CN104153981A - Method for estimating backpressure data during starting and stopping of power station circulating water pump - Google Patents
Method for estimating backpressure data during starting and stopping of power station circulating water pump Download PDFInfo
- Publication number
- CN104153981A CN104153981A CN201410056562.8A CN201410056562A CN104153981A CN 104153981 A CN104153981 A CN 104153981A CN 201410056562 A CN201410056562 A CN 201410056562A CN 104153981 A CN104153981 A CN 104153981A
- Authority
- CN
- China
- Prior art keywords
- sample
- circulating water
- data
- water pump
- sample set
- 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
Links
Landscapes
- Control Of Positive-Displacement Pumps (AREA)
Abstract
The invention discloses a method for estimating backpressure data during starting and stopping of a power station circulating water pump. The method comprises the steps that firstly, a prediction data sample set is obtained by collecting original data of corresponding units and screening original data samples; then, running data are collected, and a new data sample set is obtained by updating the original data samples or adding new samples into the original data samples; finally, training and modeling are carried out by inputting training samples into a least square support vector machine, and then the best parameter value of a model is found through the tunelssvm optimization function. According to the method for estimating the backpressure data, information is mined in a data base, effective information is screened out, a sample database is updated through self-learning, and the relevance among the data is mined through the support vector machine, so that prediction of backpressure changes during starting and stopping of the circulating water pump is achieved, and the sample database is continuously updated and corrected along with time.
Description
Technical field
The present invention relates to a kind of predictor method of backpressure data during for power station circulating water pump start and stop.
Background technique
The size of condenser vacuum has duality to the impact of unit operation.Condenser vacuum is too high, will increase the end thrust of steam turbine, causes thrust bearing shoe valve coal temperature to raise and returning-oil temperature rising, affects the safe operation of steam turbine; Condenser vacuum is too low, and discharge cylinder of steam turbine temperature raises, and hear rate increases, decrease in efficiency, and the skew of steam turbine bearing centre, also can cause steam turbine vibration when serious.Therefore, can accurately dope circulating water pump start and stop time, the variation of condenser vacuum seems extremely important.Consider in actual at the scene running environment, the factor that affects condenser vacuum is a lot, such as clean degree and the circulating water flow etc. of circulating water intake water temperature, turbine discharge flow, vapour condenser, simultaneously the pump that follows of most of power plant can not continuously adjusting flow rate, circulating water flow at present also neither one comparatively direct method can record, often do not there is the accuracy that can be applied in engineering reality by the value simply calculating.On-the-spot DCS mass historical data is accompanied by the running of unit, constantly record related data, this wherein includes and meets in a large number on-the-spot actual information, on certain theory analysis basis, can from a large amount of data, be hidden in special relationship wherein by automatic search by data mining.Support vector machine is as a kind of forecasting tool, that the VC that is based upon Statistical Learning Theory ties up on theoretical and structure risk minimum principle basis, it shows many distinctive advantages solving in small sample, non-linear and higher-dimension pattern recognition, and can promote the use of in the other machines problems concerning study such as Function Fitting.The requirement of this sample database characteristic conforms support vector machine, use it to learn this sample database and when following pump start and stop future economy predict to there is certain accuracy.
Summary of the invention
Goal of the invention: goal of the invention of the present invention is to disclose a kind of predictor method of backpressure data during for power station circulating water pump start and stop for the deficiencies in the prior art, by mined information in database and filter out effective information, upgrade sample database and excavate the relevance between data by support vector machine by self-teaching again, thus economy prediction and along with the time is constantly updated and revises sample database while reaching circulating water pump start and stop.
Technological scheme: in order to realize the goal of the invention of invention, the invention discloses a kind of predictor method of backpressure data during for power station circulating water pump start and stop, comprise the following steps:
(1) the each circulating water pump electric current measuring point to corresponding unit DCS, auxiliary mercury electric current measuring point, vacuum pump electric current measuring point, condenser vacuum measuring point, power of the assembling unit measuring point, vapour condenser inlet water temperature service data are carried out DM data sampling, and sample set is
D
0={d
1,d
2,d
3,…,d
L};
Wherein, d
i(1≤i≤L) is the data sample of different time points,
L is number of samples, and N is total number of unit circulating water pump and vacuum pump,
for circulating water pump electric current,
for vacuum pump electric current, t
ifor circulating water inlet water temperature, Pc
ifor back pressure, N
ifor the power of the assembling unit, T
ifor time mark;
(2) to sample set D
0carry out DM pretreatment, contrast the difference of all adjacent circulation water pump electric currents in each sample, if
There is start stop operation depending on circulating water pump;
Wherein,
be j circulating water pump at the current sampling data in i moment, I
1it is the start and stop setting value of circulating water pump;
(3) to sample set D
0the neighbouring sample time difference process, if
|T
i-T
i+1|>T
0;(1≤i≤L-1);
Depending on the unstable false start and stop that cause of these sample data and reject this group sample, obtain sample set D
1, its number of samples is L
1(0≤L
1≤ L);
Wherein T
iit is the specimen sample moment in i moment;
(4) to sample set D
1carry out DM pretreatment, the difference of all adjacent vacuum pump electric currents in contrast sample, if
Be considered as vacuum pump and cause back pressure to change and reject this sample, obtain sample set D
2, its number of samples is L
2(0≤L
2≤ L
1);
Wherein,
be j vacuum pump at the current sampling data in i moment, I
2the setting value that causes back pressure to change for vacuum pump;
(5) to sample set D
2middle condenser vacuum measuring point carries out curve mapping analysis and goes out the stationary value of this mutation process front back pressure and rear back pressure of circulating water pump start and stop, is recorded as prediction data sample set and is
D={d
1,d
2,d
3,…,d
M}(0≤M≤L
2);
(6) start stop operation that while detecting corresponding unit operation, circulating water pump occurs, obtain new data sample by the logical relation processing in step (2)~(5), new data sample and former data sample are carried out to interrelated logic judgement, and its interrelated logic is: establish
d′=(T′,i′
1,i′
2,…,i′
n,i′
n+1,…,i′
N,Pc′,N′,t′)
For new samples, wherein, N is total number of unit circulating water pump and vacuum pump, i '
1~i '
nfor the circulating water pump electric current of new samples, i '
n+1~i '
nfor the vacuum pump electric current of new samples, the circulating water inlet water temperature that t ' is new samples, the back pressure that Pc ' is new samples, the power of the assembling unit that N ' is new samples, the time mark that T ' is new samples; If there is this sample in prediction data sample set D in the integer range at t ' and N ' place, reject former sample and upgrade sample set, otherwise new samples is added in sample set;
(7) the sample set input least square method supporting vector machine obtaining in step (6) is trained to modeling, seek out the optimum parameter value of model by tunelssvm majorized function.
As preferably, in order to ensure the accuracy of primary data sample, ensure the efficiency of algorithm simultaneously, the sample collection time lag in the sample set in described step (1) is 30s.
As preferably, in order further to ensure efficiency and the accuracy thereof of algorithm, in described step (4), contrast is in the 20min of sample vacuum pump electric current front and back.
As preferably, in order further to ensure efficiency and the accuracy thereof of algorithm, what in described step (5), process is the condenser vacuum measuring point in 20min before and after each condenser vacuum measuring point.
Beneficial effect: the present invention compared with prior art: by mined information in database and filter out effective information, upgrade sample database and excavate the relevance between data by support vector machine by self-teaching again, thus economy prediction and along with the time is constantly updated and revises sample database while reaching circulating water pump start and stop.
Brief description of the drawings
Fig. 1 is the data screening flow chart of predictor method of the present invention;
Fig. 2 is data study and the training flow chart of predictor method of the present invention.
Embodiment
Below in conjunction with figure, the present invention is further described.
As shown in Figure 1, first each circulating water pump electric current measuring point of corresponding unit DCS, auxiliary mercury electric current measuring point, vacuum pump electric current measuring point, condenser vacuum measuring point, power of the assembling unit measuring point, vapour condenser inlet water temperature service data are carried out to DM data sampling; Then contrast the difference of all adjacent circulation water pump electric currents in each sample, if
(1≤j≤n, 1≤i≤L-1); There is start stop operation depending on circulating water pump, wherein,
be j circulating water pump at the current sampling data in i moment, I
1it is the start and stop setting value of circulating water pump; Then the neighbouring sample time difference of sample set is processed, if | T
i-T
i+1| > T
0; (1≤i≤L-1); Depending on the unstable false start and stop that cause of these sample data and reject this group sample, obtain sample set D
1, its number of samples is L
1(0≤L
1≤ L), wherein T
iit is the specimen sample moment in i moment; The difference of all adjacent vacuum pump electric currents in last contrast sample, if
(n+1≤j≤N, 1≤i≤L
1); Be considered as vacuum pump and cause back pressure to change and reject this sample, obtain sample set D
2, its number of samples is L
2(0≤L
2≤ L
1), wherein,
be j vacuum pump at the current sampling data in i moment, I
2the setting value that causes back pressure to change for vacuum pump; Complete the collecting work of primary data sample.
As shown in Figure 2, on the basis of primary data sample, the start stop operation that while detecting corresponding unit operation, circulating water pump occurs, obtain new data sample by the step identical with Fig. 1, new data sample and former data sample are carried out to interrelated logic judgement, even in the circulating water inlet water temperature of new samples and the integer range at power of the assembling unit place, in prediction data sample set, there is this sample, rejected former sample and upgrade sample set, otherwise new samples is added in sample set; Data sample is inputted to least square method supporting vector machine and train modeling.In high-dimensional feature space, construct optimum linearity decision function y (x)=sgn[w ψ (x)+b], adopt objective function to be
In formula, constraint conditio is
i=1,2 ..., n, w is weight factor, and C is penalty parameter, and b is deviate,
for mapping function.Objective function containing constraint conditio is converted to without constrained objective function by Lagrangian method, for
Wherein, y=[y
1..., y
n]
t, I
v=[1 ..., 1]
t, α=[α
1..., α
n]
t,
k (.) is kernel function, and this kernel function can be selected radial basis kernel function.
Be further described below in conjunction with embodiment.
Gather the sample of corresponding unit DCS, number of samples is 6, and its collection value is as shown in table 1.
Table 1
Obtain data by the steadiness parameter in stability analysis start and stop moment as shown in table 2.
Table 2
? | N | t | Pc 1 | Pc 2 | Δ |
1 | 308.260 | 26.268 | -92.896 | -93.303 | 0.434 |
2 | 200.417 | 26.506 | -93.8283 | -94.139 | 0.311 |
3 | 291.010 | 31.094 | -91.5896 | -92.030 | 0.441 |
4 | 189.986 | 27.243 | -93.7339 | -94.091 | 0.357 |
5 | 203.045 | 32.809 | -90.8112 | -91.111 | 0.300 |
6 | 171.827 | 29.535 | -93.4087 | -93.572 | 0.164 |
Utilize in data wherein 5 groups carry out model training as training parameter, another group obtains the Output rusults as table 3 as prediction.
Table 3
? | N | t | Actual Δ | Prediction Δ | Relative error |
1 | 308.260 | 26.268 | 0.434 | 0.4433 | 2.1% |
2 | 200.417 | 26.506 | 0.311 | 0.3192 | 2.8% |
3 | 291.010 | 31.094 | 0.441 | 0.4387 | 0.5% |
4 | 189.986 | 27.243 | 0.357 | 0.3358 | 5.9% |
5 | 203.045 | 32.809 | 0.300 | 0.3114 | 3.8% |
6 | 171.827 | 29.535 | 0.164 | 0.1674 | 2.1% |
Maximal phase mistake is poor is as can be seen from Table 3 5.9%, and this Forecasting Methodology can be predicted the operating mode of corresponding unit exactly.
Claims (4)
1. during for power station circulating water pump start and stop, a predictor method for backpressure data, is characterized in that, comprises the following steps:
(1) the each circulating water pump electric current measuring point to corresponding unit DCS, auxiliary mercury electric current measuring point, vacuum pump electric current measuring point, condenser vacuum measuring point, power of the assembling unit measuring point, vapour condenser inlet water temperature service data are carried out DM data sampling, and sample set is
D
0={d
1,d
2,d
3,…,d
L};
Wherein, d
i(1≤i≤L) is the data sample of different time points,
L is number of samples, and N is total number of unit circulating water pump and vacuum pump,
for circulating water pump electric current,
for vacuum pump electric current, t
ifor circulating water inlet water temperature, Pc
ifor back pressure, N
ifor the power of the assembling unit, T
ifor time mark;
(2) to sample set D
0carry out DM pretreatment, contrast the difference of all adjacent circulation water pump electric currents in each sample, if
There is start stop operation depending on circulating water pump;
Wherein,
be j circulating water pump at the current sampling data in i moment, I
1it is the start and stop setting value of circulating water pump;
(3) to sample set D
0the neighbouring sample time difference process, if
|T
i-T
i+1|>T
0;(1≤i≤L-1);
Depending on the unstable false start and stop that cause of these sample data and reject this group sample, obtain sample set D
1, its number of samples is L
1(0≤L
1≤ L);
Wherein T
iit is the specimen sample moment in i moment;
(4) to sample set D
1carry out DM pretreatment, the difference of all adjacent vacuum pump electric currents in contrast sample, if
Be considered as vacuum pump and cause back pressure to change and reject this sample, obtain sample set D
2, its number of samples is L
2(0≤L
2≤ L
1);
Wherein,
be j vacuum pump at the current sampling data in i moment, I
2the setting value that causes back pressure to change for vacuum pump;
(5) to sample set D
2middle condenser vacuum measuring point carries out curve mapping analysis and goes out the stationary value of this mutation process front back pressure and rear back pressure of circulating water pump start and stop, is recorded as prediction data sample set and is
D={d
1,d
2,d
3,…,d
M}(0≤M≤L
2);
(6) start stop operation that while detecting corresponding unit operation, circulating water pump occurs, obtain new data sample by the logical relation processing in step (2)~(5), new data sample and former data sample are carried out to interrelated logic judgement, and its interrelated logic is: establish
d′=(T′,i′
1,i′
2,…,i′
n,i′
n+1,…,i′
N,Pc′,N′,t′)
For new samples, wherein, N is total number of unit circulating water pump and vacuum pump, i '
1~i '
nfor the circulating water pump electric current of new samples, i '
n+1~i '
nfor the vacuum pump electric current of new samples, the circulating water inlet water temperature that t ' is new samples, the back pressure that Pc ' is new samples, the power of the assembling unit that N ' is new samples, the time mark that T ' is new samples; If there is this sample in prediction data sample set D in the integer range at t ' and N ' place, reject former sample and upgrade sample set, otherwise new samples is added in sample set;
(7) the sample set input least square method supporting vector machine obtaining in step (6) is trained to modeling, seek out the optimum parameter value of model by tunelssvm majorized function.
2. a kind of predictor method of backpressure data during for power station circulating water pump start and stop as claimed in claim 1, is characterized in that, the sample collection time lag in the sample set in step (1) is 30s.
3. a kind of predictor method of backpressure data during for power station circulating water pump start and stop as claimed in claim 1, is characterized in that, in step (4), contrast be in the 20min of sample vacuum pump electric current front and back.
4. a kind of predictor method of backpressure data during for power station circulating water pump start and stop as claimed in claim 1, is characterized in that, what in step (5), process be the condenser vacuum measuring point in each condenser vacuum measuring point front and back 20min.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410056562.8A CN104153981B (en) | 2014-02-19 | 2014-02-19 | Method for estimating backpressure data during starting and stopping of power station circulating water pump |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410056562.8A CN104153981B (en) | 2014-02-19 | 2014-02-19 | Method for estimating backpressure data during starting and stopping of power station circulating water pump |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104153981A true CN104153981A (en) | 2014-11-19 |
CN104153981B CN104153981B (en) | 2017-02-15 |
Family
ID=51879573
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410056562.8A Active CN104153981B (en) | 2014-02-19 | 2014-02-19 | Method for estimating backpressure data during starting and stopping of power station circulating water pump |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104153981B (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105569748A (en) * | 2014-10-31 | 2016-05-11 | 王砧 | Operation back pressure continuous optimized control method and system for wet cooling steam turbine |
CN106840363A (en) * | 2017-03-22 | 2017-06-13 | 南阳理工学院 | The defeated buried pipeline load identification of one kind length and safety monitoring system |
CN107524588A (en) * | 2017-07-19 | 2017-12-29 | 甘肃银光化学工业集团有限公司 | A kind of optimization method of pump apparatus operating condition |
CN107956731A (en) * | 2017-11-30 | 2018-04-24 | 中国大唐集团科学技术研究院有限公司西北分公司 | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter |
CN108802040A (en) * | 2017-05-04 | 2018-11-13 | 南京市特种设备安全监督检验研究院 | A kind of unmanned plane device and detection method for crane surface defects detection |
CN110966170A (en) * | 2019-11-27 | 2020-04-07 | 河北涿州京源热电有限责任公司 | Real-time control method for cold end back pressure of indirect air cooling generator set |
CN111058911A (en) * | 2019-11-27 | 2020-04-24 | 河北涿州京源热电有限责任公司 | Thermal generator set cold end back pressure real-time control method based on environment wet bulb temperature |
CN112132394A (en) * | 2020-08-21 | 2020-12-25 | 西安交通大学 | Power plant circulating water pump prediction state assessment method and system |
CN113392118A (en) * | 2021-06-04 | 2021-09-14 | 中电四川数据服务有限公司 | Data updating detection system and method based on machine learning |
CN113446519A (en) * | 2021-09-01 | 2021-09-28 | 蘑菇物联技术(深圳)有限公司 | Method, electronic device and storage medium for determining leakage degree of pipe network |
CN113671830A (en) * | 2021-08-10 | 2021-11-19 | 浙江浙能技术研究院有限公司 | Thermal power generating unit cold end optimization closed-loop control method based on intelligent scoring |
CN114576149B (en) * | 2022-01-28 | 2023-11-28 | 华电电力科学研究院有限公司 | Optimal operation control system and method for circulating water pump of thermal power plant |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030171876A1 (en) * | 2002-03-05 | 2003-09-11 | Victor Markowitz | System and method for managing gene expression data |
CN101907088A (en) * | 2010-05-27 | 2010-12-08 | 中国人民解放军国防科学技术大学 | Fault diagnosis method based on one-class support vector machines |
CN103400052A (en) * | 2013-08-22 | 2013-11-20 | 武汉大学 | Combined method for predicting short-term wind speed in wind power plant |
CN103498788A (en) * | 2013-09-22 | 2014-01-08 | 深圳市广前电力有限公司 | Heavy-type air compressor air flow and efficiency on-line measurement method and device |
-
2014
- 2014-02-19 CN CN201410056562.8A patent/CN104153981B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030171876A1 (en) * | 2002-03-05 | 2003-09-11 | Victor Markowitz | System and method for managing gene expression data |
CN101907088A (en) * | 2010-05-27 | 2010-12-08 | 中国人民解放军国防科学技术大学 | Fault diagnosis method based on one-class support vector machines |
CN103400052A (en) * | 2013-08-22 | 2013-11-20 | 武汉大学 | Combined method for predicting short-term wind speed in wind power plant |
CN103498788A (en) * | 2013-09-22 | 2014-01-08 | 深圳市广前电力有限公司 | Heavy-type air compressor air flow and efficiency on-line measurement method and device |
Non-Patent Citations (4)
Title |
---|
乔宗良等: "一种改进的CPSO-LSSVM软测量模型及其应用", 《仪器仪表学报》, vol. 35, no. 1, 15 January 2014 (2014-01-15), pages 234 - 240 * |
周卫庆等: "基于稳健估计的迭代型支持向量机及其在电站数据检验中的应用", 《中国电机工程学报》, vol. 31, no. 11, 15 April 2011 (2011-04-15), pages 113 - 118 * |
杨学存等: "基于参数优化支持向量机的煤泥输送管道压力预测", 《煤炭工程》, no. 12, 31 December 2013 (2013-12-31), pages 112 - 115 * |
王雷等: "基于支持向量机回归的凝汽器真空应达值确定方法的研究", 《汽轮机技术》, vol. 49, no. 1, 25 February 2007 (2007-02-25), pages 43 - 46 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105569748A (en) * | 2014-10-31 | 2016-05-11 | 王砧 | Operation back pressure continuous optimized control method and system for wet cooling steam turbine |
CN105569748B (en) * | 2014-10-31 | 2017-03-15 | 王砧 | A kind of wet cooling gas turbine operation back pressure Filled function control method and system |
CN106840363A (en) * | 2017-03-22 | 2017-06-13 | 南阳理工学院 | The defeated buried pipeline load identification of one kind length and safety monitoring system |
CN108802040A (en) * | 2017-05-04 | 2018-11-13 | 南京市特种设备安全监督检验研究院 | A kind of unmanned plane device and detection method for crane surface defects detection |
CN107524588A (en) * | 2017-07-19 | 2017-12-29 | 甘肃银光化学工业集团有限公司 | A kind of optimization method of pump apparatus operating condition |
CN107524588B (en) * | 2017-07-19 | 2019-04-26 | 甘肃银光化学工业集团有限公司 | A kind of optimization method of pump apparatus operating condition |
CN107956731A (en) * | 2017-11-30 | 2018-04-24 | 中国大唐集团科学技术研究院有限公司西北分公司 | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter |
CN111058911A (en) * | 2019-11-27 | 2020-04-24 | 河北涿州京源热电有限责任公司 | Thermal generator set cold end back pressure real-time control method based on environment wet bulb temperature |
CN110966170A (en) * | 2019-11-27 | 2020-04-07 | 河北涿州京源热电有限责任公司 | Real-time control method for cold end back pressure of indirect air cooling generator set |
CN112132394A (en) * | 2020-08-21 | 2020-12-25 | 西安交通大学 | Power plant circulating water pump prediction state assessment method and system |
CN112132394B (en) * | 2020-08-21 | 2024-03-29 | 西安交通大学 | Power plant circulating water pump predictive state evaluation method and system |
CN113392118A (en) * | 2021-06-04 | 2021-09-14 | 中电四川数据服务有限公司 | Data updating detection system and method based on machine learning |
CN113392118B (en) * | 2021-06-04 | 2022-10-18 | 中电四川数据服务有限公司 | Data updating detection system and method based on machine learning |
CN113671830A (en) * | 2021-08-10 | 2021-11-19 | 浙江浙能技术研究院有限公司 | Thermal power generating unit cold end optimization closed-loop control method based on intelligent scoring |
CN113671830B (en) * | 2021-08-10 | 2024-04-02 | 浙江浙能数字科技有限公司 | Cold end optimization closed-loop control method for thermal power generating unit based on intelligent scoring |
CN113446519A (en) * | 2021-09-01 | 2021-09-28 | 蘑菇物联技术(深圳)有限公司 | Method, electronic device and storage medium for determining leakage degree of pipe network |
CN114576149B (en) * | 2022-01-28 | 2023-11-28 | 华电电力科学研究院有限公司 | Optimal operation control system and method for circulating water pump of thermal power plant |
Also Published As
Publication number | Publication date |
---|---|
CN104153981B (en) | 2017-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104153981A (en) | Method for estimating backpressure data during starting and stopping of power station circulating water pump | |
AU2012279410B2 (en) | Automatic identification of operating parameters for power plants | |
CN103455635A (en) | Thermal process soft sensor modeling method based on least squares and support vector machine ensemble | |
CN103631681A (en) | Method for online restoring abnormal data of wind power plant | |
CN103488869A (en) | Wind power generation short-term load forecast method of least squares support vector machine | |
CN104572449A (en) | Automatic test method based on case library | |
CN103853106A (en) | Energy consumption prediction parameter optimization method of building energy supply device | |
CN102135021B (en) | Method for predicting shaft power of industrial extraction condensing steam turbine | |
CN105404280A (en) | Industrial process fault detection method based on autoregression dynamic hidden variable model | |
CN102842043B (en) | Particle swarm classifying method based on automatic clustering | |
CN102750286A (en) | Novel decision tree classifier method for processing missing data | |
CN111695666A (en) | Wind power ultra-short term conditional probability prediction method based on deep learning | |
CN103853939A (en) | Combined forecasting method for monthly load of power system based on social economic factor influence | |
CN110851577A (en) | Knowledge graph expansion method and device in electric power field | |
CN112628132B (en) | Water pump key index prediction method based on machine learning | |
CN115271154A (en) | Nonlinear regression flood element prediction method based on polynomial and partial least square coupling | |
Liu et al. | Optimizing layout of pumping well in irrigation district for groundwater sustainable use in northwest China | |
CN112183642A (en) | Method and system for detecting coal consumption of cement firing based on random forest model | |
CN115705452A (en) | Novel recovery ratio prediction method for middle and later stages of integrated sandstone reservoir development | |
CN107315845A (en) | Sulphur unit tail gas method up to standard | |
CN107180391B (en) | Wind power data span selection method and device | |
CN104899429A (en) | Multi-element-based geochemical element content value assigning method and system | |
Feder | Machine-learning approach determines spatial variation in shale decline curves | |
Nazaré et al. | Green-Box System Identification and Carbon Footprint Analysis for Sustainable Computing | |
CN114969017B (en) | Wind power data cleaning method, cleaning device and prediction method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |