Information redundancy processing method for dual pressure sensors of solid attitude and orbit control engine

Abstract

The invention discloses a solid attitude and orbit control engine dual-pressure sensor information redundancy processing method, which comprises a preprocessing step, a third information acquisition step and a redundancy processing step, and specifically comprises the following steps: firstly, carrying out smooth processing and boundary condition conformity judgment on dual-sensor signals, preliminarily eliminating invalid information, and carrying out redundancy processing; and then resolving the ballistic model in the zero dimension to obtain a pressure simulation value, processing in combination with the indication values of the double sensors to obtain third information, and when the difference between the indication values of the double sensors is within an allowable range, adopting the weighted average value of the indication values of the double sensors as a final output value, otherwise, adopting the third information to choose. According to the method, the pressure intensity of the inner ballistic trajectory is estimated through inner ballistic trajectory simulation calculation, information fusion is carried out by adopting the zero-dimensional inner ballistic trajectory model pressure intensity simulation value of the solid engine and the double pressure intensity sensors, and the problem that the solid attitude and orbit control engine can only be provided with two pressure intensity sensors at most for redundant fault-tolerant design under the limitation of the cost and the size is solved.

Description

Technical Field

[0001] The present invention relates to the field of solid attitude and orbit control engine performance simulation, and in particular to a method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine. Background Art

[0002] Solid attitude and trajectory control engines are usually composed of a gas generator, several gas valves, pressure sensors, position sensors, pressure loop controllers, position loop controllers, etc. They can perform variable pressure control and constant pressure control of the engine's internal trajectory by changing the total opening of the valve. Because they can provide direct lateral force and enable the missile to quickly correct its trajectory at the end of guidance, thereby improving the missile's hit rate, they are mostly used for trajectory control at the end of the missile's trajectory.

[0003] Solid attitude and orbit control engines rely on pressure sensor signals when working. When the pressure sensor signal is inaccurate, the output thrust may be inaccurate, affecting the flight accuracy of the aircraft. In severe cases, the pressure may exceed the safety threshold and cause the aircraft to disintegrate. Therefore, the reliability and accuracy of the pressure sensor signal of the solid attitude and orbit control engine is crucial to the success of the flight mission.

[0004] In order to ensure the reliability and accuracy of the pressure signal of the solid attitude and orbit control engine, it is preferred to install multiple sensors. However, the solid attitude and orbit control engine has the characteristic of small size due to its application scenario. Under certain volume and cost constraints, some models can only install two pressure sensors at most. When one sensor fails, the other sensor can be used as a backup. However, the working environment of the pressure sensor of the solid attitude and orbit control engine is harsh, and the failure form is complex. There will be a situation where the signals of the two pressure sensors are both in the valid range but the gap is large. Without other information assistance, it is impossible to judge the fault. Summary of the invention

[0005] In order to solve the above problems, the present invention introduces the engine zero-dimensional interior ballistic simulation calculation value to fuse the dual pressure sensor information, and provides a method for processing the redundant information of the dual pressure sensors of the solid attitude and orbit control engine, which can effectively solve the problem that the fault judgment cannot be performed when the two pressure sensor signals are both in the valid range but the gap is large. The specific technical solution is as follows:

[0006] S1. Preprocessing steps

[0007] First, the two pressure sensor signals are smoothed, and then the boundary condition compliance is judged respectively. When both pressure sensor signals are in compliance, the next step is entered to obtain the final output result. Otherwise, the sensor signal that fails the compliance judgment is excluded and will not be reused until the engine operation is finished.

[0008] Preferably, the smoothing of the two sensor signals is specifically: the two sensor signals are sequentially subjected to low-pass filtering and average filtering;

[0009] Preferably, the mathematical model of the low-pass filter is:

[0010] y(k1)=(1-θ)×y(k1-1)+θ×x(k1)

[0011] Where: k1 is the current sensor sampling period, k1-1 is the previous sensor sampling period, x(k1) is the current input initial value, y(k1-1) is the previous output value, y(k1) is the current calculated output value, and θ is the filter coefficient;

[0012] Preferably, the average value filtering is as follows: the pressure data is updated once every period of time, and an anti-pulse interference sliding average value filtering is performed, and the arithmetic mean value of the first m data is calculated in real time by discarding the maximum and minimum numbers, where m is less than or equal to the control period T, and m is greater than 2.

[0013] Preferably, the boundary condition compliance judgment is specifically: judging the value P of the pressure sensor A senserA And the pressure sensor B shows the value P senserB Is it in the valid range [P min , P max ], where P min is the estimated minimum pressure of the engine, P max Estimate the maximum pressure for the engine;

[0014] S2. Third information acquisition step

[0015] Solve the zero-dimensional interior ballistic model of the solid attitude and orbit control engine to obtain the pressure simulation value P sim , establish a dual pressure sensor based on the indication P senserA , P senserB And the pressure simulation value P sim The third information calculation model P(P senserA， P senserB， P sim ), solve the model to obtain the third information;

[0016] Preferably, the zero-dimensional interior ballistic model of the solid attitude and orbit control engine is:

[0017]

[0018] In the formula, ρ P is the propellant density, A b is the combustion area, a is the burning rate coefficient, n is the pressure index, A t is the total valve pneumatic throat area, V Cis the free volume of the combustion chamber, Γ is the function of the specific heat ratio, c * is the characteristic velocity of the propellant, P sim is the pressure simulation value, t is the time;

[0019] The third information calculation model P(P senserA， P senserB， P sim ) specifically:

[0020] Preferably, when the engine is in constant pressure operation, the pressure simulation correction value P is taken St.sim is the third information, namely P(P senserA， P senserB， P sim )=P St.sim , the calculation process of the third information calculation model is as follows:

[0021] A1: The engine ignition time is t=0, and the time of entering the constant pressure working mode is t0. i =t0+i△t (i=0, 1, 2, ...), the weighted average value P senserAVE and pressure simulation value P sim Make a difference and record the value as [t i , t i+1 ] time period, denoted as ΔP AveSim_i , that is, Δp AveSim_i =p sensorAVE (t i )-p sim (t i );

[0022] A2: Set the current pressure simulation value P sim The pressure difference ΔP during this period AveSim_i Sum and get the pressure simulation correction value P for this time period St.sim , that is, p St.sim (t) = p sensorAVE (t)+Δp AveSim_i , where t i ≤t<t i+1 ;

[0023] Preferably, when the engine is in variable pressure operation, the pressure simulation differential value ΔP is taken St.sim is the third information, namely P(P senserA， P senserB， P sim )=ΔP St.sim , pressure simulation differential value ΔP St.sim is the pressure simulation value P sim The forward difference of is expressed as:

[0024]

[0025] Among them, T is the control period, k2 is the current control period, k2-1 is the previous control period, P sim (k2) is the pressure simulation value of the current control cycle, P sim (k2) is the pressure simulation value of the previous control cycle.

[0026] S3. Redundancy processing steps

[0027] The difference between the two sensor readings is calculated. When the difference is less than or equal to the preset value ΔP AB_set When the weighted average value P of the two is taken senserAVE is the final output;

[0028] When the difference between the two sensor readings is greater than the preset value ΔP AB_set When the sensor A value P is selected according to the third information, senserA Or sensor B indicates value P senserB is the final output, sensor A is effective for n times senserA The initial value is 1. When the sensor A is taken every M times, the final output is n senserA Add 1, sensor B is valid for n times senserB The initial value is 1. When the sensor B value is taken every M times, the final output is n senserB Add 1, when n senserA or senserB If the pressure resolution is greater than the fault threshold N, the other sensor is marked as faulty and the value is no longer reused. AB_set is a configurable parameter, which means that the present invention can distinguish the difference between the dual sensor signals greater than ΔP AB_set The weighted parameter M is a settable parameter. The lower the M value, the easier it is for the weighted average of the dual sensors to accept the trusted pressure value. The fault threshold N is a settable parameter. The lower the N value, the easier it is for the dual sensor redundant processing method to permanently exclude untrusted pressure values. The M and N values ​​must be selected according to demand. If they are too low, they are easily affected by interference signals and the wrong pressure is selected. If they are too high, the fault pressure is easily selected multiple times, affecting the stability of the system.

[0029] Preferably, the weighted average value P senserAVE The calculation formula is:

[0030] p sensorAVE =(n sensorA p sensorA +n sensorB p sensorB ) / (n sensorA +n sensorB )

[0031] Among them, n senserA and nsenserB The initial value is 1.

[0032] Preferably, when the difference between the two sensor readings is greater than the pressure resolution ΔP AB_set When the sensor A value P is selected according to the third information, senserA Or sensor B indicates value P senserB The final output is:

[0033] When the engine is in constant pressure operation, calculate the two sensor values ​​and the third information pressure simulation correction value P St.sim The sensor with the smaller difference is taken as the final output.

[0034] When the engine is in constant pressure operation, calculate the forward differential value of the two sensors and the third information pressure simulation differential value ΔP St.sim The sensor value with the smaller difference is taken as the final output.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] (1) When there are only two signals, the third party information can be combined to make a fault judgment. The present invention proposes a dual sensor information redundancy processing process that introduces the third information for the solid attitude and orbit control engine. The process has a weighted constant M, a fault threshold N and a pressure resolution △P. AB_set Three adjustable parameters can adjust the resolution and sensitivity of the method to faulty sensor signals, making it suitable for solid attitude and orbit control engines with various performance mission objectives.

[0037] (2) Since there is deposition and ablation during the combustion process of the solid attitude and orbit control engine, there is a certain gap between the pressure simulation value of the zero-dimensional interior ballistic model and the actual pressure value of the engine. The present invention does not directly use the strong simulation value as the judgment standard, but designs different third information acquisition models according to the characteristics of different working modes of the solid attitude and orbit control engine. When the engine is working in variable pressure, the difference of the zero-dimensional interior ballistic model simulation value is selected as the third information. When the engine is working in constant pressure, the historical measurement value of the sensor is used to correct the pressure calculation value, and the correction value is selected as the third information. After such processing, the pressure simulation value of the zero-dimensional interior ballistic model is very close to the actual engine interior ballistic performance. The third information is used for judgment, which reduces the possibility of misjudgment of faults.

[0038] (3) When the difference between the two pressure sensors is within the allowable range, the weighted average of the two sensors is taken as the final output. The weights in this weighted average represent their respective accuracies, which are determined based on the performance of the previous values. This is more accurate than directly outputting the pressure of a certain sensor or the arithmetic average. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of a method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine provided by the present invention;

[0041] Figure 2 The principle block diagram of the solid attitude and orbit control engine system provided by the present invention;

[0042] Figure 3 The pressure sensor A and the pressure sensor B in Example 1 of the present invention and the pressure ring adoption value curve obtained after applying the present invention;

[0043] Figure 4 is the effective number n of sensor A in Example 1 of the present invention senserA And sensor B effective times n senserB curve;

[0044] Figure 5 In Example 1 of the present invention, the pressure sensor A indicates the value, the pressure sensor B indicates the value, and the pressure simulation value P sim And pressure simulation correction value P St.sim curve;

[0045] Figure 6 The pressure sensor A and the pressure sensor B in Example 2 of the present invention and the pressure ring adoption value curve obtained after applying the present invention;

[0046] Figure 7 is the effective number n of sensor A in Example 2 of the present invention senserA And sensor B effective times n senserB curve.

[0047] Specific implementation methods

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The present invention provides a method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine. Figure 1 As shown, the specific steps include:

[0050] First, the engine performance parameters are obtained, including the equivalent throat area of ​​the nozzle, the initial free volume, the change law of the charge burning surface with the combustion thickness and the propellant parameters, and then the processing flow described below is entered. The following processing flow is started and executed at a fixed time interval and can be set as the control cycle of the controller. It is worth noting that the engine performance parameters only need to obtain relatively accurate values ​​under the existing conditions, and a certain error with the actual value is allowed.

[0051] Step 1: The two sensor signals are filtered by low-pass filtering and average filtering in turn to obtain the two sensor indications P senserA and P senserB The average value filtering is preferably: the pressure data is updated once every period of time, and an anti-pulse interference sliding average filtering is performed: the maximum and minimum two numbers of the first m data are discarded in real time to calculate the arithmetic average, and the low-pass filtering is preferably y(k)=(1-θ)×y(k-1)+θ×x(k), where: x(k) is the current input value, the initial value is 0, y(k-1) is the previous output value, y(k) is the current calculated output value, and θ is the filter coefficient. The function of this step is to smooth the sensor signal and remove clutter interference;

[0052] Step 2: Determine the value P of pressure sensor A senserA And the pressure sensor B shows the value P senserB Is it in the valid range [P min , P max ], if they are all in the valid range, go to step 3; if it is detected that the indication of a certain pressure sensor is not in the valid range, the pressure sensor is judged to be faulty, and the indication of another sensor is used as the final output, and the faulty sensor is no longer reused until the work is completed; where P min is the estimated minimum pressure of the engine, P max Estimate the maximum pressure for the engine. The purpose of this step is to remove the sensor signal with obvious working failure;

[0053] Step 3: Determine the number of times sensor A is effective n senserA And sensor B effective times n senserB Is it greater than the fault threshold N? If n senserA If it is greater than N, the pressure sensor B is judged to be faulty, and the pressure sensor A is used as the final output. The faulty sensor is no longer reused until the end of the work. If n senserB If it is greater than N, the pressure sensor A is judged to be faulty, and the pressure sensor B is used as the final output, and the faulty sensor is no longer reused until the work is completed; if n senserA and n senserBIf all are not greater than N, then go to step 4; this step is used to exclude pressure sensors that have multiple erroneous values. When N is smaller, it is easier to determine that a sensor signal is a fault signal and permanently exclude the sensor.

[0054] Step 4: Determine the difference between the two sensors |P senserA -P senserB |With pressure resolution ΔP AB_set When the difference is less than or equal to the pressure resolution ΔP AB_set When the weighted average value P of the two is taken senserAVE is the final output, and then proceed to step 6;

[0055] When the difference between the two sensor readings is greater than the preset value ΔP AB_set When , go to step 5; the weights in the weighted average in this step represent the respective accuracy, and the accuracy is determined according to the previous value. The weighted average P senserAVE The calculation formula is preferably:

[0056] p sensorAVE =(n sensorA p sensorA +n sensorB p sensorB ) / (n sensorA +n sensorB )

[0057] Among them, n senserA and n senserB The initial value is 1;

[0058] The pressure resolution ΔP in this step AB_set Represents the resolution of the method, that is, it can distinguish the difference between the two sensor signals greater than ΔP AB_set Failure;

[0059] Step 5: Calculate the model P(P) based on the third information senserA， P senserA， P sim ) to make a choice, determine that a certain sensor value is correct, and take this value as the final output, n senserA The initial value is 1. When the sensor A is taken every M times, the final output is n senserA Add 1, n senserB The initial value is 1. When the sensor B value is taken every M times, the final output is n senserB Add 1. The weight parameter M in this step is a configurable parameter. The larger M is, the greater n senserA or senserB The slower the increase speed, that is, sensor A or sensor B needs to be judged correct more times to make n senserA or senserB The value of increases by the weighted average P of step 5 senserAVEFrom the calculation formula, we know that n senserA and n senserB represents the weight of the corresponding sensor value, which means that the larger the M is in this step, the greater the weighted average value P senserAVE Easier access to properly functioning sensor indications;

[0060] Step 6: Determine whether the engine has finished working, if so, then finish, otherwise return to step 1.

[0061] In this example, the calculation model P(P) according to the third information in step 5 is senserA， P senserA， P sim ) to make a choice, determine that a certain sensor value is correct, and take this value as the final output:

[0062] First, solve the zero-dimensional interior ballistic model of the solid attitude and orbit control engine to obtain the pressure simulation value P sim , the zero-dimensional interior ballistic model of the solid rocket motor is:

[0063]

[0064] In the formula, ρ P is the propellant density, A b is the combustion area, a is the burning rate coefficient, n is the pressure index, A t is the total valve aerodynamic throat area. V C is the free volume of the combustion chamber, Γ is the function of the specific heat ratio k, c * is the characteristic velocity of the propellant, P sim is the pressure in the combustion chamber, t is the time;

[0065] When the engine is working at constant pressure, take the pressure simulation correction value P St.sim is the third information, namely P(P senserA， P senserB， P sim )=P St.sim , calculate the two sensor indications and the third information pressure simulation correction value P St.sim The difference between the two values ​​is taken, and the sensor value with the smaller difference is taken as the final output. The calculation process of the third information calculation model when the engine is in constant pressure operation is as follows:

[0066] A1: Let the engine ignition time be t=0, and the time of entering the constant pressure working mode be t0. Whenever the time t is ti=t0+i△t (i=0, 1, 2, ...), the weighted average value P senserAVE and pressure simulation value P sim Make a difference and record the pressure difference in the time period [ti, ti+1), which is set as ΔP AveSim_i , whose mathematical expression is Δp AveSim_i =psensorAVE (t i )-p sim (t i )

[0067] A2: Set the current pressure simulation value P sim The pressure difference ΔP during this period AveSim_i Sum and get the pressure simulation correction value P for this time period St.sim , whose mathematical expression is p St.sim (t) = p sensorAVE (t)+Δp AveSim_i , where t i ≤t<t i+1 ;

[0068] When the engine is in variable pressure operation, take the pressure simulation differential value ΔP St.sim is the third information, namely P(P senserA， P senserB， P sim )=ΔP St.sim , calculate the forward differential value of the two sensors and the third information pressure simulation differential value ΔP St.sim The difference between the two values ​​is taken as the final output. The mathematical expression of the third information calculation model when the engine is in variable pressure operation is as follows:

[0069]

[0070] Where T is the controller control period.

[0071] Next, taking a certain type of solid attitude and orbit control engine as an example, two specific cases are given for verification, and the implementation method of the present invention is further described.

[0072] A certain type of solid attitude and orbit control engine system consists of a gas generator, a gas valve, a pressure sensor, a position sensor, a pressure loop controller, a position loop controller, etc. Figure 2 As shown, the pressure ring controller can plan the opening of each valve through the expected pressure and the pressure feedback from the pressure sensor, and then control the execution of each valve through the position ring controller, ultimately achieving the regulation of the combustion chamber pressure and nozzle thrust.

[0073] The method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine described in the present invention is applied to the pressure control loop. The engine is ignited at time 0 and enters the constant pressure working mode. After 6.2 seconds, the pressure is changed and the pressure drops. After 7 seconds, it enters the constant pressure working mode again. During the constant pressure mode of 0 to 6.2 seconds, a pressure closed-loop control algorithm is used to stabilize the pressure at a certain value.

[0074] Example 1

[0075] Sensor A works normally, and the pressure measuring port of sensor B is blocked by foreign matter at the 3rd second, causing its indication to drop continuously to atmospheric pressure from the 3rd second. The engine performance parameters are obtained as shown in Table 1. The sampling period of the pressure sensor is 0.05ms, and the pressure loop control period T is 1ms.

[0076] Table 1 Engine operating parameters

[0077] parameter unit Numeric Burning rate mm / s 7(6MPa) Characteristic speed m / s 1200 Specific heat ratio — 1.2 Propellant density <![CDATA[g / cm 3 ]]> 1.4 Burning Noodles <![CDATA[ 2 ]]> 80000 Initial free volume L 10 Stress Index — 0.5 Equivalent throat diameter mm 11.3 Pressure closed loop target pressure MPa 6.5MPa Pressure tolerance MPa ±0.5MPa

[0078] The following describes the implementation process of the present invention in this example in the order of engine time. All parameters involved in the present invention are as follows: m = 10, θ = 0.5, P min =0MPa, P max =20MPa,ΔP AB_set =1MPa, Δt=1s, M=100, N=5. For the execution results, refer to the attached manual. Figure 3 , Figure 4 and Figure 5 .

[0079] In this example, step 1 and step 2 of the implementation process of the present invention from engine ignition to the end of engine operation are the same, and the difference lies in the following steps, and step 1 and step 2 are:

[0080] Step 1: The two sensor signals are firstly subjected to low-pass filtering and average filtering in turn. The average filtering is preferably: the pressure data is updated every 0.5 ms, and an anti-pulse interference sliding average filtering is performed: the maximum and minimum two numbers of the first m data are discarded in real time to calculate the arithmetic average. The low-pass filtering is preferably y(k)=(1-θ)×y(k-1)+θ×x(k), wherein: x(k) is the current input value, the initial value is 0, y(k-1) is the previous output value, y(k) is the current calculated output value, and θ is the filter coefficient;

[0081] Step 2: Determine the value P of pressure sensor A senserA And the pressure sensor B shows the value P senserB Is it in the valid range [P min , P max ], if it is detected that the indication of a certain pressure sensor is not in the valid range, the pressure sensor is judged to be faulty, and the indication of another sensor is used as the final output, and the faulty sensor is no longer reused until the end of the work, where P min is the estimated minimum pressure of the engine, P max is the estimated maximum pressure of the engine, in this case P min Should not be higher than the engine working environment pressure, P max Should not be lower than the failure pressure of the engine casing structure. min 0MPa, P maxis 20MPa. In this case, the readings of the two sensors are between [0,20MPa];

[0082] The implementation process of steps 3, 4, 5, and 6 of the present invention at 0s to 4.3s after the engine is ignited is:

[0083] Step 3: Determine n senserA and n senserB Is it greater than the fault threshold N, n senserA and n senserB The initial value is 1 and not greater than N;

[0084] Step 4: Determine the difference between the two sensors |P senserA -P senserB |With pressure resolution ΔP AB_set The size of the pressure resolution ΔP AB_set Take 1MPa, and the difference is less than or equal to the preset value ΔP before 4.3s AB_set , take the weighted average P of the two senserAVE is the final output, and enters step 6, the weighted average is P senserAVE =0.5P senserA +0.5P senserA ;

[0085] Step 5: Not executed, skip;

[0086] Step 6: Determine whether the engine has finished working, if so, then finish, otherwise return to step 1.

[0087] The implementation process of steps 3, 4, 5, and 6 of the present invention from 4.3 seconds after engine ignition to the end of engine operation is as follows:

[0088] Step 3: Determine n senserA and n senserB Is it greater than the fault threshold N? Figure 4 , n senserA The value increases after 4.3s and is greater than N (N=5) after 4.8s, then the process goes to step 4 between 4.3s and 4.8s. After 4.8s, it is determined that sensor B is faulty, and the indication of sensor A is used, and sensor B is no longer used until the end of the work.

[0089] Step 4: Determine the difference between the two sensors |P senserA -P senserB |With pressure resolution ΔP AB_set After 4.3s, the difference is greater than the preset value ΔP AB_set (ΔP AB_set =1MPa);

[0090] Step 5: From 4.3s to 6.2s, the engine is in constant pressure operation, and the pressure simulation correction value P is taken. St.simis the third information, namely P(P senserA， P senserB， P sim )=P St.sim , calculate the two sensor indications and the third information pressure simulation correction value P St.sim The sensor with the smaller difference is taken as the final output; n senserA The initial value is 1. When the sensor A is taken every M times, the final output is n senserA Add 1, n senserB The initial value is 1. When the sensor B value is taken every M times, the final output is n senserB Add 1;

[0091] Step 6: Determine whether the engine has finished working, if so, then finish, otherwise return to step 1.

[0092] △P AB_set The value of should not be greater than the allowable error of the engine pressure. In this embodiment, the allowable error of the engine pressure is ±0.5MPa, so △P AB_set Take 1MPa. When the difference between the two sensors is less than 1MPa, the pressure control loop takes the middle value of the two for pressure closed-loop control. Therefore, there will be a certain error between the real pressure (pressure sensor A indication) and the pressure ring adopted value. When the difference between the two sensors is greater than 1MPa, by comparing with the third information pressure simulation correction value P St.sim The difference between the faulty sensor and the normal sensor pressure signal B is successfully identified. Figure 3 From the correct pressure signal (indication of pressure sensor A), it can be seen that the engine pressure fluctuation in this example is successfully controlled within ±0.5MPa.

[0093] Depend on Figure 5 It can be seen that relative to the zero-dimensional interior ballistic model pressure simulation value P sim , pressure simulation correction value P St.sim It is closer to the actual engine interior ballistic performance. This is because there is deposition and ablation in the combustion process of the solid attitude and orbit control engine. There is a certain gap between the pressure simulation value of the zero-dimensional interior ballistic model and the actual pressure value of the engine. If the pressure simulation value of the zero-dimensional interior ballistic model is directly used as the basis for judgment, for this embodiment, misjudgment is bound to occur. The present invention uses the historical measurement value of the sensor to correct the pressure calculation value when the engine works at a constant pressure, and selects the correction value as the third information, which reduces the possibility of misjudgment of faults.

[0094] Example 2

[0095] Sensor A works normally, sensor B is powered off at the 3rd second, and the final indication remains unchanged until the engine stops running.

[0096] The following describes the implementation process of the present invention in this example in the order of engine time. The values ​​of all parameters involved in the present invention are the same as those in Example 1. The implementation results refer to the attached manual. Figure 6 , Figure 7 .

[0097] In this example, steps 1 and 2 of the implementation process of the present invention from engine ignition to the end of engine operation are the same, and the difference lies in the following steps, and steps 1 and 2 are:

[0098] Step 1: The two sensor signals are firstly filtered by low-pass filtering and average filtering in turn. The average filtering is preferably: the pressure data is updated once every period of time, and an anti-pulse interference sliding average filtering is performed: the maximum and minimum two numbers of the first m data are discarded in real time to calculate the arithmetic average. The low-pass filtering is preferably y(k)=(1-θ)×y(k-1)+θ×x(k), where: x(k) is the current input value, the initial value is 0, y(k-1) is the previous output value, y(k) is the current calculated output value, and θ is the filter coefficient;

[0099] Step 2: Determine the value P of pressure sensor A senserA And the pressure sensor B shows the value P senserB Is it in the valid interval [P min , P max ], if it is detected that the indication of a certain pressure sensor is not in the valid range, the pressure sensor is judged to be faulty, and the indication of another sensor is used as the final output, and the faulty sensor is no longer reused until the end of the work, where P min is the estimated minimum pressure of the engine, P max Estimate the maximum pressure for the engine;

[0100] The implementation process of steps 3, 4, 5, and 6 of the present invention at 0s to 6.25s after the engine is ignited is:

[0101] Step 3: Determine n senserA and n senserB Is it greater than the fault threshold N, n senserA and n senserB The initial value is 1 and not greater than N;

[0102] Step 4: Determine the difference between the two sensors |P senserA -P senserB |With pressure resolution ΔP AB_set The size of the pressure resolution ΔP AB_set Take 1MPa, and the difference is less than or equal to the preset value ΔP before 4.3s AB_set , take the weighted average P of the two senserAVE is the final output, and enters step 6, the weighted average is P senserAVE =0.5P senserA+0.5P senserA ;

[0103] Step 5: Not executed, skip;

[0104] Step 6: Determine whether the engine has finished working, if so, then finish, otherwise return to step 1.

[0105] The implementation process of steps 3, 4, 5, and 6 of the present invention from 6.25 seconds after engine ignition to the end of engine operation is as follows:

[0106] Step 3: Determine n senserA and n senserB Is it greater than the fault threshold N? Figure 7 , n senserA The value increases after 6.25s and is greater than N (N=5) after 6.8s, then the process goes to step 4 from 6.25s to 6.8s. After 6.8s, it is determined that sensor B is faulty, and the indication of sensor A is used, and sensor B is no longer used until the operation is completed.

[0107] Step 4: Determine the difference between the two sensors |P senserA -P senserB |With pressure resolution ΔP AB_set After 6.25s, the difference is greater than the preset value ΔP AB_set (ΔP AB_set =1MPa);

[0108] Step 5: The engine is in variable pressure operation, and the pressure simulation differential value ΔP is taken St.sim is the third information, namely P(P senserA， P senserB， P sim )=ΔP St.sim , calculate the difference between the two sensor indications and the third information pressure simulation value ΔP St.sim The sensor with the smaller difference is taken as the final output; n senserA The initial value is 1. When the sensor A is taken every M times, the final output is n senserA Add 1, n senserB The initial value is 1. When the sensor B value is taken every M times, the final output is n senserB Add 1;

[0109] Step 6: Determine whether the engine has finished working, if so, then finish, otherwise return to step 1.

[0110] In this example, the allowable error of the engine pressure is ±0.5MPa, so △P AB_setTake 1MPa. When the difference between the two sensors is less than 1MPa, the pressure control loop takes the middle value of the two for pressure closed-loop control. Therefore, there will be a certain error between the real pressure (indication of pressure sensor A) and the value adopted by the pressure loop. When the difference between the two sensors is greater than 1MPa, by comparing with the third information pressure simulation differential value ΔP St.sim The difference between the two values ​​was eliminated, and the faulty sensor B was eliminated, and the normal sensor pressure signal was successfully identified.

[0111] Figure 6 Step 5 shows that the pressure difference value ΔP is simulated by comparing the two sensor indications with the third information St.sim The result of selecting the pressure output value by the difference. In this example, the pressure ring selects the pressure sensor B indication several times, but not more than M times, so n senserB No increase. M and N represent sensitivity. The M value represents the sensitivity of the weighted average of the dual sensors to the trusted pressure value. The lower the M value, the easier it is for the weighted average of the dual sensors to accept the trusted pressure value. The N value represents the sensitivity of the pressure control loop to permanently adopt a pressure value. The lower the N value, the easier it is for the dual sensor redundant processing method to permanently exclude non-trusted pressure values. Therefore, the M and N values ​​must be selected according to demand. If they are too low, they are easily affected by interference signals and the wrong pressure is selected. If they are too high, it is easy to select faulty pressures multiple times, affecting system stability.

[0112] The above specific implementation modes are used to illustrate the present invention. Any modifications and changes made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine, characterized in that: The method comprises the following steps: S1. Preprocessing First, the two pressure sensor signals are smoothed, and then the boundary condition compliance is judged respectively. When both pressure sensor signals are in compliance, the next step is entered to obtain the final output result. Otherwise, the sensor signal that fails the compliance judgment is excluded and will not be reused until the engine operation is finished. S2. Acquisition of Third Information Solve the zero-dimensional interior ballistic model of the solid attitude and orbit control engine to obtain the pressure simulation value P sim , establish a dual pressure sensor based on the indication P senserA , P senserB And the pressure simulation value P sim The third information calculation model P(P senserA， P senserB， P sim ), solve the model to obtain the third information; S3. Redundancy processing The difference between the two sensor readings is calculated. When the difference is less than or equal to the preset value ΔP AB_set When the weighted average value P of the two is taken senserAVE is the final output.

2. A method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 1, characterized in that: The smoothing of the two sensor signals in S1 specifically includes: subjecting the two sensor signals to low-pass filtering and average filtering in sequence.

3. A method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 2, characterized in that: The mathematical model of the low-pass filter is: y(k1)=(1-θ)×y(k1-1)+θ×x(k1) Where: k1 is the current sensor sampling period, k1-1 is the previous sensor sampling period, x(k1) is the current input initial value, y(k1-1) is the previous output value, y(k1) is the current calculated output value, and θ is the filter coefficient.

4. A method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 2, characterized in that: The average value filtering is as follows: the pressure data is updated once every period of time, and an anti-pulse interference sliding average value filtering is performed, and the arithmetic mean of the first m data is calculated by discarding the maximum and minimum numbers in real time, wherein each period of time is specifically less than or equal to the control period T, and m is greater than 2.

5. The method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 1, characterized in that: The boundary condition compliance judgment described in S1 is specifically as follows: judging the value P indicated by the pressure sensor A senserA And the pressure sensor B shows the value P senserB Is it in the valid interval [P min , P max ], where P min is the estimated minimum pressure of the engine, P max Estimate the maximum pressure for the engine.

6. A method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 1, characterized in that: The zero-dimensional interior ballistic model of the solid attitude and orbit control engine described in S2 is: In the formula, ρ P is the propellant density, A b is the combustion area, a is the burning rate coefficient, n is the pressure index, A t is the total valve pneumatic throat area, V C is the free volume of the combustion chamber, Γ is the function of the specific heat ratio, c * is the characteristic velocity of the propellant, P sim is the pressure simulation value, and t is the time.

7. The method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 1, characterized in that: The third information calculation model P(P senserA， P senserB， P sim ) Specifically: When the engine is working at constant pressure, take the pressure simulation correction value P St.sim is the third information, namely P(P senserA , P senserB， P sim )=P St.sim , the calculation process of the third information calculation model is as follows: A1: The engine ignition time is t=0, and the time of entering the constant pressure working mode is t0. i =t0+i△t (i=0, 1, 2, ...), the weighted average value P senserAVE and pressure simulation value P sim Make a difference and record the value as [t i , t i+1 ] time period, denoted as ΔP AveSim_i , that is, Δp AveSim_i =p sensorAVE (t i )-p sim (t i ); A2: Set the current pressure simulation value P sim The pressure difference ΔP during this period AveSim_i Sum and get the pressure simulation correction value P for this time period St.sim , that is, p St.sim (t) = p sensorAVE (t)+Δp AveSim_i , where t i ≤t<t i+1 ; When the engine is in variable pressure operation, take the pressure simulation differential value ΔP St.sim is the third information, namely P(P senserA , P senserB， P sim )=ΔP St.sim , pressure simulation differential value ΔP St.sim is the pressure simulation value P sim The forward difference of is expressed as: Among them, T is the control period, k2 is the current control period, k2-1 is the previous control period, P sim (k2) is the pressure simulation value of the current control cycle, P sim (k2) is the pressure simulation value of the previous control cycle.

8. The method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 1, characterized in that: The weighted average value P described in S3 senserAVE The calculation formula is: p sensorAVE =(n sensorA p sensorA +n sensorB p sensorB ) / (n sensorA +n sensorB ) Among them, n senserA and n senserB The initial value is 1.

9. The method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 1, characterized in that: The difference between the two sensor readings in S3 is greater than the preset value ΔP AB_set When the sensor A value P is selected according to the third information, senserA Or sensor B indicates value P senserB is the final output, sensor A is effective for n times senserA The initial value is 1. When the sensor A is taken every M times, the final output is n senserA Add 1, sensor B is valid for n times senserB The initial value is 1. When the sensor B value is taken every M times, the final output is n senserB Add 1, when n senserA or senserB If the pressure is greater than the fault threshold N, the other sensor is marked as faulty and the value is no longer reused; AB_set is a configurable parameter, which means that the present invention can distinguish the difference between the dual sensor signals greater than ΔP AB_set The weighted parameter M is a settable parameter. The lower the M value, the easier it is for the weighted average of the dual sensors to accept the trusted pressure value. The fault threshold N is a settable parameter. The lower the N value, the easier it is for the dual sensor redundant processing method to permanently exclude untrusted pressure values. The M and N values ​​must be selected according to demand. If they are too low, they are easily affected by interference signals and the wrong pressure is selected. If they are too high, the fault pressure is easily selected multiple times, affecting the stability of the system.

10. A method for processing redundant information of dual pressure sensors of a solid attitude and orbit control engine according to claim 9, characterized in that: When the difference between the two sensor readings is greater than the pressure resolution ΔP AB_set When the sensor A value P is selected according to the third information, senserA Or sensor B indicates value P senserB The final output is: When the engine is in constant pressure operation, calculate the two sensor values ​​and the third information pressure simulation correction value P St.sim The sensor with the smaller difference is taken as the final output. When the engine is in constant pressure operation, calculate the forward differential value of the two sensors and the third information pressure simulation differential value ΔP St.sim The sensor value with the smaller difference is taken as the final output.

Images