CN103729492A

CN103729492A - Modeling method based on behavior flow function and behavior coupling relation

Info

Publication number: CN103729492A
Application number: CN201310471429.4A
Authority: CN
Inventors: 郝泳涛; 楼狄明; 王力生
Original assignee: Tongji University
Current assignee: Tongji University
Priority date: 2013-10-10
Filing date: 2013-10-10
Publication date: 2014-04-16
Anticipated expiration: 2033-10-10
Also published as: CN103729492B

Abstract

The invention provides a modeling method based on a behavior flow function and a behavior coupling relation. Behavior flow product modeling is achieved by means of coupling between a function and a behavior, and accordingly a behavior node sequence of modeling is formed. To achieve innovation of product modeling and design automation, the basic concept of behavior flow is put forward, solving is achieved by means of the coupling relation of the function and the behavior of the concept, product modeling is achieved in the solving process, and results of product modeling are selected by means of a genetic algorithm so as to obtain the optimal solution. As is shown by instance creation of an engine model, the method can achieve automation design and innovation of a product concept model.

Description

Based on the modeling method of behavior stream function and behavior coupled relation

Technical field

The present invention relates to Modeling in Product field, particularly relate to a kind of modeling method based on behavior stream function and behavior coupled relation.

Background technology

Along with the develop rapidly of science and technology and production technology, each enterprise is faced with and continues changeable competing with uncertain globalizing market, machinery manufacturing industry is also so, but existence in keen competition how, finally still sums up in the point that the innovation of product.Because conceptual design is in the first stage of product design, determining the function of product and the structure carrier of practical function.There are some researches show, 70% of value of the product determines by conceptual phase, and the conceptual design of product is the key link that realizes products innovation as can be seen here, but do not show that conceptual design is with regard to a design proposal [3] that produces surely novelty.The essence of innovation is to have with the acquisition of unconventional method the design proposal of novelty, must obtain from different thought angle a series of solutions of same problem in order to obtain having the design of novelty.

Function-behavior-structure (Function-Behavior-Structure) model is to be proposed by Qian and Gero the earliest, and this model has been carried out to comparatively systematic research, and provided corresponding mathematical model, and the relationship description between each mapping space, as shown in Figure 1.The rationality of this design framework structure of FBS obtains numerous researchers' approval, but is not but a lot of by the innovation theory that this framed structure realizes Modeling in Product.Therefore in the face of in fierce market competition, upgrading and the R&D capability that how can complete fast at short notice product have become a successful top priority of enterprise.But study, show, in the performance history of new product, 40% is to reuse parts in the past, the 40%th, and to the modification of existing parts, 20% is brand-new design.

Summary of the invention

In view of the above, the object of the present invention is to provide a kind of modeling method based on behavior stream function and behavior coupled relation, for realizing the Automation Design and the innovation of product concept model.

For achieving the above object and other relevant objects, the invention provides a kind of modeling method based on behavior stream function and behavior coupled relation, utilize coupling between function and behavior to realize the Modeling in Product of behavior stream, form the behavior sequence node of modeling.

Preferably, the Modeling in Product that utilizes coupling between function and behavior to realize behavior stream specifically comprises the following steps:

Step 1: search from function storehouse and whether exist ready-made behavior can realize this function, if not, arrive second step, if had, carry out step 3, and this subfunction is logged in the behavior path of Modeling in Product.

Step 2: function is decomposed, decompose each subfunction always and exist in function storehouse, return to step 1.

Step 3: according to selecting corresponding behavior in the behavior of input and relevant constraint condition subordinate act storehouse, export corresponding structure according to this behavior, and this structure is logged in the behavior path of Modeling in Product.

Step 4: inquire structural table according to structure and constraint condition from structure storehouse, according to structural table judgement, whether this structure is object construction unit, if it is directly exits, and corresponding structure is contributed in the behavior path of Modeling in Product; Otherwise export function corresponding to this structure, enter step 1 and circulate successively.

Preferably, further comprising the steps of: to use genetic algorithm, utilize described behavior sequence node to recombinate preferentially to behavior stream, obtain optimum behavior stream modeling path.

Preferably, use genetic algorithm, utilize described behavior sequence node to recombinate preferentially to behavior stream, obtain optimum behavior stream modeling path and specifically comprise the following steps:

1) behavior is flow to the coding of row binary, thereby obtain the gene code of behavioural matrix;

2) carry out successively the step of selecting, intersecting, make a variation;

3) determine algorithm evaluation function and end condition; The behavior stream sequence obtaining is optimum solution.

Preferably, described product is engine.

Preferably, the general function of described engine is divided into following four subfunctions and has combined, and is respectively: breathing process, compression process, acting process and exhaust process.

As mentioned above, the present invention, on the basis based on FBS model, has proposed the modeling of behavior stream, set up functional mode, behavior model, is used knowledge templet for carrier practical function is to the mapping of behavior, to reach the object of rapid modeling and realize the preliminary innovation of Product Conceptual Design.

Because Evolutionary Design is by the sane optimization method of the high speed of natural intelligence, by the evolutionary process between Design Mode, embodies the feature of thinking in images and creative thinking, and obtain the optimum solution of design problem.Different from the design process based on pure logic reasoning, in Evolutionary Design process, by the individual random combine of evolving, obtain examination and separate, and obtain optimum solution by the parallel large-scale whole design space of search that develops.Design rate has not only been accelerated in the application of genetic algorithm in design process, and has generated the product of a lot of intention, and therefore the present invention has used genetic algorithm to carry out preferentially the behavior path in behavior stream modeling process, to choose optimum solution.

Accompanying drawing explanation

Fig. 1 is shown as the rudimentary model schematic diagram of Modeling in Product of the present invention.

Fig. 2 is shown as the corresponding relation of behavior description of the present invention and title.

Fig. 3-6 are shown as in the specific embodiment of the invention result figure of four strokes after calculating.

embodiment

Below, by specific instantiation explanation embodiments of the present invention, those skilled in the art can understand other advantages of the present invention and effect easily by the disclosed content of this instructions.The present invention can also be implemented or be applied by other different embodiment, and the every details in this instructions also can be based on different viewpoints and application, carries out various modifications or change not deviating under spirit of the present invention.

Refer to Fig. 2 to 6.It should be noted that, the diagram providing in the present embodiment only illustrates basic conception of the present invention in a schematic way, satisfy and only show with assembly relevant in the present invention in graphic but not component count, shape and size drafting while implementing according to reality, during its actual enforcement, kenel, quantity and the ratio of each assembly can be a kind of random change, and its assembly layout kenel also may be more complicated.

Behavior stream Modeling in Product

In Modeling in Product design process, the directly abstract function for needing to realize of the demand of product, and the slight modification of behavior can directly cause functional change, behavior plays very important effect in whole Modeling in Product as can be seen here.The angle of the main subordinate act stream of the present invention realizes the modeling of product, the theory of bonding behavior stream, can top-down behavior be decomposed on the one hand, by the mapping of behavioral function, realize the function of product, can carry out combinatory analysis redesign to some behavior units on the other hand, realize the innovation of product.These two kinds of ways can be implemented simultaneously, reduce the repeated labor in design setting model process, thereby raise the efficiency shortening design time.

The key concept of behavior stream modeling

Before further modeling is flowed in elaboration behavior, first key concept designed in literary composition is given an explaination, define.Behavior

Behavior (Behavior) is the basic exercise principle that forms Realization of Product scheme, shows as the basic exercise behavior that each principle builds.Principle and behavior thereof are the bridge of function to physical arrangement, do not have direct mapping between function and structure, need to work by behavior.

Because behavior is an abstract concept, the motion that presentation is object.And function can be seen the output that is input to energy flow that is made for energy flow, as the consumed energy of completing of function, be the power F institute work that acts on this structure so, because of the movement locus of behavior, by acceleration a, determined again, second law by newton can obtain F=ma, the quality that wherein m is structure, a is acceleration.Because quality is static, therefore immutable behavior movement locus can determine by acting on the suffered power of its structure, and therefore we make firmly F as the concrete manifestation form of behavior at this.

Because behavior exists simple behavior and complex behavior, we are divided into behavior unit and complex behavior by behavior in the present invention again.Their related definition is as follows:

Behavior unit (Behavior Unit) refers to that structure is referred to as behavior unit in the effect that is subject to sometime power F along the represented behavior of three-dimensional space motion track.

Complex behavior (Complicated Behavior) refers to the behavior that serial or parallel connection that can be carried out due to the sequencing in time or space by a series of behavior unit etc. is combined into.

We can be by represented available any one behavior B (t) mathematic(al) representation below:

Wherein

be expressed as this structure and be parallel to the acting force of the direction of X-axis (perpendicular to YZ plane) on t moment edge, work as power

direction while being X-axis positive dirction

get on the occasion of, when acting force is X-axis negative direction,

get negative value.

be expressed as this structure and be parallel to the power of the directive effect of Y-axis (perpendicular to XZ plane) on t moment edge, work as power

direction while being Y-axis positive dirction get on the occasion of, when the direction of acting force is Y-axis negative direction, get negative value. be expressed as this structure and be parallel to the power of the directive effect of Z axis (perpendicular to XY plane) on t moment edge, work as power

direction while being Z axis positive dirction

get on the occasion of, when the direction of acting force moves along Z axis negative direction,

get negative value.

yZ plane included angle is the power of the directive effect of θ,

be expressed as this structure t moment section along and the power of the XZ plane included angle directive effect that is θ,

be expressed as this structure t moment section along and the power of the XY plane included angle directive effect that is θ.K is Boolean, uses 0,1 sign, is to represent that behavior B is that behavior unit is undecomposable at 0 o'clock, when K value 1, represents that behavior B is that complex behavior can decompose.L _bfor the sub-behavior list of behavior B, when k=0, l is empty, and when k=1, behavior B exists sub-behavior, L _bthe sub-behavior list of decomposing for B, as B _j, B _j, B _k.

Behavior stream (Behavior Flow) is the state of a series of structures, and the effect of process behavior, according to the change of specific time sequencing and spatial order generation state, causes the order of a series of subfunctions to complete, and finally completes a specific general function.

Wherein

B_{i} = f (Δ {\bar{x}}_{i}, Δ {\bar{y}}_{i}, Δ {\bar{z}}_{i}, Δ {\bar{θ}}_{x}_{i}, Δ {\bar{θ}}_{y}_{i}, Δ {\bar{θ}}_{z}_{i}, l_{b}, k)

The behavior equation of expression behavior node i.

S _irepresent the residing state of state node i.

expression behavior acts on construction operator.

(2) formula can be converted into so:

F _{i, j}expression is from state S _ito state S _jthrough μ (B _i, B _i+1..., B _j) the subfunction that realizes of effect.

We are by μ (Bi, B _i+1..., B _j) be called the mathematical expression of behavior stream.

Because some behavior is complex behavior, can be decomposed into multiple behavior unit, behavior flow point is serial and parallel two types as can be seen here.

Parallel behavior stream can be expressed as:

Wherein B _k,lfor behavior unit, 0 < m < max (i, j); 0 < n < max (i, j); I, j>=1.

By upper we can represent behavior stream by the form of matrix:

Wherein b _n1for behavior unit, the behavior sequence of the line display serial of matrix, matrix column represents the concurrency relation of behavior sequence, we are by B _{n × n}be referred to as behavioural matrix (Behavior Matrix)

Other related notions

Definition 1: function (Function) is the description of the product particular job ability of meeting customer need.

Overall product function can be decomposed into again many subfunctions.The subfunction of product represents the conversion between input, output and state, and specific physical behavio(u)r is abstract.

Definition 2: structure (Structure) is the final form of expression of product, and the function of product, behavior are all to realize by concrete structure.

Definition 3: knowledge templet (knowledge Template) is the carrier of the information such as the function, behavior, structure of product, is the instrument of multi-angle descriptive system design object.

Knowledge templet is not independently object of a single storage file or, and it contains much information set, is present in each stage of product design, at the design object different with same stage of different design phases, has the different forms of expression.The gathering of same class knowledge templet just forms knowledge base.

The generation of behavior stream initial population

Behavior stream is an abstract dynamic concept, and the essence of modeling is exactly to realize certain function of meeting customer need.Therefore we to realize the modeling of behavior stream be also take function as starting point, utilize coupling between function and behavior to realize the Modeling in Product of behavior stream.Utilizing coupling between behaviour to carry out in the process of modeling, self-assembling formation the behavior sequence node of modeling, behavior sequence node is now the static form of expression of behavior stream.In order to utilize function and behavior coupling relation, carry out automatic production modeling, the present invention proposes take knowledge templet as function, the information carrier of behavior and structure carries out the modeling of product, to propose the data structure of functional template, behavior template and stay in place form below, and realize the related algorithm of the preliminary modeling of preliminary product to obtain initial behavior stream colony.

The data structure of functional template:

struct Function{

String funcName//function title (not reproducible)

String funcDescription//functional description

Function subFunction; // decomposable subfunction

Behavior behavior; // realize the behavior list of this function

Constraint constraint; // with the constraint condition of behavior mapping

}

The data structure of behavior template:

struct Behavior{

String behName; // behavior title (not reproducible)

String behDescription//behavior description

Function function; The function that // the behavior can realize

Structurestructure; The structure list of // behavior effect

Constraintconstraint; // with the constraint condition of structure mapping

}

The data structure of stay in place form:

struct Structure{

StringstruName; // structure title (not reproducible)

The description of StringstrDescription//structure

Behavior behavior; // act on the behavior of this structure

Functionfunction; The attainable feature list of // this structure

Constraintconstraint; // with the constraint condition of functional mapping

}

To this, we have determined functional template, behavior template, the concepts such as stay in place form and the restriction relation between them.On the clear and definite basis of customer demand, the general function that will realize that we just can be abstracted into client's demand.We are referred to as a required function completing of product the general function of this product.Due to the complexity difference of product function, we are the subfunction that several complexities are lower by the Function Decomposition of high complexity, and the structure that is beneficial to product function solves.Function Decomposition starts from the description of product general function, is then subdivided into subfunction downwards, till each subfunction decomposes always and can be realized by existing structural unit.Obtaining after the set of sub-functions of total work the subject of knowledge and the object of knowledge decomposition, deviser need to determine the action process of each subfunction, and definite mapping relations that just need to use between functional template and behavior template of these action processes.In like manner deviser determines that by behavior corresponding structure also needs the mapping relations of behavior template and stay in place form.Deviser can, according to the mapping relations between knowledge templet, carry out iteration reasoning thus, finally obtains already present atomic structure in existing knowledge base, generates the behavior path of Modeling in Product simultaneously.By the behavior path of Modeling in Product, can be met the required action process of customer demand.

Concrete implementation algorithm is as follows:

By above-mentioned algorithm, we just can obtain many output streams, wherein namely behavior sequence node of so-called output stream.It comprises will realize the sequence of steps such as the behavior structure that certain function covers.By this algorithm, just can realize the formation of the behavior stream initial population of Modeling in Product.

Genetic algorithm is in the application of behavior miscarriage product model

By above-mentioned model, we have obtained a series of behavior sequence nodes of Modeling in Product, that is behavior stream.Behavior now stream is just by the behavior node of function and structure mapping, as for behavior node, follows the practical function how which kind of movement locus can be best, explains.In this present invention, use genetic algorithm to recombinate preferentially to behavior stream, finally obtain optimum behavior stream modeling path.Why choice for use genetic algorithm is the fierceness day by day due to the progress along with scientific and technological and market competition, product homogeneity phenomenon is day by day obvious, compared with emphasizing the method for designing of pure reasoning from logic, alogical Evolutionary Design method more stresses the randomness of the individual combination of design cell, by parallel natural selection, search for whole design space, obtain design problem optimum solution.Organically combine the genetic algorithm of evolution concept and logical thought, at solution fitness function conceptual design problem convexity beyond expression of words, show superiority.Therefore the present invention has also used genetic algorithm to carry out behavior stream preferentially.

(1) the individuality coding of behavior stream

Before determining that use genetic algorithm is carried out preferentially to behavior stream, we first flow to the coding of row binary to behavior:

We represent respectively by 4 binary codings first can be illustrated in the direction of above-mentioned power, and rear three bit representation power are worth size, for example:

available 0110 represents, 0 represents

As follows to the coding of the B of behavior unit:

Note: the first place of encode fragment represents direction, rear three bit representation numerical value

Can obtain thus the gene code of behavioural matrix:

Wherein B _{n × n}for completing the behavioural matrix of expectations of customer function, line display is realized the behavior stream of certain subfunction, and the concurrency relation of behavior stream is shown in list.

By behavioural matrix B _{n × n}each row obtain (b after connecting successively ₁₁b _1nb ₂₁b _2nb _n1b _nn), its corresponding chromosome becomes:

X ₁₁,…,X _1n|,X ₂₁…X _2n|,…,|X _n1…X _nn

Wherein, gene block X _i1... X _inrepresent to realize the behavior stream phenotype of i subfunction.

For example:

Bn×n=

[\begin{matrix} 100110001100110001110101 & 111001101111001101010010 & 011110011011110011010010 \\ 110001110110001110010100 & 011010010011010010011101 & 100101010100101010010011 \\ 010010100010010100010111 & 100100111100100111000111 & 110110111110110111011101 \end{matrix}]

The chromosome coding of behavior stream is:

100110001100110001110101|111001101111001101010010|011110011011110011010010|

110001110110001110010100|011010010011010010011101|100101010100101010010011|

010010100010010100010111|100100111100100111000111|110110111110110111011101|

(2) select, intersect, variation

Be chosen as and in genetic algorithm, represent that chromosome of fine qualities should more copy, the chromosome that quality is general is answered held stationary, and poor chromosome should be withered away gradually.The solution that the selection operation of genetic algorithm makes to have high fitness value copies and remains to the next generation.Intersection is a kind of recombination operator of genetic algorithm, is mainly that it can cause the restructuring of gene model for making parent individuality produce better next son generation, thereby produces the offspring individual that may comprise premium properties.When implementing interlace operation, two chromosomes of random selection from treat Cross reaction body, carry out the exchange of chromogene information at random selected infall.The for example motor behavior of object on travelling belt can be decomposed into following three behaviors,

The stressed rotation of gear---> travelling belt and the effect of engaged gears power under motion---> object is subject to the motion of friction force effect between travelling belt and object.

The behavior of travelling belt horizontal transmission object is encoded to:

100110001100110001110101|111001101111001101010010|011110011011110011010010

The behavior that travelling belt tilts to transmit object is encoded to:

010100101010010011 above-mentioned behavior of 110001110110001110010100|011010010011010010011101|100101: it is as follows that the first row of above-mentioned two behaviors and second behavior of the second row intersect rear result:

After intersection:

100110001100110001110101|011010010011010010011101|011110011011110011010010

110001110110001110010100|111001101111001101010010|100101010100101010010011

And variation is that one is used to make the diversified recombination genetic operator of search procedure.Variation has increased the diversity of colony's gene model, thereby has increased the effect of natural selection in Swarm Evolution process, and can avoid the generation of colony's Premature convergence, thereby avoids the too early local optimum region that is absorbed in Swarm Evolution process.When mutation operation is implemented in a certain position, this value becomes 1 or become 0 from 1 from 0.As B _{n × n}first value of second genetic fragment of the first row make a variation.Same take above-mentioned said object, the behavior on travelling belt is as example, if the direction of the friction force between object travelling belt and object level can be through variation negate direction, its result is as follows:

Original coding:

100110001100110001110101|111001101111001101010010|011110011011110011010010

After variation:

100110001100110001110101|011001101111001101010010|011110011011110011010010

(3) algorithm evaluation function and end condition

The behavior track that can produce due to same structure body has diversity, how to judge that any behavior track is that optimum becomes key.We when automobile starting, make the rotating speed of engine reach more few demand that more can better meet client of time of expection rotational speed take motor car engine as example.And the longer the better consuming the distance that can travel in identical oil consumption, when consuming equal oil consumption, the power of climbing is the bigger the better.The relevant criterion that we can determine evaluation behavior stream optimum solution thus with and corresponding mathematical model:

(1) completing with the degree of coupling that realizes expectation function that the behavior flows is more high better.

K = Σ_{i = 1}^{n} f_{i} Σ_{j = 1}^{n} b_{ij};

Wherein K is the degree of coupling, f _ifor subfunction and the degree of coupling that completes this behaviour stream, b _ijfor behavior unit

(2) to realize the required time of expectation function the smaller the better.

T = \max_{0 \leq i \leq n} Σ_{j = 1}^{n} t_{ij} b_{ij};

Wherein T completes the overall time of behavior stream, t _ifor completing the time of certain behavior, b _ijfor behavior unit, get time maximal value in parallel behavior stream and be the time that this behavior stream completes.

(3) when realizing expectation function, the consumption of energy is more few better.

E = Σ_{i = 1}^{n} Σ_{j = 1}^{n} e_{ij} b_{ij};

Wherein E flows the energy consuming, e for the behavior _ijfor completing the needed energy of certain behavior, b _ijfor behavior unit

(4) in the situation that realizing expectation function and determining zero energy, the power of generation is the bigger the better.

P = Σ_{i = 1}^{n} Σ_{j = 1}^{n} e_{i} p_{i} b_{ij},

Wherein

Σ_{i = 1}^{n} e_{i} = m;

The power value that wherein P produces for completing this function, e _ifor completing the needed energy of certain behavior, b _ijfor behavior unit,

P _ithe power value producing for completing certain behavior, represent that realizing the energy that this function consumes is definite value.Total evaluation function can be expressed as:

G=λ ₁{max (K)}-λ ₁{min(T)}-λ ₃{min (E)}+λ ₄{max(P)}

Wherein λ 1 _,λ ₂, λ ₃, λ ₄represent the shared weights of corresponding above-mentioned rule, 0≤λ _i≤ 1,1≤i≤4.

In concrete operatings of genetic algorithm process, need span of given G, if G=A is meet customer need completely (wherein A represents this convergence of algorithm threshold values), i.e. behavior now stream is optimum solution, being referred to as G is absolute convergence and A, but in reality, be difficult to reach Complete Convergence, therefore we need introduce interval float value α, when the value of required G falls into interval [A-α completely, A+ α], without the cross and variation of carrying out again algorithm, as G ∈ [A-α, A+ α] algorithm termination, the behavior stream sequence that now we obtain is optimum solution.

Based on the instance analysis of engine mockup critical piece

Engine is a kind of function of using dynamic mechanically energy demand in order to meet the mankind.In the engine case that we use in the present invention, it is a kind of device that chemical energy is converted into mechanical energy.Engine general function (using F to represent) can have been combined by following 4 subfunctions, and they are respectively: breathing process, compression process, acting process, exhaust process.

By the above-mentioned generating algorithm about initial behavior stream, can obtain following behavior node: B1 piston movement, B2 inlet open, B3 gas admittance valve cuts out, B4 plug ignition, B5 vent valve is opened, B6 exhaust valve closure, B7 Crankshaft motion, B8 link motion, B9 valve motion, the motion of B10 camshaft.

According to the definition of above-mentioned behavior, above behavior is numbered, as shown in Figure 2:

According to above-mentioned behavior coding rule, above-mentioned behavior is encoded, the coding schedule that obtains above-mentioned behavior is as follows again:

By our just passable initial behavior stream population of upper table:

B ₁={0000 0000 0010 0000 0000 0000 |0001 0010 0000 0000 0000 0000|0000 0000 0110 0000 0100 0000|

0000 0000 0111 0000 0000 0000|0000 0000 0000 0000 0110 0000|0010 0010 0000 0000 0110 0000}

B ₂={0000 0000 1010 0000 0000 0000|0001 0010 0000 0000 0000 0000|0000 0000 0110 0000 0100 0000|

B ₃={0000 0000 0010 0000 0000 0000|1001 1010 1010 00000000 0000|0000 0000 0110 0000 0100 0000|0000 0000 0111 0000 0000 0000|0000 0000 0000 0000 0110 0000|0010 0010 0000 0000 0110 0000}

\begin{matrix} \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \\ \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \\ \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \end{matrix}

B ₆₂={0000 0000 1010 0000 0000 0000|0001 0010 0000 0000 0000 |0000|0000 0000 1110 0000 1100 0000 |0000 0010 1111 0000 0000 0000|0000 0010 0000 0000 | 1110 0000 |0010 0010 0000 0000 1110 0000

B ₆₃={0000 0000 0010 0000 0000 0000 ||1001 1010 1010 0000 0000 0000|0000 0000 1110 0000 11000000 |0000 0010 1111 0000 0000 0000|0000 0010 0000 0000 1110 |0000|0010 0010 0000 0000 1110 0000

B ₆₄={0000 0000 1010 0000 0000 0000|1001 1010 10100000 0000|0000|0000 0000 1110 0000 1100 0000|0000 0010 1111 0000 0000 0000|0000 0010 0000 0000 1110 0000|0010 0010 0000 0000 1110 0000

Obtaining after initial population, according to the intersection in above-mentioned genetic algorithm, variation, as follows:

B ₁={0000 0000 0010 0000 0000 0000|0001 0010 0000 0000 0000 0000|0000 0000 0110 0000 0100 0000|

0000 0000 0111 0000 0000 0000|0000 0000 0000 0000 0110 0000|0010 0010 00000000 0110 0000}

0000 0000 0111 0000 0000 0000|0000 0000 0000 0000 0110 0000|0010 001000000000 0110 0000}

B ₁with B ₂after intersection, obtain following behavior:

\begin{matrix} {\bar{B}}_{1} = {0000 & 0000 & 1010 & 0000 & 0000 & 0000 | 0001 & 0010 & 0000 & 0000 & 0000 & 0000 | 0000 & 0000 & 0110 & 0000 \\ 0100 & 0000 | \end{matrix}

\begin{matrix} {\bar{B}}_{2} = {0000 & 0000 & 1010 & 0000 & 0000 & 0000 | 0001 & 0010 & 0000 & 0000 & 0000 & 0000 | 0000 & 0000 & 0110 & 0000 \\ 0100 & 0000 | \end{matrix}

0000 0000 01110000 0000 0000|0000 0000 0000 0000 0110 0000|0010 0010 0000 0000 0110 0000}

B ₁first (direction makes a variation) of the 3rd genetic fragment obtains following behavior:

0000 0000 0111 0000 00000000|0000 0000 0000 0000 0110 0000|0010 0010 0000 0000 0110 0000}

\begin{matrix} {\bar{B}}_{1} = {0000 & 0000 & 1010 & 0000 & 0000 & 0000 | 0001 & 0010 & 0000 & 0000 & 0000 & 0000 | 0000 & 0000 & 0110 & 0000 \\ 0100 & 0000 | \end{matrix}

\begin{matrix} \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \\ \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \\ \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot & \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \cdot \end{matrix}

Owing to the general function of engine being decomposed for four subfunctions once, that is: intake stroke, pressure stroke, expansion stroke, out stroke.In this case reach that behavior stream intersects at time and number of times through certain limit, selects, can restrain as early as possible after variation, we have carried out point other definition and an adjustment with the degree of coupling that behavior is flowed at four subfunctions.While above-mentioned initial behavior stream population being carried out to above-mentioned cross and variation in algorithm, add evaluation function

G=0.3G ₁+0.4G ₂-0.2G ₃+0.1G ₄

And end condition: G ∈ [A-α, A+ α], makes A=89, and α=0.75. is G ∈ [89-0.75,89+0.75].When fitness function value G '=G after selecting of behavior stream, be referred to as behavior stream Complete Convergence demand-expected, but in this test, we only need to calculate after behavior stream fitness function fall into interval [89-0.75,89+0.75].As shown in Figures 3 to 6.

We can obtain engine and obtain the sequential coding of final behavior stream thus, as shown in the table:

The present invention is based on FBS model and proposed the concept of behavior stream Modeling in Product, and the related definition of behavior stream, and the coupled relation that utilizes function, behavior and structure has been set up the initial population of behavior stream, behavior stream has been carried out to gene code, under the constraint of fitness function, use genetic algorithm to carry out preferentially, finally the examples prove by engine the feasibility of this thinking.Effect is as follows:

1. abstract behavior can be converted into the mathematical model of the power of available measurement calculating.

2. according to the mathematical model of behavior unit, the concept of deriving behavior stream with and mathematical expression.

3. utilize the coupled relation of function, behavior, structure, use knowledge templet as medium, to complete the foundation of behavior stream initial population.

4. utilize the attribute of behavior mathematical model, behavior unit has been carried out to gene code.

5. established fitness function and the objective function of genetic algorithm, and its model has been carried out to mathematical modeling.

In sum, the present invention has effectively overcome various shortcoming of the prior art and tool high industrial utilization.

Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not for limiting the present invention.Any person skilled in the art scholar all can, under spirit of the present invention and category, modify or change above-described embodiment.Therefore, such as in affiliated technical field, have and conventionally know that the knowledgeable, not departing from all equivalence modifications that complete under disclosed spirit and technological thought or changing, must be contained by claim of the present invention.

Claims

1. the modeling method based on behavior stream function and behavior coupled relation, is characterized in that, the method at least comprises the following steps:

Utilize the coupling between function and behavior to realize the Modeling in Product that behavior is flowed, form the behavior sequence node of modeling.

According to claim 1 based on behavior stream function the modeling method with behavior coupled relation, it is characterized in that: utilize coupling between function and behavior to realize the Modeling in Product that behavior flows and specifically comprise the following steps:

3. the modeling method based on behavior stream function and behavior coupled relation according to claim 1 and 2, it is characterized in that, further comprising the steps of: to use genetic algorithm, utilize described behavior sequence node to recombinate preferentially to behavior stream, obtain optimum behavior stream modeling path.

4. the modeling method based on behavior stream function and behavior coupled relation according to claim 3, it is characterized in that, use genetic algorithm, utilize described behavior sequence node to recombinate preferentially to behavior stream, obtain optimum behavior stream modeling path and specifically comprise the following steps:

5. the modeling method based on behavior stream function and behavior coupled relation according to claim 1, is characterized in that, described product is engine.

6. the modeling method based on behavior stream function and behavior coupled relation according to claim 5, it is characterized in that, the general function of described engine is divided into following four subfunctions and has combined, and is respectively: breathing process, compression process, acting process and exhaust process.