CN112149903A - Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm - Google Patents
Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm Download PDFInfo
- Publication number
- CN112149903A CN112149903A CN202011011992.XA CN202011011992A CN112149903A CN 112149903 A CN112149903 A CN 112149903A CN 202011011992 A CN202011011992 A CN 202011011992A CN 112149903 A CN112149903 A CN 112149903A
- Authority
- CN
- China
- Prior art keywords
- frequency modulation
- primary frequency
- unit
- assessment
- index
- 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
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000005457 optimization Methods 0.000 title claims abstract description 27
- 238000004458 analytical method Methods 0.000 title claims abstract description 25
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 21
- 238000004364 calculation method Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000003745 diagnosis Methods 0.000 claims abstract description 13
- 238000003062 neural network model Methods 0.000 claims abstract description 8
- 230000009471 action Effects 0.000 claims description 64
- 230000004044 response Effects 0.000 claims description 53
- 238000011156 evaluation Methods 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 16
- 239000013598 vector Substances 0.000 claims description 14
- 230000001419 dependent effect Effects 0.000 claims description 13
- 230000001105 regulatory effect Effects 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 230000036316 preload Effects 0.000 claims description 4
- 238000003909 pattern recognition Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000013256 coordination polymer Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Feedback Control In General (AREA)
- Control Of Eletrric Generators (AREA)
Abstract
The invention discloses a thermal power generating unit primary frequency modulation analysis and optimization method based on a BP neural network algorithm, which comprises the following steps: acquiring primary frequency modulation performance index assessment data, operation data uploaded by a PMU subsystem of the thermal power generating unit and relevant operation data of units in a DCS system and a DEH system of the corresponding thermal power generating unit; respectively carrying out correlation calculation on the indexes in the primary frequency modulation performance index assessment and factors which possibly influence the primary frequency modulation assessment performance, and extracting derivative variables of the influencing factors according to the calculation result; constructing a primary frequency modulation analysis and diagnosis system by using a BP neural network algorithm, performing mode recognition on a primary frequency modulation assessment performance index result, and training system parameters; the derivative variables are varied within the constraint range and substituted into the trained BP neural network model for iterative calculation; and reversely analyzing the reason of the primary frequency modulation performance assessment index being unqualified according to the variation direction and degree of the derivative variable which meets the requirement that the output of the model is changed into qualified, so as to obtain an optimization scheme.
Description
Technical Field
The invention relates to the field of generating sets, in particular to a primary frequency modulation analysis and optimization method of a thermal power generating unit based on a BP neural network algorithm.
Background
The prior art is not strong in pertinence and practicability for comprehensively improving the primary frequency modulation performance of the thermal power generating unit. Firstly, the reason that the evaluation results of the primary frequency modulation action of each unit are not qualified is determined to be different due to different parameters, different states of operating equipment and even the difference of relevant configuration logics of the units, and the evaluation results are influenced by a single factor but are influenced by multiple factors together, so that comprehensive analysis and optimization are required to be carried out in multiple aspects. Therefore, it becomes necessary to perform targeted analysis and diagnosis on the primary frequency modulation action condition of each unit and provide a customized optimization scheme on the basis. In the prior art, only the current primary frequency modulation capability can be predicted, only the reason for examining the primary frequency modulation is searched from a single factor, or the reason for influencing the examination index of the primary frequency modulation performance is found, but a qualitative and quantitative improvement method or suggestion cannot be further provided in a targeted manner.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a thermal power generating unit primary frequency modulation analysis and optimization method based on a BP neural network algorithm, which comprises the following steps:
acquiring primary frequency modulation performance index assessment data, operation data uploaded by a PMU (phasor measurement unit) subsystem of a thermal power generating unit and relevant operation data of units in a DCS (distributed control system) system and a DEH (distributed data processing) system of the corresponding thermal power generating unit;
respectively performing correlation calculation on a 15S output response index, a 30S output response index and an electric quantity contribution index in the primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, performing correlation size sorting according to a calculation result, determining influence factors according to the correlation sorting, and extracting derivative variables of the influence factors according to the influence factors;
thirdly, constructing a primary frequency modulation analysis and diagnosis system by using a BP neural network algorithm, performing mode recognition on a primary frequency modulation assessment performance index result, and training system parameters;
step four, the variation derivative variables are substituted into the trained BP neural network model for iterative calculation until the output result of the model is changed from unqualified to qualified;
and step five, reversely analyzing the reason of the primary frequency modulation performance assessment index being unqualified according to the variation direction and degree of the derivative variable which meets the condition that the output of the model is qualified in the step four, and obtaining an optimization scheme.
Furthermore, the primary frequency modulation performance index assessment data comprise rated capacity of a generator set, start-stop time of primary frequency modulation action in the network, an initial value of primary frequency modulation action frequency, a primary frequency modulation action frequency extreme value, a 15S output response index, a 30S output response index and an electric quantity contribution index; the operation data uploaded by the PMU subsystem of the thermal power generating unit comprises active power, frequency, rotating speed, unit regulating stage pressure, a primary frequency modulation correction pre-load instruction in a coordination control system, a primary frequency modulation correction post-load instruction in the coordination control system, a valve opening signal, a primary frequency modulation action signal, a primary frequency modulation input signal and a boiler coordination signal; the relevant operation data of the units in the corresponding DCS and DEH systems of the thermal power generating unit comprise main steam pressure, main steam pressure temperature, main steam flow, high-pressure regulating valve opening instructions and high-pressure regulating valve opening feedback.
Further, the method comprises the following steps of respectively performing correlation calculation on the 15S output response index, the 30S output response index and the electric quantity contribution index in the primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, performing correlation size sorting according to correlation coefficient calculation results, determining influencing factors according to the correlation sorting, and extracting derivative variables of the influencing factors according to the influencing factors, wherein the correlation calculation method comprises the following steps:
correlation coefficient ρxyThe calculation of (2): one of the factors possibly influencing the primary frequency modulation performance is taken as a dependent variable X, one of the primary frequency modulation performance index evaluations is taken as an independent variable Y, and the calculation formula is
Wherein sigmaxIs the standard deviation, σ, of the dependent variableYIs the standard deviation of the independent variable, Cov [ X, Y]＝E[(X-u)(Y-v)]Is the covariance of the independent variable and the dependent variable, u is the expected value of the dependent variable, and v is the expected value of the independent variable;
respectively carrying out correlation calculation on a 15S output response index, a 30S output response index and an electric quantity contribution index in primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, wherein the factors possibly influencing the primary frequency modulation assessment performance comprise deviation of a unit calculation frequency and a power grid assessment frequency, main steam pressure, rotating speed unequal rate, unit valve opening, a synchronous AGC (automatic gain control) adjusting direction, unit valve flow characteristics and the like; and sequencing according to the magnitude of the correlation, and determining five factors influencing the primary frequency modulation checking performance as the main steam pressure of the unit, the unequal rate of the rotating speed of the unit, the flow characteristic linearity of a valve of the unit, the frequency deviation of the unit and the isotropy of an AGC (automatic gain control) regulation instruction of the unit according to the first 5 factors of the sequencing result of the magnitude of the correlation.
Further, the derived variables of the influencing factors include:
main steam pressure derivative variable: taking the ratio of the main steam pressure to the rated main steam pressure in the primary frequency modulation action assessment time period as a derivative variable for measuring the size of the main steam pressure;
the flow characteristic linearity derivative variable of the unit valve is as follows: taking the ratio of the percentage of the actual main steam flow variation in the primary frequency modulation action assessment time period to the percentage of the flow instruction variation in the same time period as a derivative variable for measuring the local flow characteristic of the unit;
unit frequency deviation derived variable: the frequency deviation for power grid examination is the difference between the rated frequency of the power grid and the frequency of a power grid examination reference point in a primary frequency modulation action examination time period; the frequency deviation corresponding to the response of the primary frequency modulation action of the unit is (load instruction after primary frequency modulation correction in the coordinated control system-load instruction before primary frequency modulation correction in the coordinated control system) × (50 × unit rotating speed unequal rate)/rated active power of the unit; taking the ratio of the frequency deviation responded by the unit primary frequency regulation action to the frequency deviation used for power grid examination as a derivative vector for measuring the frequency deviation measurement error;
the unit rotating speed unequal rate derivative variable is as follows: the unit rotation speed inequality rate is a set constant in the unit primary frequency modulation configuration logic, and the set range is 4% to 5%. The power grid evaluation standard is 5%, and the ratio of 5% to the rotating speed unequal rate actually set by the unit is used as a derivative variable for measuring the set rotating speed unequal rate of the unit;
unit AGC instruction isotropic derivative variable: and taking the vector ratio of the variation of the AGC command to the variation of the primary frequency modulation theoretical action in the primary frequency modulation action assessment time period as a derivative variable for measuring the isotropy of the AGC command and the primary frequency modulation action direction.
Further, the method for constructing the primary frequency modulation analysis and diagnosis system by using the BP neural network algorithm, performing pattern recognition on the primary frequency modulation assessment performance index result, and training system parameters comprises the following processes:
and (3) taking the assessment data of the dispatching center and the data after the normalization processing of the derivative vectors as input, taking the assessment result of the primary frequency modulation performance index of the corresponding unit as output, respectively building a 15S output response index model, a 30S output response index model and an electric quantity contribution index model, and training the system to obtain the weight and the threshold of the system model.
Further, the step of substituting the variant derivative variables within the constraint range into the trained BP neural network model for iterative computation until the output result of the model is changed from unqualified to qualified comprises the following steps: the single or multiple derived variables are mutated one by one, and iterative calculation is carried out again by using an expert diagnosis system obtained by training until the primary frequency modulation performance index assessment result is changed from unqualified to qualified; when the primary frequency modulation performance index assessment result is changed from unqualified to qualified, the reason that the primary frequency modulation performance assessment index is unqualified is reversely analyzed according to the variation direction and degree of the derivative variable which is varied at the moment, and then a corresponding optimization strategy is obtained.
Further, the specific process of mutating one or more derived variables one by one is as follows: the derived variable with the maximum correlation with the assessment index is mutated by setting the current value of the derived variable as an initial value alpha0If the set variation step is the single variation d, the value α of the derivative variable after the nth variation is obtainedn＝α0And + n + d, calculating again by using the value after the variation every time the variation is performed until the output result of the model is not suitableLattice becomes qualified or alphanThe value of (b) is outside the variation range; if α isnIf the value of the variable is up to the upper limit value of the variation range and the output result of the model is still unqualified, the derivative variables with the correlation smaller than the current derivative variable are sequentially selected to continue to be varied until the output result of the model is changed from unqualified to qualified or all the derivative variables are varied.
The invention has the beneficial effects that: the method has strong pertinence, original data are directly extracted from a power grid evaluation system, primary frequency modulation evaluation performance indexes of the thermal power generating unit are directly analyzed as dependent variables, and finally formed optimization strategies are set by specially aiming at improving the primary frequency modulation evaluation performance indexes. By applying the technology, the qualification rate of the primary frequency modulation action examination of the thermal power plant can be improved, and the stability and the regulation capability of the grid frequency of the power grid are enhanced.
The practicability is strong, and the extracted characteristic variables and derivative variables respectively correspond to five elements of main steam pressure, unequal rotating speed, flow characteristics of a unit valve, frequency deviation and AGC (automatic gain control) regulating instruction isotropy. By applying the technology, a targeted improvement and promotion strategy can be qualitatively and quantitatively given according to the sequence of the correlation magnitude, and the unit can be directly guided to carry out optimization work.
Drawings
FIG. 1 is a thermal power generating unit primary frequency modulation analysis and optimization method based on a BP neural network algorithm.
Detailed Description
The technical solutions of the present invention are further described in detail below with reference to the accompanying drawings, but the scope of the present invention is not limited to the following.
As shown in fig. 1, the method mainly comprises the following steps:
acquiring primary frequency modulation performance index assessment data, operation data uploaded by a PMU (phasor measurement unit) subsystem of a thermal power generating unit and relevant operation data of units in a DCS (distributed control system) system and a DEH (distributed data processing) system of the corresponding thermal power generating unit;
respectively performing correlation calculation on a 15S output response index, a 30S output response index and an electric quantity contribution index in the primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, performing correlation size sorting according to a calculation result, determining influence factors according to the correlation sorting, and extracting derivative variables of the influence factors according to the influence factors;
thirdly, constructing a primary frequency modulation analysis and diagnosis system by using a BP neural network algorithm, performing mode recognition on a primary frequency modulation assessment performance index result, and training system parameters;
step four, the variation derivative variables are substituted into the trained BP neural network model for iterative calculation until the output result of the model is changed from unqualified to qualified;
and step five, reversely analyzing the reason of the primary frequency modulation performance assessment index being unqualified according to the variation direction and degree of the derivative variable which meets the condition that the output of the model is qualified in the step four, and obtaining an optimization scheme.
Furthermore, the primary frequency modulation performance index assessment data comprise rated capacity of a generator set, start-stop time of primary frequency modulation action in the network, an initial value of primary frequency modulation action frequency, a primary frequency modulation action frequency extreme value, a 15S output response index, a 30S output response index and an electric quantity contribution index; the operation data uploaded by the PMU subsystem of the thermal power generating unit comprises active power, frequency, rotating speed, unit regulating stage pressure, a primary frequency modulation correction pre-load instruction in a coordination control system, a primary frequency modulation correction post-load instruction in the coordination control system, a valve opening signal, a primary frequency modulation action signal, a primary frequency modulation input signal and a boiler coordination signal; the relevant operation data of the units in the corresponding DCS and DEH systems of the thermal power generating unit comprise main steam pressure, main steam pressure temperature, main steam flow, high-pressure regulating valve opening instructions and high-pressure regulating valve opening feedback.
Further, the method comprises the following steps of respectively performing correlation calculation on the 15S output response index, the 30S output response index and the electric quantity contribution index in the primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, performing correlation size sorting according to correlation coefficient calculation results, determining influencing factors according to the correlation sorting, and extracting derivative variables of the influencing factors according to the influencing factors, wherein the correlation calculation method comprises the following steps:
correlation coefficient ρxyThe calculation of (2): one of the factors possibly influencing the primary frequency modulation performance is taken as a dependent variable X, one of the primary frequency modulation performance index evaluations is taken as an independent variable Y, and the calculation formula is
Wherein sigmaxIs the standard deviation, σ, of the dependent variableYIs the standard deviation of the independent variable, Cov [ Y, Y]＝E[(X-u)(Y-v)]Is the covariance of the independent variable and the dependent variable, u is the expected value of the dependent variable, and v is the expected value of the independent variable;
respectively carrying out correlation calculation on a 15S output response index, a 30S output response index and an electric quantity contribution index in primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, wherein the factors possibly influencing the primary frequency modulation assessment performance comprise deviation of a unit calculation frequency and a power grid assessment frequency, main steam pressure, rotating speed unequal rate, unit valve opening, a synchronous AGC (automatic gain control) adjusting direction, unit valve flow characteristics and the like; and sequencing according to the magnitude of the correlation, and determining five factors influencing the primary frequency modulation checking performance as the main steam pressure of the unit, the unequal rate of the rotating speed of the unit, the flow characteristic linearity of a valve of the unit, the frequency deviation of the unit and the isotropy of an AGC (automatic gain control) regulation instruction of the unit according to the first 5 factors of the sequencing result of the magnitude of the correlation.
Further, the derived variables of the influencing factors include:
main steam pressure derivative variable: taking the ratio of the main steam pressure to the rated main steam pressure in the primary frequency modulation action assessment time period as a derivative variable for measuring the size of the main steam pressure;
the flow characteristic linearity derivative variable of the unit valve is as follows: taking the ratio of the percentage of the actual main steam flow variation in the primary frequency modulation action assessment time period to the percentage of the flow instruction variation in the same time period as a derivative variable for measuring the local flow characteristic of the unit;
unit frequency deviation derived variable: the frequency deviation for power grid examination is the difference between the rated frequency of the power grid and the frequency of a power grid examination reference point in a primary frequency modulation action examination time period; the frequency deviation corresponding to the response of the primary frequency modulation action of the unit is (load instruction after primary frequency modulation correction in the coordinated control system-load instruction before primary frequency modulation correction in the coordinated control system) × (50 × unit rotating speed unequal rate)/rated active power of the unit; taking the ratio of the frequency deviation responded by the unit primary frequency regulation action to the frequency deviation used for power grid examination as a derivative vector for measuring the frequency deviation measurement error;
the unit rotating speed unequal rate derivative variable is as follows: the unit rotation speed inequality rate is a set constant in the unit primary frequency modulation configuration logic, and the set range is 4% to 5%. The power grid evaluation standard is 5%, and the ratio of 5% to the rotating speed unequal rate actually set by the unit is used as a derivative variable for measuring the set rotating speed unequal rate of the unit;
unit AGC instruction isotropic derivative variable: and taking the vector ratio of the variation of the AGC command to the variation of the primary frequency modulation theoretical action in the primary frequency modulation action assessment time period as a derivative variable for measuring the isotropy of the AGC command and the primary frequency modulation action direction.
Further, the method for constructing the primary frequency modulation analysis and diagnosis system by using the BP neural network algorithm, performing pattern recognition on the primary frequency modulation assessment performance index result, and training system parameters comprises the following processes:
and (3) taking the assessment data of the dispatching center and the data after the normalization processing of the derivative vectors as input, taking the assessment result of the primary frequency modulation performance index of the corresponding unit as output, respectively building a 15S output response index model, a 30S output response index model and an electric quantity contribution index model, and training the system to obtain the weight and the threshold of the system model.
Further, the step of substituting the variant derivative variables within the constraint range into the trained BP neural network model for iterative computation until the output result of the model is changed from unqualified to qualified comprises the following steps: the single or multiple derived variables are mutated one by one, and iterative calculation is carried out again by using an expert diagnosis system obtained by training until the primary frequency modulation performance index assessment result is changed from unqualified to qualified; when the primary frequency modulation performance index assessment result is changed from unqualified to qualified, the reason that the primary frequency modulation performance assessment index is unqualified is reversely analyzed according to the variation direction and degree of the derivative variable which is varied at the moment, and then a corresponding optimization strategy is obtained.
Further, the specific process of mutating one or more derived variables one by one is as follows: the derived variable with the maximum correlation with the assessment index is mutated by setting the current value of the derived variable as an initial value alpha0If the set variation step is the single variation d, the value α of the derivative variable after the nth variation is obtainedn＝α0And + n x d, calculating once again by using the value after the variation every time the variation is performed until the output result of the model is changed from unqualified to qualified or alphanThe value of (b) is outside the variation range; if α isnIf the value of the variable is up to the upper limit value of the variation range and the output result of the model is still unqualified, the derivative variables with the correlation smaller than the current derivative variable are sequentially selected to continue to be varied until the output result of the model is changed from unqualified to qualified or all the derivative variables are varied.
(1) Data is acquired.
And (3) calling primary frequency modulation performance index assessment data from a power grid dispatching center WARMS system database, wherein the assessment data comprise rated capacity of a generator set, start-stop time of primary frequency modulation action in a power grid, initial value of primary frequency modulation action frequency, extreme value of primary frequency modulation action frequency, 15S output response index, 30S output response index and electric quantity contribution index.
And the operation data uploaded by the PMU subsystem of the thermal power generating unit is called from a power grid dispatching center WARMS system database, and comprises active power, frequency, rotating speed, unit regulating stage pressure, a primary frequency modulation correction pre-load instruction in a coordination control system, a primary frequency modulation correction post-load instruction in the coordination control system, a valve opening signal, a primary frequency modulation action signal, a primary frequency modulation input signal and a turbine and boiler coordination signal.
And relevant operation data of the thermal power generating unit are called from a DCS (distributed control system) and a DEH (distributed data processing) system corresponding to the thermal power generating unit, wherein the relevant operation data comprise main steam pressure, main steam pressure temperature, main steam flow, high-pressure regulating valve opening instruction, high-pressure regulating valve opening feedback and the like.
(2) Qualitative and quantitative analysis of the influence factors of the primary frequency modulation examination index performance.
Respectively calculating a 15S output response index, a 30S output response index and an electric quantity contribution index in the primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, such as deviation of unit calculation frequency and power grid assessment frequency, main steam pressure, rotating speed unequal rate, unit valve opening, simultaneous AGC (automatic gain control) adjustment direction, unit valve flow characteristics and the like; then sorting is carried out according to the size of the correlation,
(3) and extracting derivative variables capable of representing the influence factors of the performance of the primary frequency modulation assessment indexes.
And according to the analysis result of the previous step, determining five influence factors of main steam pressure of the unit, unequal rate of rotating speed of the unit, flow characteristic linearity of a valve of the unit, frequency deviation of the unit and isotropy of AGC (automatic gain control) adjustment instructions of the unit.
Main steam pressure derivative variable: taking the ratio of the main steam pressure to the rated main steam pressure in the primary frequency modulation action assessment time period as a derivative variable for measuring the size of the main steam pressure;
the flow characteristic linearity derivative variable of the unit valve is as follows: taking the ratio of the percentage of the actual main steam flow variation in the primary frequency modulation action assessment time period to the percentage of the variation of the flow instruction (called as a valve opening signal in a PMU system) in the same time period as a derivative variable for measuring the local flow characteristic of the unit;
unit frequency deviation derived variable: the frequency deviation for power grid examination is the difference between the rated frequency of the power grid and the frequency of the power grid examination reference point in the primary frequency modulation action examination time period. The frequency deviation corresponding to the primary frequency modulation action response of the unit is the difference value between the real measured network frequency of the unit and the rated network frequency in the primary frequency modulation action assessment time period, and the frequency deviation corresponding to the primary frequency modulation action response of the unit (load instruction after primary frequency modulation correction in the coordinated control system-load instruction before primary frequency modulation correction in the coordinated control system) ((50) unit rotating speed unequal rate)/rated active power of the unit can also be calculated by a formula. Taking the ratio of the frequency deviation responded by the unit primary frequency regulation action to the frequency deviation used for power grid examination as a derivative vector for measuring the frequency deviation measurement error;
the unit rotating speed unequal rate derivative variable is as follows: the unit rotation speed inequality rate is a set constant in the unit primary frequency modulation configuration logic, and the set range is 4% to 5%. The power grid evaluation standard is 5%, so that the ratio of 5% to the rotating speed unequal rate actually set by the unit is used as a derivative variable for measuring the set rotating speed unequal rate of the unit;
unit AGC instruction isotropic derivative variable: and taking the vector ratio of the variation of the AGC command to the variation of the primary frequency modulation theoretical action in the primary frequency modulation action assessment time period as a derivative variable for measuring the isotropy of the AGC command and the primary frequency modulation action direction.
(4) And constructing a primary frequency modulation analysis and diagnosis system by using a BP neural network algorithm, performing mode recognition on a primary frequency modulation assessment performance index result, and training system parameters.
The data obtained by normalization processing of five characteristic derivative vectors obtained by mining of scheduling center assessment data and a big data platform in the previous step and corresponding unit primary frequency modulation performance index assessment results are used as input and output, wherein the normalization processing is to carry out linear transformation on original data through maximum-minimum standardization, the input original data in A variables is set as x, the maximum value of the A variables is max (A), the minimum value of the A variables is min (A), and then x passes through
After max-min normalization
And respectively building a 15S output response index model, a 30S output response index model and an electric quantity contribution index model, and repeatedly training the system to obtain the weight and the threshold of the system model. Taking a 15S output response index model as an example, five input layer neurons (five characteristic derivative variables) and one output layer neuron number (15S output response index assessment result) are required, eight hidden layer neurons are required, and a neuron matrix structure of [5,8,1] is established. And taking N times of primary frequency modulation action assessment data, wherein N/2 times are data with qualified primary frequency modulation action performance indexes, and the other N/2 times are data with unqualified primary frequency modulation action performance indexes, the input matrix is N rows and 5 columns [ N, 5], and the output matrix is N rows and 1 column [ N, 1 ]. The hidden layer activation function adopts an asymmetric Sigmoid function tansig, the output layer activation function adopts purelin, and the training function adopts traingf. The learning rate is 0.05, the maximum times is 10000, and the error target for terminating iteration is 0.001. The 30S output response index model and the electric quantity contribution index model are the same as the 15S output response index model.
(5) And (3) moderately mutating derivative variables in the constraint range, substituting the derivative variables into the trained BP neural network model, and performing iterative computation until the output result of the model is changed from unqualified to qualified.
And (3) changing one or more derivative variables one by one in a reasonable scope, and carrying out iterative calculation again by using an expert diagnosis system obtained by training so as to change the original primary frequency modulation performance index assessment result from unqualified one to qualified one. When the primary frequency modulation performance index assessment result is changed from unqualified to qualified, the reason that the primary frequency modulation performance assessment index is unqualified is reversely analyzed according to the variation direction and degree of the derivative variable which is varied at the moment, and then a corresponding optimization strategy is obtained.
Constrained range of moderate "variation" of derived variables:
frequency deviation variation range: if the grid frequency deviation of the actual response of the thermal power generating unit is smaller than the grid reference point frequency deviation, the frequency deviation variation range is { the grid frequency deviation of the actual response of the thermal power generating unit, the grid reference point frequency deviation }; and if the grid frequency deviation actually reflected by the thermal power generating unit is greater than or equal to the grid reference point frequency deviation, variation is not needed.
Frequency deviation variation step length: 0.001 Hz.
The variation range of the unequal rate of the rotating speed is as follows: if the actual rotating speed unequal rate is larger than or equal to 5%, the rotating speed unequal rate variation range is { 4%, 5% }; if the actual rotating speed unequal rate is less than or equal to 4 percent, the rotating speed unequal rate does not need to be changed.
Variable step length of unequal rate of rotation: 0.1 percent.
Main steam pressure variation range: if the current main steam pressure is less than 1.15 times of the main steam pressure set value corresponding to the unit slip pressure characteristic curve of the current generator active power value, taking 1.15 times of the main steam pressure set value corresponding to the unit slip pressure characteristic curve of the current generator active power value of the thermal power unit as the upper limit of the variation range, and when the value is greater than or equal to the rated main steam pressure value of the unit, taking the rated main steam pressure value of the unit as the upper limit of the variation range, wherein the main steam pressure variation range is { the current main steam pressure value, Min (the rated main steam pressure is 1.15 times of the main steam pressure set value corresponding to the unit slip pressure characteristic curve of the current generator active power value) }. If the current main steam pressure is larger than or equal to 1.15 times of the main steam pressure set value corresponding to the slip pressure characteristic curve of the unit by using the current generator active power value, variation is not needed.
Main steam pressure variation step length: 0.01 MPa.
The range of variation of the linearity of the flow characteristic of the unit is as follows: if the current flow characteristic linearity ratio is smaller than 100%, taking 100% as the upper limit of the variation range, and setting the variation range of the flow characteristic linearity ratio as { flow characteristic ratio, 100% }; if the linearity of the current flow characteristic ratio is greater than or equal to 100%, no variation is needed.
The unit flow characteristic linearity variation step length is as follows: 0.1 percent.
AGC instruction range of isotropic variation: if the vector ratio of the AGC instruction variable quantity to the primary frequency modulation theoretical variable quantity in the primary frequency modulation action examination time period is a negative value, namely the AGC instruction increasing and decreasing direction is opposite to the primary frequency modulation action direction, directly changing the AGC instruction isotropy to be 0; if the vector ratio of the AGC instruction variation and the primary frequency modulation theoretical variation in the primary frequency modulation action evaluation time period is a positive value, namely the AGC instruction increasing and decreasing direction is the same as the primary frequency modulation action direction, variation is not needed.
Default mutation order: AGC instruction isotropy variation → flow characteristic variation → main steam pressure variation → actual rotational speed inequality variation → frequency difference variation.
Actually, the weight values of all derived scalars are determined in sequence, and the variation is performed from large to small until the evaluation result of the primary frequency modulation performance index through iterative calculation is changed from unqualified to qualified, the corresponding 15S output response index is changed from less than 75% to greater than or equal to 75% (if the 15S output response index is greater than or equal to 75%, variation of the derived variable aiming at the index is not needed), the corresponding 30S output response index is changed from less than 90% to greater than or equal to 90% (if the 30S output response index is greater than or equal to 90%, variation of the derived variable aiming at the index is not needed), and the corresponding electric quantity contribution index is changed from less than 75% to greater than or equal to 75% (if the electric quantity contribution index is greater than or equal to 75%, variation of the derived variable aiming at the index is not needed).
Recording the dimension range of the derived variables for measuring the evaluation performance of the primary frequency modulation action:
AGC command isotropy RInitialValue (reproducibility) { negative infinity, positive infinity }
Linearity of flow characteristic LInitialValue (Linear) {0, positive infinity }
Main steam pressure PInitialRatio (Presure) {0, positive infinity }
Rate of rotation inequalityInitialRatio {1.0, 1.25}
Frequency deviation HInitialRatio (Hz) { negative infinity, positive infinity }
The direction and degree of variation of the derived variables that the model output becomes qualified are satisfied:
derivative variables required for the 15S force response index to pass from fail: r15S、L15S、P15S、15S、H15SDegree of variation Δ R thereof15S＝R15S-RInitial，ΔL15S＝L15S-LInitial，ΔP15S＝P15S-PInitial，Δ15S＝Initial-15S，ΔH15S＝H15S-HInitial(if any derivative variable does not need to be varied in the process, then the corresponding delta value of the term is zero),
derivative variables required for the 30S output response index to pass from fail: r30S、L30S、P30S、30S、H30SDegree of variation Δ R thereof30S＝R30S-RInitial，ΔL30S＝L30S-LInitial，ΔP30S＝P30S-PInitial，Δ30S＝Initial-30S，ΔH30S＝H30S-HInitial(if any derivative variable does not need to be varied in the process, then the corresponding delta value of the term is zero),
the derivative variables required to change the electricity contribution index from off-grade to on-grade: rElectric quantity、LElectric quantity、PElectric quantity、Electric quantity、HElectric quantityDegree of variation Δ R thereofElectric quantity＝RElectric quantity-RInitial，ΔLElectric quantity＝LElectric quantity-LInitial，ΔPElectric quantity＝PElectric quantity-PInitial，ΔElectric quantity＝Initial-Electric quantity，ΔHElectric quantity＝HElectric quantity-HInitial(if any derivative variable does not need to be varied in the process, then the corresponding delta value of the term is zero),
(6) and reversely analyzing the reason of the primary frequency modulation performance assessment index failure according to the variation direction and degree of the derivative variable which meets the condition that the output of the model is changed into qualified in the last step, and providing a targeted optimization scheme on the basis.
Analyzing the performance evaluation condition of the single primary frequency modulation action: the derivative variable and the variation degree thereof to meet the requirement of all three assessment indexes are delta R-Max (delta R)15S，ΔR30S，ΔRElectric quantity)，ΔL＝Max(ΔL15S，ΔL30S，ΔLElectric quantity)，ΔP＝Max(ΔP15S，ΔP30S，ΔPElectric quantity)，Δ＝Max(Δ15S，Δ30S，ΔElectric quantity)，ΔH＝Max(ΔH15S，ΔH30S，ΔHElectric quantity)
That is, the condition must satisfy { Max (Δ R)15S，ΔR30S，ΔRElectric quantity)∪Max(ΔL15S，ΔL30S，ΔLElectric quantity)∪Max(ΔP15S，ΔP30S，ΔPElectric quantity)∪Max(Δ15S，Δ30S，ΔElectric quantity)∪Max(ΔH15S，ΔH30S，ΔHElectric quantity)}＝{ΔR∪ΔL∪ΔP∪Δ∪ΔH}。
Statistical analysis of performance assessment conditions of multiple primary frequency modulation actions: counting the performance index evaluation and analysis condition of N/2 primary frequency modulation actions, namely traversing the [ delta ] Ri∪ΔLi∪ΔPi∪ΔTi∪Δi∪ΔHi1,2,3. Statistical Δ RiThe number of times not equal to 0, denoted as MR，0≤MRLess than or equal to N/2; statistical DeltaLiThe number of times not equal to 0, denoted as ML，0≤MLLess than or equal to N/2; statistical Δ PiThe number of times not equal to 0, denoted as MP，0≤MPLess than or equal to N/2; statistical DeltaiThe number of times not equal to 0, denoted as M，0≤MLess than or equal to N/2; statistic of Δ HiThe number of times not equal to 0, denoted as MH，0≤MHLess than or equal to N/2; calculating the frequency C of each influential triggerR＝2MR/N*100％，CL＝2ML/N*100％，CP＝2MP/N*100％，C＝2M/N*100％，CH＝2MH/N*100％。
ΔRmax＝max(ΔRi)，ΔLmax＝max(ΔLi)，ΔPmax＝max(ΔPi)，Δmax＝max(Δi)，ΔHmax＝max(ΔHi) 1,2,3. The final optimization may be { Δ Rmax∪ΔLmax∪ΔPmax∪Δmax∪ΔHmaxReference is made.
The foregoing is illustrative of the preferred embodiments of this invention, and it is to be understood that the invention is not limited to the precise form disclosed herein and that various other combinations, modifications, and environments may be resorted to, falling within the scope of the concept as disclosed herein, either as described above or as apparent to those skilled in the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (7)
1. A thermal power generating unit primary frequency modulation analysis and optimization method based on a BP neural network algorithm is characterized by comprising the following steps:
acquiring primary frequency modulation performance index assessment data, operation data uploaded by a PMU (phasor measurement unit) subsystem of a thermal power generating unit and relevant operation data of units in a DCS (distributed control system) system and a DEH (distributed data processing) system of the corresponding thermal power generating unit;
respectively performing correlation calculation on a 15S output response index, a 30S output response index and an electric quantity contribution index in the primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, performing correlation size sorting according to a calculation result, determining influence factors according to the correlation sorting, and extracting derivative variables of the influence factors according to the influence factors;
thirdly, constructing a primary frequency modulation analysis and diagnosis system by using a BP neural network algorithm, performing mode recognition on a primary frequency modulation assessment performance index result, and training system parameters;
step four, the variation derivative variables are substituted into the trained BP neural network model for iterative calculation until the output result of the model is changed from unqualified to qualified;
and step five, reversely analyzing the reason of the primary frequency modulation performance assessment index being unqualified according to the variation direction and degree of the derivative variable which meets the condition that the output of the model is qualified in the step four, and obtaining an optimization scheme.
2. The thermal power generating unit primary frequency modulation analysis and optimization method based on the BP neural network algorithm as claimed in claim 1, wherein the primary frequency modulation performance index assessment data comprises a generating unit rated capacity, a network primary frequency modulation action start-stop time, a primary frequency modulation action frequency initial value, a primary frequency modulation action frequency extreme value, a 15S output response index, a 30S output response index and an electric quantity contribution index; the operation data uploaded by the PMU subsystem of the thermal power generating unit comprises active power, frequency, rotating speed, unit regulating stage pressure, a primary frequency modulation correction pre-load instruction in a coordination control system, a primary frequency modulation correction post-load instruction in the coordination control system, a valve opening signal, a primary frequency modulation action signal, a primary frequency modulation input signal and a boiler coordination signal; the relevant operation data of the units in the corresponding DCS and DEH systems of the thermal power generating unit comprise main steam pressure, main steam pressure temperature, main steam flow, high-pressure regulating valve opening instructions and high-pressure regulating valve opening feedback.
3. The thermal power generating unit primary frequency modulation analysis and optimization method based on the BP neural network algorithm, according to claim 1, is characterized in that the 15S output response index, the 30S output response index and the electric quantity contribution index in the primary frequency modulation performance index assessment are respectively subjected to correlation calculation with factors possibly influencing the primary frequency modulation assessment performance, correlation magnitude sorting is performed according to the calculation result of the correlation coefficients, influence factors are determined according to the correlation sorting, and derivative variables of the influence factors are extracted according to the influence factors, and the method comprises the following processes:
correlation coefficient ρxyThe calculation of (2): one of the factors possibly influencing the primary frequency modulation performance is taken as a dependent variable X, one of the primary frequency modulation performance index evaluations is taken as an independent variable Y, and the calculation formula is
Wherein sigmaxIs the standard deviation, σ, of the dependent variableYIs the standard deviation of the independent variable, Cov [ X, Y]＝E[(X-u)(Y-v)]Is the covariance of the independent variable and the dependent variable, u is the expected value of the dependent variable, and v is the expected value of the independent variable;
respectively carrying out correlation calculation on a 15S output response index, a 30S output response index and an electric quantity contribution index in primary frequency modulation performance index assessment and factors possibly influencing the primary frequency modulation assessment performance, wherein the factors possibly influencing the primary frequency modulation assessment performance comprise deviation of a unit calculation frequency and a power grid assessment frequency, main steam pressure, rotating speed unequal rate, unit valve opening, a synchronous AGC (automatic gain control) adjusting direction, unit valve flow characteristics and the like; and sequencing according to the magnitude of the correlation, and determining five factors influencing the primary frequency modulation checking performance as the main steam pressure of the unit, the unequal rate of the rotating speed of the unit, the flow characteristic linearity of a valve of the unit, the frequency deviation of the unit and the isotropy of an AGC (automatic gain control) regulation instruction of the unit according to the first 5 factors of the sequencing result of the magnitude of the correlation.
4. The thermal power generating unit primary frequency modulation analysis and optimization method based on the BP neural network algorithm according to claim 1, wherein the derivative variables of the influencing factors comprise:
main steam pressure derivative variable: taking the ratio of the main steam pressure to the rated main steam pressure in the primary frequency modulation action assessment time period as a derivative variable for measuring the size of the main steam pressure;
the flow characteristic linearity derivative variable of the unit valve is as follows: taking the ratio of the percentage of the actual main steam flow variation in the primary frequency modulation action assessment time period to the percentage of the flow instruction variation in the same time period as a derivative variable for measuring the local flow characteristic of the unit;
unit frequency deviation derived variable: the frequency deviation for power grid examination is the difference between the rated frequency of the power grid and the frequency of a power grid examination reference point in a primary frequency modulation action examination time period; the frequency deviation corresponding to the response of the primary frequency modulation action of the unit is (load instruction after primary frequency modulation correction in the coordinated control system-load instruction before primary frequency modulation correction in the coordinated control system) × (50 × unit rotating speed unequal rate)/rated active power of the unit; taking the ratio of the frequency deviation responded by the unit primary frequency regulation action to the frequency deviation used for power grid examination as a derivative vector for measuring the frequency deviation measurement error;
the unit rotating speed unequal rate derivative variable is as follows: the unit rotation speed inequality rate is a set constant in the unit primary frequency modulation configuration logic, and the set range is 4% to 5%. The power grid evaluation standard is 5%, and the ratio of 5% to the rotating speed unequal rate actually set by the unit is used as a derivative variable for measuring the set rotating speed unequal rate of the unit;
unit AGC instruction isotropic derivative variable: and taking the vector ratio of the variation of the AGC command to the variation of the primary frequency modulation theoretical action in the primary frequency modulation action assessment time period as a derivative variable for measuring the isotropy of the AGC command and the primary frequency modulation action direction.
5. The thermal power generating unit primary frequency modulation analysis and optimization method based on the BP neural network algorithm according to claim 1, wherein the primary frequency modulation analysis and diagnosis system is constructed by using the BP neural network algorithm, pattern recognition is carried out on primary frequency modulation assessment performance index results, and system parameters are trained, and the method comprises the following processes:
and (3) taking the assessment data of the dispatching center and the data after the normalization processing of the derivative vectors as input, taking the assessment result of the primary frequency modulation performance index of the corresponding unit as output, respectively building a 15S output response index model, a 30S output response index model and an electric quantity contribution index model, and training the system to obtain the weight and the threshold of the system model.
6. The thermal power generating unit primary frequency modulation analysis and optimization method based on the BP neural network algorithm as claimed in claim 1, wherein the step of substituting the variant derived variables in the constraint range into the trained BP neural network model for iterative computation until the model output result is changed from unqualified to qualified comprises the following steps: the single or multiple derived variables are mutated one by one, and iterative calculation is carried out again by using an expert diagnosis system obtained by training until the primary frequency modulation performance index assessment result is changed from unqualified to qualified; when the primary frequency modulation performance index assessment result is changed from unqualified to qualified, the reason that the primary frequency modulation performance assessment index is unqualified is reversely analyzed according to the variation direction and degree of the derivative variable which is varied at the moment, and then a corresponding optimization strategy is obtained.
7. The thermal power generating unit primary frequency modulation analysis and optimization method based on the BP neural network algorithm as claimed in claim 6, wherein the specific process of one-by-one variation of single or multiple derived variables is as follows: the derived variable with the maximum correlation with the assessment index is mutated by setting the current value of the derived variable as an initial value alpha0If the set variation step is the single variation d, the value α of the derivative variable after the nth variation is obtainedn＝α0And + n + d, calculating again by using the value after the variation every time the variation is performed until the value is changedThe output result of the model is changed from unqualified to qualified or alphanThe value of (b) is outside the variation range; if α isnIf the value of the variable is up to the upper limit value of the variation range and the output result of the model is still unqualified, the derivative variables with the correlation smaller than the current derivative variable are sequentially selected to continue to be varied until the output result of the model is changed from unqualified to qualified or all the derivative variables are varied.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011011992.XA CN112149903B (en) | 2020-09-22 | 2020-09-22 | Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011011992.XA CN112149903B (en) | 2020-09-22 | 2020-09-22 | Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112149903A true CN112149903A (en) | 2020-12-29 |
| CN112149903B CN112149903B (en) | 2022-04-08 |
Family
ID=73896341
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011011992.XA Active CN112149903B (en) | 2020-09-22 | 2020-09-22 | Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112149903B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113595109A (en) * | 2021-09-06 | 2021-11-02 | 华北电力科学研究院有限责任公司 | Flywheel energy storage capacity configuration method and device |
| CN113779754A (en) * | 2021-08-02 | 2021-12-10 | 张家港宏昌钢板有限公司 | Method and system for analyzing influence factors of blast furnace, electronic device and computer-readable storage medium |
| CN113991700A (en) * | 2021-10-22 | 2022-01-28 | 中国大唐集团科学技术研究总院有限公司华东电力试验研究院 | Primary frequency modulation optimization decision method and device based on historical data |
| CN114565328A (en) * | 2022-04-29 | 2022-05-31 | 浙江中控技术股份有限公司 | AGC frequency modulation performance assessment method and device |
| CN116436160A (en) * | 2023-03-29 | 2023-07-14 | 华能威海发电有限责任公司 | AGC performance index on-line monitoring system and method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105445582A (en) * | 2015-11-26 | 2016-03-30 | 广东电网有限责任公司电力科学研究院 | Interconnection power grid primary frequency modulation responding performance assessment method |
| WO2016089475A1 (en) * | 2014-12-03 | 2016-06-09 | Saudi Arabian Oil Company | Energy performance metric in hydrocarbon-producing facilities |
| CN109638879A (en) * | 2018-12-14 | 2019-04-16 | 国网山东省电力公司电力科学研究院 | Primary frequency modulation dynamic compensation adjustment system and method based on performance indicator assessment |
| CN109784711A (en) * | 2019-01-08 | 2019-05-21 | 国网新疆电力有限公司电力科学研究院 | Fired power generating unit Primary frequency control ability assessment system, method, apparatus and storage medium |
| CN109784742A (en) * | 2019-01-23 | 2019-05-21 | 清华大学 | Unit primary frequency modulation performance comprehensive estimation method and system based on combination weights method |
| CN109858663A (en) * | 2018-11-19 | 2019-06-07 | 中国农业大学 | Distribution network failure power failure INFLUENCING FACTORS analysis based on BP neural network |
| CN109861310A (en) * | 2019-02-28 | 2019-06-07 | 上海电力学院 | Supercritical thermal power unit primary frequency control system recognizes Variable Selection |
| CN110765703A (en) * | 2019-11-04 | 2020-02-07 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Wind power plant aggregation characteristic modeling method |
| WO2020082973A1 (en) * | 2018-10-22 | 2020-04-30 | 新智数字科技有限公司 | Load prediction method and apparatus based on neural network |
| CN111396265A (en) * | 2019-01-03 | 2020-07-10 | 新疆金风科技股份有限公司 | Fault prediction method and device of wind generating set |
-
2020
- 2020-09-22 CN CN202011011992.XA patent/CN112149903B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016089475A1 (en) * | 2014-12-03 | 2016-06-09 | Saudi Arabian Oil Company | Energy performance metric in hydrocarbon-producing facilities |
| CN105445582A (en) * | 2015-11-26 | 2016-03-30 | 广东电网有限责任公司电力科学研究院 | Interconnection power grid primary frequency modulation responding performance assessment method |
| WO2020082973A1 (en) * | 2018-10-22 | 2020-04-30 | 新智数字科技有限公司 | Load prediction method and apparatus based on neural network |
| CN109858663A (en) * | 2018-11-19 | 2019-06-07 | 中国农业大学 | Distribution network failure power failure INFLUENCING FACTORS analysis based on BP neural network |
| CN109638879A (en) * | 2018-12-14 | 2019-04-16 | 国网山东省电力公司电力科学研究院 | Primary frequency modulation dynamic compensation adjustment system and method based on performance indicator assessment |
| CN111396265A (en) * | 2019-01-03 | 2020-07-10 | 新疆金风科技股份有限公司 | Fault prediction method and device of wind generating set |
| CN109784711A (en) * | 2019-01-08 | 2019-05-21 | 国网新疆电力有限公司电力科学研究院 | Fired power generating unit Primary frequency control ability assessment system, method, apparatus and storage medium |
| CN109784742A (en) * | 2019-01-23 | 2019-05-21 | 清华大学 | Unit primary frequency modulation performance comprehensive estimation method and system based on combination weights method |
| CN109861310A (en) * | 2019-02-28 | 2019-06-07 | 上海电力学院 | Supercritical thermal power unit primary frequency control system recognizes Variable Selection |
| CN110765703A (en) * | 2019-11-04 | 2020-02-07 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Wind power plant aggregation characteristic modeling method |
Non-Patent Citations (3)
| Title |
|---|
| TAT KEI CHAU等: "A Novel Control Strategy of DFIG Wind Turbines in Complex Power Systems for Enhancement of Primary Frequency Response and LFOD", 《IEEE TRANSACTIONS ON POWER SYSTEMS》 * |
| 张永军等: "火电机组一次调频性能分析及优化", 《浙江电力》 * |
| 殷建华等: "基于电网考核细则的火电机组一次调频优化", 《内蒙古电力技术》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779754A (en) * | 2021-08-02 | 2021-12-10 | 张家港宏昌钢板有限公司 | Method and system for analyzing influence factors of blast furnace, electronic device and computer-readable storage medium |
| CN113595109A (en) * | 2021-09-06 | 2021-11-02 | 华北电力科学研究院有限责任公司 | Flywheel energy storage capacity configuration method and device |
| CN113595109B (en) * | 2021-09-06 | 2024-01-19 | 华北电力科学研究院有限责任公司 | Flywheel energy storage capacity configuration method and device |
| CN113991700A (en) * | 2021-10-22 | 2022-01-28 | 中国大唐集团科学技术研究总院有限公司华东电力试验研究院 | Primary frequency modulation optimization decision method and device based on historical data |
| CN113991700B (en) * | 2021-10-22 | 2024-03-12 | 中国大唐集团科学技术研究总院有限公司华东电力试验研究院 | Primary frequency modulation optimization decision method and device based on historical data |
| CN114565328A (en) * | 2022-04-29 | 2022-05-31 | 浙江中控技术股份有限公司 | AGC frequency modulation performance assessment method and device |
| CN114565328B (en) * | 2022-04-29 | 2022-08-19 | 浙江中控技术股份有限公司 | AGC frequency modulation performance assessment method and device |
| CN116436160A (en) * | 2023-03-29 | 2023-07-14 | 华能威海发电有限责任公司 | AGC performance index on-line monitoring system and method |
| CN116436160B (en) * | 2023-03-29 | 2023-12-22 | 华能威海发电有限责任公司 | AGC performance index on-line monitoring system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112149903B (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112149903B (en) | Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm | |
| CN116757534B (en) | Intelligent refrigerator reliability analysis method based on neural training network | |
| CN108897286B (en) | Fault detection method based on distributed nonlinear dynamic relation model | |
| CN108803520B (en) | Dynamic process monitoring method based on variable nonlinear autocorrelation rejection | |
| CN111444953B (en) | Sensor fault monitoring method based on improved particle swarm optimization algorithm | |
| CN111159844B (en) | Abnormity detection method for exhaust temperature of gas turbine of power station | |
| CN107274067B (en) | Distribution transformer overload risk assessment method | |
| CN108919755B (en) | Distributed fault detection method based on multiple nonlinear cross relation models | |
| CN109740859A (en) | Transformer condition evaluation and system based on Principal Component Analysis and support vector machines | |
| CN108960309B (en) | Dynamic process monitoring method based on RBF neural network autocorrelation elimination | |
| CN110365053B (en) | Short-term wind power prediction method based on delay optimization strategy | |
| CN110530650B (en) | Method for monitoring performance state of heavy-duty gas turbine based on generalized regression neural network and box diagram analysis | |
| CN108492026B (en) | Soft measurement method based on integrated orthogonal component optimization regression analysis | |
| CN111652355A (en) | Method and device for predicting silicon content of blast furnace molten iron based on LSTM and DNN | |
| CN116146421B (en) | Intelligent control method and system based on fan state sensing | |
| Wang et al. | Deep-learning modeling and control optimization framework for intelligent thermal power plants: A practice on superheated steam temperature | |
| CN115903484A (en) | Cooperative game-based optimization method for multivariable robust controller of aircraft engine | |
| CN113095951B (en) | Intelligent gas extraction regulation and control method, equipment, device and storage medium | |
| CN114357870A (en) | Metering equipment operation performance prediction analysis method based on local weighted partial least squares | |
| CN102967393B (en) | Supercritical unit boiler separator outlet temperature on-line calibration method | |
| CN116452041A (en) | Power distribution network reliability assessment method based on cross entropy important sampling | |
| CN114384427B (en) | Quantum neural network-based lithium ion battery capacity nonlinear degradation prediction method | |
| CN114237043B (en) | Gas turbine equipment transfer function closed-loop identification method based on deep learning | |
| CN114139442A (en) | Method for optimizing steam turbine valve flow characteristic function based on K-means-LSTM | |
| CN111695581B (en) | Nonlinear dynamic process monitoring method based on autoregressive extreme learning machine |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211129 Address after: 010000 scientific research building of Inner Mongolia Electric Power Research Institute, 21 Xilin South Road, Yuquan District, Hohhot, Inner Mongolia Autonomous Region Applicant after: Inner Mongolia Electric Power Research Institute Branch of Inner Mongolia Electric Power (Group) Co.,Ltd. Applicant after: Chengdu Jiuyi Communication Technology Co., Ltd Address before: 010000 No.8 Xilin South Road, Yuquan District, Hohhot City, Inner Mongolia Autonomous Region Applicant before: Inner Mongolia Electric Power Research Institute Branch of Inner Mongolia Electric Power (Group) Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220120 Address after: 010000 Scientific Research Institute of Inner Mongolia Electric Power Research Institute, No. 21 Xilin South Road, Yuquan District, Hohhot, Inner Mongolia Autonomous Region Applicant after: Inner Mongolia Electric Power Research Institute Branch of Inner Mongolia Electric Power (Group) Co.,Ltd. Address before: 010000 scientific research building of Inner Mongolia Electric Power Research Institute, 21 Xilin South Road, Yuquan District, Hohhot, Inner Mongolia Autonomous Region Applicant before: Inner Mongolia Electric Power Research Institute Branch of Inner Mongolia Electric Power (Group) Co.,Ltd. Applicant before: Chengdu Jiuyi Communication Technology Co., Ltd |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |