CN114861808B - Intelligent sorting method for variable load test data of hydroelectric generating set - Google Patents
Intelligent sorting method for variable load test data of hydroelectric generating set Download PDFInfo
- Publication number
- CN114861808B CN114861808B CN202210546816.9A CN202210546816A CN114861808B CN 114861808 B CN114861808 B CN 114861808B CN 202210546816 A CN202210546816 A CN 202210546816A CN 114861808 B CN114861808 B CN 114861808B
- Authority
- CN
- China
- Prior art keywords
- stable
- signal
- active power
- interval
- value
- 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
- 238000012360 testing method Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000005070 sampling Methods 0.000 claims abstract description 5
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 abstract description 3
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/20—Hydro energy
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Complex Calculations (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
An intelligent sorting method for variable load test data of a hydroelectric generating set comprises the following steps: selecting signals for judging stable working conditions, wherein the signals at least comprise active power P, other signals are selectable signals, setting the maximum allowable fluctuation amount of each signal in the stable working conditions, and setting the length t of a stable interval; continuously traversing the data of the whole test duration by taking the sampling time as a step length, and obtaining the difference value between the maximum value and the minimum value of each parameter in each time window every time of traversing; stability index s of each stable interval; the m stable interval sections are classified according to the active power values, the classification principle is that values similar to the active power are classified into the same class, and k classes are finally obtained; and (3) finding out the time interval section with the minimum s value in each class, and taking the time interval section as the finally screened stable operation interval section. The method can realize automatic screening of steady-state working condition data in the variable load test process, can accurately identify the steady working condition, and is suitable for sorting large-scale data.
Description
Technical Field
The invention belongs to the field of variable load test data processing of a hydroelectric generating set, and particularly relates to an intelligent sorting method for variable load test data of a hydroelectric generating set.
Background
The variable load test of the hydroelectric generating set needs to adjust test working conditions, and the hydroelectric generating set stably operates for a period of time under each working condition during the test, then changes to the next working condition, and then stably operates for a period of time, and the test is repeated in sequence until all the test working conditions are covered. The test data are continuously collected, namely the whole variable working condition test process is collecting data, and the steady-state working condition with fixed duration is generally only needed to be analyzed during data analysis, so that a plurality of steady-state interval sections with fixed duration are needed to be sorted from the collected whole continuous signal, and only one steady-state time interval section is sorted nearby the same working condition (namely the same active power value), so that the whole continuous duration can be divided into a plurality of time interval sections for subsequent data processing analysis.
The traditional method is generally carried out by adopting manual identification, namely, the active power value is identified as a stable working condition when the active power value is stable and unchanged (or has small change) through manually observing a waveform diagram of the active power in the signal, and then the time period is manually selected.
The traditional method has lower efficiency, the sorted waveform interval is not necessarily the interval with the minimum fluctuation of active power, meanwhile, the stable working condition is limited only by one signal of the active power, complex screening conditions cannot be realized, the stability of the active power cannot completely represent the stable working condition, and when the active power is stable and other signals (such as flow signals) are not stable, the actual process still belongs to a transition process, and if the process is regarded as the stable working condition, the subsequent data analysis is influenced to a certain extent.
Disclosure of Invention
In view of the technical problems existing in the background art, the intelligent sorting method for the variable load test data of the hydroelectric generating set can automatically find out a time interval with optimal stability near the same active power value; can carry out autofilter to large batch data, computational efficiency is high.
In order to solve the technical problems, the invention adopts the following technical scheme:
an intelligent sorting method for variable load test data of a hydroelectric generating set comprises the following steps:
S1: selecting signals for judging stable working conditions, wherein the signals at least comprise active power P, other signals are selectable signals, and the code number of the selectable signals is X 1、X2...Xn; n is a natural positive integer;
S2: setting the maximum allowable fluctuation amount of each signal under the stable working condition, and if the signal has only active power P, setting the maximum allowable fluctuation amount as MP; if the signal has a selective signal, the maximum allowable fluctuation amount is MP, MX 1、MX2...MXn;
S3: setting the length t of a stable interval section;
S4: starting from the moment 0, continuously traversing the data of the whole test duration by using a time window with the length of t and taking the sampling time as a step length, and obtaining the difference value between the maximum value and the minimum value of each parameter in each time window every time; if the signal has only active power P, the difference between the maximum value and the minimum value is delta P; if the signal has a selective signal, the difference between the maximum value and the minimum value is delta P and delta X 1、ΔX2...ΔXn respectively;
If the signal has only active power P, when the signal meets delta P < M P, the time interval is considered to be a stable operation interval meeting the requirement, and all the time intervals meeting the requirement are sequentially numbered as t 1,t2...tm; m is a natural positive integer;
If the signal has a selective signal, when the delta P is simultaneously satisfied with M P,ΔX1<MX1,ΔX2<MX2...ΔXn<MXn, the time interval is considered to be a stable operation interval which satisfies the requirement, and all the time intervals which satisfy the requirement are sequentially numbered as t 1,t2...tm; m is a natural positive integer;
S5: calculating the stability index s of each stable interval section obtained in the step 4;
s6: classifying the m stable interval sections obtained in the step 4 according to the active power values, wherein the classification principle is that the values similar to the active power are classified into the same class, and finally k classes are obtained; the signals are classified according to active power no matter how many types;
s7: and (3) finding out the time interval section with the minimum s value in each class from k classes to serve as the finally screened stable operation interval section, so that k time interval sections with the best stability index meeting the condition can be obtained and serve as the finally output sorting results T1 and T2 … … Tk.
In a preferred embodiment, the calculation method in step S5 is as follows:
S5.1: the maximum value of each parameter over the entire test duration is read,
If the signal has only active power P, the maximum value is PMax;
if the signal has an optional signal, the values of the signals are PMax and X 1Max、X2Max...Xn Max in sequence;
S5.2: calculating normalized indexes of fluctuation values of parameters in each time interval meeting the conditions:
sP=ΔP/PMax,sX1=ΔX1/X1Max,sX2=ΔX2/X2Max...sXn=ΔXn/XnMax;
s5.3: the stability index s=sp+sx 1+sX2+...sXn of each stable section obtained in step 4 was obtained.
In a preferred embodiment, in the step S6, an average value PAVG of the active power in each stable interval obtained in the step 4 is calculated, and the m stable intervals obtained in the step 4 are classified according to the PAVG value by using a DBSCAN density clustering algorithm to obtain k classes.
The following beneficial effects can be achieved in this patent:
1. The method can be compiled into degree automatic operation, and has high calculation efficiency and good accuracy. The method can simultaneously consider the identification of a plurality of parameters to the stable operation condition, can automatically realize the automatic screening of the stable operation condition data in the variable load test process through a program, has strong identification accuracy and high efficiency, is suitable for sorting large-batch data, can customize the identification parameters of the stable operation condition, and can simultaneously consider the identification of a plurality of parameters to the stable operation condition;
2. a time interval with optimal stability near the same active power value can be automatically found out; can carry out autofilter to large batch data, computational efficiency is high.
Drawings
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
FIG. 1 is a schematic diagram of the data sorting result according to the present invention;
FIG. 2 is a schematic view of a continuous traversal of the present invention.
Detailed Description
The preferable scheme is as shown in fig. 1 to 2, and the intelligent sorting method for the variable load test data of the hydroelectric generating set comprises the following steps:
S1: selecting signals for judging stable working conditions, wherein the signals at least comprise active power P, other signals are selectable signals, and the code number of the selectable signals is X 1、X2...Xn; n is a natural positive integer;
S2: setting the maximum allowable fluctuation amount of each signal under the stable working condition, and if the signal has only active power P, setting the maximum allowable fluctuation amount as MP; if the signal has a selective signal, the maximum allowable fluctuation amount is MP, MX 1、MX2...MXn;
S3: setting the length t of a stable interval section;
S4: starting from the moment 0, continuously traversing the data of the whole test duration by using a time window with the length of t and taking the sampling time as a step length, and obtaining the difference value between the maximum value and the minimum value of each parameter in each time window every time; if the signal has only active power P, the difference between the maximum value and the minimum value is delta P; if the signal has a selective signal, the difference between the maximum value and the minimum value is delta P and delta X 1、ΔX2...ΔXn respectively;
If the signal has only active power P, when the signal meets delta P < M P at the same time, the time interval is considered to be a stable operation interval meeting the requirement, and all the time intervals meeting the requirement are sequentially numbered as t 1,t2...tm; m is a natural positive integer;
If the signal has a selective signal, when the delta P is simultaneously satisfied with M P,ΔX1<MX1,ΔX2<MX2...ΔXn<MXn, the time interval is considered to be a stable operation interval which satisfies the requirement, and all the time intervals which satisfy the requirement are sequentially numbered as t 1,t2...tm; m is a natural positive integer;
S5: calculating the stability index s of each stable interval section obtained in the step 4;
the calculation method in step S5 is as follows:
S5.1: the maximum value of each parameter over the entire test duration is read,
If the signal has only active power P, the maximum value is PMax;
if the signal has an optional signal, the values of the signals are PMax and X 1Max、X2Max...Xn Max in sequence;
S5.2: calculating normalized indexes of fluctuation values of parameters in each time interval meeting the conditions:
sP=ΔP/PMax,sX1=ΔX1/X1Max,sX2=ΔX2/X2Max...sXn=ΔXn/XnMax;
s5.3: the stability index s=sp+sx 1+sX2+...sXn of each stable section obtained in step 4 was obtained.
S6: classifying the m stable interval sections obtained in the step 4 according to the active power values, wherein the classification principle is that the values similar to the active power are classified into the same class, and finally k classes are obtained; the signals are classified according to active power no matter how many types;
in step S6, the average value PAVG of the active power in each stable interval obtained in step 4 is calculated, and the m stable intervals obtained in step 4 are classified according to the PAVG value by using a DBSCAN density clustering algorithm, so as to obtain k classes.
S7: and (3) finding out the time interval section with the minimum s value in each class from k classes to serve as the finally screened stable operation interval section, so that k time interval sections with the best stability index meeting the condition can be obtained and serve as the finally output sorting results T1 and T2 … … Tk.
Example 1:
The following is an example explanation when the selectable signal is the flow Q:
and step 1, selecting signal active power P and flow Q for judging stable working conditions.
And 2, setting the maximum allowable fluctuation amount MP=5MW of each signal under the stable working condition, wherein MQ=10m 3/s.
And 3, setting the length t of the stable interval section. t is taken for 30s.
And 4, starting from the moment 0, continuously traversing the data of the whole test duration by using a time window with the length of t and taking the sampling time as a step length, and obtaining the difference value between the maximum value and the minimum value of each parameter in each time window, namely delta P and delta Q (delta P=4MW and delta Q=6m 3/s is taken as an example) for each traversing, and when delta P < MP and delta Q < MQ are simultaneously satisfied, namely the time interval is considered as a section of stable operation interval meeting the requirement, and each time interval meeting the requirement is sequentially numbered as t 1,t2...tm.
And 5, calculating the stability index s of each stable section obtained in the step 4, wherein the calculation method is as follows:
Step 5.1, reading the maximum value of each parameter in the whole test duration range, wherein the maximum value is PMax=800 MW, and Qmax=600 m 3/s in sequence;
Step 5.2, obtaining the normalized index of each parameter fluctuation value in each time interval meeting the condition
Sp=Δp/pmax=0.005, sq=Δq/QMax =0.01 (taking the interval illustrated in step 4 as an example);
Step 5.3, the stability index s=sp+sq=0.015 (taking the section exemplified in step 4 as an example) of each stable section obtained in step 4 is obtained.
And 6, classifying the m stable interval sections obtained in the step 4 according to the active power values, wherein the classification principle is that the values similar to the active power are classified into the same class, and the specific implementation method is as follows:
Calculating an average value pavg=700 MW of active power in each stable interval section obtained in the step 4 (taking an interval exemplified in the step 4 as an example), and classifying m stable interval sections obtained in the step 4 according to the PAVG value by using a DBSCAN density clustering algorithm to obtain k classes.
And 7, finding out the time interval section with the minimum s value in each class of the k classes obtained in the step 6as the finally screened stable operation interval section, so that k time interval sections with the best stability index meeting the condition can be obtained and used as the finally output sorting results T1 and T2 … … Tk.
The above embodiments are only preferred embodiments of the present invention, and should not be construed as limiting the present invention, and the scope of the present invention should be defined by the claims, including the equivalents of the technical features in the claims. I.e., equivalent replacement modifications within the scope of this invention are also within the scope of the invention.
Claims (2)
1. An intelligent sorting method for variable load test data of a hydroelectric generating set is characterized by comprising the following steps:
S1: selecting signals for judging stable working conditions, wherein the signals at least comprise active power P, other signals are selectable signals, and the code number of the selectable signals is X 1、X2...Xn; n is a natural positive integer;
S2: setting the maximum allowable fluctuation amount of each signal under the stable working condition, and if the signal has only active power P, setting the maximum allowable fluctuation amount as MP; if the signal has a selective signal, the maximum allowable fluctuation amount is MP, MX 1、MX2...MXn;
S3: setting the length t of a stable interval section;
S4: starting from the moment 0, continuously traversing the data of the whole test duration by using a time window with the length of t and taking the sampling time as a step length, and obtaining the difference value between the maximum value and the minimum value of each parameter in each time window every time; if the signal has only active power P, the difference between the maximum value and the minimum value is delta P; if the signal has a selective signal, the difference between the maximum value and the minimum value is delta P and delta X 1、ΔX2...ΔXn respectively;
If the signal has only active power P, when the signal meets delta P < M P, the time interval is considered to be a stable operation interval meeting the requirement, and all the time intervals meeting the requirement are sequentially numbered as t 1,t2...tm; m is a natural positive integer;
If the signal has a selective signal, when the delta P is simultaneously satisfied with M P,ΔX1<MX1,ΔX2<MX2...ΔXn<MXn, the time interval is considered to be a stable operation interval which satisfies the requirement, and all the time intervals which satisfy the requirement are sequentially numbered as t 1,t2...tm; m is a natural positive integer;
S5: calculating the stability index s of each stable interval section obtained in the step 4;
S6: classifying the m stable interval sections obtained in the step 4 according to the active power values, wherein the classification principle is that the values similar to the active power are classified into the same class, and finally k classes are obtained;
S7: in k classes, a time interval section with the minimum s value in each class is found out and used as a finally screened stable operation interval section, so that k time interval sections with the best stability index meeting the condition can be obtained and used as finally output sorting results T1 and T2 … … Tk;
the calculation method in step S5 is as follows:
S5.1: the maximum value of each parameter over the entire test duration is read,
If the signal has only active power P, the maximum value is PMax;
if the signal has an optional signal, the values of the signals are PMax and X 1Max、X2Max...Xn Max in sequence;
S5.2: calculating normalized indexes of fluctuation values of parameters in each time interval meeting the conditions:
sP=ΔP/PMax,sX1=ΔX1/X1Max,sX2=ΔX2/X2Max...sXn=ΔXn/XnMax;
s5.3: the stability index s=sp+sx 1+sX2+...sXn of each stable section obtained in step 4 was obtained.
2. The intelligent sorting method for the variable load test data of the hydroelectric generating set according to claim 1, which is characterized by comprising the following steps of: in step S6, the average value PAVG of the active power in each stable interval obtained in step 4 is calculated, and the m stable intervals obtained in step 4 are classified according to the PAVG value by using a DBSCAN density clustering algorithm, so as to obtain k classes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210546816.9A CN114861808B (en) | 2022-05-20 | 2022-05-20 | Intelligent sorting method for variable load test data of hydroelectric generating set |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210546816.9A CN114861808B (en) | 2022-05-20 | 2022-05-20 | Intelligent sorting method for variable load test data of hydroelectric generating set |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114861808A CN114861808A (en) | 2022-08-05 |
CN114861808B true CN114861808B (en) | 2024-05-14 |
Family
ID=82639407
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210546816.9A Active CN114861808B (en) | 2022-05-20 | 2022-05-20 | Intelligent sorting method for variable load test data of hydroelectric generating set |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114861808B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012095478A (en) * | 2010-10-28 | 2012-05-17 | Kansai Electric Power Co Inc:The | Output estimation method and output estimation device for photovoltaic power generation |
CN106408008A (en) * | 2016-09-08 | 2017-02-15 | 国网江西省电力公司赣州供电分公司 | Load curve distance and shape-based load classification method |
CN112525567A (en) * | 2020-09-29 | 2021-03-19 | 华能澜沧江水电股份有限公司 | Method for detecting accumulated operation time of each stability region of hydraulic generator |
WO2021073462A1 (en) * | 2019-10-15 | 2021-04-22 | 国网浙江省电力有限公司台州供电公司 | 10 kv static load model parameter identification method based on similar daily load curves |
-
2022
- 2022-05-20 CN CN202210546816.9A patent/CN114861808B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012095478A (en) * | 2010-10-28 | 2012-05-17 | Kansai Electric Power Co Inc:The | Output estimation method and output estimation device for photovoltaic power generation |
CN106408008A (en) * | 2016-09-08 | 2017-02-15 | 国网江西省电力公司赣州供电分公司 | Load curve distance and shape-based load classification method |
WO2021073462A1 (en) * | 2019-10-15 | 2021-04-22 | 国网浙江省电力有限公司台州供电公司 | 10 kv static load model parameter identification method based on similar daily load curves |
CN112525567A (en) * | 2020-09-29 | 2021-03-19 | 华能澜沧江水电股份有限公司 | Method for detecting accumulated operation time of each stability region of hydraulic generator |
Non-Patent Citations (1)
Title |
---|
基于仿射传播聚类算法的广义负荷稳态特性建模及其应用;褚壮壮;梁军;张旭;董晓明;张永亮;;电力自动化设备;20160331;第36卷(第03期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN114861808A (en) | 2022-08-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110837866A (en) | XGboost-based electric power secondary equipment defect degree evaluation method | |
CN107038167A (en) | Big data excavating analysis system and its analysis method based on model evaluation | |
CN101226133B (en) | Method for specification and recognition of hemocyte pulse signal | |
CN111680820A (en) | Distributed photovoltaic power station fault diagnosis method and device | |
CN112749840B (en) | Method for acquiring energy efficiency characteristic index reference value of thermal power generating unit | |
CN110969306B (en) | Power distribution low-voltage transformer area load prediction method and device based on deep learning | |
CN108280236A (en) | A kind of random forest visualization data analysing method based on LargeVis | |
CN115775045A (en) | Photovoltaic balance prediction method based on historical similar days and real-time multi-dimensional study and judgment | |
CN110738232A (en) | grid voltage out-of-limit cause diagnosis method based on data mining technology | |
CN111210170A (en) | Environment-friendly management and control monitoring and evaluation method based on 90% electricity distribution characteristic index | |
CN110265996A (en) | It is a kind of suitable for photovoltaic/wind power prediction time characteristic scale modeling method | |
CN112836920A (en) | Coal electric unit energy efficiency state evaluation method and device and coal electric unit system | |
CN116796403A (en) | Building energy saving method based on comprehensive energy consumption prediction of commercial building | |
CN114595762A (en) | Photovoltaic power station abnormal data sequence extraction method | |
CN114861808B (en) | Intelligent sorting method for variable load test data of hydroelectric generating set | |
CN110991689A (en) | Distributed photovoltaic power generation system short-term prediction method based on LSTM-Morlet model | |
CN117559443A (en) | Ordered power utilization control method for large industrial user cluster under peak load | |
CN115983507B (en) | Method and system for predicting broadband oscillation risk of section of power grid of transmitting end source | |
CN113298148B (en) | Ecological environment evaluation-oriented unbalanced data resampling method | |
CN111985563B (en) | Industrial consumer power equipment classification method based on harmonic emission level | |
CN112230056B (en) | Multi-harmonic-source contribution calculation method based on OFMMK-Means clustering and composite quantile regression | |
CN114818827A (en) | Non-invasive load decomposition method based on seq2point network | |
CN114529071A (en) | Method for predicting power consumption of transformer area | |
CN110991743B (en) | Wind power short-term combination prediction method based on cluster analysis and neural network optimization | |
CN114943304B (en) | Bayesian-based 3C electronic product assembly error transfer method |
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 |