A multi-time scale response test method for a megawatt multi-composite energy storage power supply

Abstract

This invention discloses a multi-timescale response testing method for megawatt-level multi-composite energy storage power supplies. The method includes: acquiring the rated power, grid-connected voltage level, energy storage unit type, and test disturbance boundary of the power supply under test, and generating a test configuration table; generating a composite excitation signal containing multi-timescale excitation components, superimposing it onto the rated power command to obtain a composite test command; simultaneously acquiring the branch response and common-side response data of each energy storage unit to form a response data matrix; performing two-stage time-frequency component separation using complete ensemble empirical mode decomposition and wavelet packet decomposition and reconstruction to form a component attribution table; and calculating response characteristic parameters, coupling evaluation parameters, and circulating current evaluation parameters based on these parameters and writing them into a report. This method allows for the acquisition of multi-timescale response characteristics of each energy storage unit and the coupling and circulating current states between branches without disassembling the energy storage module, all through a single test.

Description

Technical Field

[0001] This invention belongs to the technical field of power energy storage equipment testing and grid connection performance evaluation, and relates to a multi-timescale response testing method for megawatt-level multi-composite energy storage power sources. Background Technology

[0002] Megawatt-level multi-functional composite energy storage power sources combine power-type and energy-type energy storage units to achieve both rapid power response and long-term energy support. As the power level of energy storage systems increases to the megawatt level, these power sources must not only meet the response speed requirements of individual branches during grid-connected operation, but also the requirements of common DC bus stability, branch current sharing, and control link coordination. Therefore, accurate testing of their multi-timescale response characteristics is a crucial prerequisite for ensuring safe grid connection and reliable operation.

[0003] Existing energy storage testing solutions primarily target single energy storage devices, dual-source composite energy storage systems, or low-voltage offline prototypes. These solutions struggle to simultaneously elicit and differentiate the fast, medium, and slow response processes of different energy storage units during the same real-world high-voltage grid-connected operation. Furthermore, without disassembling the energy storage modules, existing solutions cannot trace the collected total output response data back to a specific energy storage unit, nor can they simultaneously evaluate electrical crosstalk and branch circulating currents between energy storage units during the same test.

[0004] Therefore, there is a need for a method that can obtain the multi-timescale response characteristics of different energy storage units in multiple composite energy storage power sources through a single composite test process under megawatt-level high-voltage grid-connected conditions, and generate traceable test results including response parameters, crosstalk indices, and circulating current indices. Summary of the Invention

[0005] To address the problems existing in the background technology, this invention proposes a multi-timescale response testing method for megawatt-level multi-composite energy storage power sources.

[0006] To achieve the above objectives, the technical solution adopted by this invention is as follows: a multi-timescale response testing method for megawatt-level multi-composite energy storage power supplies, comprising: Obtain the rated power, grid connection voltage level, energy storage unit type and test disturbance boundary of the megawatt-level multi-composite energy storage power supply under test, and generate a test configuration table; A composite excitation signal containing multiple time-scale excitation components is generated according to the test configuration table, and the composite excitation signal is superimposed on the rated power command to obtain a composite test command; During the execution of the composite test command, branch response data and common-side response data of multiple energy storage units are collected simultaneously to form a response data matrix with a unified timestamp. Two-level time-frequency component separation processing is performed on the response data matrix to form a component attribution table corresponding to the multiple energy storage units; Response characteristic parameters, coupling evaluation parameters, and circulation evaluation parameters are generated based on the component attribution table and written into the multi-timescale response test report.

[0007] Furthermore, when generating the test configuration table, a channel mapping relationship is established between each energy storage unit identifier and the terminal voltage acquisition channel, branch current acquisition channel, power response acquisition channel, common DC bus voltage acquisition channel, and total output power acquisition channel, and the channel mapping relationship is written into the test configuration table.

[0008] Furthermore, when generating the composite excitation signal, short-time pulse components, medium-time fluctuation components, and long-time step components are generated according to the test disturbance boundary. Amplitude limiting and timing arrangement are performed on the short-time pulse component, the medium-time fluctuation component, and the long-time step component to form an excitation schedule; The composite excitation signal is obtained by superimposing the excitation components of each time scale according to the excitation schedule.

[0009] Furthermore, when forming the response data matrix, the branch response data and the common side response data are uniformly triggered, timestamp aligned, and abnormal sampling points removed according to the excitation schedule, so that each sampling moment corresponds to the terminal voltage, branch current, power response, common DC bus voltage, and total output power of multiple energy storage units.

[0010] Furthermore, when performing two-level time-frequency component separation processing on the response data matrix, a complete set empirical mode decomposition is performed on the total output power or the common DC bus voltage to obtain multiple intrinsic mode function components; Based on the center frequency, energy concentration range, and time span of each intrinsic mode function component, candidate response components that match the short-time pulse component, the medium-time fluctuation component, and the long-time step component are selected.

[0011] Furthermore, wavelet packet decomposition, frequency band filtering, and reconstruction are performed on the candidate response components to obtain multiple reconstructed response components; Based on the matching relationship between the plurality of reconfigured response components and the channel mapping relationship, the plurality of reconfigured response components are respectively mapped to the response tags of the corresponding energy storage units.

[0012] Furthermore, when forming the component attribution table, each response tag, corresponding energy storage unit identifier, corresponding time scale, corresponding reconstructed response component, corresponding acquisition channel identifier, and corresponding timestamp interval are associated and written into the component attribution table, so that a traceable mapping is formed between the reconstructed response component and the energy storage unit type.

[0013] Furthermore, when generating the response characteristic parameters, the response delay, rise time, overshoot, and steady-state error of each energy storage unit are calculated based on the issuance time of the composite test command and the reconstructed response component curve in the component attribution table. The difference between the moment when the preset low-proportion response amplitude is first reached and the moment of transmission is recorded as the response delay, and the difference between the moment when the preset high-proportion response amplitude is first reached and the moment when the preset low-proportion response amplitude is first reached is recorded as the rise time.

[0014] Furthermore, when generating the coupling evaluation parameters and the circulation evaluation parameters, a cross-coupling gain is formed based on the excitation change of the target energy storage unit and the response change of the non-target energy storage unit within the main excitation window; The branch circulating current evaluation value is formed based on the difference between the branch currents of different converters at the same timestamp; The cross-coupling gain and the branch circulation evaluation value are associated with the corresponding time scale in the component attribution table.

[0015] Furthermore, when writing the multi-timescale response test report, the test configuration table, the excitation schedule, the response data matrix, the component attribution table, the response characteristic parameters, the coupling evaluation parameters, and the circulation evaluation parameters are established as report entries according to the same energy storage unit identifier and the same time scale index, so that the multi-timescale response test report simultaneously presents the response speed, steady-state accuracy, coupling status between energy storage units, and branch circulation status.

[0016] Compared with the prior art, the present invention has the following advantages: by constructing a composite excitation signal containing multiple time scale excitation components and generating a unified trigger timestamp, a single test process can simultaneously cover different response time scales, and the response data of each branch and the response data of the common side have the same time reference.

[0017] By performing a two-stage time-frequency component separation process of complete set empirical mode decomposition and wavelet packet decomposition and reconstruction, and mapping the reconstructed response components to the response tags of the corresponding energy storage units, the total response data can be traced back to the specific energy storage unit without disassembling the energy storage module.

[0018] By calculating the cross-coupling gain and branch circulation current evaluation values ​​within the same test window, the test report can simultaneously reflect the response speed, steady-state accuracy, coupling status between energy storage units, and branch circulation current status of each energy storage unit. Attached Figure Description

[0019] Figure 1 This is a flowchart of a multi-timescale response testing method for megawatt-level multi-composite energy storage power supplies according to the present invention; Figure 2 This is a schematic diagram of the channel mapping relationship of the present invention; Figure 3 This is a schematic diagram illustrating the relationship between the composite excitation signal and the unified trigger timestamp of this invention; Figure 4 This is a schematic diagram of the response data component separation and tag matching data stream of the present invention; Figure 5 This is a schematic diagram illustrating the relationship between the generation of evaluation parameters and the writing of reports in this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] like Figures 1-5 As shown, the technical solution adopted in this invention is as follows: a multi-timescale response test method for megawatt-level multi-composite energy storage power sources.

[0022] Megawatt-level multi-composite energy storage power refers to a composite energy storage power with a rated power of megawatts and including at least two types of energy storage units with different response time scales.

[0023] The at least two types of energy storage units with different response time scales include high-voltage thin-film capacitor energy storage units, high-voltage supercapacitor energy storage units, and high-voltage lithium supercapacitor energy storage units.

[0024] Among the three types of energy storage units, the high-voltage thin-film capacitor energy storage unit stores and releases charge based on dielectric polarization, and its response time is the shortest. The high-voltage supercapacitor energy storage unit stores and releases charge based on the double-layer electrical method, and its response time is in the middle. The high-voltage lithium supercapacitor energy storage unit combines capacitive and lithium-ion intercalation / deintercalation characteristics, and its response time is the longest among the three types of energy storage units.

[0025] A multi-timescale response testing method for megawatt-level multi-composite energy storage power sources includes: S1: Obtain the rated power, grid connection voltage level, energy storage unit type, and test disturbance boundary of the megawatt-level multi-composite energy storage power supply under test, and generate a test configuration table.

[0026] After obtaining the rated power, grid-connected voltage level, energy storage unit type, and test disturbance boundaries of the megawatt-level multi-composite energy storage power supply under test, these parameters are written into the test configuration table. The test disturbance boundaries include allowable power disturbance boundaries and allowable bus voltage disturbance boundaries, used to limit the amplitude of subsequently generated composite excitation signals, ensuring that the test process does not exceed the safe operating range of the megawatt-level multi-composite energy storage power supply under test.

[0027] The rated power represents the maximum active power design value that the tested megawatt-level multi-composite energy storage power source can continuously output under standard grid-connected conditions, in megawatts (MW). This parameter serves as the power reference for forming the rated power command and composite test command in subsequent steps.

[0028] The grid connection voltage level refers to the nominal line voltage RMS value at the point of common connection of the megawatt-level multi-composite energy storage power supply under test to the power grid. It is used to determine the electrical boundaries and insulation coordination requirements of the test. In one embodiment, its value is 10kV or 35kV.

[0029] The energy storage unit type represents the category information of each energy storage unit in the megawatt-level multi-composite energy storage power supply under test. It is used to distinguish energy storage units with different response time scales and to provide a classification basis for the subsequent generation of component attribution tables, response labels and multi-time scale response test reports.

[0030] The permissible power disturbance boundary defines the upper limit of the excitation signal amplitude superimposed on the rated power command, and the permissible bus voltage disturbance boundary defines the permissible peak fluctuation range of the common DC bus voltage during the test.

[0031] The test configuration table is the basic data structure, which at least includes the rated power, grid-connected voltage level, energy storage unit type, and test disturbance boundary. In subsequent steps, it serves as the data source for generating composite excitation signals, forming channel mapping relationships, synchronously acquiring response data matrices, performing two-level time-frequency component separation processing, and generating multi-time-scale response test reports.

[0032] Furthermore, when generating the test configuration table, a channel mapping relationship is established between each energy storage unit identifier and the terminal voltage acquisition channel, branch current acquisition channel, power response acquisition channel, common DC bus voltage acquisition channel, and total output power acquisition channel, and the channel mapping relationship is written into the test configuration table.

[0033] When generating the test configuration table, a channel mapping relationship is established between each energy storage unit identifier and the terminal voltage acquisition channel, branch current acquisition channel, power response acquisition channel, common DC bus voltage acquisition channel, and total output power acquisition channel. This channel mapping relationship is then written into the test configuration table, so that the test configuration table simultaneously contains the electrical parameters of the object under test, disturbance boundaries, and complete data acquisition routing information.

[0034] The energy storage unit identifier is a unique machine-readable tag assigned to each independently controllable energy storage branch in the megawatt-level multi-composite energy storage power supply under test, serving as the primary key for end-to-end data traceability. The terminal voltage acquisition channel is used to collect voltage data at the corresponding port of each energy storage unit. The branch current acquisition channel is used to collect current data from each energy storage unit's branch. The power response acquisition channel is used to collect power response data for each energy storage unit during the execution of the composite test command; this data can be calculated in real-time by the acquisition system based on the synchronously acquired terminal voltage and branch current. The common DC bus voltage acquisition channel is used to collect voltage data on the common DC bus side of the megawatt-level multi-composite energy storage power supply under test; this channel is common to all energy storage units. The total output power acquisition channel is used to collect the total output power data of the overall output side of the megawatt-level multi-composite energy storage power supply under test; this channel is common to all energy storage units.

[0035] The specific method for establishing channel mapping relationships is as follows: For each energy storage unit identifier, create a record and bind this identifier to the number of the five acquisition channels mentioned above. After completing the above binding operation for all energy storage unit identifiers, the set of all records formed constitutes the channel mapping relationship set. This channel mapping relationship set is written into the test configuration table in the form of a channel mapping table.

[0036] The function of this channel mapping relationship is to establish a definite correspondence between energy storage unit identifiers and electrical quantity acquisition channels, enabling the synchronous data acquisition module to acquire branch response data and common-side response data according to the test configuration table during the execution of composite test commands. This channel mapping relationship is directly referenced in the subsequent steps of forming the response data matrix and response label mapping, so that the sampled data column of each acquisition channel corresponds to a specific energy storage unit identifier and a specific electrical quantity, thereby providing a basis for tracing the reconstructed response components back to the energy storage unit.

[0037] S2: Generate a composite excitation signal containing multiple time-scale excitation components according to the test configuration table, and superimpose the composite excitation signal onto the rated power command to obtain a composite test command.

[0038] A composite excitation signal containing multiple time-scale excitation components is generated based on the test configuration table, and this composite excitation signal is superimposed on the rated power command to obtain a composite test command. Simultaneously with sending the composite test command, a unified trigger timestamp corresponding to the composite test command is generated, serving as a common time reference for subsequent synchronous data acquisition.

[0039] like Figure 3 As shown, the relationship between the composite excitation signal and the unified trigger timestamp is as follows: the moment when the composite test command is issued is taken as the zero moment of the unified trigger timestamp, and the starting moment of each time scale excitation component in the excitation timetable is calibrated relative to the unified trigger timestamp.

[0040] The composite test command is expressed as follows: In the formula, This indicates a compound test instruction. This indicates the rated power of the megawatt-level multi-composite energy storage power supply being tested. Indicates time scale label. Indicates time scale label The corresponding excitation components.

[0041] Furthermore, when generating the composite excitation signal, short-time pulse components, medium-time fluctuation components, and long-time step components are generated according to the test disturbance boundary; amplitude limiting and timing arrangement are performed on the short-time pulse components, the medium-time fluctuation components, and the long-time step components to form an excitation schedule; the excitation components of each time scale are superimposed according to the excitation schedule to obtain the composite excitation signal.

[0042] When generating the composite excitation signal, short-time pulse components, medium-time fluctuation components, and long-time step components are generated according to the test disturbance boundary. The short-time pulse component corresponds to the first time scale and is used to excite the energy storage unit with the fastest response; the medium-time fluctuation component corresponds to the second time scale and is used to excite the energy storage unit with the medium response speed; the long-time step component corresponds to the third time scale and is used to excite the energy storage unit with the slowest response.

[0043] In one embodiment, the short-time pulse component operates on a microsecond timescale, the medium-time fluctuation component on a millisecond timescale, and the long-time step component on a second timescale. These microsecond, millisecond, and second timescales correspond to the order of response speed of the high-voltage thin-film capacitor energy storage unit, the high-voltage supercapacitor energy storage unit, and the high-voltage lithium supercapacitor energy storage unit, respectively. The specific boundaries of these timescales are determined based on the bandwidth, protection strategy, and sensor performance of the megawatt-level multi-composite energy storage power supply under test.

[0044] Amplitude limiting and timing arrangement are performed on short-time pulse components, medium-time fluctuation components, and long-time step components to form an excitation schedule. Amplitude limiting restricts the amplitude of each time-scale excitation component to within the allowable power disturbance boundary, and ensures that the disturbance to the common DC bus voltage caused by the superposition of excitation components at all time scales does not exceed the allowable bus voltage disturbance boundary. Timing arrangement records the excitation identifier, start time, duration, amplitude boundary, and target time scale label for each time-scale excitation component in the excitation schedule.

[0045] In one embodiment, the timing arrangement sequentially arranges the start time and duration of each time-scale excitation component in the order of short-time pulse component, medium-time fluctuation component, and long-time step component, ensuring that the sum of the amplitudes of the time-scale excitation components participating in the superposition at any given time does not exceed the allowable power disturbance boundary. If the sum of the amplitudes of the short-time pulse component, medium-time fluctuation component, and long-time step component exceeds the allowable power disturbance boundary at any given time, the time-scale excitation component with the later start time is postponed until the sum of the amplitudes of the time-scale excitation components at any given time does not exceed the allowable power disturbance boundary.

[0046] The composite excitation signal is obtained by superimposing the excitation components of each time scale according to the excitation schedule. The composite excitation signal is then added to the rated power command at each sampling time to obtain the composite test command, which is then sent to the hierarchical collaborative control unit of the megawatt-level multi-composite energy storage power supply under test.

[0047] S3: During the execution of the composite test command, branch response data and common-side response data of multiple energy storage units are collected simultaneously to form a response data matrix with a unified timestamp.

[0048] During the execution of the composite test command, branch response data and common-side response data from multiple energy storage units are simultaneously collected using a unified trigger timestamp as the time reference, forming a response data matrix with a unified timestamp. The branch response data includes the terminal voltage, branch current, and power response of each energy storage unit, while the common-side response data includes the common DC bus voltage and total output power. The response data matrix is ​​structured with the sampling time as the row and each electrical quantity sampling channel as the column, forming a time-stamped multidimensional data set.

[0049] In one embodiment, the sampling frequency of the synchronous acquisition is not lower than a preset sampling frequency threshold; when the short-time pulse component is a microsecond-level narrow pulse, the preset sampling frequency threshold can be taken as 1MHz.

[0050] Furthermore, when forming the response data matrix, the branch response data and the common side response data are uniformly triggered, timestamp aligned, and abnormal sampling points removed according to the excitation schedule, so that each sampling moment corresponds to the terminal voltage, branch current, power response, common DC bus voltage, and total output power of multiple energy storage units.

[0051] When forming the response data matrix, the branch response data and common-side response data are subjected to unified trigger marking, timestamp alignment, and abnormal sampling point removal according to the excitation schedule. The unified trigger marking uses the unified trigger timestamp as the zero time and adds a timestamp relative to the unified trigger timestamp to each sampling point of each acquisition channel. The timestamp alignment aligns the sampling points of each acquisition channel to the same time axis according to the timestamp, so that the same row corresponds to the same sampling time.

[0052] When the sampling rates of the acquisition channels are inconsistent, the timestamp alignment further uses the time grid corresponding to the highest sampling rate as the reference time axis and performs linear interpolation on the acquisition channel data with lower sampling rates according to the time stamp.

[0053] The abnormal sampling point removal process is as follows: For any sampling point in any acquisition channel, if the value of the sampling point exceeds the range of the corresponding acquisition channel, or if the absolute value of the difference between the sampling point and an adjacent sampling point in the same acquisition channel is greater than a preset jump threshold, then the sampling point is determined to be an abnormal sampling point and removed, and the data at the corresponding time is filled with the linear interpolation result of the adjacent valid sampling points; otherwise, the sampling point is retained. The preset jump threshold is determined based on the product of the maximum rate of change of the corresponding electrical quantity under normal operating conditions and the sampling interval. In a specific value, the preset jump threshold is taken as 20% of the range of the corresponding electrical quantity.

[0054] Through the above processing, the terminal voltage, branch current, power response, common DC bus voltage and total output power of multiple energy storage units are corresponding to each sampling moment in the response data matrix, and the data of each channel are strictly aligned on the time axis.

[0055] S4: Perform two-level time-frequency component separation processing on the response data matrix to form a component attribution table corresponding to the multiple energy storage units.

[0056] A two-stage time-frequency component separation process is performed on the response data matrix to form a component attribution table corresponding to multiple energy storage units. The two-stage time-frequency component separation process includes: a first stage, performing complete set empirical mode decomposition on the target response data to obtain intrinsic mode function components and screening candidate response components; and a second stage, performing wavelet packet decomposition, frequency band screening, and reconstruction on the candidate response components to obtain reconstructed response components and mapping the reconstructed response components to the response tags of the corresponding energy storage units.

[0057] Before performing two-stage time-frequency component separation processing, the response data matrix is ​​first subjected to unit normalization and test window truncation to obtain a preprocessed response data matrix. Unit normalization involves dividing the sampled data of each electrical quantity acquisition channel within the test window by the rated value or upper limit of the corresponding electrical quantity, converting it into a dimensionless per-unit value. Test window truncation is based on the start time and duration of each time-scale excitation component in the excitation timetable. The test window is the time interval from the earliest start time to the end of the latest time-scale excitation component, extended by a preset steady-state observation duration. The preset steady-state observation duration is not less than the duration of the long-time step component.

[0058] Furthermore, when performing two-level time-frequency component separation processing on the response data matrix, a complete set of empirical mode decomposition is performed on the total output power or the common DC bus voltage to obtain multiple intrinsic mode function components; based on the center frequency, energy concentration interval and time span of each intrinsic mode function component, candidate response components that match the short-time pulse component, the medium-time fluctuation component and the long-time step component are selected.

[0059] Complete ensemble empirical mode decomposition (CEMD) is performed on the total output power data or common DC bus voltage data in the preprocessed response data matrix to obtain multiple intrinsic mode function (EMF) components. The complete ensemble CEMD is an empirical mode decomposition method with adaptive noise, which adaptively decomposes the input non-stationary signal into several EMF components arranged from high frequency to low frequency.

[0060] The specific method for screening candidate response components is as follows: For each intrinsic mode function (IMF) component, determine the start and end times, center frequency, and time span within the test window corresponding to the energy concentration interval of the IMF component; for each time-scale excitation component in the excitation time table, if the energy concentration interval of the IMF component overlaps with the interval of the time-scale excitation component from the start time to the start time plus the preset response observation duration, and the center frequency of the IMF component falls within the preset frequency band interval corresponding to the time-scale excitation component, then the IMF component is determined as a candidate response component matching the time-scale excitation component, and a target time-scale label is assigned to the time-scale excitation component; otherwise, the IMF component is excluded.

[0061] Among them, the preset frequency band interval corresponding to the short-time pulse component is the highest frequency band interval among the three preset frequency band intervals, the preset frequency band interval corresponding to the medium-time fluctuation component is the middle frequency band interval among the three preset frequency band intervals, and the preset frequency band interval corresponding to the long-time step component is the lowest frequency band interval among the three preset frequency band intervals.

[0062] Further, wavelet packet decomposition, frequency band filtering, and reconstruction are performed on the candidate response components to obtain multiple reconstructed response components; according to the matching relationship between the multiple reconstructed response components and the channel mapping relationship, the multiple reconstructed response components are respectively mapped to the response tags of the corresponding energy storage units.

[0063] Wavelet packet decomposition, frequency band filtering, and reconstruction are performed on candidate response components to obtain multiple reconstructed response components. Specifically, for each candidate response component, wavelet packet decomposition is performed according to a preset wavelet basis and a preset decomposition level to obtain the wavelet packet coefficients of that candidate response component in each frequency band. Frequency bands with wavelet packet coefficient energy greater than a preset energy retention ratio are retained, while frequency bands with wavelet packet coefficient energy not greater than the preset energy retention ratio are set to zero. Wavelet packet reconstruction is performed on the wavelet packet coefficients of the retained frequency bands to obtain the reconstructed response component corresponding to the candidate response component. The preset wavelet basis and the preset decomposition level are determined based on the sampling frequency, target time scale label, and frequency range of the candidate response components in the test configuration table.

[0064] The specific method for mapping the reconfigured response components to the corresponding energy storage unit's response tag is as follows: For each reconfigured response component, determine its dominant energy frequency band and dominant energy time window. If the dominant energy time window of the reconfigured response component corresponds to a first time scale and the dominant energy frequency band is in the highest frequency band range, then map the reconfigured response component to the response tag of the high-voltage thin-film capacitor energy storage unit. If the dominant energy time window of the reconfigured response component corresponds to a second time scale and the dominant energy frequency band is in the middle frequency band range, then map the reconfigured response component to the response tag of the high-voltage supercapacitor energy storage unit. If the dominant energy time window of the reconfigured response component corresponds to a third time scale and the dominant energy frequency band is in the lowest frequency band range, then map the reconfigured response component to the response tag of the high-voltage lithium supercapacitor energy storage unit.

[0065] like Figure 4 As shown, a complete set of empirical mode decompositions is performed on the total output power to obtain multiple intrinsic mode function components. After candidate response component screening, wavelet packet decomposition, frequency band screening, and reconstruction, multiple reconstructed response components are obtained. The reconstructed response components corresponding to the highest frequency band interval and the first time scale are mapped to the response labels of the high-voltage thin-film capacitor energy storage unit, the reconstructed response components corresponding to the middle frequency band interval and the second time scale are mapped to the response labels of the high-voltage supercapacitor energy storage unit, and the reconstructed response components corresponding to the lowest frequency band interval and the third time scale are mapped to the response labels of the high-voltage lithium supercapacitor energy storage unit.

[0066] Furthermore, when forming the component attribution table, each response tag, corresponding energy storage unit identifier, corresponding time scale, corresponding reconstructed response component, corresponding acquisition channel identifier, and corresponding timestamp interval are associated and written into the component attribution table, so that a traceable mapping is formed between the reconstructed response component and the energy storage unit type.

[0067] When creating the component attribution table, each response tag, corresponding energy storage unit identifier, corresponding time scale, corresponding reconstructed response component, corresponding acquisition channel identifier, and corresponding timestamp interval are associated and written into the component attribution table. This establishes a traceable mapping between the reconstructed response components and the energy storage unit type. Without disassembling the megawatt-level multi-composite energy storage power supply under test, each reconstructed response component obtained by decomposing and reconstructing the total output power or common DC bus voltage can be traced back to a specific time scale, a specific energy storage unit identifier, and a specific acquisition channel identifier.

[0068] S5: Generate response characteristic parameters, coupling evaluation parameters, and circulation evaluation parameters based on the component attribution table, and write them into the multi-timescale response test report.

[0069] Response characteristic parameters, coupling evaluation parameters, and circulation evaluation parameters are generated based on the component attribution table, and these parameters are written into the multi-timescale response test report.

[0070] Furthermore, when generating the response characteristic parameters, the response delay, rise time, overshoot, and steady-state error of each energy storage unit are calculated based on the issuance time of the composite test command and the reconstructed response component curves in the component attribution table. Specifically, the difference between the time when the preset low-proportion response amplitude is first reached and the issuance time is recorded as the response delay, and the difference between the time when the preset high-proportion response amplitude is first reached and the time when the preset low-proportion response amplitude is first reached is recorded as the rise time.

[0071] When generating response characteristic parameters, the response delay, rise time, overshoot, and steady-state error of each energy storage unit are calculated based on the issuance time of the composite test command and the reconstructed response component curves in the component attribution table. The issuance time of the composite test command is the unified trigger timestamp.

[0072] The response delay is the difference between the moment when the preset low-proportion response amplitude is first reached and the moment of transmission. The rise time is the difference between the moment when the preset high-proportion response amplitude is first reached and the moment when the preset low-proportion response amplitude is first reached. The preset low-proportion response amplitude is the product of the steady-state response value and the preset low proportion, and the preset high-proportion response amplitude is the product of the steady-state response value and the preset high proportion. In a specific embodiment, the preset low proportion is 10%, and the preset high proportion is 90%.

[0073] The response delay is expressed as: In the formula, Indicates the first The response delay of each energy storage unit Indicates a unified trigger timestamp. Indicates the first The moment when the reconfiguration response component of an energy storage unit first reaches a preset low proportion of the steady-state response amplitude.

[0074] The rise time is expressed as: In the formula, Indicates the first The rise time of each energy storage unit Indicates the first The moment when the reconfiguration response component of an energy storage unit first reaches a high proportion of the preset steady-state response amplitude.

[0075] The overshoot is the ratio of the portion of the reconstructed response component curve that exceeds the steady-state response value to the steady-state response value, expressed as: In the formula, Indicates the first Overshoot of each energy storage unit Indicates the first The maximum value of the reconfiguration response component of each energy storage unit. Indicates the first Steady-state response value of each energy storage unit.

[0076] The steady-state error is the difference between the steady-state response value and the reference response value, where the reference response value is the target response value of the composite test command applied to the corresponding energy storage unit at the corresponding time scale, expressed as: In the formula, Indicates the first Steady-state error of each energy storage unit Indicates the first Reference response value for each energy storage unit.

[0077] Furthermore, when generating the coupling evaluation parameters and the circulating current evaluation parameters, a cross-coupling gain is formed based on the excitation change of the target energy storage unit and the response change of the non-target energy storage unit within the main excitation window; a branch circulating current evaluation value is formed based on the difference between the branch currents of different converters at the same timestamp; and the cross-coupling gain and the branch circulating current evaluation value are associated with the corresponding time scale in the component attribution table.

[0078] When generating coupling evaluation parameters and circulation evaluation parameters, a cross-coupling gain is formed based on the excitation change of the target energy storage unit and the response change of the non-target energy storage unit within the main excitation window. The main excitation window is the time interval from the start time of the time-scale excitation component corresponding to the target energy storage unit in the self-excitation timetable to the start time plus a preset response observation duration.

[0079] The cross-coupling gain is the ratio of the response change of the non-target energy storage unit within the main excitation window to the excitation change of the target energy storage unit within the main excitation window, expressed as: In the formula, Indicates the target energy storage unit Cross-coupling gain for non-target energy storage units, This indicates that the incentive timeline primarily targets the energy storage units. The change in incentive, This represents the change in response of a non-target energy storage unit within the same main excitation window.

[0080] A branch circulating current evaluation value is formed based on the difference between the branch currents of different converters at the same time stamp. The branch circulating current evaluation value is the absolute value of the difference between the branch currents of different converters at the same time stamp, expressed as: In the formula, Indicates energy storage unit The evaluation value of the circulating current between the corresponding converter branch and the corresponding converter branch of the energy storage unit (l). and Representing energy storage units and energy storage units The branch current of the corresponding converter branch at the same timestamp.

[0081] The cross-coupling gain and branch circulation evaluation values ​​are associated with the corresponding time scale in the component attribution table, so that the coupling evaluation parameters and circulation evaluation parameters have the same time base and the same energy storage unit index as the response characteristic parameters.

[0082] Furthermore, when writing the multi-timescale response test report, the test configuration table, the excitation schedule, the response data matrix, the component attribution table, the response characteristic parameters, the coupling evaluation parameters, and the circulation evaluation parameters are established as report entries according to the same energy storage unit identifier and the same time scale index, so that the multi-timescale response test report simultaneously presents the response speed, steady-state accuracy, coupling status between energy storage units, and branch circulation status.

[0083] When writing a multi-timescale response test report, the test configuration table, excitation schedule, response data matrix, component attribution table, response characteristic parameters, coupling evaluation parameters, and circulation evaluation parameters are compiled into report entries according to the same energy storage unit identifier and the same time scale index. Each report entry in the multi-timescale response test report uses a combination of an energy storage unit identifier and a time scale as an index, under which the corresponding reconstructed response components, response delay, rise time, overshoot, steady-state error, cross-coupling gain, and branch circulation evaluation values ​​are collected.

[0084] like Figure 5 As shown, the response delay, rise time, overshoot, and steady-state error of each energy storage unit are calculated based on the reconstructed response components with added energy storage unit tags. The cross-coupling gain is calculated based on the excitation change and response change of each energy storage unit within the same main excitation window. The branch circulating current evaluation value is calculated based on the absolute value of the difference between the branch currents of each converter at the same timestamp. The results are then written into a multi-timescale response test report according to the same energy storage unit identifier and the same timescale index.

[0085] Using the above method, the multi-timescale response test report can simultaneously present the response speed, steady-state accuracy, coupling state between energy storage units, and branch circulating current state. This allows the multi-timescale response characteristics of different energy storage units in a single composite energy storage power supply to be obtained in a single composite test process, and generates traceable test results containing response parameters, crosstalk indices, and circulating current indices.

[0086] In other embodiments of the present invention, the megawatt-level multi-composite energy storage power supply under test can be a megawatt-level multi-composite energy storage power supply with a grid-connected voltage level of 10kV, and a composite test command is generated according to the allowable power disturbance boundary and allowable bus voltage disturbance boundary corresponding to the 10kV grid-connected voltage level. Alternatively, the megawatt-level multi-composite energy storage power supply under test can be a megawatt-level multi-composite energy storage power supply with a grid-connected voltage level of 35kV, and the amplitude boundaries of the excitation components at each time scale are adjusted according to the allowable bus voltage disturbance boundary corresponding to the 35kV grid-connected voltage level. Alternatively, the megawatt-level multi-composite energy storage power supply under test can include two types of energy storage units with different response time scales, and in this case, the number of response tags in the component attribution table is adjusted according to the actual number of energy storage units included in the megawatt-level multi-composite energy storage power supply under test. Alternatively, when the megawatt-level multi-composite energy storage power supply under test is in a fault protection test scenario, a test window is generated within the allowable power disturbance boundary and allowable bus voltage disturbance boundary, and the response tags, response characteristic parameters, cross-coupling gain, and branch circulating current evaluation values ​​before and after fault triggering are written into the multi-time scale response test report.

[0087] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-timescale response testing method for megawatt-level multi-composite energy storage power sources, characterized in that, include: Obtain the rated power, grid connection voltage level, energy storage unit type and test disturbance boundary of the megawatt-level multi-composite energy storage power supply under test, and generate a test configuration table; A composite excitation signal containing multiple time-scale excitation components is generated according to the test configuration table, and the composite excitation signal is superimposed on the rated power command to obtain a composite test command; During the execution of the composite test command, branch response data and common-side response data of multiple energy storage units are collected simultaneously to form a response data matrix with a unified timestamp. Two-level time-frequency component separation processing is performed on the response data matrix to form a component attribution table corresponding to the multiple energy storage units; Response characteristic parameters, coupling evaluation parameters, and circulation evaluation parameters are generated based on the component attribution table and written into the multi-timescale response test report.

2. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 1, characterized in that, When generating the test configuration table, a channel mapping relationship is established between each energy storage unit identifier and the terminal voltage acquisition channel, branch current acquisition channel, power response acquisition channel, common DC bus voltage acquisition channel, and total output power acquisition channel, and the channel mapping relationship is written into the test configuration table.

3. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 1, characterized in that, When generating the composite excitation signal, short-time pulse component, medium-time fluctuation component and long-time step component are generated according to the test disturbance boundary; Amplitude limiting and timing arrangement are performed on the short-time pulse component, the medium-time fluctuation component, and the long-time step component to form an excitation schedule; The composite excitation signal is obtained by superimposing the excitation components of each time scale according to the excitation schedule.

4. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 3, characterized in that, When forming the response data matrix, the branch response data and the common side response data are uniformly triggered, timestamp aligned and abnormal sampling points are removed according to the excitation schedule, so that each sampling moment corresponds to the terminal voltage, branch current, power response, common DC bus voltage and total output power of multiple energy storage units.

5. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 4, characterized in that, When performing two-level time-frequency component separation processing on the response data matrix, complete set empirical mode decomposition is performed on the total output power or the common DC bus voltage to obtain multiple intrinsic mode function components; Based on the center frequency, energy concentration range, and time span of each intrinsic mode function component, candidate response components that match the short-time pulse component, the medium-time fluctuation component, and the long-time step component are selected.

6. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 5, characterized in that, Wavelet packet decomposition, frequency band filtering, and reconstruction are performed on the candidate response components to obtain multiple reconstructed response components; Based on the matching relationship between the plurality of reconfigured response components and the channel mapping relationship, the plurality of reconfigured response components are respectively mapped to the response tags of the corresponding energy storage units.

7. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 6, characterized in that, When forming the component attribution table, each response tag, corresponding energy storage unit identifier, corresponding time scale, corresponding reconstruction response component, corresponding acquisition channel identifier, and corresponding timestamp interval are associated and written into the component attribution table, so that a traceable mapping is formed between the reconstruction response component and the energy storage unit type.

8. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 7, characterized in that, When generating the response characteristic parameters, the response delay, rise time, overshoot and steady-state error of each energy storage unit are calculated based on the issuance time of the composite test command and the reconstructed response component curve in the component attribution table. The difference between the moment when the preset low-proportion response amplitude is first reached and the moment of transmission is recorded as the response delay, and the difference between the moment when the preset high-proportion response amplitude is first reached and the moment when the preset low-proportion response amplitude is first reached is recorded as the rise time.

9. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 8, characterized in that, When generating the coupling evaluation parameters and the circulation evaluation parameters, a cross-coupling gain is formed based on the excitation change of the target energy storage unit and the response change of the non-target energy storage unit within the main excitation window; The branch circulating current evaluation value is formed based on the difference between the branch currents of different converters at the same timestamp; The cross-coupling gain and the branch circulation evaluation value are associated with the corresponding time scale in the component attribution table.

10. The multi-timescale response testing method for megawatt-level multi-composite energy storage power sources according to claim 9, characterized in that, When writing the multi-timescale response test report, the test configuration table, the excitation schedule, the response data matrix, the component attribution table, the response characteristic parameters, the coupling evaluation parameters, and the circulation evaluation parameters are set up as report entries according to the same energy storage unit identifier and the same time scale index, so that the multi-timescale response test report simultaneously presents the response speed, steady-state accuracy, coupling status between energy storage units, and branch circulation status.

Images