CN116345495B - Power plant unit frequency modulation optimization method based on data analysis and modeling - Google Patents
Power plant unit frequency modulation optimization method based on data analysis and modeling Download PDFInfo
- Publication number
- CN116345495B CN116345495B CN202310368965.5A CN202310368965A CN116345495B CN 116345495 B CN116345495 B CN 116345495B CN 202310368965 A CN202310368965 A CN 202310368965A CN 116345495 B CN116345495 B CN 116345495B
- Authority
- CN
- China
- Prior art keywords
- optimal
- parameters
- parameter
- frequency modulation
- power plant
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 101
- 238000005457 optimization Methods 0.000 title claims abstract description 62
- 238000007405 data analysis Methods 0.000 title claims abstract description 21
- 230000008859 change Effects 0.000 claims abstract description 67
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 59
- 238000011217 control strategy Methods 0.000 claims abstract description 23
- 238000009826 distribution Methods 0.000 claims abstract description 22
- 239000002245 particle Substances 0.000 claims description 34
- 230000008569 process Effects 0.000 claims description 33
- 238000004364 calculation method Methods 0.000 claims description 27
- 239000013598 vector Substances 0.000 claims description 15
- 230000007423 decrease Effects 0.000 claims description 11
- 238000013139 quantization Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 6
- 230000003044 adaptive effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000010977 unit operation Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000005295 random walk Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
- H02J3/14—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by switching loads on to, or off from, network, e.g. progressively balanced loading
- H02J3/144—Demand-response operation of the power transmission or distribution network
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2310/00—The network for supplying or distributing electric power characterised by its spatial reach or by the load
- H02J2310/50—The network for supplying or distributing electric power characterised by its spatial reach or by the load for selectively controlling the operation of the loads
- H02J2310/56—The network for supplying or distributing electric power characterised by its spatial reach or by the load for selectively controlling the operation of the loads characterised by the condition upon which the selective controlling is based
- H02J2310/58—The condition being electrical
- H02J2310/60—Limiting power consumption in the network or in one section of the network, e.g. load shedding or peak shaving
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Feedback Control In General (AREA)
Abstract
The patent relates to a power plant unit frequency modulation optimization method based on data analysis and modeling. The method comprises the following steps: the method comprises the steps of collecting operation data of each unit in the power plant units, constructing a key factor weight distribution model, independently forming a set of each type of operation parameters of each unit, constructing a dimensional space body of the set corresponding to the operation data of the power plant units in a multidimensional space, finding out an optimal global solution under the dimensional space body through a gradual change simulation algorithm, taking the set of the operation parameters at the moment as a unit frequency modulation control optimal strategy, and applying the optimal control strategy to the operation of the power plant units to finish the frequency modulation optimization of the power plant units. According to the technical scheme, the frequency modulation control strategy of the power plant unit is optimized by utilizing the data analysis and modeling method, and the accuracy and efficiency of frequency modulation control are improved, so that more stable, efficient and intelligent power production is realized.
Description
Technical Field
The invention belongs to the technical field of power grid frequency modulation, and particularly relates to a power plant unit frequency modulation optimization method based on data analysis and modeling.
Background
With the rapid development of economy, the demand for electric power is increasing, and the improvement of electric power production efficiency by electric power production enterprises is urgent. The unit frequency modulation is a key link in power production, and the regulation control efficiency and the precision of the unit frequency modulation are directly related to the stability of a power grid and the quality of power supply. Therefore, the research of the frequency modulation control technology of the unit has important significance.
At present, the frequency modulation control technology of the unit is mainly divided into two types: traditional experience-based control methods and intelligent control methods based on data analysis and modeling. The traditional control method based on experience mainly depends on manual adjustment and experience summarization, the control effect is limited by personnel level and experience, and the control precision and efficiency are difficult to improve. The intelligent control method based on data analysis and modeling improves control precision and efficiency by analyzing and modeling the unit operation data, and can better adapt to the change and the requirement of power production.
In recent years, some patent documents relate to an intelligent unit frequency modulation control method based on data analysis and modeling. For example, patent document CN110801933a discloses a machine learning-based unit frequency modulation control method, which models and predicts operation data of a unit through a machine learning algorithm training model, optimizes a unit frequency modulation control strategy through a prediction result, and improves efficiency and accuracy of frequency modulation control. Patent document CN107081131a discloses a unit frequency modulation method based on model predictive control, which predicts the running state of a unit by using a model predictive algorithm and optimizes a control strategy according to a prediction result, thereby realizing more intelligent and efficient unit frequency modulation control.
However, there still exist some problems in the prior art, mainly including the following aspects:
incomplete data acquisition: the prior art mainly relies on the manual operation data who gathers the unit of power plant personnel, often has the incomplete condition of data acquisition. Such incomplete data can affect the accuracy of modeling and prediction, and thus the efficiency and accuracy of the unit frequency modulation control.
Limitations of modeling methods: the modeling method adopted in the prior art mainly comprises a neural network, a support vector machine and other methods, and the methods can solve the problem in the frequency modulation control of the unit to a certain extent, but have certain limitations in processing the problems of nonlinearity, high dimension, big data and the like. In addition, the training and adjustment of the methods consume a great deal of time and calculation resources, and cannot adapt to the frequency modulation control of the unit with high real-time requirements.
Control strategy singleness: the control strategy adopted in the prior art is always fixed and cannot adapt to the change of the running state of the unit. For example, when the unit load changes, the original control strategy may not meet the actual requirement, thereby affecting the frequency modulation control effect of the unit.
Convergence speed and search efficiency of the algorithm: algorithms adopted in the prior art, such as a simulated annealing algorithm, a genetic algorithm and the like, have certain problems in search efficiency and convergence speed, and are difficult to meet the actual requirements of unit frequency modulation control.
In summary, although the intelligent unit frequency modulation control method based on data analysis and modeling is adopted in the prior art, some problems still need to be solved. In order to improve efficiency and accuracy of frequency modulation control of a unit, more advanced technologies in aspects of data acquisition, modeling, control strategies, algorithms and the like need to be researched so as to meet actual requirements of power production. Meanwhile, the problems and challenges in the frequency modulation control of the unit are required to be continuously and deeply researched, and the development and application of the frequency modulation control technology of the unit are continuously promoted.
Disclosure of Invention
The invention mainly aims to provide the power plant unit frequency modulation optimization method based on data analysis and modeling, and the power plant unit frequency modulation optimization method based on data analysis and modeling can improve the accuracy and efficiency of power plant unit frequency modulation control, realize more intelligent and efficient unit frequency modulation control, and further improve the stability and reliability of power supply.
In order to solve the technical problems, the invention provides a power plant unit frequency modulation optimization method based on data analysis and modeling, which comprises the following steps:
a power plant unit frequency modulation optimization method based on data analysis and modeling comprises the following steps:
step S1: collecting operation data of each unit in the power plant units, wherein the types of the operation parameters at least comprise load, frequency, rotating speed and power;
step S2: constructing a key factor weight distribution model, wherein the key factor weight distribution model distributes weight values for each type of parameters in the operation parameters according to the influence of each type of parameters in the operation parameters on the frequency adjustment optimization result;
step S3: each type of operation parameters of each unit are independently formed into a set, and the set types comprise: load set, frequency set, rotation speed set and power set; each set is used as data of a certain dimension in a certain multidimensional space, and different sets correspond to different dimensions of the multidimensional space; constructing a dimension space body of a set corresponding to the operation data of the power plant unit in the multidimensional space according to the data in each set;
step S4: for each dimension space body, an optimal global solution under the dimension space body is found by setting a gradual change parameter and using a gradual change simulation algorithm, then a set of operation parameters corresponding to the optimal global solution is analyzed, and the set of operation parameters at the moment is used as a unit frequency modulation control optimal strategy;
Step S5: and applying the optimal control strategy to the operation of the power plant unit to finish the frequency modulation optimization of the power plant unit.
Further, in step 1, operation data of each unit in the power plant units is collected by means of sensors and/or unit control system software.
Further, the step S2 specifically includes:
step S2.1: preprocessing the collected operation data, including: denoising, normalizing and smoothing the data to generate preprocessing operation data;
step S2.2: the key factor weight distribution model is constructed, and the method specifically comprises the following steps: decomposing the frequency modulation optimization process of the power plant unit into two parts, wherein the first part is the type of the operation parameters, and the second part is the influence of each parameter on the frequency modulation optimization result; for each part element, pre-constructing a quantization table, and quantifying the comparison between the first part and the second part; calculating the weight of each part element to obtain a weight calculation result; consistency test is carried out on the weight calculation result to obtain a consistency ratio, if the consistency ratio is smaller than 0.1, the weight calculation result is reserved, otherwise, the weight settlement result is abandoned;
step S2.3: and distributing weights for each type of parameters in the operation parameters according to the weight calculation result.
Further, in the step S2.2, for each element of the portion, the pre-constructed quantization table is a standard table using a score of 1-9, which indicates the relative importance between the two elements, and the larger the score value, the higher the relative importance between the two elements.
Further, the method for calculating the weight of each partial element in step S2.2 to obtain the weight calculation result includes: firstly, constructing a judgment matrix A which represents comparison results among various operation parameters, wherein a ij Representing the importance of the ith parameter relative to the jth parameter:
then, a weight vector W of each parameter is calculated, wherein W i The degree of influence of the ith operating parameter on the frequency modulation optimization result is shown:
W＝(w 1 ,w 2 ,…,w n )
where n represents the number of operating parameters.
Further, the step S3 specifically includes:
step S3.1: firstly, combining each operation parameter set of each unit to obtain a vector containing all operation parameters; the dimension of the vector is equal to the number of kinds of operation parameters, and each dimension corresponds to one of a load set, a frequency set, a rotating speed set or a power set;
step S3.2: each vector is used as a point in a multi-dimensional space, and a multi-dimensional space body is constructed according to the positions of the points, wherein each point corresponds to an operation parameter set, and different operation parameter sets correspond to different dimensions in the multi-dimensional space.
Further, the step S4 specifically includes: controlling the movement of each point in the multi-dimensional space by determining a gradient parameter, wherein the higher the gradient parameter delta (t), the larger the movement of the point in the multi-dimensional space body, and the wider the search range of a gradient simulation algorithm; along with gradual reduction of the gradual change parameters, the movement of the point is gradually slowed down, and finally the point can be stabilized near a local optimal solution or a global optimal solution; the value of the fade parameter is calculated using the following formula:
wherein ,δ0 Representing the initial fade parameter, τ represents the search time constant, and t represents the current iteration number. As seen, as the number of iterations t increases, the fade parameter δ (t) gradually decreases, thereby controlling the decline of the fade parameter;
moving points in the multi-dimensional space; the moving mode of the point uses a random moving mode in a gradual change simulation algorithm, namely, randomly selecting a direction from the current position and moving a certain distance; the formula followed when the point moves is as follows:
x i (t+1)＝x i (t)+σ(t)·r i (t)；
wherein ,xi (t) represents the position of the point of the ith dimension at time t, σ (t) represents the moving distance at time t, r i (t) represents the random direction of the point of the ith dimension at time t;
in each iteration, selecting a moving distance sigma (t) to ensure that the algorithm can converge to an optimal solution and avoid sinking into a locally optimal solution; the travel distance is calculated using the following formula:
Wherein α represents a control factor of the moving distance; in the calculation formula of the moving distance, delta (t) represents the current gradual change parameter,representing the average distance a point in the multidimensional volume of space moves.
After each movement, judging whether the new position is better than the current position; if the new position is more optimal, updating the current position to the new position; otherwise, accepting the new position with a certain probability; the probability of accepting a new position is given by the following formula:
where Δe represents the energy difference between the new position and the current position, and δ (t) represents the current fade parameter. It is seen that when the energy of the new location is lower, the probability of accepting the new location is 1; when the energy of the new position is higher, the probability of accepting the new position gradually decreases along with the decrease of the gradual change parameter; repeating the point movement until an optimal global solution in the multi-dimensional space body is found; and then analyzing a set of operation parameters corresponding to the optimal global solution, and taking the set of operation parameters at the moment as an optimal strategy for unit frequency modulation control.
Further, the determining process of the gradual change parameter includes: and randomly generating an initial gradual change parameter, and recording the optimal solution and the optimal fitness under the current gradual change parameter. Probing new gradual change parameters: on the basis of the current gradient parameters, randomly generating a new gradient parameter for simulating the searching process of a gradient simulation algorithm; searching a gradual change simulation algorithm by using the new gradual change parameters, calculating the fitness of a search result, and comparing the fitness with the current optimal solution; correcting gradual change parameters: if the adaptability of the new gradient parameter is better than the current optimal solution, taking the new gradient parameter as the current gradient parameter, and updating the optimal solution and the optimal adaptability; otherwise, accept the new gradual change parameter with certain probability, go on the heuristic.
Further, the method for determining the moving distance includes: randomly generating an initial moving distance, and recording an optimal solution and an optimal fitness under the current moving distance; in the searching process, dynamically adjusting the value of the moving distance according to the current searching result and the moving distance; when the adaptability change is small, the moving distance is reduced, so that the searching precision is enhanced; when the adaptability change is large, the moving distance is increased so as to enlarge the searching range; the process of dynamically adjusting the movement distance is repeated until an optimal movement distance is found.
Further, the method for determining the control factor includes: randomly generating a set of initial particle groups in a parameter space, each particle representing a set of control factors; performing unit frequency modulation control by using a control factor of a current particle swarm, and calculating the adaptability of a control result; updating the speed and the position of each particle according to the formula of the particle swarm optimization algorithm; recording the optimal solution and the optimal fitness of each particle; the process of updating speed and position is repeated until an optimal control factor is found.
The power plant unit frequency modulation optimization method based on data analysis and modeling has the following beneficial effects:
1. The accuracy of frequency modulation control of the power plant unit is improved: according to the method, the operation data of the power plant unit is analyzed and modeled by utilizing a data analysis and modeling method, so that the unit frequency modulation control strategy is optimized, the frequency modulation control precision is improved, and the stability and reliability of power supply are improved.
2. The efficiency of frequency modulation control of the power plant unit is improved: by independently forming a set of each type of operation parameters of each unit, constructing a dimension space body of the set corresponding to the operation data of the power plant unit in the multidimensional space, and finding out the optimal global solution in the dimension space body by adopting a gradual change simulation algorithm, the frequency modulation control strategy of the power plant unit is optimized, and the frequency modulation control efficiency is improved.
3. Realize more intelligent and efficient unit frequency modulation control: the method not only can construct a key factor weight distribution model by collecting the operation data of each unit in the power plant units, but also can apply an optimal control strategy to the operation of the power plant units, thereby realizing more intelligent and efficient unit frequency modulation control.
In summary, the power plant unit frequency modulation optimization method based on data analysis and modeling provided by the patent not only can improve the accuracy and efficiency of frequency modulation control, but also can realize more intelligent and efficient unit frequency modulation control, thereby having important practical application value and social and economic benefits.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings which are required to be used in the description of the embodiments or the prior art will be briefly described below, it being obvious that the drawings in the description below are only embodiments of the invention and that other drawings are obtained according to the drawings provided without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a power plant unit frequency modulation optimization method based on data analysis and modeling according to an embodiment of the present invention.
Detailed Description
The method of the present invention will be described in further detail with reference to the accompanying drawings.
Example 1
A power plant unit frequency modulation optimization method based on data analysis and modeling comprises the following steps:
step S1: collecting operation data of each unit in the power plant units, wherein the types of the operation parameters at least comprise load, frequency, rotating speed and power;
the types of the operation parameters collected in the step S1 comprise load, frequency, rotating speed and power, and the parameters are key factors for influencing the frequency modulation optimization of the power plant unit. The load is a measure of the output electric energy of the unit of the power plant, the frequency is the frequency of the power grid, and the rotating speed and the power reflect the rotating state of the rotor of the unit and the output power. By collecting the parameters, the running state of the power plant unit can be comprehensively known, and reliable data support is provided for frequency modulation optimization in the subsequent steps.
Step S2: constructing a key factor weight distribution model, wherein the key factor weight distribution model distributes weight values for each type of parameters in the operation parameters according to the influence of each type of parameters in the operation parameters on the frequency adjustment optimization result;
the key factor weight distribution model constructed in the step S2 can distribute weight for each type of parameters in the operation parameters, so that the influence of each type of parameters on the frequency modulation optimization result is determined. This step is one of the cores of the whole method, and the construction of the key factor weight distribution model plays a vital role in the subsequent dimension space construction and optimization algorithm selection.
Step S3: each type of operation parameters of each unit are independently formed into a set, and the set types comprise: load set, frequency set, rotation speed set and power set; each set is used as data of a certain dimension in a certain multidimensional space, and different sets correspond to different dimensions of the multidimensional space; constructing a dimension space body of a set corresponding to the operation data of the power plant unit in the multidimensional space according to the data in each set;
in the step S3, each operation parameter is independently formed into a set, and each set is used as one dimension data in the multi-dimension space, so that the operation state of the power plant unit can be more intuitively represented in the multi-dimension space. Through the implementation of the step, different operation parameters can be mapped into different dimensions in a multidimensional space, so that a complete dimensional space body is formed, and the implementation of a subsequent optimization algorithm is facilitated.
Step S4: for each dimension space body, an optimal global solution under the dimension space body is found by setting a gradual change parameter and using a gradual change simulation algorithm, then a set of operation parameters corresponding to the optimal global solution is analyzed, and the set of operation parameters at the moment is used as a unit frequency modulation control optimal strategy;
the gradient simulation algorithm employed in step S4 is an optimization algorithm that searches for an optimal solution in the dimensional space by setting gradient parameters and a moving distance in the dimensional space. The algorithm has the advantages that the optimal solution can be efficiently searched in a large-scale dimension space, and the optimization result can be gradually adjusted by controlling the change rule of the gradual change parameters. Specifically, the gradual change parameter controls probability change in the optimization process, the moving distance controls step length in the optimization process, and the control factor is used for balancing the relation between the gradual change parameter and the moving distance. And through the step S4, an optimal strategy of unit frequency modulation control can be obtained, and the power plant unit frequency modulation optimization is realized.
Step S5: the optimal control strategy is applied to the operation of the power plant unit to finish the frequency modulation optimization of the power plant unit, and specifically comprises the following steps:
Step S5.1: the optimal control strategy obtained in the step S4 is applied to frequency modulation control of the power plant unit, and the optimal control effect is achieved by adjusting the load of the unit in real time;
step S5.2: in the process of applying the optimal control strategy, monitoring the running state of the unit and recording the value of each running parameter in real time;
step S5.3: by comparing with actual operation data, the effect of the optimal control strategy is estimated, and optimization and adjustment are performed;
step S5.4: in the running process of the unit, the optimal control strategy is continuously updated according to the data monitored in real time so as to adapt to different running conditions and load requirements;
step S5.5: and carrying out long-term monitoring and data analysis on the unit, collecting historical operation data, and predicting and optimizing the frequency modulation control effect of the unit by establishing a machine learning model so as to improve the frequency modulation control efficiency and stability of the unit.
In experiments, the method in the patent is applied to frequency modulation optimization of an actual power plant unit, and a series of experiments and comparative analysis are performed.
Firstly, the data acquisition and key factor weight distribution model of the method is optimized, and the acquisition precision of the data and the accuracy of the key factor weight are improved, so that a foundation is provided for the follow-up steps.
Secondly, in experiments, the method is compared with a traditional experience model, a neural network model and a support vector machine model, and the results show that the method is superior to other models in the aspect of frequency modulation optimization of a power plant unit, and frequency modulation efficiency and stability are improved.
In addition, a parameter adjustment experiment is carried out, the gradual change parameter, the moving distance and the control factor are optimally adjusted, and the optimal parameter range is determined by comparing experimental results, so that the frequency modulation optimization effect and stability are improved.
Finally, in the experiment, the comparison and analysis are carried out on different units, and the result shows that the method has strong applicability and generalization and can be effectively applied to frequency modulation optimization of power plant units of different types and scales.
Thus, the benefits of this patent include, but are not limited to: the frequency modulation efficiency and stability of the power plant unit are improved, the frequency modulation control strategy is optimized, the running efficiency of the unit is improved, the energy consumption is reduced, and the power plant unit has strong applicability and generalization.
In the implementation of step 1, the operating data of each of the power plant units may be acquired using sensors and/or unit control system software. The sensor can acquire operation data by sensing various physical quantities, such as load, frequency, rotating speed, power and other parameters, and the unit control system software can acquire the operation data through the unit monitoring system. By adopting the two modes, the operation data of the unit can be effectively obtained, and the accuracy and reliability of frequency modulation optimization of the power plant unit are improved. In practical application, the acquisition mode and specific sensor or control system software can be flexibly selected according to the specific conditions and technical requirements of the unit.
The step S2 specifically comprises the following steps:
step S2.1: preprocessing the collected operation data, including: denoising, normalizing and smoothing the data to generate preprocessing operation data; the preprocessing operation in step S2.1 is to improve the data quality, including removing noise in the data, normalizing the data to a uniform range to avoid order of magnitude differences between different parameters, and smoothing the data to eliminate abrupt changes in the data. These operations help to improve the accuracy and stability of the model.
Step S2.2: the key factor weight distribution model is constructed, and the method specifically comprises the following steps: decomposing the frequency modulation optimization process of the power plant unit into two parts, wherein the first part is the type of the operation parameters, and the second part is the influence of each parameter on the frequency modulation optimization result; for each part element, pre-constructing a quantization table, and quantifying the comparison between the first part and the second part; calculating the weight of each part element to obtain a weight calculation result; consistency test is carried out on the weight calculation result to obtain a consistency ratio, if the consistency ratio is smaller than 0.1, the weight calculation result is reserved, otherwise, the weight settlement result is abandoned;
Constructing a key factor weight distribution model is the core of the whole algorithm. The frequency modulation optimization process of the power plant unit is decomposed into two parts, and the elements of each part are quantified, so that the weight of each element can be obtained. The first part refers to the type of the operation parameters, and the second part refers to the influence of each parameter on the frequency modulation optimization result. The pre-constructed quantization table can make the comparison more operable and repeatable, making the weight calculation more accurate. When consistency test is carried out, if the consistency ratio is smaller than 0.1, the weight calculation result is reserved, otherwise, the weight is required to be readjusted so as to ensure the accuracy and consistency of the weight calculation.
Step S2.3: and distributing weights for each type of parameters in the operation parameters according to the weight calculation result. And distributing weights for each type of parameters in the operation parameters according to the weight calculation result. These weights can be used in subsequent steps to construct a multidimensional volume as weights for each set under the multidimensional space. Therefore, when the optimal solution is found, the influence of different parameters can be adjusted according to the weight of the optimal solution, so that the more accurate optimal solution is obtained.
In step S2.2, for each element of the portion, the pre-constructed quantization table is a standard table using a score of 1-9, which indicates a relative importance between the two elements, and the larger the score value, the higher the relative importance between the two elements. Such a quantization table may help us to more accurately determine the weight value of each partial element in the key factor weight distribution model, thereby providing a reliable basis for subsequent operations.
The method for calculating the weight of each partial element in the step S2.2 to obtain the weight calculation result comprises the following steps: firstly, constructing a judgment matrix A which represents comparison results among various operation parameters, wherein a ij Representing the importance of the ith parameter relative to the jth parameter:
then, a weight vector W of each parameter is calculated, wherein W i The degree of influence of the ith operating parameter on the frequency modulation optimization result is shown:
W＝(w 1 ,w 2 ,…,w n )
where n represents the number of operating parameters.
In step S2.2 of the present invention, in order to determine the extent to which each operating parameter affects the frequency modulation optimization result, a judgment matrix a needs to be constructed first. Judging each element a in matrix A ij Representing the importance of the ith parameter relative to the jth parameter, a scale of 1-9 scores is used to represent the relative importance between the two elements, with a greater score indicating a higher relative importance between the two elements.
Then, the weight vector W of each parameter is calculated from the judgment matrix a. Wherein w is i And the influence degree of the ith operation parameter on the frequency modulation optimization result is represented. The specific calculation mode is as follows: for each parameter i, the result of the comparison between it and the other parameters is calculated, i.e. a is calculated ij (i≠j) Then all a ij Dividing the sum of (i.noteq.j) by the sum of all elements in the judgment matrix A to obtain the ith element W in the weight vector W i 。
The weight calculation method carries out standardized processing on the elements in the judgment matrix A, and avoids inaccuracy of weight calculation results caused by different dimensions or scoring standards. Meanwhile, the correlations between the respective parameters are also considered in calculating the weight vector W, rather than simply considering them as mutually independent factors. Therefore, by means of the weight calculation in step S2.2, the degree of influence of each operating parameter on the frequency modulation optimization result can be determined more accurately, so that each operating parameter is optimized more specifically in the subsequent frequency modulation optimization.
The step S3 specifically includes:
step S3.1: firstly, combining each operation parameter set of each unit to obtain a vector containing all operation parameters; the dimension of the vector is equal to the number of kinds of operation parameters, and each dimension corresponds to one of a load set, a frequency set, a rotating speed set or a power set;
step S3.2: each vector is used as a point in a multi-dimensional space, and a multi-dimensional space body is constructed according to the positions of the points, wherein each point corresponds to an operation parameter set, and different operation parameter sets correspond to different dimensions in the multi-dimensional space.
The multidimensional space in the step S3.2 is an n-dimensional cube, wherein n is the number of types of unit operation parameters. The side length of the cube is equal to the maximum minus the minimum of the data ranges in each set, and the coordinates of each point represent the operating parameter values in the respective dimensions. In the process of constructing the multi-dimensional space body, a data analysis and modeling method can be used for determining the relation among the dimensions, so that the number of dimensions is reduced, and the complexity of the multi-dimensional space body is reduced. Specifically, the original operating parameters may be converted into principal components using a Principal Component Analysis (PCA) method to reduce the number of dimensions and the effect of data noise, thereby improving the efficiency of the gradient simulation algorithm and the accuracy of the optimization result. In addition, the discrete data in the operation parameters can be converted into continuous functions by using an interpolation method, so that the distribution condition of the operation parameters in the multi-dimensional space can be described more accurately, and the convergence speed of the gradual change simulation algorithm and the stability of an optimization result are improved.
The step S4 specifically comprises the following steps: controlling the movement of each point in the multi-dimensional space by determining a gradient parameter, wherein the higher the gradient parameter delta (t), the larger the movement of the point in the multi-dimensional space body, and the wider the search range of a gradient simulation algorithm; along with gradual reduction of the gradual change parameters, the movement of the point is gradually slowed down, and finally the point can be stabilized near a local optimal solution or a global optimal solution; the value of the fade parameter is calculated using the following formula:
wherein ,δ0 Representing the initial fade parameter, τ represents the search time constant, and t represents the current iteration number. As seen, as the number of iterations t increases, the fade parameter δ (t) gradually decreases, thereby controlling the decline of the fade parameter;
moving points in the multi-dimensional space; the moving mode of the point uses a random moving mode in a gradual change simulation algorithm, namely, randomly selecting a direction from the current position and moving a certain distance; the formula followed when the point moves is as follows:
x i (t+1)＝x i (t)+σ(t)·r i (t)；
wherein ,xi (t) represents the position of the point of the ith dimension at time t, σ (t) represents the moving distance at time t, r i (t) represents the random direction of the point of the ith dimension at time t;
in each iteration, selecting a moving distance sigma (t) to ensure that the algorithm can converge to an optimal solution and avoid sinking into a locally optimal solution; the travel distance is calculated using the following formula:
wherein α represents a control factor of the moving distance; in the calculation formula of the moving distance, delta (t) represents the current gradual change parameter,representing the average distance a point in the multidimensional volume of space moves.
After each movement, judging whether the new position is better than the current position; if the new position is more optimal, updating the current position to the new position; otherwise, accepting the new position with a certain probability; the probability of accepting a new position is given by the following formula:
Where Δe represents the energy difference between the new position and the current position, and δ (t) represents the current fade parameter. It is seen that when the energy of the new location is lower, the probability of accepting the new location is 1; when the energy of the new position is higher, the probability of accepting the new position gradually decreases along with the decrease of the gradual change parameter; repeating the point movement until an optimal global solution in the multi-dimensional space body is found; and then analyzing a set of operation parameters corresponding to the optimal global solution, and taking the set of operation parameters at the moment as an optimal strategy for unit frequency modulation control.
In step S4, a gradient simulation algorithm is used to find the optimal global solution in the multidimensional space. The algorithm can perform random walk in the search space and accept a non-optimal solution mode through a certain probability, so that the problem of sinking into a local optimal solution is avoided. Meanwhile, the gradual change parameters are used for controlling the searching process, and the step length can be gradually reduced in the searching process, so that the convergence speed and the accuracy of the algorithm are effectively improved. The optimal global solution found by the algorithm can be used as an optimal strategy for unit frequency modulation control, so that the aim of optimizing the power plant unit frequency modulation control is fulfilled.
The gradual change parameter determining process comprises the following steps: and randomly generating an initial gradual change parameter, and recording the optimal solution and the optimal fitness under the current gradual change parameter. Probing new gradual change parameters: on the basis of the current gradient parameters, randomly generating a new gradient parameter for simulating the searching process of a gradient simulation algorithm; searching a gradual change simulation algorithm by using the new gradual change parameters, calculating the fitness of a search result, and comparing the fitness with the current optimal solution; correcting gradual change parameters: if the adaptability of the new gradient parameter is better than the current optimal solution, taking the new gradient parameter as the current gradient parameter, and updating the optimal solution and the optimal adaptability; otherwise, accept the new gradual change parameter with certain probability, go on the heuristic.
To ensure that a globally optimal solution is found, a number of different initial fade parameters may be used and a multiple fade simulation algorithm run. For each run, the optimal solution and optimal fitness may be recorded and compared to the results of other runs. Finally, the result with the optimal fitness may be selected as the optimal solution.
When determining the fade parameters, adaptive fade parameter algorithms may also be used to dynamically adjust the fade parameters based on real-time information in the search process. The idea of the adaptive fade parameter algorithm is that if the search process is performed in a small range, the fade parameters should be small; if the search process is performed over a large range, the fade parameters should be large. According to this idea, the fade parameters may be dynamically adjusted using some adaptive algorithm, such as an adaptive dynamic parameter algorithm or an adaptive linear parameter algorithm.
In addition, some skill may be used to improve search efficiency when performing the gradient simulation algorithm. For example, local search techniques may be used to search locally for the current optimal solution during the search to find a better solution. Parallel search techniques may also be used while running a fade simulation algorithm under multiple fade parameters to speed up the search process.
The method for determining the moving distance comprises the following steps: randomly generating an initial moving distance, and recording an optimal solution and an optimal fitness under the current moving distance; in the searching process, dynamically adjusting the value of the moving distance according to the current searching result and the moving distance; when the adaptability change is small, the moving distance is reduced, so that the searching precision is enhanced; when the adaptability change is large, the moving distance is increased so as to enlarge the searching range; the process of dynamically adjusting the movement distance is repeated until an optimal movement distance is found.
More specifically, the dynamic adjustment method of the moving distance may employ an adaptive random search strategy (Adaptive Random Search, ARS). First, an initial movement distance is randomly generated and used as a start value of the search. In each searching process, the value of the moving distance is dynamically adjusted according to the current searching result and the moving distance, so as to achieve the purpose of balancing the searching range and the searching precision.
Specifically, the ARS strategy comprises the steps of:
selecting an initial movement distance: an initial movement distance is randomly generated and used as a starting value for the search.
Searching: searching is performed in the multi-dimensional space according to the current moving distance to find a better solution. And in the searching process, recording the current optimal solution and the optimal fitness.
Dynamically adjusting the moving distance: and dynamically adjusting the value of the moving distance according to the current search result and the moving distance. Specifically, if the search result is better than the current optimal solution, the moving distance is reduced by a certain proportion so as to strengthen the searching precision; conversely, if the search result becomes worse, the moving distance is increased by a certain proportion to expand the scope of the search. The adjustment ratio of the moving distance can be set according to the actual application scene, and is generally recommended to be between 0.5 and 0.9.
Repeating the step 2-3 until the optimal moving distance is found. The optimal movement distance refers to a movement distance that can reach an optimal solution and that is efficient in search. In the searching process, a certain iteration number or searching time is required to be set so as to ensure that the algorithm can find the optimal solution or approach the optimal solution in a limited time.
The moving distance is dynamically adjusted through the ARS strategy, so that the problem that an algorithm falls into a local optimal solution due to too small or too large searching range can be avoided while the searching precision is ensured, and the searching efficiency and the searching accuracy are improved.
The method for determining the control factor comprises the following steps: randomly generating a set of initial particle groups in a parameter space, each particle representing a set of control factors; performing unit frequency modulation control by using a control factor of a current particle swarm, and calculating the adaptability of a control result; updating the speed and the position of each particle according to the formula of the particle swarm optimization algorithm; recording the optimal solution and the optimal fitness of each particle; the process of updating speed and position is repeated until an optimal control factor is found.
The method comprises the following specific steps: initializing a particle swarm: a set of particles is randomly generated in the parameter space, each particle representing a set of control factors. Typically, the initial position and velocity of the particles are randomly generated.
Evaluating particle fitness: and performing unit frequency modulation control by using the control factors of the current particle swarm, and calculating the adaptability of the control result. The definition of fitness is typically related to an objective function, which refers to a function that needs to be optimized, such as minimizing frequency modulation errors or maximizing power generation efficiency, etc.
Updating particle velocity and position: and updating the speed and the position of each particle according to the formula of the particle swarm optimization algorithm. Specifically, the velocity and position of each particle is calculated from its historical optimal position, population optimal position, and current position. The update formula for speed and position is as follows:
v i (t+1)＝wv i (t)+c1r1*(pbest i -x i (t))+c2r2(gbest-x i (t))；
x i (t+1)＝x i (t)+v i (t+1)；
wherein ,vi (t) is the velocity of the ith particle at time t, x i (t) is the position of the ith particle at time t, w is the inertial weight, c1 and c2 are the acceleration constants, r1 and r2 are random numbers between 0 and 1, pbest i Is the historical optimal position of the ith particle, and gbest is the optimal position of the whole particle swarm. Recording particle history Optimal position and optimal fitness: each particle will record its own historical optimal location and optimal fitness. And if the fitness of the current position is better than that of the historical optimal position, updating the historical optimal position and the optimal fitness.
The process of updating speed and position is repeated until an optimal control factor is found. The optimal control factor is a combination of factors that enables the objective function to take a minimum or maximum value.
In summary, the invention provides a method for optimizing frequency modulation of a power plant unit, which adopts a related gradual change simulation algorithm and a key factor weight distribution model.
The gradual change simulation algorithm is a global optimization algorithm, searches for a global optimal solution through random movement of points in a multidimensional space, has the characteristics of wide search range, high convergence speed and the like, and is suitable for the optimization problem of complex multidimensional. In the invention, the unit frequency modulation problem is abstracted into an optimization problem in a multidimensional space, and the optimal solution is searched by adopting a gradual change simulation algorithm, so that an optimal unit frequency modulation control strategy is obtained.
The key factor weight distribution model is used for decomposing the unit frequency modulation optimization problem into comparison problems among all operation parameters, quantifying the importance of the operation parameters in a quantization table mode, and finally obtaining the weight of each parameter, so that a basis is provided for the formulation of a unit frequency modulation control strategy. In the invention, the key factor weight distribution model is applied to the frequency modulation problem of the unit, thereby realizing the quantitative analysis of each operation parameter in the frequency modulation of the unit and being beneficial to formulating a more accurate frequency modulation control strategy.
Therefore, the gradual change simulation algorithm and the key factor weight distribution model are related to each other, and the gradual change simulation algorithm and the key factor weight distribution model are jointly applied to frequency modulation optimization of the power plant unit so as to realize more efficient and accurate frequency modulation control strategy formulation.
While specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are by way of example only, and that various omissions, substitutions, and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the above-described method steps to perform substantially the same function in substantially the same way to achieve substantially the same result. Accordingly, the scope of the invention is limited only by the following claims.
Claims (9)
1. The power plant unit frequency modulation optimization method based on data analysis and modeling is characterized by comprising the following steps of:
step S1: collecting operation data of each unit in the power plant units, wherein the types of the operation data at least comprise load, frequency, rotating speed and power;
step S2: constructing a key factor weight distribution model, wherein the key factor weight distribution model distributes weight values for each type of parameters in the operation parameters according to the influence of each type of parameters in the operation parameters on the frequency adjustment optimization result;
Step S3: each type of operation parameters of each unit are independently formed into a set, and the set types comprise: load set, frequency set, rotation speed set and power set; each set is used as data of a certain dimension in a certain multidimensional space, and different sets correspond to different dimensions of the multidimensional space; constructing a dimension space body of a set corresponding to the operation data of the power plant unit in the multidimensional space according to the data in each set;
step S4: for each dimension space body, an optimal global solution under the dimension space body is found by setting a gradual change parameter and using a gradual change simulation algorithm, then a set of operation parameters corresponding to the optimal global solution is analyzed, and the set of operation parameters at the moment is used as a unit frequency modulation control optimal strategy;
step S5: the optimal control strategy is applied to the operation of the power plant unit, so that the frequency modulation optimization of the power plant unit is completed;
the step S4 specifically includes: controlling the movement of each point in a multi-dimensional space by determining a fade parameterThe higher the motion of the points in the multi-dimensional space body is, the wider the search range of the gradual change simulation algorithm is; along with gradual reduction of the gradual change parameters, the movement of the point is gradually slowed down, and finally the point can be stabilized near a local optimal solution or a global optimal solution; the value of the fade parameter is calculated using the following formula:
；
wherein ,representing the initial fade parameters, +.>Representing search time constant, ++>Representing the current iteration number, it is seen that as the iteration number +.>Increase, gradual change parameter->Gradually decreasing, thereby controlling the decrease of the gradual change parameter;
moving points in the multi-dimensional space; the moving mode of the point uses a random moving mode in a gradual change simulation algorithm, namely, randomly selecting a direction from the current position and moving a certain distance; the formula followed when the point moves is as follows:
；
wherein ,representation ofFirst->The point of each dimension is->Position of moment->Is indicated at->Distance of movement of time,/->Indicate->The point of each dimension is->A random direction of time;
in each iteration, a distance of movement is selectedTo ensure that the algorithm can converge to the optimal solution and avoid sinking into the local optimal solution; the travel distance is calculated using the following formula:
；
wherein ,a control factor representing a movement distance; in the calculation formula of the moving distance, < >>Representing the current fade parameters, +.>Representing the average distance a point in the multidimensional space moves;
after each movement, judging whether the new position is better than the current position; if the new position is more optimal, updating the current position to the new position; otherwise, accepting the new position with a certain probability; the probability of accepting a new position is given by the following formula:
；
wherein ,representing the energy difference between the new position and the current position,/->Representing the current gradual change parameter; it is seen that when the energy of the new location is lower, the probability of accepting the new location is 1; when the energy of the new position is higher, the probability of accepting the new position gradually decreases along with the decrease of the gradual change parameter; repeating the point movement until an optimal global solution in the multi-dimensional space body is found; and then analyzing a set of operation parameters corresponding to the optimal global solution, and taking the set of operation parameters at the moment as an optimal strategy for unit frequency modulation control.
2. The method according to claim 1, wherein in step 1, operation data of each of the power plant units is collected by means of sensors and/or unit control system software.
3. The method according to claim 1, wherein the step S2 specifically includes:
step S2.1: preprocessing the collected operation data, including: denoising, normalizing and smoothing the data to generate preprocessing operation data;
step S2.2: the key factor weight distribution model is constructed, and the method specifically comprises the following steps: decomposing the frequency modulation optimization process of the power plant unit into two parts, wherein the first part is the type of the operation parameters, and the second part is the influence of each parameter on the frequency modulation optimization result; for each part element, pre-constructing a quantization table, and quantifying the comparison between the first part and the second part; calculating the weight of each part element to obtain a weight calculation result; consistency test is carried out on the weight calculation result to obtain a consistency ratio, if the consistency ratio is smaller than 0.1, the weight calculation result is reserved, otherwise, the weight settlement result is abandoned;
Step S2.3: and distributing weights for each type of parameters in the operation parameters according to the weight calculation result.
4. A method according to claim 3, wherein the pre-constructed quantization table for each partial element in step S2.2 is a standard table using a score of 1-9, indicating a relative importance between the two elements, the larger the score value, indicating a higher relative importance between the two elements.
5. The method according to claim 4, wherein the step S2.2 of calculating the weight of each partial element to obtain the weight calculation result includes: first, a judgment matrix is constructedRepresenting the comparison result between the individual operating parameters, wherein +.>Indicate->The parameter is relative to the (th)>Importance of the individual parameters:
；
then, weight vectors of the respective parameters are calculated, wherein />Indicate->Degree of influence of individual operating parameters on the frequency-adjusting optimization result:
；
wherein ,representing the number of operating parameters.
6. The method according to claim 5, wherein the step S3 specifically includes:
step S3.1: firstly, combining each operation parameter set of each unit to obtain a vector containing all operation parameters; the dimension of the vector is equal to the number of kinds of operation parameters, and each dimension corresponds to one of a load set, a frequency set, a rotating speed set or a power set;
Step S3.2: each vector is used as a point in a multi-dimensional space, and a multi-dimensional space body is constructed according to the positions of the points, wherein each point corresponds to an operation parameter set, and different operation parameter sets correspond to different dimensions in the multi-dimensional space.
7. The method of claim 6, wherein the determining of the fade parameter comprises: randomly generating an initial gradual change parameter, and recording an optimal solution and optimal fitness under the current gradual change parameter; probing new gradual change parameters: on the basis of the current gradient parameters, randomly generating a new gradient parameter for simulating the searching process of a gradient simulation algorithm; searching a gradual change simulation algorithm by using the new gradual change parameters, calculating the fitness of a search result, and comparing the fitness with the current optimal solution; correcting gradual change parameters: if the adaptability of the new gradient parameter is better than the current optimal solution, taking the new gradient parameter as the current gradient parameter, and updating the optimal solution and the optimal adaptability; otherwise, accept the new gradual change parameter with certain probability, go on the heuristic.
8. The method of claim 7, wherein the method of determining the distance of movement comprises: randomly generating an initial moving distance, and recording an optimal solution and an optimal fitness under the current moving distance; in the searching process, dynamically adjusting the value of the moving distance according to the current searching result and the moving distance; when the adaptability change is smaller than the set value, the moving distance is reduced so as to strengthen the searching precision; when the adaptability change is larger than the set value, increasing the moving distance to expand the searching range; the process of dynamically adjusting the movement distance is repeated until an optimal movement distance is found.
9. The method of claim 8, wherein the method of determining the control factor comprises: randomly generating a set of initial particle groups in a parameter space, each particle representing a set of control factors; performing unit frequency modulation control by using a control factor of a current particle swarm, and calculating the adaptability of a control result; updating the speed and the position of each particle according to the formula of the particle swarm optimization algorithm; recording the optimal solution and the optimal fitness of each particle; the process of updating speed and position is repeated until an optimal control factor is found.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310368965.5A CN116345495B (en) | 2023-04-03 | 2023-04-03 | Power plant unit frequency modulation optimization method based on data analysis and modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310368965.5A CN116345495B (en) | 2023-04-03 | 2023-04-03 | Power plant unit frequency modulation optimization method based on data analysis and modeling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116345495A CN116345495A (en) | 2023-06-27 |
| CN116345495B true CN116345495B (en) | 2023-10-10 |
Family
ID=86880461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310368965.5A Active CN116345495B (en) | 2023-04-03 | 2023-04-03 | Power plant unit frequency modulation optimization method based on data analysis and modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116345495B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105305468A (en) * | 2015-10-19 | 2016-02-03 | 国家电网公司 | Particle swarm optimization algorithm based thermal generator set primary frequency modulation parameter optimization method |
| CN107655463A (en) * | 2017-09-21 | 2018-02-02 | 大连理工大学 | Electronic compass calibration method based on simulated annealing |
| CN109390969A (en) * | 2018-10-11 | 2019-02-26 | 国网江苏省电力有限公司泰州供电分公司 | A kind of microgrid virtual synchronous machine parameter optimization method based on parallel annealing algorithm |
| CN110472839A (en) * | 2019-07-25 | 2019-11-19 | 上海电力大学 | Thermal power plant's control system Information Security Evaluation system based on SA-PSO-AHP |
| CN112838604A (en) * | 2020-10-21 | 2021-05-25 | 国网河南省电力公司电力科学研究院 | Optimization method and system for energy storage power station group participating in AGC of power system |
| CN114037194A (en) * | 2022-01-12 | 2022-02-11 | 南京佰思智能科技有限公司 | Hydroelectric power plant power generation load prediction system and method based on machine learning |
| CN114221355A (en) * | 2021-11-29 | 2022-03-22 | 苏州科技大学 | Micro-grid finite time secondary frequency modulation control method based on event trigger mechanism |
| CN114970200A (en) * | 2022-06-17 | 2022-08-30 | 南京邮电大学 | Multi-energy system multi-target safety economic optimization scheduling method considering demand response |
-
2023
- 2023-04-03 CN CN202310368965.5A patent/CN116345495B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105305468A (en) * | 2015-10-19 | 2016-02-03 | 国家电网公司 | Particle swarm optimization algorithm based thermal generator set primary frequency modulation parameter optimization method |
| CN107655463A (en) * | 2017-09-21 | 2018-02-02 | 大连理工大学 | Electronic compass calibration method based on simulated annealing |
| CN109390969A (en) * | 2018-10-11 | 2019-02-26 | 国网江苏省电力有限公司泰州供电分公司 | A kind of microgrid virtual synchronous machine parameter optimization method based on parallel annealing algorithm |
| CN110472839A (en) * | 2019-07-25 | 2019-11-19 | 上海电力大学 | Thermal power plant's control system Information Security Evaluation system based on SA-PSO-AHP |
| CN112838604A (en) * | 2020-10-21 | 2021-05-25 | 国网河南省电力公司电力科学研究院 | Optimization method and system for energy storage power station group participating in AGC of power system |
| CN114221355A (en) * | 2021-11-29 | 2022-03-22 | 苏州科技大学 | Micro-grid finite time secondary frequency modulation control method based on event trigger mechanism |
| CN114037194A (en) * | 2022-01-12 | 2022-02-11 | 南京佰思智能科技有限公司 | Hydroelectric power plant power generation load prediction system and method based on machine learning |
| CN114970200A (en) * | 2022-06-17 | 2022-08-30 | 南京邮电大学 | Multi-energy system multi-target safety economic optimization scheduling method considering demand response |
Non-Patent Citations (2)
| Title |
|---|
| Evaluation of Frequency Regulation Performance of Energy Storage Power Plants Based on Correlation Analysis and Combined Weighting Method;Jinhua Xue;《2020 IEEE 4th Conference on Energy Internet and Energy System Integration (EI2)》;全文 * |
| 基于模糊综合评价的分布式电源调频调压能力的评价方法;刘一鸿等;《可再生能源》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116345495A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Local parameter optimization of LSSVM for industrial soft sensing with big data and cloud implementation | |
| CN110751318A (en) | IPSO-LSTM-based ultra-short-term power load prediction method | |
| CN112884236B (en) | Short-term load prediction method and system based on VDM decomposition and LSTM improvement | |
| CN111967183A (en) | Method and system for calculating line loss of distribution network area | |
| CN111832839B (en) | Energy consumption prediction method based on sufficient incremental learning | |
| CN111723982A (en) | Medium-and-long-term power load combined prediction method based on gray-Markov chain | |
| CN115759415A (en) | Power consumption demand prediction method based on LSTM-SVR | |
| CN110866633A (en) | Micro-grid ultra-short term load prediction method based on SVR support vector regression | |
| CN110807490A (en) | Intelligent prediction method for construction cost of power transmission line based on single-base tower | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN116757057A (en) | Air quality prediction method based on PSO-GA-LSTM model | |
| CN114510778A (en) | Track irregularity prediction method based on hybrid intelligent optimization LSTM | |
| CN112381315A (en) | LS-SVM intelligent platform area load prediction method and system based on PSO optimization | |
| CN116865343B (en) | Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network | |
| CN116345495B (en) | Power plant unit frequency modulation optimization method based on data analysis and modeling | |
| Guo et al. | Mobile user credit prediction based on lightgbm | |
| CN114091304B (en) | Intelligent decision-making method for processing hull plate by oxyhydrogen gas heat source | |
| CN115796327A (en) | Wind power interval prediction method based on VMD (vertical vector decomposition) and IWOA-F-GRU (empirical mode decomposition) -based models | |
| CN115688535A (en) | Power data combined interpolation method and system based on waveform similarity analysis | |
| CN115775038A (en) | Short-term load prediction method based on IGWO optimization LSSVM | |
| CN114707421A (en) | IJS-SVR model-based short-term wind power prediction method | |
| Li et al. | A Spatiotemporal Coupling Calculation-Based Short-Term Wind Farm Cluster Power Prediction Method | |
| CN114139442A (en) | Method for optimizing steam turbine valve flow characteristic function based on K-means-LSTM | |
| CN112668770A (en) | Power load prediction method based on overall similarity of information and waveform | |
| CN111222688A (en) | Daily load prediction method for commercial building |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |