CN111523749B - Intelligent identification method for hydroelectric generating set model - Google Patents
Intelligent identification method for hydroelectric generating set model Download PDFInfo
- Publication number
- CN111523749B CN111523749B CN202010129510.4A CN202010129510A CN111523749B CN 111523749 B CN111523749 B CN 111523749B CN 202010129510 A CN202010129510 A CN 202010129510A CN 111523749 B CN111523749 B CN 111523749B
- Authority
- CN
- China
- Prior art keywords
- whale
- particle
- population
- fitness
- generating set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 241000283153 Cetacea Species 0.000 claims abstract description 110
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 75
- 238000005457 optimization Methods 0.000 claims abstract description 42
- 230000008569 process Effects 0.000 claims abstract description 28
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 27
- 230000001105 regulatory effect Effects 0.000 claims abstract description 21
- 230000004044 response Effects 0.000 claims abstract description 21
- 230000005284 excitation Effects 0.000 claims abstract description 9
- 239000002245 particle Substances 0.000 claims description 52
- 230000006870 function Effects 0.000 claims description 48
- 239000003795 chemical substances by application Substances 0.000 claims description 39
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000004088 simulation Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 abstract description 6
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 abstract description 2
- 230000007246 mechanism Effects 0.000 description 7
- 230000003044 adaptive effect Effects 0.000 description 5
- 238000013459 approach Methods 0.000 description 4
- 238000013178 mathematical model Methods 0.000 description 4
- 230000002028 premature Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000002068 genetic effect Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000011664 nicotinic acid Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 235000014036 Castanea Nutrition 0.000 description 1
- 241001070941 Castanea Species 0.000 description 1
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000019637 foraging behavior Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000013095 identification testing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 244000062804 prey Species 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000012546 transfer Methods 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses an intelligent identification method for a hydroelectric generating set model, which comprises the following steps: establishing a corresponding identification system model according to the water turbine speed regulating system, and acquiring an actual response signal output by the water turbine speed regulating system under the excitation of a given input signal and an analog response signal output by the identification system under the excitation of the given input signal; defining a difference value between an actual response signal and a simulated response signal as a target function, and performing iterative optimization on the parameters to be identified by adopting a whale optimization algorithm to minimize the target function to obtain the optimal identification parameters of the hydroelectric generating set; in the iteration process, the global search probability is increased by balancing random search and optimal search; the method increases the global search probability in the traditional whale optimization algorithm, fuses immune operators, adjusts the search space by adopting a self-adaptive correction method, improves the optimization efficiency, has the advantages of high convergence speed, short calculation time and high efficiency, and effectively improves the identification precision.
Description
Technical Field
The invention belongs to the technical field of water turbine parameter identification, and particularly relates to an intelligent identification method for a hydroelectric generating set model.
Background
Water is the source of life, and electricity is the guarantee of social development. Hydropower stations play a crucial role in the supply of electric power in daily life, industry and agriculture. Therefore, the research on the water turbine has important significance on the stable and safe operation of the water turbine generator set. However, due to the complexity and non-linearity of engineering systems, it has been difficult to obtain accurate models and parameters. The advent of identification technology has provided a new approach to this problem and has evolved into one of the most critical research branches for water turbines.
The identification of turbines has advanced in great length over the past few decades. The rapid development of the population intelligent Algorithm provides huge power for the field of water turbine model parameter identification, such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Gravity Search Algorithm (GSA), Whale Optimization Algorithm (WOA), etc.; it has become a trend to apply the above population intelligence methods to various identification tests.
The whale optimization algorithm is a heuristic bionic optimization algorithm based on hunting behaviors of whales in nature, the search subject is a whale population and a prey population, and the prey is moved and searched by surrounding prey and spiral bubble net of the whales, so that the target is optimized. However, as with the genetic algorithm and the particle swarm algorithm, the WOA algorithm also has the problems of premature phenomenon and slow convergence speed, and is difficult to meet the requirement of parameter identification of the current water turbine speed regulating system.
Disclosure of Invention
The invention provides an intelligent identification method for a hydroelectric generating set model, aiming at the condition that the convergence rapidity and accuracy of the conventional whale optimization algorithm are insufficient, the global search probability is increased in the conventional whale optimization algorithm, and the phenomenon that the model is early trapped in local optimum is avoided; immune operators are fused, and the diversity of the population is increased; in the identification process, an adaptive correction method is adopted to solve the uncertainty of the estimated parameter range, the search space is adjusted in a self-adaptive manner, and the optimization efficiency is improved; the reliability and the accuracy of the model identification of the hydroelectric generating set are greatly improved.
In order to achieve the above object, according to an aspect of the present invention, there is provided a method for intelligently identifying a model of a hydroelectric generating set, the method including the steps of:
establishing a corresponding identification system model according to a water turbine speed regulating system, and acquiring an actual response signal output by the water turbine speed regulating system under the excitation of a given input signal and an analog response signal output by the identification system under the excitation of the given input signal;
defining the difference value between the actual response signal and the simulated response signal as an objective function, and performing iterative optimization on the parameter to be identified of the water turbine speed regulating system by adopting a whale optimization algorithm to minimize the objective function to obtain the optimal identification parameter of the hydroelectric generating set; in the iterative optimization process, the global search probability is increased by balancing the random search and the optimal search:
if the ratio of the current iteration times to the maximum iteration times is larger than a preset value, updating the position of the whale population by a method of randomly selecting the whale position to search for prey;
and if the ratio of the current iteration times to the maximum iteration times is not greater than a preset value, updating the position of the whale population by adopting a spiral bubble net moving method.
Preferably, the intelligent identification method for the hydroelectric generating set model further comprises the following steps:
and carrying out immune operator variation operation on the whale population in the iterative optimization process, and updating the fitness function value of each whale particle in the population and the optimal search agent in the population by adopting fitness and concentration indexes as evaluation conditions.
Preferably, the intelligent identification method for the hydroelectric generating set model further comprises the step of performing adaptive correction on the range of the parameter to be identified:
if the position of the optimal search agent generated under the current iteration number is within an expected range, reducing the expected range according to a preset scaling factor;
and if the position of the optimal search agent generated under the current iteration number is at the boundary of the expected range, expanding the expected range according to the scaling factor.
Preferably, the intelligent identification method for the hydroelectric generating set model adopts a whale optimization algorithm to perform iterative optimization on the to-be-identified parameters of the hydraulic turbine speed regulating system, and specifically comprises the following steps:
(1) randomly initializing a first generation whale population, setting parameters to be identified and starting iteration;
(2) calculating an objective function value of each whale particle in the population, finding the whale particle with the minimum objective function value, and recording an optimal search agent and the position thereof under the current iteration times; updating the parameters to be identified, and executing the next iteration;
(3) in the iteration process, if p is less than 0.5 and | E | ≧ 1, or p is more than or equal to 0.5 and a is less than 1, updating the position of the whale by adopting a random hunting mode;
if p is more than or equal to 0.5 and a is more than or equal to 1, updating the position of the whale by adopting a spiral bubble net moving mode;
if p <0.5 and | E | <1 are satisfied, updating the position of the whale in a manner of surrounding the prey;
wherein, E-2 ar-a, a-2 (1-T/T);
p represents a random probability; e represents a random variable; a represents a convergence factor; r is a random number between 0 and 1; t is the current iteration number; t is the maximum iteration number;
(4) and when the termination condition is met, stopping iteration and outputting the optimal identification parameters.
Preferably, the intelligent identification method for the hydroelectric generating set model, which uses the fitness and the concentration indexes as evaluation conditions to update the fitness function value of each whale particle in the population and the optimal search agent in the population, specifically comprises the following steps:
(1) in each iteration process, acquiring fitness function values of whale particles in a whale population and summing the fitness function values to obtain a population fitness sum;
(2) traversing each whale particle in the whale population, calculating a difference value between the fitness function value of the ith whale particle and the fitness function of any whale particle, and adding 1 to a counter when the absolute value of the difference value is smaller than a preset minimum distance; counting the counter value of the ith whale particle; wherein i is 1 to Nm，NmRepresents the total number of whale particles in the population;
(3) calculating the fitness index of the whale particle according to the fitness function value of the ith whale particle and the fitness sum of the population, Pf＝Pi/Psum(ii) a Calculating concentration index according to the value of the counter of the ith whale particle and the total number of the whale particles, Pd＝Nc/Nm；
Wherein, Pf、PdRespectively representing a fitness index and a concentration index;
Psumrepresenting the sum of fitness of the population;
Nca counter value representing a whale population;
(4) calculating the selection probability of the ith whale particle according to the fitness index and the concentration index, Pr＝α*Pf+(1-α)*Pd(ii) a Alpha is a proportionality coefficient;
(5) when the selection probability of the ith whale particle is larger than the preset replacement probability, calculating the position of the latest search agent according to the preset expected range;
(6) if the fitness function value of the latest search agent is larger than the fitness function value of the ith whale particle, updating the position of the ith whale particle to the position of the latest search agent;
(7) and if the fitness function value of the latest search agent is larger than the fitness function value of the optimal search agent under the current iteration times, updating the position of the optimal search agent under the current iteration times to be the position of the latest search agent.
Preferably, in the method for intelligently identifying a hydroelectric generating set model, the termination condition is a preset maximum continuous iteration number and/or a variation between objective functions at a certain interval iteration number, and the iteration is stopped when the iteration number reaches the maximum continuous iteration number or the variation is smaller than a preset threshold value.
Preferably, in the intelligent identification method for the hydroelectric generating set model, the given input signal is a frequency step signal, and the actual/simulated response signal is a guide vane opening and/or a set frequency signal.
Preferably, in the method for intelligently identifying a hydroelectric generating set model, the objective function is a weighted sum of a difference between an actual guide vane opening and a simulated guide vane opening and a difference between an actual generating set frequency signal and a simulated generating set frequency signal.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
(1) in the traditional whale optimization algorithm, the global search probability is increased, the convergence speed in the later iteration stage of the identification process is increased, and a better solution can be obtained;
(2) according to the method, the immune operator is introduced in the iterative optimization process, and is applied to whale populations, so that the problem that the algorithm is easy to fall into local optimization is solved;
(3) in the identification process, the self-adaptive correction method is adopted for the parameter range to be identified, so that the problem of uncertainty of the estimated parameter range is solved, the search space is adjusted in a self-adaptive manner, and the optimization efficiency is improved;
(4) the invention can be realized by adopting MATLAB language with simple and flexible usage, and the algorithm has low complexity, is easy to program and has good adaptability in engineering application.
Drawings
Fig. 1 is a schematic structural diagram of a water turbine governing system provided by an embodiment of the invention;
FIG. 2 is a schematic diagram of a mathematical model of a turbine governor system according to an embodiment of the present invention;
FIG. 3 is a flow chart of an identification process provided by an embodiment of the present invention;
FIG. 4 is a flow chart of the WOA algorithm;
FIG. 5 is a flow chart of an improved whale optimization algorithm (IWOA algorithm) provided by an embodiment of the present invention;
FIG. 6 is a convergence curve for the best results provided by an embodiment of the present invention for the IWOA, WOA, IA, and GA algorithms;
FIG. 7 is a graph illustrating the coincidence of measured data and simulated data with optimal results for IWOA, according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
According to the intelligent identification method for the hydroelectric generating set model, provided by the invention, on the basis of a traditional whale optimization algorithm, the global search probability is increased, so that the algorithm can be quickly converged in the later iteration stage. Meanwhile, in order to avoid the situation of local optimum, an immune mutation operator is introduced in the iteration process to process the whale population; and after obtaining the new particles through variation, calculating the fitness value of the new particles, and further updating the population and the optimal search agent according to the fitness value of the new particles. In addition, in order to improve the optimization efficiency, the invention also provides a self-adaptive correction method for estimating the parameter range by utilizing the position information and the scaling factor of the optimal search agent, and the finally determined optimal search agent is the optimal solution, namely the optimal identification parameter required by the scheme. The improved whale optimization algorithm is applied to parameter identification of the hydroelectric generating set model, and reliability and accuracy of hydroelectric generating set model identification can be greatly improved.
The following describes the identification method provided by the present invention in detail with reference to the embodiments and the accompanying drawings.
The intelligent identification method for the hydroelectric generating set model provided by the embodiment specifically comprises the following steps:
s1: establishing a corresponding identification system model according to a water turbine speed regulating system, and acquiring an actual response signal output by the water turbine speed regulating system under the excitation of a given input signal and an analog response signal output by the identification system under the excitation of the given input signal;
fig. 1 is a schematic structural diagram of a hydraulic turbine governing system provided in this embodiment, where the hydraulic turbine governing system includes three components, namely a PID controller, an execution structure and a hydro-generator; fig. 2 is a mathematical model of the hydraulic turbine governing system according to the present embodiment, and model parameters are set, and the definitions of the parameter variables are shown in table 1:
TABLE 1 model parameter definition for turbine governing system
The model parameters corresponding to the PID controller and the execution structure are known quantities, and what needs to be identified is the model parameter of the hydraulic generator, and further, the model parameter is the coefficient a of the transfer function of the hydraulic generator1～AnAnd B0～Bn-1(ii) a In this embodiment, the selected parameter to be identified is a vector θ ═ a4 A3 A2 A1 B3 B2 B1 B0]。
In the embodiment, taking a no-load working condition as an example, given input signals of a water turbine speed regulating system and an identification model thereof are selected as same given frequency step signals according to engineering practical requirements, and system output guide vane opening and unit frequency signals are collected as output response to construct a target function of a parameter to be identified; of course, the user may actually select the required output response signal to construct the objective function according to the engineering application, and is not limited to the guide vane opening and the unit frequency signal provided in the embodiment. The measurement time for the dynamic process was 40s and the sampling time was 0.005 s.
S2: defining the difference value between the actual response signal and the simulated response signal as an objective function, and performing iterative optimization on the parameter to be identified of the water turbine speed regulating system by adopting a whale optimization algorithm to minimize the objective function to obtain the optimal identification parameter of the hydroelectric generating set;
the present embodiment defines the objective function as:
wherein, N is the sample number, and x and y are the actual measurement data of the unit frequency and the guide vane opening output by the water turbine speed regulating system;
and
the simulation data of the unit frequency and the guide vane opening degree output by the identification model.
Under the excitation of the same given input signal, the water turbine speed regulating system and the identification model thereof generate respective output responses, continuously correct the parameter to be identified, minimize the error between actual measurement data and simulation data, namely the size of the target function, and lead the parameter to be identified to be close to the real parameter of the water turbine speed regulating system. The flow chart of the identification process provided by the embodiment is shown in fig. 3.
In the embodiment, a whale optimization algorithm is adopted to perform iterative optimization on the parameter to be identified of the water turbine speed regulating system, and the WOA algorithm is a bionic algorithm and simulates the foraging behavior of whales in the sea. The algorithm is proposed based on three steps, namely, enclosing prey, moving a spiral bubble net and searching prey; in the randomly generated initial whale population, each whale represents a particle and has certain fitness; and finding out the optimal particles according to the size of the fitness value of each particle in the population. And updating the positions of all whales in the population by combining three behaviors of whale surrounding prey, spiral bubble net movement and random hunting, and then obtaining a new population of N individuals. And calculating the fitness value of each individual in the population, and updating the optimal particles. And continuously and circularly repeating until the standard of stopping iteration is reached, and exiting the iteration cycle, wherein the optimal particles are the optimal result.
The mathematical model of the WOA algorithm is described as follows:
(1) surrounding prey
When whales find a prey, they can lock in place and enclose the prey. Thus, whale movement is easily affected by the location of the prey, and the optimal search agent is generally considered to be a prey. Thus, the location of the search agent may be updated by some surrounding rule, the mathematical model of which may be described as:
Xd(t+1)＝Xd *(t)-E·|C·Xd *(t)-Xd(t′)| (2)
wherein X ═ X1,…,Xd,…,Xdm]X represents the location of the search agent; d is 1, …, dm; dm represents the dimension of the search agent location; t denotes the number of current iterations, X*Representing the location of the current optimal search agent; e and C represent random variables, respectively:
wherein, a represents convergence factor, a ═ 2 (1-T/T); t is the maximum number of iterations and r is a random number between 0 and 1.
(2) Spiral bubble net movement
The bubble net attack method mainly comprises two mechanisms: a shrink wrap mechanism and a spiral update location mechanism; wherein:
a. the wrapping mechanism is realized by linearly decreasing the variable a;
b. the spiral updating position is realized through a spiral equation, the distance between whale particles and the current optimal search agent position is calculated, and then whales are simulated to capture preys in a spiral mode; it can let whale approach prey and shorten the distance between them, and this process can be expressed as follows:
Xd(t+1)＝|Xd *(t)-Xd(t)|·ebl·cos(2π·l)+Xd *(t) (4)
wherein, | | Xd *(t)-Xd(t) | represents the distance between the ith whale and the current optimal location; b is a constant for defining the form of a logarithmic spiral, and l represents [ -1,1 ]]A random number in between; set b ═ 1, l ═ -r (2+ T/T) + 1.
It should be noted that, whale needs to shrink the surrounding circle while surrounding the prey in a spiral form, so in order to realize the synchronization model, the surrounding mechanism and the spiral position update are shrunk through random probability p information; when p <0.5, a shrink wrap mechanism is performed; when p is more than or equal to 0.5, the spiral position updating mechanism is executed.
(3) Searching prey
In the searching process, whale can not only approach to prey, but also randomly search for better solution in the global scope, but the randomly selected search agent replaces the optimal search agent, and can be described as the following formula:
wherein, XrandIs a randomly selected position, E ranges from [ - | E |, ] non-calculation]It decreases as the current iteration number t increases.
These three steps are balanced by the values of the random probability p and the random variable E, and the flow chart of the WOA algorithm is shown in fig. 4.
The whale optimization algorithm updates the position of whale by surrounding prey, moving a spiral bubble net and searching prey in the optimizing search process, and has the advantages of high running speed, high solving efficiency, good convergence and the like. But when dealing with complex problems, it may suffer from "premature" phenomena and other deficiencies. In order to accelerate the convergence rate of the algorithm and improve the accuracy of the algorithm, the invention increases the global search probability by balancing random search and optimal search in the iterative optimization process:
if the ratio of the current iteration times T to the maximum iteration times T is larger than a preset value, updating the position of a whale population by a method of randomly selecting a whale position to search for prey;
and if the ratio of the current iteration time T to the maximum iteration time T is not greater than a preset value, updating the position of the whale population by adopting a spiral bubble net moving method.
The WOA algorithm is balanced between optimal search and random search, and the original algorithm is less in random search at the later stage of iteration, so that the algorithm is easy to fall into local optimal. The scheme improves the whale optimization algorithm structure, and increases the global search probability by expanding the convergence factor a; the specific algorithm is as follows:
fig. 4 is a flowchart of an improved whale optimization algorithm (IWOA algorithm for short) in the embodiment, and referring to fig. 5, the process of iteratively optimizing the parameter to be identified of the water turbine speed regulating system by using the whale optimization algorithm specifically includes:
(1) randomly initializing a first generation whale population, setting parameters to be identified and starting iteration;
(2) calculating an objective function value of each whale particle in the population, finding the whale particle with the minimum objective function value, and recording an optimal search agent and the position thereof under the current iteration times; updating the parameters to be identified, and executing the next iteration;
(3) in the iteration process, if p is less than 0.5 and | E | ≧ 1, or p is more than or equal to 0.5 and a is less than 1, updating the position of the whale by adopting a random hunting mode;
if p is more than or equal to 0.5 and a is more than or equal to 1, updating the position of the whale by adopting a spiral bubble net moving mode;
if p <0.5 and | E | <1 are satisfied, updating the position of the whale in a manner of surrounding the prey;
(4) and when the termination condition is met, stopping iteration and outputting the optimal identification parameters.
In order to avoid unnecessary long iteration or premature stopping of iteration, a self-adaptive termination strategy is adopted, and a maximum continuous iteration time T is preset in the strategy; meanwhile, considering the change threshold of the objective function; herein, the variation threshold of the objective function is a relative deviation value of the fitness; setting the maximum continuous iteration time T as 100 and the variation threshold delta as 10-5The change threshold of the objective function is compared every third iteration. And when the iteration times reach the maximum continuous iteration times T or the variable quantity threshold is smaller than a preset threshold delta, meeting a termination condition, and stopping iteration.
As a further preferred scheme of this embodiment, the method further performs immune operator mutation operation on the whale population in the iterative optimization process, and updates the fitness function value of each whale particle in the population and the optimal search agent in the population by using the fitness and concentration indexes as evaluation conditions. The fusion immune operator can effectively adjust the selection pressure and keep the species diversity; the method comprises the following specific steps:
wherein, X*Representing an optimal search agent; lb and ub are the upper and lower bounds of the desired range, respectively; n is a radical ofmIs the total number of whale particles in the population; p isiA fitness function value representing an ith whale particle; dminRepresents a minimum distance; preRepresenting a replacement probability; pf、PdAnd alpha is a fitness index, a concentration index and a proportionality coefficient respectively; prRepresenting a selection probability; .*Represents a dot product;
in this embodiment, the minimum distance D is setminReplacement probability P ═ 1re＝0.5，ub＝[200 200 200 50 50 50 50 50]，lb＝[0 0 0-50-50 0 0]。
As a further preferred scheme of the embodiment, the method performs adaptive correction on the range of the parameter to be identified in the iterative optimization process to solve the problem of uncertainty of the range of the parameter to be identified; specifically, the method comprises the following steps:
if the position of the optimal search agent generated under the current iteration number is within an expected range, reducing the expected range according to a preset scaling factor;
and if the position of the optimal search agent generated under the current iteration number is at the boundary of the expected range, expanding the expected range according to the scaling factor.
The corresponding specific algorithm is as follows:
where k% is the scaling factor and lb _ new and ub _ new are the lower and upper limits of the newly generated desired range.
In order to verify the effectiveness of the method, a parameter identification experiment of the water turbine speed regulating system under the no-load working condition is carried out, GA, IA, WOA and IWOA algorithms are respectively adopted to identify model parameters of the hydroelectric generating set, and the amplitude of step disturbance with given frequency is set to be 50-52 Hz. For each algorithm, the number of chestnuts in the population and the number of iterations were 100 and 50, respectively, and the experiment was repeated 21 times. In this identification process, for the IWOA algorithm, the scaling factor k% is 5%, the number of start iterations DS is 20, the number of iteration intervals dds is 10, and the threshold va is 10. For genetic algorithms, the crossover probability pc0.7, probability of mutation pm＝0.06。
Table 2 shows the adaptive values, the number of iterations, the total iteration time, and the average iteration time of the optimal solution, the intermediate solution, and the worst solution for each algorithm; wherein Cost represents an adaptive value, Iteration represents the Iteration number, To _ time represents the total Iteration time, and Ave _ time represents the average time of each Iteration;
TABLE 2 comparison of relevant parameters for optimal, intermediate and worst solutions of different algorithms
As can be seen from table 2, the number of iterations of the IA and GA algorithms is much smaller than those of the other algorithms, which indicates that it is difficult for the IA and GA algorithms to get rid of local optima. The results also show that the IWOA algorithm has enhanced search capabilities. Although IWOA takes a little more time per iteration than the WOA algorithm, it has a great advantage in the cost, iteration and total time of the three solutions.
Fig. 6 shows the convergence curves for IWOA, WOA, IA and GA algorithms. From fig. 6 we can see that the IA and GA algorithms have stopped iterating before the 15 th iteration due to premature convergence of the algorithm. When the number of iterations approaches 20, the IWOA algorithm can achieve a good solution. However, the WOA algorithm is a good solution until the 30 th iteration, which shows that the IWOA algorithm provided by the present solution has a faster convergence speed.
Table 3 describes the minimum, median, maximum, mean and variance of the fitness value, iteration number, total iteration time and mean iteration time of the 21 recognition results. Obviously, the cost index of the IWOA algorithm is the best among the four algorithms, which means that the IWOA algorithm has a high accuracy. The number of iterations and the value of To _ time show that the IWOA algorithm has a high efficiency. In addition, the variance results of the various methods show good stability of the IWOA algorithm. Although the value of Ave _ time for the IWOA algorithm is slightly larger than the value of the WOA algorithm, it is still within an acceptable range.
TABLE 3 comparison of the parameters associated with the adaptation values, the number of iterations, the total time of the iterations and the average iteration time of the different algorithms
Fig. 7 shows the coincidence curve of the measured data and the simulated data with the optimal results of IWOA, and it can be seen from fig. 7 that the simulated output data obtained by the IWOA algorithm provided by the present invention matches well with the measured data.
The results show that the hydropower unit estimation model identified by the IWOA algorithm provided by the invention is closest to the real model, and the accuracy is highest. The average convergence curve of the IWOA algorithm is optimized at a fast convergence rate and gives a better solution than other algorithms. The comparison result of the simulation output of the identification model and the collected actual output shows that the coincidence degree of the estimation curve and the original curve obtained by the identification of the IWOA algorithm is high, and the superiority of the method is shown.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (6)
1. An intelligent identification method for a hydroelectric generating set model is characterized by comprising the following steps:
acquiring an actual response signal output by a water turbine speed regulating system under the excitation of a given input signal and a simulation response signal output by an identification model created by the water turbine speed regulating system under the excitation of the given input signal;
defining the difference value between the actual response signal and the simulated response signal as an objective function, and performing iterative optimization on the parameter to be identified of the water turbine speed regulating system by adopting a whale optimization algorithm to minimize the objective function to obtain the optimal identification parameter of the hydroelectric generating set; in the iterative optimization process, the global search probability is increased by balancing the random search and the optimal search:
if the ratio of the current iteration times to the maximum iteration times is larger than a preset value, updating the position of the whale population in a mode of randomly selecting the whale position to search for prey;
if the ratio of the current iteration times to the maximum iteration times is not larger than a preset value, updating the position of the whale population in a spiral bubble network moving mode; the iterative optimization process of the parameters to be identified of the water turbine speed regulating system by adopting the whale optimization algorithm specifically comprises the following steps:
(1) randomly initializing a first generation whale population, setting parameters to be identified and starting iteration;
(2) calculating an objective function value of each whale particle in the population, finding the whale particle with the minimum objective function value, and recording an optimal search agent and the position thereof under the current iteration times; updating the parameters to be identified, and executing the next iteration;
(3) in the iteration process, if p is less than 0.5 and | E | ≧ 1, or p is more than or equal to 0.5 and a is less than 1, updating the position of the whale by adopting a random hunting mode;
if p is more than or equal to 0.5 and a is more than or equal to 1, updating the position of the whale by adopting a spiral bubble net moving mode;
if p <0.5 and | E | <1 are satisfied, updating the position of the whale in a manner of surrounding the prey;
wherein, E-2 ar-a, a-2 (1-T/T);
p represents a random probability; e represents a random variable; a represents a convergence factor; r is a random number between 0 and 1; t is the current iteration number; t is the maximum iteration number;
(4) and when the termination condition is met, stopping iteration and outputting the optimal identification parameters.
2. The hydroelectric generating set model intelligent identification method of claim 1, further comprising the steps of:
carrying out immune operator variation operation on whale populations in an iterative optimization process, and updating fitness function values of whale particles in the populations and optimal search agents in the populations by adopting fitness and concentration indexes as evaluation conditions, wherein the immune operator variation operation specifically comprises the following steps:
(1) in each iteration process, acquiring fitness function values of whale particles in a whale population and summing the fitness function values to obtain a population fitness sum;
(2) traversing each whale particle in the whale population, calculating a difference value between the fitness function value of the ith whale particle and the fitness function of any whale particle, and adding 1 to a counter when the absolute value of the difference value is smaller than a preset minimum distance; counting the counter value of the ith whale particle; wherein i is 1 to Nm，NmRepresents the total number of whale particles in the population;
(3) calculating the fitness index of the whale particle according to the fitness function value of the ith whale particle and the fitness sum of the population, Pf＝Pi/Psum(ii) a Calculating concentration index according to the value of the counter of the ith whale particle and the total number of the whale particles, Pd＝Nc/Nm；
Wherein, Pf、PdRespectively representing a fitness index and a concentration index; piA fitness function value representing an ith whale particle;
Psumrepresenting the sum of fitness of the population;
Nca counter value representing a whale population;
(4) calculating the selection probability P of the ith whale particle according to the fitness index and the concentration indexr，Pr＝α*Pf+(1-α)*Pd(ii) a Alpha is a proportionality coefficient;
(5) when the selection probability of the ith whale particle is larger than the preset replacement probability, calculating the position of the latest search agent according to the preset expected range;
(6) if the fitness function value of the latest search agent is larger than the fitness function value of the ith whale particle, updating the position of the ith whale particle to the position of the latest search agent;
(7) and if the fitness function value of the latest search agent is larger than the fitness function value of the optimal search agent under the current iteration times, updating the position of the optimal search agent under the current iteration times to be the position of the latest search agent.
3. The intelligent identification method of the hydroelectric generating set model according to claim 1 or 2, further comprising the step of adaptively correcting the range of the parameter to be identified:
if the position of the optimal search agent generated under the current iteration number is within an expected range, reducing the expected range according to a preset scaling factor;
and if the position of the optimal search agent generated under the current iteration number is at the boundary of the expected range, expanding the expected range according to the scaling factor.
4. The hydroelectric generating set model intelligent identification method of claim 1, wherein the termination condition is a preset maximum number of continuous iterations and/or a variation between objective functions at certain interval iterations, and the iteration is stopped when the number of iterations reaches the maximum number of continuous iterations or the variation is smaller than a preset threshold.
5. The hydroelectric generating set model intelligent identification method of claim 1, wherein the given input signal is a frequency step signal, and the actual/simulated response signal is a guide vane opening and/or a set frequency signal.
6. The hydroelectric generating set model intelligent identification method of claim 5, wherein the objective function is a weighted sum of a difference between an actual guide vane opening and a simulated guide vane opening and a difference between an actual generating set frequency signal and a simulated generating set frequency signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010129510.4A CN111523749B (en) | 2020-02-28 | 2020-02-28 | Intelligent identification method for hydroelectric generating set model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010129510.4A CN111523749B (en) | 2020-02-28 | 2020-02-28 | Intelligent identification method for hydroelectric generating set model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111523749A CN111523749A (en) | 2020-08-11 |
| CN111523749B true CN111523749B (en) | 2022-05-13 |
Family
ID=71900898
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010129510.4A Expired - Fee Related CN111523749B (en) | 2020-02-28 | 2020-02-28 | Intelligent identification method for hydroelectric generating set model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111523749B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113014170B (en) * | 2021-04-21 | 2022-08-12 | 昆明理工大学 | Permanent magnet synchronous motor minimum loss control method based on whale optimization algorithm |
| CN113411006B (en) * | 2021-05-19 | 2022-07-12 | 南昌大学 | Whale algorithm-based charging and discharging control method for energy storage bidirectional converter in grid-connected mode |
| CN113887691B (en) * | 2021-08-24 | 2022-09-16 | 杭州电子科技大学 | Whale evolution system and method for service combination problem |
| CN114114922B (en) * | 2021-11-26 | 2023-10-24 | 云南电力技术有限责任公司 | Control parameter optimization method for water turbine adjusting system |
| CN114325516B (en) * | 2021-12-16 | 2024-03-01 | 华中科技大学 | Method and system for measuring magnetic characteristics of permanent magnet material |
| CN115793472B (en) * | 2023-02-13 | 2023-06-20 | 华东交通大学 | Modeling method, modeling system, control method and control system of heavy-duty train |
| CN117193404A (en) * | 2023-09-28 | 2023-12-08 | 武汉船舶职业技术学院 | Multi-objective optimization control method for hydroelectric generating set model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106681133A (en) * | 2016-12-29 | 2017-05-17 | 中国南方电网有限责任公司电网技术研究中心 | Method for identifying hydroelectric generating set model improved type subspace closed loop |
| CN107798199A (en) * | 2017-11-09 | 2018-03-13 | 华中科技大学 | A kind of Hydropower Unit parameter closed-loop identification method |
| CN109886589A (en) * | 2019-02-28 | 2019-06-14 | 长安大学 | A method of low-carbon Job-Shop is solved based on whale optimization algorithm is improved |
| CN110147611A (en) * | 2019-05-20 | 2019-08-20 | 福州大学 | A kind of Structural Damage Identification based on IWOA |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9026257B2 (en) * | 2011-10-06 | 2015-05-05 | Avista Corporation | Real-time optimization of hydropower generation facilities |
-
2020
- 2020-02-28 CN CN202010129510.4A patent/CN111523749B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106681133A (en) * | 2016-12-29 | 2017-05-17 | 中国南方电网有限责任公司电网技术研究中心 | Method for identifying hydroelectric generating set model improved type subspace closed loop |
| CN107798199A (en) * | 2017-11-09 | 2018-03-13 | 华中科技大学 | A kind of Hydropower Unit parameter closed-loop identification method |
| CN109886589A (en) * | 2019-02-28 | 2019-06-14 | 长安大学 | A method of low-carbon Job-Shop is solved based on whale optimization algorithm is improved |
| CN110147611A (en) * | 2019-05-20 | 2019-08-20 | 福州大学 | A kind of Structural Damage Identification based on IWOA |
Non-Patent Citations (5)
| Title |
|---|
| A modified whale optimization algorithm based adaptive fuzzy logic PID controller for load frequency control of autonomous power generation systems;Raghuraman Sivalingam,Subramani Chinnamuthu,Subhransu SekharDash;《AUTOMATIKA》;20171231;410-421 * |
| Synthesis of Nonuniform sparse Linear Array Antenna using Whale Optimization Algorithm;Pengliang Yuan, Chenjiang Guo, Jun Ding and Yan Qu;《2017 Sixth Asia-Pacific Conference on Antennas and Propagation 》;20171019;1-3 * |
| 基于改进螺旋更新位置模型的鲸鱼优化算法;吴泽忠等;《系统工程理论与实践》;20191125(第11期);全文 * |
| 基于自适应调整权重和搜索策略的鲸鱼优化算法;孔芝等;《东北大学学报(自然科学版)》;20200115(第01期);全文 * |
| 收敛因子非线性变化的鲸鱼优化算法;龙文,伍铁斌,唐 斌;《工程科技Ⅱ辑》;20171213;102-107 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111523749A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111523749B (en) | Intelligent identification method for hydroelectric generating set model | |
| CN107798199B (en) | Hydroelectric generating set parameter closed-loop identification method | |
| CN111353582B (en) | Particle swarm algorithm-based distributed deep learning parameter updating method | |
| CN112700060B (en) | Station terminal load prediction method and prediction device | |
| CN114066333A (en) | Data processing method and device | |
| Yong | An improved artificial fish swarm algorithm for optimal operation of cascade reservoirs | |
| Lu et al. | Study on the optimal hydropower generation of Zhelin reservoir | |
| CN111144663A (en) | Ultra-short-term wind power prediction method for offshore wind farm considering output fluctuation process | |
| CN111709511A (en) | Harris eagle optimization algorithm based on random unscented Sigma point variation | |
| CN112670984A (en) | Power system economic load distribution method based on improved whale algorithm | |
| CN115017787A (en) | Wind power plant voltage ride through characteristic equivalent modeling method and system based on intelligent algorithm | |
| CN109857459A (en) | A kind of E grades of supercomputer ocean model transplants optimization method and system automatically | |
| CN114330119B (en) | Deep learning-based extraction and storage unit adjusting system identification method | |
| CN114880806A (en) | New energy automobile sales prediction model parameter optimization method based on particle swarm optimization | |
| CN115563882A (en) | Non-real-time observation active power distribution network optimization method based on improved simulated learning | |
| CN109146131A (en) | A kind of wind-power electricity generation prediction technique a few days ago | |
| CN116306306A (en) | Micro-grid inertia constant estimation method based on African bald-Condition algorithm | |
| CN115912367A (en) | Intelligent generation method for operation mode of power system based on deep reinforcement learning | |
| CN114912828A (en) | Optimization method based on numerical control machine tool technological parameters | |
| CN117540642A (en) | Multi-objective optimization method and system | |
| CN115808952B (en) | Energy system maximum power tracking control method based on improved particle swarm optimization | |
| CN114995105A (en) | Water turbine regulating system PID parameter optimization method based on improved genetic algorithm | |
| CN113076696A (en) | Load short-term prediction method and prediction system based on IPSO-chaotic BP network | |
| CN107171310A (en) | A kind of distributed light storage electricity generation system voltage pulsation inhibition index Forecasting Methodology of power distribution network | |
| CN114841472B (en) | GWO optimization Elman power load prediction method based on DNA hairpin variation |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220513 |