_2+.0 and t 1 _1≠t 2 And 1, carrying out exchange positions on the gene bit values to be exchanged in different phase regions, storing the gene bit values into corresponding arrays, and ending the chromosome mutation operation.
Further, when the accumulated value exceeds the pre-prepared reselection rate when the cyclic body is executed for not more than half times, the calculation of the genetic algorithm is carried out again through a reselection mechanism: sequencing the Pop_data gene values from large to small; the maximum number of the same gene values was counted and recorded as num.
Further, each household table corresponds to the discrimination result of each row of the best_pop [ m_num ] [6], and the discrimination result comprises a 0 th bit representative household table number; bit 1=1/2/3, respectively representing a phase a, a phase B, and a phase C; bit 2 represents the degree of reliability of discrimination=0/1, respectively representing unreliable/trusted; bits 3,4 and 5 respectively represent the numbers of phases A, B and C in the group
The beneficial effects are that: compared with the prior art, the invention has the advantages that: the artificial intelligence algorithm based on the electrical characteristics has certain universality, can be popularized to the fields of power transmission, transformation and distribution, lays a foundation for the analysis of big data of a power grid and the deep application of artificial intelligence, can be widely applied to the field of the ubiquitous Internet of things of electric power, and brings considerable social and economic benefits. As follows:
1. the problem of low-voltage transformer area topology identification can be solved by utilizing the existing data of the mining system, namely, other equipment is not required to be installed, and a person is not required to be dispatched to perform actual measurement and investigation on site, so that the large expense is saved. Meanwhile, the efficiency is greatly improved, and the operation management level is improved.
2. The operation condition of the station area can be effectively monitored, and more reliable basis and means are provided for monitoring, analyzing and treating the low-voltage three-phase imbalance.
3. And acquiring data according to the user table, and providing auxiliary reference for the deep analysis of the load characteristics of the user and the tracking of the dynamic operation characteristics of the load.
Drawings
FIG. 1 is a schematic diagram of a system for use;
FIG. 2 is a functional block diagram of the present invention;
FIG. 3 is a flow chart of a method implementation in the present invention;
FIG. 4 is a flow chart of a master function in the present invention;
FIG. 5 is a flow chart of the cumulative number of times a user belongs to an A/B/C user facies;
FIG. 6 is a flow chart of statistical analysis of 10 output subscriber table belonging to a zone discrimination;
FIG. 7 is a flow chart of reliability discrimination;
FIG. 8 is a schematic diagram of a chromosome set to be crossed;
FIG. 9 is a schematic diagram of crossover rules for i chromosome crossover to i+1 chromosome;
FIG. 10 is a schematic diagram of crossover rules for crossing i+1 chromosomes to i chromosomes.
Detailed Description
The technical scheme of the invention is described in detail below through the drawings, but the protection scope of the invention is not limited to the embodiments.
The invention is based on a large amount of data accumulated by the acquisition system, adopts an artificial intelligence Genetic Algorithm (GA), automatically infers the low-voltage topological connection relation of the platform area, automatically generates the household-phase topological relation, and can achieve the accurate recognition rate of not less than 97% according to the accumulated two month data of the existing user acquisition system, thereby basically meeting the commercial demand.
The genetic algorithm is an adaptive global optimization probability search algorithm formed by simulating the genetic and evolutionary processes of organisms in natural environments, and mainly comprises 3 processes of selection, crossing and variation.
The mining system as shown in fig. 1 comprises:
(1) Low pressure side: refers to a low voltage user intelligent device, comprising: automatic equipment such as household watches, leakage protection, piezoelectric capacitors, SVG/SVC and the like of residential users; the sampled information includes: voltage, current, active, reactive, and electrical metrics, etc.; various communication means may be employed to communicate with the cell concentrator, such as: LORA\broadband carrier\RS 485\network\narrowband carrier, etc.; the sampling time interval can be 1 point/5 min, 1 point/15 min, 1 point/60 min and 1 point/day according to different communication modes and requirements.
(2) Station area side: the system refers to an automatic device such as a station area acquisition concentrator, a distribution transformer low-voltage outgoing line multifunctional ammeter and the like; the sampled information includes: voltage, current, active, reactive, and electrical metrics, etc.; typically by means of a wireless public network.
(3) Master station side: the system is a master station system of the mining system and is used for collecting and managing uploaded data and analyzing and managing the uploaded data.
The invention installs a set of low-voltage area topology intelligent analysis software on the side of the main station system, reads the historical data stored by the acquisition system through the forward physical isolation equipment, analyzes the historical data by adopting an artificial intelligent algorithm, judges and infers the household-to-household topological relation of the related area, and performs graph-model display according to a unified area information access model defined based on IEC 61850 standard. The functional structure is shown in fig. 2.
According to the invention, a set of low-voltage area topology intelligent analysis software is installed, the collected data information is analyzed by adopting an artificial intelligent algorithm, the household-phase topology relation of the relevant area is judged and inferred, and the graph-model integrated display is carried out.
1. Data selection, data cleaning and data screening method based on electrical quantity trend self-adaptive adjustment
In order to improve the convergence stability and the discrimination precision of the algorithm, special processing needs to be carried out on the collected data of the existing active power, and the uniqueness processing is mainly embodied in the following aspects.
1. The data selection method comprises the following steps:
in the multidimensional data (current, voltage, reactive power and electric energy) of electric quantity, the invention takes active power as a study object of a data sample, on one hand, the change rate characteristic of the data is more suitable for the identification of a genetic algorithm than other data, and the precision is 5-10% better than other data.
2. The data cleaning method comprises the following steps:
three pretreatment modes are mainly adopted:
first, reject all 0 family table sets: for a sample with a sampling value of 0, this indicates that the user is not powered up for the period of the homologous data, and the user is rejected.
Second, reject the weak value family table set: the user is rejected for data values that are relatively small for which the sampled value exceeds 1/3 to 0 or the whole period.
Thirdly, eliminating an invalid time sample space set in the valid user table set: and if no indication exists at a certain moment when a certain user table exists in the plurality of moment sample spaces, all the corresponding moment data of all the corresponding effective user tables are removed.
3. The data screening method comprises the following steps:
the specific screening procedure is as follows (electric power P is the content of the study):
1) The electrical quantity data of the outgoing line side of the transformer in the transformer area is P a ,P b ,P c ,P Total-variation And the like, generally 1 point/15 minutes and the like;
2) The data sampled by the ammeter at the user side comprises P a ,P b ,P c ,P General-household Etc. since the user side meter is typically a single-phase meter and the specific acquisition phase is not known, there is typically only P General-household The data are represented as single-phase data;
3) P for taking outgoing line of distribution transformer Total-variation And accumulating the user table P corresponding to the same sampling time:
calculate error Δp=abs ((P) Total-variation -P General-household )/P Total-variation )×100；
Setting a parameter epsilon, and when delta P is less than or equal to epsilon, reserving the distribution transformer at the moment and the sampling data of the user ammeter, and adding the distribution transformer and the sampling data of the user ammeter into an effective data sample set; otherwise, the data is considered as bad data, and the time sampling data is discarded.
4) Based on the previous step epsilon, once the data start time period and epsilon values are determined, the amount of time point data after screening is then determined. Since the subsequent intelligent reasoning algorithm is related to epsilon for the value participating in calculation, the following steps are mainly adopted: the data amount of the effective time point is required, so that the adaptive selection mode needs to be adopted for v, namely the data amount K of the effective time point needs to satisfy: the value of K is related to epsilon and mainly comprises the following steps:
ε≤10％,K≥300；
ε≤15％,K≥400；
ε≤20％,K≥600；
ε≤25％,K≥1200；
ε≤35％,K≥2000。
with the control accuracy of the automatic adjustment epsilon, the conditions met by the data quantity K are different, and the intelligent recognition system can be started once the data preparation conditions are met. The processed data sample eliminates the mutation burr of the data, eliminates the bad data, effectively eliminates the influence of the sampling abnormal point on the algorithm convergence, and the processed sample data can be applied to intelligent reasoning calculation.
2. Random population construction method based on electrical node sample mapping
A combined set sample set of the distribution transformer of the transformer area, three phases and a user single phase P, wherein the user single phase sample set is put into the P Total (S) Returning to a storage list to store ValidDataSet:
decomposing to obtain a distribution side single-phase active one-dimensional matrix:
A/B/C phase total active one-dimensional matrix of the distribution transformer:
for the total active power of the distribution transformer side,
where j=0, 1,2,; m=j, j+3,..4 x validcount_num-3;
a phase A active one-dimensional matrix of the distribution transformer:
for the active power of the phase A of the distribution transformer,
where j=0, 1,2, validcount_num-1; m=j+1, j+4,..4 x validcount_num-2;
the configuration transformer B phase active one-dimensional matrix:
for the active power of the phase B of the distribution transformer,
where j=0, 1,2, validcount_num-1; m=j+2, j+5,..4 x validcount_num-1;
distribution transformer C-phase active one-dimensional matrix:
for the active power of the phase C of the distribution transformer side,
where j=0, 1,2, …, validcount_num-1; m=j+3, j+6, …,4 x validcount_num;
single-phase active matrix at the household table level:
wherein i=0, 1,2, ·m_num-1; j=0, 1,2,.,. Validcount_num-1; m=i; n=j, j+3,..4 x validcount_num-3;
randomly generating an initialized chromosome string structure data matrix, wherein each string structure data is called a chromosome, and M chromosomes form a population:
wherein M is the number of constructed chromosomes, N is the length of each chromosome
For a household-phase topology discrimination algorithm, the specific construction mode is as follows: m is the number of chromosome initial population constructed, the value m=100, N is the length of each chromosome, n=3×m_num,3 represents that the table has only A, B, C phases of discrimination possibility, the regions are all performed under a specific single region, and m_num is the total table number.
Uniqueness operation process:
1) Initial chromosome population
2) i=0; i++; each chromosome in the i.ltoreq.M-1///family is recycled;
3) j=0; j++; j < N-1)//// number of each gene per chromosome;
4) Taking a random integer of 1.ltoreq.t_num.ltoreq.N-1, when t_num is unique in pop_data [ i, j ], the ith gene is the only one (each chromosome is not repeated with a non-zero value), then: pop_data [ i, j ] = [ i, t_num ];
5) Returning to 3);
6) Returning to 4);
7) Random chromosome populations are produced.
Finally, each single-phase household table can be arranged in
The unique corresponding attribution position can be found in each chromosome, the specific position is randomly generated, and finally, after the iteration of the GA algorithm, the falling position corresponds to a certain phase position in the A/B/C three phases.
And the user table and a certain phase of the corresponding A/B/C phase are ingeniously and uniquely determined through a random number generation mode.
3. Construction method based on electric quantity trend deviation fitness function
The fitness function construction of the chromosome is the most important loop in the coding process, and the fitness indicates the goodness of an individual or a solution. Different problems, the adaptation function is defined differently.
For the user-phase, according to the coding mode, each m_num cis-position gene represents the user access condition of one A/B/C, so the whole chromosome sequence is divided into 3 partitions (if the relation is a user-variable relation, the number of the partitions is determined by the number of the platform regions), the power of the user of each partition is calculated according to the user number on the gene position in the partition, and the power error of each partition is obtained by calculating the difference between the power error and the acquired total power of each known partition, and the error is recorded as err. Since the present invention seeks a chromosome where err=0 and the fitness value is proportional to the quality of the chromosome, the fitness function fitness=max (err) -err is constructed such that the fitness function value is consistent with the evolution direction.
The fitness function is constructed as follows:
1. change rate of the three phases A/B/C of the distribution transformer:
a phase A active one-dimensional matrix of the distribution transformer:
for the active power of the phase A of the distribution transformer,
where j=0, 1,2, validcount_num-1; m=j+1, j+4,..4 x validcount_num-2;
rate of change of phase a of the ligand:
for the active change rate of the phase A of the distribution transformer side,
where i=0, 1,2, validcount_num-2;
the configuration transformer B phase active one-dimensional matrix:
for the active power of the phase B of the distribution transformer,
where j=0, 1,2, validcount_num-1; m=j+2, j+5,..4 x validcount_num-1;
rate of change of the ligand B phase:
for the active change rate of the phase B of the distribution transformer side,
where i=0, 1,2, validcount_num-2;
distribution transformer C-phase active one-dimensional matrix:
for the active power of the phase C of the distribution transformer side,
where j=0, 1,2, validcount_num-1; m=j+3, j+6,..4 x validcount_num;
rate of change of the ligand-transformer C phase:
active rate of change for phase C of the distribution transformer side, where i=0, 1,2,..;
2. rate of change delta E of user meter P 1-a1 、ΔE 1-b1 、ΔE 1-c1 :
Single-phase active matrix at the household table level:
wherein i=0, 1,2, ·m_num-1; j=0, 1,2,.,. Validcount_num-1; m=i; n=j, j+3,..4 x validcount_num-3;
the method comprises the following steps: pop_data_a [ i, j ] matrix:
because the length of Pop_data is 3 times of the length of A/B/C, the front is respectively corresponding to A phase and the middle is respectively corresponding to B phase
And phase, and finally corresponds to phase C. The corresponding phase a is now treated according to the effective gene code m_num, exceeding the gene code zero, leaving a reserve.
ifb ij ≤m_num,c ij ＝b ij ，else c ij ＝0，
i∈[1,Pop_num],j∈[1,validcount_num]
The method comprises the following steps: pop_data_b [ i, j ] matrix:
because the length of Pop_data is 3 times of the length of A/B/C, the front is respectively corresponding to A phase and the middle is respectively corresponding to B phase
And phase, and finally corresponds to phase C. The corresponding phase B is treated according to the effective gene code m_num, and the gene code is set to zero, so that the reserved is left.
if b ij ≤m_num,d ij ＝b ij ，else d ij ＝0，
i∈[1,Pop_num],j∈[validcount_num,2*validcount_num-1]
3. The method comprises the following steps: pop_data_c [ i, j ] matrix:
because the length of Pop_data is 3 times of the length of A/B/C, the front is respectively corresponding to A phase and the middle is respectively corresponding to B phase
And phase, and finally corresponds to phase C. The corresponding phase C is now treated according to the effective gene code (m_num), exceeding the gene code zero, leaving a reserve.
if b ij ≤m_num,e ij ＝b ij ，else e ij ＝0，
i∈[1,Pop_num],j∈[2*validcount_num,3*validcount_num-1]
4. The method comprises the following steps:the conversion matrix process is as follows:
wherein i=0, 1,2, ·m_num-1; j=0, 1,2,.,. Validcount_num-1; m=i; n=j, j+3,..4 x validcount_num-3;
initializing:
each corresponding row in pop_data_a is taken as the active user table with non-zero t_num and t_num as E φ []And then accumulating the sampling points of the columns corresponding to the index rows.
Each corresponding row in pop_data_b is taken as a valid user table with non-zero t_num and t_num as E φ []And then accumulating the sampling points of the columns corresponding to the index rows.
Each corresponding row in pop_data_c is taken as the active user table with non-zero t_num and t_num as E φ []And then accumulating the sampling points of the columns corresponding to the index rows.
5. The method comprises the following steps: e (E) 1-a1 The rate of change matrix (one dimension shorter than the previous conversion matrix) is processed as follows:
subtracting the previous term from the next term of the matrix and dividing the previous term by the next term
6. The method comprises the following steps: e (E) 1-b1 The rate of change matrix (one dimension shorter than the previous conversion matrix) is processed as follows:
subtracting the previous term from the next term of the matrix and dividing the previous term by the next term
7. The method comprises the following steps: e (E) 1-c1 The rate of change matrix (one dimension shorter than the previous conversion matrix) is processed as follows: subtracting the previous term from the next term of the matrix and dividing the previous term by the next term
8. Calculation of ΔE a ，ΔE b ，ΔE c ：
For delta E a The one-dimensional array is composed of Pop_num values, each value is an accumulated sum value of the distribution change rate and the corresponding user table change rate after taking absolute values, and a calculation formula is as follows:
a is a multiplication factor, generally A=1, m E [0, pop_num-1], and ABS represents taking absolute value
For delta E b The one-dimensional array is composed of Pop_num values, each value is an accumulated sum value of the distribution change rate and the corresponding user table change rate after taking absolute values, and a calculation formula is as follows:
a is a multiplication factor, typically A=1, m.epsilon.0, pop_num-1]
For delta E c The one-dimensional array is composed of Pop_num values, each value is an accumulated sum value of the distribution change rate and the corresponding user table change rate after taking absolute values, and a calculation formula is as follows:
a is a multiplication factor, typically A=1, m.epsilon.0, pop_num-1]
9. Calculation of ΔE 1 ：
ΔE 1 The degree of proximity between the distribution transformer and the corresponding user table in each chromosome is reflected, the smaller the value is, the closer the value is to the true value, and the calculation formula is as follows:
ΔE 1 ＝ΔE a +ΔE b +ΔE c
10. let ΔE 1 Conversion to Fitness, which measures proximity in each chromosome:
fit [ p ]]＝ΔE 1 [p],p＝1,2,…Pop_num-1
Calculating the Fit_max maximum value: fit_max=max (Fit [ p ])
Fitness[P]＝Fit_max-Fit[P],P＝1,2,…,Pop_num-1
Fitness and DeltaE 1 The conversion relation is that the larger the corresponding Fitness value in each subsequent chromosome is, the correspondingThe closer the chromosome set is to the actual true value of the ligand transformation.
4. Selection function construction method based on elite optimization mechanism
And (3) screening chromosomes in the population by a elite strategy according to the calculation result of Fitness Fitness. The elite strategy is to construct probability intervals based on fitness, and if the fitness is higher, the probability is higher and the probability is selected for being a member of the next generation population for multiple times, so that chromosomes with smaller fitness are eliminated, and the specific steps are as follows:
4.1 Normalized processing procedure:
fitness normalization process:
P_fitness[j]＝Fitness[j]/Sumfit_ness，j＝0，1，…Pop_num-1
4.2 Chromosome preservation of top five ranks
The 5 bigger chromosomes corresponding to the P_fitness [ j ] from big to small are put in the following order: in the fit_max5[ w ] array, w=0, 1,2,3,4, j=0, 1,2, …, pop_num-1;
4.3 An elite policy screening construction method:
sequencing Pop_data from small to large according to P_fitness [ j ], j=0, 1,2, …, and correspondingly obtaining D_Popdata [ m ] [ n ], m=0, 1,2 … and Pop_num-1 after sequencing; n=0, 1,2, …, poplenge-1;
the ms, ms epsilon [0,1] value is selected randomly, so that the range of random selection is reduced in order to meet the success and effectiveness of each selection, and the conditions are met: ms is less than or equal to max (P_fitness [ j ]), j=0, 1,2, …, pop_num-1, and pop_num is selected in total.
If the j+1 chromosome corresponding to D_Popdata is selected when the probability ms is between j and j+1 chromosomes, namely:
P_fitness[j]≤ms≤P_fitness[j+1]，j＝0，1，2，…，Pop_num-1；
and saving the j+1 chromosome corresponding to the selected D_Popdata to the NewPopdata, and after Pop_num is selected for a plurality of times ms, generating a new array NewPopdata, wherein the new array NewPopdata contains chromosomes with more preferable Pop_num selection results.
Searching whether Fit_max5[ w ] is contained in the NewPopdata, wherein w=0, 1,2,3 and 4, if not, calculating P_fitness [ j ] of the NewPopdata, j=0, 1,2, … and Pop_num-1, and replacing the minimal corresponding chromosome sample of the P_fitness by the chromosome sample which does not appear in Fit_max5[ w ] and w=0, 1,2,3 and 4, thereby effectively preserving the previous 5 better chromosomes and eliminating the chromosome with smaller P_fitness value.
5. Construction method of chromosome set crossover operator based on electrical quantity data
Generating a new array NewPopdata containing Pop_num chromosomes, and combining two adjacent chromosomes as a group of chromosomes to be crossed in sequence, wherein the two adjacent chromosomes can form a totalA downward rounding of the chromosome set.
Next, taking j, i=0, 1,2, …, the pop_num-2 chromosome and i+1, i=0, 1,2, …, the pop_num-2 chromosome are subjected to pairwise crossover operation:
(1) Before the two groups of chromosomes to be crossed are crossed, an operation random number c_rand E (0, 1) is selected randomly, if c_rand is more than or equal to pc (pc is a cross probability factor, and the value pc E (0.8,1) is generally taken so that most of the chromosomes can meet the condition of the cross operation), the two groups of chromosomes are not crossed, and the chromosome groups keep the original genes unchanged.
(2) If c_rand < pc is satisfied, the chromosome set performs a crossover operation, specifically as follows:
taking two random integers:
t 1 ，t 2 and t 1 <t 2 ，t 1 ∈(0，Poplenge-1)，t 2 ∈(0，Poplenge-1)
Poplenge＝3*m_num
Will [ t ] of the ith strip 1 ，t 2 ]Genes of the segment and [ t ] of item i+1 1 ，t 2 ]Genes were crossed according to the following rules:
let the parent i and i+1 chromosomes be, where the color is the corresponding t 1 ，t 2 ]The portions of the segments to be intersected are shown in figure 8,
then the sub-generation chromosomes generated by crossing the i chromosome to the i+1 chromosome are as shown in figure 9:
1) Mapping i chromosome [ t ] 1 ，t 2 ]The position gene completely inherits the sub-generation of the corresponding position;
2) Find out i+1 chromosome and i chromosome t 1 ，t 2 ]The gene sites with the same segments are positioned in the frame selection part;
3) I+1 chromosome and i chromosome [ t ] 1 ，t 2 ]The gene sites with different segments are sequentially inserted into the vacant positions.
Similarly, another sub-generation dyeing generated by crossing i+1 chromosomes to i chromosomes can be obtained, and the result is shown in six of FIG. 10, and the construction method of chromosome band boundary mutation operator based on electrical quantity data
The variation of the research is to randomly change a certain gene position of a chromosome according to a variation probability factor pm (pm is less than or equal to 10%), randomly select one user, exchange a gene corresponding to the user to another randomly selected position, namely randomly change the access position of the certain user, wherein the position of the gene position change is provided with a precondition: because the position of the position is within the region (such as the corresponding three sections of A/B/C), the variation effect is not achieved, and therefore, the basic precondition design requirement is still met when the corresponding position is selected.
The pop_num chromosomes were subjected to the same rule as follows.
The mutation procedure was performed on a single chromosome as follows:
5) Judging whether mutation operation is needed or not: when the random number count_rand is more than or equal to pm, the count_rand E (0, 1) is not operated at this time; when count_rand < pm, the next operation is continued.
6) Selecting a random integer t 1 The_rand is less than or equal to Poplenge-1, androunding downwards, when t 1 1=0 represents the gene to which phase a belongs; when t 1 1=1 represents the gene to which phase B belongs; when t 1 1=2 represents the gene to which phase C belongs. Let t 1 The value of the_rand gene is put in the variable t_sample_1;
the same applies to the above steps, taking the second random number t 2 _rand≤Poplenge-1，t 2 _rand≠t 1 And (d) rand, and rounding downwards, when t 2 1=0 represents the gene to which phase a belongs; when t 2 1=1 represents
7) A gene of phase B; when t 2 1=2 represents the gene to which phase C belongs. Let t 2 The value of the_rand gene is put in the variable t_sample_2;
8) The discriminating process comprises the following steps:
4.1 If t_pattern_1+t_pattern_2=0, the gene bit to be exchanged belongs to an invalid bit, the mutation operation is not performed, and the process returns to 3) to continue random selection;
4.2 If t_temp_1+t_temp_2+.0 and t 1 _1≠t 2 And 1, carrying out exchange positions on the gene bit values to be exchanged in different phase regions, storing the gene bit values into corresponding arrays, and ending the chromosome mutation operation.
7. Construction method of reselection mechanism for preventing premature ripening
Considering the possibility of the judgment accuracy degradation caused by the early maturing condition, namely, entering a local convergence point for a special condition to terminate the circulation condition, the accumulated value exceeds the pre-prepared reselection rate pw within a short period of time (lower than 2500 times) of executing the circulation body, and the calculation of the genetic algorithm is carried out again through a reselection mechanism.
The specific construction method comprises the following steps:
1. the maximum chromosome number of the same-number gene composition in Pop_data_A is found, and the main operation is two steps:
1.1 Ordering the gene values in Pop_data_A from big to small;
1.2 Counting the maximum number of the same gene value, and marking as num_a;
2. the maximum chromosome number of the same-number gene composition in Pop_data_B is found, and the main operation is two steps:
1.3 Ordering pop_data_b from big to small;
1.4 Counting the maximum number of the same gene value, and marking as num_b;
3. the maximum chromosome number of the same-number gene composition in Pop_data_C is found, and the main operation is two steps:
1.5 Ordering pop_data_c from big to small;
1.6 The maximum number of the same gene values was counted and noted as num_c.
The judgment basis is as follows:
setting the total cycle times as T (generally T is more than or equal to 5000), and under the condition that the cycle body execution is not more than 2500 times, satisfying (num_a+num_b+num_c)/3 x Pop_num is more than or equal to pw, wherein pw E [0.8,1] takes a value, restarting the reselection mechanism, namely restarting the genetic algorithm from the random function reselection.
8. A non-signal injection type family phase topological relation flow based on a genetic algorithm:
and (3) designing a general control flow:
1) Preset parameters t=10000, pc=0.8, pm=0.1, pw=0.8, pop_num=100
2) According to the electric quantity data information, obtaining relevant data content after data cleaning, including:
E 1-a [],E 1-b [],E 1-c [],m_num，validcount_num
3) Initializing Pop_data, and randomly generating Pop_data [ ] according to m_num;
4) Enter t=10000 loop process
5) Calling the fitness function module, returning to the Fit _ less,ΔE 1
6) Entering into a selection function call of a genetic algorithm to obtain
NewPopdata,Fit_max5，MBest_pop[]＝max(Fit_max5)
7) Entering a cross function call of a genetic algorithm to obtain updated newPopdata;
8) Entering a cross function call of a genetic algorithm to obtain updated newPopdata;
9) When t=500, calling a reset function call entering a genetic algorithm, if the premature condition is satisfied, entering a step 3), otherwise, continuing to execute the operation downwards;
10 Return 4)
11 Obtaining the final newpop data, and obtaining best_pop through the following determined mapping relation, wherein the specific judgment is as follows:
12 The discrimination result of each row of each household table corresponding to best_pop [ m_num ] [6] comprises a 0 th bit representative household table number; bit 1=1/2/3, respectively representing a phase a, a phase B, and a phase C; bit 2 represents the degree of reliability of discrimination=0/1, respectively representing unreliable/trusted; bits 3,4 and 5 represent the numbers of phases A, B and C belonging to the group.
Specific running test example:
1. station area: 173 table area
2. Data type: the actual 2019 power data of 4 and 5 months of power, and partial screenshot of the data source are shown as follows
3. Number of users: 80
4. Total cycle: 10000 times
5. The number of cycles to obtain the minimum ΔE1 is 2026; Δe1 minimum value 17552.573203380536
6. The 0 th bit represents the user table number (has corresponding unique relation with the actual user table number); bit 1=1/2/3, respectively representing a phase a, a phase B, and a phase C; bit 2 represents the degree of reliability of discrimination=0/1, respectively representing unreliable/trusted; bits 3,4 and 5 represent the numbers of phases A, B and C belonging to the group. The running test results are as follows (phase A, phase B and phase C are 23, 25 and 22 respectively, wherein the 11 family meter is not judged due to zero or no electricity consumption, and the subsequent electricity consumption is waited to be accurately judged):
[4,3,1,0,0,100]
[5,3,1,0,0,100]
[7,2,1,0,100,0]
[8,2,1,0,100,0]
[9,3,1,0,0,100]
[11,3,1,0,0,100]
[13,2,1,1,99,0]
[14,3,1,1,0,99]
[15,1,1,100,0,0]
[16,1,1,100,0,0]
[17,1,1,99,0,1][19,2,1,1,99,0][20,1,1,92,8,0][21,2,1,0,99,1][22,1,1,100,0,0][23,2,1,0,100,0][24,2,1,0,100,0][26,3,1,0,0,100][27,2,1,0,100,0][28,1,1,100,0,0][29,1,1,99,1,0][30,2,1,0,100,0][31,3,1,0,0,100][32,1,1,100,0,0][33,3,1,0,34,66][34,1,1,99,1,0][35,3,1,0,0,100][36,2,1,0,100,0][37,2,1,0,99,1][38,1,1,100,0,0][39,1,1,100,0,0][41,2,1,0,100,0][42,2,1,0,100,0][43,3,1,0,0,100][44,3,1,1,0,99][45,3,1,0,0,100][46,2,1,0,100,0][47,2,1,1,99,0][48,2,1,0,100,0][49,1,1,100,0,0][50,1,1,100,0,0]
[51,1,1,100,0,0]
[52,1,1,100,0,0]
[53,3,1,0,0,100]
[54,2,1,0,100,0]
[55,2,1,0,100,0]
[56,3,1,0,0,100]
[57,2,1,0,100,0]
[58,1,1,100,0,0]
[59,1,1,100,0,0]
[61,2,1,0,100,0]
[62,3,1,0,0,100]
[63,2,1,1,99,0]
[64,3,1,0,1,99]
[65,3,1,0,1,99]
[66,3,1,0,1,99]
[67,3,1,0,0,100]
[68,1,1,100,0,0]
[69,1,1,100,0,0]
[70,1,1,99,0,1]
[72,2,1,0,100,0]
[73,3,1,0,0,100]
[74,2,1,0,100,0]
[75,2,1,0,100,0]
[76,1,1,99,1,0]
[77,3,1,0,0,100]
[78,1,1,99,1,0]
[79,2,1,0,100,0]
[80,3,1,0,0,100]
[81,1,1,100,0,0]。
the artificial intelligence algorithm based on the electrical characteristics has certain universality, can be popularized to the fields of power transmission, transformation and distribution, lays a foundation for the analysis of big data of a power grid and the deep application of artificial intelligence, can be widely applied to the field of the ubiquitous Internet of things of electric power, and brings considerable social and economic benefits. As follows:
1. the problem of low-voltage transformer area topology identification can be solved by utilizing the existing data of the mining system, namely, other equipment is not required to be installed, and a person is not required to be dispatched to perform actual measurement and investigation on site, so that the large expense is saved. Meanwhile, the efficiency is greatly improved, and the operation management level is improved.
2. The operation condition of the station area can be effectively monitored, and more reliable basis and means are provided for monitoring, analyzing and treating the low-voltage three-phase imbalance.
3. And acquiring data according to the user table, and providing auxiliary reference for the deep analysis of the load characteristics of the user and the tracking of the dynamic operation characteristics of the load.
As described above, although the present invention has been shown and described with reference to certain preferred embodiments, it is not to be construed as limiting the invention itself. Various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (6)
1. A non-signal injection type household phase topological relation identification method based on a genetic algorithm is characterized by comprising the following steps:
(1) According to electric quantity data information, obtaining effective data after data cleaning, wherein the electric quantity data is any one of current, voltage, reactive power, electric energy and active power, the data cleaning method comprises the steps of eliminating all 0 user table sets, eliminating weak user table sets, eliminating invalid time sample space sets in the effective user table sets, returning the effective data after data cleaning to a storage list ValidDataSet, wherein each qualified sample comprises a combined set sample set of a station distribution transformer and three-phase and user single-phase electric quantity, m_num in the storage list ValidDataSet represents the total user table number, validcount_num represents the qualified data sample number, and obtaining a distribution side total electric quantity matrix E according to the storage list ValidDataSet data Total of distribution transformer [j]Distribution side A, B, C single-phase electric quantity matrix E 1-a [j]、E 1-b [j]、E 1-c [j]Single-phase electric quantity matrix for sum household meter level
(2) According to the effective data, randomly generating an initialized chromosome population data matrix pop_data, wherein M is the number of chromosome initial populations, N=3×m_num,3 represents the discrimination result of the household table with only A, B, C phases, and randomly generating chromosome populations according to m_num and pop_num
The effective data is obtained by screening the cleaned data, and the electric quantity data of the outgoing line side of the transformer in the transformer area comprises P a ,P b ,P c ,P Total-variation The data obtained from the user side ammeter sample has P a ,P b ,P c ,P General-household Obtaining P of outgoing line of distribution transformer Total-variation And accumulating and calculating an error delta P corresponding to the user table P at the same sampling moment, setting a parameter epsilon, and when delta P is less than or equal to epsilon, reserving the distribution transformer at the moment and the sampling data of the user electric meter, adding the distribution transformer and the sampling data of the user electric meter into an effective data sample set, otherwise, giving up epsilon and the quantity K of the effective moments satisfy the following conditions:
ε≤10％,K≥300；
ε≤15％,K≥400；
ε≤20％,K≥600；
ε≤25％,K≥1200；
ε≤35％,K≥2000；
(3) Entering a circulation process:
(4) Calling an Fitness function, returning the proximity delta E between the distribution transformer side and the corresponding user table in each chromosome, and converting the delta E into Fitness for measuring the proximity degree in each chromosome;
the fitness function construction process is as follows:
calculating the change rate delta E of the A/B/C three phases of the distribution transformer 1-a [i]、ΔE 1-b [i]、ΔE 1-c [i]Calculating the change rate delta E of the user meter P 1-a1 、ΔE 1-b1 、ΔE 1-c1 ；
Obtaining Pop_data_A [ i, j ], pop_data_B [ i, j ], pop_data_C [ i, j ] matrix;
obtaining E 1-a1 Rate of change matrix, E 1-b1 Rate of change matrix, E 1-c1 A rate of change matrix;
calculating the cumulative sum delta E of the distribution change rate and the absolute value of the corresponding household table change rate a ，ΔE b ，ΔE c ；
Proximity delta E between distribution transformer and corresponding user table in each chromosome 1 ＝ΔE a +ΔE b +ΔE c ；
Calculating Fit [ p ]]＝ΔE 1 [p],p＝1,2,…Pop_num-1，Fit_max＝max(Fit[p])，
Fitness[P]＝Fit_max-Fit[P],P＝1,2,…,Pop_num-1；
Dividing the whole chromosome into partitions sequentially, calculating power of users in each partition according to the user numbers on the gene positions in the partitions, calculating difference between the calculated power and the total power of each partition to obtain power error of each partition, marking the power error as err, and constructing a fitness function fitness=max (err) -err, so that the value of the fitness function is consistent with the evolution direction;
(5) Calling a selection function, a cross function and a variation function of a genetic algorithm to obtain updated newPopdata, wherein the updated newPopdata contains chromosomes with Pop_num selection results;
(6) Setting an early ripening condition, entering a reset function call of a genetic algorithm, if the early ripening condition is met, entering a step (2), otherwise, continuing a circulation process;
(7) And obtaining a final NewPopdata, and obtaining a discrimination result best_Pop of each user table corresponding to each row through a mapping relation.
2. The genetic algorithm-based non-signal injection type user phase topological relation identification method is characterized by comprising the following steps of: the method for constructing the selection function comprises the following steps of
Fitness normalization:
P_fitness[j]＝Fitness[j]/Sumfit_ness，j＝0,1,…,Pop_num-1；
the 5 chromosomes corresponding to the P_fitness [ j ] from big to small are placed in the first five chromosomes to be ranked to store a Fit_max5[ w ] array;
randomly selecting a probability value ms, ms epsilon [0,1], and meeting the conditions:
ms≤max(P_fitness[j]),j＝0,1,2,…,Pop_num-1。
3. the genetic algorithm-based non-signal injection type user phase topological relation identification method is characterized by comprising the following steps of: the cross function is: selecting an operation random number c_rand epsilon (0, 1) and crossover probability factors pc and pc epsilon (0.8,1), and if c_rand is more than or equal to pc, not performing crossover operation on the two groups of chromosomes, wherein the chromosome groups keep the original genes unchanged; if c_rand < pc is satisfied, the chromosome set performs a crossover operation.
4. A method for identifying a non-signal injection type user phase topological relation based on a genetic algorithm according to claim 3, wherein the method comprises the following steps: the variation function is:
1) Judging whether mutation operation is needed or not: when the random number count_rand is more than or equal to pm, the count_rand epsilon (0, 1), pm is a variation probability factor, the variation is not operated at this time, and when the count_rand is less than pm, the next operation is continued;
2) Selecting a random integer t 1 The_rand is less than or equal to Poplenge-1, androunding down, m_num is the total household table number, when t 1 1=0 represents the gene of phase a, when t 1 1=1 represents the gene of phase B, when t 1 1=2 represents the gene of the C phase, t is taken as 1 The value of the_rand gene is put in the variable t_sample_1;
3) The same steps as above, the second random number t is taken 2 _rand≤Poplenge-1，t 2 _rand≠t 1 And (d) rand, and rounding downwards, when t 2 1=0 represents the gene of phase a, when t 2 1=1 represents the gene of phase BWhen t 2 1=2 represents the gene of the C phase, t is taken as 2 The value of the_rand gene is put in the variable t_sample_2;
4) The discriminating process comprises the following steps:
4.1 If t_pattern_1+t_pattern_2=0, the gene bit to be exchanged belongs to an invalid bit, the mutation operation is not performed, and the process returns to 3) to continue random selection;
4.2 If t_temp_1+t_temp_2+.0 and t 1 _1≠t 2 And 1, carrying out exchange positions on the gene bit values to be exchanged in different phase regions, storing the gene bit values into corresponding arrays, and ending the chromosome mutation operation.
5. The genetic algorithm-based non-signal injection type user phase topological relation identification method is characterized by comprising the following steps of: when the accumulated value exceeds the preset reselection rate when the circulation body is executed for not more than half times, the calculation of the genetic algorithm is carried out again through a reselection mechanism: sequencing the Pop_data gene values from large to small; the maximum number of the same gene values was counted and recorded as num.
6. The genetic algorithm-based non-signal injection type user phase topological relation identification method is characterized by comprising the following steps of: each household table corresponds to the discrimination result of each row of the best_pop, and the discrimination result comprises a 0 th representative household table number; bit 1=1/2/3, respectively representing a phase a, a phase B, and a phase C; bit 2 represents the degree of reliability of discrimination=0/1, respectively representing unreliable/trusted; bits 3,4 and 5 represent the numbers of phases A, B and C belonging to the group.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011126494.XA CN112036553B (en) | 2020-10-20 | 2020-10-20 | Genetic algorithm-based non-signal injection type household phase topological relation identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011126494.XA CN112036553B (en) | 2020-10-20 | 2020-10-20 | Genetic algorithm-based non-signal injection type household phase topological relation identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112036553A CN112036553A (en) | 2020-12-04 |
| CN112036553B true CN112036553B (en) | 2024-04-09 |
Family
ID=73573358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011126494.XA Active CN112036553B (en) | 2020-10-20 | 2020-10-20 | Genetic algorithm-based non-signal injection type household phase topological relation identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112036553B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112132283B (en) * | 2020-10-20 | 2024-04-05 | 江苏其厚智能电气设备有限公司 | Genetic algorithm-based non-signal injection type household transformer topological relation identification method |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010278369A (en) * | 2009-06-01 | 2010-12-09 | Tokyo Electric Power Co Inc:The | System and program for estimating apparatus constant of transformer |
| CN106127304A (en) * | 2016-06-30 | 2016-11-16 | 国网山东省电力公司无棣县供电公司 | One is applicable to power distribution network Network Topology Design method |
| CN108776869A (en) * | 2018-06-07 | 2018-11-09 | 国网河南省电力公司电力科学研究院 | Taiwan area Three-Phase Transformer imbalance administering method based on intelligent electric meter big data analysis |
| CN109995059A (en) * | 2019-05-21 | 2019-07-09 | 华北水利水电大学 | Low-voltage distribution system three-phase load based on genetic algorithm optimizes balanced arrangement method |
| CN110147884A (en) * | 2019-04-02 | 2019-08-20 | 佛山市顺德区中山大学研究院 | A kind of efficient circuits topological structure evolvement method and its system |
| CN110837965A (en) * | 2019-11-07 | 2020-02-25 | 国网山东省电力公司威海供电公司 | Method for rapidly distinguishing station area house change relations |
| CN110867906A (en) * | 2019-12-02 | 2020-03-06 | 南京工程学院 | Power grid topological relation identification method based on power balance index |
| CN111132178A (en) * | 2019-12-25 | 2020-05-08 | 广州供电局有限公司计量中心 | Electric power wireless sensor network design method based on edge calculation |
| CN111478349A (en) * | 2020-04-16 | 2020-07-31 | 国网冀北电力有限公司电力科学研究院 | Power three-phase balance strategy method based on power utilization equipment intelligent network regulation and control Internet of things |
| CN111680374A (en) * | 2020-05-21 | 2020-09-18 | 合肥泽众城市智能科技有限公司 | Method for checking and repairing topological relation of pipe network |
| CN111784067A (en) * | 2020-07-09 | 2020-10-16 | 北京理工大学 | Two-stage power distribution network planning method based on genetic algorithm |
| CN112132283A (en) * | 2020-10-20 | 2020-12-25 | 江苏其厚智能电气设备有限公司 | Non-signal injection type user variable topological relation identification method based on genetic algorithm |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9728971B2 (en) * | 2012-12-10 | 2017-08-08 | The Research Foundation For The State University Of New York | Apparatus and method for optimal phase balancing using dynamic programming with spatial consideration |
| US20140365148A1 (en) * | 2013-04-01 | 2014-12-11 | University Of Connecticut | Methods And Systems For Test Power Analysis |
-
2020
- 2020-10-20 CN CN202011126494.XA patent/CN112036553B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010278369A (en) * | 2009-06-01 | 2010-12-09 | Tokyo Electric Power Co Inc:The | System and program for estimating apparatus constant of transformer |
| CN106127304A (en) * | 2016-06-30 | 2016-11-16 | 国网山东省电力公司无棣县供电公司 | One is applicable to power distribution network Network Topology Design method |
| CN108776869A (en) * | 2018-06-07 | 2018-11-09 | 国网河南省电力公司电力科学研究院 | Taiwan area Three-Phase Transformer imbalance administering method based on intelligent electric meter big data analysis |
| CN110147884A (en) * | 2019-04-02 | 2019-08-20 | 佛山市顺德区中山大学研究院 | A kind of efficient circuits topological structure evolvement method and its system |
| CN109995059A (en) * | 2019-05-21 | 2019-07-09 | 华北水利水电大学 | Low-voltage distribution system three-phase load based on genetic algorithm optimizes balanced arrangement method |
| CN110837965A (en) * | 2019-11-07 | 2020-02-25 | 国网山东省电力公司威海供电公司 | Method for rapidly distinguishing station area house change relations |
| CN110867906A (en) * | 2019-12-02 | 2020-03-06 | 南京工程学院 | Power grid topological relation identification method based on power balance index |
| CN111132178A (en) * | 2019-12-25 | 2020-05-08 | 广州供电局有限公司计量中心 | Electric power wireless sensor network design method based on edge calculation |
| CN111478349A (en) * | 2020-04-16 | 2020-07-31 | 国网冀北电力有限公司电力科学研究院 | Power three-phase balance strategy method based on power utilization equipment intelligent network regulation and control Internet of things |
| CN111680374A (en) * | 2020-05-21 | 2020-09-18 | 合肥泽众城市智能科技有限公司 | Method for checking and repairing topological relation of pipe network |
| CN111784067A (en) * | 2020-07-09 | 2020-10-16 | 北京理工大学 | Two-stage power distribution network planning method based on genetic algorithm |
| CN112132283A (en) * | 2020-10-20 | 2020-12-25 | 江苏其厚智能电气设备有限公司 | Non-signal injection type user variable topological relation identification method based on genetic algorithm |
Non-Patent Citations (5)
| Title |
|---|
| A genetic algorithm approach for the identification of microgrids partitioning into distribution networks;Korjani Saman 等;《IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society》;21-25 * |
| HVDC Grid Fault Current Limiting Method Through Topology Optimization Based on Genetic Algorithm;Tao Yan 等;《IEEE Journal of Emerging and Selected Topics in Power Electronics》;第09卷(第06期);7045-7055 * |
| 主动配电网网络优化技术研究;杨铭;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》(第06期);C042-697 * |
| 结合遗传算法的人工神经网络在电力变压器故障诊断中的应用;徐文 等;《中国电机工程学报》(第02期);109-112 * |
| 配电网故障区段定位及拓扑重构方法;孙孔明;《中国博士学位论文全文数据库 (工程科技Ⅱ辑)》(第11期);C042-23 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112036553A (en) | 2020-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110082699B (en) | Low-voltage transformer area intelligent electric energy meter operation error calculation method and system | |
| CN112103988B (en) | Method for establishing cluster division double-layer model combined with network reconstruction | |
| CN107578288B (en) | Non-invasive load decomposition method considering user power consumption mode difference | |
| CN112132283B (en) | Genetic algorithm-based non-signal injection type household transformer topological relation identification method | |
| CN108599172B (en) | Transmission and distribution network global load flow calculation method based on artificial neural network | |
| CN113126019B (en) | Remote estimation method, system, terminal and storage medium for error of intelligent ammeter | |
| CN104123684B (en) | Power distribution network economy reconstructing method considering network loss and reliability | |
| CN111612273A (en) | Regional-level comprehensive energy system partition design method and system | |
| CN108764588A (en) | A kind of temperature influence power prediction method based on deep learning | |
| CN112036553B (en) | Genetic algorithm-based non-signal injection type household phase topological relation identification method | |
| CN117578434A (en) | Power distribution network flexibility evaluation method and device considering flexible resource adjustability | |
| CN113595071A (en) | Transformer area user identification and voltage influence evaluation method | |
| CN113949079B (en) | Power distribution station user three-phase unbalance prediction optimization method based on deep learning | |
| CN118210603A (en) | Cloud resource scheduling method based on enhanced growth optimizer | |
| CN117974395A (en) | Environment-friendly community management method and system based on carbon emission monitoring | |
| CN115577770A (en) | Continuous electrical appliance load prediction method based on load decomposition and deep learning | |
| CN116502894A (en) | Photovoltaic transformer area power failure risk assessment method and device, electronic equipment and storage medium | |
| CN111985718B (en) | Data-driven distribution transformer model selection recommendation method | |
| CN116093984A (en) | Parallel three-phase imbalance optimization method based on space and user load time sequence characteristics | |
| CN115759418A (en) | Adjustable capacity prediction method and system for electric power retail customer | |
| CN116128117A (en) | Distribution line loss prediction method and device based on digital twinning | |
| CN115564106A (en) | Carbon emission measuring and calculating method based on electric power data | |
| CN113705885A (en) | Power distribution network voltage prediction method and system integrating VMD, XGboost and optimized TCN | |
| CN111948446A (en) | Self-adaptive load identification method and intelligent electric energy meter | |
| CN114243695B (en) | Power load prediction method based on bidirectional long-short-term memory neural network |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |