CN107956731A - The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter - Google Patents
The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter Download PDFInfo
- Publication number
- CN107956731A CN107956731A CN201711242338.8A CN201711242338A CN107956731A CN 107956731 A CN107956731 A CN 107956731A CN 201711242338 A CN201711242338 A CN 201711242338A CN 107956731 A CN107956731 A CN 107956731A
- Authority
- CN
- China
- Prior art keywords
- sample set
- sampling
- unit
- sample
- samples
- 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.)
- Pending
Links
- 238000001816 cooling Methods 0.000 title claims abstract description 79
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 20
- 238000012706 support-vector machine Methods 0.000 claims abstract description 17
- 238000005070 sampling Methods 0.000 claims description 167
- 238000010977 unit operation Methods 0.000 claims description 23
- 238000006243 chemical reaction Methods 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 8
- 230000007613 environmental effect Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 5
- 238000012843 least square support vector machine Methods 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04D—NON-POSITIVE-DISPLACEMENT PUMPS
- F04D27/00—Control, e.g. regulation, of pumps, pumping installations or pumping systems specially adapted for elastic fluids
- F04D27/001—Testing thereof; Determination or simulation of flow characteristics; Stall or surge detection, e.g. condition monitoring
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01D—NON-POSITIVE DISPLACEMENT MACHINES OR ENGINES, e.g. STEAM TURBINES
- F01D21/00—Shutting-down of machines or engines, e.g. in emergency; Regulating, controlling, or safety means not otherwise provided for
- F01D21/003—Arrangements for testing or measuring
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Structures Of Non-Positive Displacement Pumps (AREA)
Abstract
The invention belongs to fired power generating unit control technology field, disclose the Forecasting Methodology that a kind of Air Cooling Fan for Power Station frequency change influences unit back pressuce parameter, the initial data of corresponding unit is gathered by DCS history station, and primary data sample is screened to obtain prediction data sample set；Operation data are acquired again, then by the way that new data sample set is updated or had increased access to primary data sample；Modeling is trained finally by by training sample input support vector machines, then logical majorized function seeks out the optimum parameter value of model.The predictor method is by mined information in the database and filters out effective information, sample database is updated by self study again and the relevance between data is excavated by support vector machines, so as to reach unit economy prediction when changing to air cooling blower fan frequency and correct sample database as the time is constantly updated.
Description
Technical Field
The invention belongs to the technical field of thermal power unit control, and particularly relates to a method for predicting influence of frequency change of an air cooling fan of a power station on a unit backpressure parameter.
Background
In the production process of the thermal power generating unit, the backpressure of the unit is too low, and the axial thrust of the steam turbine is increased; the back pressure is too high, the temperature of a steam turbine exhaust cylinder rises, the cylinder efficiency is reduced, the heat consumption is increased, and the steam turbine can vibrate in severe cases.
The frequency of the air cooling fan of the direct air cooling unit has direct influence on the backpressure of the unit, the backpressure has higher sensitivity to the frequency change (including starting and stopping) of the air cooling fan, and meanwhile, the air cooling fan consumes a large amount of service power. In the actual production process, the frequency of a certain air cooling fan may need to be started or stopped or changed according to factors such as unit load, environmental temperature and equipment failure, the back pressure of the unit changes along with the frequency, and the safety and economy of the unit are affected, so that the change of the back pressure of the unit when the frequency of the air cooling fan changes is accurately predicted, and the safe and economical operation of the unit is guided.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a method for predicting the influence of the frequency change of the air cooling fan in the power station on the unit backpressure parameter, which can accurately position the influence of the specific frequency change of the specific air cooling fan on the unit backpressure, so as to guide the unit operation, and has important significance for improving the unit safety and economy.
The technical idea of the technical scheme of the invention is as follows: firstly, acquiring original data of a corresponding unit through a DCS (distributed control system) historical station, and screening original data samples to obtain a final data sample set; collecting operation data, and updating or adding original data samples to obtain a new data sample set; and finally, inputting the training sample into a support vector machine for training and modeling, and finding out the optimal parameter value of the model by an optimization function. The prediction method comprises the steps of mining information in a database, screening effective information, updating a sample database through self-learning, mining the relevance between data through a support vector machine, and accordingly predicting the change of the backpressure of the unit when the frequency of the air-cooling fan changes and continuously updating and correcting the sample database along with time.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme.
A prediction method for influence of frequency change of air cooling fans of a power station on backpressure parameters of a unit, wherein the unit comprises a plurality of air cooling fans and a plurality of vacuum pumps, and the prediction method comprises the following steps:
step 1, obtaining a historical data sample of unit operation at each sampling time in a period of time, and recording the historical data sample as a first sample set, wherein the historical data sample of unit operation comprises: the sampling method comprises the following steps of (1) sampling a frequency value of each air cooling fan, a current sampling value of each vacuum pump, a unit power sampling value, a unit backpressure sampling value, an environment temperature sampling value and a sampling moment;
step 2, screening out a sample of the air cooling fan subjected to frequency conversion operation from the first sample set according to a frequency sampling value of any one air cooling fan in historical data samples of unit operation at every two adjacent sampling moments, and marking the sample as a second sample set;
step 3, according to two sampling moments in the historical data samples of the unit operation at every two adjacent sampling moments, eliminating samples with spurious frequency conversion operation caused by sampling moment difference in the second sample set, and recording the eliminated sample set as a third sample set;
step 4, according to the current sampling value of any vacuum pump in the historical data samples of the unit operation at every two adjacent sampling moments, eliminating samples of unit backpressure change caused by the vacuum pump current change in the third sample set, and recording the eliminated sample set as a final data sample set; the samples in the final data sample set are samples of unit back pressure change caused by frequency change of an air cooling fan under different unit power and ambient temperature conditions;
and 5, dividing samples in the final data sample set into training samples and testing samples, taking a frequency sampling value of an air cooling fan, a current sampling value of a vacuum pump, a power sampling value of a thermal power generating unit and an environment temperature sampling value corresponding to each training sample as input parameters of a support vector machine, taking a backpressure sampling value of a turboset corresponding to the training sample as output parameters of the support vector machine, obtaining the relationship among the frequency of the air cooling fan, the current of the vacuum pump, the power of the thermal power generating unit and the environment temperature and the backpressure of the turboset through all the training samples, and obtaining a predicted value of the frequency change of the air cooling fan on the backpressure of the turboset according to the relationship among the frequency of the air cooling fan, the current of the vacuum pump, the power of the thermal power generating unit and the environment temperature and the backpressure of.
The technical scheme of the invention has the characteristics and further improvements that:
(1) after step 4 and before step 5, the method further comprises: adjusting the unit backpressure of the samples in the final data sample set; the method specifically comprises the following steps:
and taking samples with the same conditions in the final data sample set as a group of samples, wherein the samples with the same conditions indicate that the same air cooling fan has the same frequency change, the same vacuum pump has the same current value, the unit power is the same, and the ambient temperature is the same, determining the unit backpressure values corresponding to different sampling moments according to the group of samples, thereby determining the unit backpressure stable value of the group of samples, and updating the unit backpressure sampling value of each sample in the group of samples to be the unit backpressure stable value.
(2) After step 4 and before step 5, the method further comprises:
acquiring a current data sample of unit operation at each sampling time in a current time period, and recording the current data sample as a current sample set, wherein the current data sample of unit operation comprises: the sampling method comprises the following steps of (1) sampling a frequency value of each air cooling fan, a current sampling value of each vacuum pump, a unit power sampling value, a unit backpressure sampling value, an environment temperature sampling value and a sampling moment;
and screening the current sample set to obtain a final data sample set of the current sample set, and updating the final data sample set of the historical data samples by adopting the final data sample set of the current sample set to obtain an updated sample set.
(3) Updating the final data sample set of the historical data samples by adopting the final data sample set of the current sample set, which specifically comprises the following steps:
if the environmental temperature sampling value and the unit power sampling value of a first sample in the final data sample set of the current sample set are respectively and correspondingly the same as the environmental temperature sampling value and the unit power sampling value of a second sample in the final data sample set of the historical data sample, replacing the second sample with the first sample; the first sample is any sample in a final data sample set of the current sample set, and the second sample is any sample in a final data sample set of the historical data samples.
(4) The step 2 specifically comprises the following substeps:
(2a) record the first sample set as D0＝{d1，d2，...，di，...，dM}；
Wherein d isi(1 ≦ i ≦ M) representing the ith historical data sample, M representing the total number of samples in the first sample set;
and is
Wherein, TiThe sampling time of the ith historical data sample is shown, N represents the total number of air coolers and vacuum pumps in the unit,respectively representing the frequency sampling values corresponding to the 1 st to the nth air cooling fans at the ith sampling moment,respectively represents current sampling values, P, corresponding to the (N + 1) th vacuum pump to the Nth vacuum pump at the ith sampling momentiRepresenting the backpressure sampling value of the unit corresponding to the ith sampling moment, NiRepresenting the unit power sampling value t corresponding to the ith sampling momentiRepresenting an environment temperature sampling value corresponding to the ith sampling moment;
(2b) in the first sample set D0Selecting samples satisfying the following conditions, and recording the samples as a second sample set D1：
Namely, it is
Wherein,the sampling value of the frequency corresponding to the jth air-cooled fan at the ith sampling moment is shown,indicating the frequency sampling value, I, corresponding to the jth air cooling fan at the (I + 1) th sampling moment1The frequency change value of the frequency conversion operation of the air cooling fan is shown.
(5) In step 3, eliminating samples with spurious frequency conversion operation caused by the difference of two sampling moments, specifically comprising:
eliminating samples meeting the following conditions from the second sample set, and recording the eliminated sample set as a third sample set:
|Ti-Ti+1|＞T0(1≤i≤M-1)
wherein, TiIndicating the sampling time, T, of the ith historical data samplei+1Represents the sampling time, T, of the i +1 th historical data sample0The time difference threshold value which represents the occurrence of false frequency conversion caused by the sampling time difference is set artificially;
the number of samples in the third sample set is M1(0≤M1≤M)。
(6) The step 4 specifically comprises the following steps:
eliminating samples meeting the following conditions in the third sample set, and recording the proposed sample set as a final data sample set:
wherein,the current sampling value corresponding to the jth vacuum pump at the ith sampling moment is represented, the current sampling value corresponding to the jth vacuum pump at the (I + 1) th sampling moment is represented, and I2The current value which represents the set current value caused by the change of the backpressure of the unit due to the change of the current of the vacuum pump is set manually;
the number of samples in the final data sample set is M2(O≤M2≤M1)。
According to the technical scheme, effective information is mined and screened from a massive operation database, the sample database is updated through self-learning, and the relevance between data is mined through a support vector machine, so that the purpose of predicting the backpressure change when the frequency of the air-cooled fan changes is achieved, the sample database is continuously updated and corrected along with time, the accuracy and operability are high, the influence of the specific frequency change of the specific air-cooled fan on the backpressure of the unit can be accurately positioned, the unit operation is guided, and the method and the system have important significance for improving the safety and economy of the unit.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for predicting the influence of the frequency change of the air cooling fan of the power station on the backpressure parameter of the unit, according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The mass historical data of the field DCS is recorded continuously in the operation process of the unit, the relevant data comprises a large amount of information which accords with the actual field, and the special relationships hidden in the mass data can be automatically searched out from the mass data through data mining on the basis of certain theoretical analysis in sequence, so that a basis is provided for realizing accurate prediction.
The embodiment of the invention provides a method for predicting the influence of frequency change of air cooling fans of a power station on backpressure parameters of a unit, wherein the unit comprises a plurality of air cooling fans and a plurality of vacuum pumps, and as shown in figure 1, the prediction method comprises the following steps:
step 1, obtaining a historical data sample of unit operation at each sampling time in a period of time, and recording the historical data sample as a first sample set, wherein the historical data sample of unit operation comprises: the sampling value of the frequency of each air cooling fan, the sampling value of the current of each vacuum pump, the sampling value of the power of the unit, the sampling value of the back pressure of the unit, the sampling value of the environmental temperature and the sampling time.
And 2, screening out a sample of the air cooling fan subjected to frequency conversion operation from the first sample set according to a frequency sampling value of any one air cooling fan in historical data samples of unit operation at every two adjacent sampling moments, and recording the sample as a second sample set.
The step 2 specifically comprises the following substeps:
(2a) record the first sample set as D0＝{d1，d2，...，di，...，dM}；
Wherein d isi(1 ≦ i ≦ M) representing the ith historical data sample, M representing the total number of samples in the first sample set;
and is
Wherein, TiThe sampling time of the ith historical data sample is shown, N represents the total number of air coolers and vacuum pumps in the unit,respectively representing the frequency sampling values corresponding to the 1 st to the nth air cooling fans at the ith sampling moment,respectively represents current sampling values, P, corresponding to the (N + 1) th vacuum pump to the Nth vacuum pump at the ith sampling momentiRepresenting the backpressure sampling value of the unit corresponding to the ith sampling moment, NiRepresenting the unit power sampling value t corresponding to the ith sampling momentiRepresenting an environment temperature sampling value corresponding to the ith sampling moment;
(2b) in the first sample set D0Selecting samples satisfying the following conditions, and recording the samples as a second sample set D1：
Namely, it is
It should be noted thatSample of conditions, consider that an air-cooled fan has occurredA variable frequency operation capable of preventing fine fluctuation of the frequency of the air-cooling fan from being considered as the variable frequency operation; wherein,the sampling value of the frequency corresponding to the jth air-cooled fan at the ith sampling moment is shown,indicating the frequency sampling value, I, corresponding to the jth air cooling fan at the (I + 1) th sampling moment1The frequency change value of the frequency conversion operation of the air cooling fan is shown.
And 3, according to two sampling moments in the historical data samples of the unit operation at every two adjacent sampling moments, eliminating samples with spurious frequency conversion operation caused by the difference of the sampling moments in the second sample set, and recording the eliminated sample set as a third sample set.
The step 3 specifically comprises the following steps:
eliminating samples meeting the following conditions from the second sample set, and recording the eliminated sample set as a third sample set:
|Ti-Ti+1|＞T0(1≤i≤M-1)
in addition, it is necessary to satisfy | Ti-Ti+1|＞T0If the condition is satisfied, the sample data is considered to be unstable to cause false frequency conversion, and the group of samples are rejected; wherein, TiIndicating the sampling time, T, of the ith historical data samplei+1Represents the sampling time, T, of the i +1 th historical data sample0The time difference threshold value which represents the occurrence of false frequency conversion caused by the sampling time difference is set artificially;
the number of samples in the third sample set is M1(0≤M1≤M)。
Step 4, according to the current sampling value of any vacuum pump in the historical data samples of the unit operation at every two adjacent sampling moments, eliminating samples of unit backpressure change caused by the vacuum pump current change in the third sample set, and recording the eliminated sample set as a final data sample set; and the samples in the final data sample set are samples of unit backpressure changes caused by frequency changes of the air cooling fan under different unit power and ambient temperature conditions.
The step 4 specifically comprises the following steps:
eliminating samples meeting the following conditions in the third sample set, and recording the proposed sample set as a final data sample set:
it should be noted thatThe sample of the condition is regarded as that the back pressure of the unit is changed due to the vacuum pump, and the sample is rejected; wherein,the current sampling value corresponding to the jth vacuum pump at the ith sampling moment is represented, the current sampling value corresponding to the jth vacuum pump at the (I + 1) th sampling moment is represented, and I2The current value which represents the set current value caused by the change of the backpressure of the unit due to the change of the current of the vacuum pump is set manually;
the number of samples in the final data sample set is M2(O≤M2≤M1)。
Further, after step 4 and before step 5, the method further comprises: adjusting the unit backpressure of the samples in the final data sample set; the method specifically comprises the following steps:
and taking samples with the same conditions in the final data sample set as a group of samples, wherein the samples with the same conditions indicate that the same air cooling fan has the same frequency change, the same vacuum pump has the same current value, the unit power is the same, and the ambient temperature is the same, determining the unit backpressure values corresponding to different sampling moments according to the group of samples, thereby determining the unit backpressure stable value of the group of samples, and updating the unit backpressure sampling value of each sample in the group of samples to be the unit backpressure stable value.
Further, after step 4 and before step 5, the method further comprises:
acquiring a current data sample of unit operation at each sampling time in a current time period, and recording the current data sample as a current sample set, wherein the current data sample of unit operation comprises: the sampling method comprises the following steps of (1) sampling a frequency value of each air cooling fan, a current sampling value of each vacuum pump, a unit power sampling value, a unit backpressure sampling value, an environment temperature sampling value and a sampling moment;
and screening the current sample set to obtain a final data sample set of the current sample set, and updating the final data sample set of the historical data samples by adopting the final data sample set of the current sample set to obtain an updated sample set.
Illustratively, the updating of the final data sample set of the historical data samples by using the final data sample set of the current sample set includes:
if the environmental temperature sampling value and the unit power sampling value of a first sample in the final data sample set of the current sample set are respectively and correspondingly the same as the environmental temperature sampling value and the unit power sampling value of a second sample in the final data sample set of the historical data sample, replacing the second sample with the first sample; the first sample is any sample in a final data sample set of the current sample set, and the second sample is any sample in a final data sample set of the historical data samples.
Specifically, assume that the first sample in the final data sample set of the current sample set is:
d′＝(T′inin，...，im，im+1，...，jNp ', N ', t '), where N is the total number of air cooling fans and vacuum pumps of the unit, in～imAir-cooled Fan frequency, i, as a first samplen+1～iNThe vacuum pump current of the first sample is T ', the ambient temperature of the first sample is T ', the backpressure of the first sample is P ', the unit power of the first sample is N ', and the time mark of the first sample is T '; and if the t 'and the N' exist in the integer range of the historical data sample, removing the original second sample and updating the final data sample set of the historical data sample, otherwise, adding the first sample into the final data sample set of the historical data sample.
And 5, dividing samples in the final data sample set into training samples and testing samples, taking a frequency sampling value of an air cooling fan, a current sampling value of a vacuum pump, a power sampling value of a thermal power generating unit and an environment temperature sampling value corresponding to each training sample as input parameters of a support vector machine, taking a backpressure sampling value of a turboset corresponding to the training sample as output parameters of the support vector machine, obtaining the relationship among the frequency of the air cooling fan, the current of the vacuum pump, the power of the thermal power generating unit and the environment temperature and the backpressure of the turboset through all the training samples, and obtaining a predicted value of the frequency change of the air cooling fan on the backpressure of the turboset according to the relationship among the frequency of the air cooling fan, the current of the vacuum pump, the power of the thermal power generating unit and the environment temperature and the backpressure of.
It should be noted that, values of the frequency of a plurality of groups of air cooling fans, the current of the vacuum pump, the unit power, the unit backpressure and the ambient temperature can be obtained through all the steps before the step 5, wherein the same air cooling fan may be changed in frequency and then corresponds to the change of the current of a specific group of vacuum pump, the unit power, the ambient temperature and the unit backpressure; or after the same frequency changes are respectively carried out on the same plurality of air cooling units, the changes of the current of a group of specific vacuum pumps, the power of the units, the ambient temperature and the back pressure of the units can be caused; or after the same air cooling units respectively carry out different frequency changes, the air cooling units correspond to various combinations such as the change of a group of specific vacuum pump current, the unit power, the ambient temperature, the unit backpressure and the like; therefore, training can be carried out through the support vector machine according to the data, and a relation model of the frequency of the air cooling fan, the current of the vacuum pump, the unit power, the ambient temperature and the unit backpressure is obtained.
In the final sample set, M in the final sample set2And (3) taking 20 groups of samples as training samples, inputting the training samples into a support vector machine for training and modeling, specifically, taking the frequency of an air cooling fan, the current of a vacuum pump, the unit power N and the ambient temperature t as input vectors of the support vector machine, taking the unit backpressure P as an output vector of the support vector machine, taking the rest 20 groups of samples as test samples, and testing the accuracy of the model.
Training samples in the set of data samples utilize non-linear mappingMapping samples from an original space to a feature spaceThereby realizing the conversion of the nonlinear regression in the input space into the linear regression of the high-order characteristic space, and the objective function of the least square support vector machine isWherein the constraint condition isi is 1,2, …, n, ξ is the slack variable, C is the penalty parameter, b is the bias value,for the mapping function, w is the normal vector of the hyperplane. Converting the target function containing constraint conditions into an unconstrained target function by a Lagrange method, and finally solvingThe optimization problem of the solution is converted into the solutionWherein y ═ y1,…,yn]T，IV＝[1,…,1]T，Ω＝{Ωij|i,j＝1,…,n}，K is kernel function, and linear regression is obtained by using least square method to obtain a and bAnd finally, obtaining a relation prediction model of the frequency of the air cooling fan, the current of the vacuum pump, the unit power, the ambient temperature and the unit backpressure.
Because the air cooling fans are rough in starting and stopping and frequency setting in the actual production site of the direct air cooling unit in the prior art, operators basically set the running number and the running frequency of the fans according to experience, and the influence on the back pressure of the unit when the frequency of some air cooling fans is changed by the traditional heat balance method is not feasible. According to the embodiment of the invention, effective information is mined and screened in the massive operation database, the sample database is updated through self-learning, and the relevance between the data is mined through the support vector machine, so that the purpose of predicting the backpressure change when the frequency of the air-cooled fan changes is achieved, the sample database is continuously updated and corrected along with time, the accuracy and operability are higher, the influence of the specific frequency change of the specific air-cooled fan on the backpressure of the unit can be accurately positioned, the operation of the unit is further guided, and the method and the device have important significance for improving the safety and economy of the unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (7)
1. A prediction method for influence of frequency change of air cooling fans of a power station on backpressure parameters of a unit, wherein the unit comprises a plurality of air cooling fans and a plurality of vacuum pumps, and the prediction method is characterized by comprising the following steps:
step 1, obtaining a historical data sample of unit operation at each sampling time within a set period of time, and recording the historical data sample as a first sample set, wherein the historical data sample of unit operation comprises: the method comprises the following steps of (1) sampling a frequency value of each air cooling fan, a current sampling value of each vacuum pump, a power sampling value of a thermal power generating unit, a backpressure sampling value of a steam turbine unit, an environment temperature sampling value and a sampling time;
step 2, screening out a sample of the air cooling fan subjected to frequency conversion operation from the first sample set according to a frequency sampling value of any one air cooling fan in historical data samples of unit operation at every two adjacent sampling moments, and marking the sample as a second sample set;
step 3, in the second sample set, according to two sampling moments in the historical data samples of the unit operation at every two adjacent sampling moments, eliminating samples with spurious frequency conversion operation caused by the difference of the two sampling moments, and recording the eliminated sample set as a third sample set;
step 4, in the third sample set, according to the current sampling value of any vacuum pump in the historical data samples of the unit operation at every two adjacent sampling moments, eliminating the sample of the backpressure change of the steam turbine unit caused by the current change of the vacuum pump, and recording the eliminated sample set as a final data sample set; the samples in the final data sample set are samples of backpressure changes of the steam turbine unit caused by frequency changes of the air cooling fan under different thermal power unit powers and different environmental temperature conditions;
step 5, dividing the samples in the final data sample set into training samples and testing samples, taking the frequency sampling value of the air cooling fan, the current sampling value of the vacuum pump, the power sampling value of the thermal power generating unit and the environment temperature sampling value corresponding to each training sample as input parameters of the support vector machine, and the back pressure sampling value of the steam turbine set corresponding to the training sample is used as the output parameter of the support vector machine, thereby obtaining the relationship between the frequency of the air cooling fan, the current of the vacuum pump, the power of the thermal power generating unit and the ambient temperature and the backpressure of the steam turbine unit through all training samples, and obtaining a predicted value of the air cooling fan frequency change to the steam turbine unit backpressure according to the frequency of the air cooling fan, the current of the vacuum pump, the power of the thermal power generating unit and the relation between the ambient temperature and the steam turbine unit backpressure, wherein the support vector machine adopts a least square support vector machine.
2. The method for predicting the influence of the frequency change of the air cooling fan of the power station on the unit backpressure parameter as claimed in claim 1, wherein the steam turbine unit backpressure of the sample in the final data sample set in the step 4 is corrected; the method specifically comprises the following steps:
taking samples with the same condition at different sampling time instants in the final data sample set as a group of samples, wherein the condition represents that: the thermal power generating units have the same power and the same ambient temperature, the same air cooling fans have the same frequency change, and the same vacuum pumps have the same current value;
and determining the back pressure values of the steam turbine set corresponding to different sampling moments according to the group of samples, thereby determining the back pressure stable value of the steam turbine set of the group of samples, and updating the back pressure sampling value of the steam turbine set of each sample in the group of samples into the back pressure stable value of the steam turbine set.
3. The method for predicting the influence of the frequency change of the air cooling fan of the power station on the backpressure parameter of the unit as claimed in claim 1, wherein after the step 4 and before the step 5, the method further comprises the following steps:
acquiring a current data sample of unit operation at each sampling time in a current time period, and recording the current data sample as a current sample set, wherein the current data sample of unit operation comprises: the method comprises the following steps of (1) sampling a frequency value of each air cooling fan, a current sampling value of each vacuum pump, a power sampling value of a thermal power generating unit, a backpressure sampling value of a steam turbine unit, an environment temperature sampling value and a sampling time;
and screening the current sample set to obtain a final data sample set of the current sample set, and updating the final data sample set of the historical data samples by adopting the final data sample set of the current sample set to obtain an updated sample set.
4. The method for predicting the influence of the frequency change of the air cooling fan of the power station on the backpressure parameter of the unit as claimed in claim 3, wherein the final data sample set of the historical data samples is updated by using the final data sample set of the current sample set, and specifically comprises the following steps:
if the environment temperature sampling value and the thermal power unit power sampling value of the first sample in the final data sample set of the current sample set are respectively and correspondingly the same as the environment temperature sampling value and the thermal power unit power sampling value of the second sample in the final data sample set of the historical data sample, replacing the second sample with the first sample; the first sample is any sample in a final data sample set of the current sample set, and the second sample is any sample in a final data sample set of the historical data samples.
5. The method for predicting the influence of the frequency change of the air cooling fan of the power station on the backpressure parameter of the unit as claimed in claim 1, wherein the step 2 specifically comprises the following substeps:
(2a) record the first sample set as D0＝{d1，d2，...，di，...，dM}；
Wherein d isi(1 ≦ i ≦ M) representing the ith historical data sample, M representing the total number of samples in the first sample set;
and is
Wherein, TiThe sampling time of the ith historical data sample is shown, N represents the total number of air coolers and vacuum pumps in the unit,respectively representing the frequency sampling values corresponding to the 1 st to the nth air cooling fans at the ith sampling moment,respectively represents current sampling values, P, corresponding to the (N + 1) th vacuum pump to the Nth vacuum pump at the ith sampling momentiRepresenting the sampling value of the back pressure of the steam turbine set corresponding to the ith sampling moment, NiThermal power generating unit power acquisition corresponding to ith sampling momentSample value, tiRepresenting an environment temperature sampling value corresponding to the ith sampling moment;
(2b) in the first sample set D0Selecting samples satisfying the following conditions, and recording the samples as a second sample set D1：
<mrow> <mo>|</mo> <mrow> <msubsup> <mi>i</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>-</mo> <msubsup> <mi>i</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>j</mi> </msubsup> </mrow> <mo>|</mo> <mo>></mo> <msub> <mi>I</mi> <mn>1</mn> </msub> <mo>,</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>&le;</mo> <mi>j</mi> <mo>&le;</mo> <mi>n</mi> <mo>,</mo> <mn>1</mn> <mo>&le;</mo> <mi>i</mi> <mo>&le;</mo> <mi>M</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
Namely, it is
Wherein,the sampling value of the frequency corresponding to the jth air-cooled fan at the ith sampling moment is shown,the frequency sampling value l corresponding to the (i + 1) th sampling time of the jth air cooling fan is represented1The frequency change value of the frequency conversion operation of the air cooling fan is shown.
6. The method for predicting the influence of the frequency change of the air cooling fan of the power station on the backpressure parameter of the unit as claimed in claim 1, wherein in the step 3, samples with false frequency conversion operation caused by the difference of two sampling moments are removed, and the method specifically comprises the following steps:
eliminating samples meeting the following conditions from the second sample set, and recording the eliminated sample set as a third sample set:
|Ti-Ti+1|＞T0(1≤i≤M-1)
wherein, TiIndicating the sampling time, T, of the ith historical data samplei+1Represents the sampling time, T, of the i +1 th historical data sample0The time difference threshold value which represents the occurrence of false frequency conversion caused by the sampling time difference is set artificially;
the number of samples in the third sample set is M1(0≤M1≤M)。
7. The method for predicting the influence of the frequency change of the air cooling fan of the power station on the backpressure parameter of the unit as claimed in claim 1, wherein the step 4 specifically comprises:
eliminating samples meeting the following conditions in the third sample set, and recording the proposed sample set as a final data sample set:
<mrow> <mo>|</mo> <mrow> <msubsup> <mi>i</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>-</mo> <msubsup> <mi>i</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>j</mi> </msubsup> </mrow> <mo>|</mo> <mo>></mo> <msub> <mi>I</mi> <mn>2</mn> </msub> <mo>,</mo> <mrow> <mo>(</mo> <mi>n</mi> <mo>+</mo> <mn>1</mn> <mo>&le;</mo> <mi>j</mi> <mo>&le;</mo> <mi>N</mi> <mo>,</mo> <mn>1</mn> <mo>&le;</mo> <mi>i</mi> <mo>&le;</mo> <msub> <mi>M</mi> <mn>1</mn> </msub> <mo>)</mo> </mrow> </mrow>
wherein,the current sampling value corresponding to the jth vacuum pump at the ith sampling moment is represented, the current sampling value corresponding to the jth vacuum pump at the (I + 1) th sampling moment is represented, and I2The current value which represents the backpressure change of the steam turbine set caused by the current change of the vacuum pump and is set manually;
the number of samples in the final data sample set is M2(O≤M2≤M1)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711242338.8A CN107956731A (en) | 2017-11-30 | 2017-11-30 | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711242338.8A CN107956731A (en) | 2017-11-30 | 2017-11-30 | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107956731A true CN107956731A (en) | 2018-04-24 |
Family
ID=61962694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711242338.8A Pending CN107956731A (en) | 2017-11-30 | 2017-11-30 | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107956731A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108800954A (en) * | 2018-05-31 | 2018-11-13 | 中冶华天工程技术有限公司 | Ring cold machine air quantity control method based on sound source and system |
| CN111130418A (en) * | 2019-12-24 | 2020-05-08 | 绍兴市上虞区理工高等研究院 | Method and device for detecting angular speed of rotor of permanent magnet synchronous motor |
| CN118249683A (en) * | 2024-05-21 | 2024-06-25 | 国家电投集团江西电力有限公司新昌发电分公司 | Intelligent excitation control method and system for thermal power generating unit based on autonomous controllable algorithm |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000019105A1 (en) * | 1998-09-25 | 2000-04-06 | Honeywell Inc. | Multiple centrifugal air compressor system control |
| CN101464964A (en) * | 2007-12-18 | 2009-06-24 | 同济大学 | Pattern recognition method capable of holding vectorial machine for equipment fault diagnosis |
| US7608353B2 (en) * | 2003-10-29 | 2009-10-27 | Gm Global Technology Operations, Inc. | Method for detecting compressor surge in a fuel cell system using a mass flow meter |
| CN102663412A (en) * | 2012-02-27 | 2012-09-12 | 浙江大学 | Power equipment current-carrying fault trend prediction method based on least squares support vector machine |
| CN103488869A (en) * | 2013-08-23 | 2014-01-01 | 上海交通大学 | Wind power generation short-term load forecast method of least squares support vector machine |
| CN104153981A (en) * | 2014-02-19 | 2014-11-19 | 东南大学 | Method for estimating backpressure data during starting and stopping of power station circulating water pump |
-
2017
- 2017-11-30 CN CN201711242338.8A patent/CN107956731A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000019105A1 (en) * | 1998-09-25 | 2000-04-06 | Honeywell Inc. | Multiple centrifugal air compressor system control |
| US7608353B2 (en) * | 2003-10-29 | 2009-10-27 | Gm Global Technology Operations, Inc. | Method for detecting compressor surge in a fuel cell system using a mass flow meter |
| CN101464964A (en) * | 2007-12-18 | 2009-06-24 | 同济大学 | Pattern recognition method capable of holding vectorial machine for equipment fault diagnosis |
| CN102663412A (en) * | 2012-02-27 | 2012-09-12 | 浙江大学 | Power equipment current-carrying fault trend prediction method based on least squares support vector machine |
| CN103488869A (en) * | 2013-08-23 | 2014-01-01 | 上海交通大学 | Wind power generation short-term load forecast method of least squares support vector machine |
| CN104153981A (en) * | 2014-02-19 | 2014-11-19 | 东南大学 | Method for estimating backpressure data during starting and stopping of power station circulating water pump |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108800954A (en) * | 2018-05-31 | 2018-11-13 | 中冶华天工程技术有限公司 | Ring cold machine air quantity control method based on sound source and system |
| CN111130418A (en) * | 2019-12-24 | 2020-05-08 | 绍兴市上虞区理工高等研究院 | Method and device for detecting angular speed of rotor of permanent magnet synchronous motor |
| CN118249683A (en) * | 2024-05-21 | 2024-06-25 | 国家电投集团江西电力有限公司新昌发电分公司 | Intelligent excitation control method and system for thermal power generating unit based on autonomous controllable algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Self-attention ConvLSTM and its application in RUL prediction of rolling bearings | |
| CN110837226B (en) | Thermal power generating unit operation optimization method based on intelligent optimization algorithm and related device | |
| CN107956731A (en) | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter | |
| CN113048807B (en) | Air cooling unit backpressure abnormity detection method | |
| CN112801388B (en) | Power load prediction method and system based on nonlinear time series algorithm | |
| CN117096504A (en) | Temperature control method and device, equipment and storage medium | |
| CN114881101B (en) | Bionic search-based power system typical scene association feature selection method | |
| CN115936676A (en) | Equipment health management method and system based on digital twins | |
| CN116205508A (en) | Distributed photovoltaic power generation abnormality diagnosis method and system | |
| CN109214610B (en) | Saturated power load prediction method based on long-term and short-term memory neural network | |
| CN117318033B (en) | Power grid data management method and system combining data twinning | |
| Zhang et al. | Condition based maintenance and operation of wind turbines | |
| Gong et al. | Machine learning-enhanced loT and wireless sensor networks for predictive analysis and maintenance in wind turbine systems | |
| Fu et al. | Digital twin-driven vibration amplitude simulation for condition monitoring of axial blowers in blast furnace ironmaking | |
| CN116720079A (en) | Wind driven generator fault mode identification method and system based on multi-feature fusion | |
| CN116720983A (en) | Power supply equipment abnormality detection method and system based on big data analysis | |
| CN115952914A (en) | Big data-based electric power metering operation and maintenance work judgment planning method | |
| CN112926801B (en) | Load curve combined prediction method and device based on quantile regression | |
| JP6880864B2 (en) | Energy management system and energy management method | |
| Jia et al. | Remaining useful life prediction of equipment based on xgboost | |
| CN114297582A (en) | Modeling method of discrete counting data based on multi-probe locality sensitive Hash negative binomial regression model | |
| CN113348473A (en) | Installation foundation for managing artificial intelligence module | |
| CN116227367B (en) | Back pressure prediction model construction method, back pressure prediction method and back pressure prediction device of direct air cooling system | |
| Li et al. | Power plant fan fault warning based on bidirectional feature compression and state estimation | |
| CN113447813B (en) | Fault diagnosis method and equipment for offshore wind generating set |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180424 |