Flexible DC power transmission system oscillation risk assessment method based on phase-space reconstruction

Abstract

The invention discloses a flexible direct current power transmission system oscillation risk assessment method based on phase-space reconstruction, and belongs to the technical field of power system oscillation risk assessment, and the method comprises the steps: collecting an oscillation output signal comprising an oscillation component when a flexible direct current power transmission system is unstable, and carrying out the preprocessing of the oscillation output signal; performing phase-space trajectory reconstruction on the preprocessed oscillation output signal by adopting a time delay embedding method; on the basis of the reconstructed phase space trajectory, predicting the phase space trajectory of the oscillation output signal of the power transmission system at the next time by using a linear autoregression model as a prediction model or a prediction model representing the geometrical characteristics of the phase space trajectory on polar coordinates; and analyzing the evolution trajectory of the predicted phase space trajectory, and evaluating the oscillation risk of the power transmission system. According to the method, oscillation risk assessment and early warning are realized based on real-time measurement data, the method does not depend on accurate modeling of the system, the complexity is low, the operation cost is low, and the method has relatively high adaptability under different working conditions.

Description

Technical Field

[0001] This invention belongs to the field of power system oscillation risk assessment technology, specifically relating to a method for oscillation risk assessment of flexible DC transmission systems based on phase space reconstruction. Background Technology

[0002] With its advantages of flexibility, low harmonic content, convenient voltage boosting and capacity expansion, and adaptability to uncertain energy access, flexible DC transmission technology is widely used in new energy grid connection and large-scale power grid interconnection. However, the large-scale commissioning of flexible DC transmission projects has led to an increase in the proportion of power electronic equipment in the power system. Because power electronic control introduces high-frequency nonlinear elements, the interactions between various heterogeneous power electronic devices such as flexible DC converters, new energy converters, and energy storage devices, as well as between these devices and the AC grid, bring numerous stability problems to the dynamic processes of the power grid, among which oscillation instability is particularly prominent. Oscillation problems not only reduce power quality and endanger equipment safety, but may also further lead to system instability and divergence, seriously threatening the safe and stable operation of the power system.

[0003] Currently, methods for assessing the oscillation risk of DC transmission systems mainly focus on the oscillation risk assessment of AC systems such as thermal power units connected to flexible DC grids or wind power connected to the grid. Some studies have used unit action coefficients and time-domain simulation to analyze and assess the risk of subsynchronous oscillations induced by the Luxi back-to-back hybrid DC transmission project; however, this method cannot deeply analyze the impact of changes in system operating parameters on oscillation risk. Other studies have used a doubly-fed induction generator (DFIG) wind farm connected to a flexible DC grid as an example, employing a frequency scanning method and utilizing the equivalent circuit damping criterion to quantitatively assess the impact of changes in control parameters on the potential broadband oscillation risk of the system. However, this method requires additional equipment to inject disturbances into the system, which is detrimental to normal system operation, and the scanning process is time-consuming. The above-mentioned oscillation risk assessment methods require the establishment of an accurate system model and the assessment of the system's oscillation risk based on this model; significant misjudgments may occur when there are errors in the estimation of operating parameters. To reduce reliance on system models, some scholars have proposed data-driven oscillation risk assessment methods, constructing a wide-area monitoring and analysis system for subsynchronous oscillations. These methods combine damping identification and pooling impedance models to determine whether subsynchronous oscillations can be stabilized. However, most of these methods assume that the system oscillates under large disturbances. In reality, large disturbances are not common, and since the system has already oscillated, they cannot serve as a risk warning. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a method for assessing the oscillation risk of flexible DC transmission systems based on phase space reconstruction. Based on real-time measurement data, it achieves oscillation risk assessment and early warning, without relying on precise system modeling. It has low complexity, low operating costs, and strong adaptability under different operating conditions.

[0005] This invention provides the following technical solution: Firstly, a method for assessing the oscillation risk of flexible DC transmission systems based on phase space reconstruction is provided, including: The oscillation output signal, including the oscillation component, is collected when the flexible DC transmission system becomes unstable, and then preprocessed. The phase space trajectory of the preprocessed oscillating output signal is reconstructed using a time delay embedding method. Based on the reconstructed phase space trajectory, a linear autoregressive model or a prediction model that characterizes the geometric features of the phase space trajectory in polar coordinates is used to predict the phase space trajectory of the power transmission system oscillation output signal at the next time step. The evolution trajectory of the predicted phase space trajectory is analyzed to assess the oscillation risk of the power transmission system.

[0006] Optionally, the step of reconstructing the phase space trajectory of the preprocessed oscillation output signal using the time delay embedding method specifically includes: The autocorrelation coefficient method is used to determine the delay time of phase space reconstruction and the spurious nearest neighbor method is used to determine the embedding dimension of phase space reconstruction. Based on Takens' theorem, the phase space trajectory of the preprocessed oscillation output signal is reconstructed.

[0007] Optionally, based on the reconstructed phase space trajectory, a linear autoregressive model is used as the prediction model to predict the phase space trajectory of the transmission system oscillation output signal at the next time step, specifically: Using the phase variables reconstructed from the phase space as the state vector, and minimizing the one-step prediction error as the objective function, the parameters of the linear autoregressive model are estimated using the least squares method. The linear autoregressive model is as follows: ;in, and These are the embedding vectors of the phase space trajectories at the current time t and the next time t+1, respectively. Here is the state transition matrix. For bias, For noise; objective function for: ;in, , The number of valid samples. This represents the number of consecutive observations starting from the initial moment. For embedded dimensions, For delay time; Based on the solved state transition matrix, real-time acquisition, and preprocessed oscillation output signal, the phase space trajectory of the power transmission system oscillation output signal in the next time step is predicted.

[0008] Optionally, the phase space trajectory of the predicted next time-space oscillation output signal is represented as: ; in, The oscillating output signal at the predicted time The embedding vector of the phase space trajectory. , The timing for implementing suppression strategies in the power transmission system. and These are the set number of steps and time step, respectively. The oscillation output signal is at the input time. The phase space trajectory embedding vector, The bias term of the estimated linear autoregressive model, The state transition matrix of the estimated linear autoregressive model is shown in the superscript. i and Representing a matrix i power and Power of 1.

[0009] Optionally, based on the reconstructed phase space trajectory, a prediction model that characterizes the geometric features of the phase space trajectory in polar coordinates is used to predict the phase space trajectory of the transmission system oscillation output signal at the next time step, specifically: Map the reconstructed phase space trajectory to a lower-dimensional space; Minimizing the one-step prediction error is used as the objective function to solve for the parameters of the discrete prediction model combining radial quantity and intersection vector. Based on the real-time acquisition and preprocessed oscillation output signal, the phase space trajectory of the oscillation output signal in the next time step is predicted. The discrete prediction model for the combination of the radial quantity and the intersection vector is as follows: ; in, The radius of the phase space trajectory. This represents the mapping of points in the phase trajectory to a lower-dimensional space. For phase space trajectory phase, α , β , γ and ω 0 represents four parameters characterizing the phase space trajectory of the oscillation signal in a flexible DC transmission system; objective function for: ,in, , The number of valid samples. This represents the number of consecutive observations starting from the initial moment. For embedded dimensions, This is a delay time.

[0010] Optionally, the phase space trajectory of the predicted next time-space oscillation output signal is represented as: ; in, , , , , and The oscillating output signal at the predicted time The phase space trajectory radius and phase, , The timing for implementing suppression strategies in the power transmission system. and These are the set number of steps and time step, respectively. and The oscillation output signal at the input time is respectively The phase space trajectory radius and phase.

[0011] Optionally, the oscillation risk of the transmission system is assessed, specifically: if the phase space trajectory of the predicted oscillation output signal exceeds the activation conditions of the relay protection, then the transmission system has an oscillation risk; otherwise, the transmission system does not have an oscillation risk.

[0012] Optionally, in the oscillation output signal including oscillation components when the flexible DC transmission system becomes unstable, the flexible DC transmission system includes a flexible DC converter, and the oscillation output signal is: the current deviation signal between the d-axis and q-axis in the current control loop of the flexible DC converter. and .

[0013] In a second aspect, a computer device is provided, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the oscillation risk assessment method for flexible DC transmission systems based on phase space reconstruction as described in any one of the first aspects.

[0014] Thirdly, a computer-readable storage medium is provided for storing a computer program; when the computer program is executed by a processor, it implements the steps of the method for assessing the oscillation risk of a flexible DC transmission system based on phase space reconstruction as described in any one of the first aspects.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention achieves oscillation risk assessment and early warning by collecting measured system data and estimating the system phase trajectory in real time, exhibiting strong adaptability under different operating conditions. The prediction model of this invention is based on online data and does not rely on precise modeling of the transmission system. Compared with traditional model-based flexible DC system oscillation risk assessment methods, it avoids repetitive modeling and frequency sweeping when operating conditions change, resulting in lower algorithm complexity. Furthermore, by reconstructing the system phase space through measured data and achieving oscillation risk assessment based on the predicted phase trajectory, compared with existing purely data-driven risk assessment methods, it eliminates the need to detect large system disturbances before determining stability, enabling system oscillation risk prediction and early warning with low operating costs. Attached Figure Description

[0016] Figure 1 This is a flowchart of the oscillation risk assessment method for flexible DC transmission systems based on phase space reconstruction according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of the oscillation risk assessment method for flexible DC transmission systems based on phase space reconstruction according to Embodiment 2 of the present invention; Figure 3 The effectiveness of the oscillation risk assessment method proposed in Embodiment 2 of this invention is verified by time-domain simulation when there is no oscillation risk in the power transmission system. Figure 4 The effectiveness of the oscillation risk assessment method proposed in Embodiment 2 of this invention is verified by time-domain simulation when there is no oscillation risk in the power transmission system. Figure 5 This is a structural diagram of the flexible DC transmission system to which this invention is directed, and a schematic diagram of the application location of the proposed oscillation risk assessment method. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the scope of protection of the present invention. It should be noted that the term "comprising" and any variations thereof in the specification, claims and the above-mentioned drawings of the present invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or devices.

[0018] Example 1 like Figure 1 As shown, a method for assessing the oscillation risk of a flexible DC transmission system based on phase space reconstruction includes: Step S1: Collect the oscillation output signal, including the oscillation component, when the flexible DC transmission system becomes unstable, and preprocess it.

[0019] In this embodiment, the flexible DC transmission system includes a flexible DC converter, and the oscillation output signal is the current deviation signal between the d-axis and q-axis in the current control loop of the flexible DC converter. and .

[0020] Preprocessing includes two parts: sampling filtering and outlier data processing. First, the acquired oscillation output signal is input to a low-pass filter to remove high-frequency noise from the original signal. Based on this, the outlier data is further processed to minimize the interference introduced during sampling, thus avoiding any impact on subsequent phase space reconstruction. The specific steps are as follows: S11: Select a suitable low-pass filter cutoff frequency.

[0021] The low-pass filter stage must ensure that it retains the fundamental frequency and oscillation frequency band signals while filtering out high-frequency noise. For the oscillation problem of flexible DC transmission systems, the oscillation frequency is generally within 10kHz, so the cutoff frequency of the low-pass filter can be set to 10kHz.

[0022] S12: Detect abnormal data points.

[0023] For the collected data points, it is necessary to determine whether outliers exist. This invention uses an improved Z-score method for outlier detection: for a given data point in the collected data... Calculate the local average of the window consisting of the point and six surrounding data points. and local standard deviation Take three from the front and three from the back. Then it is believed This is an outlier.

[0024] S13: Handle abnormal data points.

[0025] Common methods for handling outliers include deletion, correction, and replacement. Based on an improved Z-score method for outlier detection, outlier data is... Replace with local window average .

[0026] Step S2: The phase space trajectory of the preprocessed oscillation output signal is reconstructed using the time delay embedding method.

[0027] In this embodiment, the autocorrelation coefficient method is used to determine the delay time of phase space reconstruction and the spurious nearest neighbor method is used to determine the embedding dimension of phase space reconstruction. Based on Takens' theorem, the phase space trajectory of the preprocessed oscillation output signal is reconstructed.

[0028] Step S2 specifically includes: S21: The dynamics of a flexible DC transmission system are defined as follows: ;in, t Moment n 3D state space is defined as ; It is a nonlinear mapping.

[0029] S22: After sufficient time, record that all system states converge to Z Tight manifold .

[0030] S23: Regarding , It is a smooth vector field. It is a smooth function, when integers L >2 Z At that time, mapping It is an embedding: ; in, t The system state at any given time is ,set up , , Then the above mapping can be expressed in the following form: ; S24: Based on Takens' theorem, when the dynamic behavior of a system is determined by a set of state variables, given a sufficiently long time series, a phase space structure identical to the original system can be reconstructed by selecting appropriate delay times and embedding dimensions. That is, a one-dimensional time series can be used to reconstruct the system. Different delay times To build d 3D phase space vector: ; in, For the first time series data i Dimensional data.

[0031] S25: The delay time for phase space reconstruction is determined based on the autocorrelation coefficient method, and its expression is as follows: ; in, N This indicates the number of delayed coordinates embedded in the Takens, i.e., the number of consecutive time steps from the initial moment. It is the autocorrelation function. Representing data y The average value.

[0032] S26: Determine the embedding dimension of the phase space reconstruction based on the spurious nearest neighbor method. The specific steps are as follows: S261: Regarding dEach vector in the phase space Each of them has a nearest neighbor point at Euclidean distance. The distance is: ; S262: Order ; like ,but for false proximity points, For the threshold; S263: From the embedding dimension d Starting with a minimum value of 2, calculate the proportion of false nearest neighbors, and then gradually increase the dimension. d Until the proportion of false nearest neighbors is less than 5% or the proportion of false nearest neighbors no longer increases. d When the value increases and decreases, it can be considered that the chaotic attractor is fully open. d This is the embedding dimension.

[0033] S27: Based on the embedded parameters determined above, the phase trajectory of the flexible DC transmission system is reconstructed, where the embedded vector can be defined as: ; Point set { v t That is, the system in the phase space plane. The trajectory.

[0034] Step S3: Based on the reconstructed phase space trajectory, use a linear autoregressive model as the prediction model to predict the phase space trajectory of the power transmission system oscillation output signal at the next time step.

[0035] S31: Using the phase variables reconstructed from the phase space as the state vector, minimizing the one-step prediction error as the objective function, the parameters of the linear autoregressive model are estimated using least squares with multiple known sets of observation data.

[0036] The linear autoregressive model is as follows: ; in, and These are the embedding vectors of the phase space trajectories at the current time t and the next time t+1, respectively. Here is the state transition matrix. , For bias, It is noise.

[0037] objective function for: ; in, , The number of valid samples. This represents the number of consecutive observations starting from the initial moment. For embedded dimensions, This is a delay time.

[0038] Using the objective function When solving a linear autoregressive model, the closed-form solution in matrix form can be written as: ; The superscript T stands for transpose.

[0039] S32: Based on the solved state transition matrix, real-time acquisition and preprocessing of the oscillation output signal, predict the phase space trajectory of the power transmission system oscillation output signal in the next time step.

[0040] The predicted phase space trajectory of the next time-space oscillation output signal is represented as: ; in, The oscillating output signal at the predicted time The embedding vector of the phase space trajectory. , The timing for implementing suppression strategies in the power transmission system. and These are the set number of steps and time step, respectively. The oscillation output signal is at the input time. The phase space trajectory embedding vector, The bias term of the estimated linear autoregressive model, The state transition matrix of the estimated linear autoregressive model is shown in the superscript. i and Representing a matrix i power and Power of 1.

[0041] Step S4: Analyze the evolution trajectory of the predicted phase space trajectory to assess the oscillation risk of the power transmission system.

[0042] The oscillation risk of the power transmission system is assessed as follows: if the phase space trajectory of the predicted oscillation output signal exceeds the activation conditions of the relay protection, then the power transmission system has an oscillation risk; otherwise, the power transmission system does not have an oscillation risk.

[0043] In this embodiment, the activation conditions of the flexible DC transmission relay protection (such as overcurrent protection setting values) are mapped to the phase plane space to obtain the vector of the relay protection device in the phase space. ,in and This represents the projection onto the phase plane coordinate axes. Furthermore, the following formula is used to determine whether the system has an oscillation risk.

[0044] ; Here, || represents the vector modulus operation.

[0045] Based on this, if If the condition is met, the system phase trajectory is within a safe range and there is no risk of system oscillation; otherwise, if the phase trajectory exceeds the relay protection activation condition in the future, there is a risk of system oscillation.

[0046] Example 2 like Figure 2 As shown, the difference between Example 2 and Example 1 is that step S3 is: based on the reconstructed phase space trajectory, a prediction model that characterizes the geometric features of the phase space trajectory on polar coordinates is used to predict the phase space trajectory of the power transmission system oscillation output signal at the next time.

[0047] The specific steps are as follows: Step S3-1: Map the reconstructed phase space trajectory to a low-dimensional space.

[0048] Specifically, by using Singular Value Decomposition (SVD) to map the system's phase space trajectory to a lower-dimensional space, we can obtain: ; in, and Corresponding point set The direction of the maximum singular value, with the superscript T representing transpose. This represents the mapping of points in the phase trajectory to a lower-dimensional space.

[0049] Step S3-2: Using minimizing the one-step prediction error as the objective function, solve for the parameters of the discrete prediction model combining radial quantity and intersection vector, and predict the phase space trajectory of the oscillation output signal in the next time step based on the real-time acquisition and preprocessed oscillation output signal.

[0050] Step S3-2-1: Based on the flow characteristics of the oscillation signal trajectory in phase space, construct a discrete candidate model composed of radial quantities and angular vectors, the expression of which is: ; in, The radius of the phase space trajectory. For phase space trajectory phase, α , β , γ and Four parameters characterizing the phase space trajectory of the oscillation signal in a flexible DC transmission system were determined through fitting.

[0051] Step S3-2-2 Objective Function for: ,in, , The number of valid samples. This represents the number of consecutive observations starting from the initial moment. For embedded dimensions, For the delay time, based on the objective function The parameters of a discrete candidate model, which is a combination of radial and angular vectors, can be obtained using a heuristic algorithm.

[0052] Sub-step S3-2-3: The phase space trajectory of the predicted next time-space oscillation output signal is represented as: ; in, , , , , and The oscillating output signal at the predicted time The phase space trajectory radius and phase, , The timing for implementing suppression strategies in the power transmission system. and These are the set number of steps and time step, respectively. and The oscillation output signal at the input time is respectively The phase space trajectory radius and phase.

[0053] Step S4: Analyze the evolution trajectory of the predicted phase space trajectory to assess the oscillation risk of the power transmission system.

[0054] The oscillation risk of the power transmission system is assessed as follows: if the phase space trajectory of the predicted oscillation output signal exceeds the activation conditions of the relay protection, then the power transmission system has an oscillation risk; otherwise, the power transmission system does not have an oscillation risk.

[0055] In this embodiment, the activation conditions of the flexible DC transmission relay protection (such as overcurrent protection setting values) are mapped to the phase plane space to obtain the boundaries of the relay protection device on each coordinate axis of the phase space. and Furthermore, the following formula is used to determine whether the system has a risk of oscillation.

[0056] ; Here, & represents the logical AND.

[0057] Based on this, if If the condition is met, the system phase trajectory is within a safe range and there is no risk of system oscillation; otherwise, if the phase trajectory exceeds the relay protection activation condition in the future, there is a risk of system oscillation.

[0058] Example 3 The method of Example 2 is used to assess the oscillation risk of a flexible DC transmission system.

[0059] like Figure 5 As shown, the verification example used is a flexible DC transmission system including an AC system and an AC filter. The flexible DC transmission system adopts a single-unit equivalent model and is connected to the 500kV AC system via two transformers: a 500 / 220kV transformer and a 220 / 110kV transformer. An AC filter is installed at the PCC point where the flexible DC transmission system connects to the AC system. In this example, the signal used for oscillation risk assessment is the current deviation signal between the d-axis and q-axis in the current control loop of the flexible DC converter. and .

[0060] Built on MATLAB / Simulink platform Figure 2 The time-domain simulation model of the flexible DC transmission system is shown. The operating conditions are set as follows: t When the current is between 1 and 1.01 s, a small disturbance signal with an amplitude of 1% is introduced into the current control loop of the flexible DC transmission system.

[0061] When selecting the proportional coefficient of the current control element of the flexible DC converter =1.54, the system phase trajectory estimated using the method of Example 2 is as follows Figure 3 As shown in (a) above, the light-colored portion represents the actual measured data, and the dark-colored portion represents the phase trajectory predicted based on the phase space reconstruction model. It can be seen that the system phase trajectory exhibits a convergent trend, indicating that the system has no oscillation risk under this operating condition. Verification was performed in the time-domain simulation model, and the system output power waveform is shown below. Figure 3 As shown in (b) in the figure, it also shows a convergence trend, which verifies the correctness of the oscillation risk assessment.

[0062] When taking the proportional coefficient of the current control circuit of the flexible DC converter =0.54, the system phase trajectory diagram estimated using the method of Example 2 is as follows. Figure 4 As shown in (a), the light-colored portion represents the actual measured data, and the dark-colored portion represents the phase trajectory predicted based on the phase space reconstruction model. It can be seen that the system phase trajectory exhibits a divergent trend, indicating that the system has an oscillation risk under this operating condition. Verification was performed in the time-domain simulation model, and the system output power waveform is as follows: Figure 4 As shown in (b), the trend also diverges, verifying the correctness of the oscillation risk assessment.

[0063] Example 4 The present invention provides a computer device, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the above-described method for assessing the oscillation risk of flexible DC transmission systems based on phase space reconstruction.

[0064] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0065] Example 5 The present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the steps of the above-described method for assessing the oscillation risk of flexible DC transmission systems based on phase space reconstruction.

[0066] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0067] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The systems, devices, and storage media disclosed in the embodiments are described simply because they correspond to the methods disclosed in the embodiments; relevant details can be found in the method section.

[0068] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0069] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for assessing oscillation risk of a flexible HVDC power transmission system based on phase space reconstruction, characterized in that, The method comprises the following steps: Collecting an oscillation output signal including an oscillation component when a flexible HVDC power transmission system is unstable, and preprocessing the oscillation output signal; Reconstructing a phase space trajectory of the preprocessed oscillation output signal by using a time delay embedding method; Based on the reconstructed phase space trajectory, predicting a phase space trajectory of an oscillation output signal of the power transmission system at the next time by using a linear autoregressive model as a prediction model or a prediction model representing geometric features of the phase space trajectory in polar coordinates; Analyzing an evolution trajectory of the predicted phase space trajectory to evaluate an oscillation risk of the power transmission system.

2. The phase space reconstruction based flexible HVDC system oscillation risk assessment method according to claim 1, characterized in that, The step of reconstructing the phase space trajectory of the preprocessed oscillation output signal by using the time delay embedding method comprises the following steps: Determining a delay time of phase space reconstruction by using an autocorrelation coefficient method and determining an embedding dimension of phase space reconstruction by using a false nearest neighbor method, and reconstructing the phase space trajectory of the preprocessed oscillation output signal based on Takens theorem.

3. The phase space reconstruction based flexible HVDC system oscillation risk assessment method according to claim 1, wherein, Based on the reconstructed phase space trajectory, predicting a phase space trajectory of an oscillation output signal of the power transmission system at the next time by using a linear autoregressive model as a prediction model, which comprises the following steps: Taking the phase variable of the reconstructed phase space as a state vector, taking minimization of one-step prediction error as an objective function, and estimating parameters of the linear autoregressive model by using a least square method; The linear auto-regressive model is: ; wherein, and are the embedding vectors of the phase space trajectory at the current time t and the next time t+1, respectively, is a state transition matrix, is a bias, is a noise; Objective function is: ; wherein, , is the number of effective samples, is the number of time steps observed continuously from the initial time, is the embedding dimension, is the delay time; Based on the solved state transition matrix and the real-time collected and preprocessed oscillation output signal, predicting the phase space trajectory of the oscillation output signal of the power transmission system at the next time.

4. The phase space reconstruction based flexible HVDC system oscillation risk assessment method according to claim 3, characterized in that, The predicted phase space trajectory of the oscillation output signal at the next time is represented as: ； wherein is an embedding vector of the phase space trajectory of the oscillatory output signal at the prediction time instant , , is the input time instant at which the damping strategy is executed for the power transmission system, and are a set number of steps and a time step, respectively, is an embedding vector of the phase space trajectory of the oscillatory output signal at the input time instant , is a bias term of the estimated linear auto-regressive model, is a state transition matrix of the estimated linear auto-regressive model, the superscripts i and denote the i th power and the th power of a matrix, respectively.

5. The phase space reconstruction based flexible direct current power transmission system oscillation risk assessment method according to claim 1, wherein, Based on the reconstructed phase space trajectory, predicting a phase space trajectory of an oscillation output signal of the power transmission system at the next time by using a prediction model representing geometric features of the phase space trajectory in polar coordinates, which comprises the following steps: Mapping the reconstructed phase space trajectory to a low-dimensional space; Taking minimization of one-step prediction error as an objective function, solving parameters of a discrete prediction model of a radial vector and a cross vector combination, and predicting the phase space trajectory of the oscillation output signal at the next time based on the real-time collected and preprocessed oscillation output signal; The discrete prediction model of the radial vector and the cross vector combination is: ； wherein, is the phase space trajectory radius, is the mapping of a point in the phase trajectory into a lower dimensional space, is the phase space trajectory phase, α , β , γ and ω 0 are four parameters characterizing the phase space trajectory of the oscillation signal of a flexible HVDC power transmission system; Objective function is: wherein, , is the number of effective samples, is the number of time steps observed continuously from the initial time, is the embedding dimension, is the delay time.

6. The phase space reconstruction based flexible direct current power transmission system oscillation risk assessment method according to claim 5, characterized in that, The predicted phase space trajectory of the oscillation output signal at the next time is represented as: ； wherein , , , , and are the radius and phase of the phase space trajectory of the oscillation output signal at the prediction time , , is the time of input of the damping strategy by the power transmission system, and are the set number of steps and time step, respectively, and are the radius and phase of the phase space trajectory of the oscillation output signal at the input time .

7. The phase space reconstruction based flexible direct current power transmission system oscillation risk assessment method according to claim 1, characterized in that, The step of evaluating the oscillation risk of the power transmission system comprises the following step: if the predicted phase space trajectory of the oscillation output signal exceeds a starting condition of a relay protection, the power transmission system has an oscillation risk, otherwise, the power transmission system does not have an oscillation risk.

8. The phase space reconstruction based flexible direct current power transmission system oscillation risk assessment method according to claim 1, characterized in that, The oscillation output signal including an oscillation component when the flexible HVDC power transmission system is unstable, the flexible HVDC power transmission system comprising a flexible HVDC converter, the oscillation output signal being: a current deviation signal of d-axis and q-axis in a current control loop of the flexible HVDC converter and .

9. A computer device, comprising: The system comprises a processor and a memory; when the processor executes a computer program stored in the memory, the steps of the method for evaluating an oscillation risk of a flexible HVDC power transmission system based on phase space reconstruction according to any one of claims 1-8 are implemented.

10. A computer-readable storage medium, characterized in that, The computer program is stored in the memory; when the computer program is executed by the processor, the steps of the method for evaluating an oscillation risk of a flexible HVDC power transmission system based on phase space reconstruction according to any one of claims 1-8 are implemented.

Images