Ignition auxiliary starting method for turboprop engine

Abstract

The invention discloses an ignition auxiliary starting method for a turboprop engine, and aims to actively start an ignition auxiliary device to realize engine excitation ignition, guarantee stable starting of the engine and improve the use efficiency of the engine when an engine ignition exciter is abnormal. The electrical parameters of the ignition exciter and the working temperature of the ignition plug are collected through a built-in detection unit of the auxiliary starting device, the time domain feature vector of each detection value is extracted, and an accurate decision result is obtained by using a data fusion algorithm, so that the accuracy of state detection of the ignition exciter is improved. According to the turboprop engine ignition auxiliary starting method, interaction between systems and control instruction transmission are completed through an electrical interface and a cable, when an engine EEC system receives a state abnormal result, a relay contact switch is driven to start an ignition auxiliary device excitation power switch, and the active control ignition process of an engine control system is completed.

Description

Technical Field

[0001] The invention relates to the field of engines, and in particular to a redundant design method which can assist in realizing the ignition and starting of a turboprop engine when an ignition exciter has a working abnormality. Background Art

[0002] The turboprop engine ignition system is an important starting component in the engine system, including the ignition exciter, high-voltage transmission cable and ignition plug. It is mainly used to release spark energy to burn the mixture of fuel and air input from the main oil manifold, thereby achieving a reliable ignition and starting process of the engine. With the continuous improvement of propeller aircraft's requirements for engine performance and the increasingly harsh ignition and starting environment, the internal circuit components of the ignition exciter are prone to performance failures. Long-term use may cause abnormal spark frequency, insufficient spark energy and insufficient spark duration. In severe cases, the engine cannot ignite effectively, resulting in a failure to start the aircraft, which has an adverse impact on the conduct of flight missions.

[0003] At present, the design of the health status monitoring system and online fault detection method of the engine ignition exciter is very mature. The system monitors the electrical parameters of the internal circuit of the ignition exciter and only feedbacks the fault information to the operator. There is no synchronous redundant design of the ignition device. When the ignition exciter fails, it needs to be disassembled and the fault analysis is performed before the flight mission can continue, which greatly reduces the efficiency of the ignition exciter. In addition, the system lacks the fusion analysis process of the monitoring data. The single data comparison analysis will lead to the wrong judgment of the ignition exciter status, and it is impossible to comprehensively monitor the working status of the turboprop engine ignition exciter. Summary of the invention

[0004] The purpose of the present invention is to provide an ignition auxiliary starting system and method based on multi-detection data fusion decision in order to solve the above problems. The ignition auxiliary device synchronously collects the electrical parameters of the ignition exciter and the working temperature of the ignition plug, extracts the high-dimensional feature vectors of each detection data set, performs feature fusion judgment after feature alignment, and outputs the state decision result of the ignition exciter. The decision result is fed back to the turboprop engine EEC to control the switching of the relay contact switch, power the discharge circuit of the auxiliary starting device, and perform high-voltage auxiliary excitation, thereby realizing fast and stable ignition starting of the engine.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0006] The invention provides an ignition auxiliary starting system for a turboprop engine, comprising an engine electronic control system (EEC), an ignition auxiliary starting device, an airborne power supply, a relay switch, an ignition exciter, a main oil circuit ignition plug, and a sensor module. The ignition auxiliary starting device is installed in parallel with the ignition exciter and connected to the EEC. The sensor module is responsible for measuring and transmitting signals such as temperature, voltage, and current to the ignition auxiliary starting device. The power supply of the ignition auxiliary starting system consists of an internal power supply of the EEC and an external airborne power supply. The internal power distribution network of the EEC outputs a 5V DC voltage to power a main control unit, and the external airborne power supply provides a 28V DC voltage for high-voltage discharge output power supply. The relay switch is connected to the airborne power supply and the EEC respectively. In order to reduce voltage drop loss and noise interference and improve transmission speed, detection signals and communication instructions are transmitted through a high-speed shielded cable.

[0007] As a preferred embodiment of the present invention, the ignition auxiliary starting device includes a signal processing circuit, a main control unit, an energy storage capacitor, a discharge output circuit, a power processing module, a power supply and a communication interface.

[0008] The present invention also provides a turboprop engine ignition auxiliary starting method, comprising the following steps:

[0009] Step S1, the power parameters of the ignition exciter are collected by the front-end high-sensitivity detection circuit and transmitted to the signal processing circuit;

[0010] Step S2, the original measurement signal is processed by low-pass filtering, operational amplification and biasing, and then sent to the single-chip microcomputer chip of the ignition auxiliary starting device for A / D conversion, and the processed signal is fused and decision-made;

[0011] Step S3, the EEC is responsible for decision-making processing of the received feedback, outputting a control instruction to drive the relay to operate, switching the 28V DC power supply to power the discharge output circuit, and completing the excitation process of the ignition auxiliary device.

[0012] As a preferred embodiment of the present invention, Figure 3 As shown, the step S2 performs fusion and decision processing on the processed signal.

[0013] Specifically, it includes the following contents: the entire ignition actuator fault judgment process is divided into the initial state determination of a single detection quantity, data fusion and state diagnosis.

[0014] After the main control unit obtains the target measurement data such as temperature, current, and voltage, it obtains the observation equation of the detection data: A(t, a) = B(t, b) + E(t, e);

[0015] Where A is the value of each sensor from t1 to t nThe n-dimensional measurement vector at time, let A = [a1, a2, a3, ..., a n ] T , B is the period from t1 to t n The actual measurement (true value) vector of n dimensions at time instant, E is the n-dimensional measurement noise vector, which includes external interference noise and internal circuit noise. Let E = [e1, e2, e3, …, e n ] T .

[0016] Since the temperature, current, and voltage sensor data are heterogeneous measurement data and cannot be directly fused, a measurement data isomorphism determination method is proposed here to determine the ignition exciter fault characterization state of each single detection quantity, and implement two threshold determination methods at the same time. The specific implementation of the method is as follows:

[0017] Step S21, based on the spatiotemporal autocorrelation of the sensor, its own data has similar trends at adjacent moments, and a determination threshold Δ1 is set. If the difference in the detection amount at adjacent moments exceeds the threshold, there is a probability of a fault anomaly.

[0018] Step S22, extract the feature vector of the detection data vector and calculate the conditional mean d of each measurement vector element n and variance σ n 2 , and the voltage characteristic vector Y0[d y ,σ1 2 ], current characteristic vector L0[d l ,σ2 2 ], temperature characteristic vector W0[d w ,σ3 2 ]. The characteristic vector V[d0,σ0] of the data set is calibrated based on the sensor data of the ignition actuator in normal working state. 2 ], and compare the extracted feature vector with the threshold interval Δ2. The determination method is as follows:

[0019]

[0020] When any of the above judgment conditions is met, the state result N is set to 1, indicating a preliminary normal state, and N is set to 2, indicating a preliminary abnormal state. Δ1 and Δ2 are determined based on prior knowledge, thereby obtaining the preliminary state feature vectors of each sensor detection quantity after isomorphism.

[0021] Among them, the voltage initial state characteristic vector Y[N y ,σ1 2 ], the initial state characteristic vector of the current L[N l ,σ2 2 ], temperature initial state characteristic vector W[N w ,σ32 ].

[0022] Step S23, combining the Kalman filter algorithm and adopting the iterative fusion method, fuses the state feature vectors and makes an optimal estimate of the judgment result. The calculation method is as follows:

[0023] Firstly, data fusion of current and voltage preliminary state determination vectors is performed.

[0024]

[0025] The state vector after data fusion is then fused with the data of the preliminary state determination vector of temperature.

[0026]

[0027] Thus, the final estimated vector Z[d r1 , σ r1 ]. The optimal state vector is calculated, which has a smaller error than the original data and a state mean closer to the real data.

[0028] The final estimated vector is identified according to the following formula:

[0029]

[0030] Step S24, determine the state decision result of the ignition actuator. If the decision result is that the ignition actuator is in a faulty state, the level signal output to the engine system is output as a high level, the EEC outputs a control instruction to drive the relay contact to close, and the ignition starting auxiliary device discharge circuit power supply is turned on.

[0031] The beneficial effects of the present invention are as follows: the present invention aims at the existing ignition exciter health status monitoring system which only feeds back fault information, lacks the fusion analysis process of monitoring data, cannot accurately determine the fault status of the ignition exciter, and does not carry out the deficiency of redundant design of auxiliary ignition. An ignition auxiliary starting method based on multi-detection data feature fusion decision is proposed, an ignition starting auxiliary device is added to the engine control system, multi-data fusion and decision analysis methods are introduced into the ignition exciter status monitoring process, and based on the decision result, the auxiliary device is actively controlled by the engine EEC to start the ignition. It is a new starting method that improves the redundancy of the engine ignition mechanism and enhances the stability of the ignition excitation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the overall structure of the turboprop engine ignition device including the ignition auxiliary starting device of the present invention;

[0033] Figure 2 It is a schematic diagram of the internal circuit structure of the ignition auxiliary starting device;

[0034] Figure 3 It is a schematic diagram of the multi-detection data fusion and state decision steps;

[0035] Figure 4 This is a schematic diagram of the working process of the ignition auxiliary starting device. DETAILED DESCRIPTION

[0036] In order to make the objectives, technical solutions and advantages of the present invention clearer, the implementation method of the present invention will be described in detail below in conjunction with the accompanying drawings and embodiments, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0037] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0038] Example 1

[0039] As attached Figure 1-2 As shown, embodiment 1 of the present invention provides an ignition auxiliary starting system for a turboprop engine, including an engine electronic control system (EEC), an ignition auxiliary starting device, an airborne power supply, a relay switch, an ignition exciter, a main oil circuit ignition plug, and a sensor module. The ignition auxiliary starting device is installed in parallel with the ignition exciter and connected to the EEC. The sensor module is responsible for measuring and transmitting temperature, voltage, current and other signals to the ignition auxiliary starting device; the power supply of the ignition auxiliary starting system consists of two parts: an internal power supply of the EEC and an external airborne power supply. The internal power distribution network of the EEC outputs a 5V DC voltage to power the main control unit, and the external airborne power supply provides a 28V DC voltage for high-voltage discharge output power supply; the relay switch is connected to the airborne power supply and the EEC respectively. In order to reduce voltage drop loss and noise interference and improve transmission speed, detection signals and communication instructions are transmitted through high-speed shielded cables.

[0040] As a preferred embodiment of the present invention, the ignition auxiliary starting device includes a signal processing circuit, a main control unit, an energy storage capacitor, a discharge output circuit, a power processing module, a power supply and a communication interface.

[0041] Example 2

[0042] As attached Figure 3-4 As shown, embodiment 2 of the present invention provides a turboprop engine ignition auxiliary starting method, comprising the following steps:

[0043] Step S1, the power parameters of the ignition exciter are collected by the front-end high-sensitivity detection circuit and transmitted to the signal processing circuit;

[0044] Step S2, the original measurement signal is processed by low-pass filtering, operational amplification and biasing, and then sent to the single-chip microcomputer chip of the ignition auxiliary starting device for A / D conversion, and the processed signal is fused and decision-made;

[0045] Step S3, the EEC is responsible for decision-making processing of the received feedback, outputting a control instruction to drive the relay to operate, switching the 28V DC power supply to power the discharge output circuit, and completing the excitation process of the ignition auxiliary device.

[0046] As a preferred embodiment of the present invention, Figure 3 As shown, the step S2 performs fusion and decision processing on the processed signal.

[0047] Specifically, it includes the following contents: the entire ignition actuator fault judgment process is divided into the initial state determination of a single detection quantity, data fusion and state diagnosis.

[0048] After the main control unit obtains the target measurement data such as temperature, current, and voltage, it obtains the observation equation of the detection data: A(t, a) = B(t, b) + E(t, e);

[0049] Where A is the value of each sensor from t1 to t n The n-dimensional measurement vector at time, let A = [a1, a2, a3, ..., a n ] T , B is the period from t1 to t n The actual measurement (true value) vector of n dimensions at time instant, E is the n-dimensional measurement noise vector, which includes external interference noise and internal circuit noise. Let E = [e1, e2, e3, …, e n ] T .

[0050] Since the temperature, current, and voltage sensor data are heterogeneous measurement data and cannot be directly fused, a measurement data isomorphism determination method is proposed here to determine the ignition exciter fault characterization state of each single detection quantity, and implement two threshold determination methods at the same time. The specific implementation of the method is as follows:

[0051] Step S21, based on the spatiotemporal autocorrelation of the sensor, its own data has similar trends at adjacent moments, and a determination threshold Δ1 is set. If the difference in the detection amount at adjacent moments exceeds the threshold, there is a probability of a fault anomaly.

[0052] Step S22, extract the feature vector of the detection data vector and calculate the conditional mean d of each measurement vector element n and variance σ n 2 , and the voltage characteristic vector Y0[d y ,σ12 ], current characteristic vector L0[d l ,σ2 2 ], temperature characteristic vector W0[d w ,σ3 2 ]. The characteristic vector V[d0,σ0] of the data set is calibrated based on the sensor data of the ignition actuator in normal working state. 2 ], and compare the extracted feature vector with the threshold interval Δ2. The determination method is as follows:

[0053]

[0054] When any of the above judgment conditions is met, the state result N is set to 1, indicating a preliminary normal state, and N is set to 2, indicating a preliminary abnormal state. Δ1 and Δ2 are determined based on prior knowledge, thereby obtaining the preliminary state feature vectors of each sensor detection quantity after isomorphism.

[0055] Among them, the voltage initial state characteristic vector Y[N y ,σ1 2 ], the initial state characteristic vector of the current L[N l ,σ2 2 ], temperature initial state characteristic vector W[N w ,σ3 2 ].

[0056] Step S23, combining the Kalman filter algorithm and adopting the iterative fusion method, fuses the state feature vectors and makes an optimal estimate of the judgment result. The calculation method is as follows:

[0057] Firstly, data fusion of current and voltage preliminary state determination vectors is performed.

[0058]

[0059] The state vector after data fusion is then fused with the data of the preliminary state determination vector of temperature.

[0060]

[0061] Thus, the final estimated vector Z[d r1 , σ r1 ]. The optimal state vector is calculated, which has a smaller error than the original data and a state mean closer to the real data.

[0062] The final estimated vector is identified according to the following formula:

[0063]

[0064] Step S24, determine the state decision result of the ignition actuator. If the decision result is that the ignition actuator is in a faulty state, the level signal output to the engine system is output as a high level, the EEC outputs a control instruction to drive the relay contact to close, and the ignition starting auxiliary device discharge circuit power supply is turned on.

[0065] In order to verify the accuracy of the multi-data fusion strategy, an abnormal state judgment test of the ignition actuator was carried out. Two ignition start auxiliary devices were connected to the normal ignition actuator end. One device was written with a single measurement data judgment strategy (when one of the sensor data is abnormal, the auxiliary start is started), and the other device was written with a multi-data fusion strategy. A total of 20 engine steady-state ignitions were carried out.

[0066] Table 1 Experimental data

[0067]

[0068] Two ignition start assist devices were connected to the fault ignition actuator. One device was written with a single measurement data judgment strategy (when one of the sensor data is abnormal, the assist start is started), and the other device was written with a multi-data fusion strategy. A total of 20 engine steady-state ignitions were carried out.

[0069] Table 2 Experimental data

[0070]

[0071] The test results show that compared with fault judgment based on a single amount of data, the fault judgment accuracy of the ignition actuator using multi-sensor data fusion is higher, the engine assisted starting process is more stable, the waste of onboard resources is avoided, and the working efficiency of the entire machine is improved.

[0072] Figure 4 This is the specific workflow of the ignition auxiliary starting device. After processing the original detected signal, the main control unit performs fusion analysis on the processed data to obtain the result of the working state determination of the ignition exciter. If the decision result is that the ignition exciter is in a faulty state, the level signal output to the engine system is 1, the EEC outputs a control instruction to drive the relay contacts to close, the power supply of the ignition starting auxiliary device discharge circuit is turned on, and the capacitor is gradually charged until the energy stored in it can ionize the spark gap, and then discharges through the ignition plug to complete the auxiliary ignition excitation. If the decision result is that the ignition exciter is in a normal state, the level signal 0 is returned, the EEC does not make an active control instruction, the relay contacts do not act, and the data fusion process is returned to continue to determine the working state of the ignition exciter.

[0073] Although the present invention is described herein with reference to the illustrative embodiments of the present invention, the above embodiments are only preferred embodiments of the present invention, and the embodiments of the present invention are not limited to the above embodiments. It should be understood that those skilled in the art can design many other modifications and embodiments, which will fall within the scope and spirit of the principles disclosed in this application.

Claims

1. A turboprop engine ignition auxiliary starting system, characterized in that: It includes EEC, ignition auxiliary starting device, airborne power supply, relay switch, ignition exciter, main oil circuit ignition plug, and sensor module. The ignition auxiliary starting device is installed in parallel with the ignition exciter and connected to the EEC. The sensor module is responsible for measuring and transmitting temperature, voltage, current and other signals to the ignition auxiliary starting device. The power supply of the ignition auxiliary starting system consists of two parts: the internal power supply of EEC and the external airborne power supply. The internal power distribution network of EEC outputs 5V DC voltage to power the main control unit, and the external airborne power supply provides 28V DC voltage for high-voltage discharge output power supply. The relay switch is connected to the airborne power supply and EEC respectively. In order to reduce voltage drop loss and noise interference and improve transmission speed, detection signals and communication commands are transmitted through high-speed shielded cables.

2. A turboprop engine ignition auxiliary starting system according to claim 1, characterized in that: The ignition auxiliary starting device includes a signal processing circuit, a main control unit, an energy storage capacitor, a discharge output circuit, a power processing module, a power supply and a communication interface.

3. A turboprop engine ignition auxiliary starting method, characterized in that: The steps include: Step S1, the power parameters of the ignition exciter are collected by the front-end high-sensitivity detection circuit and transmitted to the signal processing circuit; Step S2, the original measurement signal is processed by low-pass filtering, operational amplification and biasing, and then sent to the single-chip microcomputer chip of the ignition auxiliary starting device for A / D conversion, and the initial state of the single detection quantity is determined, and the processed signal is fused and decision-made; Step S3, the EEC is responsible for decision-making processing of the received feedback, outputting a control instruction to drive the relay to operate, switching the 28V DC power supply to power the discharge output circuit, and completing the excitation process of the ignition auxiliary device.

4. A turboprop engine ignition auxiliary starting method according to claim 3, characterized in that: The initial state of the single detection quantity in step S2 is determined, specifically including the following contents: after the main control unit obtains the target measurement data such as temperature, current, voltage, etc., it obtains the observation equation of the detection data: A(t, a)=B(t, b)+E(t, e); Where A is the value of each sensor from t1 to t n The n-dimensional measurement vector at time, let A = [a1, a2, a3, ..., a n ] T , B is the period from t1 to t n The actual measurement (true value) vector of n dimensions at time instant, E is the n-dimensional measurement noise vector, which includes external interference noise and internal circuit noise. Let E = [e1, e2, e3, …, e n ] T .

5. A turboprop engine ignition auxiliary starting method according to claim 4, characterized in that: The processed signals in step S2 are fused and processed for decision making, specifically comprising the following steps: Step S21, based on the spatiotemporal autocorrelation of the sensor, the data of the sensor has similar trends at adjacent moments, and a determination threshold Δ1 is set. If the difference between the detection values ​​at adjacent moments exceeds the threshold, there is a probability of a fault anomaly; Step S22, extract the feature vector of the detection data vector and calculate the conditional mean d of each measurement vector element n and variance σ n 2 , and the voltage characteristic vector Y0[d y ,σ1 2 ], current characteristic vector L0[d l ,σ2 2 ], temperature characteristic vector W0[d w ,σ3 2 ]; According to the sensor data of the ignition actuator in normal working state, the characteristic vector V[d0,σ0 2 ], compare the extracted feature vector with the threshold interval Δ2; Step S23, combining the Kalman filter algorithm and adopting the iterative fusion method to fuse the state feature vectors, make an optimal estimate of the determination result, and determine the final estimation vector after data fusion; Step S24, performing state recognition on the final estimated vector: Determine the state decision result of the ignition actuator.

6. A turboprop engine ignition auxiliary starting method according to claim 5, characterized in that: The determination method for comparing with the threshold interval Δ2 in step S22 is as follows: When any of the above judgment conditions is met, the state result N is set to 1, indicating a preliminary normal state, and N is set to 2, indicating a preliminary abnormal state, where Δ1 and Δ2 are determined based on prior knowledge, thereby obtaining the preliminary state feature vectors of each sensor detection quantity after isomorphism; Among them, the voltage initial state characteristic vector Y[N y ,σ1 2 ], the initial state characteristic vector of the current L[N l ,σ2 2 ], temperature initial state characteristic vector W[N w ,σ3 2 ].

7. A turboprop engine ignition auxiliary starting method according to claim 6, characterized in that: The calculation method of the Kalman filter algorithm in step S22 is as follows: First, the data fusion of the current and voltage preliminary state determination vectors is performed. Then the state vector after data fusion is fused with the data of the preliminary state determination vector of temperature. Thus, the final estimated vector Z[d r1 , σ r1 ].

Images