AU2021106109A4 - Evaluation index screening strategy for lean management of power system line loss under big data environment - Google Patents
Evaluation index screening strategy for lean management of power system line loss under big data environment Download PDFInfo
- Publication number
- AU2021106109A4 AU2021106109A4 AU2021106109A AU2021106109A AU2021106109A4 AU 2021106109 A4 AU2021106109 A4 AU 2021106109A4 AU 2021106109 A AU2021106109 A AU 2021106109A AU 2021106109 A AU2021106109 A AU 2021106109A AU 2021106109 A4 AU2021106109 A4 AU 2021106109A4
- Authority
- AU
- Australia
- Prior art keywords
- rate
- evaluation index
- line loss
- index
- power system
- 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.)
- Ceased
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 88
- 238000012216 screening Methods 0.000 title claims abstract description 30
- 238000007637 random forest analysis Methods 0.000 claims abstract description 26
- 238000007726 management method Methods 0.000 claims description 45
- 238000004088 simulation Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 7
- 230000005611 electricity Effects 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 6
- 238000009795 derivation Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 238000012797 qualification Methods 0.000 claims description 3
- 230000008439 repair process Effects 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 101100406575 Arabidopsis thaliana ORTH4 gene Proteins 0.000 claims 1
- 101100263687 Xenopus laevis vim4 gene Proteins 0.000 claims 1
- 238000011084 recovery Methods 0.000 claims 1
- 238000000034 method Methods 0.000 abstract description 15
- 238000011161 development Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 238000013278 delphi method Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000003763 carbonization Methods 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- 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
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Algebra (AREA)
- Water Supply & Treatment (AREA)
- Medical Informatics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Public Health (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an evaluation index screening strategy for lean management of
power system line loss under big data environment, which comprises the following steps: step
1, combining multiple methods to generate a power system line loss lean management
evaluation index library; step 2, use the stochastic forest model to screen the evaluation index
of power system line loss lean management. According to the invention, an objective and
reasonable evaluation index system is established by screening evaluation indexes through
random forest learning.
113
Figures
C Evluaionndexoeficint of line loss lean management
4
Establishment of evaluationindex system.
ICollect line loss data andinformation, and establish theI
Ievaluationindex library of line loss lean managementI
Combining multi-methodI
Se ~ ee&Dnfev1uatum idex SFnff dtnsetI
I Screening of evaluaftionindex based on random forestI
Figure1I
Description
Figures
C Evluaionndexoeficint of line loss lean management
4 Establishment of evaluationindex system.
ICollect line loss data andinformation, and establish theI Ievaluationindex libraryof line loss lean managementI Combining multi-methodI
Se ~ ee&Dnfev1uatum idex SFnff dtnsetI
I Screening of evaluaftionindex based on random forestI
Figure1I
Evaluation index screening strategy for lean management of power system
line loss under big data environment
The embodiment of the invention relates to the field of evaluation management
in power systems, in particular to an evaluation index screening strategy for lean
management of power system line loss under big data environment.
Energy shortage makes low carbonization of economic development one of the
urgent global problems. Power grid enterprises, as a bridge connecting power
production departments and users, play an important role in achieving the national
goal of carbon neutrality and achieving green and low-carbon development in 2060.
Line loss management is the core task of power grid enterprises. Line loss occurs in
every link of transmission, substation, distribution and sales, runs through the whole
power grid value chain, and comprehensively reflects the management level of power
grid planning and design, dispatching operation, production, operation and marketing.
However, in actual business, there are the following outstanding problems in line
loss management, such as: 1) The line loss management system is not detailed enough,
and the corresponding directions of loss reduction are not detailed enough and
systematic enough. Line loss management is a systematic work involving power
equipment, power supply users, distribution networks, etc., but the current power
supply enterprises do not have a complete system to effectively connect them for
management; 2) The basic files of line loss management are chaotic, and the
professional quality of the line loss management personnel is insufficient, which leads
to the ineffective implementation of some work in the line loss management network.
The line loss rate is one of the key strategic indexs to comprehensively measure
the effect and level of the implementation of the company's medium-and long-term
development strategy. Taking the lean management of line loss as the key direction of
the enterprise's transformation to lean management helps power supply enterprises
achieve the goal of "optimal technical line loss and minimum management line loss",
and promotes the management level of three core businesses, namely customer
service, power grid operation and power grid development. It is the fundamental
embodiment of the company's implementation of lean management transformation
and the development direction proposed by the medium-and long-term development
strategy of power grid companies.
At present, many institutions and scholars at home and abroad have begun to
study the evaluation index system of line loss lean management and formulate
relevant certification standards. The existing evaluation index screening methods
mainly include subjective screening method and objective screening method.
Subjective screening method is a method of screening evaluation indexs through
the collective wisdom of experts, which is simple and does not need complicated
calculation. However, subjective screening method has strong subjective factors and
lacks correlation analysis among evaluation indexes, which has strong interference to
the evaluation results and reduces the credibility of the results.
The objective screening method is to scientifically and objectively screen out
important evaluation indexs and eliminate redundant or irrelevant evaluation indexs
through specific data analysis. However, the calculation results of the objective
screening method cannot tell the evaluator how many evaluation indexes should be
kept, and the evaluator needs to decide the final evaluation index system according to
the calculation results.
The above two kinds of screening methods have advantages and disadvantages,
but they can complement each other in some working conditions. How to synthesize
the excellent characteristics of the two kinds of methods to get a comprehensive
evaluation index screening strategy is a direction worth studying.
The purpose of the invention is to provide an evaluation index screening strategy
for power system line loss lean management under the big data environment.
In order to achieve the purpose of the invention, the invention adopts the
following technical scheme: a screening strategy for the evaluation index of power
system line loss lean management under the big data environment comprises the
following steps,
Step 1, a multi-method combination is adopted to generate a power system line
loss lean management evaluation index library, which comprises the following steps:
Step 1.1, use frequency analysis method to make frequency statistics on line loss
evaluation standards and evaluation indexes of related specifications under big data
environment, and select evaluation indexes with higher frequency as the most original
evaluation index library of power system line loss lean management;
Step 1.2, preliminarily screen the evaluation index library of power system line
loss lean management by Delphi method;
Step 1.3, collect the sample data of the preliminarily screened evaluation index
library of power system line loss lean management, including: 1) quantitative
evaluation index data preprocessing: defining the dimension and magnitude of each
quantitative evaluation index, taking the physical quantity value of evaluation index
as the scoring value; 2) qualitative evaluation index data preprocessing: the scoring value of qualitative evaluation index is determined according to the scoring rules of relevant specifications or expert groups;
Step 1.4, determine an evaluation index library of lean management of power
system line los by using LMS algorithm; the steps are as follows:
A) for n selected power supply enterprises (Si, S2, ... , Sn), the line loss of each
power supply enterprise can be measured by the observed m evaluation indexes
values xij (i = 1, 2, ... , n; j = 1, 2, ... , m), then the evaluation index xj is calculated
according to the sample unbiased variance of n power supply enterprises' line losses,
and the formula is:
s. = x.. - X. 2 2,j=,- - ,m n i=1i In which, x, is the sample mean value of the evaluation index xj according to the
line losses of n power supply enterprises, and the calculation formula is:
1 Xi,= Ix,,, j=1,2,- -.,m
B) if there is ko (), so that
k~O
Then the corresponding evaluation index X, of Sko can be deleted;
After the above steps, the evaluation index library of power system line loss lean
management is obtained as follows:
First-level First-level Second-levelindex index Second-levelindex index Name Name Name Name Qualified rate of Planning power supply radius Manage Proportion of old low dimension of distribution dimensions voltage watt-hour meters network
Capacity-load ratio Measurement fault error of main transformer rate Qualified rate of reactive power Abnormal handling rate allocation in of line loss substation 10kV distribution transformer reactive Complete rate of four power compensation types of terminals rate Proportion of cross section that is too Abnormal rate of line loss small Unplanned outage Line loss in public rate of distribution transformer substation lines area Outage rate of public transformer Mediumvoltagelineloss Complaint rate of ten thousand households Compliance rate of service time limit for low voltage industry expansion Repair work order rate of 100 households Electricity sales growth rate Proportion of line Electricity charge weight over light recoveryrate load Proportion of heavy Measure the standard rate andlightloadof of asset management transformer Comprehensive Success rate of electricity voltage qualification consumption information rate collection Run Pass rate of power Abnormal data handling . n factor rate of marketing system dimension Sum of unbalance Low-voltage customer rates of bus power intelligent payment coverage rate Lattice rate Success rate of automatic fee control execution Station power Energy-saving main consumption rate Technical transformer ratio Three-phase dimension Proportion of main unbalance rate transformer for on-load voltage regulation
Overload station Proportion of old area ratio equipment Low voltage station High loss distribution area ratio ratio
Step 2, screening evaluation indexes of power system line loss lean management
by using a random forest model, which comprises the following steps:
Step 2.1, select random forest classifier model to determine the classification
effect of power system line loss lean management of evaluation index set, and all
model parameters adopt default values;
Step 2.2, starting from the evaluation index library in step 1, every random forest
simulation training is performed, one minimum importance evaluation index is
eliminated, and different evaluation index subsets are generated; Evaluation index
importance calculation method is:
a) for c category labels, the Gini value calculation formula of data set D of a
decision tree node is:
Gini(D) =1- p2 in the above formula, pi represents the proportion of class i samples in data set D;
b) the ith tree node j containing the data set D in the random forest is split under
the action of evaluation index xk to generate two new tree nodes, the data sets of the
two new tree nodes are Di and Dr respectively, and the Gini index of the data set D
under the action of evaluation index xk is calculated as follows:
Gini-index(D, x,)= x Gini(D,)+ rx Gini(D,) D D
in which, D , and ID,I are the total number of samples of data sets D, Di and
Dr respectively; c) according to the steps a) and b), the Gini index of the ith tree node j containing the data set D in the random forest under the action of the evaluation index Xk is obtained, and then the Gini gain is obtained by calculating its variation, and the calculation formula is:
VIMg"' = Gini(D) - Gini-index(D, x,) d) assuming that the nodes in the ith tree with the evaluation index xk as the
splitting index all exist in the set J, the calculation formula of the importance of the
evaluation index xk in the ith tree is as follows:
VJini VIM ' " = YVJVIM;G'Y n '
jEJ e) for m evaluation indexes, a random forest model of n trees is generated in total,
and the importance of evaluation index xk is calculated by normalization:
VIM k = YYVIMyin titl t=1 i=1
Through the above steps, the random forest model of the current m evaluation
indexes is trained. When constructing each tree, different bootstrap sample(Random n
extraction with retur) are used for the training set. Therefore, for each tree (assuming
for the kth tree), about one third of the training examples are not involved in the
generation of the kth tree, which is called oob(out-of-bag) samples of the kth tree.
Such sampling characteristics allow oob estimation, and it is calculated as
follows:(Note: take sample as unit)
1) for each sample, calculate its classification as a tree of oob samples (about 1/3
trees);
2) then a simple majority vote is taken as the classification result of the sample;
3) finally, the ratio of the number of misclassification to the total number of
samples is used as the oob misclassification rate of random forest.
Step 2.3, perform curve fitting on the estimation accuracy of all out-of-bag
samples obtained by random forest simulation, determine the inflection point of the
curve change of the estimation accuracy of out-of-bag samples, and take the
evaluation index subset corresponding to the inflection point as the final evaluation
index of power system line loss lean management.
2. The evaluation index screening strategy for lean management of power system
line loss under big data environment according to claim 1 is characterized in that in
step 2.3, the step of determining the inflection point of curve change of estimation
accuracy of out-of-bag samples is as follows:
a) deriving the fitting curve formulas of all the estimation accuracy of the out-of
bag samples obtained by simulation, and obtaining the formula after derivation of the
fitting curve.
b) taking logarithm of the formula derived from the fitting curve, and taking the
simulation times corresponding to the minimum logarithm value as the inflection
point of curve change of estimation accuracy of out-of-bag samples.
Compared with the prior art, the power system line loss lean management
evaluation index screening strategy under the big data environment has the
advantages that:
1. Multi-method combination (frequency analysis method, Delphi method and
LMS algorithm) is adopted to establish the evaluation index database of power system
line loss lean management, which avoids the dimension disaster caused by huge
evaluation index data in random forest mode11.
2. Combining the evaluation system and artificial intelligence technology, the
evaluation index system of line loss lean management in power system is established
by selecting the evaluation index model based on random forest.
3. Applying the random forest to the evaluation index screening overcomes the
deficiency that the traditional evaluation index screening method relies too much on
the collective wisdom of experts.
In order to explain the technical scheme of the embodiment of the present
invention more clearly, the following will briefly introduce the drawings that need to
be used in the embodiment or related technical description.
Fig. 1 is the basic flow of establishing the evaluation index system of power
system line loss lean management.
Fig. 2 is the screening strategy of evaluation index based on random forest.
Fig. 3 is a distribution graph of estimation accuracy of 38 out-of-bag samples.
Fig. 4 is a logarithmic graph after derivation of the fitting curve of the estimation
accuracy of out-of-bag samples.
In order to understand the above objects, features and advantages of the present
invention more clearly, the technical scheme of the present invention will be further
described in detail with reference to the drawings and specific embodiments.
In this embodiment, 38 line loss evaluation index data of 104 power supply
enterprises are selected for simulation analysis, and the line loss lean management
evaluation index screening work of power system is completed.
According to the screening process of evaluation index of random forest as
shown in fig. 1 and fig. 2, this example has carried out 38 random forest simulations,
thus obtaining the estimation accuracy of 38 out-of-bag samples as shown in fig. 3.
Considering that the fluctuation of estimation accuracy of out-of-bag samples in
38 random forest simulations in this example is not obvious, it is difficult to visually determine the inflection point of estimation accuracy of out-of-bag samples. This example determines the inflection point of estimation accuracy curve of out-of-bag samples in 38 random forest simulations by curve fitting, then taking the logarithm of the fitting curve. The logarithmic value after derivation of the fitting curve for the estimation accuracy of out-of-bag samples is shown in Figure 4.
From Figure 3, it can be determined that the inflection point of the change of
estimation accuracy of 38 out-of-bag samples appears in the 27th random forest
simulation, and the corresponding 12 evaluation indexes can be used as the evaluation
index system of power system line loss lean management, as shown in Table 2.
Table 2:
First-level index Second-level index No. Name No. Name Qualified rate of power C1l supply radius of Planning distribution network C1 Plamn Reactive power allocation dimension C12 ratio of substation
C13 Proportion of cross section that is too small Complete rate of data Manage C21 acquisition for four types C2 dimensions of terminals C22 Abnormal rate of line loss
C23 Electricity sales growth rate C24 Repair work order rate of 100 households C31 Proportion of heavy and light load of transformer C3 Run C32 Comprehensive voltage dimension qualification rate C33 Three-phase unbalance rate C41 Energy-saving main C4 Technical transformer ratio dimension C42 Proportion of old equipment
Claims (2)
1. an evaluation index screening strategy for lean management of power system
line loss under big data environment is characterized by comprising the following
steps:
Step 1, establish an evaluation index library of power system line loss lean
management;
First-level First-level Second-levelindex idx index Second-level index index Seon-lveeide Name Name Name Name Qualified rate of power supply radius Proportion of old low of distribution voltage watt-hour meters network Capacity-load ratio Measurement fault error of main transformer rate Qualified rate of reactive power Abnormal handling rate allocation in of line loss substation 10kV distribution transformer reactive Complete rate of four power compensation types of terminals rate Proportion of cross Planning section that is too Abnormal rate of line loss dimension small Manage Unplanned outage dimensions Line loss in public rate of distribution transformer substation lines area Outage rate of public transformer Mediumvoltagelineloss Complaint rate of ten thousand households Compliance rate of service time limit for low voltage industry expansion Repair work order rate of 100 households Electricity sales growth rate Run Proportion of line Electricity charge dimension weight over light recovery rate load Proportion of heavy Measure the standard rate andlightloadof of asset management transformer Comprehensive Success rate of electricity voltage qualification consumption information rate collection Pass rate of power Abnormal data handling factor rate of marketing system Sum of unbalance Low-voltage customer rates of bus power intelligent payment coverage rate Lattice rate Success rate of automatic fee control execution Station power Energy-saving main consumption rate transformer ratio Three-phase Proportion of main unbalance rate Technical transformer for on-load . dimension voltage regulation Overload station Proportion of old area ratio equipment Low voltage station High loss distribution area ratio ratio
Step 2, screening evaluation indexes of power system line loss lean management
by using a random forest model, which comprises the following steps:
Step 2.1, select random forest classifier model to determine the classification
effect of power system line loss lean management of evaluation index set, and all
model parameters adopt default values;
Step 2.2, starting from the evaluation index library in step 1, every random forest
simulation training is performed, one minimum importance evaluation index is
eliminated, and different evaluation index subsets are generated; Evaluation index
importance calculation method is:
a) for c category labels, the Gini value calculation formula of data set D of a
decision tree node is:
Gini(D) =1- c= p2 in the above formula, pi represents the proportion of class i samples in data set D; b) the ith tree node j containing the data set D in the random forest is split under the action of evaluation index Xkto generate two new tree nodes, the data sets of the two new tree nodes are Di and Dr respectively, and the Gini index of the data set D under the action of evaluation index Xk is calculated as follows:
Gii ndx(D,| R- Giniindex(D, xk)= |D,| x Gini(D,)+ x Gini(D,) D D
in which, D , and ID,I are the total number of samples of data sets D, Di and
Dr respectively;
c) according to the steps a) and b), the Gini index of the ith tree node j containing
the data set D in the random forest under the action of the evaluation index xk is
obtained, and then the Gini gain is obtained by calculating its variation, and the
calculation formula is:
VIMY"I = Gini(D)- Gini-index(D, x,) d) assuming that the nodes in the ith tree with the evaluation index xk as the
splitting index all exist in the set J, the calculation formula of the importance of the
evaluation index xk in the ith tree is as follows:
VJini VIM VIM4,ji' ' " = YVIM '
jEJ e) for m evaluation indexes, a random forest model of n trees is generated in total,
and the importance of evaluation index xk is calculated by normalization:
VIMk" VIM = n Y VIMj" t=1 i=1
Step 2.3, perform curve fitting on the estimation accuracy of all out-of-bag
samples obtained by random forest simulation, determine the inflection point of the
curve change of the estimation accuracy of out-of-bag samples, and take the evaluation index subset corresponding to the inflection point as the final evaluation index of power system line loss lean management.
2. The evaluation index screening strategy for lean management of power system
line loss under big data environment according to claim 1 is characterized in that in
step 2.3, the step of determining the inflection point of curve change of estimation
accuracy of out-of-bag samples is as follows:
a) deriving the fitting curve formulas of all the estimation accuracy of the out-of
bag samples obtained by simulation, and obtaining the formula after derivation of the
fitting curve.
b) taking logarithm of the formula derived from the fitting curve, and taking the
simulation times corresponding to the minimum logarithm value as the inflection
point of curve change of estimation accuracy of out-of-bag samples.
Figures 1/3
Figure 1
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2021106109A AU2021106109A4 (en) | 2021-08-20 | 2021-08-20 | Evaluation index screening strategy for lean management of power system line loss under big data environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2021106109A AU2021106109A4 (en) | 2021-08-20 | 2021-08-20 | Evaluation index screening strategy for lean management of power system line loss under big data environment |
Publications (1)
Publication Number | Publication Date |
---|---|
AU2021106109A4 true AU2021106109A4 (en) | 2021-10-28 |
Family
ID=78179680
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU2021106109A Ceased AU2021106109A4 (en) | 2021-08-20 | 2021-08-20 | Evaluation index screening strategy for lean management of power system line loss under big data environment |
Country Status (1)
Country | Link |
---|---|
AU (1) | AU2021106109A4 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114897108A (en) * | 2022-07-08 | 2022-08-12 | 浙江省标准化研究院(金砖国家标准化(浙江)研究中心、浙江省物品编码中心) | Biocidal market admission evaluation method and system |
CN116089782A (en) * | 2022-11-23 | 2023-05-09 | 国网甘肃省电力公司临夏供电公司 | Multi-guide matched power distribution network refined line loss data system and application thereof |
-
2021
- 2021-08-20 AU AU2021106109A patent/AU2021106109A4/en not_active Ceased
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114897108A (en) * | 2022-07-08 | 2022-08-12 | 浙江省标准化研究院(金砖国家标准化(浙江)研究中心、浙江省物品编码中心) | Biocidal market admission evaluation method and system |
CN114897108B (en) * | 2022-07-08 | 2022-09-13 | 浙江省标准化研究院(金砖国家标准化(浙江)研究中心、浙江省物品编码中心) | Biocidal market admission evaluation method and system |
CN116089782A (en) * | 2022-11-23 | 2023-05-09 | 国网甘肃省电力公司临夏供电公司 | Multi-guide matched power distribution network refined line loss data system and application thereof |
CN116089782B (en) * | 2022-11-23 | 2023-11-21 | 国网甘肃省电力公司临夏供电公司 | Multi-guide matched power distribution network refined line loss data system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113159339B (en) | One-area one-index line loss management method and system based on big data | |
AU2021106109A4 (en) | Evaluation index screening strategy for lean management of power system line loss under big data environment | |
CN112131441B (en) | Method and system for rapidly identifying abnormal electricity consumption behavior | |
CN104504508B (en) | The taiwan area closed-loop data analysis method returned based on step analysis and small echo | |
CN111507013A (en) | Line loss fault positioning implementation method for power system | |
CN104331773A (en) | Comprehensive assessment method for power network planning schemes | |
CN110910026B (en) | Cross-provincial power transmission line loss intelligent management and decision method and system | |
CN110826228B (en) | Regional power grid operation quality limit evaluation method | |
CN112785060A (en) | Lean operation and maintenance level optimization method for power distribution network | |
CN112508232A (en) | Short-circuit current limitation measure evaluation method based on multi-level fuzzy comprehensive evaluation model | |
CN110533291B (en) | Medium voltage distribution network weak link identification method based on risk assessment | |
CN116011827A (en) | Power failure monitoring analysis and early warning system and method for key cells | |
CN115600934A (en) | Investment efficiency benefit evaluation method and device for power distribution network engineering | |
Shaghaghi et al. | Proposing a new optimized forecasting model for the failure rate of power distribution network thermal equipment for educational centers | |
Lin et al. | Voltage sag severity analysis based on improved FP-Growth algorithm and AHP algorithm | |
CN109245109B (en) | Power supply low-voltage early warning system and method | |
Li et al. | Distribution transformer mid-term heavy load and overload pre-warning based on logistic regression | |
CN114971272A (en) | Medium-voltage distribution network gridding planning evaluation method and system | |
CN110717725B (en) | Power grid project selection method based on big data analysis | |
CN112016631A (en) | Improvement scheme related to low-voltage treatment | |
Li et al. | Construction and application of intelligent evaluation indicator system of line loss lean management based on knowledge graph | |
Yang et al. | AI-Based Detection of Power Consumption Behavior of People in a Smart City | |
Yan et al. | Comprehensive energy efficiency rating evaluation model of enterprise power based on grid big data | |
Lingang et al. | Research on integrated calculation method of theoretical line loss of MV and LV distribution Network based on Adaboost integrated learning | |
Zhong et al. | [Retracted] Analysis and Design of the Project Risk Management System Based on the Fuzzy Clustering Algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FGI | Letters patent sealed or granted (innovation patent) | ||
MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |