CN113568386A - Thermal power generating unit all-working-condition big data analysis method based on interval estimation - Google Patents

Thermal power generating unit all-working-condition big data analysis method based on interval estimation Download PDF

Info

Publication number
CN113568386A
CN113568386A CN202110868393.8A CN202110868393A CN113568386A CN 113568386 A CN113568386 A CN 113568386A CN 202110868393 A CN202110868393 A CN 202110868393A CN 113568386 A CN113568386 A CN 113568386A
Authority
CN
China
Prior art keywords
thermal power
unit
time window
power generating
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110868393.8A
Other languages
Chinese (zh)
Other versions
CN113568386B (en
Inventor
赵章明
高林
王林
李军
高海东
肖勇
李海滨
李华
周俊波
王明坤
侯玉婷
郭亦文
王文毓
陆晨旭
金国强
昌鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian Thermal Power Research Institute Co Ltd
Huaneng Qinmei Ruijin Power Generation Co Ltd
Original Assignee
Xian Thermal Power Research Institute Co Ltd
Huaneng Qinmei Ruijin Power Generation Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xian Thermal Power Research Institute Co Ltd, Huaneng Qinmei Ruijin Power Generation Co Ltd filed Critical Xian Thermal Power Research Institute Co Ltd
Priority to CN202110868393.8A priority Critical patent/CN113568386B/en
Publication of CN113568386A publication Critical patent/CN113568386A/en
Application granted granted Critical
Publication of CN113568386B publication Critical patent/CN113568386B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41885—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32339—Object oriented modeling, design, analysis, implementation, simulation language
    • 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
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Investigating Or Analyzing Materials Using Thermal Means (AREA)

Abstract

The invention discloses a thermal power unit all-working-condition big data analysis method based on interval estimation, which is characterized in that the load of a thermal power unit is taken as a characteristic variable for steady-state judgment, in a certain time window, a confidence interval of a load difference value sample overall mean value of the thermal power unit contains a zero value under a significance level alpha, and the absolute value of a difference value between a confidence lower limit and a confidence upper limit is stable under a limited condition, the thermal power unit is judged to be in a steady-state working condition in the time window, otherwise, the thermal power unit is judged to be in an unsteady-state working condition; when the steady-state and unsteady-state working condition of the thermal power generating unit is analyzed, the running state of the thermal power generating unit can be accurately judged under the condition that the steady-state and unsteady-state working conditions frequently and alternately appear.

Description

Thermal power generating unit all-working-condition big data analysis method based on interval estimation
Technical Field
The invention relates to the field of intelligent data processing of thermal power generating units, in particular to a thermal power generating unit all-working-condition big data analysis method based on interval estimation.
Background
With the advance of the digital transformation of thermal power generating units, intelligent process analysis and control technology is applied to more and more thermal power generating units, and in terms of long-term development trend, the advanced digital technology is deeply integrated with the power generation production control process, so that the intelligent process analysis and control technology becomes an important driving force for power generation enterprises to practice intelligent transformation. On one hand, in the field of energy conservation and emission reduction in the power production process, technologies such as big data and artificial intelligence play more and more important roles, and deep modeling mining is performed on the generation process and a thermal system by using the technologies such as the big data and the artificial intelligence, so that the method is an important means for breaking through the bottleneck of energy conservation and emission reduction of a unit. On the other hand, the thermal power generating set is gradually changed from an electric quantity type power supply to an adjusting type power supply, the set is switched between a steady-state operation mode and an unsteady-state operation mode for a long time, the physical characteristics of the set under steady-state and unsteady-state working conditions are greatly different, the operation states of the set are greatly different, and correspondingly, the problems that the fluctuation range is large, irregular deviation, the occurrence frequency of partial characteristics is few, the data distribution difference is large and the like are shown on operation data are solved, so that the instability of a modeling result is increased for most of statistics-based big data and artificial intelligence algorithms. Therefore, steady-state and non-steady-state working conditions are distinguished, the steady-state and non-steady-state working conditions are respectively modeled by utilizing big data and an artificial intelligence algorithm and are organically and uniformly fused, and the safety and the stability of the power production process are improved.
At present, working condition division methods of thermal power generating units mainly comprise the following two methods:
1) and (3) according to the load change range of the thermal power generating unit, adopting an equal-width dividing method to divide the working condition of the thermal power generating unit into a plurality of kinds in an equal-width manner. Although the method can cover all the operating conditions of the thermal power generating unit, the actual distribution characteristics of the operating conditions of the thermal power generating unit cannot be completely reflected. In addition, when the non-steady-state working condition and the steady-state working condition alternately appear in the historical operation data of the unit, the operation mode cannot be effectively distinguished from the data, and the stability and the accuracy of a modeling result are greatly influenced for big data and an artificial intelligent modeling algorithm.
2) The working conditions of the thermal power generating units are divided through historical data mining, an unsupervised machine learning clustering algorithm is adopted, and historical operating data sets of the thermal power generating units are divided through similarity among samples, so that the working conditions of the thermal power generating units are divided. Although the method overcomes the limitation of the equal-width division method, the clustering algorithm cannot effectively determine the specific type of the working condition, for example, the number of clustering clusters of the K-means clustering algorithm needs to be set manually, more prior knowledge is needed, the requirement on the related professional background of modeling personnel is higher, and most of large data and professional personnel of the artificial intelligent algorithm do not have the professional background of thermal power.
In summary, the existing thermal power generating unit working condition division method cannot effectively divide steady-state and unsteady-state working conditions, and in order to solve the problem, the invention provides a universal steady-state and unsteady-state working condition analysis method, which effectively makes up for the defects of other methods at present, realizes effective identification and differentiation of data in different operation modes, and provides a good data processing method for accurate modeling of a system.
Disclosure of Invention
The invention provides a thermal power generating unit all-condition big data analysis method based on interval estimation.
Further explained below, the present invention adopts the following technical scheme:
a thermal power unit all-working-condition big data analysis method based on interval estimation is characterized in that thermal power unit load is used as a characteristic variable for steady-state judgment, in a certain time window, a confidence interval of a load difference value sample overall mean value of a thermal power unit contains a zero value under a significance level alpha, and the absolute value of a difference value between a confidence lower limit and a confidence upper limit is stable under a limited condition, the thermal power unit is judged to be in a steady-state working condition in the time window, otherwise, the thermal power unit is judged to be in an unsteady-state working condition; the method specifically comprises the following steps:
step 1: unit load sequence data { X) is obtained from DCS (distributed control system) of thermal power generating unit0,X1,X2,…,Xn}; calculating to obtain unit load difference value sequence data { d by adopting a first-order difference method1,d2,d3,…,dn}; the first order difference method is used for calculating the unit load difference sequence data, and the specific calculation formula is as follows:
dj＝Xj-Xj-1j is 1,2,3, …, n is 1
In formula 1, XjUnit load representing time j, djRepresenting the unit load difference value of the moment j and the moment j-1;
step 2: when it is assumed thatThe size of the window between
Figure BDA0003188099870000031
Presentation pair
Figure BDA0003188099870000032
Rounding, and for j (n-k +1) with k being more than or equal to j, generating unit load difference sequence data { d1,d2,d3,…,dnMoving the time window from left to right by taking the step length as 1;
and step 3: for each time window, adopting an interval estimation method, and under a given significance level alpha, estimating a confidence interval of the unit load difference value sample overall mean value m in the time window
Figure BDA0003188099870000033
And
Figure BDA0003188099870000034
satisfy the requirement of
Figure BDA0003188099870000035
At a given significance level α, the mean of the unit load difference samples within the time window is first calculated:
Figure BDA0003188099870000041
in the formula 2, the first step is,
Figure BDA0003188099870000042
mean value, d, representing unit load difference samples within a time windowjRepresenting the unit load difference value of the moment j and the moment j-1;
calculating the standard deviation of the unit load difference value in the time window:
Figure BDA0003188099870000043
calculating the confidence upper limit and the confidence lower limit of the confidence interval of the unit load difference value sample overall mean value m in the time window:
Figure BDA0003188099870000044
Figure BDA0003188099870000045
in the formulae 4 and 5, the compound represented by the formula,
Figure BDA0003188099870000046
representing a distribution of t (n-1)
Figure BDA0003188099870000047
Quantile division;
and 4, step 4: judging the working condition of the unit: when j is k or j is n-k +1, if
Figure BDA0003188099870000048
And is
Figure BDA0003188099870000049
Judging that the unit is in a stable working condition in the time window, otherwise, judging that the unit is in an unstable working condition in the time window; when k is<j<(n-k +1), if
Figure BDA00031880998700000410
And is
Figure BDA00031880998700000411
And judging that the unit is in a stable working condition at the moment j, otherwise, judging that the unit is in an unstable working condition at the moment j.
Figure BDA00031880998700000412
The smaller the sum significance level α, the smaller the uncertainty of the determination result, and otherwise, the larger the uncertainty of the determination result.
The significance level a is 0.05 or 0.01.
Compared with the prior art, the invention has the following characteristics:
when the steady-state and unsteady-state working condition of the thermal power generating unit is analyzed, the running state of the thermal power generating unit can be accurately judged under the condition that the steady-state and unsteady-state working conditions frequently and alternately appear. The method not only can judge the operating condition of the thermal power generating unit, but also can quantitatively analyze the uncertainty of the judgment result by using the confidence coefficient and the confidence interval of the interval estimation. In addition, the method does not need a specific professional background, and can be used for performing discriminant analysis on the steady-state and unsteady-state working conditions of the thermal power generating unit only by setting the size and the significance level of the sliding window. The method has strong practical significance for the application of artificial intelligence and big data algorithm in power generation enterprises.
Drawings
FIG. 1 is a flow chart of the present invention.
Fig. 2 is a result of analyzing a steady-state and non-steady-state condition of a 1000MW unit in a certain time period according to an embodiment of the present invention.
Detailed Description
The present invention is described in further detail below with reference to specific embodiments.
Example (b):
as shown in fig. 1, the specific steps of distinguishing the unit operating conditions of a 1000MW unit of a power plant by a large data analysis method of steady-state and unsteady-state operating conditions of a thermal power unit based on interval estimation are as follows:
step 1, acquiring unit load sequence data { X) from the unit DCS system0,X1,X2,…,Xn}; calculating to obtain unit load difference value sequence data { d by adopting a first-order difference method1,d2,d3,…,dn}; the first order difference method is used for calculating the unit load difference sequence data, and the specific calculation formula is as follows:
dj＝Xj-Xj-1j is 1,2,3, …, n is 1
In formula 1, XjUnit load representing time j, djRepresenting the unit load difference between time j and time j-1.
Step 2, assuming a time window of size
Figure BDA0003188099870000061
Presentation pair
Figure BDA0003188099870000062
And (6) taking the whole. For k is less than or equal to j less than or equal to (n-k +1), the unit load difference sequence data { d1,d2,d3,…,dnOn, move the time window from left to right with step size 1.
And 3, for each time window, estimating a confidence interval of the unit load difference value sample overall mean value m in the time window by adopting an interval estimation method under a given significance level alpha
Figure BDA0003188099870000063
And
Figure BDA0003188099870000064
satisfy the requirement of
Figure BDA0003188099870000065
Step 3.1, firstly, calculating the average value of the unit load difference value samples in the time window:
Figure BDA0003188099870000066
in the formula 2, the first step is,
Figure BDA0003188099870000067
mean value, d, representing unit load difference samples within a time windowjRepresenting the unit load difference between time j and time j-1.
Step 3.2, calculating the standard deviation of the unit load difference value in the time window:
Figure BDA0003188099870000068
step 3.3, calculating a confidence upper limit and a confidence lower limit of a confidence interval of the unit load difference value sample overall mean value m in the time window:
Figure BDA0003188099870000069
Figure BDA00031880998700000610
in the formulae 4 and 5, the compound represented by the formula,
Figure BDA00031880998700000611
representing a distribution of t (n-1)
Figure BDA00031880998700000612
And (4) quantile number.
Step 4, when j is k or j is n-k +1, if
Figure BDA00031880998700000613
And is
Figure BDA00031880998700000614
Judging that the unit is in a stable working condition in the time window, otherwise, judging that the unit is in an unstable working condition in the time window; when k is<j<(n-k +1), if
Figure BDA0003188099870000071
And is
Figure BDA0003188099870000072
Judging that the unit is in a stable working condition at the moment j, otherwise, judging that the unit is in an unstable working condition at the moment j;
in this embodiment, the time window size k is 20, the significance level α is 0.05, and the confidence is
Figure BDA0003188099870000073
As shown in FIG. 2, the operation condition of a 1000MW unit of a power plant within a period of time is judged, and experimental results show that the method can effectively judge whether the unit is in a steady state or an unsteady state.
When the steady-state unsteady-state working condition of the thermal power generating unit is analyzed, the operation working condition of the thermal power generating unit can be judged, and the uncertainty of the judgment result can be quantitatively analyzed by using the confidence coefficient and the confidence interval of the interval estimation, so that the method has strong practical significance for the application of artificial intelligence and big data algorithm in power generation enterprises.

Claims (2)

1. A thermal power generating unit all-working-condition big data analysis method based on interval estimation is characterized by comprising the following steps: taking the load of the thermal power generating unit as a characteristic variable for steady-state judgment, and in a certain time window, if a confidence interval of the load difference sample population mean value of the thermal power generating unit contains a zero value under a significance level alpha, and the absolute value of the difference value between a confidence lower limit and a confidence upper limit is stable under a limited condition, judging that the thermal power generating unit is in a steady-state working condition in the time window, otherwise, judging that the thermal power generating unit is in an unsteady-state working condition; the method specifically comprises the following steps:
step 1: unit load sequence data { X) is obtained from DCS (distributed control system) of thermal power generating unit0，X1，X2，...，Xn}; calculating to obtain unit load difference value sequence data { d by adopting a first-order difference method1，d2，d3，...，dn}; the first order difference method is used for calculating the unit load difference sequence data, and the specific calculation formula is as follows:
dj＝Xj-Xj-11,2,3, n formula 1
In formula 1, XjUnit load representing time j, djRepresenting the unit load difference value of the moment j and the moment j-1;
step 2: assuming that the time window size is k,
Figure FDA0003188099860000011
Figure FDA0003188099860000012
presentation pair
Figure FDA0003188099860000013
Rounding, and for j (n-k +1) with k being more than or equal to j, generating unit load difference sequence data { d1，d2，d3，...，dnMoving the time window from left to right by taking the step length as 1;
and step 3: for each time window, adopting an interval estimation method, and under a given significance level alpha, estimating a confidence interval of the unit load difference value sample overall mean value m in the time window
Figure FDA0003188099860000014
Figure FDA0003188099860000015
And
Figure FDA0003188099860000016
satisfy the requirement of
Figure FDA0003188099860000017
At a given significance level α, the mean of the unit load difference samples within the time window is first calculated:
Figure FDA0003188099860000021
in the formula 2, the first step is,
Figure FDA0003188099860000022
mean value, d, representing unit load difference samples within a time windowjRepresenting the unit load difference value of the moment j and the moment j-1;
calculating the standard deviation of the unit load difference value in the time window:
Figure FDA0003188099860000023
calculating the confidence upper limit and the confidence lower limit of the confidence interval of the unit load difference value sample overall mean value m in the time window:
Figure FDA0003188099860000024
Figure FDA0003188099860000025
in the formulae 4 and 5, the compound represented by the formula,
Figure FDA0003188099860000026
representing a distribution of t (n-1)
Figure FDA0003188099860000027
Quantile division;
and 4, step 4: judging the working condition of the unit: when j is k or j is n-k +1, if
Figure FDA0003188099860000028
And is
Figure FDA0003188099860000029
Judging that the unit is in a stable working condition in the time window, otherwise, judging that the unit is in an unstable working condition in the time window; when k < j < (n-k +1), if
Figure FDA00031880998600000210
And is
Figure FDA00031880998600000211
And judging that the unit is in a stable working condition at the moment j, otherwise, judging that the unit is in an unstable working condition at the moment j.
2. The thermal power generating unit all-operating-condition big data analysis method based on interval estimation is characterized in that: the significance level a is 0.05 or 0.01.
CN202110868393.8A 2021-07-30 2021-07-30 Thermal power generating unit all-working-condition big data analysis method based on interval estimation Active CN113568386B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110868393.8A CN113568386B (en) 2021-07-30 2021-07-30 Thermal power generating unit all-working-condition big data analysis method based on interval estimation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110868393.8A CN113568386B (en) 2021-07-30 2021-07-30 Thermal power generating unit all-working-condition big data analysis method based on interval estimation

Publications (2)

Publication Number Publication Date
CN113568386A true CN113568386A (en) 2021-10-29
CN113568386B CN113568386B (en) 2023-02-28

Family

ID=78169436

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110868393.8A Active CN113568386B (en) 2021-07-30 2021-07-30 Thermal power generating unit all-working-condition big data analysis method based on interval estimation

Country Status (1)

Country Link
CN (1) CN113568386B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070153958A1 (en) * 2005-12-30 2007-07-05 Russell William E Ii Method of determining margins to operating limits for nuclear reactor operation
CN101645599A (en) * 2009-08-25 2010-02-10 广东电网公司电力科学研究院 Pretreatment unit capable of automatically generating power to control target loads
CN104750973A (en) * 2015-02-28 2015-07-01 河北省电力建设调整试验所 Thermal power generating unit load (quasi) steady-state working condition clustering algorithm based on data smoothness functions
CN105469325A (en) * 2015-12-21 2016-04-06 云南电网有限责任公司电力科学研究院 Method and system for determining load stability state of thermal power generating unit
CN106529161A (en) * 2016-10-28 2017-03-22 东南大学 Method for determining ascending and descending load speed on basis of thermal power unit operation data

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070153958A1 (en) * 2005-12-30 2007-07-05 Russell William E Ii Method of determining margins to operating limits for nuclear reactor operation
CN101645599A (en) * 2009-08-25 2010-02-10 广东电网公司电力科学研究院 Pretreatment unit capable of automatically generating power to control target loads
CN104750973A (en) * 2015-02-28 2015-07-01 河北省电力建设调整试验所 Thermal power generating unit load (quasi) steady-state working condition clustering algorithm based on data smoothness functions
CN105469325A (en) * 2015-12-21 2016-04-06 云南电网有限责任公司电力科学研究院 Method and system for determining load stability state of thermal power generating unit
CN106529161A (en) * 2016-10-28 2017-03-22 东南大学 Method for determining ascending and descending load speed on basis of thermal power unit operation data

Also Published As

Publication number Publication date
CN113568386B (en) 2023-02-28

Similar Documents

Publication Publication Date Title
CN107682319A (en) A kind of method of data flow anomaly detection and multiple-authentication based on enhanced angle Outlier factor
CN111638707B (en) Intermittent process fault monitoring method based on SOM clustering and MPCA
CN112749840B (en) Method for acquiring energy efficiency characteristic index reference value of thermal power generating unit
CN110348488B (en) Modal identification method based on local density peak clustering
CN106067034B (en) Power distribution network load curve clustering method based on high-dimensional matrix characteristic root
CN110083896B (en) Standard reaching conformity degree evaluation method for wind turbine generator operating power curve
CN109933040B (en) Fault monitoring method based on hierarchical density peak clustering and most similar mode
CN112149714A (en) Method for determining energy efficiency characteristic index reference value of coal-electric unit based on data mining
CN112364560A (en) Intelligent prediction method for working hours of mine rock drilling equipment
CN111476303A (en) Line loss analysis method of fuzzy C-means clustering based on quantum optimization particle swarm
CN110348540B (en) Clustering-based method and device for screening transient power angle stability faults of power system
CN106547899B (en) Intermittent process time interval division method based on multi-scale time-varying clustering center change
CN111950840A (en) Intelligent operation and maintenance knowledge retrieval method and system for metrological verification device
CN110020680B (en) PMU data classification method based on random matrix theory and fuzzy C-means clustering algorithm
CN113568386B (en) Thermal power generating unit all-working-condition big data analysis method based on interval estimation
CN108537249B (en) Industrial process data clustering method for density peak clustering
CN107844679B (en) Sectional type blast furnace molten iron silicon content prediction method and device
CN112765746A (en) Turbine blade top gas-thermal performance uncertainty quantification system based on polynomial chaos
CN116821828A (en) Multi-dimensional time sequence prediction method based on industrial data
CN112508320B (en) Automatic process stage division workflow for batch production
Shah et al. Fuzzy based selection of PWARX model for the nonlinear hybrid dynamical systems
CN116757337B (en) House construction progress prediction system based on artificial intelligence
CN113033683B (en) Industrial system working condition monitoring method and system based on static and dynamic joint analysis
Niu et al. Modeling of penicillin fermentation process based on FCM and improved Just-in-Time learning algorithm
CN109193624B (en) Load classification method based on improved K-means clustering

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