Complex task actual flight test horizontal dynamic space design optimization method and computer product

Abstract

The invention discloses a complex task actual flight test horizontal dynamic space design optimization method and a computer product, and the method comprises the steps: S1, decomposing a test task profile, and constructing a test task outward tree; s2, mapping each test task sub-profile in the constructed task outbound tree to obtain a corresponding test area, and forming an initial task area association symbol graph; s3, dynamically evolving a task area association symbol graph; s4, performing dynamic modeling on the relationship evolution of the test area to obtain a weight updating function and a symbol updating function; s5, expanding the weight updating function and the symbol updating function to the multivariate relation network to obtain the weight updating function and the symbol updating function of the multiple test areas; and S6, performing optimization based on the weight updating function and the symbol updating function of the multiple test regions to obtain an optimized area. The problems of test resource allocation and environmental adaptability are solved, and the method has remarkable advantages in the aspects of space saving, stability improvement, complex event processing and the like.

Description

Technical Field

[0001] This invention relates to the field of flight test environment construction technology, specifically to a method for dynamic spatial design optimization of complex mission flight tests and a computer product. Background Technology

[0002] The planning and design of flight tests mainly includes test requirements analysis, test mission design, test environment design, test and measurement scheme design, and organizational implementation scheme design. For simple flight tests, the environment is relatively simple (i.e., location, scope, etc.), so the design of flight test missions for simple missions often focuses on the design and planning of test procedures, test sample size, test and measurement methods, and organizational implementation schemes. However, for complex and comprehensive test missions, there are multiple mission profiles, each with different environmental requirements, and usually time-series requirements. The demand for test space is dynamically changing, which poses challenges to resource scheduling, risk management, and cost and time control in flight tests. Therefore, researching the design and optimization methods of dynamic space for complex flight tests and finding the optimal solution for test space design is an urgent problem to be solved. Summary of the Invention

[0003] This invention provides a method and computer product for dynamic spatial design optimization of complex mission flight tests, which solves the problem in the prior art that it is difficult to perform dynamic spatial design and optimization for complex mission flight tests.

[0004] Specifically, the present invention is implemented as follows:

[0005] According to a first aspect, the present invention provides a method for dynamic space design optimization of complex mission flight tests, comprising:

[0006] Step S1: Decompose the experimental task profile and construct the experimental task outbound tree;

[0007] Step S2: Transform each experimental task sub-profile in the constructed task outward tree. Mapping out corresponding and independent test areas ,each Required area By analyzing the correlation and opposition of the initial regions of each task sub-profile using symbolic graphs, a symbolic graph of the initial task regions is formed. ;

[0008] Step S3: Flowchart over time The dynamic evolution symbol map is obtained by associating the task region with the changes. :

[0009]

[0010] In the formula:

[0011] For the set of elements in the task region, ,, Number of task areas;

[0012] For the time process Changing edge set, ;

[0013] The sign function for the edge. : ;

[0014] Define the adjacency matrix related to the test procedure. for:

[0015]

[0016] In the formula:

[0017] For time Time test area and The strength of the relationship, i.e., the relationship weight;

[0018] Step S4: Time-based process Changes were observed, and dynamic modeling of the evolution of relationships within the experimental area was performed to obtain... Weight update function at time step And edge sign update function ;

[0019] Step S5: Update the weight function And edge sign update function Extending to a multi-relationship network, a weight update function for multiple test areas is obtained. And edge sign update function ;

[0020] Step S6: Optimize the initial area of ​​the experiment based on the weight update function of the multiple test regions and the sign update function of the edges to obtain the optimized area.

[0021] Furthermore, step S1 also includes:

[0022] Let the experimental task be The task profile outward tree corresponding to this experimental task is: (in It is a directed tree. for The vertex set; during task profile decomposition, if And for each different vertex There exists a unique source arrive If there is a directed path, then the subtask profile decomposition is correct, and it is called... For the root The task profile outward tree.

[0023] Furthermore, in step S2, the initial task region association symbol map... for:

[0024]

[0025] In the formula:

[0026] For the set of elements in the task region, , Number of task areas;

[0027] For edge set,

[0028] The sign function for the edge. : ;

[0029] In the diagram, positive edges represent regions that are related, negative edges represent regions that are mutually exclusive, and no edges represent regions that are independent and have no direct relationship.

[0030] Furthermore, in step S4, it is assumed that the experimental process... Changing edges weight and corresponding symbolic functions The dynamic evolution formulas are respectively and ,but Any test region at any time and The dynamic evolution equation of the relationship is analyzed as follows:

[0031]

[0032]

[0033] Time-weight update function for:

[0034]

[0035] In the formula:

[0036] This represents the weighted memory coefficient for a certain experimental area, i.e., the proportion of historical relationships retained in the area. ;

[0037] The sensitivity of a certain test area to external influences; ;

[0038] The influence of external events / activities on the strength of relationships within the test area;

[0039] At time, the edge sign update function for:

[0040]

[0041] In the formula:

[0042] This is the threshold for switching the polarity of regional relationships, i.e., the critical value that triggers relationship reversal;

[0043] like Sensitive to regional relations;

[0044] like Stable regional relations;

[0045] This is the triggering condition for the polarity switching of the relationship in the test area, and .

[0046] Furthermore, in step S5, when the number of test areas... For multiple times In addition to considering the influence of the relationship between any two test areas, the interaction relationships between multiple test areas also need to be considered. Therefore, the above model must be extended from a binary relationship to a multi-way relationship network. During the experiment, Each test area needs to be modified based on process changes, emergencies, environmental changes, etc. A linear model is used to update the weights of the test areas. The design concept is as follows: new weight = regional historical memory + event / environmental change response. Indicates the process over time The regional relationship weight matrix, then the test area For the region and The indirect impact of the relationship is determined by the weight tensor update equation as follows:

[0047]

[0048] in:

[0049]

[0050] At this point, the weight update function for multiple test regions for:

[0051]

[0052] In the formula:

[0053] This represents the coupling coefficient of the regional triangular relationship. ;

[0054] This is an aggregation term representing regional triangular relationships;

[0055] In time At that time, the test area and The strength of the relationship, i.e., the region and Relationship weights;

[0056] In time At that time, the test area and The strength of the relationship, i.e., the region and Relationship weights.

[0057] Symbol update function for edges of multiple test regions for:

[0058]

[0059] In the formula:

[0060] This is the threshold for switching the polarity of regional relationships, i.e., the critical value that triggers relationship reversal;

[0061] This represents the regional group influence coefficient. ;

[0062] This is the effect term in the regional group formula. When it is greater than 0, it indicates that most task regions support the current relationship; when it is less than 0, it indicates that most regions are mutually exclusive with the current region.

[0063] In time At that time, the test area and Relationship symbols, i.e., regions and The nature of the relationship;

[0064] In time At that time, the test area and Relationship symbols, i.e., regions and The nature of the relationship.

[0065] Furthermore, in step S6, if each experimental task sub-profile Mapping out corresponding and independent test areas ,each The required initial area is The maximum experimental area that actual resources can provide is ,and Therefore, the initial experimental area needs to be optimized before the experiment can be carried out.

[0066] The methods for optimizing the initial area of ​​the experiment based on the weight update function and the sign update function of the edges in multiple test regions include any of the following:

[0067] (1) Optimize based on optimal physical space resources;

[0068] (2) Optimize based on the optimal stability of the experimental tissue;

[0069] (3) Optimize by taking into account both physical space and experimental organization stability.

[0070] Furthermore, when optimizing for optimal physical space resources, the optimization model is as follows:

[0071]

[0072] In the formula:

[0073] The optimized initial area;

[0074] For task sub-section Corresponding to the required area The initial area;

[0075] When optimizing to achieve optimal tissue stability, the optimization model is as follows:

[0076]

[0077] In the formula:

[0078] The optimized initial area;

[0079] A sign update function for edges in multiple test regions;

[0080] This is an indicator function; it takes the value 1 when the relational property changes, and 0 otherwise.

[0081] For all and The resulting region pairs are combined;

[0082] When considering both physical space and experimental organization stability for optimization, the optimization model is as follows:

[0083]

[0084]

[0085] In the formula:

[0086] The optimized initial area;

[0087] Task self-profile Corresponding to the required area The initial area;

[0088] This is a trade-off coefficient for the stability of the mission area. ,when Prioritize minimizing the area when Prioritize minimizing changes in regional relations;

[0089] This is an indicator function; it takes the value 1 when the relational property changes, and 0 otherwise.

[0090] A sign update function for edges in multiple test regions;

[0091] This is the triggering condition for the polarity switching of the relationship in the test area, and .

[0092] Furthermore, The sum should satisfy:

[0093]

[0094] Task self-profile Corresponding to the required area The area;

[0095] The maximum test area that can be provided.

[0096] According to a second aspect, the present invention also provides a computer program product, including computer program instructions that, when executed by a processor, implement the steps of the above-described method.

[0097] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0098] This invention solves the problems of experimental resource allocation and environmental adaptability through outward tree decomposition, symbolic graph dynamic evolution and multi-objective optimization. It has significant advantages in space saving, stability improvement and complex event handling, and is applicable to high-risk and high-cost experimental fields such as aerospace, deep sea and military industries. Attached Figure Description

[0099] Figure 1 This is a schematic diagram of the outward tree of the experimental task profile in Example 1;

[0100] Figure 2 This is a symbol diagram showing the initial region association for each task sub-section in Example 1;

[0101] Figure 3 This is a symbol diagram of the initial task region in Example 1. Detailed Implementation

[0102] The present invention will be further described in detail below through specific embodiments.

[0103] Example 1

[0104] This embodiment is an example of an implementation of the technical solution of a dynamic space design optimization method for complex mission flight test, as described in the invention.

[0105] First, the experimental task profile is decomposed, and an outward tree of the experimental tasks is constructed. Let a certain experimental task be... The outward tree for this experimental task is The experimental task profile of the outward-facing tree is as follows Figure 1 As shown, after decomposing the task profile, the task sub-profile is as follows: ,in Powertrain calibration Communication relay test Collaborative target tracking Obstacle avoidance algorithm verification, initial region association symbols for each task sub-profile are as follows: Figure 2 As shown,

[0106] The constructed task profile outward tree Sub-profiles of each experimental mission Mapping out corresponding and independent test areas Experimental task sub-profile and initial region area The demand mapping relationship is shown in the table below:

[0107]

[0108] Initial task region association matrix The symbol diagram is as follows: Figure 3 As shown.

[0109]

[0110] When t=1, assume an external event occurs: a sudden strong wind and wave, causing the following changes in the influencing parameters:

[0111]

[0112]

[0113] With α=0.85 and β=0.6, the weights are updated as follows:

[0114]

[0115]

[0116] Taking θ = 0.4, since 0.355 > 0 and 0.56 > 0, meaning a positive correlation is maintained, the sign is updated as follows:

[0117]

[0118]

[0119] When t=2, assuming a new event: radar interference affecting communication, γ=0.07, the test area weight update function is as follows:

[0120]

[0121] The test area symbol matrix update function is:

[0122]

[0123] Finally, dynamic optimization of the overall area is performed.

[0124]

[0125] The total area constraint is as follows:

[0126]

[0127] The optimization results are as follows:

[0128]

[0129] The optimized total area is reduced by 1.5. (9.7%), the number of regional relationship changes was 1, which is relatively stable.

[0130] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A complex task real sea trial level dynamic space design optimization method, characterized in that, Comprising: Step S1: decompose the test task profile, and construct the test task outward tree; Step S2: mapping out corresponding and independent test areas mapping out corresponding and independent test areas forming an initial task area correlation symbol map by analyzing the correlation and opposition of each task sub-profile initial area through signed graph ; Step S3: with the test time flow The dynamic evolution task area correlation symbol map is changed to obtain a dynamic evolution symbol map : ； In the formula: for a task area element set, , is the number of task areas; To flow over time Changing edge sets, ; a sign function for edges, : ; Defining an adjacency matrix related to a test flow is: ； In the formula: for time time trial area and strength of the relationship, i.e. relationship weight; Step S4: Time flow Variations, test area relationship evolution dynamic modeling, get The weight update function at the moment And the symbol update function of the edge ; Step S5: updating the weight function and the sign function of the edge to the multi-relational network, obtaining the weight updating function of the multi-test region and the sign function of the edge ; Step S6: based on the weight update function of the multi-test area and the sign update function of the edge, the initial area of the test is optimized to obtain the optimized area.

2. The method of claim 1, wherein, The step S1 further comprises: Let the experimental task be The task profile outward tree corresponding to this experimental task is: , for The vertex set; during task profile decomposition, if And for each different vertex There exists a unique source arrive If there is a directed path, then the subtask profile decomposition is correct, and it is called... For the root The task profile outward tree.

3. The method of claim 1, wherein, In the step S2, the initial task area association symbol map is generated as follows: is: ； In the formula: for a task area element set, , for a task area quantity; is an edge set, a sign function for edges, : ; In the figure, the positive edge represents the area association, the negative edge represents the area mutual exclusion, and the no edge represents the area independent without direct relationship.

4. The method of claim 1, wherein, In the step S4, Time weight update function is: ； In the formula: a weight memory coefficient for a certain test area, i.e. a retention proportion of the area history relationship, ; sensitivity of a certain test area to external influences; ; The influence of external events / activities on the strength of the relationship of the trial area; At time t, the sign update function for an edge is: ； In the formula: To be in time The relationship symbol, i.e. the relational property, of the areas And at time a region relationship polarity switching threshold, i.e. a critical value triggering a relationship inversion; If : Zone relationship sensitive; If : the region relationship is stable; To test the zone relationship polarity switch trigger condition, and .

5. The method of claim 1, wherein, In the step S5, the weight updating function of the multiple test regions is: ； In the formula: for the area triangle relationship coupling coefficient, ; to the area triangulation aggregation item; to be in time with the test area and the relationship strength, i.e. the relationship weight, of the area and to be in time To determine the relationship strength between the time , the test area and the relationship weight between the area and ; Symbol update function for multiple trial region edges is: ； In the formula: a region relationship polarity switching threshold, i.e. a critical value triggering a relationship inversion; for the regional population impact coefficient, ; For the regional group formula effect term, when it is > 0, it indicates that the majority of the task regions support the current relationship, and when it is < 0, it indicates that the majority of the regions are mutually exclusive with the current regional relationship; For the time , the test area and the relationship symbol, that is, the relationship between the area and ; To be in time The relationship symbol, i.e. the relationship property of the areas and at the time and .

6. The method of claim 1, wherein, In the step S6, the optimization mode of the initial area of the test based on the weight update function of the multi-test area and the sign update function of the edge comprises any one of the following: (1) optimizing in the physical space resource optimization; (2) optimizing in the test organization stability optimization; (3) optimizing by comprehensively considering the physical space and the test organization stability.

7. The method of claim 6, wherein, When optimizing in the physical space resource optimization, the optimization model is: ； In the formula: To optimize area; for the task sub-profile corresponding to the required area area; When optimizing in the test organization stability optimization, the optimization model is: ； In the formula: To optimize area; A function to update the sign of the edges; is an indicator function that takes the value 1 if the relationship changes and 0 otherwise; for all and constitute a combination; When optimizing by comprehensively considering the physical space and the test organization stability, the optimization model is: ； ； In the formula: To optimize area; for the task profile corresponding to the desired area area; a trade-off coefficient for the area stability of the task region, when , the area is prioritized to be minimized, when , the region relation change is prioritized to be reduced; is an indicator function which takes the value 1 if the relationship changes and 0 otherwise; a sign update function for edges; To test the zone relationship polarity switch trigger condition, and .

8. The method of claim 7, wherein, The sum of the steps S6, should satisfy: ； for the task profile corresponding to the desired area area; The maximum test area available.

9. A computer product comprising computer program instructions, characterised in that, The computer program instructions are executed by the processor to realize the steps of the method in any one of claims 1-8.

Images