CN111470075A - Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm - Google Patents

Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm Download PDF

Info

Publication number
CN111470075A
CN111470075A CN202010300579.9A CN202010300579A CN111470075A CN 111470075 A CN111470075 A CN 111470075A CN 202010300579 A CN202010300579 A CN 202010300579A CN 111470075 A CN111470075 A CN 111470075A
Authority
CN
China
Prior art keywords
pressure
flow
gas
output
oxygen
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.)
Granted
Application number
CN202010300579.9A
Other languages
Chinese (zh)
Other versions
CN111470075B (en
Inventor
尹文娟
宋涛
王猛杰
林震
焦焱
刘学
薛有
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.)
Beijing Institute of Control Engineering
Original Assignee
Beijing Institute of Control Engineering
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 Beijing Institute of Control Engineering filed Critical Beijing Institute of Control Engineering
Priority to CN202010300579.9A priority Critical patent/CN111470075B/en
Publication of CN111470075A publication Critical patent/CN111470075A/en
Application granted granted Critical
Publication of CN111470075B publication Critical patent/CN111470075B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B64—AIRCRAFT; AVIATION; COSMONAUTICS
    • B64G—COSMONAUTICS; VEHICLES OR EQUIPMENT THEREFOR
    • B64G99/00—Subject matter not provided for in other groups of this subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Remote Sensing (AREA)
  • Combined Controls Of Internal Combustion Engines (AREA)

Abstract

The invention discloses an on-orbit thrust prediction method of a spacecraft based on an artificial intelligence algorithm. The method comprises the following steps: combining ground test data to obtain a neural network model of the pressure reducer and the one-way valve; acquiring the current oxygen path and fuel path flow through the pressure and temperature of the storage tank at the previous moment; obtaining the current pressure of the storage tank through the flow data; obtaining the current output pressure of the pressure reducer and the output flow of the one-way valve by combining the current pressure of the gas cylinder and the pressure of the storage tank with a constructed neural network model, and further calculating to obtain the pressure of the gas cylinder at the next moment; and when the calculated time is larger than the ignition time, ending the steps and calculating to obtain the predicted thrust of the engine. The neural network model can be further modified by using on-orbit data, so that the modified predicted thrust can be obtained. The method achieves the purpose of predicting the on-orbit thrust with high precision by an artificial intelligence algorithm, and avoids the problem of construction of a conventional mathematical model.

Description

Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm
Technical Field
The invention relates to a spacecraft two-component propulsion system technology, in particular to a spacecraft on-orbit thrust prediction method based on an artificial intelligence algorithm, which is suitable for on-orbit thrust prediction of a two-component propulsion system satellite.
Background
The geostationary orbit satellite generally adopts a two-component propulsion system, and selects a 490N engine (dinitrogen tetroxide is used as an oxidant and methylhydrazine is used as a combustion agent) to carry out orbit transfer, and attitude control during the orbit transfer is carried out by a 10N thruster.
Before the satellite orbit transfer, the thrust of an engine needs to be predicted so as to determine a final orbit transfer strategy. In addition, due to the deviation between the ground test and the on-orbit performance, the thrust of the engine is needed to be calibrated again by depending on the on-orbit control effect of one time, and then the thrust of the engine controlled by the next time is corrected according to the calibration result, so that the accuracy of the orbit change is improved, and the waste of the propellant is avoided. Typically, performance needs to be calculated by constructing a complex mathematical model of the propulsion system to obtain the predicted thrust. The mode often needs to pertinently establish a mathematical model for products, particularly pressure reducers and one-way valves selected by a gas path system, and the universality is poor. When model correction is performed, the number of parameters involved is large, and the randomness is large.
Disclosure of Invention
The technical problem solved by the invention is as follows: compared with the prior art, the method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm is provided, and the purpose of rapidly and accurately predicting the thrust of the engine for propulsion systems with different structures is achieved.
The above object of the present invention is achieved by the following technical solutions: an on-orbit thrust prediction method of a spacecraft based on an artificial intelligence algorithm comprises the following steps:
(1) input pressure P combined with pressure reducergOutput pressure PregOutput flow rate QgObtaining the relation P through a neural networkreg＝net1(Pg,Qg)+a1，a1If the value is the value to be corrected, taking 0 if the correction is not needed; net1() A function representing the input pressure and output flow of the pressure reducer;
(2) the output pressure of the pressure reducer, the output pressure of the one-way valve and the output flow are combined, and the relational expression Q is obtained through the neural networkgo＝net2(Preg-Po)×b1，Qgf＝net3(Preg-Pf)×b2，QgoFor the output flow of the oxygen check valve, PoIs the oxygen tank pressure, QgfFor combustion of the flow, P, of the check valvefTo the combustion chamber pressure, b1、b2If the coefficient is not required to be corrected, 1 is selected; net2() Representing a function, net, related to the output pressure of the pressure reducer, the output flow of the oxygen check valve3() A function representing the output pressure of the pressure reducer and the output flow of the fuel check valve;
(3) obtaining the current oxygen path flow Q through the current pressure and temperature of the storage tank and combining with a small deviation equation of the engine flowoFuel flow Qf；
(4) By flow data Qo、QfObtaining the current oxidant tank PoAnd combustion agent tank pressure Pf；
(5) Through neural network net1Obtaining the current output pressure P of the pressure reducerreg(ii) a Through neural network net2、net3Calculating to obtain the current output flow Q of the oxygen check valve and the fuel check valvego、Qgf；
(6) Calculating the pressure P of the gas cylinder at the next moment through the output pressure of the pressure reducer and the output flow of the one-way valveg；
(7) If the current time tc< track Change time TfAnd (4) circulating the steps (3) to (6); otherwise, the calculation is finished, and the average pressure value P of the oxidant storage tank and the combustion agent storage tank is obtainedoa，PfaAnd then obtaining the predicted thrust F of the engine by a small deviation equationc；
(8) If the on-track data is needed to be used for correction, selecting a coefficient a to be corrected1、b1、b2For the argument, an objective function J is set1＝|Poa-poa|，J2＝|Pfa-pfaAnd (4) repeating the steps (1) to (7), and solving by combining a multivariable multi-target optimization algorithm to obtain the coefficient a to be corrected1、b1、b2The optimal solution of (2);
wherein p isoa、pfaReal measurement data of average pressure of an oxidant storage tank and a combustion agent storage tank are obtained;
(9) a is to1、b1、b2Substituting the step (1) and the step (2), and repeating the step (3) to the step (7) to obtain the corrected predicted thrust F of the next orbital transfercc。
The neural network model construction method in the step (1) and the step (2) comprises the following steps: and selecting a 3-layer BP neural network, and selecting purelin type functions for the transmission functions of the hidden layer and the output layer.
In the step (3), the method for calculating the flow rates of the oxygen path and the fuel path comprises the following steps:
Qo＝a11Peo+b11Pef+c11To+d11Tf+e11，
Qf＝a12Peo+b12Pef+c12To+d12Tf+e12，
in the formula, the coefficient a11、b11、c11、d11、e11、a12、b12、c12、d12、e12All obtained by a rail-controlled engine ground test; peo、ToRespectively, oxidant inlet pressure, temperature of the engine; pef、TfThe combustion agent inlet pressure and temperature of the engine, respectively;
Peo＝Po-ΔPlo，Pef＝Pf-ΔPlf；
ΔPlo、ΔPlfrespectively oxygen path and fuel path flow resistance.
In the step (4), the current oxidant storage tank PoAnd combustion agent tank pressure PfThe calculation formula of (2) is as follows:
Figure BDA0002453834570000031
Figure BDA0002453834570000032
wherein gamma is the adiabatic index of helium,
Poi、Pfipressure of oxidant tank and combustion agent tank, Q, respectivelygoi、QoiRespectively the oxygen one-way valve flow and the oxygen path flow at the previous moment, Qgfi、QfiRespectively the flow of the fuel check valve and the flow of the fuel path at the previous moment, rhoregi、ρoi、ρfiRespectively outputting the densities of gas, gas in an oxygen box and gas in a combustion box by the pressure reducer at the previous moment, wherein delta t is a time scale;
Vgo、Vgfthe gas volumes in the oxygen box and the fuel box are respectively obtained by calculating according to the following relational expressions:
Figure BDA0002453834570000033
Figure BDA0002453834570000041
Vgoi、Vgfithe volumes of the gases in the oxygen tank and the fuel tank at the last moment are respectively.
In the step (6), the pressure P of the gas cylinder at the next momentgThe calculation method comprises the following steps:
Figure BDA0002453834570000042
in the formula, VgIs the volume of the gas cylinder,
Figure BDA0002453834570000044
is the heat flux density S of gas in the gas cylinder to the unit area of the pipe wallgThe contact surface area of the gas in the gas cylinder and the pipe wall is shown; pgiRepresenting the cylinder pressure at the previous moment;
flow rate Q of pressure reducer at last momentgiSatisfies the following conditions: qgi＝Qgoi+Qgfi；
Gas density in gas cylinders ρgSatisfies the following conditions:
Figure BDA0002453834570000043
ρgirepresenting the gas density in the cylinder at the previous moment;
in the step (7), the thrust force FcThe calculation formula of (2) is as follows:
Fc＝aPoa+bPfa+cTo+dTf+e；
in the formula, the coefficients a, b, c, d and e are obtained by a ground test of an orbit control engine.
The correction coefficient a in the step (8)1、b1、b2The following constraints are satisfied: -0.3. ltoreq. a1≤0.3；0.5≤b1≤2；0.5≤b2≤2。
Compared with the prior art, the invention has the following beneficial effects:
the spacecraft on-orbit thrust prediction method based on the artificial intelligence algorithm combines the ground test data, and replaces a complex valve model by selecting a proper neural network structure, so that the prediction model is simpler and more practical and has strong universality;
the method can utilize the on-orbit data and correct the model through a multi-target multi-parameter optimization method, thereby further improving the prediction precision of the on-orbit thrust, being beneficial to improving the orbital transfer efficiency and avoiding the waste of the propellant;
the method is suitable for conventional double-component propulsion systems, including propulsion systems with parallel storage box structures, and has wide application value and popularization prospect.
Drawings
FIG. 1 is a block diagram of a two-component propulsion system in an embodiment of the present invention;
fig. 2 is a flowchart of a spacecraft on-orbit thrust prediction method based on an artificial intelligence algorithm in an embodiment of the invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Fig. 1 is a block diagram of a two-component propulsion system according to an embodiment of the present invention, and as shown in fig. 1, a typical two-component propulsion system includes gas cylinders He1, He2, He3, an oxidant tank OX, a combustion agent tank FU, a pressure reducer PR, check valves CV1, CV2, an engine L AE, an attitude control thruster RCT, pressure sensors PT1, PT2, PT3, normally closed electric explosion valves PV1, PV2, PV3, adding valves MV1, MV2, MV3, MV4, MV5, MV6, and gas test interfaces TP1, TP2, TP 3.
The gas cylinder is used for storing high-pressure gas (usually helium) from He1, He1 and He1, the gas cylinder outlet is provided with a pressure sensor PT1 and a charging and discharging valve MV1 for monitoring the pressure of the gas cylinder and charging the gas cylinder respectively, tank OX is used for storing oxidant and helium, tank FU is used for storing combustion agent and helium, propellant tanks OX and FU each comprise an air inlet and an air outlet, the air inlet and the air outlet are located at the top, the air inlet and the liquid outlet are located at the bottom, upstream and downstream liquid ports of tank OX are respectively provided with charging and discharging valves MV1 and MV1 for charging or discharging gas and propellant respectively, tank OX downstream is provided with a pressure sensor PT1 for monitoring the pressure of oxidant tank OX, upstream and downstream liquid ports of tank FU are respectively provided with charging and discharging valves MV1 and MV1, and PV1 for charging or discharging gas and propellant tanks PV 72, respectively, and helium tanks PV for preventing the gas from being blown out from the gas cylinder, PV, and propellant when the pressure of the gas cylinder is measured by a one-way pressure sensor PR, PV and PV are respectively installed between the gas cylinder PV, PV 72 for preventing the gas cylinder PV, PR for testing the gas cylinder, PV, PR for testing the gas cylinder, PV and PV for testing the gas cylinder, PV, PR for testing the gas cylinder, PV.
Fig. 2 is a flowchart of a spacecraft on-orbit thrust prediction method based on an artificial intelligence algorithm in an embodiment of the present invention, where the method is based on a two-component propulsion system shown in fig. 1, and referring to fig. 2, the spacecraft on-orbit thrust prediction method based on the artificial intelligence algorithm provided in this embodiment specifically includes the following steps:
(1) combined with surface data (input pressure P of pressure reducer)gOutput pressure PregOutput flow rate Qg) Obtaining the relation p through a neural networkreg＝net1(Pg,Qg)+a1，a1If the correction value is not needed, the value is 0. Net1() Representing a function related to the input pressure, output flow of the reducer. The neural network model selects a 3-layer BP neural network, and the transmission functions of the hidden layer and the output layer both select purelin type functions.
(2) The relation Q is obtained by combining the ground data (the output pressure of the pressure reducer, the output pressure of the one-way valve and the output flow) through the neural networkgo＝net2(Preg-Po)×b1，Qgf＝net3(Preg-Pf)×b2，QgoFor the output flow of the oxygen check valve, PoIs the oxygen tank pressure, QgfFor combustion of the flow, P, of the check valvefTo the combustion chamber pressure, b1、b2And 1 is taken if the coefficient is not required to be corrected. Net2() Representing a function, net, related to the output pressure of the pressure reducer, the output flow of the oxygen check valve3() A function relating to the output pressure of the pressure reducer and the output flow rate of the fuel check valve is shown. The neural network model selects a 3-layer BP neural network, and the transmission functions of the hidden layer and the output layer both select purelin type functions.
(3) Obtaining the current oxygen path flow Q through the current pressure and temperature of the storage tank and combining with a small deviation equation of the engine flowoFuel flow Qf. The specific calculation method comprises the following steps:
Qo＝a11Peo+b11Pef+c11To+d11Tf+e11
Qf＝a12Peo+b12Pef+c12To+d12Tf+e12
in the formula, the coefficient a11、b11、c11、d11、e11、a12、b12、c12、d12、e12All obtained by rail control engine ground test. Peo、ToRespectively, oxidant inlet pressure, temperature, P of the engineef、TfRespectively, the combustion agent inlet pressure, temperature of the engine. Wherein, the temperature To、TfThe tank head temperature is typically selected. The inlet pressure of the engine being obtained by subtracting the line flow resistance from the reservoir pressure, i.e.
Peo＝Po-ΔPlo，Pef＝Pf-ΔPlf
ΔPlo、ΔPlfRespectively oxygen path and fuel path flow resistance.
(4) By flow data Qo、QfObtaining the current oxidant tank PoAnd combustion agent tank pressure Pf. The specific calculation method comprises the following steps:
Figure BDA0002453834570000071
Figure BDA0002453834570000072
where γ is the adiabatic index of helium, typically 1.67.
Poi、PfiThe pressures of the oxidant storage tank and the combustion agent storage tank, the densities of the output gas, the gas in the oxygen tank and the gas in the combustion tank, Qgoi、QoiRespectively the oxygen one-way valve flow and the oxygen path flow at the previous moment, Qgfi、QfiRespectively the flow of the fuel check valve and the flow of the fuel path at the previous moment, rhoregi、ρoi、ρfiRespectively outputting the densities of gas, gas in an oxygen box and gas in a combustion box by the pressure reducer at the previous moment, wherein delta t is a time scale;
Vgo、Vgfthe gas volumes in the oxygen box and the fuel box are respectively obtained by calculating according to the following relational expressions:
Figure BDA0002453834570000073
Figure BDA0002453834570000074
Vgoi、Vgfithe volumes of the gases in the oxygen tank and the fuel tank at the last moment are respectively.
(5) Through neural network net1Obtaining the current output pressure p of the pressure reducerreg(ii) a Through neural network net2、net3Calculating to obtain the current output flow Q of the oxygen check valve and the fuel check valvego、Qgf。
(6) Tong (Chinese character of 'tong')The output pressure of the over-pressure reducer and the output flow of the one-way valve are calculated to obtain the pressure P of the gas cylinder at the next momentg. The specific calculation method comprises the following steps:
Figure BDA0002453834570000081
in the formula, VgIs the volume of the gas cylinder,
Figure BDA0002453834570000083
is the heat flux density S of gas in the gas cylinder to the unit area of the pipe wallgThe contact surface area of the gas in the gas cylinder and the pipe wall. PgiRepresenting the cylinder pressure at the previous moment;
flow rate Q of pressure reducer at last momentgiSatisfies the following conditions: qgi＝Qgoi+Qgfi；
Gas density in gas cylinders ρgSatisfies the following conditions:
Figure BDA0002453834570000082
ρgirepresenting the gas density in the cylinder at the previous moment;
(7) if the current time tc< track Change time TfAnd (4) circulating the steps (3) to (6); otherwise, the calculation is finished, and the average pressure value P of the oxidant storage tank and the combustion agent storage tank is obtainedoa，PfaAnd then obtaining the predicted thrust F of the engine by a small deviation equationc. Thrust force FcThe specific calculation method comprises the following steps:
Fc＝aPoa+bPfa+cTo+dTf+e
in the formula, the coefficients a, b, c, d and e are obtained by a ground test of an orbit control engine.
(8) If the on-track data is needed to be used for correction, selecting a coefficient a to be corrected1、b1、b2Selecting a certain time of orbital transfer parameter as an initial condition for independent variable, wherein the initial condition comprises the parameters of the propulsion system before orbital transfer and during orbital transfer, and the average pressure p of an oxidant storage tank during orbital transferoaAverage pressure of oxidant storage tankpfaTo target, an objective function J is set1＝|Poa-poa|，J2＝|Pfa-pfa|。
Furthermore, the correction coefficient a1、b1、b2The following constraints are satisfied:
-0.3≤a1≤0.3；0.5≤b1≤2；0.5≤b2≤2
repeating the steps (1) to (7), combining a multivariate multi-target optimization algorithm NSGA-II algorithm, and when all target functions converge to a threshold value (generally taking 1KPa) meeting the precision requirement, stopping calculation to obtain a coefficient a to be corrected1、b1、b2The optimal solution of (2);
(9) a is to1、b1、b2The step (1) and the step (2) are carried in, and the corrected predicted thrust F of the next orbital transfer can be obtained through the step (3) to the step (7)cc。
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims (7)

1. An on-orbit thrust prediction method of a spacecraft based on an artificial intelligence algorithm is characterized by comprising the following steps:
(1) input pressure P combined with pressure reducergOutput pressure PregOutput flow rate QgObtaining the relation P through a neural networkreg＝net1(Pg,Qg)+a1，a1If the value is the value to be corrected, taking 0 if the correction is not needed; net1() Indicating input pressure to pressure reducerForce, output flow related functions;
(2) the output pressure of the pressure reducer, the output pressure of the one-way valve and the output flow are combined, and the relational expression Q is obtained through the neural networkgo＝net2(Preg-Po)×b1，Qgf＝net3(Preg-Pf)×b2，QgoFor the output flow of the oxygen check valve, PoIs the oxygen tank pressure, QgfFor combustion of the flow, P, of the check valvefTo the combustion chamber pressure, b1、b2If the coefficient is not required to be corrected, 1 is selected; net2() Representing a function, net, related to the output pressure of the pressure reducer, the output flow of the oxygen check valve3() A function representing the output pressure of the pressure reducer and the output flow of the fuel check valve;
(3) obtaining the current oxygen path flow Q through the current pressure and temperature of the storage tank and combining with a small deviation equation of the engine flowoFuel flow Qf；
(4) By flow data Qo、QfObtaining the current oxidant tank PoAnd combustion agent tank pressure Pf；
(5) Through neural network net1Obtaining the current output pressure P of the pressure reducerreg(ii) a Through neural network net2、net3Calculating to obtain the current output flow Q of the oxygen check valve and the fuel check valvego、Qgf；
(6) Calculating the pressure P of the gas cylinder at the next moment through the output pressure of the pressure reducer and the output flow of the one-way valveg；
(7) If the current time tc< track Change time TfAnd (4) circulating the steps (3) to (6); otherwise, the calculation is finished, and the average pressure value P of the oxidant storage tank and the combustion agent storage tank is obtainedoa，PfaAnd then obtaining the predicted thrust F of the engine by a small deviation equationc；
(8) If the on-track data is needed to be used for correction, selecting a coefficient a to be corrected1、b1、b2For the argument, an objective function J is set1＝|Poa-poa|，J2＝|Pfa-pfaAnd (4) repeating the steps (1) to (7), and solving by combining a multivariable multi-target optimization algorithm to obtain the coefficient a to be corrected1、b1、b2The optimal solution of (2);
wherein p isoa、pfaReal measurement data of average pressure of an oxidant storage tank and a combustion agent storage tank are obtained;
(9) a is to1、b1、b2Substituting the step (1) and the step (2), and repeating the step (3) to the step (7) to obtain the corrected predicted thrust F of the next orbital transfercc。
2. The method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm, according to the claim 1 or 2, is characterized in that: the neural network model construction method in the step (1) and the step (2) comprises the following steps: and selecting a 3-layer BP neural network, and selecting purelin type functions for the transmission functions of the hidden layer and the output layer.
3. The method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm, according to claim 2, is characterized in that: in the step (3), the method for calculating the flow rates of the oxygen path and the fuel path comprises the following steps:
Qo＝a11Peo+b11Pef+c11To+d11Tf+e11，
Qf＝a12Peo+b12Pef+c12To+d12Tf+e12，
in the formula, the coefficient a11、b11、c11、d11、e11、a12、b12、c12、d12、e12All obtained by a rail-controlled engine ground test; peo、ToRespectively, oxidant inlet pressure, temperature of the engine; pef、TfThe combustion agent inlet pressure and temperature of the engine, respectively;
Peo＝Po-ΔPlo，Pef＝Pf-ΔPlf；
ΔPlo、ΔPlfrespectively oxygen path and fuel path flow resistance.
4. The method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm, according to claim 3, is characterized in that: in the step (4), the current oxidant storage tank PoAnd combustion agent tank pressure PfThe calculation formula of (2) is as follows:
Figure FDA0002453834560000021
Figure FDA0002453834560000022
wherein gamma is the adiabatic index of helium,
Poi、Pfipressure of oxidant tank and combustion agent tank, Q, respectivelygoi、QoiRespectively the oxygen one-way valve flow and the oxygen path flow at the previous moment, Qgfi、QfiRespectively the flow of the fuel check valve and the flow of the fuel path at the previous moment, rhoregi、ρoi、ρfiRespectively outputting the densities of gas, gas in an oxygen box and gas in a combustion box by the pressure reducer at the previous moment, wherein delta t is a time scale;
Vgo、Vgfthe gas volumes in the oxygen box and the fuel box are respectively obtained by calculating according to the following relational expressions:
Figure FDA0002453834560000031
Figure FDA0002453834560000032
Vgoi、Vgfiare respectively the last oneThe gas volumes in the oxygen tank and the fuel tank at the moment.
5. The method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm, according to claim 4, is characterized in that: in the step (6), the pressure P of the gas cylinder at the next momentgThe calculation method comprises the following steps:
Figure FDA0002453834560000033
in the formula, VgIs the volume of the gas cylinder,
Figure FDA0002453834560000034
is the heat flux density S of gas in the gas cylinder to the unit area of the pipe wallgThe contact surface area of the gas in the gas cylinder and the pipe wall is shown; pgiRepresenting the cylinder pressure at the previous moment;
flow rate Q of pressure reducer at last momentgiSatisfies the following conditions: qgi＝Qgoi+Qgfi；
Gas density in gas cylinders ρgSatisfies the following conditions:
Figure FDA0002453834560000035
ρgirepresenting the gas density in the cylinder at the previous time.
6. The method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm, according to claim 5, is characterized in that: in the step (7), the thrust force FcThe calculation formula of (2) is as follows:
Fc＝aPoa+bPfa+cTo+dTf+e；
in the formula, the coefficients a, b, c, d and e are obtained by a ground test of an orbit control engine.
7. The method for predicting the on-orbit thrust of the spacecraft based on the artificial intelligence algorithm, according to claim 6, is characterized in that: correction in the step (8)Coefficient a1、b1、b2The following constraints are satisfied: -0.3. ltoreq. a1≤0.3；0.5≤b1≤2；0.5≤b2≤2。
CN202010300579.9A 2020-04-16 2020-04-16 Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm Active CN111470075B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010300579.9A CN111470075B (en) 2020-04-16 2020-04-16 Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010300579.9A CN111470075B (en) 2020-04-16 2020-04-16 Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm

Publications (2)

Publication Number Publication Date
CN111470075A true CN111470075A (en) 2020-07-31
CN111470075B CN111470075B (en) 2021-09-07

Family

ID=71753710

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010300579.9A Active CN111470075B (en) 2020-04-16 2020-04-16 Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm

Country Status (1)

Country Link
CN (1) CN111470075B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111914366A (en) * 2020-08-05 2020-11-10 湖南航天机电设备与特种材料研究所 Method for acquiring cylinder outlet speed of high-pressure cold air launching aircraft
CN113239644A (en) * 2021-04-30 2021-08-10 北京控制工程研究所 Working point determining method suitable for double-component propulsion system
CN115324772A (en) * 2022-07-28 2022-11-11 北京控制工程研究所 Method for predicting mixing ratio of propellant of double-component thruster

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2150411C1 (en) * 1998-04-24 2000-06-10 Криворотов Александр Семенович Multistage rocket for change of trajectory of motion of asteroid
US20030024235A1 (en) * 2001-07-31 2003-02-06 Pisano Alan David Control strategy for gas turbine engine
CN107776916A (en) * 2017-09-14 2018-03-09 北京控制工程研究所 A kind of method based on the adjustment discharge without gas bypassing propulsion system
US10384813B2 (en) * 2016-11-16 2019-08-20 Jerome Drexler Spacecraft landing and site-to-site transport for a planet, moon or other space body
CN110525694A (en) * 2019-07-23 2019-12-03 中国空间技术研究院 A kind of propulsion system mixture ratio control method considering tank pressure gap in parallel

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2150411C1 (en) * 1998-04-24 2000-06-10 Криворотов Александр Семенович Multistage rocket for change of trajectory of motion of asteroid
US20030024235A1 (en) * 2001-07-31 2003-02-06 Pisano Alan David Control strategy for gas turbine engine
US10384813B2 (en) * 2016-11-16 2019-08-20 Jerome Drexler Spacecraft landing and site-to-site transport for a planet, moon or other space body
CN107776916A (en) * 2017-09-14 2018-03-09 北京控制工程研究所 A kind of method based on the adjustment discharge without gas bypassing propulsion system
CN110525694A (en) * 2019-07-23 2019-12-03 中国空间技术研究院 A kind of propulsion system mixture ratio control method considering tank pressure gap in parallel

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111914366A (en) * 2020-08-05 2020-11-10 湖南航天机电设备与特种材料研究所 Method for acquiring cylinder outlet speed of high-pressure cold air launching aircraft
CN111914366B (en) * 2020-08-05 2024-04-26 湖南航天机电设备与特种材料研究所 Method for acquiring cylinder-out speed of high-pressure cold air emission aircraft
CN113239644A (en) * 2021-04-30 2021-08-10 北京控制工程研究所 Working point determining method suitable for double-component propulsion system
CN113239644B (en) * 2021-04-30 2023-12-12 北京控制工程研究所 Working point determining method suitable for double-component propulsion system
CN115324772A (en) * 2022-07-28 2022-11-11 北京控制工程研究所 Method for predicting mixing ratio of propellant of double-component thruster
CN115324772B (en) * 2022-07-28 2024-05-31 北京控制工程研究所 Method for predicting mixing ratio of propellants of two-component thruster

Also Published As

Publication number Publication date
CN111470075B (en) 2021-09-07

Similar Documents

Publication Publication Date Title
CN111470075B (en) Spacecraft on-orbit thrust prediction method based on artificial intelligence algorithm
CA2616990C (en) Fuel cell system and generation control device
Whitmore et al. Engineering model for self-pressurizing saturated-N2O-propellant feed systems
CN111426364A (en) Propellant residual quantity measuring method based on multi-source data fusion
US11092290B2 (en) Method for measuring fluid level in liquid hydrogen tank and liquid hydrogen storage system
CN112378670B (en) Rocket engine fault detection method based on improved particle filtering
Sukharev et al. Identification of model flow parameters and model coefficients with the help of integrated measurements of pipeline system operation parameters
CN112613116A (en) Petri net quantum Bayesian fault diagnosis method for liquid rocket engine starting stage
CN115344019A (en) Natural gas metering flow adjusting process based on composite intelligent algorithm
CN113759726B (en) Natural gas odorizing control system and control method thereof
CN112768735B (en) Estimation method of tail exhaust hydrogen concentration of fuel cell system
CN105402454B (en) Gas supply adjusting device based on pressure reducer
US20230098469A1 (en) Method and system for computing a transition parameter of a liquefied gas storage medium
US11598289B1 (en) Propellant volume and mixture ratio control
CN112502856B (en) Liquid oxygen and oxygen double-path adjustable supply system
CN113239644A (en) Working point determining method suitable for double-component propulsion system
Iannetti et al. Development of model-based fault diagnosis algorithms for MASCOTTE cryogenic test bench
Kazmi et al. A nonlinear observer for PEM fuel cell system
Tomi et al. Development of air supply controller for FCV based on model-based development approach
Bettocchi et al. Gas turbine health indices determination by using neural networks
Han et al. Air flow trajectory and surge avoidance of centrifugal compressor under variable pressure operation of automotive fuel cells
Ma et al. Study on air pressure-flow decoupling control in fuel cell system based on feedforward algorithm
US20230145651A1 (en) Method for controlling a fuel cell system having a hydrogen fuel injector/ejector
CN113609754B (en) Air quantity distribution control method for sewage aeration
Yin et al. On-orbit Engine Thrust Prediction Algorithm for Geosynchronous Satellites based on Neural Network

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