31 results about "Forced oscillation" patented technology

The invention discloses an electric power system multi-source forced oscillation positioning method based on a graph attention network and variational mode decomposition. The method comprises the following steps: acquiring power grid topology data of a target electric power system and attribute data of each node; inputting the power grid topology data and the attribute data of each node into a pre-trained target network model, wherein the topology perception space decoupling network processes the power grid topology data and the attribute data of each node and outputs the energy contribution degree of each node; the frequency domain time decoupling network processes the attribute data of each node and the corresponding energy contribution degree and outputs a synchronous machine component and a fan component; and the physical constraint oscillation classification network processes the synchronous machine component and the fan component and outputs a multi-source forced oscillation positioning result. According to the invention, the modeling of the electrical connection relation between the power grid nodes is realized through the topology perception space decoupling network, and the decoupling of the synchronous machine and the fan component is realized through the frequency domain time decoupling network, so that the precision of oscillation source space positioning is improved.

The invention relates to the field of power systems, in particular to a method and system for improving the low-frequency oscillation suppression capability of a power system stabilizer, and the method comprises the steps: building an excitation regulator hardware-in-loop simulation platform to form a closed-loop test environment; constructing a model containing prime mover forced oscillation and generator special operation conditions; configuring simulation parameters based on the actual constant value of the unit; injecting multi-modal disturbance and acquiring a dynamic signal; calculating a multi-dimensional index for evaluating the low-frequency oscillation suppression capability of the PSS; generating a full-working-condition adaptability judgment result through a dynamic weight distribution algorithm; and if not, optimizing the PSS parameters by taking the weighted comprehensive score as a target, and repeating the test until the requirements are met. According to the invention, the problem that the low-frequency oscillation suppression capability of the traditional PSS parameter setting is insufficient under a special working condition is solved.

The invention discloses a power system forced oscillation source positioning method based on ICEEMDAN-ITEO, and belongs to the technical field of power system stability analysis and control. In a signal adaptive decomposition and modal extraction part, an ICEEMDAN algorithm is utilized to decompose a wide-area measurement signal, suppress modal aliasing and noise interference and obtain a modal component with physical significance; the energy weight of each modal component is calculated by adopting ITEO, the identification capability of the dominant forced oscillation component is enhanced through an adjustable resolution parameter, and the time domain waveform of the dominant forced oscillation component is reconstructed; in the oscillation source judgment part based on the dissipation mode energy flow, the dissipation mode energy flow of each generator port is calculated according to the dominant oscillation mode component, and accurate judgment of a disturbance source is realized through linear fitting of slope characteristics of the dissipation mode energy flow; in the verification and application part of the oscillation source positioning result, the accuracy and robustness of the method in different noise and disturbance scenes are verified through a test system comprising a network-forming fan.

The invention discloses a hybrid CGAN-based power system forced oscillation source positioning method, and the method comprises the steps: inputting to-be-recognized data into a trained CNN classifier, and obtaining an oscillation source recognition result; wherein the CNN classifier is trained according to the mixed data set; the mixed data set is constructed according to FO generation data generated by the trained mixed CGAN and real data; the FO generation data is false forced oscillation data; each group of real data comprises active power, reactive power, voltage amplitude and voltage phase angle; the hybrid CGAN is obtained according to the CNN, the BiLSTM and the attention mechanism. According to the method, the convolutional neural network, the bidirectional long-short-term memory network and the attention mechanism are fused into the generator and the discriminator of the CGAN, so that the similarity between the FO generation data and the real data is improved, the FO generation data and the real data are used for training, and the precision of forced oscillation source positioning can be effectively improved.

The invention discloses a power grid synchronous broadband forced oscillation detection and positioning method and system and a medium, and the method comprises the steps: obtaining an oscillation data sequence of a power grid and a power system topological graph, and inputting the oscillation data sequence and the power system topological graph into a pre-trained multi-task graph sequence network, so as to obtain an oscillation source positioning detection result; comprising the following steps: performing information transmission on an oscillation data sequence through a convolutional layer, a pooling layer and an output feature obtained by a created embeddable sequence autoregression module and a topological graph of a power system through a constructed multi-task graph sequence network, and obtaining node features of each node through a gate control cycle unit GRU, and inputting a graph-level detection branch and a node-level detection branch to realize event type classification of the oscillation source and positioning of the oscillation source. The method aims at reducing interference of other information on broadband oscillation information, data are dynamically analyzed from the two dimensions of time and space, and broadband oscillation source positioning is rapidly and accurately achieved.

The invention discloses an oscillating source station single-station positioning and removing method based on meta-learning, electronic equipment and a storage medium, and relates to the technical field of power systems and automation thereof. Comprising the following steps: acquiring oscillation operation data of a single station, the oscillation operation data comprising electrical characteristics of a station exit; performing feature extraction on the oscillation operation data through a pre-trained neural network model, and judging whether the single station is an oscillation source or not; generating an optimal initial parameter of the neural network model based on a meta-learning process, wherein the meta-learning process is trained by using a forced oscillation sample; and if the single-field station is an oscillation source, triggering a circuit breaker to execute cutting operation, and suppressing oscillation propagation. Compared with a traditional oscillation source positioning method, the method avoids a data transmission process between stations, avoids communication difficulty, and has engineering practicability and timeliness.

The application discloses a parameter setting method, device and system for a swing reactance of an excitation regulator, wherein the parameter setting method adds forced oscillation to a speed regulator link in a closed-loop simulation test environment, so as to stimulate equal-amplitude oscillation of mechanical power and unit frequency in a set frequency range; the error between a real signal value output by a real excitation device and a simulation signal value output by a simulation test model is calculated, the initial value of the swing reactance parameter is selected according to the error, the adaptability and effectiveness of the swing reactance are verified based on the initial value, the test result is obtained through a voltage step test under a set working condition, and the swing reactance parameter is modified according to the test result, so that the active power fluctuation when the PSS is put into operation meets the set condition. The parameter setting method provided by the application can reduce the risk of power oscillation accidents caused by large unit speed calculation error due to unreasonable swing reactance parameter setting and weak damping or even negative damping of the PSS.

The utility model relates to the technical field of magnetic oscillation, and discloses a novel magnetic oscillator which comprises a base, a circuit board is installed in the base, the bottom of the circuit board is in threaded connection with a coil framework, a central column is sleeved with the circuit board, and front plates at the two ends of the magnetic base are sleeved with the surface of the central column. The surface of the central column is sleeved with a magnetic base middle-end iron plate, and the upper end and the lower end of the magnetic base middle-end iron plate are provided with circular magnets. Compared with a traditional oscillator which adopts a cam eccentricity principle and is driven by a motor, a magnetic force driving mode does not need a large motor and a large transmission mechanism, the overall structure is more compact, the occupied space is small, the first guide column and the second guide column are arranged and are fixed through the pin and the second bolt, and the operation is convenient. And the use of mechanical transmission parts is reduced, the problem of serious abrasion of a transmission mechanism and a supporting rod after long-time movement is avoided, the service life of the device is prolonged, and the stability of the device is improved.

The present application belongs to the technical field of fan torsional vibration prediction, and particularly relates to a grid-connected direct-drive fan torsional vibration prediction method, device, equipment and storage medium. The method comprises: constructing a small signal model of the grid-connected direct-drive fan; based on a shafting double-mass block model, deducing the relationship between the port voltage and current signals of the direct-drive fan and the torsional vibration of the shafting of the direct-drive fan when forced oscillation occurs; when forced oscillation occurs, by constructing a wind farm s-domain node admittance model, the transfer function between different grid node ports is obtained, so as to calculate the required grid node port voltage and current signals, taking the calculated voltage and current signals as the input of the small signal model of the grid-connected direct-drive fan, and combining the relationship between the port voltage and current signals of the grid-connected direct-drive fan and the torsional vibration of the shafting of the grid-connected direct-drive fan to calculate the torsional vibration of the shafting of the grid-connected direct-drive fan caused by forced oscillation. It is helpful to timely find and handle potential torsional vibration problems and avoid the occurrence of faults.

The present application relates to power distribution network technical field, especially a kind of forced oscillation detection and positioning method and system for power distribution network. Including using spectral graph theory, cut set partition is carried out to power grid, and cut set segmentation is repeatedly carried out based on subnetwork energy transmission value;Based on the key line of the line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of cut set of the key line of

The invention discloses a forced oscillation suppression method combining phase shift compensation and active damping, and belongs to the technical field of stable operation analysis of a power system. According to the forced oscillation suppression method combining phase shift compensation and active damping, most inter-harmonic disturbance sources are suppressed through direct-current voltage phase shift compensation, system damping is improved in a small-amplitude mode by combining active damping control, and the forced oscillation amplitude is suppressed under the condition that system performance is not affected. The mechanism that the inter-harmonic disturbance source triggers forced oscillation of the forced system is elaborated by using an equivalent disturbance equation, the influence of the amplitude of the inter-harmonic disturbance source and the damping of the forced system on the amplitude of forced oscillation is considered, and a forced oscillation suppression strategy of phase shift compensation and active damping is combined. The amplitude of an inter-harmonic disturbance source is suppressed through phase shift compensation, and the weak damping state of a forced system is improved through active damping, so that the amplitude of forced oscillation is suppressed.

The invention discloses a CPSD-based power system forced oscillation source equipment level tracing method, and relates to the technical field of power system forced oscillation disturbance source equipment level positioning, and the method comprises the steps: collecting and calculating the active power, reactive power, frequency and voltage amplitude measurement data of a generator node during a forced oscillation period; obtaining a dynamic component of the relative steady-state value; calculating a frequency domain dissipation energy spectrum, and confirming a forced oscillation dominant oscillation frequency and a forced oscillation source; respectively calculating the frequency domain cross-power spectral density of the speed regulation system and the excitation system during the forced oscillation period; calculating amplitude spectrums and phase spectrums of each generator speed regulation system and excitation system at the forced oscillation frequency, and cooperatively positioning forced oscillation disturbance equipment by the amplitude spectrums and the phase spectrums; and calculating the energy contribution rate of each generator speed regulation system and excitation system, and determining the dominant forced oscillation disturbance equipment. Therefore, by adopting the method, the control system where the disturbance source is located can be accurately positioned, and forced oscillation equipment-level positioning is realized.

The invention discloses a forced oscillation hydrodynamic experiment device for covering a buoyancy block rigid riser, which comprises an actuation module, a physical quantity sensing module, a rigid riser and a buoyancy block model, and is characterized in that the physical quantity sensing module is mounted on an actuation control module, the rigid riser is mounted on the physical quantity sensing module, the buoyancy block model is mounted on the rigid riser, and the buoyancy block model is mounted on the rigid riser. Therefore, the motion response of the rigid riser and the hydrodynamic load of the rigid riser can be measured in real time. The high-Reynolds-number forced oscillation experiment can be carried out on the rigid cylinder covered with the buoyancy block, the hydrodynamic coefficient of the rigid cylinder covered with the buoyancy block in a high-Reynolds-number flow field is obtained, and then a hydrodynamic coefficient library used for forecasting is established. The experimental device is simple in structure and easy to build and disassemble, meanwhile, the rigid stand pipe model is easy to disassemble, other stand pipe accessory structures can be replaced, and the experimental mode is flexible.

The invention discloses a progressive positioning method for a forced oscillation source of a power system considering a grid-forming type double-fed fan, and the method comprises the steps: carrying out the spectrum Granger causal analysis of the time series of the output active power of each synchronous machine which is strongly related to oscillation and the grid-forming type double-fed fan, and judging the causal relationship between power generation units; preliminarily positioning a power generation unit where the forced oscillation disturbance source is located; furthermore, an unknown input observer based on performance constraint is adopted to perform positioning verification, and the unknown input observer with the performance constraint is designed based on linear models of power generation units such as a synchronous machine with a potential forced oscillation source and a grid-forming double-fed fan. And establishing a structured residual set capable of effectively representing the forced oscillation disturbance source in the power generation unit by using the observer, and comparing each residual frequency spectrum in the residual set to realize accurate positioning of the forced oscillation disturbance source. Therefore, rapidity, convenience and accuracy required by positioning can be met.

This application relates to the field of marine engineering technology, and discloses a floating breakwater with a resonant cavity and its hydrodynamic performance analysis method. The method includes: constructing a floating breakwater model; determining the target length and target width of the resonant cavity, the wave-facing open-type wave-dissipating chamber, and the wave-repelling open-type wave-dissipating chamber on the leeward side; dividing the entire fluid domain of the floating breakwater into multiple regions; obtaining the velocity potential of each region; calculating the wave force of the floating breakwater; using the velocity potential of each region to express the added mass and radiation damping of the floating breakwater when it undergoes unit forced oscillation, so as to construct the motion response equation of the floating breakwater in the frequency domain under wave action; obtaining the transmission coefficient and energy dissipation coefficient of the floating breakwater through the equation set, and evaluating the shielding performance and wave-dissipating performance of the floating breakwater based on each coefficient. The synergy of the resonant cavity and the open-type wave-dissipating chamber can improve the wave-dissipating efficiency, and the analysis method can provide an efficient and reliable basis for the design of breakwaters.

The invention relates to a forming and / or cutting device (10) and a method for operating it. The forming and / or cutting device (10) is configured to plastically deform and / or cut a workpiece (11). The forming and / or cutting device (10) has a ram (15) mounted linearly on a machine frame (13), on which a tool (16) is arranged. The ram (15) can be moved by means of a ram drive (21) in a working direction (A) and in the opposite direction to the working direction (A) in a retraction direction (R) between two reversal points (U1, U2). During or after the tool (16) hits the workpiece (11) and in particular before reaching the second reversal point (U2), a rotary movement of a motor (22) of the plunger drive (21) is modified at least during a deceleration phase (P) such that the rotational speed of the motor (22) is first reduced and then increased again.This allows for active damping of occurring forces and force oscillations.

The invention discloses a new energy power system subsynchronous / super-synchronous oscillation dominant frequency rapid prediction and identification method and system based on transfer learning, and relates to the field of power systems and automation thereof, and the method comprises the steps: constructing forced oscillation data and subsynchronous / super-synchronous oscillation data; selecting electrical operation data in an oscillation initial stage from the forced oscillation data as characteristic data, taking dominant frequency in an oscillation stable stage in the forced oscillation data as a label, and constructing a forced oscillation sample; similarly, constructing a sub-synchronous / super-synchronous oscillation sample based on the sub-synchronous / super-synchronous oscillation data; taking a forced oscillation sample as a source domain, taking a sub / super-synchronous oscillation sample as a target domain, and training the machine learning neural network model through a transfer learning method to obtain an oscillation identification model; and performing rapid prediction and identification on the dominant frequency of the sub / super-synchronous oscillation of the new energy grid-connected point by using the oscillation identification model. The method has good timeliness for prediction and identification of the dominant frequency of the sub / super-synchronous oscillation.

This invention discloses a method for suppressing low-frequency oscillations in energy storage based on a damped controller and a resonant controller, specifically relating to the field of low-frequency oscillation suppression. The method involves identifying the type of low-frequency oscillation and then adaptively selecting and online tuning the corresponding damped or resonant controller to generate compensation power. This method constructs a dual-discrimination path for negatively damped oscillations and forced oscillations, effectively identifying the type of low-frequency oscillation and overcoming the deficiency of lacking effective oscillation type identification capability. By adaptively selecting the controller according to the oscillation type, targeted suppression is applied to different types of oscillations, avoiding ineffective adjustment of a single control strategy, reducing resource waste in the energy storage system, improving suppression efficiency, slowing down the lifespan decay of the energy storage device, and curbing the propagation of forced oscillation energy.

The invention relates to the technical field of ocean engineering, and discloses an energy dissipation floating breakwater with a resonance type cavity and a hydrodynamic performance analysis method of the energy dissipation floating breakwater. Respectively determining target lengths and target widths of the resonant cavity, the wave head side perforated wave dissipation chamber and the wave back side perforated wave dissipation chamber; the whole fluid domain of the floating breakwater is divided into a plurality of areas; solving the speed potential of each region; calculating the wave force of the floating breakwater, and expressing the additional mass and radiation damping when the floating breakwater performs unit forced oscillation by utilizing the speed potential of each region so as to construct a motion response equation of the floating breakwater in a frequency domain under the action of waves; the transmission coefficient and the energy dissipation coefficient of the floating breakwater are obtained through the equation set, and the shielding performance and the wave dissipation performance of the floating breakwater are evaluated according to all the coefficients. The cooperation of the resonant cavity and the perforated wave dissipation chamber can improve the wave dissipation efficiency, and the analysis method can provide an efficient and reliable basis for the design of the breakwater.

In absence of electrical approaches for realization of highly stable RF oscillator, opto-electronic oscillators (OEO) techniques are provided, where self-forced oscillation techniques using long optical delays demonstrate significant short-term and long-term frequency stability. Fully integrated opto-electronic oscillator chip (IOEC) may be the most efficient realization of an RF frequency synthesizer in terms of operation frequency (covering microwave and millimeter wave), size (<10 cm3), ruggedness to environmental effects of temperature (−40 to 80 C), vibration (up to 40 g), low timing jitter (<5 fs for 40 GHz carrier), and wall-plug efficiency (output power >10 dBm from under 1 W power). A free-running III-V (primarily InP) based multi-mode laser (MML) diodes is designed with large mode number (e.g., over 60 modes) and intermodal oscillation frequency compatible with desired RF carrier signal (e.g., 1-40 GHz).

A computer-implemented method includes providing, via a user interface of a respiratory therapy system, a flow of air to an airway of a user. The method also includes superimposing a pattern of pseudo-random varying pressure on the flow of air. The method also includes generating, via a pressure sensor of the respiratory therapy system, pressure data associated with the flow of air. The method also includes determining, based at least in part on the pressure data, physiological data indicative of one or more airway characteristic of the user.

The application discloses a power system forced oscillation positioning method based on a space-time attention mechanism, solves the problems of complex space-time dependence on an accurate model, noise sensitivity and insufficient positioning accuracy in a multi-source or complex oscillation scene in the prior art, and realizes high-precision single-source forced oscillation positioning; the method comprises the following steps: acquiring power system node time series data, and preprocessing to obtain a standardized space-time input matrix; a weighted normalized adjacency matrix is constructed based on a topological structure; the space-time input data is linearly mapped to a high-dimensional space to obtain a high-dimensional feature tensor; node attributes and time series encoding are embedded to obtain a space-time embedded feature tensor; a space-enhanced feature tensor is obtained based on the adjacency matrix; time dimension attention calculation and feature aggregation are performed on the space-enhanced tensor to obtain a space-time fusion feature tensor; global average pooling and full connection classification are performed on the space-time fusion tensor to output node oscillation source probability distribution and locate the oscillation source position.

This application discloses a method for identifying and locating forced oscillations using multimodal information fusion and an autoencoder. First, the logarithmic change in voltage amplitude and the change in active power are calculated from the collected data and fused using the entropy weight method to comprehensively characterize the oscillation signal. Then, the fused signal is decomposed into modal components of different frequencies using a fully integrated empirical mode decomposition and adaptive noise (CEEMDAN) algorithm, reducing data complexity. Zero-padded Hanning window fast Fourier transform (ZP-Hanning-FFT) is used to improve spectral resolution, calculate the frequency, amplitude, and energy of the modal components, and output the identification results, thus improving identification accuracy. Next, the mean, standard deviation, and energy of the modal components are calculated, and multi-scale convolution is used to extract feature information at different scales, which is then input into the autoencoder for training, improving the localization accuracy.

The invention discloses an energy storage suppression method and device for forced power oscillation, a program and a medium, and relates to the technical field of power system stability control, and the method comprises the following steps: when forced power oscillation occurs in a system, executing oscillation mode analysis to identify a dominant oscillation mode, and quantifying forced oscillation response characteristics under distributed disturbance injection; modeling is carried out based on isomorphic characteristics of an energy storage system and a disturbance source on a power grid power injection level, and an energy storage reverse phase compensation power reference value required by oscillation suppression is calculated; the method comprises the following steps: for a distributed energy storage collaborative scene, fusing wide area measurement system data, and constructing a power distribution model with energy storage resource global optimization configuration as a target when oscillation is completely suppressed; a feedback control module is introduced for calculation errors, and energy storage output power is dynamically corrected through real-time oscillation state information until oscillation is completely converged; and maintaining the dynamic power output of the energy storage system until the distributed disturbance source is removed. And the transient stability margin of the power system is obviously improved.

The invention discloses an electric power system low-frequency oscillation type identification and tracing method based on measurement, and relates to the technical field of electric power system low-frequency oscillation monitoring. Comprising the following steps: acquiring measurement data of a plurality of generators in a power system through a synchronous phasor measurement unit (PMU); performing detrending processing on the measurement data, and calculating the deviation value of each measurement relative to the steady state trend; performing spectral analysis on the at least one deviation value, and extracting a dominant oscillation frequency of the system; filtering the deviation value by using a band-pass filter, and extracting a characteristic signal taking the dominant oscillation frequency as a central frequency band; calculating a port oscillation energy component of each generator based on the filtered characteristic signal; according to the trend characteristic of the port oscillation energy component, identifying whether the low-frequency oscillation type of the system is weak damped oscillation or forced oscillation; and calculating the port oscillation energy flow of each generator based on the filtered characteristic signal, and preliminarily positioning the unit generating the oscillation source according to the comparison result of the change slope of the port oscillation energy flow.

The invention discloses a system weak point identification method and system for forced oscillation, and the method comprises the steps: collecting the time series data of the rotor angle, angular velocity and voltage phase / frequency of a generator of a power system based on a PMU, and constructing a measurement matrix and a time derivative matrix after preprocessing the collected data; constructing a feature matrix, solving a coefficient matrix, obtaining a state transition matrix and an input matrix from the coefficient matrix, and extracting to obtain a system dynamic equation; calculating a sensitivity matrix to represent the influence of the input on the system state; defining an output matrix, and calculating an influence matrix; quantifying the total contribution of each input to oscillation based on the sensitivity matrix and the influence matrix, and completing the recognition of the system weak point; the method can extract the dynamic equation of the system without depending on specific parameter information of the power system, identifies the weak point of the system through quantitative analysis, and improves the simplicity and reliability of the method.

The invention belongs to the technical field of forced oscillation, and provides a pulse oscillation technology system error-based multi-point calibration method, which comprises the steps of theoretical impedance data calculation, fitting data collection, compensation curved surface fitting model construction and impedance data correction. According to the method, the compensation curved surface fitting model is constructed, so that the measurement accuracy in high-frequency and high-impedance ranges is improved, errors caused by factors such as system asymmetry can be processed, more accurate calibration can be realized for full-frequency bands and various impedance ranges, physiological conditions of different patients can be adapted, and the reliability of measurement results is ensured.

Techniques for determining a damping rate of unforced oscillations of a meniscus are disclosed. An example method includes receiving input describing a shape of a container, physical parameters of a liquid inside the container, and an equilibrium shape of the meniscus. The method also includes generating a mesh conforming to the shape of the container and generating a discrete version of a continuous eigenvalue problem based on this mesh. The method also includes computing, at appropriate mesh nodes, values for pressure, velocity components, and meniscus surface deformation corresponding to a suitable number of least-damped late-time oscillation modes of the liquid and computing an angular frequency and damping rate of these least-damped late-time oscillation modes from the discrete version of the continuous eigenvalue problem. The method also includes identifying the mode that has the lowest damping rate and computing a liquid relaxation time by inverting the damping rate of the identified mode.

The application discloses a power grid synchronous broadband forced oscillation detection and positioning method and system and a medium. The method comprises the following steps: obtaining oscillation data sequences and a power system topology graph of a power grid, and inputting the pre-trained multi-task graph sequence network to obtain an oscillation source positioning detection result. The oscillation data sequences are input into a convolution layer, a pooling layer and an output feature obtained by a created embeddable sequence autoregressive module. The power system topology graph is input into a multi-task graph sequence network to perform information transmission, and a node feature of each node is obtained through a gated recurrent unit (GRU). Then, the method is input into a graph-level detection branch and a node-level detection branch to realize event type classification of an oscillation source and positioning of the oscillation source. The application aims to reduce the interference of other information on broadband oscillation information, dynamically analyze data from two dimensions of time and space, and quickly and accurately realize broadband oscillation source positioning.

The present application belongs to the field of direct connection hypersonic inlet / isolator ground wind tunnel experiment, and particularly relates to a forced oscillation generating device for hypersonic inlet / isolator and a design method thereof. Step 1: determining the static pressure and Mach number of the incoming flow gas; Step 2: based on the static pressure and Mach number of the incoming flow gas in step 1, determining the material adopted by the shell of the oscillation generating device and the oscillation back pressure generating mechanism; Step 3: based on the material selected in step 2, determining the type of the driving device in the oscillation back pressure generating mechanism; Step 4: based on the type of the driving device in step 3, determining the transmission ratio of the gear and the throttling ratio range of the rotating blade in the oscillation back pressure generating mechanism; Step 5: based on the transmission ratio and the throttling ratio in step 4, processing the forced oscillation device experimental piece and installing it on the ground wind tunnel experiment table, selecting the throttling ratio and the oscillation frequency and conducting the experiment, and the forced oscillation device will generate the expected oscillation back pressure at the outlet of the isolator. To solve the problem of flow separation in the inlet in the prior art, and further affect the power and control performance of the aircraft.