CN110083902B - An Inverse Design Method of Temperature Distortion Atlas Based on Discrete Sequence - Google Patents

Abstract

Temperature distortion map inverse design based on discrete sequenceThe method comprises the steps of determining the incoming flow velocity and the distance between an upstream distortion section and a downstream map section, numbering the arrangement distribution of nozzles, setting the heat flow combination sequence of each nozzle as a vector α, dividing a downstream target map section into units with the magnitude N, sorting the divided micro units outwards by taking the horizontal direction as a zero-angle line, and marking as a map column vector β, and respectively constructing a column vector Q₁，...，Q_i，...，Q_MThe method comprises the steps of measuring a downstream temperature distortion map, taking a generalized inverse matrix B from the matrix A to determine the distribution of heat flow, obtaining the states of all nozzles at the upstream according to a vector α, namely realizing the specific temperature distortion map by adjusting the injection intensity of the heat flow, and providing data support for temperature distortion under the actual working condition by the obtained result, so that the column vector α is reversely solved by using the matrix A and the column vector β.

Description

An Inverse Design Method of Temperature Distortion Atlas Based on Discrete Sequence

technical field

The invention relates to aerodynamic stability evaluation technology of aero-engine, in particular to a discrete sequence-based temperature distortion map inverse design method for estimating the distribution of heat sources according to the temperature distortion map.

Background technique

Aero-engine aerodynamic stability is an important indicator for evaluating engine performance, which requires the engine to be able to resist the interference of destabilizing factors in the entire flight envelope in addition to its superior performance in key design states, and to ensure sufficient available stability margin ( [1] Zhao Yunsheng. Research on aero-engine aerodynamic stability analysis system [D]. Nanjing University of Aeronautics and Astronautics, 2013). However, in the process of aircraft formation flight, missile launch, etc., the engine inlet inevitably absorbs heat flow. The inhalation of heat flow will bring obvious temperature distortion to the engine inlet. Serious temperature distortion will cause the engine to enter an unstable working state. The in-depth understanding of the problem will be of great significance to the design and development of future aero-engines.

In view of the important influence of temperature distortion on the stable operation of the engine, major aviation powers such as the United States, the United Kingdom, China, and Russia have carried out mechanistic and systematic research on temperature distortion, and formulated corresponding temperature distortion generation and simulator design methods ([[ 2] Dai Bing, Ye Wei. Comparative Analysis of American and Russian Aero-Engine Stability Standards [J]. Aviation Standardization and Quality, 2009(02):44-48). Existing temperature distortion is mainly to set a certain number of nozzles in the intake passage, perform ignition combustion or heat flow injection at the nozzles to achieve a local increase in airflow temperature, and control the heat flow intensity of the nozzles to achieve the purpose of customizing the distortion measurement cross-sectional map. . At present, in order to obtain the specified temperature distortion map, the main engineering practice is to study the combination of the heat flow intensity of the nozzles through numerical simulation or experiment on the basis of the determined number of nozzles and nozzle distribution. Due to the lack of a custom design method for the temperature distortion map at present, it will consume a lot of manpower and material resources in the fine debugging of the temperature distortion map, and there are problems such as long development cycle, complicated design, and large randomness of experiments. It is very necessary to obtain a temperature distortion design method that can obtain a complex temperature distortion map and ensure the accuracy of the map.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method for inverse design of temperature distortion maps based on discrete sequences by estimating the distribution of heat sources according to the existing temperature distortion maps accurately design methods and other deficiencies.

The present invention includes the following steps:

1) According to the temperature map distribution of the given measurement section, determine the flow velocity and the distance between the upstream distortion section and the downstream map section;

2) Generate the number of nozzles M and the nozzle distribution according to the temperature, sequentially number the arrangement and distribution of the nozzles, and set the heat flow combination sequence of each nozzle as a vector α;

3) Divide the downstream target atlas section into units of magnitude N respectively, sort the obtained tiny units with zero angle in the horizontal direction, and sort the tiny units outward one by one, and use interpolation based on the temperature distortion atlas. The temperature is assigned to N units in the way of , and it is recorded as the atlas column vector β;

4) Construct column vectors Q ₁ ,...,Q _i ,...,Q _M respectively, where Q _i is denoted as (q ₁ ,...,q _i ,...,q _M ) ^T , except that q ₁ is set to 1, Other values are set to 0, which means that except for the i-th nozzle that is in the maximum heat flow state, the other M-1 nozzles are all closed. Among them, the maximum heat flow state is recorded as 1, and the minimum heat flow state is recorded as 0;

5) Carry out numerical simulation or experimental research with the ith nozzle in the maximum heat flow and the other M-1 in the closed state, measure the downstream temperature distortion map, and refer to the number in step 3) to obtain the vector (a _1i , a _2i ) , a _3i ... a _Ni ) ^T , a total of M nozzles form a matrix

6) According to Aα=β, take the generalized inverse matrix _B for the matrix A, where BA=EM , EM is the _M -order unit matrix, and reverse the heat flow distribution vector α, that is, α=Bβ, so as to determine the heat flow distribution, where , the maximum value is 1 and the minimum value is 0 in the vector α;

7) According to the obtained vector α, the state of each upstream nozzle is obtained, that is, a specific temperature distortion map can be realized by adjusting the heat flow injection intensity. The obtained results provide data support for exploring the temperature distortion under real working conditions. Therefore, the inverse design method is Reverse column vector α using matrix A and column vector β.

The present invention can be analyzed through the construction and solution of vectors and matrices, wherein the upstream discrete sequence distribution of heat flow is vector α, and the downstream flow temperature distribution is vector β, and the relationship between the two is matrix A, vectors α, β and matrix A For the construction method, please refer to the specific implementation section.

The present invention can more accurately reproduce the inverse design theoretical method of the real flow field shown by the temperature distortion map, that is, taking the existing temperature distortion map as the target, in a specific temperature nozzle arrangement state, reversely obtain the temperature that can form the temperature shown in the map The heat flow intensity distribution of the distortion field, considering that the temperature distortion map is generally given in a specific incoming flow velocity state, the distance between the upstream distortion simulator and the temperature distortion measurement section is generally a fixed value, and the number of temperature generating nozzles in engineering and the nozzle distribution are generally given, and the solution of the upstream temperature distortion simulator for the temperature distortion map can be understood as: given the downstream flow temperature distribution, construct the relationship between the upstream heat flow and the downstream temperature distribution, and reverse the upstream heat flow discrete sequence.

The present invention takes into account that the number of nozzles and nozzle distribution of the upstream section temperature generation in engineering are generally given, and the spatial distribution of downstream temperature distortion is mainly processed by adjusting the heat flow intensity of the nozzle. The relationship between them can be understood as vector operations between different discrete sequences. Combined with the influence law of upstream heat flow on the temperature distribution of the downstream measurement section, different discrete sequences of nozzle states are constructed, and the positive influence law of the upstream temperature distortion simulator on the downstream temperature distortion map can be analyzed. The so-called inverse design is to find the discrete sequence combination of the upstream nozzle from the known temperature distortion map, so as to realize the design of the temperature distortion simulator.

The present invention has the following outstanding technical effects:

The present invention can simulate the heat flow distribution under real working conditions and obtain a temperature distortion map with high precision. Because the influence of each nozzle in the combustion section on the temperature measuring surface is quantified, the waste of manpower and material resources caused by a large number of random attempts in the test is avoided, and the test period is greatly shortened. At the same time, the inverse design method only needs to establish a database of matrix A with a mapping law, and can quickly obtain the corresponding heat flow distribution from different temperature distortion maps, which significantly reduces the test cost compared with the traditional method.

Description of drawings

Figure 1 is a schematic top view of the temperature distortion test device.

FIG. 2 is a schematic diagram of a distribution of ignition nozzles in the combustion section in the test section in FIG. 1 (M=21).

FIG. 3 is a schematic diagram of a distribution of temperature measurement surfaces downstream of the test section in FIG. 1 (N=40).

FIG. 4 is a schematic diagram of a temperature distortion map obtained through experimental simulation on the temperature measuring surface downstream of the test section in FIG. 1 .

The marks in the figure are: 1 indicates the combustion section in the test section, 2 indicates the fuel pipe in the combustion section 1 in the test section, 3 indicates the mixing section in the test section, 4 indicates the temperature measuring surface downstream of the test section, 5 indicates The gas pipe in the combustion section 1 in the test section, 6 indicates the fixed rod in the combustion section 1 in the test section, 7 indicates the air inlet upstream of the test section, and 8 indicates the cross section of the ignition nozzle in the test section to the temperature measuring surface. Distance, 9 represents the ignition nozzle, 10 represents the temperature measurement point in the temperature measurement surface, and 11 represents the high temperature distortion area in the temperature distortion map.

Detailed ways

The following embodiments will further illustrate the present invention in conjunction with the accompanying drawings.

The embodiment of the present invention includes the following steps:

1) According to the temperature map distribution of the given measurement section, determine the flow velocity and the distance between the upstream distortion section and the downstream map section;

2) Generate the number of nozzles M and the nozzle distribution according to the temperature, sequentially number the arrangement and distribution of the nozzles, and set the heat flow combination sequence of each nozzle as a vector α;

3) Divide the downstream target atlas section into units of magnitude N respectively, sort the obtained tiny units with zero angle in the horizontal direction, and sort the tiny units outward one by one, and use interpolation based on the temperature distortion atlas. The temperature is assigned to N units in the way of , and it is recorded as the atlas column vector β;

4) Construct column vectors Q ₁ ,...,Q _i ,...,Q _M respectively, where Q _i is denoted as (q ₁ ,...,q _i ,...,q _M ) ^T , except that q ₁ is set to 1, Other values are set to 0, which means that except for the ith nozzle that is in the maximum heat flow state, the other M-1 nozzles are all closed, and the maximum heat flow state is recorded as 1, and the minimum heat flow state is recorded as 0;

5) Carry out numerical simulation or experimental research with the ith nozzle in the maximum heat flow and the other M-1 in the closed state, measure the downstream temperature distortion map, and refer to the number in step 3) to obtain the vector (a _1i , a _2i ) , a _3i ... a _Ni ) ^T , a total of M nozzles form a matrix

6) According to Aα=β, take the generalized inverse matrix _B for the matrix A, where BA=EM , EM is the _M -order unit matrix, and reverse the heat flow distribution vector α, that is, α=Bβ, so as to determine the heat flow distribution, where , the maximum value is 1 and the minimum value is 0 in the vector α;

7) According to the obtained vector α, the state of each upstream nozzle is obtained, that is, a specific temperature distortion map can be realized by adjusting the heat flow injection intensity. The obtained results provide data support for exploring the temperature distortion under real working conditions. Therefore, the inverse design method is Reverse column vector α using matrix A and column vector β.

Figure 1 shows a schematic top view of the temperature distortion test device, which is basically the same as the traditional test device, the ignition nozzle 9 (see Figure 2) is mutually stabilized by the fixed rod 6 in the combustion section 1 in the test section, and the combustion section in the test section is stabilized with each other. The fuel pipe 2 in 1 and the gas pipe 5 in the combustion section 1 in the test section communicate with each nozzle; the length of the mixing section 3 in the test section is the cross section of the ignition nozzle 9 to the temperature measurement surface downstream of the test section The distance from the cross section of the ignition nozzle to the temperature measuring surface in the test section of 4. The heat flow distribution deduced by the inverse design method is mainly manifested in the cross section of the ignition nozzle in Figure 2. According to the result vector α of the inverse design method, the ignition nozzle 9 in the combustion section 1 in the test section is adjusted, and the temperature measuring surface in the downstream of the test section is adjusted. 4 using the temperature measuring point 10 (see Figure 3) in the temperature measurement surface to obtain data, after processing the data, obtain the temperature distortion simulation map of Figure 4, and compare it with the known temperature distortion map, if the degree of closeness is good , the simulated heat flow field can represent the real heat flow field.

The temperature distortion test section of the present invention has two main sections, namely the cross section of the upstream ignition nozzle and the downstream temperature measurement surface. In actual operation, according to the given temperature distortion map, the temperature value of each unit is determined on the temperature measurement surface, and then the heat flow distribution in the real working condition is simulated according to the inverse design method, as follows:

Definition: The combustion intensity of the i-th ignition nozzle is α _i , and the obtained heat flow distribution is a vector α; the temperature value obtained at each measuring point on the downstream temperature measuring surface is denoted as b _j , and all the temperature values constitute the atlas column vector β .

Assume that the heat flow distribution at a certain moment is α=(α ₁ ,α ₂ ,...,α _i ,...,α _M ) ^T , and the temperature of a certain point j on the temperature measuring surface at this moment is based on other than the heat flow α _j1 , α _j2 ,...,α _ji ,...,α _jM , which are affected by each nozzle respectively, these values are measured from experiments, in which it is only necessary to know that the three ignition nozzles on a certain axis have an impact on the downstream measurement. The influence of the temperature surface can be calculated to obtain the values of the remaining nozzles. According to the principle of linear superposition of data points, the influence of point j under other factors is α _j1 +α _j2 +α _ji +…+α _jM , and each term of the above formula is multiplied by the combustion intensity α of each nozzle at this time _i can obtain the total influence effect b _j of the temperature n at this time point, that is, b _j =α ₁ *α _j1 +α ₂ *α _j2 +α _i *α _ji +...+α _M *α _jM . Therefore, according to this idea and so on, the matrix A can be constructed to reflect the mapping law of the temperature influence of the upstream heat source on the downstream temperature measurement surface:

According to Aα=β, take the generalized inverse matrix _B for matrix A, where BA=EM, EM is the _M -order unit matrix, so the heat flow distribution vector α can be reversed, that is, α=Bβ, so as to determine the heat flow distribution.

The invention avoids the waste of manpower and material resources caused by a large number of random attempts in the test, and greatly shortens the test period; at the same time, the inverse design method only needs to establish a database of the matrix A with the mapping law, and the distortion map from different temperature can be obtained. The corresponding heat flow distribution can be obtained quickly, which significantly reduces the test cost compared with the traditional method.

As shown in Figures 1 to 4, the present invention includes a combustion section 1 in the test section, a fuel pipe 2 in the combustion section 1 in the test section, a mixing section 3 in the test section, a temperature measuring surface 4 downstream of the test section, and a test section. The gas pipe 5 in the combustion section 1 in the test section, the fixed rod 6 in the combustion section 1 in the test section, the air inlet 7 upstream of the test section, the distance from the cross section of the ignition nozzle in the test section to the temperature measuring surface 8, ignition The nozzle 9, the temperature measurement point 10 in the temperature measurement surface, and the high temperature distortion area 11 in the temperature distortion map.

First, according to the model of the engine, determine the distance 8 from the cross section of the ignition nozzle to the temperature measuring surface. Then, according to the working state of the engine, the 12 ignition nozzles 9 are sequentially numbered according to their arrangement and distribution, and the combustion intensity function curve of each ignition nozzle 9 is calculated.

The numbering rules are as follows:

On the circular section at the heat source, take the center of the circle as the vertex and draw a ray horizontally to the right. This ray is set as the horizontal reference line, that is, the 0° line, and the counterclockwise rotation is positive. Rotate the horizontal reference line along the positive direction and search for the ignition nozzles, wherein, for the ignition nozzles 9 on the same ray, from the center of the circle outward, the radial direction is sequentially numbered. According to the above numbering method, the M ignition nozzles 9 are sequentially ordered.

The combustion intensity of each ignition nozzle is obtained by adjusting the gas supply and oil supply. Theoretically, the maximum combustion intensity is set to 1, the minimum value is 0, and the remaining values are obtained between 0 and 1.

Divide the downstream target atlas section into units of magnitude N respectively, sort the obtained tiny units with zero angle in the horizontal direction, and sort the tiny units outward one circle at a time, and use interpolation based on the temperature distortion map. Assign the temperature to N units, and denote it as the map column vector β.

The i-th ignition nozzle is controlled to operate, and the remaining M-1 ignition nozzles are closed, that is, Q _i =(q ₁ ,...,q _i ,...,q _M ) ^T , where q _i is set to 1, and other values are set is 0, and the maximum heat flow state is recorded as 1, and the minimum heat flow state is recorded as 0. Carry out numerical simulation or experimental research in which the i-th ignition nozzle is in the maximum heat flow and the other M-1 are in the closed state, measure the downstream temperature distortion map, and refer to the above-mentioned numbering rule of the downstream target map section to obtain the vector (α _1i , α _2i ,...,α _3i ,...,α _Ni ) ^T . A total of M ignition nozzles 9 can form a matrix

According to Aα=β, a generalized inverse matrix B is taken for the matrix A, where BA=EM , and _EM is an _M -order unit matrix. Therefore, the heat flow distribution vector α can be reversely obtained, that is, α=Bβ, and the current gas and oil supply state of each ignition nozzle 9, that is, the current heat flow distribution, can be calculated through the combustion intensity curve obtained by discrete point fitting.

The heat flow distribution obtained by the above inverse design method is reversed. After the simulation test, the temperature distortion map in Figure 4 is obtained. By comparing with the known temperature distortion map, especially the area and orientation of the high temperature distortion zone 11, the real heat flow distribution can be simulated. . The invention avoids the waste of manpower and material resources caused by a large number of random attempts in the test, and greatly shortens the test period; at the same time, the inverse design method only needs to establish a database of the matrix A with the mapping law, and the distortion map from different temperature can be obtained. The corresponding heat flow distribution can be obtained quickly, which significantly reduces the test cost compared with the traditional method.

Claims (1)

1. A temperature distortion map inverse design method based on discrete sequences is characterized by comprising the following steps:

1) determining the incoming flow velocity and the distance between the upstream distortion section and the downstream map section according to the given temperature map distribution of the measurement section;

2) generating the number M of nozzles and the nozzle distribution according to the temperature, sequentially numbering the arrangement distribution of the nozzles, and setting a heat flow combination sequence of each nozzle as a vector α;

3) respectively dividing the downstream target map cross section into units of magnitude N, sequencing the divided micro units in a circle outward by taking the horizontal direction as a zero-angle line, performing temperature assignment on the N units in an interpolation mode based on a temperature distortion map, and recording as a map column vector β;

4) respectively constructing a column vector Q₁,…,Q_i,…,Q_MWherein Q is_iIs described as (q)₁,…,q_i,…,q_M)^TRemoving q₁Except for setting to 1, other values are set to 0, namely, the other M-1 nozzles are in the off state except that the ith nozzle is in the maximum heat flow state, wherein the maximum heat flow state is marked as 1, and the minimum heat flow state is marked as 0;

5) respectively carrying out numerical simulation or experimental study of the ith nozzle in the maximum heat flow state and other M-1 nozzles in the closed state, measuring a downstream temperature distortion map, and referring to the serial number of the step 3) to obtain a vector (a)_1i，a_2i，a_3i…a_Ni)^TA total of M nozzles forming a matrix

6) According to A α - β, a generalized inverse matrix B is taken for the matrix A, wherein BA-E_M，E_MDetermining the distribution of the heat flow by reversely calculating a heat flow distribution vector α, namely α -B β for the M-order unit matrix, wherein the maximum value 1 and the minimum value 0 in the vector α;

7) according to the obtained vector α, the states of the nozzles at the upstream are obtained, namely, a specific temperature distortion map is realized by adjusting the injection intensity of the heat flow, and the obtained result provides data support for researching the temperature distortion under the real working condition, so that the reverse design method is to reversely solve the column vector α by using the matrix A and the column vector β.

Images