CN108388673B - Computer system for economic management analysis data - Google Patents

Computer system for economic management analysis data Download PDF

Info

Publication number
CN108388673B
CN108388673B CN201810250063.0A CN201810250063A CN108388673B CN 108388673 B CN108388673 B CN 108388673B CN 201810250063 A CN201810250063 A CN 201810250063A CN 108388673 B CN108388673 B CN 108388673B
Authority
CN
China
Prior art keywords
module
economic management
data
search
particle
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
Application number
CN201810250063.0A
Other languages
Chinese (zh)
Other versions
CN108388673A (en
Inventor
罗萍
王小茜
陈渊远
罗衡宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Polytechnic College
Original Assignee
Shandong Polytechnic College
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong Polytechnic College filed Critical Shandong Polytechnic College
Priority to CN201810250063.0A priority Critical patent/CN108388673B/en
Publication of CN108388673A publication Critical patent/CN108388673A/en
Application granted granted Critical
Publication of CN108388673B publication Critical patent/CN108388673B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/242—Query formulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Economics (AREA)
  • Evolutionary Biology (AREA)
  • Strategic Management (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Educational Administration (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Software Systems (AREA)
  • Genetics & Genomics (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Game Theory and Decision Science (AREA)
  • Physiology (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention belongs to the technical field of data analysis, and discloses a computer system for economic management and analysis of data, which comprises: the system comprises an input module, a genetic algorithm module, an output module, a linearization transformation module, a calling module, an economic management database and a modeling prediction module; the input module is connected with the genetic algorithm module and the linearization transformation module; the genetic algorithm module is connected with the output module; the linear transformation module is connected with the calling module and the modeling prediction module; the calling module is connected with the economic management database. The invention provides a systematic and logical analysis method, which comprehensively considers several factors such as requirements, production, cost, market and the like and adopts a genetic algorithm to calculate the optimal solution of a multivariable problem; and lographo transformation and modeling can be performed according to data in the database, so that economic management data can be predicted.

Description

Computer system for economic management analysis data
Technical Field
The invention belongs to the technical field of data search, and particularly relates to a computer system for economic management and analysis of data.
Background
Currently, the current state of the art commonly used in the industry is such that:
the management economics is a branch of the application economics, provides a systematic and logical search method for the management decisions, focuses on influencing daily decisions and the economic power of long-term planning decisions, is the application of the microscopic economics in the management practice, is a bridge for communicating the economics theory with the enterprise management decisions, provides a search tool and a search method for enterprise decision and management, and is mainly provided by the theory around several factors such as demand, production, cost, market and the like. At present, a computer system for data search is low in efficiency and accuracy, a more advanced calculation method is not adopted, and the development trend is difficult to predict.
The interactive matting technology is widely applied to the fields of image and video editing, three-dimensional reconstruction and the like, and has extremely high application value under the limited prospect of matting images under user interaction. In the recent matting technology, a Laplace matrix gives a linear relation among pixels on an alpha image, and plays an important role in estimating the alpha image. Interactive matting is the computation of an alpha map of the foreground under limited user interaction, separating the foreground from the background. The input of the matting problem is an original image I and a trimap image provided by a user, and the output is an alpha image, a foreground F and a background B, so that the matting problem is a typical ill-conditioned problem and needs to introduce an assumed condition to solve the alpha image. Matting algorithms can be divided into three categories: a sampling-based method, a propagation-based method, a combined sampling and propagation method.
The linear relation among alpha values of neighborhood pixels is given by a Laplace matting matrix deduced by the prior art, and the method is widely applied to a matting algorithm; the Laplace matting matrix has the limitation, the Laplace matting matrix represents the relationship among pixels in the spatial neighborhood, but cannot reflect the relationship among pixels in non-neighborhoods; the Laplace matting matrix is established on the basis of the assumption of space continuity, and in some regions with abrupt change of foreground and background components, the Laplace matting matrix is difficult to obtain ideal effects.
The existing genetic algorithm has strong dependence on the quality and the size of an initial population, requires a large proportion of feasible search programs in the initial population, and may not obtain the optimal search program or even may not converge. The selection difficulty of the pheromone residual coefficient and the parameter in the transition probability formula is high when the ant colony algorithm searches, the convergence rate of the algorithm is not ideal, and the algorithm is easy to fall into a local optimal solution. The discrete optimization problem is not well processed and is easy to fall into local optimization.
In summary, the prior art has the following problems:
at present, a computer system for data search has low efficiency and insufficient accuracy, does not adopt a more advanced calculation method, and is difficult to predict development trends.
The derived Laplace matrix in the prior art cannot reflect the relationship between pixels in non-neighborhoods; in some regions with abrupt changes of foreground and background components, the Laplace matrix is difficult to obtain the ideal effect.
The existing search procedure is not ideal in speed.
Disclosure of Invention
In view of the problems of the prior art, the present invention provides a computer system for economic management of analytical data.
The present invention is thus implemented, a computer system for economic management of analytical data, comprising:
the input module is connected with the genetic algorithm module and the linearization transformation module and is used for inputting economic management data;
the genetic algorithm module is connected with the output module and used for calculating the input data by adopting a genetic algorithm;
the genetic algorithm module constructs an interference matrix according to the incidence relation among the analyzed economic management data to obtain a feasible analysis program; constructing a fitness function suitable for a universal gravitation search algorithm;
redefining and modifying a calculation formula of the universal gravitation search algorithm to construct a new universal gravitation search calculation formula; iteratively solving the economic management data to be analyzed by adopting a new universal gravitation search calculation formula to obtain an optimal and feasible analysis program;
the output module is connected with the genetic algorithm module and used for outputting the calculation result of the genetic algorithm module;
the linear transformation module is connected with the calling module and the modeling prediction module and used for carrying out logatio transformation on input data and linearizing a nonlinear problem;
the linear transformation module comprises a logritio transformation module, a mobile least square method modeling module, an optimization module, a computer processing module and a statistical analysis module;
the logrito transformation module is used for reflecting the change situation of the absolute quantity of the real data value of the economic management analysis data through an Euclidean matrix and reflecting the relative change situation of the data;
the mobile least square method modeling module finds out the corresponding linear correlation function according to the independent variable and the dependent variable in the data by a mobile least square method, and analyzes the economic management picture data according to the fitted linear function;
the method specifically comprises the following steps: moving the least squares modeling Module in the grayscale image, Window wiThe alpha value in the neighborhood satisfies the local linear condition, and the local linear relation is solved by using a moving least square method, which is expressed as follows:
Figure GDA0002955780080000031
weight ω, ω in equation (1)iIs the neighborhood wkThe weight value in (1); formula (1) is represented in the form of the following matrix:
Figure GDA0002955780080000032
for each neighborhood wk,GkIs defined as | wkA | × 2 matrix; gkEach row comprising a vector (I)i,1)，WkIs the weight omega corresponding to each row vectoriVector of composition, Gk' is GkW of (2)kWeighting, the corresponding per-row vector is represented as (W)k.Ii,Wk)，
Figure GDA0002955780080000033
The vector is composed of alpha values corresponding to all pixels in the neighborhood;
coefficient ak,bkThe solution is as follows:
Figure GDA0002955780080000034
order to
Figure GDA0002955780080000041
J (α) is represented by the following formula:
Figure GDA0002955780080000042
Figure GDA0002955780080000043
δi,jis the Kronecker delta function, mukAnd σ2Respectively a small window wkInner based on WkWeighted mean and variance, | wkII is the number of pixels in the window, L is the moving Las matting matrix;
introducing a weight omegaiThe method is applied to a color model, and the moving least square matting method under the color model is as follows:
the linear relationship between the channels of the color image is represented by:
Figure GDA0002955780080000044
c is the number of channels of the color image, and after considering the information of each channel, equation (1) is converted into the following equation:
Figure GDA0002955780080000045
after the formula (2) is simplified, the mobile Laplace matrix under the color model is obtained by solving the following formula:
J(α)＝αLαT；
Figure GDA0002955780080000046
in formula (3), I is a matrix composed of 3 × 1 color vectors corresponding to all pixels in a small neighborhood, and μkW of IkWeighted average, ΣkIs I at WkA covariance matrix under weighting;
the optimization module converts the economic management analysis problem into a single-target solution by linearizing data, establishes different mathematical models according to different independent variables and dependent variables, optimizes various solution sets into a single optimal solution by optimization, and eliminates inferior solution sets which do not meet actual requirements;
the computer processing module is used for receiving, storing, converting, transmitting, releasing and calculating the independent variable, the dependent variable, the established mathematical model and the obtained optimal solution;
the statistical analysis module is used for counting the intermediate data and inferior solution set generated in the modeling, optimizing and calculating processes of the computer and analyzing the reason of generating the intermediate data in a unified way;
and the calling module is connected with the economic management database and is used for calling the data in the economic management database.
Further, the interference matrix is as follows:
Figure GDA0002955780080000051
wherein Ma is an interference matrix, a is the economic management data searching direction, and a belongs to { +/-, +/-y, +/-z }; c1 C2 ... CnRepresenting each economic management data individual to be searched; n is the number of economic management data to be searched; c ij1 is indicated in the economic management data CiWhen searching along the direction a, the economic management data C is associated withjConfusion occurs; economic management data not to be confused with itself, Cii＝0。
Further, the fitness function is:
Figure GDA0002955780080000052
where Fit (t) is a fitness function, f (X)i) Represents the search time of the economic management data i; qi(k,k+1)Represents the search time taken to complete the search process from the k-th economic management data to the k + 1-th economic management data, Qi(k,k+1)＝d·Di(k,k+1)+k·Ti(k,k+1)+l·Li(k,k+1)；Di(k,k+1)For the number of changes of search direction, Ti(k,k+1)For the number of times of replacement of search tools, Li(k,k+1)For the number of changes of search type, k ∈ [1, N-1 ]](ii) a d is a weight coefficient in the search time at the time of reorientation of the search direction, k is a weight coefficient in the search time at which the search tool is replaced, l is a weight coefficient in the total search time at which the change of the search type is satisfied, and d + k + l is 1.
Further, the new universal gravitation search calculation formula is as follows:
Figure GDA0002955780080000053
wherein, the economic management data i to be searched is represented by particles i, Fi d(t) is the resultant force of universal attraction of the particles i, and Rand represents a random number whose value range is [0, 1 ]]，
Figure GDA0002955780080000054
Wherein xi d(t) is the position of particle i in d-dimensional space at time t, FijShowing that the particle i is subjected to the universal gravitation of the particle j, G (t) is a universal gravitation constant,
Figure GDA0002955780080000061
alpha is the attenuation coefficient, G0Is an initial gravitational constant, T is a time period, ε is a small value constant, MpiRepresenting mass of passive attraction, MajRepresenting the mass of active gravity, Rij(t) is the Euclidean distance between particle i and particle j at time t,
Figure GDA0002955780080000062
i, j ═ 1, 2.., n, where x isi，xjIs the position of the particle i, j in space;
the iterative solution of the economic management data to be analyzed by adopting a new universal gravitation search calculation formula comprises the following steps:
1) population size determination and initialization
The economic management data to be searched is provided with N economic management data which form an N-dimensional search space, and the population is marked as X ═ X1，x2，x3，…xN) The ith particle position is labeled: xi＝(xi 1，xi 2，xi 3，…，xi d，…xi N)i＝1,2,3,…N；
2) Setting the maximum iteration times and calculating the quality:
setting the initial iteration value T as 0 and the maximum iteration time T as 100, and calculating the Fit of the particle at the time T according to the fitness function formulai(t) defining the minimum sorting rule for solving the problem, and solving the values of the minimum problems Worst (t) and best (t) in the calculation process according to a new universal gravitation search calculation formula, wherein:
Figure GDA0002955780080000063
Figure GDA0002955780080000064
Figure GDA0002955780080000065
Figure GDA0002955780080000066
best (t) is the best fitness value of the population at time t, Worst (t) is the worst fitness value of the population at time t, Fitj(t) is the fitness value of the individual i at time t, Mi(t) is the particle inertial mass;
3) determining universal gravitation constant and calculating universal gravitation resultant force
Figure GDA0002955780080000067
Wherein, the maximum iteration number T is taken as 100, and the initial gravitational constant G0100, an attenuation coefficient alpha of 20,
Figure GDA0002955780080000071
taking the epsilon as 5, and taking the epsilon as the index,
Figure GDA0002955780080000072
4) calculating the acceleration a
Figure GDA0002955780080000073
5) Updating particle velocity and position
vi d(t+1)＝Rand×vi d(t)+ai d(t)
xi d(t+1)＝xi d(t)+vi d(t+1)
Wherein v isi d(t +1) is the speed of the particle i in the d-dimensional space at the moment of t +1, and Rand represents the value range of [0, 1%]Random number of vi d(t) is the velocity of particle i in d-dimensional space at time t, ai d(t) is the acceleration of the particle i in the d-dimensional space at time t;
xi d(t +1) is the position of particle i in the d-dimensional space at time t +1, xi d(t) is the position of the particle i in the d-dimensional space at time t, vi d(t +1) is the velocity of particle i in the d-dimensional space at time t + 1;
6) judging whether an iteration end condition is reached or not, and outputting the most feasible search program;
when the preset maximum iteration number is reached, the circulation is stopped, and the position value x of each particle at the moment is outputi dWhile simultaneously applying x to each particlei dAnd sequencing the output values from small to large, wherein the obtained sequencing sequence is the optimal search sequence.
Further, the computer system for economic management of search data further comprises: and the economic management database is connected with the calling module and used for storing economic management data.
Further, the computer system for economic management of analytical data further comprises: and the modeling prediction module is connected with the linear transformation module and used for establishing a regression model and predicting the economic management condition at a certain moment according to the calculation result.
Further, the basic operation process of the genetic algorithm in the genetic algorithm module is as follows:
a) initialization: setting an evolution algebra counter T to be 0, setting a maximum evolution algebra T, and randomly generating M individuals as an initial population P (0);
b) individual evaluation: calculating the fitness of each individual in the population P (t);
c) selecting operation, namely acting the selecting operator on the group;
d) and (3) cross operation: applying a crossover operator to the population;
e) and (3) mutation operation: acting mutation operators on the population; the group P (t) is subjected to selection, intersection and mutation operation to obtain a next generation group P (t + 1);
f) and (5) judging a termination condition, namely if T is T, outputting the individual with the maximum fitness obtained in the evolution process as an optimal solution, and terminating the calculation.
Furthermore, the input module is provided with a base, an input keyboard is arranged in the middle of the base, the left side of the input keyboard is provided with USB jacks, the right side of the input keyboard is provided with a handwriting input part, and the number of the USB jacks on the left side of the input keyboard is 4.
In summary, the advantages and positive effects of the invention are:
the invention provides a systematic and logical search method, which comprehensively considers several factors such as requirements, production, cost, market and the like and adopts a genetic algorithm to calculate the optimal solution of a multivariable problem; and lographo transformation and modeling can be performed according to data in the database, so that economic management data can be predicted.
The method provided by the invention has the advantages that the method can obtain better effects in the areas with complex foreground and areas with complex mixed foreground and background. Deriving a moving Laplace matrix by using minimum moving quadratic multiplication instead of a least square method; compared with the least square method, the linear condition solved by the moving least square method is more accurate; the KNN neighborhood is used to replace the spatial neighborhood so that the Laplace matrix can reflect the relationship of the alpha values of the non-inter-neighborhood pixels. The method uses the moving least square method to solve the alpha image according to the matrix, so that the foreground matting processing can be carried out on the image under the complex background, compared with the prior method, the method is more effective, the more accurate alpha image can be solved, and good effects can be obtained in the areas with complex foreground and background in the image, particularly in the mixed areas of the foreground and background colors and the areas with local holes and large changes.
The present invention focuses search agents together due to their mutual attraction to each other according to newton's law of gravity and newton's second law of motion. Experimental results show that the gravity search algorithm has high superiority in solving various nonlinear functions. In the process of continuously updating the individual mass in the mobile search, the individual mass with excellent fitness value is larger, the attraction generated in the interaction motion process is larger, the mass of the individual mass is larger, the movement is slow, and the movement of the individual with small mass is relatively quick. Therefore, the whole population continuously moves towards individuals with excellent fitness values, and the purposes of realizing information interaction and excellent individual guide search among individuals and moving the whole population towards an excellent solution direction are achieved. The search quality and speed are improved.
Drawings
FIG. 1 is a block diagram of a computer system for economic management of analytical data according to an embodiment of the present invention;
FIG. 2 is a block diagram of an input module of a computer system for economic management of analytical data according to an embodiment of the present invention;
FIG. 3 is a block diagram of a linear transformation module of a computer system for economic management of analytical data according to an embodiment of the present invention;
in the figure: 1. an input module; 2. a genetic algorithm module; 3. an output module; 4. a linearization transformation module; 5. calling a module; 6. an economic management database; 7. a modeling prediction module; 8. a base; 9. a USB jack; 10. an input keyboard; 11. a handwriting input section; 12. a logrito transform module; 13. a mobile least square method modeling module; 14. an optimization module; 15. a computer processing module; 16. and a statistic searching 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 to 3, a computer system for economic management of analysis data according to an embodiment of the present invention includes: the system comprises an input module 1, a genetic algorithm module 2, an output module 3, a linearization transformation module 4, a calling module 5, an economic management database 6 and a modeling prediction module 7.
The input module 1 is connected with the genetic algorithm module 2 and the linearization transformation module 4 and is used for inputting economic management data;
the input module 1 is provided with a base 8, an input keyboard 10 is arranged in the middle of the base 8, a USB jack 9 is arranged on the left side of the input keyboard 10, and a handwriting input part 11 is arranged on the right side of the input keyboard 10.
The genetic algorithm module 2 is connected with the output module 3 and used for calculating the input data by adopting a genetic algorithm;
the output module 3 is connected with the genetic algorithm module 2 and is used for outputting the calculation result of the genetic algorithm module 2;
the linearization transformation module 4 is connected with the calling module 5 and the modeling prediction module 7 and is used for carrying out logritio transformation on input data and linearizing a nonlinear problem;
the linearization transformation module 4 includes: a logrito transformation module 12, a least square modeling module 13, an optimization module 14, a computer processing module 15 and a statistic search module 16.
The logrito transformation module 12 is used for reflecting the change situation of the absolute quantity of the real data value of the economic management search data through an Euclidean matrix, and simultaneously reflecting the relative change situation of the data.
The mobile least square method modeling module 13 finds out the corresponding linear correlation function by the independent variable and the dependent variable in the data through the mobile least square method, so that the disordered data which seems to be irrelevant becomes more specific and transparent, and searches the economic management data through the fitted linear function, so that the data are more linear, and the model is more stable.
The optimization module 14 converts the economic management search problem into a single-target solution problem by linearizing the data, can establish different mathematical models according to different independent variables and dependent variables, and optimizes various solution sets into a single optimal solution by optimization, thereby eliminating inferior solution sets which do not meet the actual production requirements.
And the computer processing module 15 is used for performing information calculation processing such as receiving, storing, converting, transmitting, issuing and the like on the independent variable, the dependent variable, the established mathematical model and the obtained optimal solution, and is finally suitable for solving real practical problems.
The statistical search module 16 is used for counting a series of intermediate data and inferior solution sets generated in the modeling, optimizing and calculating processes of the computer, counting the characteristic conditions of the data, searching the reasons for generating the intermediate data, avoiding the intermediate data, reducing the inferior solution sets, accelerating the generation of the optimal solution and reducing the time cost when processing the searched data next time.
The calling module 5 is connected with the economic management database 6 and used for calling data in the economic management database 6;
the economic management database 6 is connected with the calling module 5 and used for storing economic management 6 data;
and the modeling prediction module 7 is connected with the linearization transformation module 4 and used for establishing a regression model and predicting the economic management condition at a certain moment according to the calculation result.
The basic operation process of the genetic algorithm in the genetic algorithm module 2 provided by the invention is as follows:
a) initialization: setting an evolution algebra counter T to be 0, setting a maximum evolution algebra T, and randomly generating M individuals as an initial population P (0).
b) Individual evaluation: calculating the fitness of each individual in the population P (t).
c) And (4) selecting operation, namely acting a selection operator on the group. The purpose of selection is to inherit optimized individuals directly to the next generation or to generate new individuals by pairwise crossing and then to inherit them to the next generation. The selection operation is based on fitness evaluation of individuals in the population.
d) And (3) cross operation: the crossover operator is applied to the population. What plays a core role in genetic algorithms is the crossover operator.
e) And (3) mutation operation: and (4) acting mutation operators on the population. I.e., to vary the gene values at certain loci of the individual strings in the population.
And (t) obtaining a next generation group P (t +1) after selection, crossing and mutation operations of the group P (t).
f) And (5) judging a termination condition, namely if T is equal to T, outputting the individual with the maximum fitness obtained in the evolution process as an optimal solution, and terminating the calculation.
In the invention, a user inputs economic management data from an input module 1, a genetic algorithm module 2 adopts a genetic algorithm to calculate the input data, and then an optimal solution of a multivariable problem calculated based on the comprehensive consideration of several factors such as demand, production, cost, market and the like is output in an output module 3. The linearization transformation module 4 can perform logritio transformation on the input data, linearize the nonlinear problem, establish a regression model in the modeling prediction module 7, and predict the economic management condition at a certain moment according to the calculation result. The invention provides a systematic and logical search method, which comprehensively considers several factors such as requirements, production, cost, market and the like and adopts a genetic algorithm to calculate the optimal solution of a multivariable problem; and lographo transformation and modeling can be performed according to data in the database, so that economic management data can be predicted.
The invention is further described below with reference to specific assays.
The genetic algorithm module constructs an interference matrix according to the incidence relation among the analyzed economic management data to obtain a feasible analysis program; constructing a fitness function suitable for a universal gravitation search algorithm;
redefining and modifying a calculation formula of the universal gravitation search algorithm to construct a new universal gravitation search calculation formula; iteratively solving the economic management data to be analyzed by adopting a new universal gravitation search calculation formula to obtain an optimal and feasible analysis program;
the output module is connected with the genetic algorithm module and used for outputting the calculation result of the genetic algorithm module;
the linear transformation module is connected with the calling module and the modeling prediction module and used for carrying out logatio transformation on input data and linearizing a nonlinear problem;
the linear transformation module comprises a logritio transformation module, a mobile least square method modeling module, an optimization module, a computer processing module and a statistical analysis module;
the logrito transformation module is used for reflecting the change situation of the absolute quantity of the real data value of the economic management analysis data through an Euclidean matrix and reflecting the relative change situation of the data;
the mobile least square method modeling module finds out the corresponding linear correlation function according to the independent variable and the dependent variable in the data by a mobile least square method, and analyzes the economic management picture data according to the fitted linear function;
the method specifically comprises the following steps: moving the least squares modeling Module in the grayscale image, Window wiThe alpha value in the neighborhood satisfies the local linear condition, and the local linear relation is solved by using a moving least square method, which is expressed as follows:
Figure GDA0002955780080000121
weight ω in equation (1)，ωiIs the neighborhood wkThe weight value in (1); formula (1) is represented in the form of the following matrix:
Figure GDA0002955780080000131
for each neighborhood wk,GkIs defined as | wkA | × 2 matrix; gkEach row comprising a vector (I)i,1)，WkIs the weight omega corresponding to each row vectoriVector of composition, Gk' is GkW of (2)kWeighting, the corresponding per-row vector is represented as (W)k.Ii,Wk)，
Figure GDA0002955780080000132
The vector is composed of alpha values corresponding to all pixels in the neighborhood;
coefficient ak,bkThe solution is as follows:
Figure GDA0002955780080000133
order to
Figure GDA0002955780080000134
J (α) is represented by the following formula:
Figure GDA0002955780080000135
Figure GDA0002955780080000136
δi,jis the Kronecker delta function, mukAnd σ2Respectively a small window wkInner based on WkWeighted mean and variance, | wkII is the number of pixels in the window, L is the moving Las matting matrix;
introducing a weight omegaiApplication ofTo the color model, the moving least square matting method under the color model is as follows:
the linear relationship between the channels of the color image is represented by:
Figure GDA0002955780080000137
c is the number of channels of the color image, and after considering the information of each channel, equation (1) is converted into the following equation:
Figure GDA0002955780080000138
after the formula (2) is simplified, the mobile Laplace matrix under the color model is obtained by solving the following formula:
J(α)＝αLαT；
Figure GDA0002955780080000139
in formula (3), I is a matrix composed of 3 × 1 color vectors corresponding to all pixels in a small neighborhood, and μkW of IkWeighted average, ΣkIs I at WkA covariance matrix under weighting;
the optimization module converts the economic management analysis problem into a single-target solution by linearizing data, establishes different mathematical models according to different independent variables and dependent variables, optimizes various solution sets into a single optimal solution by optimization, and eliminates inferior solution sets which do not meet actual requirements;
the computer processing module is used for receiving, storing, converting, transmitting, releasing and calculating the independent variable, the dependent variable, the established mathematical model and the obtained optimal solution;
the statistical analysis module is used for counting the intermediate data and inferior solution set generated in the modeling, optimizing and calculating processes of the computer and analyzing the reason of generating the intermediate data in a unified way;
and the calling module is connected with the economic management database and is used for calling the data in the economic management database.
Further, the interference matrix is as follows:
Figure GDA0002955780080000141
wherein Ma is an interference matrix, a is the economic management data searching direction, and a belongs to { +/-, +/-y, +/-z }; c1 C2 ... CnRepresenting each economic management data individual to be searched; n is the number of economic management data to be searched; c ij1 is indicated in the economic management data CiWhen searching along the direction a, the economic management data C is associated withjConfusion occurs; economic management data not to be confused with itself, Cii＝0。
Further, the fitness function is:
Figure GDA0002955780080000142
where Fit (t) is a fitness function, f (X)i) Represents the search time of the economic management data i; qi(k,k+1)Represents the search time taken to complete the search process from the k-th economic management data to the k + 1-th economic management data, Qi(k,k+1)＝d·Di(k,k+1)+k·Ti(k,k+1)+l·Li(k,k+1)；Di(k,k+1)For the number of changes of search direction, Ti(k,k+1)For the number of times of replacement of search tools, Li(k,k+1)For the number of changes of search type, k ∈ [1, N-1 ]](ii) a d is a weight coefficient in the search time at the time of reorientation of the search direction, k is a weight coefficient in the search time at which the search tool is replaced, l is a weight coefficient in the total search time at which the change of the search type is satisfied, and d + k + l is 1.
Further, the new universal gravitation search calculation formula is as follows:
Figure GDA0002955780080000151
wherein, the economic management data i to be searched is represented by particles i, Fi d(t) is the resultant force of universal attraction of the particles i, and Rand represents a random number whose value range is [0, 1 ]]，
Figure GDA0002955780080000152
Wherein xi d(t) is the position of particle i in d-dimensional space at time t, FijShowing that the particle i is subjected to the universal gravitation of the particle j, G (t) is a universal gravitation constant,
Figure GDA0002955780080000153
alpha is the attenuation coefficient, G0Is an initial gravitational constant, T is a time period, ε is a small value constant, MpiRepresenting mass of passive attraction, MajRepresenting the mass of active gravity, Rij(t) is the Euclidean distance between particle i and particle j at time t,
Figure GDA0002955780080000154
i, j ═ 1, 2.., n, where x isi，xjIs the position of the particle i, j in space;
the iterative solution of the economic management data to be analyzed by adopting a new universal gravitation search calculation formula comprises the following steps:
1) population size determination and initialization
The economic management data to be searched is provided with N economic management data which form an N-dimensional search space, and the population is marked as X ═ X1，x2，x3，…xN) The ith particle position is labeled: xi＝(xi 1，xi 2，xi 3，…，xi d，…xi N)i＝1,2,3,…N；
2) Setting the maximum iteration times and calculating the quality:
setting the initial iteration value T as 0 and the maximum iteration time T as 100, and calculating according to the fitness function formulaFit of particle at time ti(t) defining the minimum sorting rule for solving the problem, and solving the values of the minimum problems Worst (t) and best (t) in the calculation process according to a new universal gravitation search calculation formula, wherein:
Figure GDA0002955780080000161
Figure GDA0002955780080000162
Figure GDA0002955780080000163
Figure GDA0002955780080000164
best (t) is the best fitness value of the population at time t, Worst (t) is the worst fitness value of the population at time t, Fitj(t) is the fitness value of the individual i at time t, Mi(t) is the particle inertial mass;
3) determining universal gravitation constant and calculating universal gravitation resultant force
Figure GDA0002955780080000165
Wherein, the maximum iteration number T is taken as 100, and the initial gravitational constant G0100, an attenuation coefficient alpha of 20,
Figure GDA0002955780080000166
taking the epsilon as 5, and taking the epsilon as the index,
Figure GDA0002955780080000167
4) calculating the acceleration a
Figure GDA0002955780080000168
5) Updating particle velocity and position
vi d(t+1)＝Rand×vi d(t)+ai d(t)
xi d(t+1)＝xi d(t)+vi d(t+1)
Wherein v isi d(t +1) is the speed of the particle i in the d-dimensional space at the moment of t +1, and Rand represents the value range of [0, 1%]Random number of vi d(t) is the velocity of particle i in d-dimensional space at time t, ai d(t) is the acceleration of the particle i in the d-dimensional space at time t;
xi d(t +1) is the position of particle i in the d-dimensional space at time t +1, xi d(t) is the position of the particle i in the d-dimensional space at time t, vi d(t +1) is the velocity of particle i in the d-dimensional space at time t + 1;
6) judging whether an iteration end condition is reached or not, and outputting the most feasible search program;
when the preset maximum iteration number is reached, the circulation is stopped, and the position value x of each particle at the moment is outputi dWhile simultaneously applying x to each particlei dAnd sequencing the output values from small to large, wherein the obtained sequencing sequence is the optimal search sequence.
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 (8)

1. A computer system for economic management of analytical data, the computer system for economic management of analytical data comprising:
the input module is connected with the genetic algorithm module and the linearization transformation module and is used for inputting economic management data;
the genetic algorithm module is connected with the output module and used for calculating the input data by adopting a genetic algorithm;
the genetic algorithm module constructs an interference matrix according to the incidence relation among the analyzed economic management data to obtain a feasible analysis program; constructing a fitness function suitable for a universal gravitation search algorithm;
redefining and modifying a calculation formula of the universal gravitation search algorithm to construct a new universal gravitation search calculation formula; iteratively solving the economic management data to be analyzed by adopting a new universal gravitation search calculation formula to obtain an optimal and feasible analysis program;
the output module is connected with the genetic algorithm module and used for outputting the calculation result of the genetic algorithm module;
the linear transformation module is connected with the calling module and the modeling prediction module and used for carrying out logatio transformation on input data and linearizing a nonlinear problem;
the linear transformation module comprises a logritio transformation module, a mobile least square method modeling module, an optimization module, a computer processing module and a statistical analysis module;
the logrito transformation module is used for reflecting the change situation of the absolute quantity of the real data value of the economic management analysis data through an Euclidean matrix and reflecting the relative change situation of the data;
the mobile least square method modeling module finds out the corresponding linear correlation function according to the independent variable and the dependent variable in the data by a mobile least square method, and analyzes the economic management picture data according to the fitted linear function;
the method specifically comprises the following steps: moving the least squares modeling Module in the grayscale image, Window wiThe alpha value in the neighborhood satisfies the local linear condition, and the local linear relation is solved by using a moving least square method, which is expressed as follows:
Figure FDA0002955780070000011
weight ω, ω in equation (1)iIs the neighborhood wkThe weight value in (1); formula (1) is represented in the form of the following matrix:
Figure FDA0002955780070000021
for each neighborhood wk,GkIs defined as | wkA | × 2 matrix; gkEach row comprising a vector (I)i,1)，WkIs the weight omega corresponding to each row vectoriVector of composition, Gk' is GkW of (2)kWeighting, the corresponding per-row vector is represented as (W)k.Ii,Wk)，
Figure FDA0002955780070000022
The vector is composed of alpha values corresponding to all pixels in the neighborhood;
coefficient ak,bkThe solution is as follows:
Figure FDA0002955780070000023
order to
Figure FDA0002955780070000024
J (α) is represented by the following formula:
Figure FDA0002955780070000025
Figure FDA0002955780070000026
δi,jis the Kronecker delta function, mukAnd σ2Respectively a small window wkInner based on WkIs added withWeight mean and variance, | wkII is the number of pixels in the window, L is the moving Las matting matrix;
introducing a weight omegaiThe method is applied to a color model, and the moving least square matting method under the color model is as follows:
the linear relationship between the channels of the color image is represented by:
Figure FDA0002955780070000027
c is the number of channels of the color image, and after considering the information of each channel, equation (1) is converted into the following equation:
Figure FDA0002955780070000028
after the formula (2) is simplified, the mobile Laplace matrix under the color model is obtained by solving the following formula:
J(α)＝αLαT；
Figure FDA0002955780070000031
in formula (3), I is a matrix composed of 3 × 1 color vectors corresponding to all pixels in a small neighborhood, and μkW of IkWeighted average, ΣkIs I at WkA covariance matrix under weighting;
the optimization module converts the economic management analysis problem into a single-target solution by linearizing data, establishes different mathematical models according to different independent variables and dependent variables, optimizes various solution sets into a single optimal solution by optimization, and eliminates inferior solution sets which do not meet actual requirements;
the computer processing module is used for receiving, storing, converting, transmitting, releasing and calculating the independent variable, the dependent variable, the established mathematical model and the obtained optimal solution;
the statistical analysis module is used for counting the intermediate data and inferior solution set generated in the modeling, optimizing and calculating processes of the computer and analyzing the reason of generating the intermediate data in a unified way;
and the calling module is connected with the economic management database and is used for calling the data in the economic management database.
2. The computer system for economic management of analytical data of claim 1,
the interference matrix is as follows:
Figure FDA0002955780070000032
wherein Ma is an interference matrix, a is the economic management data searching direction, and a belongs to { +/-, +/-y, +/-z }; c1 C2 ... CnRepresenting each economic management data individual to be searched; n is the number of economic management data to be searched; cij1 is indicated in the economic management data CiWhen searching along the direction a, the economic management data C is associated withjConfusion occurs; economic management data not to be confused with itself, Cii＝0。
3. The computer system for economic management of analytical data of claim 1, wherein the fitness function is:
Figure FDA0002955780070000033
where Fit (t) is a fitness function, f (X)i) Represents the search time of the economic management data i; qi(k,k+1)Represents the search time taken to complete the search process from the k-th economic management data to the k + 1-th economic management data, Qi(k,k+1)＝d·Di(k,k+1)+k·Ti(k,k+1)+l·Li(k,k+1)；Di(k,k+1)For the number of changes of search direction, Ti(k,k+1)As search toolsNumber of replacements of, Li(k,k+1)For the number of changes of search type, k ∈ [1, N-1 ]](ii) a d is a weight coefficient in the search time at the time of reorientation of the search direction, k is a weight coefficient in the search time at which the search tool is replaced, l is a weight coefficient in the total search time at which the change of the search type is satisfied, and d + k + l is 1.
4. The computer system for economic management of analytical data of claim 1, wherein the new gravitational search calculation formula is as follows:
Figure FDA0002955780070000041
wherein, the economic management data i to be searched is represented by particles i, Fi d(t) is the resultant force of universal attraction of the particles i, and Rand represents a random number whose value range is [0, 1 ]]，
Figure FDA0002955780070000042
Wherein xi d(t) is the position of particle i in d-dimensional space at time t, FijShowing that the particle i is subjected to the universal gravitation of the particle j, G (t) is a universal gravitation constant,
Figure FDA0002955780070000043
alpha is the attenuation coefficient, G0Is an initial gravitational constant, T is a time period, ε is a small value constant, MpiRepresenting mass of passive attraction, MajRepresenting the mass of active gravity, Rij(t) is the Euclidean distance between particle i and particle j at time t,
Figure FDA0002955780070000044
i, j ═ 1, 2.., n, where x isi，xjIs the position of the particle i, j in space;
the iterative solution of the economic management data to be analyzed by adopting a new universal gravitation search calculation formula comprises the following steps:
1) population size determination and initialization
The economic management data to be searched is provided with N economic management data which form an N-dimensional search space, and the population is marked as X ═ X1，x2，x3，…xN) The ith particle position is labeled: xi＝(xi 1，xi 2，xi 3，...，xi d，...xi N)i＝1,2,3,···N；
2) Setting the maximum iteration times and calculating the quality:
setting the initial iteration value T as 0 and the maximum iteration time T as 100, and calculating the Fit of the particle at the time T according to the fitness function formulai(t) defining the minimum sorting rule for solving the problem, and solving the values of the minimum problems Worst (t) and best (t) in the calculation process according to a new universal gravitation search calculation formula, wherein:
Figure FDA0002955780070000051
Figure FDA0002955780070000052
Figure FDA0002955780070000053
Figure FDA0002955780070000054
best (t) is the best fitness value of the population at time t, Worst (t) is the worst fitness value of the population at time t, Fitj(t) is the fitness value of the individual i at time t, Mi(t) is the particle inertial mass;
3) determining universal gravitation constant and calculating universal gravitation resultant force
Figure FDA0002955780070000055
Wherein, the maximum iteration number T is taken as 100, and the initial gravitational constant G0100, an attenuation coefficient alpha of 20,
Figure FDA0002955780070000056
taking the epsilon as 5, and taking the epsilon as the index,
Figure FDA0002955780070000057
4) calculating the acceleration a
Figure FDA0002955780070000058
5) Updating particle velocity and position
vi d(t+1)＝Rand×vi d(t)+ai d(t)
xi d(t+1)＝xi d(t)+vi d(t+1)
Wherein v isi d(t +1) is the speed of the particle i in the d-dimensional space at the moment of t +1, and Rand represents the value range of [0, 1%]Random number of vi d(t) is the velocity of particle i in d-dimensional space at time t, ai d(t) is the acceleration of the particle i in the d-dimensional space at time t;
xi d(t +1) is the position of particle i in the d-dimensional space at time t +1, xi d(t) is the position of the particle i in the d-dimensional space at time t, vi d(t +1) is the velocity of particle i in the d-dimensional space at time t + 1;
6) judging whether an iteration end condition is reached or not, and outputting the most feasible search program;
when the preset maximum iteration number is reached, the circulation is stopped, and the current time is outputPosition value x of each particlei dWhile simultaneously applying x to each particlei dAnd sequencing the output values from small to large, wherein the obtained sequencing sequence is the optimal search sequence.
5. The computer system for economic management of analytical data of claim 1, wherein the computer system for economic management of search data further comprises: and the economic management database is connected with the calling module and used for storing economic management data.
6. The computer system for economic management of analytical data of claim 1, wherein the computer system for economic management of analytical data further comprises: and the modeling prediction module is connected with the linear transformation module and used for establishing a regression model and predicting the economic management condition at a certain moment according to the calculation result.
7. The computer system for economic management of analytical data of claim 1, wherein the basic operation process of the genetic algorithm in the genetic algorithm module is as follows:
a) initialization: setting an evolution algebra counter T to be 0, setting a maximum evolution algebra T, and randomly generating M individuals as an initial population P (0);
b) individual evaluation: calculating the fitness of each individual in the population P (t);
c) selecting operation, namely acting the selecting operator on the group;
d) and (3) cross operation: applying a crossover operator to the population;
e) and (3) mutation operation: acting mutation operators on the population; the group P (t) is subjected to selection, intersection and mutation operation to obtain a next generation group P (t + 1);
f) and (5) judging a termination condition, namely if T is T, outputting the individual with the maximum fitness obtained in the evolution process as an optimal solution, and terminating the calculation.
8. The computer system for economic management of analytical data according to claim 1, wherein the input module is provided with a base, an input keyboard is installed in the middle of the base, USB jacks are provided on the left side of the input keyboard, a handwriting input part is provided on the right side of the input keyboard, and the number of the USB jacks on the left side of the input keyboard is 4.
CN201810250063.0A 2018-03-26 2018-03-26 Computer system for economic management analysis data Active CN108388673B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810250063.0A CN108388673B (en) 2018-03-26 2018-03-26 Computer system for economic management analysis data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810250063.0A CN108388673B (en) 2018-03-26 2018-03-26 Computer system for economic management analysis data

Publications (2)

Publication Number Publication Date
CN108388673A CN108388673A (en) 2018-08-10
CN108388673B true CN108388673B (en) 2021-06-15

Family

ID=63072149

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810250063.0A Active CN108388673B (en) 2018-03-26 2018-03-26 Computer system for economic management analysis data

Country Status (1)

Country Link
CN (1) CN108388673B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109491312A (en) * 2018-11-15 2019-03-19 广东水利电力职业技术学院（广东省水利电力技工学校） A kind of CNC Robot motion control core implementation method and control system based on Codesys
CN110413662B (en) * 2019-08-05 2021-11-26 中国地质大学（北京） Multichannel economic data input system, acquisition system and method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102506863A (en) * 2011-11-07 2012-06-20 北京航空航天大学 Universal gravitation search-based unmanned plane air route planning method
CN103646178A (en) * 2013-12-18 2014-03-19 中国石油大学(华东) Multi-objective optimization method based on improved gravitation search algorithm
CN104794278A (en) * 2015-04-21 2015-07-22 西安电子科技大学 Optimizing method for product assembly sequences

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8010535B2 (en) * 2008-03-07 2011-08-30 Microsoft Corporation Optimization of discontinuous rank metrics

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102506863A (en) * 2011-11-07 2012-06-20 北京航空航天大学 Universal gravitation search-based unmanned plane air route planning method
CN103646178A (en) * 2013-12-18 2014-03-19 中国石油大学(华东) Multi-objective optimization method based on improved gravitation search algorithm
CN104794278A (en) * 2015-04-21 2015-07-22 西安电子科技大学 Optimizing method for product assembly sequences

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于改进万有引力搜索算法的微网优化运行;李鹏等;《中国电机工程学报》;20140705;第34卷(第19期);第3073-3079页 *

Also Published As

Publication number Publication date
CN108388673A (en) 2018-08-10

Similar Documents

Publication Publication Date Title
CN109165664B (en) Attribute-missing data set completion and prediction method based on generation of countermeasure network
CN106779087A (en) A kind of general-purpose machinery learning data analysis platform
Li et al. Dynamic structure embedded online multiple-output regression for streaming data
CN111339818A (en) Face multi-attribute recognition system
CN108388673B (en) Computer system for economic management analysis data
Li et al. Task relation networks
CN113221475A (en) Grid self-adaption method for high-precision flow field analysis
Fan et al. Adaptive partition intuitionistic fuzzy time series forecasting model
CN116629352A (en) Hundred million-level parameter optimizing platform
Zhang et al. Task Relatedness-Based Multitask Genetic Programming for Dynamic Flexible Job Shop Scheduling
Gong et al. Evolutionary computation in China: A literature survey
AL-Behadili Classification algorithms for determining handwritten digit
Bi et al. Accurate prediction of workloads and resources with multi-head attention and hybrid LSTM for cloud data centers
CN111708919B (en) Big data processing method and system
Chang et al. Multi-task learning for emotion descriptors estimation at the fourth abaw challenge
CN113095466A (en) Algorithm of satisfiability model theoretical solver based on meta-learning model
Artemov et al. Subsystem for simple dynamic gesture recognition using 3DCNNLSTM
Zhang et al. Learning to search efficient densenet with layer-wise pruning
Balazs et al. Hierarchical-interpolative fuzzy system construction by genetic and bacterial memetic programming approaches
CN116415177A (en) Classifier parameter identification method based on extreme learning machine
CN113835964B (en) Cloud data center server energy consumption prediction method based on small sample learning
Bo et al. Working set selection using functional gain for LS-SVM
CN116167492A (en) Closed-loop optimization self-adaptive scheduling method for semiconductor production line
Hu et al. Learning multi-expert distribution calibration for long-tailed video classification
Zheng et al. An evolutionary multitasking optimization algorithm via reference-point based nondominated sorting approach

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