CN106921158B - demand coefficient analysis method for historical collected data based on time sequence of distribution transformer - Google Patents
demand coefficient analysis method for historical collected data based on time sequence of distribution transformer Download PDFInfo
- Publication number
- CN106921158B CN106921158B CN201710070453.5A CN201710070453A CN106921158B CN 106921158 B CN106921158 B CN 106921158B CN 201710070453 A CN201710070453 A CN 201710070453A CN 106921158 B CN106921158 B CN 106921158B
- Authority
- CN
- China
- Prior art keywords
- prediction
- data
- coefficient
- sequence
- distribution transformer
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 6
- 238000000034 method Methods 0.000 claims description 19
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000011160 research Methods 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 2
- 238000012886 linear function Methods 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 238000011161 development Methods 0.000 abstract description 3
- 238000005065 mining Methods 0.000 abstract 1
- 238000012549 training Methods 0.000 description 5
- 230000009466 transformation Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- 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
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses demand coefficient analysis methods for historical collected data based on a time sequence of a distribution transformer, which are used for mining the intrinsic rule of the historical data by analyzing the historical data of the demand coefficient based on the time sequence and further predicting the future development trend of the historical data.
Description
Technical Field
The invention relates to a calculation method for load rates of distribution network transformers, in particular to a prediction method for demand coefficients of distribution network resident load transformers.
Background
In recent years, with the rapid development of the economy of China, the power consumption is increased rapidly, the requirement of users on the quality of electric energy is continuously improved, the use condition of residential power transformers of a power distribution network is widely concerned by , in the actual operation of the power distribution network, the reported capacity of the residential power load transformers is not , all the residential power load transformers are connected to the power distribution network, the capacity of the connected power distribution network changes along with time, the ratio of the actual connected capacity of the power distribution transformers to the reported capacity of distribution transformers is researched, the power distribution network can be better guided to be scientifically and reasonably scheduled, the safe operation of the power distribution network is guaranteed, the power supply capacity of the existing power distribution network can be evaluated, the calculation accuracy of the capacity of a feeder line is improved, the calculation accuracy of the capacity of the feeder.
At present, the research of the demand coefficient mainly focuses on the relation between the demand coefficient and the number of the residential community, namely, the electrical design specification of the residential building gives the value range of the demand coefficient corresponding to different numbers of the residential building, however, the value of the demand coefficient given in the specification is large, the value of the demand coefficient is corrected by according to the methods of a numerical regression analysis method, a mathematical statistics method, a nuclear density estimation Parzen window method and the like in the related research, but the dynamic characteristic of the demand coefficient changing along with time is not discussed.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides methods for analyzing the demand coefficient of historical collected data based on the time sequence of the distribution transformer, which can accurately predict the future development trend of the distribution transformer.
need coefficient analysis method based on distribution transformer time sequence history collected data, which comprises (1) data collection, collecting distribution transformer reporting capacity and need coefficient related variable, providing data base for next research;
(2) need coefficient calculation: analyzing original data, and obtaining dynamic data of distribution transformer required coefficients based on time series by using required coefficient definition and a related formula;
(3) grey prediction: according to the actual situation with less sampling data and the excellent characteristics of gray prediction in small sample prediction, a gray prediction GM (1,1) model is constructed, and gray prediction fitting values are obtained through the steps of accumulation, subtraction and the like;
(4) BP neural network prediction: the fitting value obtained by the grey prediction model reflects the size of the actual value to a great extent, and due to the excellent fitting and approximation characteristics of the BP neural network in the aspect of nonlinear function relationship, the BP neural network is used for finding out the correlation relationship between the actual value and obtaining the final prediction result.
The BP neural network is divided into 3 layers: input layer, hidden layer, and intermediate layer. The invention adopts sigmod function for hidden layer and linear function for output layer. The SCG algorithm is adopted for training, and the calculation efficiency is improved.
In summary, compared with the prior art, the invention has the following advantages:
the method fully considers the dynamic characteristic of the change of the demand coefficient along with time, obtains the predicted value of the demand coefficient in the next months by using a data mining algorithm, improves the load rate evaluation capability of the distribution transformer, and provides data support for the relevant work of the power grid part .
Drawings
FIG. 1 is a flow chart of a method for analyzing demand coefficients of historically collected data based on time series of distribution transformers according to the present invention.
FIG. 2 is a comparison graph of prediction error for the combined prediction method of the present invention and the uncombined single method.
Detailed Description
The present invention will be described in more detail with reference to examples.
Example 1
In this embodiment, a single transformer in a city of a certain provincial meeting is taken as an example to explain the specific method of the invention. The original data is a time sequence consisting of the reporting capacity of the transformer and the load variable related to the demand coefficient. The time series ranged from 13 months to 48 months of commissioning for 36 sets of monthly data.
1. And (6) data acquisition.
The distribution network power distribution transformer comprises reporting capacity of each resident power distribution transformer of the distribution network, monthly three-phase maximum current, phase voltage, current transformer transformation ratio CT and voltage transformer transformation ratio PT.
2. Coefficient calculations are required.
The coefficient required calculation is as shown in equation (1):
in the formula, L is the maximum value of the monthly load of the load user transformer, and L is the reported capacity of the user load transformer.
The maximum monthly load value l is calculated as shown in formula (2):
l＝U×PT×(IAmax+IBmax+ICmax)×CT (2)
in the formula, U is the voltage measured by the transformer, PT and CT are the transformation ratio of the voltage-very-current transformer, IAmax,IBmax,ICmaxThe maximum values of the three-phase measured currents of the transformer A, B, C are respectively.
The values of the distribution required coefficients calculated are shown in table 1:
TABLE 1 distribution transformer requirement coefficient
3. And predicting grey.
A grey predictive GM (1,1) model was constructed. Firstly, accumulating original time sequence data one by one to generate a new sequence; then, solving the parameters of the sequence differential equation generated by accumulation by using a least square parameter estimation method; then, solving the differential equation, namely accumulating the fitting values of the generated sequence; and finally, sequentially subtracting and reducing the fitting values of the accumulated generated sequence to obtain the fitting values of the original time sequence.
4. And constructing a BP neural network training sample set and a test sample set.
And (3) corresponding the gray prediction fitting value of the required coefficient to the actual value to obtain 36 pairs of data, dividing the 36 pairs of data into 9 groups according to the time sequence, wherein each group comprises 4 pairs of data, the first 8 groups are used as training sample sets, and the last groups are used as test sample sets.
BP neural network prediction.
Firstly, training by using a training sample set to obtain a BP neural network prediction model: taking the grey prediction fitting value of each group for 4 months as neural network input, and taking the corresponding actual value of 4 months as output; and then, applying the trained model to predict: and inputting the 9 th group of gray prediction fitting values of 4 months into the trained neural network model to obtain neural network prediction data. The predicted results are shown in table 2 below. To illustrate the prediction accuracy of the combined method of the present invention, the following table 2 simultaneously lists the prediction results of the gray prediction model and the BP neural network prediction model using two separate prediction methods.
Table 23 model prediction results
6. Error analysis and prediction result evaluation
The effect and the precision of the combined prediction algorithm are evaluated by adopting the relative error epsilon:
in the formula (d)preAnd drealThe specific numerical values are the prediction errors of the combined model and the two single models in the figure 2.
From fig. 2, the error of the combined prediction method is generally lower than that of the other two methods, which shows that the combined prediction model can improve the prediction accuracy and is superior to the two single prediction models, thereby illustrating the superiority of the algorithm of the present invention.
The parts not described in the present embodiment are the same as those in the prior art.
Claims (2)
- methods for analyzing the need coefficient of historical collected data based on the time sequence of distribution transformer, which is characterized by comprising the steps of (1) data collection, namely collecting distribution transformer reporting capacity and the related variable of the need coefficient, and providing a data base for the next research;(2) coefficient calculation is required: analyzing original data, and obtaining dynamic data of distribution transformer required coefficients based on time series by using required coefficient definition and a related formula;(3) and grey prediction: according to the actual situation that sampling data are few and the excellent characteristic of gray prediction in small sample prediction, a gray prediction GM (1,1) model is constructed, and firstly, the original time sequence data are accumulated one by one to generate a new sequence; then, solving the parameters of the sequence differential equation generated by accumulation by using a least square parameter estimation method; then, solving the differential equation, namely accumulating the fitting values of the generated sequence; finally, sequentially subtracting and reducing the fitting values of the accumulated generated sequence to obtain the fitting values of the original time sequence;(4) BP neural network prediction: the fitting value obtained by the grey prediction model reflects the size of the actual value, and because of the excellent fitting and approximation characteristics of the BP neural network in the aspect of nonlinear function relationship, the BP neural network is used for finding out the correlation relationship between the actual value and obtaining the final prediction result.
- 2. The method of claim 1 for demand coefficient analysis of historically collected data based on time series of distribution transformers, wherein: the BP neural network is divided into 3 layers: the device comprises an input layer, a hidden layer and an intermediate layer, wherein the hidden layer adopts a sigmod function, and the output layer adopts a linear function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710070453.5A CN106921158B (en) | 2017-02-09 | 2017-02-09 | demand coefficient analysis method for historical collected data based on time sequence of distribution transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710070453.5A CN106921158B (en) | 2017-02-09 | 2017-02-09 | demand coefficient analysis method for historical collected data based on time sequence of distribution transformer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106921158A CN106921158A (en) | 2017-07-04 |
| CN106921158B true CN106921158B (en) | 2020-01-31 |
Family
ID=59454633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710070453.5A Active CN106921158B (en) | 2017-02-09 | 2017-02-09 | demand coefficient analysis method for historical collected data based on time sequence of distribution transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106921158B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256724B (en) * | 2017-11-27 | 2021-12-07 | 国网浙江省电力公司嵊泗县供电公司 | Power distribution network open capacity planning method based on dynamic industry coefficient |
| CN108197773A (en) * | 2017-12-08 | 2018-06-22 | 囯网河北省电力有限公司电力科学研究院 | Methods of electric load forecasting, load forecast device and terminal device |
| CN108258683A (en) * | 2018-01-19 | 2018-07-06 | 国网江苏省电力有限公司苏州供电分公司 | Power distribution network transformer capacity service condition Forecasting Methodology |
| CN109543769A (en) * | 2018-11-30 | 2019-03-29 | 国网山东省电力公司电力科学研究院 | A kind of transformer station high-voltage side bus shortage of data mending method based on function type principal component analysis and wavelet transformation |
| CN112114215A (en) * | 2020-09-17 | 2020-12-22 | 陕西师范大学 | Transformer aging evaluation method and system based on error back propagation algorithm |
| CN112528371A (en) * | 2020-12-09 | 2021-03-19 | 四川蓉信开工程设计有限公司 | Electrical design method for water treatment project |
| CN112487665B (en) * | 2020-12-18 | 2022-09-09 | 天津博迈科海洋工程有限公司 | Method for calculating actual demand coefficient of ocean engineering electrical equipment based on probability statistics |
| CN115248906B (en) * | 2022-09-22 | 2023-01-06 | 武汉格蓝若智能技术有限公司 | State error identification method and system for double current transformers on outgoing line of generator |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105447328A (en) * | 2015-12-25 | 2016-03-30 | 国网浙江诸暨市供电公司 | Method for determining demand factor in residential building electrical load calculation |
| CN106056234A (en) * | 2016-05-18 | 2016-10-26 | 北京博锐尚格节能技术股份有限公司 | Transformer capacity determination method and device |
-
2017
- 2017-02-09 CN CN201710070453.5A patent/CN106921158B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105447328A (en) * | 2015-12-25 | 2016-03-30 | 国网浙江诸暨市供电公司 | Method for determining demand factor in residential building electrical load calculation |
| CN106056234A (en) * | 2016-05-18 | 2016-10-26 | 北京博锐尚格节能技术股份有限公司 | Transformer capacity determination method and device |
Non-Patent Citations (2)
| Title |
|---|
| 基于串联灰色神经网络的电力负荷预测方法;张大海等;《系统工程理论与实践》;20041231;第130页倒数后11行-第131页前5行 * |
| 基于核密度估计的需要系数求取方法;孟明等;《建筑电气》;20160531;第43页第2栏第2段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106921158A (en) | 2017-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106921158B (en) | demand coefficient analysis method for historical collected data based on time sequence of distribution transformer | |
| Chai et al. | Coal consumption in China: how to bend down the curve? | |
| He et al. | Forecasting China's total energy demand and its structure using ADL-MIDAS model | |
| CN104914327B (en) | Transformer fault maintenance Forecasting Methodology based on real-time monitoring information | |
| CN106779277B (en) | Classified evaluation method and device for network loss of power distribution network | |
| CN103020459B (en) | A kind of cognitive method of various dimensions electricity consumption behavior and system | |
| CN104616079A (en) | Temperature change based power grid daily electricity consumption prediction method | |
| CN106980910B (en) | Medium-and-long-term power load measuring and calculating system and method | |
| CN110070282A (en) | A kind of low-voltage platform area line loss analysis of Influential Factors method based on Synthesis Relational Grade | |
| CN110969306B (en) | Power distribution low-voltage transformer area load prediction method and device based on deep learning | |
| CN104268402A (en) | Power system load clustering method based on fuzzy c-means algorithm | |
| Xu et al. | A water level prediction model based on ARIMA-RNN | |
| CN113723844A (en) | Low-voltage transformer area theoretical line loss calculation method based on ensemble learning | |
| CN104092215B (en) | Distribution transformer capacity control method and system | |
| CN103018611A (en) | Non-invasive load monitoring method and system based on current decomposition | |
| CN104112204A (en) | Evaluation method for efficient operation of power supply quality | |
| CN113642136A (en) | Power grid topology identification method based on HPLC broadband carrier module | |
| CN105005623A (en) | Power demand prediction method based on keyword retrieval index correlation analysis | |
| CN104008493B (en) | A kind of data capture method and device | |
| CN107846025A (en) | System and method for adjusting power system voltage | |
| CN110414776B (en) | Quick response analysis system for power utilization characteristics of different industries | |
| CN112421608A (en) | Family load prediction method based on Markov from bottom to top | |
| CN111275292A (en) | Ground grid state evaluation method based on fuzzy analytic hierarchy process | |
| CN111552911B (en) | Quantitative analysis method for technical line loss influence factors based on multi-scene generation | |
| CN105654208B (en) | Empirical mode decomposition method for determining maximum value of cellular load in space load prediction |
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 |