Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

40 results about "Compressor stall" patented technology

A compressor stall is a local disruption of the airflow in the compressor of a gas turbine or turbocharger. A stall that results in the complete disruption of the airflow through the compressor is referred to as a compressor surge. The severity of the phenomenon ranges from a momentary power drop barely registered by the engine instruments to a complete loss of compression in case of a surge, requiring adjustments in the fuel flow to recover normal operation.

Axial flow compressor stall surge prediction method based on deep learning

ActiveCN112001128AImprove the performance of active controlImprove forecast accuracyGeometric CADSustainable transportationAviationData set
The invention discloses an axial flow compressor stall surge prediction method based on deep learning, and belongs to the technical field of aero-engine modeling and simulation. The method comprises the steps of: firstly, preprocessing aircraft engine surge data, and dividing a test data set and a training data set from experimental data; secondly, sequentially constructing an LR branch network module, a WaveNet branch network module and an LR-WaveNet prediction model; and finally, performing real-time prediction on the test data: firstly, preprocessing the test set data in the same way, and adjusting the data dimension according to the input requirement of the LR-WaveNet prediction model; according to the time sequence, adopting an LR-WaveNet prediction model to give the surge predictionprobability of each sample; and adopting an LR-WaveNet prediction model to give the surge probability of data with noise points along with time, and testing the anti-interference performance of the model. According to the method, the time domain statistical characteristics and the change trend are integrated, the prediction precision is improved, and certain anti-interference performance is achieved; and the active control performance of the engine can be improved, and certain universality is achieved.
Owner:DALIAN UNIV OF TECH

Axial flow compressor stall surge prediction method based on deep learning

InactiveCN111737910AImprove the performance of active controlImprove forecast accuracyGeometric CADSustainable transportationAviationData set
The invention discloses an axial flow compressor stall surge prediction method based on deep learning, and belongs to the technical field of aero-engine modeling and simulation. The method comprises:firstly, preprocessing aircraft engine surge data, and dividing a test data set and a training data set from experimental data; secondly, sequentially constructing an LR branch network module, a WaveNet branch network module and an LR-WaveNet prediction model; and finally, performing real-time prediction on the test data: firstly, preprocessing the test set data in the same way, and adjusting thedata dimension according to the input requirement of the LR-WaveNet prediction model; according to the time sequence, by adopting an LR-WaveNet prediction model, giving the surge prediction probability of each sample; and by adopting an LR-WaveNet prediction mode, giving the surge probability of data with noise points along with time, and testng the anti-interference performance of the model. According to the method, the time domain statistical characteristics and the change trend are integrated, the prediction precision is improved, and certain anti-interference performance is achieved; and the active control performance of the engine can be improved, and certain universality is achieved.
Owner:DALIAN UNIV OF TECH

Prediction method of multistage axial-flow compressor stall margin on the basis of characteristic value theory

The invention relates to a prediction method of a multistage axial-flow compressor stall margin. The prediction method comprises the following steps: according to a situation of the rotating stall inception of an axial-flow compressor, adopting a small perturbation theory to establish a three-dimensional compressible Euler equation which describes a flow field; applying harmonic decomposition and a dispersion relation to establish a boundary condition on the hub and the leaf opex of each drainage basin; adopting a three-dimensional semi-actuator model to carry out modelling on a stator and a rotor, and adopting a mode matching technology, the conservation law and the representation condition of compressor loss characteristics on an interface to obtain a characteristic value problem which solves a linearized flow field; and solving the characteristic value problem to obtain the characteristic disturbance frequency of a compression system, and judging system stability through the characteristic disturbance frequency, wherein a judgement standard is that the disturbance frequency [Omega] is a complex number, [Omega] is [Omega]R+i[Omega]I, disturbance is attenuated along with time evolution when the imaginary part [Omega]I of the frequency is greater than 0, the system is stable, otherwise, [Omega]I is less than 0, the disturbance is amplified along with time, and the system loses stability. The prediction method can be used for predicting the rotating stall inception of the multistage axial-flow compressor.
Owner:BEIHANG UNIV

Axial flow compressor stall surge prediction device based on frequency characteristic change

The invention discloses an axial flow compressor stall surge prediction device based on frequency characteristic change. An inlet dynamic pressure/flow sensor is mounted at an inlet casing wall of a compressor, an outlet dynamic pressure/flow sensor is mounted on the periphery of an outlet casing wall of the compressor, the inlet dynamic pressure/flow sensor and the outlet dynamics/flow pressure sensor are correspondingly connected with a signal acquisition card at the same time, the signal acquisition card is connected with a single-chip microcomputer established with a spectrum characteristic database, a noise energy ratio threshold valve is arranged in the spectrum characteristic database, and the single-chip microcomputer is connected with a controller conducting the surge early warning when the radiation noise energy ratio of the outlet end of the compressor reaches a set noise energy ratio threshold valve. By establishing the spectrum characteristic database, the pressure fluctuation frequency change in a working process of detecting and monitoring the compressor in real time, the frequency spectrum characteristics of the inlet air pressure and the outlet air pressure can becompared, the early warning is effectively conducted on the occurring of the surge before the surge occurs, and the reliability is high.
Owner:JIANGXI UNIV OF SCI & TECH

Compressor stall pre-warning signal reconstruction and identification method based on single-path sensor signal

The invention provides a compressor stall pre-warning signal reconstruction and identification method based on a single-path sensor signal. The method includes obtaining the pressure signal through asingle-path sensor installed in the circumferential direction of a front-stage case of a compressor; and then according to the rotation characteristic of the compressor stall pre-warning signal, simulating the output of a multi-path sensor in the circumferential direction by adopting a signal translation method, and realizing the multi-path stall pre-warning signal reconstruction of the compressor. The reconstructed multi-path stall pre-warning signal is expressed as the superposition of the torus average and the circumferential pressure disturbance in the vector form, and the circumferentialpressure disturbance vector is subjected to the spatial Fourier transform to be decomposed into the superposition of multi-order modal waves; the first-order modal amplitude is compared with the set threshold value, and the stall is considered to occur when the threshold value is exceeded, and the identification of the stall pre-warning signal is realized. A single sensor can be used to replace aplurality of sensors, so that the defect that a plurality of sensors are installed is overcome, and the method has a large engineering application value.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

Prediction Method of Stall Boundary of Multistage Axial Compressor Based on Eigenvalue Theory

The invention relates to a prediction method of a multistage axial-flow compressor stall margin. The prediction method comprises the following steps: according to a situation of the rotating stall inception of an axial-flow compressor, adopting a small perturbation theory to establish a three-dimensional compressible Euler equation which describes a flow field; applying harmonic decomposition and a dispersion relation to establish a boundary condition on the hub and the leaf opex of each drainage basin; adopting a three-dimensional semi-actuator model to carry out modelling on a stator and a rotor, and adopting a mode matching technology, the conservation law and the representation condition of compressor loss characteristics on an interface to obtain a characteristic value problem which solves a linearized flow field; and solving the characteristic value problem to obtain the characteristic disturbance frequency of a compression system, and judging system stability through the characteristic disturbance frequency, wherein a judgement standard is that the disturbance frequency [Omega] is a complex number, [Omega] is [Omega]R+i[Omega]I, disturbance is attenuated along with time evolution when the imaginary part [Omega]I of the frequency is greater than 0, the system is stable, otherwise, [Omega]I is less than 0, the disturbance is amplified along with time, and the system loses stability. The prediction method can be used for predicting the rotating stall inception of the multistage axial-flow compressor.
Owner:BEIHANG UNIV

Axial seam processing casing

The invention provides an axial slit processing casing, which belongs to the technical field of air compressors. The processing casing includes: the moving blades of the compressor rotor are arranged in the processing casing, and the edges of the moving blades face the inner wall of the processing casing; a plurality of edges are provided on the inner wall of the processing casing from the upstream to the downstream corresponding to the edge of the moving blades. The axially extending slit structure of the processing casing, and a plurality of slit structures are uniformly and misplaced distributed along the circumferential direction of the processing casing. The treatment casing provided by the present invention can use the pressure difference between upstream and downstream in the passage to suck the low-energy fluid blocked on the top of the moving blade into the suction slot through the axially misaligned slot structure, and at the same time, it can maximize the flow close to the downstream of the passage. The low-energy fluid is sucked into the slot, and this part of the fluid is ejected upstream of the moving blade, improving the flow condition entering the top of the moving blade, delaying the occurrence of compressor stall, widening the margin of the compressor, and negatively affecting the peak efficiency of the compressor smaller.
Owner:INST OF ENGINEERING THERMOPHYSICS - CHINESE ACAD OF SCI

Compressor Stall Premonition Signal Reconstruction and Recognition Method Based on Single Sensor Signal

The invention provides a compressor stall pre-warning signal reconstruction and identification method based on a single-path sensor signal. The method includes obtaining the pressure signal through asingle-path sensor installed in the circumferential direction of a front-stage case of a compressor; and then according to the rotation characteristic of the compressor stall pre-warning signal, simulating the output of a multi-path sensor in the circumferential direction by adopting a signal translation method, and realizing the multi-path stall pre-warning signal reconstruction of the compressor. The reconstructed multi-path stall pre-warning signal is expressed as the superposition of the torus average and the circumferential pressure disturbance in the vector form, and the circumferentialpressure disturbance vector is subjected to the spatial Fourier transform to be decomposed into the superposition of multi-order modal waves; the first-order modal amplitude is compared with the set threshold value, and the stall is considered to occur when the threshold value is exceeded, and the identification of the stall pre-warning signal is realized. A single sensor can be used to replace aplurality of sensors, so that the defect that a plurality of sensors are installed is overcome, and the method has a large engineering application value.
Owner:NORTHWESTERN POLYTECHNICAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products