New energy power system inertia online evaluation method, system and equipment under environmental disturbance measurement

Abstract

The embodiment of the invention provides a new energy power system inertia online evaluation method, system and device under environmental disturbance measurement, and the method comprises the steps: constructing an active-frequency dynamic model of a new energy power system under environmental disturbance, the system equivalent inertia of the new energy system as a to-be-evaluated parameter is embedded in the active-frequency dynamic model; collecting time sequence data of a frequency deviation value and an active power deviation value of a specific time window of the new energy power system in a steady-state operation environment; constructing a loss function based on the active-frequency dynamic model; carrying out iterative optimization on the loss function by adopting a self-adaptive gradient descent algorithm until a preset convergence condition is met, and determining the equivalent inertia of a target system through the loss function after iterative optimization; and determining a system inertia online evaluation value of the new energy power system based on the target system equivalent inertia.

Description

Technical Field

[0001] This invention relates to the field of power system operation status monitoring and safety assessment technology, and in particular to a method, system and equipment for online assessment of the inertia of a new energy power system under environmental disturbance measurement. Background Technology

[0002] A decline in system inertia severely threatens the frequency stability of the power grid. Frequency is a core indicator for measuring the instantaneous balance between power generation and consumption load, while inertia is the first line of defense for maintaining this balance and resisting disturbances. Insufficient inertia can lead to excessively large frequency change rates and excessively low or high frequency extremes when the system experiences power deficits or surpluses. This can easily trigger the activation of protection devices such as low-frequency load shedding or high-frequency generator tripping, and in severe cases, may lead to system collapse and large-scale blackouts. Post-incident analysis of major blackouts in recent years identified insufficient system inertia as one of the key factors contributing to the escalation of accidents and the rapid deterioration of frequency. Therefore, real-time and accurate online assessment of the inertia level of new energy power systems is a crucial prerequisite for achieving frequency security early warning and formulating effective prevention and control strategies, and has extremely significant practical implications.

[0003] To monitor system inertia levels in real time, researchers have developed various measurement-based inertia assessment methods using high-precision, high-refresh-rate synchronous phasor data provided by widely deployed phasor measurement units. Existing methods can be broadly categorized into three types: The first category is based on large disturbance events. This method relies on significant power surge events occurring in the system, such as large generator tripping or line faults. By measuring the system frequency change and active power imbalance after the event, and combining this with simplified rotor motion equations, the equivalent inertia of the system is directly calculated. The advantage of this type of method is its intuitive principle, but its fatal flaw lies in its reliance on uncontrollable and rare large system disturbances. It cannot be continuously monitored during normal steady-state operation of the system, resulting in severely limited practicality.

[0004] The second category is based on signal injection identification methods. To overcome the dependence on large disturbance events, this method actively injects specific, controllable disturbance signals (such as pseudo-random sequences, step signals, etc.) into the power system, and then identifies the inertia by analyzing the dynamic response characteristics of the system. Although this type of method supports online applications, the actively injected signals themselves may adversely affect power quality, system stability, and user equipment, thus limiting its application scope.

[0005] The third category is based on environmental disturbance measurement methods. This method is considered the most promising online assessment solution. It utilizes the small, naturally occurring disturbances that always exist in the power system during steady-state operation, such as random load fluctuations and natural changes in renewable energy output (collectively referred to as "environmental disturbances"). By continuously analyzing the frequency and active power response data under these environmental disturbances, non-intrusive and continuous monitoring of system inertia can be achieved.

[0006] However, existing inertia assessment methods based on environmental perturbations still have significant limitations. Some methods, such as those based on power spectral density analysis, covariance matrix, or Markov chain Monte Carlo methods, require a detailed model of the system and are subject to strict theoretical assumptions, making them difficult to apply in complex real-world systems. Other data-driven methods, such as autoregressive moving average models, reduce the requirements for mathematical models, but their performance heavily depends on the selection of hyperparameters such as the model order. During steady-state operation, frequency fluctuations are very small, and inappropriate hyperparameters can easily lead to rounding errors, numerical underflow, and underfitting, causing numerical stability problems, significantly increasing the error of the assessment results, and even causing the algorithm to diverge.

[0007] In summary, the following technical problems urgently need to be solved in the existing technology: (1) How to get rid of the dependence on large disturbance events or externally injected signals in the system and realize non-intrusive online evaluation; (2) How to reduce the dependence on accurate mathematical models and enhance the applicability of the method; (3) How to overcome the defects of hyperparameter sensitivity and poor numerical stability in data-driven methods and improve the accuracy and robustness of evaluation results. Summary of the Invention

[0008] To address the aforementioned technical problems, embodiments of the present invention provide a method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement, comprising: An active-frequency dynamic model of a new energy power system under environmental disturbances is constructed, wherein the active-frequency dynamic model is embedded with the system equivalent inertia of the new energy system as a parameter to be evaluated. Time-series data of frequency deviation and active power deviation values ​​of the new energy power system under steady-state operation environment are collected for a specific time window. A loss function is constructed based on the active-frequency dynamic model. The loss function is iteratively optimized using an adaptive gradient descent algorithm until a preset convergence condition is met. The equivalent inertia of the target system is then determined using the iteratively optimized loss function. The online evaluation value of the system inertia of the new energy power system is determined based on the equivalent inertia of the target system.

[0009] In one embodiment, constructing the active-frequency dynamic model of the new energy power system under environmental disturbances includes: Based on the rotor motion equations of the new energy power system and the Ornstein-Uhlenbeck stochastic process algorithm, an active-frequency dynamic model is constructed to describe the stochastic power fluctuations of the new energy power system under environmental disturbances.

[0010] In one embodiment, the expression for the active-frequency dynamic model includes: ; in, H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. The frequency deviation value representing the center point of the region's inertia; n Represents the total number of units in the entire region; The mean regression time constant represents OU The rate at which the process fluctuates to its average value; The diffusion coefficient is denoted as . Refers to the Wiener process.

[0011] In one embodiment, constructing the loss function based on the active-frequency dynamic model includes: The active-frequency dynamic model is discretized based on the forward Euler method to obtain the first... t The predicted frequency deviation value at any given time; The loss function is constructed by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point.

[0012] In one embodiment, the expression for the frequency prediction value includes: ; yes t The predicted frequency deviation at time [time]. H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. for t- The active power deviation input value at time 1.

[0013] In one embodiment, constructing the loss function by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point includes: loss function L Defined as: ; in, N The total number of data points within the time window. fort The true frequency deviation at any given time. yes t The predicted frequency deviation value at time.

[0014] In one embodiment, the step of iteratively optimizing the loss function using an adaptive gradient descent algorithm includes: iteratively optimizing the loss function using the AdaGrad algorithm, wherein the iterative optimization includes parameter initialization, gradient calculation, adaptive learning rate adjustment, parameter update, and convergence determination.

[0015] In one embodiment, the window length of the specific window is 30s-35s, and the sampling frequency is 100Hz-150Hz.

[0016] Another embodiment of the present invention also provides an online evaluation system for the inertia of a new energy power system under environmental disturbance measurement, comprising: The first construction module is used to construct an active-frequency dynamic model of a new energy power system under environmental disturbances. The active-frequency dynamic model embeds the system equivalent inertia of the new energy system as a parameter to be evaluated. The acquisition module is used to acquire time-series data of frequency deviation and active power deviation values ​​of the new energy power system under steady-state operation conditions within a specific time window. The second construction module is used to construct a loss function based on the active-frequency dynamic model; The first determining module is used to iteratively optimize the loss function using an adaptive gradient descent algorithm until a preset convergence condition is met, and then determine the equivalent inertia of the target system through the iteratively optimized loss function. The second determining module is used to determine the online evaluation value of the system inertia of the new energy power system based on the equivalent inertia of the target system.

[0017] Another embodiment of the present invention also provides an electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the online evaluation method for the inertia of new energy power systems under environmental disturbance measurement as described in any one of the above descriptions.

[0018] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0019] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the online evaluation method for the inertia of a new energy power system under environmental disturbance measurement in an embodiment of the present invention.

[0022] Figure 2 This is a flowchart illustrating the online evaluation method for the inertia of a new energy power system under environmental disturbance measurement, as described in another embodiment of the present invention.

[0023] Figure 3 This is a structural diagram of the improved IEEE-39 node system with new energy power generation in an embodiment of the present invention.

[0024] Figure 4 This is a structural block diagram of the online inertia assessment system for new energy power systems under environmental disturbance measurement, as described in an embodiment of the present invention. Detailed Implementation

[0025] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but these are not intended to limit the scope of the invention.

[0026] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the following description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope of this disclosure will be apparent to those skilled in the art.

[0027] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present disclosure and, together with the general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.

[0028] These and other features of the invention will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0029] It should also be understood that although the invention has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of the invention, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0030] The above and other aspects, features and advantages of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0031] Specific embodiments of the present disclosure are described thereafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure and can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the present disclosure. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but merely to serve as the basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of substantially any suitable detailed structures.

[0032] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in still another embodiment,” all of which may refer to one or more of the same or different embodiments according to this disclosure.

[0033] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0034] like Figure 1 As shown, this embodiment of the invention provides a method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement, including: S1: Construct an active-frequency dynamic model of a new energy power system under environmental disturbances, wherein the active-frequency dynamic model is embedded with the system equivalent inertia of the new energy system as a parameter to be evaluated; S2: Collect time-series data of frequency deviation and active power deviation values ​​of the new energy power system under steady-state operation environment at a specific time window; S3: Construct a loss function based on the active-frequency dynamic model; S4: The loss function is iteratively optimized using an adaptive gradient descent algorithm until the preset convergence condition is met. The equivalent inertia of the target system is then determined using the iteratively optimized loss function. S5: Determine the online evaluation value of the system inertia of the new energy power system based on the equivalent inertia of the target system.

[0035] The method in this embodiment is applicable to both unit-level and region-level inertia assessment. Unit-level assessment collects power and frequency information from the unit's ports; region-level assessment collects the frequency of the region's inertia center point and the region's net external exchange power. Applying this method, even when the measurement data contains 10% Gaussian white noise of signal energy, the relative error of the inertia assessment result can still be kept within 2%.

[0036] Based on the above, it can be seen that the solution in this embodiment achieves non-intrusive online evaluation without large disturbances or external signal injection. The adaptive gradient descent algorithm effectively solves the problems of hyperparameter sensitivity and poor numerical stability of traditional data-driven methods. While ensuring high accuracy, it has excellent noise robustness and high computational efficiency. It can be flexibly applied to real-time inertia evaluation at the unit level and regional level, providing a more reliable and practical technical means for frequency security early warning and prevention of new energy power systems.

[0037] In one embodiment, constructing the active-frequency dynamic model of the new energy power system under environmental disturbances includes: S101: An active-frequency dynamic model for describing the random power fluctuations of the new energy power system under environmental disturbances, constructed based on the rotor motion equation of the new energy power system and the Ornstein-Uhlenbeck stochastic process algorithm.

[0038] The expression for the active-frequency dynamic model includes: ; in, H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. The frequency deviation value representing the center point of the region's inertia; n Represents the total number of generating units in the entire region, including the number of synchronous motor and new energy generating units; The mean regression time constant represents OU The speed at which the process fluctuates back to its average value is typically tens or hundreds of seconds; The diffusion coefficient is denoted as . The Wiener process is usually represented by Gaussian white noise in the actual operation of power systems.

[0039] Furthermore, in this embodiment, the window length of the specific window is 30s-35s, and the sampling frequency is 100Hz-150Hz. Data acquisition can be performed, but is not limited to, through phasor measurement units deployed at the ports of new energy power system units or regional inertia centers.

[0040] Furthermore, such as Figure 2 As shown, the construction of the loss function based on the active-frequency dynamic model includes: S301: Discretize the active-frequency dynamic model based on the forward Euler method to obtain the first... t The predicted frequency deviation value at any given time; S302: Construct the loss function by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point.

[0041] In this embodiment, the loss function is the mean square error between the frequency prediction value output by the active-frequency dynamic model and the actual frequency measurement value collected, thereby transforming the system inertia assessment problem into finding a loss function that minimizes the error. L The optimization problem of minimizing the equivalent inertia of the system.

[0042] Specifically, the expression for the frequency prediction value includes: ; yes t The predicted frequency deviation at time [time]. H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. for t- The active power deviation input value at time 1.

[0043] The loss function is constructed by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point, including: loss function L Defined as: ; in, N The total number of data points within the time window. for t The true frequency deviation at any given time. yes t The predicted frequency deviation value at time.

[0044] In another embodiment, the step of iteratively optimizing the loss function using an adaptive gradient descent algorithm includes: S401: Iteratively optimizing the loss function using the AdaGrad algorithm, wherein the iterative optimization includes parameter initialization, gradient calculation, adaptive learning rate adjustment, parameter update, and convergence determination.

[0045] Specifically, 1) Parameter initialization: parameters to be estimated H (System equivalent inertia) and D (System overall equivalent damping coefficient), algorithm state variables including iteration counter k First-order moment estimation and Second-order moment estimation and And hyperparameters including the global learning rate First-order moment decay rate Second-order moment decay rate sum of numerical stability constants δ Perform initialization; 2) Gradient calculation: in the first... kIn the next iteration, the loss function is calculated. L For parameters H and D gradient and ; 3) Adaptive learning rate adjustment and parameter update: Updating the first-order moment estimate and the second-order moment estimate: ; ; Bias corrections are applied to the first and second moment estimates: ; Update parameters to be estimated: ; 4) Convergence Check: Repeat the gradient calculation and parameter update steps until the preset convergence condition is met. The convergence condition may be: the change in the loss function value is less than a minimum threshold, the number of iterations reaches the upper limit, or the norm of the parameter gradient is lower than a specific threshold.

[0046] 5) Output inertia evaluation results: After the algorithm converges, the optimal parameters H^ and D^ obtained in the current iteration are used as the final evaluation results. Among them, H^ is the equivalent inertia evaluation value of the system or unit under the current operating state.

[0047] To fully verify the effectiveness of the method proposed in this invention, the classic IEEE-39-bus system in power system research can be used as a test benchmark, and improvements can be made to simulate a high-proportion renewable energy integration scenario. The improvements are as follows: Figure 3 As shown: The original synchronous generators G3, G7, G8, and G9 are replaced with four wind farms of equal capacity, denoted as WF1, WF2, WF3, and WF4, respectively. All retained synchronous generators (G1, G2, G4, G5, G6, and G10) in the system are equipped with automatic voltage regulators. The system comprises 19 load nodes and 46 transmission lines. To analyze regional inertia, the system is divided into four interconnected regions, with region 4 represented by the equivalent synchronous machine G10, representing the external power grid. To simulate environmental disturbances in a real power grid, random fluctuations following an Ornstein-Uhlenbeck random process are superimposed on the active power of all 19 load nodes. OU The average regression time constant of the process θ Set to 20 seconds, diffusion coefficient σAdjustments were made to ensure that the system frequency fluctuations were limited to the dead zone of the synchronous generator governor (typically ±0.05 Hz), thereby ensuring the system operated in a steady state. The PMU sampling frequency was set to 100 Hz to provide high-precision input data for the method in this embodiment. Unless otherwise specified, the time window length used for inertia assessment was set to 30 seconds by default.

[0048] Specifically, the method of this embodiment includes: Step 1: Model building and data acquisition, including building a stochastic dynamic model of the system's active power-frequency, which uses inertia H and damping D as the core parameters to be identified.

[0049] Step 2: Parameter Initialization. The initial value of the parameter to be estimated, inertia H, is set to 40 seconds (close to the true value), and damping... D The initial value was set to 1.0 pu based on experience. The AdaGrad algorithm was used for iterative optimization of the model and loss function. The state variables of the AdaGrad algorithm included: iteration counter k = 0; first-order moment estimates and second-order moment estimates were both set to 0. Among the algorithm hyperparameters, the global learning rate was 0.1; the first-order moment decay rate was 0.9; the second-order moment decay rate was 0.999; and the numerical stability constant was... These hyperparameters use the common settings of the AdaGrad algorithm and do not require fine-tuning.

[0050] Step 3: Iterative Optimization Process. Discretize the continuous model using the forward Euler method to construct the loss function; the algorithm enters an iterative loop, performing gradient calculation and adaptive updates; calculate the loss function. L right H and D The gradient is calculated through backpropagation; adaptive updates are performed to update the first and second moments, and bias correction is applied. Finally, the parameters are updated using an adaptive learning rate. H and D This embodiment sets two convergence conditions: a maximum number of iterations of 5000, or a change in the loss function between two consecutive iterations of less than 100%. The iteration stops when any of the conditions is met.

[0051] Step 4: Output Results. After the iteration stops, the algorithm outputs the final optimized parameters H and D.

[0052] To verify the reliability of the proposed algorithm, different initial values ​​(20s, 40s, 60s, and 100s) were set, and the inertia evaluation values ​​after 5000 iterations were observed. Both the initial value and the number of iterations affect the inertia evaluation results. The closer the initial value is to the theoretical inertia value, the faster the evaluation results converge to near the theoretical true value. When the initial value is far from the theoretical inertia value, a sufficient number of iterations are required for convergence. Simultaneously, the inclusion of an adaptive learning rate mechanism avoids manual adjustment of hyperparameters, ensuring that the final inertia evaluation results converge to close to the theoretical true value, thus mitigating numerical stability issues caused by improper initial value settings. Data comparison shows that the evaluation data obtained by the method in this embodiment are close to the actual required values, fully demonstrating the robustness of the method in this embodiment.

[0053] like Figure 4 As shown, another embodiment of the present invention also provides an online evaluation system for the inertia of a new energy power system under environmental disturbance measurement, comprising: The first construction module is used to construct an active-frequency dynamic model of a new energy power system under environmental disturbances. The active-frequency dynamic model embeds the system equivalent inertia of the new energy system as a parameter to be evaluated. The acquisition module is used to acquire time-series data of frequency deviation and active power deviation values ​​of the new energy power system under steady-state operation conditions within a specific time window. The second construction module is used to construct a loss function based on the active-frequency dynamic model; The first determining module is used to iteratively optimize the loss function using an adaptive gradient descent algorithm until a preset convergence condition is met, and then determine the equivalent inertia of the target system through the iteratively optimized loss function. The second determining module is used to determine the online evaluation value of the system inertia of the new energy power system based on the equivalent inertia of the target system.

[0054] In one embodiment, constructing the active-frequency dynamic model of the new energy power system under environmental disturbances includes: Based on the rotor motion equations of the new energy power system and the Ornstein-Uhlenbeck stochastic process algorithm, an active-frequency dynamic model is constructed to describe the stochastic power fluctuations of the new energy power system under environmental disturbances.

[0055] In one embodiment, the expression for the active-frequency dynamic model includes: ; in, H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. The frequency deviation value representing the center point of the region's inertia; n Represents the total number of units in the entire region; The mean regression time constant represents OU The rate at which the process fluctuates to its average value; The diffusion coefficient is denoted as . Refers to the Wiener process.

[0056] In one embodiment, constructing the loss function based on the active-frequency dynamic model includes: The active-frequency dynamic model is discretized based on the forward Euler method to obtain the first... t The predicted frequency deviation value at any given time; The loss function is constructed by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point.

[0057] In one embodiment, the expression for the frequency prediction value includes: ; yes t The predicted frequency deviation at time [time]. H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. for t- The active power deviation input value at time 1.

[0058] In one embodiment, constructing the loss function by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point includes: loss function L Defined as: ; in, N The total number of data points within the time window. for t The true frequency deviation at any given time. yes t The predicted frequency deviation value at time.

[0059] In one embodiment, the step of iteratively optimizing the loss function using an adaptive gradient descent algorithm includes: iteratively optimizing the loss function using the AdaGrad algorithm, wherein the iterative optimization includes parameter initialization, gradient calculation, adaptive learning rate adjustment, parameter update, and convergence determination.

[0060] In one embodiment, the window length of the specific window is 30s-35s, and the sampling frequency is 100Hz-150Hz.

[0061] Another embodiment of the present invention also provides an electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the online evaluation method for the inertia of new energy power systems under environmental disturbance measurement as described in any one of the above descriptions.

[0062] Furthermore, one embodiment of the present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the online evaluation method for the inertia of a new energy power system under environmental disturbance measurement as described above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above-described method embodiments, and will not be repeated here.

[0063] Furthermore, embodiments of the present invention also provide a computer program product, which is tangibly stored on a computer-readable medium and includes computer-readable instructions, which, when executed, cause at least one processor to perform an online evaluation method for the inertia of a new energy power system under environmental disturbance measurement as described in the embodiments above.

[0064] It should be noted that the computer storage medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access storage medium (RAM), a read-only storage medium (ROM), an erasable programmable read-only storage medium (EPROM or flash memory), an optical fiber, a portable compact disk read-only storage medium (CD-ROM), an optical storage medium, a magnetic storage medium, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program configured for use by or in connection with an instruction execution system, system, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, antenna, optical fiber, RF, etc., or any suitable combination thereof.

[0065] Furthermore, those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0066] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0067] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0068] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of protection of this application is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this application as described above, which are not provided in detail for the sake of brevity.

Claims

1. A method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement, characterized in that, include: An active-frequency dynamic model of a new energy power system under environmental disturbances is constructed, wherein the active-frequency dynamic model is embedded with the system equivalent inertia of the new energy system as a parameter to be evaluated. Time-series data of frequency deviation and active power deviation values ​​of the new energy power system under steady-state operation environment are collected for a specific time window. A loss function is constructed based on the active-frequency dynamic model. The loss function is iteratively optimized using an adaptive gradient descent algorithm until a preset convergence condition is met. The equivalent inertia of the target system is then determined using the iteratively optimized loss function. The online evaluation value of the system inertia of the new energy power system is determined based on the equivalent inertia of the target system.

2. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 1, characterized in that, The active-frequency dynamic model of the new energy power system under environmental disturbances includes: Based on the rotor motion equations of the new energy power system and the Ornstein-Uhlenbeck stochastic process algorithm, an active-frequency dynamic model is constructed to describe the stochastic power fluctuations of the new energy power system under environmental disturbances.

3. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 2, characterized in that, The expression for the active-frequency dynamic model includes: ； in, H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. The frequency deviation value representing the center point of the region's inertia; n Represents the total number of units in the entire region; The mean regression time constant represents OU The rate at which the process fluctuates to its average value; The diffusion coefficient is denoted as . Refers to the Wiener process.

4. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 1, characterized in that, The loss function constructed based on the active-frequency dynamic model includes: The active-frequency dynamic model is discretized based on the forward Euler method to obtain the first... t The predicted frequency deviation value at any given time; The loss function is constructed by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point.

5. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 4, characterized in that, The expression for the frequency prediction value includes: ； yes t The predicted frequency deviation at time [time]. H The system's overall equivalent inertia. D The system's overall equivalent damping coefficient. for t- The active power deviation input value at time 1.

6. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 4, characterized in that, The loss function is constructed by combining the predicted frequency deviation value at each time point with the actual frequency deviation value at the corresponding time point, including: loss function L Defined as: ； in, N The total number of data points within the time window. for t The true frequency deviation at any given time. yes t The predicted frequency deviation value at time.

7. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 4, characterized in that, The step of iteratively optimizing the loss function using the adaptive gradient descent algorithm includes: iteratively optimizing the loss function using the AdaGrad algorithm, wherein the iterative optimization includes parameter initialization, gradient calculation, adaptive learning rate adjustment, parameter update, and convergence determination.

8. The method for online evaluation of the inertia of a new energy power system under environmental disturbance measurement according to claim 1, characterized in that, The window length of the specific window is 30s-35s, and the sampling frequency is 100Hz-150Hz.

9. An online evaluation system for the inertia of a new energy power system under environmental disturbance measurement, characterized in that, include: The first construction module is used to construct an active-frequency dynamic model of a new energy power system under environmental disturbances. The active-frequency dynamic model embeds the system equivalent inertia of the new energy system as a parameter to be evaluated. The acquisition module is used to acquire time-series data of frequency deviation and active power deviation values ​​of the new energy power system under steady-state operation conditions within a specific time window. The second construction module is used to construct a loss function based on the active-frequency dynamic model; The first determining module is used to iteratively optimize the loss function using an adaptive gradient descent algorithm until a preset convergence condition is met, and then determine the equivalent inertia of the target system through the iteratively optimized loss function. The second determining module is used to determine the online evaluation value of the system inertia of the new energy power system based on the equivalent inertia of the target system.

10. An electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the online evaluation method for the inertia of new energy power systems under environmental disturbance measurement as described in any one of claims 1-8.

Images