A performance testing system and method for a dual-mode integrated engine

Abstract

This invention relates to the field of engine testing technology. It discloses a dual-modal integrated engine performance testing system, comprising a hardware testing module, a data acquisition and preprocessing module, a central control and analysis module, a result display and storage module, a predictive analysis submodule, and a preprocessing execution submodule. The invention also discloses a dual-modal integrated engine performance testing method, comprising the following steps: Step 1: System initialization; Step 2: Engine start-up and modal identification; Step 3: Parameter acquisition and preprocessing; Step 4: Data processing, performance evaluation, and predictive analysis; Step 5: Preprocessing decision-making and execution; Step 6: Result display and storage; Step 7: Real-time control. Through the above settings, potential faults can be avoided in advance, reducing the probability of fault occurrence and improving the stability and safety of engine operation.

Description

Technical Field

[0001] This invention relates to the field of engine testing technology, and in particular to a performance testing system and method for a dual-mode integrated engine. Background Technology

[0002] Currently, existing technologies in the field of dual-mode integrated engine performance testing have certain limitations.

[0003] For example, an engine testing system and method disclosed on the China Patent Network (patent publication number CN113670624A) mainly utilizes a crankcase pressure measurement subsystem and a CNC center console. After acquiring piston leakage data, the CNC center console controls an air compressor to create a simulated piston leakage environment within a crankcase fixture. This allows the CNC center console and air compressor to recreate the actual blow-by situation in the crankcase of the engine under test. The crankcase pressure obtained from this simulation can then be considered the actual pressure in the crankcase of the engine under test. This achieves simultaneous monitoring of engine crankcase pressure and piston leakage. However, this testing method suffers from redundant information in the preprocessing stage, making conflict elimination difficult. It lacks effective means to predict potential future engine performance problems, often only addressing issues passively after they occur, rather than reducing the probability of failure through preprocessing.

[0004] Based on this, the present invention proposes a novel performance testing system and method for a dual-mode integrated engine. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a performance testing system and method for a dual-mode integrated engine.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A performance testing system for a dual-mode integrated engine includes a hardware testing module, a data acquisition and preprocessing module, a central control and analysis module, a result display and storage module, a predictive analysis submodule, and a preprocessing execution submodule.

[0008] The hardware detection module includes a group of sensors for temperature, pressure, flow rate, speed, torque, vibration, noise, and modal recognition, used to detect various performance parameters of the engine.

[0009] The data acquisition and preprocessing module is connected to the hardware detection module and is responsible for acquiring and preprocessing various parameters.

[0010] The central control and analysis module is connected to the data acquisition and preprocessing module, the engine control unit, the predictive analysis submodule and the preprocessing execution submodule, respectively. After receiving the preprocessed data, it analyzes and processes it according to the preset strategy and has a built-in dual-mode engine performance evaluation model.

[0011] The results display and storage module is connected to the central control and analysis module and is used to display results and store data;

[0012] The predictive analysis submodule connects the central control and analysis module and the data acquisition and preprocessing module, receives data, and predicts potential performance problems of the engine in the future.

[0013] The preprocessing execution submodule is connected to the central control and analysis module and the engine control unit, receives preprocessing instructions, and performs preprocessing operations on the engine.

[0014] Preferably, the temperature sensor group is distributed in the engine combustion chamber, intake port, and exhaust port; the pressure sensor group detects intake pressure, exhaust pressure, and cylinder pressure; the flow sensor group detects fuel flow and intake flow; and the modal recognition sensor is used to identify the current engine mode.

[0015] Preferably, the data acquisition and preprocessing module comprises two sub-units: a data acquisition unit and a preprocessing module.

[0016] The data acquisition unit is only responsible for establishing real-time communication with the hardware detection module, collecting raw performance parameters of temperature, pressure, and flow rate and modal identification information according to a preset cycle, without performing data processing, and directly transmitting the raw data to the preprocessing module.

[0017] The preprocessing module consists of a basic processing layer and an optimization decision layer, with the following specific structure and functions:

[0018] Basic processing layer: includes a data fusion unit and a state mapping unit. The data fusion unit is used to fuse real-time and historical data to eliminate redundancy and conflicts.

[0019] The state mapping unit integrates real-time performance parameters and prediction results, eliminates redundancy and conflicts, extracts effective information, and extracts the implicit deviation state between real-time monitoring data and the benchmark map through multi-dimensional difference analysis based on the ideal state map of the engine. The decision generation unit maps these deviation values ​​into the adjustment amount of specific parameters through "inverse model transformation".

[0020] The optimized decision-making layer includes a decision generation unit, an evaluation unit, a fluctuation adaptive unit, and an effect simulation unit.

[0021] The evaluation unit constructs a two-dimensional evaluation model for candidate adjustment values, which quantifies potential benefits horizontally and evaluates execution risks vertically. It calculates the benefit / risk ratio through equilibrium game analysis and assigns execution priority weights to ensure that resources are tilted toward high-cost-performance operations.

[0022] The fluctuation adaptive unit relies on a historical fluctuation feature library and combines it with real-time parameter fluctuation trends. It generates dynamic compensation factors through advanced predictive analysis, and corrects the target parameter adjustment values ​​in real time, thereby eliminating uncertainties in the execution process and ensuring that the deviation between the actual effect and the expected effect is controlled within a preset range.

[0023] Preferably, the information extracted by the data fusion unit assists the central control and analysis module in judging performance trends; the preprocessing instructions of the decision generation unit can reduce the probability of failure; the evaluation results of the effect simulation unit provide data support for system upgrades, simulate adjustment effects, and feed back to the central control and analysis module.

[0024] Preferably, the central control and analysis module can comprehensively evaluate the engine's power, economy, emissions performance and reliability based on different modal parameters, monitor the dynamic response of mode switching and analyze its smoothness and coordination, and send commands to the engine control unit to achieve real-time regulation.

[0025] Preferably, the predictive analysis submodule includes a data integration unit, a model training and invocation unit, and a result output unit;

[0026] The data integration unit collects historical and real-time detection data and operating condition information, and forms a unified dataset after cleaning and standardization.

[0027] The model training and invocation unit has a built-in prediction model based on big data and machine learning. It is trained and optimized using historical data and invokes the model for analysis and computation during prediction.

[0028] The result output unit organizes and formats the prediction results, generates a prediction report, and sends it to the central control and analysis module.

[0029] Preferably, the preprocessing operations of the preprocessing execution submodule include adjusting the fuel injection advance angle, changing the intake air volume, and optimizing the lubrication system operating parameters.

[0030] Preferably, a performance testing method for a dual-mode integrated engine based on the above-described dual-mode integrated engine performance testing system includes the following steps:

[0031] Step 1: System initialization, ensuring all modules are working properly, setting detection and prediction strategies and preprocessing trigger conditions and parameters;

[0032] Step 2: The engine starts, and the modal recognition sensor determines the current mode and sends information;

[0033] Step 3: The data acquisition and preprocessing module collects parameters and modal information periodically, preprocesses them, and sends them to the relevant modules;

[0034] Step 4: The central control and analysis module calls the corresponding model to evaluate performance, monitor parameter thresholds, and analyze mode switching; the predictive analysis submodule calls the model to predict performance issues and sends the results.

[0035] Step 5: When the prediction result meets the triggering conditions, the central control and analysis module generates a preprocessing instruction, which is executed by the preprocessing execution submodule and feedback is provided.

[0036] Step Six: The results display and storage module displays and stores the relevant results and data;

[0037] Step 7: The central control and analysis module can send control commands to adjust the engine in real time. After adjustment, repeat steps 3 to 6.

[0038] The present invention has the following beneficial effects:

[0039] 1. This invention, by setting up a hardware detection module, includes a sensor group for temperature, pressure, flow rate, speed, torque, vibration, noise, and modal recognition, which can comprehensively detect various performance parameters of the engine. The sensor group is distributed in the engine combustion chamber and air intake, which can accurately acquire parameter information from different parts. Combined with the dual-modal engine performance evaluation model built into the central control and analysis module, the engine's power, economy, emissions performance, and reliability can be comprehensively evaluated based on different modal parameters. At the same time, the dynamic response of modal switching is monitored and the stability and coordination are analyzed, so as to achieve a comprehensive and accurate evaluation of engine performance.

[0040] 2. This invention includes a predictive analysis submodule, which comprises a data integration unit, a model training and calling unit, and a result output unit. It can collect historical and real-time detection data and operating condition information, and predict potential performance problems of the engine in the future through a predictive model based on big data and machine learning. The preprocessing execution submodule can receive preprocessing instructions from the central control and analysis module to perform preprocessing operations on the engine in advance, such as adjusting the fuel injection advance angle, thereby avoiding potential faults in advance, reducing the probability of fault occurrence, and improving the stability and safety of engine operation.

[0041] 3. The data acquisition and preprocessing module of this invention can preprocess the acquired parameters. The state mapping unit in the preprocessing module can extract effective information and latent deviation states. The evaluation unit can perform dual-dimensional evaluation of the adjustment value and assign execution priority weights. The fluctuation adaptive unit can generate dynamic compensation factors to correct the adjustment value, ensuring that the actual effect and the expected deviation are within the preset range. At the same time, the central control and analysis module can realize real-time control of the engine. Combined with the result display and storage module for data storage, it provides reliable data support for subsequent system upgrades and engine performance optimization. Attached Figure Description

[0042] Figure 1 This is a flowchart of the method for publishing the present invention;

[0043] Figure 2 This is a graph showing the noise detection data of the present invention;

[0044] Figure 3 This is a diagram showing the vibration amplitude under normal conditions detected by the present invention;

[0045] Figure 4 This is a diagram showing the vibration amplitude under fan blade wear conditions as described in this invention. Detailed Implementation

[0046] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0047] A performance testing system for a dual-mode integrated engine includes a hardware testing module, a data acquisition and preprocessing module, a central control and analysis module, a result display and storage module, a predictive analysis submodule, and a preprocessing execution submodule.

[0048] The hardware testing module is used to detect various performance parameters of the dual-modal integrated engine, including a temperature sensor group, a pressure sensor group, a flow sensor group, a speed sensor, a torque sensor, a vibration sensor, a noise sensor, and a modal recognition sensor. In practical applications, the temperature sensor group is distributed in the engine's combustion chamber, intake port, and exhaust port, capable of measuring temperature changes in different parts; the pressure sensor group is used to detect engine intake pressure, exhaust pressure, and cylinder pressure; the flow sensor group is used to detect fuel flow and intake air flow; and the modal recognition sensor is used to accurately identify the engine's current mode, providing a basis for subsequent performance testing and analysis.

[0049] The data acquisition and preprocessing module is connected to the hardware detection module and is used to acquire various parameters detected by the hardware detection module and preprocess the acquired data. The data acquisition and preprocessing module consists of two sub-units: a data acquisition unit and a preprocessing module. The data acquisition unit is only responsible for establishing real-time communication with the hardware detection module, acquiring raw performance parameters of temperature, pressure, and flow rate, as well as modal identification information, according to a preset cycle. It does not perform data processing and directly transmits the raw data to the preprocessing module.

[0050] The data acquisition and preprocessing module comprises two primary sub-units: a data acquisition unit and a preprocessing module. The preprocessing module further includes a basic processing layer and an optimization decision layer. The data acquisition unit establishes real-time communication only with the hardware detection module, collecting raw performance parameters and modal information at preset intervals without performing data processing, directly transmitting the raw data to the basic processing layer of the preprocessing module. The basic processing layer includes a data fusion unit and a state mapping unit. The data fusion unit fuses real-time and historical data to eliminate redundancy and conflicts; the state mapping unit integrates real-time performance parameters and prediction results, eliminates redundancy and conflicts, extracts effective information, and, based on the engine's ideal state map, extracts the implicit deviation states between the real-time monitoring data and the benchmark map using a multi-dimensional difference algorithm. The decision generation unit maps these deviation values ​​into specific parameter adjustment amounts through "inverse model transformation."

[0051] The optimized decision-making layer comprises a decision generation unit, an evaluation unit, a fluctuation adaptation unit, and an effect simulation unit. The evaluation unit constructs a two-dimensional evaluation model for candidate adjustment values, quantifying potential benefits horizontally and assessing execution risks vertically. It calculates the benefit / risk ratio using an equilibrium game algorithm and assigns execution priority weights to ensure resources are allocated to cost-effective operations. The fluctuation adaptation unit relies on a historical fluctuation feature library and combines it with real-time parameter fluctuation trends. Through an advanced prediction model, it generates dynamic compensation factors to correct the target parameter adjustment values ​​in real time, mitigating uncertainties during execution and ensuring that the deviation between actual results and expectations is controlled within a preset range.

[0052] The information extracted by the data fusion unit assists the central control and analysis module in judging performance trends; the preprocessing instructions of the decision generation unit can reduce the probability of failure; the evaluation results of the effect simulation unit provide data support for system upgrades, simulate adjustment effects, and feed back to the central control and analysis module.

[0053] The central control and analysis module is connected to the data acquisition and preprocessing module, the engine control unit, the predictive analysis submodule, and the preprocessing execution submodule. It receives preprocessed data, analyzes and processes the data according to a preset detection strategy, and incorporates a dual-mode engine performance evaluation model. The central control and analysis module can comprehensively evaluate engine power, economy, emissions performance, and reliability based on different modal parameters, emphasizing corresponding performance indicators under different modes. Simultaneously, it monitors the dynamic response of mode switching and analyzes its smoothness and coordination. When the engine undergoes mode switching, it tracks changes in relevant parameters in real time, determines whether the switching process is smooth, and can send commands to the engine control unit for real-time control.

[0054] The predictive analytics submodule connects to the central control and analysis module and the data acquisition and preprocessing module to receive relevant data and predict potential future engine performance issues. The predictive analytics submodule includes a data integration unit, a model training and invocation unit, and a result output unit. The data integration unit collects historical detection data, real-time detection data, and engine operating condition information. It cleans the data using a statistical outlier detection algorithm and standardizes it using a standardization algorithm to create a unified dataset, providing high-quality input data for the predictive model. The model training and invocation unit incorporates a predictive model based on big data and machine learning. It uses a distributed computing framework to build the big data-based predictive model, employs a gradient boosting tree and recurrent neural network algorithm to build the machine learning predictive model, and uses cross-validation algorithms for model optimization. It continuously trains and optimizes the model using historical datasets to enable it to predict engine performance issues. During prediction, it calls the well-trained model to analyze and process the integrated data. The result output unit organizes and formats the prediction results obtained from the model calculations. It uses a rule-based reasoning algorithm to organize the prediction results and a template matching algorithm for formatting, generating a clear and easy-to-understand prediction report which is then sent to the central control and analysis module.

[0055] The preprocessing execution submodule connects to the central control and analysis module and the engine control unit. It receives preprocessing commands and performs preprocessing operations on the engine in advance. The preprocessing operations performed by the preprocessing execution submodule include adjusting the fuel injection advance angle, changing the intake air volume, and optimizing the lubrication system operating parameters. When it receives preprocessing commands from the central control and analysis module, it can execute the corresponding operations to optimize the engine's operating state and reduce the possibility of malfunctions.

[0056] A performance testing method for a dual-modal integrated engine based on the aforementioned performance testing system includes the following steps:

[0057] Step 1: System Initialization

[0058] Start the performance testing system and initialize the hardware testing module, data acquisition and preprocessing module, central control and analysis module, result display and storage module, predictive analysis submodule, and preprocessing execution submodule to ensure that each module works properly. At the same time, set the threshold range of testing parameters, testing cycle, prediction time window, and other testing and prediction strategies, as well as the trigger conditions and execution parameters of preprocessing operations.

[0059] Step 2: Engine Start-up and Modal Recognition

[0060] When the dual-modal integrated engine is started, the modal recognition sensor in the hardware detection module detects the engine's operating status in real time, determines the current mode of the engine, and sends the modal information to the data acquisition and preprocessing module.

[0061] Step 3: Parameter Acquisition and Preprocessing

[0062] The data acquisition and preprocessing module acquires various performance parameters and modal information detected by the hardware detection module according to the set detection cycle. Then, it performs preprocessing operations such as filtering, noise reduction, and data format conversion on the acquired data, and then sends the preprocessed data to the central control and analysis module and the predictive analysis submodule.

[0063] Step 4: Data Processing, Performance Evaluation, and Predictive Analysis

[0064] The central control and analysis module receives preprocessed data, calls the corresponding performance evaluation model based on the engine's current mode, and evaluates various aspects of the engine's performance. During the evaluation process, it monitors in real time whether the engine parameters are within the preset threshold range. If abnormal parameters occur, it issues an early warning signal in a timely manner. At the same time, when the engine switches modes, the central control and analysis module focuses on analyzing the dynamic response parameters during the mode switching process to evaluate the smoothness and coordination of the mode switching. The predictive analysis submodule receives preprocessed data and real-time analysis data from the central control and analysis module, calls the predictive model to predict potential future performance problems of the engine, and sends the prediction results to the central control and analysis module.

[0065] Step 5: Pre-processing decision and execution

[0066] The central control and analysis module receives the prediction results from the predictive analysis submodule. When the prediction results meet the preset preprocessing trigger conditions, it generates corresponding preprocessing instructions based on the predicted problem type and severity, and sends them to the preprocessing execution submodule. After receiving the preprocessing instructions, the preprocessing execution submodule performs preprocessing operations on the engine according to the instructions and feeds back the execution status to the central control and analysis module.

[0067] Step Six: Results Display and Storage

[0068] The results display and storage module receives analysis results, evaluation reports, early warning signals, prediction results, and the execution status of the preprocessing execution submodule sent by the central control and analysis module. It displays these results in an intuitive way, allowing operators to understand the engine's performance status in real time. At the same time, it stores the detection data, analysis results, prediction results, and preprocessing execution records, providing data support for subsequent data analysis, system optimization, and troubleshooting.

[0069] Step 7: Real-time Adjustment

[0070] If the central control and analysis module finds room for optimization in engine performance during the performance evaluation process, or if further adjustments are needed after preprocessing, it can send control commands to the engine control unit based on the analysis results to adjust the engine's operating parameters in real time to improve engine performance. After adjustment, steps three to six are repeated to re-detect, evaluate, predict, and record the engine performance, forming a closed-loop detection and control process to ensure that the engine is always in good operating condition.

[0071] Based on the specific implementation method described above, the system and testing method are applied to the performance testing of dual-mode integrated engines in the aerospace field.

[0072] Application Case 1: The performance testing system of this invention is used to test the dual-mode integrated engine in the aviation field. The temperature sensor group in the hardware testing module uses high-temperature resistant thermocouple sensors, which are distributed in key parts such as the engine combustion chamber and turbine blades. They can withstand a temperature range of -50℃ to 1200℃ with an accuracy of ±0.3℃. The pressure sensor group uses piezoelectric pressure sensors to detect intake pressure, exhaust pressure and combustion chamber pressure. The measurement range is 0 to 20MPa with an accuracy of ±0.1%FS.

[0073] The testing process is as follows:

[0074] Step 1: During system initialization, the detection period is set to 0.05s, the prediction time window is 4 hours, and the preprocessing trigger condition is that the probability of performance problems occurring exceeds 50%.

[0075] Step 2: After the engine starts, the modal recognition sensor accurately identifies the engine's current high-altitude, high-speed mode by collecting parameters such as engine speed (0-20000 r / min) and thrust.

[0076] The formula for the modality recognition algorithm is:

[0077]

[0078] In the formula, Modal identifiers (1 indicates high-altitude, high-speed mode; -1 indicates low-altitude, low-speed mode). The current rotational speed, The critical speed value is 15000 r / min. For current thrust, The critical thrust value is 80kN. It is a symbolic function.

[0079] Step 3: The data acquisition and preprocessing module acquires various parameters at a cycle of 0.05s, such as combustion chamber temperature of 900℃ and intake pressure of 1.2MPa, and performs filtering and noise reduction processing.

[0080] The filtering process uses the Kalman filter algorithm, and the formula is:

[0081]

[0082]

[0083] In the formula for The filtered value at time step for Predicted value at any time For Kalman gain, for Time measurement value, For the observation matrix, To predict the error covariance, To measure the noise covariance, It is an identity matrix.

[0084] Step 4: The central control and analysis module calls the performance evaluation model corresponding to the high-altitude, high-speed mode. The evaluation shows that the engine's power performance meets the requirements, but the vibration parameters have slight fluctuations. The formula for power performance evaluation (thrust coefficient) is:

[0085] In the formula, is the thrust coefficient; F is the actual thrust; air density; This refers to the intake speed; This represents the engine's air intake cross-sectional area.

[0086] The predictive analysis submodule combines historical data to predict the probability of increased vibration. The formula for this prediction algorithm is as follows:

[0087] In the formula, This represents the probability of increased vibration. This is the current effective value of vibration velocity. This represents the effective value of the initial vibration velocity. This is the current dominant vibration frequency. This is the initial dominant vibration frequency; Weighting coefficients , For correction factor, .

[0088] Step 5: The central control and analysis module generates preprocessing instructions, and the preprocessing execution submodule controls the engine's damping system to adjust parameters.

[0089] The formula for adjusting the damping parameters is:

[0090] In the formula, For the new damping stiffness, For the original damping stiffness, This represents the probability of increased vibration.

[0091] Step Six: The results display and storage module displays the status of various parts of the engine and the preprocessing effect in the form of a 3D model, and stores the relevant data.

[0092] Step 7: After preprocessing, subsequent testing showed that the vibration parameters had stabilized and no further adjustment was needed.

[0093] Application Case 2: Passenger Aircraft Engine Performance Testing in High-Altitude and Low-Temperature Environments;

[0094] In the low-temperature environment at an altitude of 10,000 meters, the dual-mode integrated engine, as the core of the passenger aircraft's power, must simultaneously meet the requirements of high thrust output and low fuel consumption. After the performance testing system is started, it first enters the system initialization phase. In response to the extreme low-temperature environment at high altitudes, the lower limit of the detection threshold of the temperature sensor group is adjusted to -60°C, and the detection cycle is shortened to 0.5 seconds to ensure real-time capture of sudden temperature changes.

[0095] After engine startup, the modal recognition sensor quickly identifies that it is in "high-altitude cruise mode," in which the engine operates at a high bypass ratio. In the hardware detection module, temperature sensors distributed throughout the combustion chamber monitor combustion temperature in real time, the pressure sensor group focuses on detecting intake pressure and in-cylinder pressure, and the flow sensor group accurately records fuel flow and lean intake volume. After data acquisition, the data acquisition and preprocessing module uses a moving average filtering algorithm to eliminate vibration interference caused by high-altitude turbulence, as shown in the following formula:

[0096]

[0097] in, for The vibration value after time smoothing For the past 5 sampling points The original vibration data, sampling interval The processed data is then synchronously sent to the central control and analysis module.

[0098] The central control and analysis module invoked the "high-altitude performance evaluation model" and found that the cylinder pressure was slightly lower than the standard value by 1.2%. Combined with the minute high-frequency vibrations detected by the vibration sensor, it was determined that the low temperature was causing a decrease in fuel atomization efficiency. Simultaneously, the predictive analysis submodule integrated historical data from nearly 300 high-altitude flights and used a gradient boosting tree algorithm to predict changes in fuel consumption rate over the next 10 minutes.

[0099]

[0100] in, The increase in fuel consumption rate over the next 10 minutes represents the prediction made by the predictive analysis submodule. "Change in engine fuel consumption rate over the next 10 minutes from this point onwards," with positive values ​​indicating an increase in fuel consumption rate and negative values ​​indicating a decrease.

[0101] The number of decision trees in the gradient boosting tree is a hyperparameter determined after training and optimization based on historical high-altitude flight data from 300 flights. For the first The weight coefficients of each decision tree. For the first The output value of each decision tree.

[0102] Where the input feature vector (In-cylinder pressure, combustion temperature, vibration value) The prediction results show that fuel consumption rate may increase by 0.8%. The central control and analysis module then triggers the preprocessing execution submodule to calculate the adjustment amount according to the fuel injection advance angle adjustment formula:

[0103]

[0104] Substitute the cylinder pressure deviation Standard value, proportionality coefficient Integral coefficient The fuel injection advance angle is obtained. The adjustment command is to advance the temperature by 2°. Simultaneously, the preheating power is increased using the intake preheating power adjustment formula:

[0105]

[0106] Reference power Temperature coefficient Target temperature Actual temperature The preheating power was calculated. Adjusted to 6.5kW.

[0107] The results display module presents the parameter changes before and after the adjustment in the form of dynamic curves, while the storage module records the data of the entire process from detection to regulation, including the mode switching response time and the fuel efficiency improvement value after preprocessing.

[0108] Application Case 3: When civilian sightseeing seaplanes perform low-altitude sightseeing missions in coastal areas (salt spray concentration 8-12 g / m³), both flight stability and fuel economy must be considered. During the initialization of the performance testing system, corrosion warning parameters and noise control parameters are incorporated into the testing strategy for moderate salt spray environments along the coast. The testing cycle is set to 1 second, and the preprocessing trigger condition is "salt spray concentration exceeding 10 g / m³ and engine running continuously for 30 minutes".

[0109] After the engine starts, the modal recognition sensor identifies it as being in "low-altitude cruise mode" (flight altitude 500-800 meters, speed 200-250 km / h). In the hardware detection module, a noise sensor strictly monitors engine operating noise, a temperature sensor group monitors exhaust temperature, and a pressure sensor group monitors fuel line pressure.

[0110] The data acquisition and preprocessing module performs min-max standardization on the acquired parameters to eliminate fluctuation interference caused by coastal gusts, as shown in the following formula:

[0111] The original value of salt spray concentration value range .

[0112] Standardized for use in subsequent analysis.

[0113] The central control and analysis module invoked the "Sightseeing Flight Performance Evaluation Model," which assessed that the exhaust temperature was 4°C higher than the standard value. Combined with salt spray concentration data, this indicated a small amount of salt crystals adhering to the exhaust outlet. The predictive analysis submodule used a gradient boosting tree algorithm to predict noise changes over the next 20 minutes.

[0114]

[0115] Input features (Salt spray concentration, exhaust temperature, fuel pressure) The prediction results show that the noise may increase by 2 decibels.

[0116] The central control and analysis module triggers the preprocessing execution submodule, which fine-tunes the lubrication system operating parameters according to the lubricating oil flow adjustment formula:

[0117]

[0118] The reference lubricating oil flow rate Noise figure Substituting the predicted noise increment of 2dB, it is determined that the lubricating oil flow rate needs to be increased by 10% (adjusted to 2.2L / min) to reduce mechanical friction noise. Simultaneously, the anti-crystallization procedure is activated according to the exhaust flow rate adjustment formula.

[0119]

[0120] Substitute the salt spray coefficient Actual salt spray concentration The target flow rate was found to be 1.15 times the baseline flow rate (a short-term increase of 15% in exhaust flow rate).

[0121] The results display module shows the effects of noise, temperature, and other parameters on regulation in a real-time dashboard format. The storage module records the smoothness data of mode switching during this test and the changes in fuel consumption after preprocessing. Subsequently, the system continues to collect and analyze parameters according to the test cycle to ensure that the engine remains in a stable and efficient operating state throughout the entire coastal sightseeing flight.

[0122] Specific Figure 2 To detect noise data, the trend of normal data without faults and high-frequency noise fault data in terms of speed parameters over time (0-60s) is shown.

[0123] The normal operating speed data without faults is basically stable within the range of 1480-1520 RPM, with small fluctuations, reflecting the stability of the engine speed under normal operating conditions.

[0124] The speed of the high-frequency noise fault data is also in the range of 1480-1520, but the fluctuation frequency of the curve is higher. This reflects that when the engine has a high-frequency noise fault, the speed parameter is disturbed and shows more frequent small fluctuations, which can be used as one of the bases for judging whether the engine has a high-frequency noise fault.

[0125] Figure 3 This is a diagram showing the vibration amplitude of the fan blades under fault-free conditions, recording the changes in the vibration value of the fan blades over one minute.

[0126] Under fault-free conditions, the vibration amplitude of the fan blades is generally small, fluctuating mainly within the range of -0.6 to 0.6, and mostly within the range of -0.3 to 0.3. The vibration is stable and low in amplitude. This curve reflects the vibration characteristics of the engine fan blades under normal operating conditions and can serve as a benchmark vibration spectrum for fault-free fan blades, providing a reference for determining whether any abnormalities have occurred in the fan blades.

[0127] Figure 4 The diagram shows the vibration amplitude under blade wear conditions, illustrating the change in vibration value over one minute during blade wear.

[0128] Compared to the fault-free state, the vibration amplitude during blade wear increases significantly, with the maximum vibration value reaching 1.2 and the minimum decreasing to -0.8. The fluctuation range is significantly expanded, and the overall vibration intensity is enhanced. This curve intuitively demonstrates the impact of blade wear on vibration parameters. By comparing it with the vibration amplitude diagram under fault-free conditions, the presence of wear problems in the blades can be quickly identified, providing crucial information for engine performance evaluation and fault diagnosis.

[0129] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A performance detection system for a dual-mode integrated engine, characterized in that, The system comprises a hardware detection module, a data acquisition and preprocessing module, a central control and analysis module, a result display and storage module, a predictive analysis submodule and a preprocessing execution submodule. The hardware detection module comprises temperature, pressure, flow rate, rotation speed, torque, vibration, noise and modal recognition sensor groups for detecting various performance parameters of the engine. The data acquisition and preprocessing module is connected to the hardware detection module and is responsible for collecting and preprocessing various parameters. The central control and analysis module is connected to the data acquisition and preprocessing module, the engine control unit, the predictive analysis submodule and the preprocessing execution submodule. The result display and storage module is connected to the central control and analysis module for displaying results and storing data. The predictive analysis submodule is connected to the central control and analysis module and the data acquisition and preprocessing module for receiving data and predicting possible future performance problems of the engine. The preprocessing execution submodule is connected to the central control and analysis module and the engine control unit for receiving preprocessing instructions and performing preprocessing operations on the engine in advance. The data acquisition and preprocessing module comprises a data acquisition unit and a preprocessing module. The data acquisition unit is responsible for establishing real-time communication with the hardware detection module, collecting temperature, pressure, flow rate and modal recognition information at a preset period, and transmitting the raw data to the preprocessing module. The preprocessing module comprises a basic processing layer and an optimized decision-making layer. The basic processing layer comprises a data fusion unit and a state mapping unit. The state mapping unit integrates real-time performance parameters and prediction results, eliminates redundancies and conflicts, extracts effective information, and extracts implicit deviation states of real-time monitoring data and benchmark graphs through multi-dimensional difference analysis based on the ideal state graph of the engine. The optimized decision-making layer comprises a decision-making unit, an evaluation unit, a fluctuation adaptive unit and an effect simulation unit. The evaluation unit constructs a two-dimensional evaluation model for candidate adjustment values, quantizes potential benefits horizontally and evaluates execution risks vertically, calculates the benefit / risk ratio through balanced game analysis, and assigns execution priority weights to ensure that resources are tilted towards high-performance operations. The fluctuation adaptive unit generates dynamic compensation factors through advanced prediction analysis based on historical fluctuation characteristics and real-time parameter fluctuation trends, and adjusts target parameter adjustment values in real time to eliminate uncertainties in the execution process and ensure that the actual effect and expected deviation are controlled within a preset range.

2. The performance detection system of a dual-mode integrated engine according to claim 1, characterized in that, Temperature sensors are distributed in the combustion chamber, air inlet and exhaust outlet of the engine; pressure sensors detect intake pressure, exhaust pressure and cylinder pressure; flow rate sensors detect fuel flow and intake flow; and modal recognition sensors are used to identify the current mode of the engine.

3. The performance detection system of a dual-mode integrated engine according to claim 1, characterized in that, The information extracted by the data fusion unit assists the central control and analysis module in determining performance trends; The pre-processing instructions of the decision generation unit can reduce the probability of failure; the evaluation results of the effect simulation unit provide data support for system upgrade, simulate adjustment effects, and feedback to the central control and analysis module.

4. The performance detection system of a dual-mode integrated engine according to claim 1, wherein, The central control and analysis module can comprehensively evaluate the engine's power, economy, emission performance, and reliability according to different modal parameters, monitor modal switching dynamic response, and analyze stability and coordination, and can also send instructions to the engine control unit to achieve real-time regulation and control.

5. The performance detection system of a dual-mode integrated engine according to claim 1, wherein, The predictive analysis submodule includes a data integration unit, a model training and calling unit, and a result output unit; The data integration unit collects historical and real-time detection data and operating condition information, and forms a unified data set after cleaning and standardization processing; The model training and calling unit has built-in prediction models based on big data and machine learning, which are trained and optimized through historical data, and are called for analysis and calculation during prediction; The result output unit formats the prediction results, generates a prediction report, and sends it to the central control and analysis module.

6. The performance detection system of a dual-mode integrated engine according to claim 1, wherein The pre-processing operations of the pre-processing execution submodule include adjusting the fuel injection advance angle, changing the intake air volume, and optimizing the lubrication system operating parameters.

7. A method of performance detection of a system of performance detection of a dual- mode integrated engine according to any one of claims 1 to 6, characterized in that, The method comprises the following steps: Step one: system initialization, ensure that each module works normally, set detection and prediction strategies and pre-processing trigger conditions and parameters; Step two: engine start, modal recognition sensor determines the current mode and sends information; Step three: the data acquisition and pre-processing module acquires parameters and modal information according to the cycle, and sends them to the relevant modules after pre-processing; Step four: the central control and analysis module calls the corresponding model to evaluate the performance, monitors the parameter threshold, and analyzes the modal switching situation; the predictive analysis submodule calls the model to predict performance problems and sends the results; Step five: the central control and analysis module generates pre-processing instructions when the prediction results meet the trigger conditions, which are executed by the pre-processing execution submodule and fed back; Step six: the result display and storage module displays and stores relevant results and data; Step seven: the central control and analysis module can send control instructions to regulate the engine in real time, and repeat steps three to six after regulation.

Images