CN109670227B - Method for estimating parameter pairs of simulation mathematical model based on big data - Google Patents
Method for estimating parameter pairs of simulation mathematical model based on big data Download PDFInfo
- Publication number
- CN109670227B CN109670227B CN201811503393.2A CN201811503393A CN109670227B CN 109670227 B CN109670227 B CN 109670227B CN 201811503393 A CN201811503393 A CN 201811503393A CN 109670227 B CN109670227 B CN 109670227B
- Authority
- CN
- China
- Prior art keywords
- big data
- module
- data
- mathematical model
- simulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 47
- 238000013178 mathematical model Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000005457 optimization Methods 0.000 claims abstract description 31
- 238000003860 storage Methods 0.000 claims abstract description 21
- 238000013500 data storage Methods 0.000 claims abstract description 13
- 239000002245 particle Substances 0.000 claims description 30
- 238000004422 calculation algorithm Methods 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 12
- 230000003044 adaptive effect Effects 0.000 claims description 7
- 238000009826 distribution Methods 0.000 claims description 3
- 230000006872 improvement Effects 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 abstract description 6
- 238000004364 calculation method Methods 0.000 description 9
- 230000002776 aggregation Effects 0.000 description 5
- 238000004220 aggregation Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000007906 compression Methods 0.000 description 3
- 230000006835 compression Effects 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 238000013523 data management Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000009776 industrial production Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 230000006837 decompression Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of simulation, and discloses a method for estimating a simulation mathematical model parameter pair based on big data, wherein the system for estimating the simulation mathematical model parameter pair based on the big data comprises the following steps: the device comprises a big data acquisition module, a main control module, a parameter setting module, an optimization module, an estimation module, a data storage module and a display module. According to the invention, the big data acquisition module can ensure that the industrial big data can be safely transmitted in real time, so that the pressure of network transmission is relieved; most of distortion data can be filtered through the estimation module, and the unbiased property of the parameter estimation result is improved; meanwhile, a row key consisting of the measuring point ID and the time sequence is designed through the data storage module to store industrial big data according to rows, so that the data with time correlation and measuring point correlation in business logic are adjacently arranged according to rows in physical storage, the read-write performance is optimized, and the balance between the query efficiency and the write-in efficiency is realized.
Description
Technical Field
The invention belongs to the technical field of simulation, and particularly relates to a method for estimating a pair of simulation mathematical model parameters based on big data.
Background
When the simulation mathematical model is applied to a specific industrial production device, targeted parameter estimation needs to be carried out. Simulation (Simulation), i.e. using project models to translate uncertainty specific to a particular level into their impact on the target, which is expressed at the level of the project Simulation project as a whole. Project simulation utilizes a computer model and risk estimation of a specific level, and generally adopts a Monte Carlo method to carry out simulation. The models are used to reproduce the essential processes occurring in the actual system and to study the existing or designing systems by means of experiments on the system models, also called simulations. The models referred to herein include physical and mathematical, static and dynamic, continuous and discrete models. The system is also very wide, and comprises electric, mechanical, chemical, hydraulic, thermal and other systems, and also comprises social, economic, ecological, management and other systems. Simulation is a particularly effective means of study when the system under study is expensive, the risk of experimentation is great, or it takes a long time to understand the consequences of system parameter changes. An important tool for simulation is the computer. The difference between simulation and numerical calculation and solution is that it is an experimental technique. The simulation process comprises two main steps of establishing a simulation model and carrying out a simulation experiment. However, the data volume generated by the existing simulation data source in real time is huge, the data of the industrial control system or the intelligent sensor is changed almost in second level, and under the conditions of small data packet, large quantity and high frequency, the pressure on the acquisition server and the transmission network is very large, and the acquisition efficiency is low; and, industrial production data often contains various noise errors, and some data may be distorted; the interference of the data with larger residual errors on parameter calculation is increased, and the calculation result is biased; meanwhile, the data scale in the industrial process is larger and larger, the data volume is larger and larger, the acquisition and storage of massive large data face huge pressure, and the traditional relational database or real-time database can not meet the application requirements of the industrial large data.
In summary, the problems of the prior art are as follows: the existing simulation data source has huge data amount generated in real time, high data storage cost and low data management and scheduling capability; the data of an industrial control system or an intelligent sensor is changed almost in second level, and under the conditions of small data packets, large quantity and high frequency, the pressure on an acquisition server and a transmission network is very large, and the acquisition efficiency is low; and, industrial production data often contains various noise errors, and some data may be distorted; the interference of the data with larger residual errors on parameter calculation is increased, and the calculation result is biased; meanwhile, the data scale in the industrial process is larger and larger, the data volume is larger and larger, the acquisition and storage of massive large data face huge pressure, and the traditional relational database or real-time database can not meet the application requirements of the industrial large data; the display has large noise of the displayed image and poor image texture display effect.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method for estimating the parameter pair of a simulation mathematical model based on big data.
The invention is realized in such a way that a method for estimating the parameter pair of the simulation mathematical model based on the big data comprises the following steps:
acquiring industrial big data information by using a data acquisition interface;
setting a pair of simulation mathematical model parameters by using simulation software through a parameter setting module;
step three, optimizing the acquired big data by using a big data optimization algorithm based on an optimized particle swarm optimization algorithm; estimating operation of simulation mathematical model parameter pairs by using simulation software;
step four, storing the acquired industrial big data by using a memory;
and fifthly, displaying the acquired industrial big data and the estimation result by using a display for image denoising based on the global adaptive fractional integral image denoising algorithm.
Further, in the fifth step, a big data optimization algorithm based on an optimized particle swarm optimization algorithm is as follows:
in a D-dimension big data cloud storage clustering feature space, m particles form a population, a data clustering problem is converted into a multi-objective optimization problem, and big data are clustered in cloud storage:
minF(x)＝[f 1 (x)，f 2 (x)，…，f n (x)]
s.t.g i (x) I =1,2, \ 8230n ≥ 0 (or ≥ 0)
h j (x)＝0j＝1，2，…，m；
Wherein f is i (x) (i =1, 2.. Times.n) is an objective function, g i (x) The system has two unstable 1 cycle points x =0 and x =1-1/μ, h j (x) Is an equality constraint; introducing a chaos particle swarm disturbance concept to obtain a characteristic solution of a clustering center dominated by a decision variable x, wherein the characteristic solution comprises the following steps:
for each big data information feature vector X i And (4) archiving:
l i (k)＝(1-ρ)l i (k-1)+γf(x i (k))；
wherein f is i Is a Pareto optimal solution, P ij (k) Represents the ith decision variable at time k, the inequality f i (X*)≤f i (X) holds, where i =1, 2.., N, sets the threshold N of the clustering th When N is present eff ＜N th When searching for O of the region α And O β The probability that the clustering of the two intervals is correct is:
updating particles in a storage library by adopting a particle swarm hop count improvement mechanism;
updating the spatial position of each particle in the population of particles:
wherein x is k To search for inertial weights within the region, a is the non-inferior solution of the cluster center, d e For the distance from the extreme point to the non-inferior solution, when evaluating the uniformity degree of the solution set distribution, calculating the vector function q (x) according to the optimal clustering center i k /x i k-1 ) And according to the updating iteration sequence in the module group, obtaining:
∑ τ ＝diag(max(σ i -τ，0)；
the particle fitness function of the big data cluster in the cloud storage is obtained as follows:
where { α, β } is the diversity convergence objective function.
Further, the image denoising algorithm based on the global adaptive fractional integral in the fifth step includes:
setting the average value of the gradient amplitudes of each pixel point in 8 directions in the image f (i, j) as M (i, j), normalizing, and solving the integral order corresponding to the pixel point; taking the maximum value of M (i, j) as Y and the minimum value as X, normalizing the gradient amplitude of the pixel point, and then solving the dynamic fractional order integral order:
another object of the present invention is to provide a big-data-based simulative mathematical model parameter pair estimation system applying the big-data-based simulative mathematical model parameter pair estimation method, wherein the big-data-based simulative mathematical model parameter pair estimation system comprises:
the big data acquisition module is connected with the main control module and is used for acquiring industrial big data information through the data acquisition interface;
the main control module is connected with the big data acquisition module, the parameter setting module, the optimization module, the estimation module, the data storage module and the display module and is used for controlling each module to normally work through the single chip microcomputer;
the parameter setting module is connected with the main control module and used for setting a simulation mathematical model parameter pair through simulation software;
the optimization module is connected with the main control module and used for optimizing the acquired big data through an optimization algorithm;
the estimation module is connected with the main control module and is used for estimating the parameters of the simulation mathematical model through simulation software;
the data storage module is connected with the main control module and used for storing the acquired industrial big data through the memory;
and the display module is connected with the main control module and used for displaying the acquired industrial big data and the estimation result through the display.
Another object of the present invention is to provide an information data processing terminal applying the method for estimating the parameter pairs of the big data based simulation mathematical model.
The invention has the advantages and positive effects that: according to the invention, the big data acquisition module acquires industrial big data according to a multi-process concurrent form, and real-time data acquisition with second frequency can be realized, so that the pressure of a data acquisition interface is relieved; by carrying out uniform protocol conversion and packet uploading on the industrial big data, the industrial big data can be ensured to be transmitted safely in real time, so that the pressure of network transmission is relieved; the accuracy of parameter estimation can be improved more or less by the estimation module. How much the accuracy can be improved is still related to the data distortion caused by the noise in the data. The method can filter most of distortion data, and improves the unbiased property of the parameter estimation result; meanwhile, a row key consisting of the measuring point ID and the time sequence is designed through the data storage module to store industrial big data according to rows, so that the data with time correlation and measuring point correlation on business logic are adjacently arranged according to rows on physical storage, the read-write performance is optimized, and the balance between the query efficiency and the write-in efficiency is realized; the global adaptive fractional integral image denoising algorithm is used for reducing image noise and reserving enhanced image textures; the big data optimization algorithm based on the particle swarm optimization algorithm is beneficial to optimizing data, reducing storage cost and improving data management and scheduling capability.
Drawings
Fig. 1 is a flowchart of a method for measuring a pair of parameters of a simulation mathematical model based on big data according to an embodiment of the present invention.
FIG. 2 is a schematic diagram of a measurement system for large data-based pairs of parameters of a simulated mathematical model according to an embodiment of the present invention;
in the figure: 1. a big data acquisition module; 2. a main control module; 3. a parameter setting module; 4. an optimization module; 5. an estimation module; 6. a data storage module; 7. and a display module.
Detailed Description
In order to further understand the contents, features and effects of the present invention, the following embodiments are illustrated and described in detail with reference to the accompanying drawings.
The structure of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in FIG. 1, the method for measuring the pair of parameters of the big data-based simulation mathematical model provided by the invention comprises the following steps:
step S101, collecting industrial big data information by using a data collection interface;
step S102, setting a simulation mathematical model parameter pair by using simulation software through a parameter setting module;
step S103, optimizing the acquired big data by using a big data optimization algorithm based on an optimized particle swarm optimization algorithm; estimating operation of simulation mathematical model parameter pairs by using simulation software;
step S104, storing the collected industrial big data by using a memory;
and S105, displaying the acquired industrial big data and the estimation result by using a display for image denoising based on the global adaptive fractional integral image denoising algorithm.
In step S105, the big data optimization algorithm based on the optimized particle swarm optimization algorithm provided in the embodiment of the present invention is as follows:
on the premise that m particles form a population in a D-dimensional big data cloud storage clustering feature space, when a disturbance sequence is added into the population, clustering precision is influenced, therefore, the data clustering problem is converted into a multi-target optimization problem, and the mathematics of big data clustering in cloud storage is described as follows:
min F(x)＝[f 1 (x)，f 2 (x)，…，f n (x)]
s.t.g i (x) I =1,2, \ 8230n ≥ 0 (or ≥ 0)
h j (x)＝0j＝1，2，…，m
Wherein f is i (x) (i =1, 2.. Times.n) is an objective function, g i (x) The system has two unstable 1-cycle points x =0 and x =1-1/μ, h j (x) Is an equality constraint. Here, a chaos particle swarm disturbance concept is introduced, and a characteristic solution of a clustering center dominated by a decision variable x is obtained as follows:
to avoid trapping of particles in local optima, the feature vector X is used for each large data information i Archiving is carried out, and is:
l i (k)＝(1-ρ)l i (k-1)+γf(x i (k))
wherein f is i Is a Pareto optimal solution, P ij (k) Represents the ith decision variable at time k, the inequality f i (X*)≤f i (X) holds, where i =1,2, \ 8230;, N, sets the threshold N for clustering th When N is present eff <N th When searching for O of the region α And O β The probability that the clustering of the two intervals is correct is:
updating particles in a storage library by adopting a particle swarm hop count improvement mechanism;
updating the spatial position of each particle in the population of particles:
wherein x is k To search for inertial weights within the region, a is the non-inferior solution of the cluster center, d e Calculating the distance from the extreme point to the non-inferior solution according to the optimal clustering center vector when evaluating the uniformity degree of the solution set distribution
Function q (x) i k /x i k-1 ) And according to the updating iteration sequence in the module group, obtaining:
∑τ＝diag(max(σ i -τ，0)
the particle fitness function of the big data cluster in the cloud storage is obtained as follows:
where { α, β } is the diversity convergence objective function.
The big data optimization algorithm based on the particle swarm optimization algorithm is beneficial to optimizing data, reducing storage cost and improving data management and scheduling capability.
In step S105, the global adaptive fractional order integral-based image denoising algorithm provided in the embodiment of the present invention includes the following steps:
and setting the average value of the gradient amplitudes of each pixel point in 8 directions in the image f (i, j) as M (i, j), and normalizing to obtain the integral order corresponding to the pixel point. Taking the maximum value of M (i, j) as Y and the minimum value as X, normalizing the gradient amplitude of the pixel point, and then calculating the dynamic fractional order integral order:
therefore, the noise reduction method can realize that the part (regarded as a noise point) with a larger gradient mean value has a smaller negative order, and the fractional integral of the order has a larger attenuation effect on the noise; integration orders with corresponding sizes at the middle and small positions (regarded as image texture points) of the gradient amplitude have certain enhancement and retention effects on the image texture.
As shown in FIG. 2, the system for measuring the parameter pair of the big data-based simulation mathematical model provided by the invention comprises: the device comprises a big data acquisition module 1, a main control module 2, a parameter setting module 3, an optimization module 4, an estimation module 5, a data storage module 6 and a display module 7.
The big data acquisition module 1 is connected with the main control module 2 and is used for acquiring industrial big data information through a data acquisition interface;
the main control module 2 is connected with the big data acquisition module 1, the parameter setting module 3, the optimization module 4, the estimation module 5, the data storage module 6 and the display module 7 and is used for controlling each module to normally work through the singlechip;
the parameter setting module 3 is connected with the main control module 2 and used for setting a simulation mathematical model parameter pair through simulation software;
the optimization module 4 is connected with the main control module 2 and used for optimizing the acquired big data through an optimization algorithm;
the estimation module 5 is connected with the main control module 2 and is used for estimating the parameters of the simulation mathematical model through simulation software;
the data storage module 6 is connected with the main control module 2 and used for storing the acquired industrial big data through a memory;
and the display module 7 is connected with the main control module 2 and is used for displaying the acquired industrial big data and the estimation result through a display.
The big data acquisition module 1 provided by the invention has the following acquisition method:
1) Acquiring the industrial big data at a data acquisition interface in a multi-process concurrent mode, and performing protocol conversion and primary packet uploading on the industrial big data;
2) And receiving the industrial big data uploaded by the first package at a data receiving interface, carrying out data analysis and numerical value compression on the industrial big data, and writing the processed industrial big data into a database.
The receiving the industrial big data uploaded by the first package, performing data analysis and numerical compression processing on the industrial big data, and writing the processed industrial big data into a database comprises the following steps:
receiving the industrial big data uploaded by the primary packet at a data convergence interface, and performing data summarization, data compression, data encryption and secondary packet uploading on the industrial big data;
and receiving the industrial big data uploaded by the secondary package at the data receiving interface, carrying out data decryption, data decompression, data analysis and numerical value compression on the industrial big data, and writing the processed industrial big data into a database.
The data acquisition method provided by the invention further comprises the following steps: and after the industrial big data is acquired in the multi-process concurrent mode, data filtering is carried out on the industrial big data, and only data with the timestamp and the data value changing simultaneously can be stored in a first shared memory area in the data acquisition interface.
The data acquisition method provided by the invention further comprises the following steps: the data is transmitted by adopting a breakpoint resume technology, and the breakpoint resume technology comprises the following steps:
when the communication between the data acquisition interface and the data aggregation interface fails, marking the data which is not transmitted; when the communication is recovered to normal, under the condition of not influencing normal data uploading, the untransmitted data is uploaded to the data aggregation interface, and/or when the communication between the data aggregation interface and the data receiving interface fails, the untransmitted data is marked; and when the communication is recovered to be normal, uploading the untransmitted data to the data receiving interface under the condition of not influencing the uploading of normal data.
The data acquisition interface uploads the industrial big data uploaded by the first package to the data aggregation interface through a UDP network transmission protocol; and uploading the industrial big data uploaded by the secondary packet to the data receiving interface through a TCP network transmission protocol at the data aggregation interface.
The estimation method of the estimation module 5 provided by the invention comprises the following steps:
setting dependent variable y and p independent variables x 1 ,x 2 ,…,x p There is a functional relationship: y = beta 0 +β 1 x 1 +…+β p x p Wherein beta is 0 ,β 1 ,…,β p Is a parameter to be estimated, characterized in that it is carried out in the following manner:
(1) Let n be x in data samples participating in least square calculation i1 ,x i2 ,…,x ip ,y i I =1,2, \ 8230;, n, using
Obtained by weighted least squares regression
Wherein->
Is beta 0 ,β 1 ,…,β p An estimated value of (d);
(2) Calculating the average deviation of the sample data:
(3) From i =1 \ 8230 \ 8230n cycle the following treatments were carried out:
calculating epsilon i ＝abs(y i -β 0 -β 1 x i 1-…-β p x ip )；
If epsilon i If the distortion data is too large, judging whether the distortion data is distortion data, and filtering if the distortion data is distortion data;
wherein the distortion determination is performed as follows:
ε i >ε*x s ，
wherein x s Can be an empirical constant or can be automatically adjusted by a program, x s >1；
(4) Obtaining the number n of the filtered data samples 1 Let the total number of samples initially involved in the calculation be n0
When n is 1 <0.67 × n0 or n = n 1 If so, finishing the parameter estimation calculation;
otherwise: let n = n 1 And returning to (1) with this n 1 Re-performing least square calculation on the data samples
The storage method of the data storage module 6 provided by the invention is as follows:
firstly, acquiring a measuring point name and a measuring point time of the industrial big data;
secondly, respectively acquiring corresponding measuring point IDs and time sequences according to the measuring point names and the measuring point times;
and then, storing the industrial big data according to the measuring point ID and the time sequence.
The method for acquiring the corresponding time sequence according to the measuring point time comprises the following steps:
converting the character string of the measuring point time into date and time;
and acquiring a time interval between the measuring point time and preset time, wherein the hours corresponding to the time interval are the time sequence.
The step of storing the industrial big data according to the measuring point ID and the time sequence comprises the following steps: and storing the industrial big data from the same measuring point ID according to the time sequence.
The storage method provided by the invention further comprises the following steps:
merging industrial big data from the same measuring point ID into the same data area;
and merging the industrial big data from the same time sequence in the same data area into the same data file.
The storage method provided by the invention further comprises the following steps: automatically filtering industrial big data from the same measuring point ID and the same measuring point time, and storing the industrial big data if the industrial big data is not stored; if the industry big data is stored, the industry big data is abandoned.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications, equivalent changes and modifications made to the above embodiment according to the technical spirit of the present invention are within the scope of the technical solution of the present invention.
Claims (5)
1. A method for estimating a pair of parameters of a big data-based simulation mathematical model is characterized by comprising the following steps:
acquiring industrial big data information by using a data acquisition interface;
setting a simulation mathematical model parameter pair by using simulation software through a parameter setting module;
step three, optimizing the acquired big data by using a big data optimization algorithm based on an optimized particle swarm optimization algorithm; estimating operation of simulation mathematical model parameter pairs by using simulation software;
step four, storing the acquired industrial big data by using a memory;
and fifthly, displaying the acquired industrial big data and the estimation result by using a display for carrying out image denoising based on the global adaptive fractional order integral image denoising algorithm.
2. The method for estimating the parameter pair of the big data-based simulation mathematical model according to claim 1, wherein the big data optimization algorithm based on the particle swarm optimization algorithm in the fifth step is as follows:
in a D-dimension big data cloud storage clustering feature space, m particles form a population, a data clustering problem is converted into a multi-objective optimization problem, and big data are clustered in cloud storage:
min F(x)＝[f 1 (x)，f 2 (x)，…，f n (x)]
s.t.g i (x) Less than or equal to 0 or g i (x)≥0 i＝1，2，…n
h j (x)＝0 j＝1，2，…，m；
Wherein f is i (x) (i =1,2.. Multidot., n) is an objective function, g i (x) The system has two unstable 1-cycle points x =0 and x =1-1/μ, h j (x) Is an equality constraint; introducing a chaos particle swarm disturbance concept to obtain a characteristic solution of a clustering center dominated by a decision variable x, wherein the characteristic solution is as follows:
for each big data information feature vector X i And (4) archiving:
l i (k)＝(1-ρ)l i (k-1)+γf(x i (k))；
wherein, f i Is a Pareto optimal solution, P ij (k) Represents the ith decision variable at time k, inequality f i (X*)≤f i (X) holds, where i =1, 2.. Times, N, sets the threshold N for the cluster th When N is present eff ＜N th When searching for O of the region α And O β The probability that the clustering of the two intervals is correct is:
updating particles in a storage library by adopting a particle swarm hop count improvement mechanism;
updating the spatial position of each particle in the population of particles:
wherein x is k To search for inertial weights within the region, a is the non-inferior solution of the cluster center, d e For the distance from the extreme point to the non-inferior solution, when evaluating the uniformity of the solution set distribution, calculating the vector function q (x) according to the optimal clustering center i k /x i k-1 ) And according to the updating iteration sequence in the module group, obtaining:
∑ i ＝diag(max(σ i -τ，0)；
the particle fitness function of the big data cluster in the cloud storage is obtained as follows:
where { α, β } is the diversity convergence objective function.
3. The method for estimating the pair of parameters of the big data-based simulation mathematical model as claimed in claim 1, wherein the global adaptive fractional order integral-based image denoising algorithm in the fifth step comprises:
setting the average value of the gradient amplitudes of each pixel point in 8 directions in the image f (i, j) as M (i, j), normalizing, and solving the integral order corresponding to the pixel point; taking the maximum value of M (i, j) as Y and the minimum value as X, normalizing the gradient amplitude of the pixel point, and then solving the dynamic fractional order integral order:
4. a big-data-based simulation mathematical model parameter pair estimation system applying the big-data-based simulation mathematical model parameter pair estimation method of claim 1, wherein the big-data-based simulation mathematical model parameter pair estimation system comprises:
the big data acquisition module is connected with the main control module and used for acquiring industrial big data information through the data acquisition interface;
the main control module is connected with the big data acquisition module, the parameter setting module, the optimization module, the estimation module, the data storage module and the display module and is used for controlling each module to normally work through the single chip microcomputer;
the parameter setting module is connected with the main control module and used for setting a simulation mathematical model parameter pair through simulation software;
the optimization module is connected with the main control module and used for optimizing the acquired big data through an optimization algorithm;
the estimation module is connected with the main control module and is used for estimating the parameters of the simulation mathematical model through simulation software;
the data storage module is connected with the main control module and used for storing the acquired industrial big data through the memory;
and the display module is connected with the main control module and used for displaying the acquired industrial big data and the estimation result through the display.
5. An information data processing terminal applying the method for estimating the parameter pair of the big data based simulation mathematical model according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811503393.2A CN109670227B (en) | 2018-12-10 | 2018-12-10 | Method for estimating parameter pairs of simulation mathematical model based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811503393.2A CN109670227B (en) | 2018-12-10 | 2018-12-10 | Method for estimating parameter pairs of simulation mathematical model based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109670227A CN109670227A (en) | 2019-04-23 |
| CN109670227B true CN109670227B (en) | 2023-04-07 |
Family
ID=66143583
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811503393.2A Active CN109670227B (en) | 2018-12-10 | 2018-12-10 | Method for estimating parameter pairs of simulation mathematical model based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109670227B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112953913A (en) * | 2021-01-29 | 2021-06-11 | 江苏提米智能科技有限公司 | Method for realizing rapid access, aggregation, storage and release of industrial Internet of things intelligent equipment data |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6775818B2 (en) * | 2002-08-20 | 2004-08-10 | Lsi Logic Corporation | Device parameter and gate performance simulation based on wafer image prediction |
| CN103105778B (en) * | 2013-02-05 | 2015-07-15 | 中原工学院 | Estimation method for industrial process simulation mathematical model parameters |
| US10204200B2 (en) * | 2016-06-29 | 2019-02-12 | Oracle International Corporation | Variation-aware design analysis |
| CN107480394A (en) * | 2017-08-28 | 2017-12-15 | 河南科技学院 | A kind of methods of evaluation of simulation mathematical model parameter pair |
-
2018
- 2018-12-10 CN CN201811503393.2A patent/CN109670227B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109670227A (en) | 2019-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711609B (en) | Photovoltaic power station output power prediction method based on wavelet transformation and extreme learning machine | |
| CN112990478B (en) | Federal learning data processing system | |
| CN105184394B (en) | Optimal control method based on CPS online data mining of power distribution network | |
| Holly et al. | Evaluation of hyperparameter-optimization approaches in an industrial federated learning system | |
| CN118211268A (en) | Heterogeneous federal learning privacy protection method and system based on diffusion model | |
| Zhou et al. | Student's $ t $ VAR Modeling With Missing Data Via Stochastic EM and Gibbs Sampling | |
| CN110544047A (en) | Bad data identification method | |
| CN106952077B (en) | Work order processing strategy generation method and device | |
| CN109670227B (en) | Method for estimating parameter pairs of simulation mathematical model based on big data | |
| CN116993548A (en) | Incremental learning-based education training institution credit assessment method and system for LightGBM-SVM | |
| CN110008914A (en) | A kind of pattern recognition system neural network based and recognition methods | |
| CN101206727B (en) | Data processing apparatus, data processing method | |
| Wang et al. | Online reconstruction from big data via compressive censoring | |
| CN110995770B (en) | Fuzzy test application effect comparison method | |
| CN114238504B (en) | E-government oriented cross-chain data query and consensus optimization method | |
| Purnawansyah et al. | K-Means clustering implementation in network traffic activities | |
| CN106897553B (en) | Home intelligent power method for establishing model based on single-sensor | |
| CN109978138A (en) | The structural reliability methods of sampling based on deeply study | |
| CN115952895A (en) | Wind power device power prediction method, system and storage medium | |
| Zhao et al. | Realization of intrusion detection system based on the improved data mining technology | |
| CN114117349A (en) | Power system random variable probability analysis method, system, equipment and storage medium | |
| CN109493065A (en) | A kind of fraudulent trading detection method of Behavior-based control incremental update | |
| CN112738724A (en) | Method, device, equipment and medium for accurately identifying regional target crowd | |
| Tan et al. | Research on Fault Prediction Model Based on 5G Data Center | |
| CN116541252B (en) | Computer room fault log data processing method and device |
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 |