CN115793472A - Modeling method, modeling system, control method and control system of heavy-duty train - Google Patents
Modeling method, modeling system, control method and control system of heavy-duty train Download PDFInfo
- Publication number
- CN115793472A CN115793472A CN202310101097.4A CN202310101097A CN115793472A CN 115793472 A CN115793472 A CN 115793472A CN 202310101097 A CN202310101097 A CN 202310101097A CN 115793472 A CN115793472 A CN 115793472A
- Authority
- CN
- China
- Prior art keywords
- heavy
- data
- control force
- control
- speed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000004088 simulation Methods 0.000 claims abstract description 37
- 241000283153 Cetacea Species 0.000 claims abstract description 23
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 238000005457 optimization Methods 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims description 48
- 230000005284 excitation Effects 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 230000000875 corresponding effect Effects 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 5
- 238000012795 verification Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000036963 noncompetitive effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000000418 atomic force spectrum Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 230000000116 mitigating effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002243 precursor Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention relates to a modeling method, a modeling system, a control method and a control system of a heavy-duty train, and belongs to the technical field of heavy-duty train control. The modeling method comprises the following steps: acquiring the operation data of the heavy-duty train; clustering speed data and control force data in the operation data respectively to obtain a speed clustering result and a control force clustering result; constructing an interval two-type fuzzy model based on the speed clustering result and the control force clustering result; and optimizing parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-load train. According to the method, the running data of the heavy-duty train is obtained, the interval type-two fuzzy model is established in a data driving mode, and the parameters of the interval type-two fuzzy model are optimized by a whale optimization algorithm to obtain a high-precision simulation model, so that the high-precision control of the heavy-duty train is realized.
Description
Technical Field
The invention relates to the technical field of heavy-duty train control, in particular to a modeling method, a modeling system, a control method and a control system of a heavy-duty train.
Background
Due to the advantages of environmental protection, rapidness, large transportation capacity and the like, heavy-duty trains are more and more emphasized, and the development of the heavy-duty trains becomes a trend. The heavy-duty train is a large and complex system, and the difficulty of train driving is very high. Particularly in harsh environments, frequent braking and mitigation results in a dramatic increase in driver fatigue and operational risk factors. Therefore, in order to improve the efficiency of train operation on the premise of ensuring safety, a reasonably designed automatic driving control system for heavy-duty trains is needed, which means that the operation process of heavy-duty trains needs to be described more accurately and a proper control method needs to be selected.
In the existing modeling research, most scholars establish a mechanism model based on system stress analysis and an existing empirical formula. Although the mechanism modeling can meet the train operation control requirement to a certain extent, a large amount of prior knowledge and empirical formulas are required, and the accuracy of the model is verified through multiple field tests, so that the experiment cost is very high. And, along with the operation mobile unit and the track of train receive wearing and tearing consumption easily, this will lead to there being great error between mechanism model and the actual train, and the automatic driving control strategy can no longer satisfy the train operation requirement. How to improve the simulation precision of heavy haul trains and further improve the effect of automatic driving control becomes a technical problem to be solved urgently.
Disclosure of Invention
The invention aims to provide a modeling method, a modeling system, a control method and a control system of a heavy-duty train, so as to improve the simulation precision of the heavy-duty train and further improve the effect of automatic driving control.
In order to achieve the purpose, the invention provides the following scheme:
the invention provides a modeling method of a heavy-duty train, which comprises the following steps:
acquiring the running data of the heavy-load train; the operational data includes speed data and control force data;
clustering speed data and control force data in the operation data respectively to obtain a speed clustering result and a control force clustering result; specifically, the speed clustering result includes n speed classes, and the control force clustering result includes n control force classes;
constructing an interval two-type fuzzy model based on the speed clustering result and the control force clustering result;
and optimizing the parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-duty train.
Optionally, the clustering speed data and control force data in the operating data respectively to obtain a speed clustering result and a control force clustering result specifically includes:
and respectively clustering the speed data and the control force data in the operating data by adopting a fuzzy C-means clustering method to obtain a speed clustering result and a control force clustering result.
Optionally, the two-interval fuzzy model is:
wherein v (t + 1) represents the velocity at time t +1, y k Denotes the kth submodel, f k Represents the weight of the kth sub-model, M represents the number of sub-classes,
the coefficient for the speed representing the kth sub-class,
a coefficient regarding a control force indicating a kth sub-class; v (t) represents the velocity at time t, u (t) represents the control force at time t,
and
are respectively provided withIndicating the lower excitation intensity and the upper excitation intensity of the kth sub-class,
and
respectively representing the lower excitation intensity and the upper excitation intensity of the jth subclass,
、
and
respectively representing the center of a lower membership function, the center of an upper membership function and the variance of the velocity class in the kth sub-class,
、
and
respectively representing the center of a lower membership function, the center of an upper membership function and the variance of the control force class in the kth sub-class,
、
and
respectively a lower membership function center, an upper membership function center and a variance of the velocity class in the jth sub-class,
、
and
respectively representing the lower membership function center, the upper membership function center and the variance of the control force class in the jth sub-class; n represents the number of speed classes in the speed clustering result or the number of control force classes in the control force clustering result, each subclass comprises a speed class and a control force class,
n speed classes and n control force classes are combined to obtain M sub-classes.
A modeling system of a heavy haul train, which is applied to the modeling method, comprises:
the operation data acquisition module is used for acquiring the operation data of the heavy-duty train; the operational data includes speed data and control force data;
the clustering module is used for clustering the speed data and the control force data in the operating data respectively to obtain a speed clustering result and a control force clustering result;
the interval type two fuzzy model building module is used for building an interval type two fuzzy model based on the speed clustering result and the control force clustering result;
and the interval type two fuzzy model optimization module is used for optimizing the parameters of the interval type two fuzzy model by adopting a whale optimization algorithm to obtain an optimized interval type two fuzzy model which is used as a simulation model of the heavy-load train.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the modeling method described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when executed, implements the above-described modeling method.
A method for controlling the operation of a heavy-duty train comprises the following steps:
obtaining a simulation model of the heavy-duty train by adopting the modeling method;
constructing an iterative learning active disturbance rejection controller based on the simulation model;
and controlling the heavy-duty train based on an iterative learning active disturbance rejection controller.
Optionally, the iterative learning active disturbance rejection controller includes an active disturbance rejection controller and an iterative learning controller, the active disturbance rejection controller includes a linear state extended state observer and a state error feedback control law, and the simulation model is set in the iterative learning controller.
An operation control system of a heavy haul train, the operation control system of the heavy haul train comprising:
the simulation model building module is used for obtaining a simulation model of the heavy-duty train by adopting the modeling method;
the active disturbance rejection controller building module is used for building an iterative learning active disturbance rejection controller based on the simulation model;
and the control module is used for controlling the heavy-duty train based on the iterative learning active disturbance rejection controller.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention discloses a modeling method of a heavy-duty train, which comprises the following steps: acquiring the operation data of the heavy-duty train; the operational data includes speed data and control force data; clustering speed data and control force data in the operation data respectively to obtain a speed clustering result and a control force clustering result; constructing an interval two-type fuzzy model based on the speed clustering result and the control force clustering result; and optimizing the parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-duty train. According to the method, the running data of the heavy-duty train is obtained, the interval type-two fuzzy model is established in a data driving mode, and the parameters of the interval type-two fuzzy model are optimized by a whale optimization algorithm to obtain a high-precision simulation model, so that the high-precision control of the heavy-duty train is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 is a flowchart of a modeling method for a heavy haul train according to an embodiment of the present invention;
FIG. 2 is a block diagram of a two-type interval fuzzy system according to an embodiment of the present invention;
FIG. 3 is a diagram of a range two-type membership function according to an embodiment of the present invention;
FIG. 4 is a verification diagram of a simulation model provided by an embodiment of the present invention;
FIG. 5 is an error diagram of a simulation model provided by an embodiment of the present invention;
fig. 6 is a flowchart of a control method for a heavy-duty train according to an embodiment of the present invention;
fig. 7 is a schematic diagram of an iterative learning active-disturbance-rejection controller according to an embodiment of the present invention;
FIG. 8 is a velocity tracking diagram of a control method provided by an embodiment of the present invention;
FIG. 9 is a velocity tracking error diagram of a control method according to an embodiment of the present invention;
fig. 10 is a control diagram of a control method according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a modeling method, a modeling system, a control method and a control system of a heavy-duty train, so as to improve the simulation precision of the heavy-duty train and further improve the effect of automatic driving control.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Example 1
In continuous research and research, the introduction of data-driven technology in the industrial process of metallurgy, which is difficult to establish an accurate mechanism model, can effectively improve the control effect. Therefore, for a heavy-duty train, data-driven modeling may have a better effect, and a large amount of operation records generated in the train operation process can also provide sufficient data support for a data-driven method, thereby providing a powerful basis for realizing more optimal control. As shown in fig. 1, based on this, embodiment 1 of the present invention provides a modeling method for a heavy haul train, where the modeling method includes the following steps:
s1, acquiring running data of a heavy-duty train; the operational data includes speed data and control force data.
And S2, clustering the speed data and the control force data in the operation data respectively to obtain a speed clustering result and a control force clustering result.
Illustratively, the invention S2 adopts a fuzzy C-means clustering method to cluster two data sets of the speed and the control force of the heavy-duty train respectively, and divides the data sets into a plurality of subclasses respectively.
The fuzzy C-means clustering method clusters data by a continuous iteration updating method, can process a large amount of data and quickly converge to give a classification result, and comprises the following steps:
s201 randomly selecting n cluster centers and initializing a matrix determined by a membership function
, wherein U0 Is composed of
Order matrix, N being samplesC is the order of the data (i.e., the number of clusters or classes, which is equal to n);
represents the degree of membership of the s-th cluster (class) of the i-th sample, and
；
s202 calculates a value of the cluster center:
wherein ,ms Is the cluster center of the s-th cluster, x i The ith sample is the sample, and the sum of the membership degrees of the ith sample for each cluster is 1;
s203, updating the membership function matrix:
wherein ,mc Represents the cluster center of the c-th cluster, a represents a constant;
if it is not
If yes, the iteration is finished, otherwise, the step returns to S202 for recalculation; wherein, U f and Uf+1 Membership function matrices for the f-th iteration and the f + 1-th iteration respectively,
indicating the iteration accuracy.
And S3, constructing an interval two-type fuzzy model based on the speed clustering result and the control force clustering result.
S3 specifically comprises the following steps
1. Data processing:
before modeling, firstly, the operation data (speed and control force) of the heavy-duty train are processed, and the specific steps are as follows:
s301 determines the number of cluster centers n for the two input data sets.
S302, clustering processing is carried out on input data respectively by adopting an FCM (Fuzzy C-means) algorithm to obtain 2 x n groups of clustering data, wherein the clustering data comprises n speed classes and n control force classes.
S303, calculating the variance sigma corresponding to each cluster data and recording
、
Respectively representing the variance of the ith speed class and the variance of the ith control force class, wherein
，
；
S304, each group of clustering data is respectively subjected to 30-time non-repeated 50% random extraction and the average value is calculated, and the maximum value and the minimum value are taken as the centers of the upper membership function and the lower membership function of the group of clustering data. Note the book
、
And
、
the upper and lower membership function centers of the l speed class and the o control force class are respectively;
the variance and the clustering center obtained by the steps form a precursor parameter of the interval type two fuzzy model together.
2. Interval type two fuzzy system
The interval type-two fuzzy system is an extension of the one-type fuzzy system, and the membership function is popularized to one domain from a determined value, so that the problems of language ambiguity and data noise are solved better, and the interval type-two set modeling method is more suitable for modeling heavy-duty trains with high uncertainty, strong nonlinearity and serious noise.
The structure of the interval type two fuzzy system is further described, and the system structure diagram is shown in FIG. 2.
The first layer is an input layer, the input value is a clear signal, and the control force and the speed of the heavy-duty train are input in embodiment 1 of the invention;
the second layer is a membership function layer, firstly, two input data are respectively clustered according to the set number of clustering centers and a clustering method, and then, an upper membership function and a lower membership function corresponding to each group of subclass data are respectively obtained according to the selected membership function type so as to determine the range and necessary parameters of a membership domain.
The third layer is an excitation intensity layer, and n is obtained by calculation 2 Upper and lower excitation intensities of individual subclasses (each subclass characterizing a sub-rule)
、
；
The fourth layer of membership function layer realizes an inference mechanism on the basis of Takagi-Sugeno rules and iterative Karnik-Mendel algorithm, and obtains the weight of each rule through calculation
, wherein 
；
And finally, multiplying all the submodels by corresponding weights by a fifth-layer output layer, and summing to obtain final output. The final output of the system can be written as:
wherein ,
and
respectively an upper membership function and a lower membership function of the ith velocity class,
and
respectively an upper membership function and a lower membership function of the o-th control force class.
The gaussian membership function is chosen for the calculation, which in the type one fuzzy is generally of the form:
wherein ,
is membership, m is the clustering center; sigma is the clustering width, namely the variance corresponding to the clustering data; and x is a sample. While in interval two type ambiguity, it is generalized to an uncertain coverage domain, and its general form becomes:
the membership function image is shown in fig. 3.
In the above-mentioned process, the first step,
、
respectively an upper membership function and a lower membership function;
、
is upper and lower membership functionThe center of the number.
Then, the modeling data is divided according to two input categories to obtain
And the sub-model parameters obtained by fitting all the sub-class data of the data of each subclass form the back piece parameters of the interval two-type fuzzy model. Wherein, the data of each subclass corresponds to a sub-rule. By the first
The individual sub-rules are examples, and the corresponding expressions are shown below.
wherein ,Rj A rule representing the jth sub-class; v. of j(t) and uj (t) respectively representing the speed and control force of the jth subclass at the moment t;
the ith cluster type representing the first input (velocity), wherein
；
The o-th cluster type representing the second input (control force), wherein
(ii) a And is provided with
；
The coefficient with respect to speed of the jth sub-class,
coefficient on control force for the jth subclass.
In summary, a general form of the interval type two fuzzy model can be obtained.
Where v (t + 1) represents the velocity at time t +1, y k Denotes the kth submodel, f k Represents the weight of the kth sub-model, M represents the number of sub-classes,
the coefficient for the speed representing the kth sub-class,
a coefficient regarding a control force indicating a kth sub-class; v (t) represents the velocity at time t, u (t) represents the control force at time t,
and
respectively representing the lower excitation intensity and the upper excitation intensity of the kth sub-class,
and
respectively representing the lower excitation intensity and the upper excitation intensity of the jth subclass,
、
and
respectively representing the center of a lower membership function, the center of an upper membership function and the variance of the velocity class in the kth sub-class,
、
and
respectively representing the center of a lower membership function, the center of an upper membership function and the variance of the control force class in the kth sub-class,
、
and
respectively a lower membership function center, an upper membership function center and a variance of the velocity class in the jth sub-class,
、
and
respectively representing the lower membership function center, the upper membership function center and the variance of the control force class in the jth sub-class; n represents the number of speed classes in the speed clustering result or the number of control force classes in the control force clustering result, each subclass comprises a speed class and a control force class,
n speed classes and n control force classes are combined to obtain M sub-classes.
And S4, optimizing the parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-duty train.
After the interval type-two fuzzy model and the initial parameters of the model are obtained, based on model verification errors, a WOA Algorithm (Whale Algorithm) is adopted to locally optimize the former parameters of the model, and the influence of data errors on model precision is reduced.
S401Setting the number r of whales and the initial position of each whale in the D dimension space
And the number of iterations;
s402 every whale has 50% probability of surrounding the prey and 50% probability of forming a bubble net to expel the prey to update the position of the whale.
S403 surrounds the prey: (1) when in use
When the temperature of the water is higher than the set temperature,
(ii) a (2) When the temperature is higher than the set temperature
Time-piece
. Wherein A is (-b) 0 ,b 0 ) A random number in between, and b 0 With the number of updates gradually decreasing from 2 to 0.
Refers to the position of the r-th whale at the time t,
for the optimal whale position at the moment t,
the position of a whale is randomized at the moment t. L is a random number uniformly distributed among (0, 2),
the position of the r-th whale at the time t + 1,
the position of the ith whale at time t + 1,
the location of the ith whale at time t.
S404 bubble net prey expelling article：
. Wherein R is
Uniformly distributed random numbers in between.
S405 determines whether all iterations have been completed, and if not, repeats S402.
The former part parameters of the model are optimized through the method to obtain final model parameters, and then the established interval type-II fuzzy model, namely the optimized interval type-II fuzzy model, is obtained. In the modeling process, the overall nonlinear problem is divided into a plurality of local linear problems to be solved, so that the parameters can be kept near the global optimum, the WOA algorithm with strong local optimizing capability can find the optimum point more easily, and the heavy-duty train model can be accurately described.
The modeling method of embodiment 1 of the invention is verified by taking an HXD1 type heavy-duty train as an object.
The modeling uses 40738 groups of actual operation data of the HXD1 type heavy-duty train in a certain line in China, 30000 groups are randomly selected as modeling data, and 400 groups are randomly selected as model verification data. The parameters of the front and back parts obtained by the modeling and optimizing process are shown in tables 1 and 2. And substituting the two fuzzy models into an equation (7) to obtain the established interval two-type fuzzy model.
TABLE 1 front part parameters
wherein 
Represents the first speed class and the second speed class,
represents the first
A control force class;
、
respectively as upper and lower membership function centers; σ is the variance corresponding to each cluster data.
TABLE 2 Back part parameters
Next, the obtained simulation model is verified using the modeling data, and a model verification map and a model error map are shown in fig. 4 and 5.
From the above results, it can be found that the modeling error is stabilized
The effect is very good.
Example 2
For the control of heavy-duty trains, many advanced methods such as sliding mode control, adaptive control and the like are proposed. The sliding mode control has good robustness and stability, and the buffeting problem can be solved only by sacrificing a tracking error. The adaptive control can be automatically adjusted according to the data characteristics, the control effect is good, but the adaptive control method is still a model-based control method, and the control effect is not good for a heavy-load train which is a variable strong nonlinear system.
In embodiment 2 of the present invention, based on the simulation model obtained in embodiment 1, high-precision operation control of the heavy-duty train is realized by an iterative learning self-interference rejection controller, as shown in fig. 6.
As shown in fig. 6, an embodiment 2 of the present invention provides an operation control method for a heavy haul train, where the operation control method for the heavy haul train includes the following steps:
a simulation model of the heavy-duty train is obtained by the modeling method of the embodiment 1.
And constructing an iterative learning active disturbance rejection controller based on the simulation model.
And controlling the heavy-duty train based on an iterative learning active disturbance rejection controller.
The iterative learning active disturbance rejection controller mainly comprises two parts: active disturbance rejection control and iterative learning control. The active disturbance rejection control is mainly used for estimating and compensating disturbance, and the iterative learning control is mainly used for making up for the incompleteness of the periodic disturbance compensation of the active disturbance rejection controller.
The method regards the running of the train in a fixed interval as non-strict repeated periodic action, and compensates the interference which is difficult to observe in the active disturbance rejection control by using a data-driven control method through iterative learning control, so that the dependence on parameters in the active disturbance rejection control is greatly reduced. Meanwhile, the compensation effect of the active disturbance rejection control on non-competitive disturbance and uncertain factors can make up the defect that iterative learning control is difficult to process the non-competitive disturbance and uncertain factors.
The invention adopts a P-type learning law in iterative learning, and has the function of correcting the output quantity of a controller in the previous period by utilizing historical period errors so as to ensure that the errors are converged cycle by cycle. The periodic learning law is as follows:
wherein u (t-N) and e (t-N) are respectively the control quantity of the last period at the time t and the error between the model output and the target output; phi is a learning parameter; n is the number of sampling points in one control period,
the control law of the whole control system at the time t.
The active disturbance rejection controller mainly comprises a Linear Extended State Observer (LESO) and a state error feedback control law (NLSEF).
(1) Linear Extended State Observer (LESO)
The LESO estimates the total disturbance f (t) + bXu (t) in real time on line and compensates the total disturbance f (t) + bXu (t), so that the suppression of the disturbance and the high-precision operation tracking are realized, wherein f (t) is the nominal disturbance at the moment of t, and b is a control coefficient. Further, the nominal model may utilize known terms of modeling to reduce the burden of disturbance estimation, increasing controller performance. Based on the established heavy-duty train model, the following linear state extended state observer is designed:
among these are the following modeling known items, all of which are part of the simulation model.
And is provided with
Is composed of
To pair
The tracking error of (2);
actual system output at time t;
the control law of the whole control system at the moment t;
is composed of
The tracking signal of (a) is detected,
a tracking signal actually output by the system at the moment of t + 1; h is the sampling period, where
Second;
and
respectively estimating values of total disturbance at the time t and the time t + 1;
、
is a coefficient;
for sudden disturbances, v (t-1) represents the velocity at time t-1.
(2) State error feedback control law (NLSEF)
NLESF is used to obtain the control law of active disturbance rejection control, which can reflect the disturbance compensation of the linear state extended state observer on the control output, and finally realize the effect of active disturbance rejection:
wherein ,
the error between the system output tracking signal and the target signal for time t + 1,
is the target signal at time t + 1,
the tracking signal is output for the system at time t + 1,
is a velocity factor, whose value is positively correlated to the tracking velocity, fal () is a non-linear function,
is a coefficient;
is a non-linear factor;
a control law of an active disturbance rejection control part at the moment t + 1;
in order to calculate an intermediate quantity of the control law,
sign () is a sign function for the nominal perturbation at time t + 1. In summary, a structure diagram of the active disturbance rejection controller according to embodiment 2 of the present invention is shown in fig. 7.
Based on the two control parts, the final output control law of the whole controller is the sum of the outputs of the two parts:
wherein ,
the control law of the whole control system at the time t + 1,
and iteratively learning the control law of the control part for the time t + 1.
In the invention, a speed-displacement tracking target curve is obtained on the basis of the operation data on a certain line section in China, and the performance of the controller is judged through the tracking effect. The history data of 5 operation cycles is used as the history support data in the iterative learning control.
Fig. 8 is a velocity tracking diagram, and it can be seen that the controller has a very good tracking effect. FIG. 9 is a plot of velocity tracking error with tracking error at
Within the range, the error is very small, and the tracking precision of the controller is proved numerically. Fig. 10 is a control force curve, and it can be seen that the control force is very smooth and meets the actual conditions and requirements.
Example 3
An embodiment 3 of the present invention provides a modeling system for a heavy haul train, where the modeling system is applied to the modeling method of embodiment 1, and the modeling system includes:
the operation data acquisition module is used for acquiring the operation data of the heavy-duty train; the operational data includes speed data and control force data.
And the clustering module is used for clustering the speed data and the control force data in the operating data respectively to obtain a speed clustering result and a control force clustering result.
And the interval type two fuzzy model building module is used for building an interval type two fuzzy model based on the speed clustering result and the control force clustering result.
And the interval type two fuzzy model optimization module is used for optimizing the parameters of the interval type two fuzzy model by adopting a whale optimization algorithm to obtain an optimized interval type two fuzzy model which is used as a simulation model of the heavy-load train.
Further, the computer program in the memory described above may be stored in a computer-readable storage medium when it is implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: u disk, removable hard disk, read only memory, random access memory, magnetic or optical disk, etc. for storing program codes.
Example 4
An embodiment 4 of the present invention provides an operation control system for a heavy haul train, where the operation control system for the heavy haul train includes:
the simulation model building module is used for obtaining a simulation model of the heavy-duty train by adopting the modeling method;
the active disturbance rejection controller building module is used for building an iterative learning active disturbance rejection controller based on the simulation model;
and the control module is used for controlling the heavy-duty train based on the iterative learning active disturbance rejection controller.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention discloses a modeling method of a heavy-duty train, which comprises the following steps: acquiring the operation data of the heavy-duty train; the operational data includes speed data and control force data; clustering speed data and control force data in the operation data respectively to obtain a speed clustering result and a control force clustering result; constructing an interval type two fuzzy model based on the speed clustering result and the control force clustering result; and optimizing the parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-duty train. According to the method, the running data of the heavy-duty train is obtained, the interval type-two fuzzy model is established in a data driving mode, and the parameters of the interval type-two fuzzy model are optimized by a whale optimization algorithm to obtain a high-precision simulation model, so that the high-precision control of the heavy-duty train is realized.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The principle and the embodiment of the present invention are explained by applying specific examples, and the above description of the embodiments is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (9)
1. A modeling method of a heavy-duty train is characterized by comprising the following steps:
acquiring the operation data of the heavy-duty train; the operational data includes speed data and control force data;
clustering speed data and control force data in the operation data respectively to obtain a speed clustering result and a control force clustering result;
constructing an interval two-type fuzzy model based on the speed clustering result and the control force clustering result;
and optimizing the parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-duty train.
2. The modeling method of a heavy-duty train according to claim 1, wherein said clustering speed data and control force data in said operation data respectively to obtain a speed clustering result and a control force clustering result specifically comprises:
and respectively clustering the speed data and the control force data in the operating data by adopting a fuzzy C-means clustering method to obtain a speed clustering result and a control force clustering result.
3. The modeling method of a heavy-duty train according to claim 1, wherein said block type two fuzzy model is:
wherein v (t + 1) represents t +1Velocity of time, y k Denotes the kth submodel, f k Represents the weight of the kth sub-model, M represents the number of sub-classes,
the coefficient for speed representing the kth sub-class,
a coefficient regarding a control force indicating a kth sub-class; v (t) represents the velocity at time t, u (t) represents the control force at time t,
and
respectively representing the lower excitation intensity and the upper excitation intensity of the kth sub-class,
and
respectively representing the lower excitation intensity and the upper excitation intensity of the jth subclass,
、
and
respectively representing the center of a lower membership function, the center of an upper membership function and the variance of the velocity class in the kth sub-class,
、
and
respectively representing the lower membership function center, the upper membership function center and the variance of the control force class in the kth sub-class,
、
and
respectively a lower membership function center, an upper membership function center and a variance of the velocity class in the jth sub-class,
、
and
respectively representing the center of a lower membership function, the center of an upper membership function and the variance of a control force class in the jth sub-class; n represents the number of speed classes in the speed clustering result or the number of control force classes in the control force clustering result, each subclass comprises a speed class and a control force class,
n speed classes and n control force classes are combined to obtain M sub-classes.
4. A modeling system for a heavy haul train, wherein the modeling system is applied to the modeling method according to any one of claims 1 to 3, and the modeling system comprises:
the operation data acquisition module is used for acquiring the operation data of the heavy-duty train; the operational data includes speed data and control force data;
the clustering module is used for clustering the speed data and the control force data in the operating data respectively to obtain a speed clustering result and a control force clustering result;
the interval type II fuzzy model building module is used for building an interval type II fuzzy model based on the speed clustering result and the control force clustering result;
and the interval type-II fuzzy model optimization module is used for optimizing parameters of the interval type-II fuzzy model by adopting a whale optimization algorithm to obtain the optimized interval type-II fuzzy model which is used as a simulation model of the heavy-duty train.
5. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the modeling method of any of claims 1-3 when executing the computer program.
6. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when executed, implements a modeling method according to any of claims 1-3.
7. The operation control method of the heavy haul train is characterized by comprising the following steps of:
obtaining a simulation model of the heavy-duty train by using the modeling method of any one of claims 1 to 3;
constructing an iterative learning active disturbance rejection controller based on the simulation model;
and controlling the heavy-duty train based on an iterative learning active disturbance rejection controller.
8. The operation control method of the heavy-duty train according to claim 7, wherein the iterative learning active-disturbance-rejection controller includes an active-disturbance-rejection controller and an iterative learning controller, the active-disturbance-rejection controller includes a linear state expansion state observer and a state error feedback control law, and the simulation model is provided in the iterative learning controller.
9. An operation control system of a heavy haul train, comprising:
a simulation model construction module for obtaining a simulation model of the heavy-duty train by using the modeling method of any one of claims 1 to 3;
the active disturbance rejection controller building module is used for building an iterative learning active disturbance rejection controller based on the simulation model;
and the control module is used for controlling the heavy haul train based on the iterative learning active disturbance rejection controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310101097.4A CN115793472B (en) | 2023-02-13 | 2023-02-13 | Modeling method, modeling system, control method and control system of heavy-duty train |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310101097.4A CN115793472B (en) | 2023-02-13 | 2023-02-13 | Modeling method, modeling system, control method and control system of heavy-duty train |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115793472A true CN115793472A (en) | 2023-03-14 |
| CN115793472B CN115793472B (en) | 2023-06-20 |
Family
ID=85430877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310101097.4A Active CN115793472B (en) | 2023-02-13 | 2023-02-13 | Modeling method, modeling system, control method and control system of heavy-duty train |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115793472B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050197994A1 (en) * | 2004-03-03 | 2005-09-08 | Shigeru Fujii | Intelligent robust control system for motorcycle using soft computing optimizer |
| CN102981408A (en) * | 2012-12-10 | 2013-03-20 | 华东交通大学 | Running process modeling and adaptive control method for motor train unit |
| CN103019267A (en) * | 2012-12-10 | 2013-04-03 | 华东交通大学 | Predicative control method for modeling and running speed of adaptive network-based fuzzy inference system (ANFIS) of high-speed train |
| CN110989342A (en) * | 2019-11-19 | 2020-04-10 | 华北电力大学 | Real-time T-S fuzzy modeling method for combined cycle unit heavy-duty gas turbine |
| CN111523749A (en) * | 2020-02-28 | 2020-08-11 | 华中科技大学 | Intelligent identification method for hydroelectric generating set model |
| CN112394729A (en) * | 2020-11-09 | 2021-02-23 | 华东交通大学 | T-S fuzzy modeling and fuzzy tracking control method for motor train unit operation process |
| CN112967420A (en) * | 2021-04-01 | 2021-06-15 | 华东交通大学 | Interval two-type based heavy-duty train operation process monitoring method and system |
| CN114690754A (en) * | 2020-12-25 | 2022-07-01 | 天津大学 | Track tracking control algorithm for iterative learning self-optimization-seeking feedforward of repeatedly operated unmanned vehicle |
-
2023
- 2023-02-13 CN CN202310101097.4A patent/CN115793472B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050197994A1 (en) * | 2004-03-03 | 2005-09-08 | Shigeru Fujii | Intelligent robust control system for motorcycle using soft computing optimizer |
| CN102981408A (en) * | 2012-12-10 | 2013-03-20 | 华东交通大学 | Running process modeling and adaptive control method for motor train unit |
| CN103019267A (en) * | 2012-12-10 | 2013-04-03 | 华东交通大学 | Predicative control method for modeling and running speed of adaptive network-based fuzzy inference system (ANFIS) of high-speed train |
| CN110989342A (en) * | 2019-11-19 | 2020-04-10 | 华北电力大学 | Real-time T-S fuzzy modeling method for combined cycle unit heavy-duty gas turbine |
| CN111523749A (en) * | 2020-02-28 | 2020-08-11 | 华中科技大学 | Intelligent identification method for hydroelectric generating set model |
| CN112394729A (en) * | 2020-11-09 | 2021-02-23 | 华东交通大学 | T-S fuzzy modeling and fuzzy tracking control method for motor train unit operation process |
| CN114690754A (en) * | 2020-12-25 | 2022-07-01 | 天津大学 | Track tracking control algorithm for iterative learning self-optimization-seeking feedforward of repeatedly operated unmanned vehicle |
| CN112967420A (en) * | 2021-04-01 | 2021-06-15 | 华东交通大学 | Interval two-type based heavy-duty train operation process monitoring method and system |
Non-Patent Citations (2)
| Title |
|---|
| YATING FU ET AL.: "Linear active disturbance rejection control of heavy-haul train operation based on an interval type-2 fuzzy logic system model" * |
| 杨辉等: "列车运行建模与速度控制方法综述" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115793472B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102242516B1 (en) | Train machine learning models on multiple machine learning tasks | |
| CN110647042B (en) | Robot robust learning prediction control method based on data driving | |
| CN108284442B (en) | Mechanical arm flexible joint control method based on fuzzy neural network | |
| CN111047085B (en) | Hybrid vehicle working condition prediction method based on meta-learning | |
| US20120259600A1 (en) | Method of identifying hammerstein models with known nonlinearity structures using particle swarm optimization | |
| CN112930541A (en) | Determining a control strategy by minimizing delusional effects | |
| CN114077776A (en) | Structural reliability robust optimization design method based on active learning agent model | |
| CN111931973A (en) | Method for improving short-time load prediction precision of cable line | |
| CN111241749B (en) | Permanent magnet synchronous motor chaos prediction method based on reserve pool calculation | |
| CN116303786B (en) | Block chain financial big data management system based on multidimensional data fusion algorithm | |
| CN115793472B (en) | Modeling method, modeling system, control method and control system of heavy-duty train | |
| CN112488399A (en) | Power load prediction method and device | |
| CN114830137A (en) | Method and system for generating a predictive model | |
| CN114911157A (en) | Robot navigation control method and system based on partial observable reinforcement learning | |
| CN111738407B (en) | Clock error prediction method, device, medium and terminal based on deep learning | |
| CN114372558A (en) | Residential electricity load prediction method, medium and equipment based on multi-model fusion | |
| Jin et al. | Expectation-maximization algorithm based identification of Hammerstein nonlinear ARMAX systems with Gaussian mixture noises | |
| CN113112092A (en) | Short-term probability density load prediction method, device, equipment and storage medium | |
| CN113191526A (en) | Short-term wind speed interval multi-objective optimization prediction method and system based on random sensitivity | |
| JP3276035B2 (en) | A sequential accelerated learning method for neural network models | |
| CN116340766B (en) | Sliding window-based lithium battery SOC online prediction method and related equipment | |
| Mahootchi et al. | Oppositional extension of reinforcement learning techniques | |
| Ma et al. | Research on robot SLAM of RBPF improved with weight accumulation | |
| CN114624994B (en) | Novel active disturbance rejection control method and system of high-order flexible linear system | |
| CN113343366B (en) | Method for determining main section parameters of vehicle body and related equipment |
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 |