Short-circuit parameter measurement optimization method suitable for multiple scenes

Abstract

The invention relates to the technical field of power system automation, and discloses a short-circuit parameter measurement optimization method suitable for multiple scenes, which comprises the following steps: acquiring electric data and environmental parameters of a power system, and carrying out standardized preprocessing; constructing a scene feature vector to represent the current system operation state based on the preprocessed data, establishing a basic measurement model for calculating the short-circuit impedance, and evaluating the precision of an initial measurement model; configuring model parameters through an optimization algorithm, and designing a fitness function to guide an optimization process; scene feature changes are monitored in real time, a collaborative optimization objective function is constructed based on scene switching judgment, and measurement model parameters are dynamically adjusted based on a gradient parameter updating method; and outputting an optimized short-circuit parameter measurement result. According to the method, through dynamic scene modeling and intelligent closed-loop optimization, the influence of environmental interference and working condition change on measurement can be eliminated, and the core problems that a traditional short-circuit parameter measurement method is insufficient in precision and poor in adaptability in a complex scene are solved.

Description

Technical Field

[0001] This invention relates to the field of power system automation technology, and in particular to a method for optimizing short-circuit parameter measurement that is adaptable to multiple scenarios. Background Technology

[0002] With the rapid development of power systems, accurate measurement of short-circuit parameters is crucial for the safe operation of these systems. Traditional short-circuit parameter measurement methods mainly rely on test data from single scenarios, and their measurement accuracy is often difficult to guarantee when facing different operating conditions. Furthermore, while existing measurement methods have a certain degree of adaptability, they are prone to significant errors in complex environments or extreme operating conditions, making it difficult to meet the needs of multi-scenario applications. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention is proposed.

[0004] Therefore, this invention provides an optimized method for short-circuit parameter measurement that is adaptable to multiple scenarios, which can solve the problems of poor scenario adaptability, low measurement accuracy, and insufficient versatility in short-circuit parameter measurement.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a short-circuit parameter measurement optimization method adaptable to multiple scenarios, comprising: collecting electrical data and environmental parameters of the power system and performing standardized preprocessing; constructing a scenario feature vector based on the preprocessed data to characterize the current system operating state, establishing a basic measurement model for calculating short-circuit impedance, and evaluating the accuracy of the initial measurement model using a weighted root mean square error method; configuring model parameters through an optimization algorithm, designing a fitness function to guide the optimization process, and adjusting the weights of model accuracy error and scenario adaptability in the optimization objective through a balance coefficient; monitoring changes in scenario characteristics in real time, constructing a collaborative optimization objective function based on scenario switching judgment, and dynamically adjusting the measurement model parameters based on a gradient-based parameter update method; quantitatively evaluating the performance of the measurement method, and outputting the optimized short-circuit parameter measurement results.

[0006] As a preferred embodiment of the short-circuit parameter measurement optimization method adapted to multiple scenarios described in this invention, the construction of the scenario feature vector includes balancing the importance of different features through a feature weight matrix and controlling the sensitivity of feature mapping by combining a scenario adjustment factor.

[0007] The system state vector is transformed into a scene feature space using a feature mapping function.

[0008] As a preferred embodiment of the short-circuit parameter measurement optimization method adapted to multiple scenarios described in this invention, the selection of the measurement model includes establishing a short-circuit impedance calculation equation based on electrical measurement principles;

[0009] The measurement errors of voltage and current are compensated by introducing correction factors;

[0010] Measurement accuracy is assessed using weighted root mean square error.

[0011] As a preferred embodiment of the short-circuit parameter measurement optimization method adapted to multiple scenarios described in this invention, the step of configuring model parameters through optimization algorithms includes adopting a scenario adaptation mechanism and quantitatively evaluating scenario characteristic deviations through an exponential decay function.

[0012] Based on scene feature deviations, scene adaptability indicators are generated to guide parameter adjustments.

[0013] As a preferred embodiment of the short-circuit parameter measurement optimization method adaptable to multiple scenarios described in this invention, the optimization algorithm is an improved particle swarm optimization algorithm, which achieves the optimal configuration of measurement parameters through an intelligent search strategy.

[0014] The particle position update equation is:

[0015]

[0016] in, and Let represent the position vectors of the i-th particle at iterations k+1 and k, respectively; to guide the particle's search direction, the velocity update equation is defined as:

[0017]

[0018] in, and Let be the velocity vectors of the i-th particle at iterations k+1 and k, respectively; ω be the inertia weight; and c1 and c2 be the learning factors, which adjust the particle's trajectory toward its optimal position P. i The degree of learning with the global optimal position G, where r1 and r2 are random numbers in the interval [0,1].

[0019] As a preferred embodiment of the short-circuit parameter measurement optimization method adaptable to multiple scenarios described in this invention, the improved particle swarm optimization algorithm includes further optimization of measurement accuracy and adaptability through a fitness function, specifically,

[0020] F = min{α1ε + α2(1-A)}

[0021] In this context, F is the fitness function, and α1 and α2 are balance coefficients used to weigh the importance of measurement error ε and scene adaptability index A.

[0022] As a preferred embodiment of the short-circuit parameter measurement optimization method adaptable to multiple scenarios described in this invention, the real-time monitoring of scene feature changes includes, through the collaborative design of scene feature analysis and optimization objectives, determining when the scene feature vector change meets the scene switching judgment criteria.

[0023] D = ||S t -S t-1 ||>θ

[0024] Among them, S t and S t-1 Let be the scene feature vectors at the current time and the previous time, respectively; θ be the scene switching judgment threshold; and ||·|| denote the vector norm. Based on this judgment criterion, a collaborative optimization objective function is constructed:

[0025]

[0026] Where, μ i Let ε be the weight coefficient for the i-th scene. i Let λ represent the measurement error in the corresponding scenario, λ be the weight of the constraint term, and c be the weight of the constraint term. j Let be the j-th constraint, u be the number of scenarios, and m be the number of constraints.

[0027] As a preferred embodiment of the short-circuit parameter measurement optimization method adapted to multiple scenarios described in this invention, the dynamic adjustment of measurement model parameters is achieved by employing a gradient-based parameter update method. This method dynamically updates the parameter configuration by real-time monitoring of the measurement status and constructs a basic equation for parameter update based on deep learning theory.

[0028]

[0029] Among them, P new and P old These are the parameter values ​​before and after the update, respectively, where η is the learning rate. Let L(P) be the gradient of the loss function with respect to the parameter P; the dynamic adjustment formula for the learning rate is:

[0030] η t =η0(1+δT) -β

[0031] Where η0 is the initial learning rate, δ is the time decay coefficient, T is the number of iterations, and β is the decay exponent, used to adjust the nonlinear decay characteristics.

[0032] It should be noted that the measurement method's performance characteristics are comprehensively reflected from three dimensions: measurement accuracy, relative accuracy, and scenario adaptability, providing a reliable evaluation basis for method optimization and practical application. In practical applications, these three indicators can be weighted and combined according to specific needs to form a comprehensive evaluation index, further guiding the improvement and optimization of the measurement method. Through this multi-dimensional evaluation mechanism, not only can the overall performance of the measurement method be accurately reflected, but also aspects requiring key improvement can be identified, providing directional guidance for the continuous optimization of the measurement method.

[0033] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a short-circuit parameter measurement optimization method adapted to multiple scenarios.

[0034] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a short-circuit parameter measurement optimization method adapted to multiple scenarios.

[0035] The beneficial effects of this invention are as follows: By integrating scene feature modeling and dynamic parameter adjustment mechanisms, it effectively overcomes the accuracy limitations of traditional methods under complex operating conditions; it achieves stable and reliable measurement results in diverse scenarios such as steady-state, transient, and extreme environments, significantly reducing system errors. Based on scene deviation quantification evaluation and real-time feedback mechanisms, it achieves millisecond-level dynamic parameter tuning; it greatly improves the response speed and adaptability to sudden changes in the operating environment (such as load fluctuations and fault recovery). Through multi-dimensional collaborative optimization strategies, it meets real-time constraints while ensuring measurement accuracy; the algorithm has strong robustness and can be directly applied to key scenarios such as smart grid monitoring and fault diagnosis. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of a short-circuit parameter measurement optimization method adapted to multiple scenarios, provided as an embodiment of the present invention. Detailed Implementation

[0038] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0039] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a short-circuit parameter measurement optimization method adaptable to multiple scenarios, including:

[0040] S1: Collect electrical data and environmental parameters of the power system, perform standardized preprocessing, and construct scene feature vectors.

[0041] S2: Based on the scene feature vector constructed from the preprocessed data, select the measurement model and configure the model parameters through the optimization algorithm.

[0042] S3: Real-time monitoring of scene feature changes and dynamic adjustment of measurement model parameters.

[0043] S4: Output the optimized short-circuit parameter measurement results.

[0044] It should be noted that by constructing a multi-dimensional scenario feature model and a dynamic closed-loop optimization mechanism, the core defects of traditional short-circuit parameter measurement methods, such as poor scenario adaptability, sensitivity to environmental interference, response lag, and a single evaluation system, are fundamentally solved.

[0045] Its advantages are: breaking through the limitations of traditional single working conditions, achieving accurate measurement from steady state to extreme working conditions through real-time perception of scene feature vectors and exponential decay adaptive mechanism; integrating gradient-driven parameter updates and improved particle swarm optimization algorithm to dynamically compensate for nonlinear errors caused by environmental factors such as temperature and humidity; automatically triggering model switching based on quantified scene deviation to eliminate delays in manual intervention and ensure real-time performance in key scenarios such as fault recovery; creating a unique multi-scene collaborative objective function to coordinate measurement accuracy, computational efficiency, and engineering constraints, achieving a balance between theoretical performance and practical application; and establishing a three-dimensional evaluation system for accuracy / precision / scene adaptability to provide closed-loop feedback for algorithm optimization.

[0046] Example 2 is an embodiment of the present invention, which provides a short-circuit parameter measurement optimization method that is adaptable to multiple scenarios. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0047] Furthermore, in this embodiment of the application, step S1 collects electrical data and environmental parameters of the power system, performs standardized preprocessing, and constructs a scene feature vector. Specific steps include A1-A2:

[0048] A1: To achieve accurate measurement of short-circuit parameters in multiple scenarios, a mathematical description of the system's operating state is constructed. The system operating state vector is defined as:

[0049] S={V i ,I i Z i ,θ i ,P i}

[0050] In the formula, V i I represents the voltage value at the i-th node, reflecting the system voltage level. i Z represents the current value of the i-th line, characterizing the system load state; i θ represents the short-circuit impedance value, indicating the system impedance characteristics. i P is a phase angle parameter that describes the phase relationship. i These are scene characteristic parameters, including environmental factors such as temperature and humidity.

[0051] A2: Based on the state vector, establish a scene feature description function:

[0052] F s =f(S,W,α)

[0053] Where W is the feature weight matrix, used to balance the importance of different features; α is the scene adjustment factor, controlling the sensitivity of feature mapping; and f is the feature mapping function, realizing the transformation from the state vector to the scene feature space. This modeling method fully considers the multidimensional characteristics of system operation, laying the foundation for subsequent scene adaptability analysis.

[0054] Furthermore, in this embodiment, step S2 selects a measurement model based on the scene feature vector constructed from the preprocessed data and configures the model parameters through an optimization algorithm. Specific steps include B1-B3:

[0055] B1: Based on scene feature modeling, a measurement model for short-circuit parameters is designed. This model achieves high-precision parameter measurement through reasonable mathematical modeling and error compensation. Based on electrical measurement principles, the basic equations for parameter measurement are:

[0056] Z sc =g(V m ,I m ,φ)+j·h(V m ,I m ,β)

[0057] In the formula, Z sc V is the short-circuit impedance to be measured. m and I mφ and β are the measured voltage and current values, respectively; φ and β are correction coefficients used to compensate for measurement errors; g and h are the measurement functions for the real and imaginary parts, respectively, describing the amplitude and phase angle characteristics of the impedance.

[0058] B2: To assess measurement accuracy, a weighted root mean square error is introduced.

[0059]

[0060] Among them, Z i This is the actual short-circuit impedance value (Ω). For the measured impedance value (Ω), w i Here, represents the weighting coefficient, and n represents the number of measurement samples. This measurement model achieves accurate measurement of short-circuit parameters through function mapping and error evaluation.

[0061] In this embodiment, the measurement accuracy is evaluated by weighted root mean square error;

[0062] In an alternative embodiment, measurement accuracy can also be assessed using a weighted average absolute error (ARE). Specifically, the absolute difference between the actual impedance value and the model measurement value for each measurement sample is calculated. Weighting coefficients are applied to these absolute differences to calculate the weighted average absolute error for all samples. This value directly reflects the average absolute magnitude of the deviation between the measured result and the actual value. The evaluation criterion is that the smaller the average value, the higher the measurement accuracy.

[0063] In another alternative embodiment, measurement accuracy can be evaluated through interval-based average relative error statistics. Specifically, the measurement samples are pre-divided into several intervals based on the actual impedance values. Within each interval, the average relative error of all sample measurements relative to the actual values ​​is calculated. The final evaluation result consists of the average relative error values ​​of each interval. The evaluation criteria are whether the model's accuracy performance is consistent across different impedance intervals and whether the average relative error of each interval is sufficiently small.

[0064] B3: To improve the accuracy of short-circuit parameter measurements, an optimization model based on an improved particle swarm optimization algorithm is proposed. This model achieves optimal parameter configuration through an intelligent search strategy. The particle position update equation is:

[0065]

[0066] in, and Let represent the position vectors of the i-th particle at iterations k+1 and k, respectively, reflecting the optimized state of the measurement parameters. To guide the particle's search direction, the velocity update equation is defined as:

[0067]

[0068] In the formula, and Let be the velocity vectors of the i-th particle at iterations k+1 and k, respectively; ω is the inertia weight, controlling the degree to which the particle maintains its original velocity; c1 and c2 are learning factors, adjusting the particle's trajectory towards its optimal position P. i The degree of learning from the global optimal position G; r1 and r2 are random numbers in the interval [0,1]. The fitness function for the optimization process is designed as follows:

[0069] F = min{α1ε + α2(1-A)}

[0070] Among them, α1 and α2 are balance coefficients used to weigh the importance of measurement error ε and scenario adaptability index A, so as to achieve comprehensive optimization of measurement accuracy and adaptability.

[0071] Furthermore, in this embodiment, step S3 involves real-time monitoring of scene feature changes and dynamic adjustment of measurement model parameters, specifically including steps C1-C4:

[0072] C1: To improve the scene adaptability of the measurement method, a scene adaptation mechanism based on exponential decay is proposed. This mechanism achieves intelligent adaptation of the measurement method to different operating environments by quantitatively evaluating scene characteristics. The evaluation metrics are:

[0073]

[0074] In the formula, λ i γ represents the weighting coefficients for different scenarios, reflecting the importance of each scenario; γ is the decay factor, controlling the rate at which adaptability decays with deviation; d i , where represents the scene deviation, characterizing the degree of difference between the current scene and the standard scene; n represents the number of scene features.

[0075] C2: Scene deviation is calculated using normalization:

[0076]

[0077] Among them, S i S represents the feature value of the current scene. ref The difference between the reference scene feature values ​​and the normalized value reflects the degree of scene change. This adaptive mechanism provides a basis for adaptive adjustment of measurement parameters by quantitatively assessing scene differences, ensuring the reliability of the measurement method under different scenarios.

[0078] In this embodiment, a scene adaptation mechanism is adopted, which quantitatively assesses scene feature deviation by combining the normalized feature difference of the exponential decay function with the dynamic adjustment of the weight of the decay factor.

[0079] In an optional embodiment, scene feature deviation can also be assessed using a fixed threshold interval grading method. Specifically, a fixed threshold interval is predefined for each scene feature, and the absolute difference between the current scene feature value and the reference value is directly compared with these preset thresholds to determine the deviation level of the feature. The final "scene deviation" is represented by a discrete level identifier to indicate the degree of deviation for each feature. The overall scene deviation can be comprehensively evaluated by statistically analyzing the deviation level scores of each feature.

[0080] In another optional embodiment, scene feature deviation can also be assessed using a simple weighted absolute percentage difference. Specifically, the absolute percentage difference between the current value and the reference value of each scene feature is directly calculated, and a weighted average of the absolute percentage differences of all features is calculated based on pre-set static weight coefficients that reflect the importance of each feature. This weighted average serves as the final quantitative indicator of "scene deviation".

[0081] C3: During short-circuit parameter measurement, dynamic changes in system operating status and environmental conditions can affect measurement accuracy. By monitoring the measurement status in real time and dynamically updating parameter configurations, the accuracy and reliability of the measurement are ensured. Based on deep learning theory, the basic equation for parameter updating is constructed:

[0082]

[0083] In the formula, P new and P old These are the parameter values ​​before and after the update, respectively; η is the learning rate. Let L(P) be the gradient of the loss function with respect to the parameter P. The dynamic adjustment equation for the learning rate is:

[0084] η t =η0(1+δt) -β

[0085] Where η0 is the initial learning rate; δ is the time decay coefficient, controlling the rate at which the learning rate decays over time; t is the number of iterations; and β is the decay exponent, used to adjust the nonlinear decay characteristics. This adaptive adjustment mechanism can automatically adjust parameters according to the dynamic characteristics of the measurement process, improving the robustness of the measurement.

[0086] In this embodiment, the gradient-based parameter update method dynamically adjusts the measurement model parameters;

[0087] In an optional embodiment, dynamic adjustment of the measurement model parameters can also be achieved using gradient descent with a fixed learning rate. Specifically, a fixed learning rate is used for parameter updates. In each iteration, the gradient of the loss function with respect to the current parameters is directly calculated, and this gradient is multiplied by a preset, constant learning rate coefficient. This product is used to adjust the parameter values. The learning rate needs to be set empirically and remains constant throughout the entire measurement parameter update process.

[0088] In an optional embodiment, dynamic adjustment of the measurement model parameters can also be achieved through a parameter smoothing update method based on exponential moving averages. Specifically, each time new measurement data or status information is obtained, a proposed parameter value is calculated based on the new data. The current actual parameter value is then weighted and averaged with this new proposed value, with the weights controlled by a fixed smoothing factor. The smoothing factor determines the degree of influence of the new proposed value on the final update result, and the final updated parameters are the result of this weighted average.

[0089] C4: The complexity of the power system operating environment requires measurement methods to be adaptable to multiple scenarios. Through scenario feature analysis and the collaborative design of optimization objectives, comprehensive optimization of the measurement method under different scenarios was achieved. First, a judgment criterion for scenario switching was designed:

[0090] D = ||S t -S t-1 ||>θ

[0091] In the formula, S t and S t-1 Let be the scene feature vectors at the current time and the previous time, respectively; θ be the scene switching judgment threshold; and ||·|| represent the vector norm. Based on this judgment criterion, a collaborative optimization objective function is constructed:

[0092]

[0093] Where, μ i Let ε be the weight coefficient for the i-th scene. i Let λ represent the measurement error in the corresponding scenario, λ be the weight of the constraint term, and c be the weight of the constraint term. j Let be the j-th constraint, u be the number of scenarios, and m be the number of constraints.

[0094] This collaborative optimization strategy can ensure measurement accuracy in various scenarios while meeting the constraints of practical engineering applications.

[0095] Furthermore, in this embodiment, step S4 outputs the optimized short-circuit parameter measurement results, specifically including steps D1-D3:

[0096] D1: To comprehensively evaluate the performance of short-circuit parameter measurement methods, a multi-dimensional evaluation system was constructed to quantitatively assess the performance of the measurement methods from different perspectives, providing a scientific basis for method improvement and optimization. First, the root mean square error index is introduced:

[0097]

[0098] In the formula, E rms The root mean square error (Ω) reflects the overall accuracy level of the measurement; Z i Let be the actual short-circuit impedance value (Ω) at the i-th measurement point; This corresponds to the measured value (Ω). This index provides a quantitative assessment of measurement accuracy by calculating the root mean square of the difference between the measured value and the actual value.

[0099] D2: To further evaluate the relative accuracy of the measurement, an accuracy index is introduced:

[0100]

[0101] Where R represents the measurement accuracy (%); e i Let be the absolute error (Ω) at the i-th measurement point, calculated as follows: |Z i | represents the absolute value of the actual impedance. This indicator normalizes the measurement error to the actual value, facilitating comparison of measurement results of different magnitudes.

[0102] D3: Considering the scene adaptability characteristics of the measurement method, design scene adaptability evaluation indicators:

[0103]

[0104] In the formula, A s κ is the scene adaptability index, ranging from [0,1], with a larger value indicating better adaptability; κ is the scene sensitivity coefficient, used to adjust the sensitivity of the adaptability assessment; d i The deviation of the i-th scene feature reflects the degree of difference between the current scene and the standard scene. This indicator achieves a non-linear quantitative assessment of scene adaptability through the properties of an exponential function.

[0105] In this embodiment, by designing a scene adaptability index, the scene feature deviation is transformed into an adaptability score in the [0,1] interval using the characteristics of the exponential function, thereby achieving a quantitative evaluation of the performance of the measurement method.

[0106] In an optional embodiment, the performance of the measurement method can also be quantitatively evaluated through a hierarchical accuracy evaluation. Specifically, the basic accuracy layer calculates the absolute error distribution between the measurement result and the actual value, replacing the absolute value evaluation of the root mean square error; the relative accuracy layer statistically analyzes the maximum relative error percentage within different impedance ranges, replacing the overall accuracy index.

[0107] In another optional embodiment, the performance of the measurement method can be quantitatively evaluated through dynamic weighted comprehensive scoring. Specifically, weight coefficients are assigned to different measurement points according to the importance of the short circuit location, and the weight ratio of accuracy and adaptability indicators is adaptively adjusted. The accuracy score is calculated based on the proportion of samples with measurement errors within a preset safety threshold. The adaptability score is graded according to the number of iterations required for parameter adjustment after scene switching and the accuracy recovery speed. The accuracy and adaptability scores are combined according to dynamic weights to generate a performance index of 0-100 points.

[0108] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0109] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that:

[0110] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0112] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0113] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0114] Example 4, an embodiment of the present invention, provides a short-circuit parameter measurement optimization method adaptable to multiple scenarios. Simulation verification was performed using MATLAB R2023a, based on operational data from a 330kV substation from 2022 to 2023. The data acquisition process strictly followed power system measurement specifications. The sampling frequency was set to 10kHz, and basic electrical quantities such as voltage and current, as well as relevant system state parameters, were recorded at each measurement point. To ensure data representativeness, data under different operating conditions were collected, including steady-state data during normal operation, transient data during load changes, dynamic data during system faults and recovery processes, and special operating condition data under extreme weather conditions.

[0115] The experimental design employed a comparative experiment method, selecting three representative existing methods as the control group. The traditional method uses the classic short-circuit parameter measurement method; Improved Method 1 and Improved Method 2 represent two mainstream improved methods, both introducing different optimization mechanisms based on the traditional method.

[0116] Performance evaluation employs a multi-dimensional indicator system. RMSE reflects the absolute accuracy of the measurement; relative error characterizes the relative accuracy of the measurement; scene adaptability assessment method demonstrates its adaptability to environmental changes; and computation time measurement method measures real-time performance. Furthermore, key indicators such as measurement error, switching speed, settling time, and accuracy recovery rate are recorded in detail for performance under different scenarios. The measurement results are shown in the table below:

[0117] Table 1: Comparison of measurement performance of different methods

[0118] method RMSE(Ω) Relative error (%) Scene adaptability Calculation time (ms) Traditional Law 0.856 3.25 0.756 45 Improved Method 1 0.634 2.48 0.823 62 Improved Method 2 0.512 2.15 0.867 78 This patent 0.385 1.64 0.912 85

[0119] Table 2: Measurement Error Analysis under Different Scenarios

[0120] Scene type Traditional method (%) Improved method 1 (%) Improved method 2 (%) This patent (%) Normal operation 2.85 2.23 1.95 1.45 Load mutation 4.12 3.56 3.12 2.28 Fault recovery 3.95 3.25 2.86 2.15 Extreme working conditions 5.23 4.45 3.85 2.95

[0121] Table 3: Scene Adaptability Analysis

[0122] Adaptation indicators Traditional Law Improved Method 1 Improved Method 2 This patent Switching speed (ms) 85 72 65 48 Settling time (ms) 156 134 112 95 Accurate recovery rate (%) 92.5 94.8 96.2 98.5

[0123] Experimental results demonstrate that this patented method exhibits significant advantages in measurement accuracy and scenario adaptability. It improves the RMSE (Real-Time Error) index by 55% compared to traditional methods, reducing the relative error to 1.64%. Particularly under extreme conditions, the measurement error is controlled within 3%, outperforming existing methods. Scene switching time is reduced to 48ms, stabilization recovery time is shortened to 95ms, and the accuracy recovery rate reaches 98.5%, fully validating the effectiveness and practical value of the method. The entire testing process includes four main stages: data acquisition, model training, performance testing, and result analysis. The measurement response time is less than 0.1 seconds, meeting the needs of practical engineering applications.

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing short-circuit parameter measurement in multiple scenarios, characterized in that: include, Collect electrical data and environmental parameters from the power system and perform standardized preprocessing; Based on the preprocessed data, a scenario feature vector is constructed to represent the current system operating state. A basic measurement model for calculating short-circuit impedance is established, and the accuracy of the initial measurement model is evaluated using the weighted root mean square error method. By optimizing the algorithm to configure model parameters, designing a fitness function to guide the optimization process, and using a balancing coefficient to adjust the weights of model accuracy error and scene adaptability in the optimization objective; Real-time monitoring of scene feature changes, judgment based on scene switching, construction of collaborative optimization objective function, and dynamic adjustment of measurement model parameters based on gradient-based parameter update method; The performance of the measurement method is quantitatively evaluated, and the optimized short-circuit parameter measurement results are output.

2. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 1, characterized in that: The construction of the scene feature vector includes balancing the importance of different features through a feature weight matrix and controlling the sensitivity of feature mapping by combining a scene adjustment factor. The system state vector is transformed into a scene feature space using a feature mapping function.

3. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 2, characterized in that: The selected measurement model includes establishing a short-circuit impedance calculation equation based on electrical measurement principles; The measurement errors of voltage and current are compensated by introducing correction factors; Measurement accuracy is assessed using weighted root mean square error.

4. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 3, characterized in that: The optimization algorithm configuration of model parameters includes adopting a scene adaptation mechanism and quantitatively evaluating scene feature deviations through an exponential decay function. Based on scene feature deviations, scene adaptability indicators are generated to guide parameter adjustments.

5. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 4, characterized in that: The optimization algorithm is an improved particle swarm optimization algorithm, which achieves the optimal configuration of measurement parameters through an intelligent search strategy; The particle position update equation is: in, and Let represent the position vectors of the i-th particle at iterations k+1 and k, respectively; to guide the particle's search direction, the velocity update equation is defined as: in, and Let be the velocity vectors of the i-th particle at iterations k+1 and k, respectively; ω be the inertia weight; and c1 and c2 be the learning factors, which adjust the particle's trajectory toward its optimal position P. i The degree of learning with the global optimal position G, where r1 and r2 are random numbers in the interval [0,1].

6. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 5, characterized in that: The improved particle swarm optimization algorithm includes further optimization of measurement accuracy and adaptability through a fitness function, specifically, F = min{α1ε + α2(1-A)} In this context, F is the fitness function, and α1 and α2 are balance coefficients used to weigh the importance of measurement error ε and scene adaptability index A.

7. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 6, characterized in that: The real-time monitoring of scene feature changes includes, through scene feature analysis and the collaborative design of optimization targets, determining when changes in the scene feature vector meet the criteria for scene switching. D=‖S t -S t-1 ‖>θ Among them, S t and S t-1 Let be the scene feature vectors at the current time and the previous time, respectively; θ be the scene switching judgment threshold; and ||·|| denote the vector norm. Based on this judgment criterion, a collaborative optimization objective function is constructed: Where, μ i Let ε be the weight coefficient for the i-th scene. i Let λ represent the measurement error in the corresponding scenario, λ be the weight of the constraint term, and c be the weight of the constraint term. j Let be the j-th constraint, u be the number of scenarios, and m be the number of constraints.

8. The short-circuit parameter measurement optimization method adaptable to multiple scenarios as described in claim 7, characterized in that: The dynamic adjustment of measurement model parameters employs a gradient-based parameter update method. This method involves real-time monitoring of the measurement status and dynamic updates to the parameter configuration. Based on deep learning theory, a fundamental equation for parameter update is constructed. Among them, P new and P old These are the parameter values ​​before and after the update, respectively, where η is the learning rate. Let L(P) be the gradient of the loss function with respect to the parameter P; the dynamic adjustment formula for the learning rate is: or t =η0(1+δT) -β Where η0 is the initial learning rate, δ is the time decay coefficient, T is the number of iterations, and β is the decay exponent, used to adjust the nonlinear decay characteristics.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the short-circuit parameter measurement optimization method adapted to multiple scenarios as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the short-circuit parameter measurement optimization method adapted to multiple scenarios as described in any one of claims 1 to 7.

Images