Multi-mode seamless switching control method of PCS energy storage converter and related equipment

By collecting and analyzing the operating status parameters of the PCS energy storage converter in real time, identifying potential switching modes and performing transient stability analysis, the problems of lag and misjudgment in the switching decision in the existing PCS control strategy are solved, achieving seamless switching of multiple modes and improving the system's operational reliability and flexibility.

CN120855406AInactive Publication Date: 2025-10-28SHENZHEN SYD NETWORK TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202511098684.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-08-06
Publication Date
2025-10-28
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing PCS control strategies lack the ability to dynamically perceive and analyze real-time operating status, resulting in delayed or misjudged mode switching decisions, making it difficult to assess stability risks during the switching process, and limiting the application potential of energy storage systems in high-dynamic demand scenarios.

Method used

By collecting the operating status parameters of the PCS energy storage converter in real time, identifying potential switching modes based on the mode switching trigger conditions, performing transient stability analysis, generating transient response characteristic curves, and prioritizing these curves, seamless multi-mode switching control is achieved.

Benefits of technology

This improves the reliability and flexibility of the PCS energy storage converter under complex operating conditions, reduces energy disturbances and equipment stress during switching, and ensures the continuity and reliability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120855406A_ABST
    Figure CN120855406A_ABST
Patent Text Reader

Abstract

The invention relates to a multi-mode seamless switching control method of a PCS energy storage converter and related equipment, and the method comprises the following steps: collecting the current operation state parameters of the PCS energy storage converter; obtaining a potential switching mode corresponding to a mode switching trigger condition based on a matching relationship between the current operation state parameter and the preset mode switching trigger condition; performing transient stability analysis on the PCS energy storage converter based on the potential switching modes to obtain a transient response characteristic curve of each potential switching mode; performing priority ranking on the potential switching modes based on the transient response characteristic curve to obtain a switching priority sequence; and performing multi-mode seamless switching control on the PCS energy storage converter on the basis of the switching priority sequence, thereby solving the technical problem that in a multi-mode PCS control strategy, dynamic sensing and analysis capability for a real-time operation state is lacked, so that a switching decision is lagged or misjudged.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of PCS energy storage converter technology, and in particular to a multi-mode seamless switching control method and related equipment for PCS energy storage converter. Background Art

[0002] Currently, research on multi-mode PCS control strategies commonly employs fixed logical judgments or experience-based settings for mode switching, lacking the ability to dynamically perceive and analyze real-time operating conditions. This leads to delayed or misjudged switching decisions. Furthermore, existing methods often overlook the differences in transient responses under different mode switching paths, making it difficult to effectively assess stability risks during the switching process. This can potentially trigger system oscillations or even protective shutdowns, limiting the application potential of energy storage systems in high-dynamic demand scenarios.

[0003] Therefore, there is an urgent need for a multi-mode seamless switching control method that integrates real-time operational status monitoring, potential switching mode identification, transient stability assessment, and optimized sequencing. This method should not only possess rapid response capabilities and good dynamic adaptability, but also predict system behavior before switching to ensure a smooth and disturbance-free switching process. This would improve the reliability and flexibility of the PCS energy storage converter under complex operating conditions, meeting the urgent demand of new power systems for high-performance control of energy storage devices. Summary of the Invention

[0004] The main objective of this invention is to provide a multi-mode seamless switching control method for PCS energy storage converters, which solves the technical problem that the lack of dynamic perception and analysis capability of real-time operating status in multi-mode PCS control strategies leads to delayed or misjudged switching decisions.

[0005] To achieve the above objectives, the present invention provides a multi-mode seamless switching control method for a PCS energy storage converter, comprising the following steps: The current operating status parameters of the PCS energy storage converter are collected in real time through preset electrical sensors. Based on the matching relationship between the current operating status parameters and the preset mode switching trigger conditions, the potential switching modes corresponding to the mode switching trigger conditions are obtained. Transient stability analysis of the PCS energy storage converter is performed based on the potential switching modes to obtain the transient response characteristic curves of each potential switching mode. Based on the transient response characteristic curve, the potential switching modes are prioritized to obtain a switching priority sequence; Based on the switching priority sequence, the PCS energy storage converter is subjected to multi-mode seamless switching control.

[0006] Furthermore, the step of obtaining the potential switching mode corresponding to the mode switching trigger condition based on the matching relationship between the current operating state parameters and the preset mode switching trigger condition includes: Detect the threshold boundary of the current operating status parameters and identify a subset of abnormal parameters that exceed the set threshold; Each parameter type in the subset of abnormal parameters is matched with the trigger rule field in the preset mode switching trigger condition to obtain multiple mode switching trigger condition entries corresponding to the subset of abnormal parameters. The multiple mode switching trigger condition entries are associated with a predefined operating mode mapping table to obtain one or more operating mode identifiers corresponding to each trigger condition entry. Based on the operating mode identifiers, the potential switching modes corresponding to the mode switching trigger conditions are determined. The potential switching modes include at least four different functional types of operating modes: grid-connected inverter mode, off-grid V / f control mode, constant power charging mode, and reactive power support mode.

[0007] Furthermore, the transient stability analysis of the PCS energy storage converter based on the potential switching modes is performed to obtain the transient response characteristic curves of each potential switching mode, including: Small-signal disturbance injection is performed on the PCS energy storage converter based on the potential switching modes to obtain voltage, current and power fluctuation data under each potential switching mode. Frequency domain characteristic analysis is then performed based on the voltage, current and power fluctuation data to obtain the frequency domain response characteristics of each mode. By using state-space modeling, the dynamic characteristics of the PCS energy storage converter under each potential switching mode are simulated, and the transient response of each potential switching mode is analyzed in the time domain based on the dynamic characteristics to obtain the time domain response characteristic parameters. Based on the frequency domain response characteristics and the time domain response characteristic parameters, a transient stability evaluation matrix for each potential switching mode is constructed, and a multi-objective optimization calculation is performed on the transient stability evaluation matrix to obtain a comprehensive stability index. The comprehensive stability index is normalized to obtain a normalized stability index, and transient response characteristic curves of each potential handover mode are generated based on the normalized stability index. The transient response characteristic curves reflect the dynamic performance and stability of each mode during the handover process.

[0008] Furthermore, the step of prioritizing the potential handover modes based on the transient response characteristic curve to obtain a handover priority sequence includes: Numerical integration is performed on the transient response characteristic curve to obtain the energy fluctuation curve of each potential switching mode, and the energy loss of the PCS energy storage converter in each potential switching mode is calculated based on the energy fluctuation curve to obtain the energy loss matrix. The transient response characteristic curves are aligned in the time domain using a dynamic time warping algorithm to obtain a standardized mode response sequence. The mode switching cost matrix between each potential switching mode is then calculated based on the standardized mode response sequence. The mode switching cost matrix quantifies the ease or difficulty of switching between different modes. The energy loss matrix and the mode switching cost matrix are weighted and fused to obtain a comprehensive switching evaluation matrix, and a mode switching topology map is constructed based on the comprehensive switching evaluation matrix, wherein the mode switching topology map represents the switching relationship between each potential switching mode; Based on the mode switching topology map, the potential switching modes are prioritized to obtain switching paths. The feasibility of the switching paths is verified and conflicts are detected and eliminated to obtain a switching priority sequence.

[0009] Furthermore, the calculation of the mode switching cost matrix between potential handover modes based on the standardized mode response sequence includes: Extract the multidimensional response feature vectors of each potential switching mode from the standardized mode response sequence, and perform principal component analysis on the multidimensional response feature vectors to obtain the principal component loading matrix of each mode. The multidimensional response feature vector is reduced in dimensionality by the principal component loading matrix to obtain the principal component score vector of each potential switching mode. Based on the principal component score vector, feature space mapping is performed to obtain the projection coordinate points of each mode in the feature space. The Mahalanobis distance between the projected coordinate points is calculated to obtain the Mahalanobis distance matrix between each potential switching mode. The mode switching cost is quantified based on the Mahalanobis distance matrix to obtain the mode switching cost matrix.

[0010] Furthermore, the step of prioritizing the potential switching modes based on the mode switching topology map to obtain the switching path includes: Graph theory analysis is performed on the mode switching topology to obtain the adjacency matrix of the mode switching topology, and the connectivity score of each potential switching mode is calculated based on the adjacency matrix. The mode switching topology graph is divided into communities using spectral clustering to obtain the mode switching community structure. Based on the mode switching community structure, the modularity coefficient of each potential switching mode is estimated to obtain the mode clustering degree. A multi-criteria decision analysis is performed on the connectivity score and pattern clustering to obtain a comprehensive handover priority index. Based on the comprehensive handover priority index, the potential handover patterns are prioritized to obtain an initial handover priority sequence. Based on the initial switching priority sequence, a multi-source shortest path search is performed on the mode switching topology to obtain the switching path.

[0011] The present invention also provides a multi-mode seamless switching control device for a PCS energy storage converter, comprising: The acquisition module is used to acquire the current operating status parameters of the PCS energy storage converter in real time through preset electrical sensors; The matching module is used to obtain the potential switching mode corresponding to the mode switching trigger condition based on the matching relationship between the current running status parameters and the preset mode switching trigger condition; The analysis module is used to perform transient stability analysis on the PCS energy storage converter based on the potential switching modes, and obtain the transient response characteristic curves of each potential switching mode. The sorting module is used to prioritize the potential switching modes based on the transient response characteristic curve to obtain a switching priority sequence. The control module is used to perform multi-mode seamless switching control on the PCS energy storage converter based on the switching priority sequence.

[0012] The present invention also 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 any of the methods described above.

[0013] The present invention also 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 any of the methods described above.

[0014] This invention provides a multi-mode seamless switching control method for a PCS energy storage converter, comprising the following steps: acquiring current operating state parameters of the PCS energy storage converter; obtaining potential switching modes corresponding to the mode switching trigger conditions based on the matching relationship between the current operating state parameters and preset mode switching trigger conditions; performing transient stability analysis on the PCS energy storage converter based on the potential switching modes to obtain transient response characteristic curves for each potential switching mode; prioritizing the potential switching modes based on the transient response characteristic curves to obtain a switching priority sequence; and performing multi-mode seamless switching control on the PCS energy storage converter based on the switching priority sequence. This method solves the technical problem of lacking dynamic perception and analysis capabilities of real-time operating status in multi-mode PCS control strategies, leading to delayed or misjudged switching decisions. It achieves the technical effect of prioritizing potential switching modes based on transient response characteristics, ensuring the priority execution of switching paths with higher stability and smoother transitions, significantly reducing energy disturbances and equipment stress during switching, and improving the continuity and reliability of system operation. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the steps of a multi-mode seamless switching control method for a PCS energy storage converter in one embodiment of the present invention; Figure 2 This is a structural block diagram of a multi-mode seamless switching control device for a PCS energy storage converter in one embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0016] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] like Figure 1 As shown, Figure 1 This invention provides a multi-mode seamless switching control method for a PCS energy storage converter, comprising the following steps: Step S1: Real-time acquisition of the current operating status parameters of the PCS energy storage converter through preset electrical sensors.

[0019] Specifically, in this scheme, the current operating status parameters of the PCS energy storage converter are collected in real time through preset electrical sensors. This process is the foundation for subsequent mode recognition and switching control. Specifically, during the operation of the PCS energy storage converter, the system relies on pre-deployed electrical sensor devices such as voltage sensors, current sensors, and frequency detection modules to continuously monitor key operating parameters such as DC-side voltage, AC-side current, power output, frequency fluctuations, and temperature changes, and transmits this data to the control system in real time. Through continuous collection and analysis of these operating status parameters, the system can dynamically grasp the current operating state of the PCS and compare it with preset multiple mode switching trigger conditions, thus providing a basis for identifying potential switching modes. For example, in scenarios where the grid frequency shifts or load changes, after the system collects data on frequency drop and power fluctuations through sensors, it can determine whether the trigger conditions for switching from grid-connected mode to islanded mode are met, and then proceed to the next step of transient stability assessment to ensure that the switching action is both timely and safe. This step not only enhances the system's perception capabilities but also provides reliable data support for subsequent intelligent decision-making.

[0020] Step S2: Based on the matching relationship between the current running state parameters and the preset mode switching trigger conditions, obtain the potential switching mode corresponding to the mode switching trigger conditions.

[0021] Specifically, in this step, the system analyzes the matching relationship between the current operating status parameters of the PCS energy storage converter, which are collected in real time by preset electrical sensors, and various preset mode switching trigger conditions. This determines whether the current system meets the activation conditions for a certain mode switching and thus obtains the potential switching mode corresponding to that trigger condition. In practice, the control system compares the collected operating status parameters such as voltage, current, frequency, and power with various mode switching thresholds stored in the control logic in real time. When one or more parameters reach and exceed a preset trigger threshold, the current system state is considered to match the mode switching trigger condition, thereby identifying the potential switching mode corresponding to that trigger condition. For example, when the grid frequency continuously decreases and falls below a set safety threshold, the system determines that it is in an abnormal operating condition. In this case, the grid-connected operation mode may no longer be applicable, and the control system will identify "islanding operation mode" as one of the potential switching modes, providing a decision-making basis for subsequent transient stability analysis. This process achieves dynamic perception of the operating status and intelligent identification of mode switching, ensuring that the switching strategy can adapt to complex and ever-changing actual operating environments.

[0022] Step S3: Perform transient stability analysis on the PCS energy storage converter based on the potential switching modes to obtain the transient response characteristic curves of each potential switching mode.

[0023] Specifically, in this step, based on the previously identified potential switching modes, the system performs stability analysis on the transient behavior of the PCS energy storage converter during mode switching, thereby obtaining transient response characteristic curves for each potential switching mode. Specifically, a mathematical model and simulation logic for switching between different operating modes are pre-established in the control system. When a potential switching mode is identified, the system calls the corresponding dynamic model, using the current operating state parameters as initial conditions, to simulate the trends of key variables such as voltage, current, and power changing over time during the switching process, and plots the corresponding transient response characteristic curves accordingly. For example, in the scenario where a drop in grid frequency triggers a switch from grid-connected mode to islanded mode, the system simulates AC-side voltage fluctuations, DC-side power oscillations, and the system frequency recovery process at the moment of switching, generating response curves reflecting system stability to assess whether the switching path will lead to system instability or equipment overload. Through this analysis process, the system can predict the dynamic behavior before actually executing the switching operation, providing a quantitative basis for subsequent mode priority ranking, thereby improving the safety and reliability of switching control.

[0024] Step S4: Based on the transient response characteristic curve, prioritize the potential switching modes to obtain a switching priority sequence.

[0025] Specifically, in this step, the system quantitatively evaluates the dynamic stability of different switching modes during the switching process based on the transient response characteristic curves of each potential switching mode obtained above, and prioritizes these potential switching modes accordingly, thereby generating a switching priority sequence with practical control significance. In practice, the control system comprehensively evaluates the impact of each switching mode on the operational stability of the PCS energy storage converter based on key indicators in each transient response characteristic curve, such as overshoot, recovery time, oscillation amplitude, and steady-state error, and sorts the modes according to the principle of higher stability and smaller disturbance. For example, in application scenarios where a decrease in grid frequency necessitates a switch from grid-connected mode to islanded mode, if multiple islanded switching paths exist, the system can identify the switching mode with the smoothest transition by comparing the differences in voltage fluctuation amplitude and recovery time under each path, and assign it a higher switching priority. This process ensures that, when multiple modes are available, the system can prioritize the switching path with the least impact on the operating state and the most stable transient process, providing a scientific decision-making basis for subsequent seamless multi-mode switching control.

[0026] Step S5: Based on the switching priority sequence, perform multi-mode seamless switching control on the PCS energy storage converter.

[0027] Specifically, in this step, the system, based on the aforementioned generated switching priority sequence, performs seamless switching control between multiple modes of the PCS energy storage converter in descending order of priority, ultimately achieving a stable operating state after the switch. Specifically, after receiving the switching command, the control system dynamically adjusts the control parameters, reference signals, and power flow strategy of the PCS energy storage converter according to the optimal or second-best switching path determined in the switching priority sequence. This ensures that key variables such as voltage, current, and power can smoothly transition when switching between different operating modes, avoiding drastic fluctuations or system instability. For example, in application scenarios where a drop in grid frequency triggers a switch from grid-connected mode to islanded mode, the system will prioritize the islanded switching path, which offers higher stability and smoother transient response. During the switching process, closed-loop control is used to adjust the inverter output voltage and frequency in real time, quickly matching them with local load characteristics to achieve seamless switching and ultimately enter a new stable operating state. This process not only improves the dynamic adaptability of the PCS under complex operating conditions but also significantly enhances the continuous power supply capability and operational reliability of the energy storage system under abnormal grid conditions.

[0028] In a specific embodiment, obtaining the potential switching mode corresponding to the mode switching trigger condition based on the matching relationship between the current operating state parameters and the preset mode switching trigger condition includes: Detect the threshold boundary of the current operating status parameters and identify a subset of abnormal parameters that exceed the set threshold; Each parameter type in the subset of abnormal parameters is matched with the trigger rule field in the preset mode switching trigger condition to obtain multiple mode switching trigger condition entries corresponding to the subset of abnormal parameters. The multiple mode switching trigger condition entries are associated with a predefined operating mode mapping table to obtain one or more operating mode identifiers corresponding to each trigger condition entry, and potential switching modes are determined based on the operating mode identifiers. The potential switching modes include at least four different functional types of operating modes: grid-connected inverter mode, off-grid V / f control mode, constant power charging mode, and reactive power support mode.

[0029] Specifically, in this step, the system further identifies a set of potential switching modes corresponding to the current abnormal operating condition by deeply analyzing the matching relationship between the previously collected current operating status parameters and the preset mode switching trigger conditions. The specific implementation process includes three key steps: First, the threshold boundaries in the current operating status parameters are detected to identify whether there is a subset of abnormal parameters that exceed the set safety range; second, the types of these abnormal parameters are semantically matched with the trigger rule fields in the preset mode switching trigger conditions to obtain a set of conditions that may trigger switching; finally, these trigger condition entries are correlated with a predefined operating mode mapping table to obtain the operating mode identifier corresponding to each trigger condition, and one or more potential switching modes are generated accordingly. In actual operation, the control system continuously receives data streams from preset electrical sensors and monitors the changes in the values ​​of key parameters such as voltage, current, frequency, and power in real time. When one or more parameters exceed their set safety thresholds, such as when the grid frequency is below 49.5Hz or the DC side voltage is above the rated value by 10%, the system will classify these parameters as part of the abnormal parameter subset. Subsequently, the system compares the field names of these abnormal parameters (such as "frequency" and "voltage") with the trigger rule fields stored in the mode switching trigger condition database within the control logic. This comparison is based not only on the consistency of field names but also on the semantic consistency of parameter types, units, and trigger logic, thereby ensuring the accuracy and applicability of the matching results. After completing the field-level matching, the system obtains a set of mode switching trigger condition entries corresponding to the current subset of abnormal parameters. For example, in an application scenario where the grid frequency is continuously decreasing, the system may identify the trigger condition entry "frequency exceeding limit". Next, the system uses this entry as an index to query a predefined operating mode mapping table, which records the mapping relationship between each trigger condition entry and one or more operating modes. For example, "frequency exceeding limit" may correspond to "off-grid V / f control mode" and "reactive power support mode", indicating that the system currently has a need to switch from grid-connected operation to islanded operation, while also possessing the ability to provide grid voltage support. Ultimately, based on the aforementioned mapping relationship, the system generates a set of potential switching modes. These modes include at least four different functional operating modes: grid-connected inverter mode, off-grid V / f control mode, constant power charging mode, and reactive power support mode. This process realizes a closed-loop logic construction from raw data to mode decision-making, enabling the PCS energy storage converter to intelligently identify suitable switching targets based on real-time status in complex and ever-changing operating environments. This provides an accurate input basis for subsequent transient stability assessment and priority ranking, thereby significantly improving the system's adaptability and operational safety.

[0030] In a specific embodiment, the transient stability analysis of the PCS energy storage converter based on the potential switching modes, to obtain the transient response characteristic curves of each potential switching mode, includes: Small-signal disturbance injection is performed on the PCS energy storage converter based on the potential switching modes to obtain voltage, current and power fluctuation data under each potential switching mode. Frequency domain characteristic analysis is then performed based on the voltage, current and power fluctuation data to obtain the frequency domain response characteristics of each mode. By using state-space modeling, the dynamic characteristics of the PCS energy storage converter under each potential switching mode are simulated, and the transient response of each potential switching mode is analyzed in the time domain based on the dynamic characteristics to obtain the time domain response characteristic parameters. Based on the frequency domain response characteristics and the time domain response characteristic parameters, a transient stability evaluation matrix for each potential switching mode is constructed, and a multi-objective optimization calculation is performed on the transient stability evaluation matrix to obtain a comprehensive stability index. The comprehensive stability index is normalized to obtain a normalized stability index, and transient response characteristic curves of each potential handover mode are generated based on the normalized stability index. The transient response characteristic curves reflect the dynamic performance and stability of each mode during the handover process.

[0031] Specifically, in this step, based on the previously identified potential switching modes, the system conducts an in-depth analysis of the transient processes that may be triggered when the PCS energy storage converter switches between different operating modes, and generates transient response characteristic curves for each switching mode accordingly. This process aims to evaluate the dynamic stability performance of the system under different switching paths, providing a scientific basis for subsequent priority ranking. Specifically, this analysis process includes four main steps: small-signal disturbance injection and fluctuation data acquisition, frequency domain characteristic analysis, state-space modeling and time-domain response parameter extraction, stability evaluation matrix construction and multi-objective optimization calculation, and finally, through normalization processing, obtains transient response characteristic curves reflecting the switching performance of each mode. First, after identifying the potential switching modes, the system will perform small-signal disturbance injection operations for each potential switching mode (such as grid-connected inverter mode, off-grid V / f control mode, constant power charging mode, and reactive power support mode). This operation simulates the small disturbances that may occur during actual operation, such as slight fluctuations in grid frequency or sudden load changes, thereby stimulating the system's dynamic response at the moment of switching. Subsequently, the system collects voltage, current, and power variation data under each switching mode, recording their fluctuation characteristics after a disturbance occurs. This raw data forms the basis for subsequent frequency and time domain analysis. Next, the system performs frequency domain characteristic analysis on the collected fluctuation data, using methods such as Fast Fourier Transform (FFT) to convert the time-domain signal into a frequency-domain signal, thereby obtaining the frequency domain response characteristics under each potential switching mode. This characteristic spectrum reflects key information such as the system's oscillation frequency distribution, harmonic content, and damping characteristics under different switching modes. For example, during the switch from grid-connected mode to off-grid V / f control mode, the system may exhibit low-frequency oscillations. Frequency domain analysis can accurately identify the oscillation frequency and its amplitude, providing a basis for judging system stability. Simultaneously, the system also simulates the dynamic behavior under each potential switching mode using state-space modeling. A state-space model is a mathematical tool that can describe the changes of internal state variables of a system over time, suitable for the dynamic analysis of nonlinear, multi-input multi-output systems. Based on this, the system can acquire time-domain response characteristic parameters during the switching process of each mode, such as rise time, overshoot, settling time, and steady-state error. These parameters directly reflect the dynamic transition performance and stability level during the switching process. Furthermore, the system integrates the aforementioned frequency-domain response characteristics with the time-domain response characteristic parameters to construct a comprehensive transient stability evaluation matrix. This matrix uses each potential switching mode as a row and various stability indicators as columns, forming a multi-dimensional evaluation system. Subsequently, the system performs multi-objective optimization calculations on this matrix, comprehensively considering multiple factors such as stability, response speed, and energy loss during the switching process, to derive a comprehensive stability index reflecting overall stability. The higher this index, the smoother and safer the switching mode is during the dynamic process.Finally, to facilitate comparison of stability differences between different switching modes, the system normalizes the comprehensive stability index of all potential switching modes, unifying it within a standardized range (e.g., 0~1), thereby generating corresponding transient response characteristic curves. These curves not only visually demonstrate the dynamic performance changes of each mode during the switching process but also provide a quantitative basis for subsequent switching priority ranking. For example, in an application scenario where the grid frequency is continuously decreasing, the system identifies "off-grid V / f control mode" and "reactive power support mode" as two potential switching modes. Through small-signal disturbance injection, frequency domain, and time domain analysis of these two modes, the system finds that the "off-grid V / f control mode" exhibits faster voltage recovery and smaller oscillation amplitude during switching, resulting in a higher comprehensive stability index; while the "reactive power support mode" shows some low-frequency oscillation, leading to a relatively lower stability index. Therefore, on the transient response characteristic curve, the curve of the "off-grid V / f control mode" is smoother and more stable, indicating that it is more suitable for the switching requirements under the current operating conditions. Through this analysis process, the system can accurately assess the stability performance of each mode before actual switching, thereby making more scientific and safer switching decisions.

[0032] In a specific embodiment, prioritizing the potential switching modes based on the transient response characteristic curve to obtain a switching priority sequence includes: Numerical integration is performed on the transient response characteristic curve to obtain the energy fluctuation curve of each potential switching mode, and the energy loss of the PCS energy storage converter in each potential switching mode is calculated based on the energy fluctuation curve to obtain the energy loss matrix. The transient response characteristic curves are aligned in the time domain using a dynamic time warping algorithm to obtain a standardized mode response sequence. The mode switching cost matrix between each potential switching mode is then calculated based on the standardized mode response sequence. The mode switching cost matrix quantifies the ease or difficulty of switching between different modes. The energy loss matrix and the mode switching cost matrix are weighted and fused to obtain a comprehensive switching evaluation matrix, and a mode switching topology map is constructed based on the comprehensive switching evaluation matrix, wherein the mode switching topology map represents the switching relationship between each potential switching mode; Based on the mode switching topology map, the potential switching modes are prioritized to obtain a set of switching paths. The feasibility of the switching path set is verified and conflict detection and elimination are performed to obtain a switching priority sequence.

[0033] Specifically, in this step, the system further prioritizes these modes based on the transient response characteristic curves of each potential switching mode generated above, ultimately forming a switching priority sequence with practical control significance. This process includes several key calculation and analysis steps: first, the energy fluctuations of the transient response characteristic curves are quantified; second, the switching cost relationship between different modes is extracted using a dynamic time warping algorithm; third, energy loss and switching cost are fused and evaluated to construct a mode switching topology map; and finally, the generation of the switching path set and conflict detection are completed on this basis, thereby obtaining the final switching priority sequence. In the specific implementation process, the system first performs numerical integration on each transient response characteristic curve to obtain the energy fluctuation curve of each potential switching mode during the switching process. By integrating the changing trends of variables such as voltage, current, and power at the instant of switching, the system can obtain the total energy change experienced by the PCS energy storage converter during the switching period under each mode, and construct an energy loss matrix accordingly. This matrix records the energy loss that may occur when switching from the current operating mode to each potential switching mode, serving as an important indicator for measuring the economy and stability of the switching process. Meanwhile, to more accurately compare the dynamic behavior differences between different switching modes, the system employs the Dynamic Time Warping (DTW) algorithm to align the transient response characteristic curves in the time domain. Since response curves under different modes may exhibit inconsistencies on the time axis (e.g., one mode recovers quickly while another recovers slowly), direct comparison would introduce errors. Therefore, the DTW algorithm effectively eliminates the influence of time offset, thus obtaining standardized mode response sequences. Based on these standardized sequences, the system further calculates the similarity distance between each potential switching mode, i.e., the mode switching cost matrix, to characterize the ease or difficulty of switching between different modes—for example, switching from grid-connected inverter mode to off-grid V / f control mode has a lower cost, while switching to constant power charging mode may involve significant control logic adjustments, resulting in a higher switching cost. Next, the system weightedly fuses the two matrices—the energy loss matrix and the mode switching cost matrix—to generate a comprehensive switching evaluation matrix. This matrix comprehensively considers energy consumption and control complexity during the switching process, ensuring that the ranking results both meet the system's stability requirements and reflect the feasibility of actual operation. Based on this comprehensive evaluation matrix, the system further constructs a mode switching topology graph. This graph uses nodes to represent potential switching modes and edges to represent the switching cost and energy loss level between two nodes, thus intuitively displaying the relationship structure between all potential switching modes and the advantages and disadvantages of switching paths. Subsequently, the system performs graph traversal algorithm analysis based on this mode switching topology graph to identify all possible switching path sets, and verifies the feasibility of these paths and detects and eliminates conflicts.For example, in application scenarios where a drop in grid frequency triggers a switch from grid-connected mode to islanded mode, although multiple switching paths exist, some paths may be unfeasible due to control parameter conflicts, equipment protection mechanism limitations, or excessive energy loss. The system needs to eliminate infeasible paths and retain the truly feasible and optimal switching scheme. Ultimately, based on the comprehensive switching evaluation results and the set of feasible switching paths after conflict detection, the system prioritizes each potential switching mode, generating a final switching priority sequence. This sequence not only reflects the stability and energy consumption characteristics of each mode during the switching process but also embodies the implementation difficulty and safety margin of its switching path, providing a scientific decision-making basis for subsequent seamless multi-mode switching control. For instance, in the aforementioned application scenario, the "off-grid V / f control mode" is given higher priority due to its lower energy loss and lower switching cost, becoming the preferred switching target; while the "reactive power support mode," although providing some grid support capability, is ranked second best due to its higher switching cost and less stable transient response. Through this process, the system achieves complete closed-loop control from multi-mode recognition to intelligent sequencing, significantly improving the PCS energy storage converter's adaptability and switching safety in complex operating environments.

[0034] In a specific embodiment, calculating the mode switching cost matrix between potential switching modes based on the standardized mode response sequence includes: Extract the multidimensional response feature vectors of each potential switching mode from the standardized mode response sequence, and perform principal component analysis on the multidimensional response feature vectors to obtain the principal component loading matrix of each mode. The multidimensional response feature vector is reduced in dimensionality by the principal component loading matrix to obtain the principal component score vector of each potential switching mode. Based on the principal component score vector, feature space mapping is performed to obtain the projection coordinate points of each mode in the feature space. The Mahalanobis distance between the projected coordinate points is calculated to obtain the Mahalanobis distance matrix between each potential switching mode. The mode switching cost is quantified based on the Mahalanobis distance matrix to obtain the mode switching cost matrix.

[0035] Specifically, in this step, the system further calculates the similarity distance between potential switching modes using the standardized mode response sequences obtained above, and constructs a mode switching cost matrix to measure the ease of switching between different operating modes. This process involves several key mathematical processing steps: first, multidimensional feature extraction and principal component analysis are performed on the standardized response sequences; second, feature space mapping is performed on the dimensionality-reduced principal component score vectors; then, Mahalanobis distance is used to calculate the statistical differences between modes; finally, the cost of different switching paths is quantified based on this distance matrix, thereby generating a mode switching cost matrix with practical control significance. Specifically, after completing the time-domain alignment of the transient response characteristic curves using the Dynamic Time Warping (DTW) algorithm, the system obtains standardized mode response sequences for each potential switching mode. These sequences typically consist of data on multiple variables such as voltage, current, and power changing over time, exhibiting high-dimensional and nonlinear characteristics. To more effectively compare the dynamic behavior differences between different modes, the system first extracts multidimensional features from these response sequences, such as maximum value, rise time, overshoot, recovery time, and oscillation frequency, forming a multidimensional response feature vector corresponding to each mode. Next, the system performs Principal Component Analysis (PCA) on these multidimensional response feature vectors to reduce data dimensionality and remove redundant information. PCA is a commonly used statistical method that projects the original high-dimensional data into a new low-dimensional space by calculating the eigenvalues ​​and eigenvectors of the covariance matrix, where each dimension represents a major direction of change in the original data. By performing PCA on the multidimensional response feature vectors, the system obtains the principal component loading matrix for each potential switching mode, which reflects the weight distribution of the original features in the new space. Subsequently, the system uses the principal component loading matrix to reduce the dimensionality of the original feature vectors, obtaining the principal component score vector for each mode. This score vector is the projection of the original features into the principal component space, preserving the most representative dynamic behavior information. Based on this, the system further maps these principal component score vectors to a unified feature space to obtain the projected coordinates of each potential switching mode in this space. These coordinates form the basis for subsequent distance calculations. Next, the system calculates Mahalanobis distance for all projected coordinates, obtaining the Mahalanobis distance matrix between each potential switching mode. Mahalanobis distance is a statistical distance metric that considers the correlation between variables and can effectively reflect the differences in dynamic behavior between different modes. Compared to Euclidean distance, Mahalanobis distance is better suited to address collinearity and scaling inconsistencies among multiple variables, making it more suitable for assessing the similarity or differences between different switching modes. Finally, based on the Mahalanobis distance matrix and considering actual control requirements, the system performs cost quantification modeling for different switching paths, thereby generating a mode switching cost matrix.Each element in this matrix represents the control adjustment cost required to switch from one potential switching mode to another, including but not limited to the difficulty of controller parameter reconfiguration, the complexity of energy flow reconfiguration, and system stability risks. For example, in application scenarios with declining grid frequency, if the system identifies "off-grid V / f control mode" and "reactive power support mode" as two potential switching modes, this matrix can determine that switching from the current operating mode to "off-grid V / f control mode" has a lower cost, while switching to "reactive power support mode" incurs a higher cost due to the need to reconfigure power reference values ​​and control strategies. In summary, this step, by extracting multidimensional features of the standardized response sequence, performing principal component analysis and feature space mapping, calculating Mahalanobis distance, and constructing a mode switching cost matrix, achieves precise quantification of the switching difficulty between different potential switching modes. This not only provides a key input basis for subsequent priority ranking but also enhances the intelligent decision-making capability of the control system in complex operating environments, enabling the PCS energy storage converter to achieve safer, more efficient, and stable switching operations between multiple modes.

[0036] In a specific embodiment, the step of prioritizing the potential switching modes based on the mode switching topology map to obtain a switching path includes: Graph theory analysis is performed on the mode switching topology to obtain the adjacency matrix of the mode switching topology, and the connectivity score of each potential switching mode is calculated based on the adjacency matrix. The mode switching topology graph is divided into communities using spectral clustering to obtain the mode switching community structure. Based on the mode switching community structure, the modularity coefficient of each potential switching mode is estimated to obtain the mode clustering degree. A multi-criteria decision analysis is performed on the connectivity score and pattern clustering to obtain a comprehensive handover priority index. Based on the comprehensive handover priority index, the potential handover patterns are prioritized to obtain an initial handover priority sequence. Based on the initial switching priority sequence, a multi-source shortest path search is performed on the mode switching topology to obtain the switching path.

[0037] Specifically, in this step, the system prioritizes the identified potential switching modes based on the previously constructed mode-switching topology graph and ultimately generates a switching path with practical control significance. This process involves multiple mathematical modeling and graph theory analysis steps: first, adjacency matrix extraction and connectivity score calculation are performed on the mode-switching topology graph; second, community structure partitioning is completed using spectral clustering and the modularity coefficient of each mode is evaluated; then, multi-criteria decision analysis is performed by combining connectivity and clustering to form a comprehensive switching priority index; finally, shortest path search is performed on this basis to obtain the switching path. In the specific implementation process, the system first transforms the mode-switching topology graph into a mathematical expression in the form of an adjacency matrix, where each element in the matrix represents whether there is a direct switching relationship between two potential switching modes and the magnitude of their switching cost. Based on this adjacency matrix, the system further calculates the connectivity score of each potential switching mode. This score reflects the degree of connection between a mode and other modes in the entire switching network—for example, if a mode (such as "off-grid V / f control mode") can be easily switched to multiple other modes, its connectivity score is high, indicating that it occupies a key node position in the network and has high flexibility and adaptability. Meanwhile, to gain a deeper understanding of the intrinsic relationships between different switching modes, the system employs spectral clustering to divide the entire switching topology into communities. Spectral clustering is a clustering algorithm based on the eigenvalue decomposition of the graph Laplacian matrix, which can effectively identify subgraph structures with high cohesion and low coupling characteristics in the topology. Using this method, the system can divide all potential switching modes into several "mode switching communities" with similar dynamic behaviors or control strategies, thus revealing the functional aggregation relationships between different operating modes. For example, in application scenarios where grid frequency decreases, "off-grid V / f control mode" and "constant power charging mode" may be grouped into the same community, indicating a certain similarity in their control logic; while "reactive power support mode" may belong to another community, reflecting its difference in functional positioning. Based on the obtained switching community structure, the system further estimates the modularity coefficient of each potential switching mode, i.e., the mode clustering degree. This coefficient measures the difference between the connection strength of a mode within its own community and its overall distribution in the entire topology. The higher the value, the more concentrated the mode is in a specific functional area, exhibiting stronger internal consistency. For example, if most of the switching paths in the "off-grid V / f control mode" occur within the local community, its modularity coefficient is high, indicating relatively stable control behavior and making it suitable as a priority switching option in the current state. Next, the system uses the two indicators mentioned above—connectivity score and pattern clustering degree—as key evaluation dimensions for priority ranking, conducting multi-criteria decision analysis. This analysis typically employs a weighted linear combination method, assigning different weights based on actual control requirements to generate a comprehensive switching priority index.The higher the index, the better the potential switching mode is perceived to be within the overall switching network, exhibiting both high reachability (high connectivity) and clear functional affiliation (strong clustering). Therefore, it should be assigned a higher switching priority. Based on this comprehensive switching priority index, the system sorts all potential switching modes, generating an initial switching priority sequence. This sequence represents the optimal order of switching modes based on a comprehensive consideration of factors such as stability, controllability, and functionality, starting from the current operating state. For example, in the aforementioned application scenario of declining grid frequency, the "off-grid V / f control mode" is ranked first due to its high connectivity and modularity scores, becoming the preferred switching target; while the "reactive power support mode," although possessing some grid support capability, is ranked second best due to its lower connectivity and higher switching cost. Finally, based on this initial switching priority sequence, the system executes a multi-source shortest path search algorithm (such as Dijkstra's algorithm or multi-starting-point breadth-first search) on the mode switching topology to find switching paths from the current operating mode to each potential switching mode. These paths not only meet the minimum switching cost requirement but also take into account practical operating constraints such as energy loss and transient response stability. For example, during the switch from grid-connected inverter mode to "off-grid V / f control mode," the system may identify a path that passes through "constant power charging mode." Although this path has additional hops, it is still considered one of the optimal paths because it has the lowest overall switching cost and the most stable transient response. In summary, this step, through graph theory analysis, community partitioning, multi-criteria decision ranking, and shortest path search, among other mathematical and intelligent processing procedures, achieves a systematic prioritization of potential switching modes and ultimately generates a set of switching paths with practical control value. This not only improves the adaptive capability and switching safety of the PCS energy storage converter in complex operating environments but also provides scientific and precise decision support for subsequent seamless switching control, ensuring efficient and stable operation under various conditions.

[0038] In a specific embodiment, the step of dividing the mode switching topology graph into communities using spectral clustering to obtain the mode switching community structure includes: The mode switching topology is constructed by Laplace matrix to obtain the feature spectrum distribution of the mode switching network, and eigenvalue decomposition is performed based on the feature spectrum distribution to obtain the spectral feature vector group of the mode switching network. A similarity measurement function is constructed based on the spectral feature vector group, and node similarity is calculated on the mode switching topology graph based on the similarity measurement function to obtain a similarity matrix between mode nodes. Adaptive threshold segmentation is then performed on the similarity matrix to obtain an initial community boundary set. The initial community boundary set is identified by density peak detection to obtain the core node set of the pattern community, and the affiliation of community members is calculated based on the core node set of the pattern community to obtain the community structure division result. The community structure division results are analyzed to obtain a community switching relationship map. Based on the community switching relationship map, the community structure is optimized and reorganized to obtain a mode-switching community structure.

[0039] Specifically, in this crucial step, the system uses spectral clustering to conduct in-depth analysis of the mode-switching topology and ultimately delineates the mode-switching community structure with inherent correlations. The entire process comprises four main stages: Laplace matrix construction and eigenvalue decomposition, similarity metric function construction and node similarity calculation, density peak detection and community member affiliation calculation, and analysis and optimization of inter-community switching relationships. First, to achieve community partitioning of the mode-switching network, the system needs to mathematically model the mode-switching topology, specifically by constructing its corresponding Laplace matrix. The Laplace matrix is ​​a special matrix form obtained from the adjacency matrix through certain transformations; it reflects the connection strength and topological relationships between nodes in the network. Based on this Laplace matrix, the system further performs eigenvalue decomposition to extract the feature spectral distribution of the mode-switching network and its corresponding spectral feature vector sets. These feature vector sets not only reveal the geometric characteristics within the network but also provide an important data foundation for subsequent community partitioning. For example, in applications where grid frequency decreases, Laplace matrix analysis of the switching relationships between different operating modes of the PCS energy storage converter reveals strong coupling between some modes, while others are relatively independent. Next, the system uses the obtained spectral feature vector set to construct a similarity metric function. This function quantifies the similarity between different nodes and is typically defined using methods such as Euclidean distance or cosine similarity. Based on this similarity metric function, the system performs pairwise comparisons of all nodes, calculating the similarity matrix between mode nodes. Then, by applying adaptive threshold segmentation technology to the similarity matrix, an appropriate segmentation threshold is automatically determined based on the actual distribution of values ​​in the matrix, thus dividing the similarity matrix into multiple parts to form an initial community boundary set. For example, in the aforementioned application scenario, if the "off-grid V / f control mode" and the "constant power charging mode" have high similarity, they may be grouped into the same initial community boundary set, while the "reactive power support mode" may form a separate set. Subsequently, the system uses a density peak detection algorithm to process the initial community boundary set to identify the core node set of each community. Density peak detection is an effective clustering algorithm that identifies potential cluster centers by finding points with high local density and combining this information with distance information. In this process, the system can accurately locate core nodes within a pattern community. For example, in power grid frequency management, the "off-grid V / f control mode" might be a core node in a particular community because it has low switching costs and good stability compared to other modes. Based on the identified core node set, the system further performs community member affiliation calculations to determine the specific community to which each node belongs, thus obtaining a complete community structure division result. The final step is a detailed analysis of the community structure division results, focusing particularly on the switching relationships between communities.This involves constructing a community switching relationship graph, which details the switching paths or connections between different communities. By analyzing this graph, the system can understand which communities have lower switching costs and which are relatively isolated, thus providing a basis for further optimization of the community structure. For example, in the context of frequent grid frequency fluctuations, if the switching path from "grid-connected inverter mode" to "off-grid V / f control mode" is found to be common and stable, then the connectivity between these two should be given special attention during optimization and reorganization. Based on this analysis, the system optimizes and reorganizes the existing community structure to ensure that the internal compactness of each community and the differentiation between communities reach an optimal state, ultimately forming a reasonable mode switching community structure. In summary, the process of dividing the mode switching topology into communities using spectral clustering is not only a deep analysis of the complex network structure but also a precise capture of the inherent connections between different operating modes. This method not only improves the adaptability of the PCS energy storage converter in the face of changing operating conditions but also provides scientific decision support for the control system in actual operation, ensuring that the system can flexibly respond to various challenges while maintaining high efficiency and stability. For example, in the event of abnormal grid frequency, the system can quickly identify the most suitable switching path for the current situation and rapidly adjust to the most appropriate operating mode to ensure the continuity and reliability of power supply.

[0040] The multi-mode seamless switching control method for the PCS energy storage converter in the embodiments of the present invention has been described above. The multi-mode seamless switching control device for the PCS energy storage converter in the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 2 One embodiment of the multi-mode seamless switching control device for the PCS energy storage converter in this invention includes: The acquisition module 21 is used to acquire the current operating status parameters of the PCS energy storage converter in real time through preset electrical sensors; Matching module 22 is used to obtain the potential switching mode corresponding to the mode switching trigger condition based on the matching relationship between the current running state parameters and the preset mode switching trigger condition; Analysis module 23 is used to perform transient stability analysis on the PCS energy storage converter based on the potential switching modes, and obtain the transient response characteristic curves of each potential switching mode; The sorting module 24 is used to sort the potential switching modes by priority based on the transient response characteristic curve to obtain a switching priority sequence; Control module 25 is used to perform multi-mode seamless switching control on the PCS energy storage converter based on the switching priority sequence.

[0041] In this embodiment, the specific implementation of each unit in the above device embodiment is described in the above method embodiment, and will not be repeated here.

[0042] Reference Figure 3 This invention also provides a computer device whose internal structure can be as follows: Figure 3 As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0043] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0044] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0045] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0047] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A multi-mode seamless switching control method for a PCS energy storage converter, characterized in that, Includes the following steps: The current operating status parameters of the PCS energy storage converter are collected in real time through preset electrical sensors. Based on the matching relationship between the current operating status parameters and the preset mode switching trigger conditions, the potential switching modes corresponding to the mode switching trigger conditions are obtained. Transient stability analysis of the PCS energy storage converter is performed based on the potential switching modes to obtain the transient response characteristic curves of each potential switching mode. Based on the transient response characteristic curve, the potential switching modes are prioritized to obtain a switching priority sequence; Based on the switching priority sequence, the PCS energy storage converter is subjected to multi-mode seamless switching control.

2. The multi-mode seamless switching control method for PCS energy storage converter according to claim 1, characterized in that, The step of obtaining the potential switching mode corresponding to the mode switching trigger condition based on the matching relationship between the current operating state parameters and the preset mode switching trigger condition includes: Detect the threshold boundary of the current operating status parameters and identify a subset of abnormal parameters that exceed the set threshold; Each parameter type in the subset of abnormal parameters is matched with the trigger rule field in the preset mode switching trigger condition to obtain multiple trigger condition entries corresponding to the subset of abnormal parameters. The multiple trigger condition entries are associated with a predefined operation mode mapping table to obtain one or more operation mode identifiers corresponding to each trigger condition entry, and the potential switching mode corresponding to the mode switching trigger condition is determined based on the operation mode identifier.

3. The multi-mode seamless switching control method for the PCS energy storage converter according to claim 1, characterized in that, The transient stability analysis of the PCS energy storage converter based on the potential switching modes yields transient response characteristic curves for each potential switching mode, including: Small-signal disturbance injection is performed on the PCS energy storage converter based on the potential switching modes to obtain voltage, current and power fluctuation data under each potential switching mode. Frequency domain characteristic analysis is then performed based on the voltage, current and power fluctuation data to obtain the frequency domain response characteristics of each mode. By using state-space modeling, the dynamic characteristics of the PCS energy storage converter under each potential switching mode are simulated, and the transient response of each potential switching mode is analyzed in the time domain based on the dynamic characteristics to obtain the time domain response characteristic parameters. Based on the frequency domain response characteristics and the time domain response characteristic parameters, a transient stability evaluation matrix for each potential switching mode is constructed, and a multi-objective optimization calculation is performed on the transient stability evaluation matrix to obtain a comprehensive stability index. The comprehensive stability index is normalized to obtain the normalized stability index, and the transient response characteristic curves of each potential switching mode are generated based on the normalized stability index.

4. The multi-mode seamless switching control method for PCS energy storage converter according to claim 1, characterized in that, The step of prioritizing the potential handover modes based on the transient response characteristic curve to obtain a handover priority sequence includes: Numerical integration is performed on the transient response characteristic curve to obtain the energy fluctuation curve, and the energy loss of the PCS energy storage converter in each potential switching mode is calculated based on the energy fluctuation curve to obtain the energy loss matrix. The transient response characteristic curves are aligned in the time domain by a dynamic time warping algorithm to obtain a standardized mode response sequence, and the mode switching cost matrix between each potential switching mode is calculated based on the standardized mode response sequence. The energy loss matrix and the mode switching cost matrix are weighted and fused to obtain a comprehensive switching evaluation matrix, and a mode switching topology map is constructed based on the comprehensive switching evaluation matrix. Based on the mode switching topology map, the potential switching modes are prioritized to obtain switching paths. The feasibility of the switching paths is verified and conflicts are detected and eliminated to obtain a switching priority sequence.

5. The multi-mode seamless switching control method for the PCS energy storage converter according to claim 4, characterized in that, The calculation of the mode switching cost matrix between potential switching modes based on the standardized mode response sequence includes: Extract the multidimensional response feature vectors of each potential switching mode from the standardized mode response sequence, and perform principal component analysis on the multidimensional response feature vectors to obtain the principal component loading matrix of each mode. The multidimensional response feature vector is reduced in dimensionality by the principal component loading matrix to obtain the principal component score vector of each potential switching mode. Based on the principal component score vector, feature space mapping is performed to obtain the projection coordinate points of each mode in the feature space. The Mahalanobis distance between the projected coordinate points is calculated to obtain the Mahalanobis distance matrix between each potential switching mode. The mode switching cost is quantified based on the Mahalanobis distance matrix to obtain the mode switching cost matrix.

6. The multi-mode seamless switching control method for the PCS energy storage converter according to claim 4, characterized in that, The step of prioritizing the potential switching modes based on the mode switching topology map to obtain the switching path includes: Graph theory analysis is performed on the mode switching topology to obtain the adjacency matrix of the mode switching topology, and the connectivity score of each potential switching mode is calculated based on the adjacency matrix. The mode switching topology graph is divided into communities using spectral clustering to obtain the mode switching community structure. Based on the mode switching community structure, the modularity coefficient of each potential switching mode is estimated to obtain the mode clustering degree. A multi-criteria decision analysis is performed on the connectivity score and pattern clustering to obtain a comprehensive handover priority index. Based on the comprehensive handover priority index, the potential handover patterns are prioritized to obtain an initial handover priority sequence. Based on the initial switching priority sequence, a multi-source shortest path search is performed on the mode switching topology to obtain the switching path.

7. A multi-mode seamless switching control device for a PCS energy storage converter, characterized in that, include: The acquisition module is used to acquire the current operating status parameters of the PCS energy storage converter in real time through preset electrical sensors; The matching module is used to obtain the potential switching mode corresponding to the mode switching trigger condition based on the matching relationship between the current running status parameters and the preset mode switching trigger condition; The analysis module is used to perform transient stability analysis on the PCS energy storage converter based on the potential switching modes, and obtain the transient response characteristic curves of each potential switching mode. The sorting module is used to prioritize the potential switching modes based on the transient response characteristic curve to obtain a switching priority sequence. The control module is used to perform multi-mode seamless switching control of the PCS energy storage converter based on the switching priority sequence.

8. 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 method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Citation Information

Cited By

  • Automatic power protection and grid-connected and off-grid switching method and system of mobile automatic power protection device

    CN121395570A