A supercapacitor energy storage control method and system

Abstract

The application relates to the technical field of energy storage control, and discloses a super capacitor energy storage control method and system, wherein the method comprises the following steps: acquiring computing power stage data, electrical operation data and capacitor state data; performing capacitor state identification on the capacitor state data to obtain a capacitor state result; performing change direction extraction on the computing power stage data to obtain a candidate paired event containing a first change direction and a second change direction; performing consistency verification on the candidate paired event based on the electrical operation data to obtain a trusted paired event; and performing margin interval calculation according to the capacitor state result and the trusted paired event to obtain a bidirectional voltage interval. The paired event identification and consistency verification technical scheme can achieve the technical effect of identifying continuous opposite load changes in advance, realize unified reservation of energy storage margins, and solve the problem that control actions are dependent on bus deviation or single-step power changes.

Description

Technical Field

[0001] This invention relates to the field of energy storage control technology, specifically to a supercapacitor energy storage control method and system, electronic equipment, and storage medium. Background Technology

[0002] exist In scenarios involving rack cluster power supply and supercapacitor-assisted energy storage control, task phase switching can trigger paired power changes, such as a sudden drop in load followed by a rebound, or a rapid increase in load followed by a drop. These changes are opposite in direction and occur at short intervals. If compensation is initiated after the bus voltage has already shifted, the supercapacitor may not be able to establish matching power in time, leading to significant fluctuations in the DC bus. Therefore, it is necessary to identify paired load events and arrange charging and discharging margins in advance.

[0003] Currently, existing technologies typically employ bus voltage threshold control, transient power filtering control, or single-step load prediction control. After detecting bus voltage deviation, power difference, or predicted power change, the supercapacitor is controlled to charge or discharge, and the supercapacitor voltage is restored to a fixed intermediate state after transient compensation.

[0004] However, existing technologies have significant limitations in practical applications. On the one hand, control actions rely on existing bus deviations or single-step power changes, making it difficult to identify whether consecutive load changes in opposite directions belong to the same task phase switching event, and it is impossible to reserve energy storage margin in advance. On the other hand, immediately returning to a fixed intermediate voltage after the compensation of the previous change edge will prematurely occupy the reverse direction margin required for the next change edge, and may introduce additional power disturbances to the bus. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a supercapacitor energy storage control method and system, electronic device, and storage medium, to at least partially solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a supercapacitor energy storage control method, comprising:

[0008] Acquire computing power phase data, electrical operation data, and capacitor status data;

[0009] The capacitor state data is used to identify the capacitor state, and the capacitor state result is obtained.

[0010] The computing power stage data is subjected to edge extraction to obtain candidate paired events containing the first edge and the second edge;

[0011] Based on the electrical operation data, the candidate paired events are subjected to consistency verification to obtain reliable paired events;

[0012] Based on the capacitor state results and reliable paired events, a margin interval is calculated to obtain the bidirectional voltage interval.

[0013] The capacitor voltage is preset in the bidirectional voltage range to obtain the preset voltage value;

[0014] Based on the trusted paired events, bidirectional voltage range and preset voltage value, power compensation is performed on the first change edge, and the required energy storage margin in the opposite direction of the second change edge is locked.

[0015] When the second change occurs, the matching margin lock is released and power compensation is performed in the opposite direction to obtain the event compensation amount;

[0016] The capacitor state is restored based on the event compensation amount, and the energy storage control result is generated.

[0017] Preferably, the computational power stage data is subjected to edge extraction to obtain candidate paired events containing a first edge and a second edge, including:

[0018] The computing power stage data is used to identify stage changes and obtain stage switching positions;

[0019] The direction of computing power change between adjacent stages is determined based on the stage switching position, and the change path is obtained;

[0020] According to the time sequence, the changing edges are paired up, and two changing edges that appear in opposite directions and consecutively are respectively identified as the first changing edge and the second changing edge;

[0021] The first and second changing edges are combined to obtain candidate paired events.

[0022] Preferably, based on the electrical operation data, consistency verification is performed on the candidate paired events to obtain trusted paired events, including:

[0023] The occurrence time, direction of change, and magnitude of change of the first and second change edges are determined based on the candidate paired events.

[0024] The electrical operation data is analyzed to identify the direction of change, thereby obtaining the electrical change direction that matches the occurrence time.

[0025] The electrical operating data is analyzed to identify the magnitude of the change, and the electrical change magnitude that matches the occurrence time is obtained.

[0026] The change directions of the first change edge and the second change edge are compared with the electrical change direction for consistency, and the change amplitude is compared with the electrical change amplitude for consistency.

[0027] When both the first and second change edges satisfy the conditions of having the same direction and the same amplitude, the candidate paired events are determined as credible paired events.

[0028] Preferably, a bidirectional voltage range is obtained by calculating the margin range based on the capacitance state results and reliable paired events, including:

[0029] Based on the trusted paired events, determine the power compensation direction of the first change along the matched power compensation direction and the power compensation direction of the second change along the opposite direction of the matched power compensation direction.

[0030] Calculate the required energy storage margin for the first and second transition edges based on the trusted paired events;

[0031] The voltage range within which the capacitor can be used for charging and discharging is determined based on the capacitor status results.

[0032] Using the energy storage margin required for the first and second change edges as constraints, calculate the voltage range that simultaneously satisfies the charging margin and the discharging margin within the voltage range.

[0033] The voltage range that simultaneously satisfies both charging and discharging margins is defined as the bidirectional voltage range.

[0034] Preferably, power compensation is performed on the first change edge, and the required energy storage margin in the opposite direction is locked for the second change edge, including:

[0035] Based on the trusted paired events, determine the direction of change of the first change edge and the opposite direction of change of the second change edge;

[0036] Based on the preset voltage value and the bidirectional voltage range, determine the available power compensation margin for the first change and the required energy storage margin for the second change in the opposite direction.

[0037] When the first change edge occurs, power compensation is performed by calling the power compensation margin according to the change direction of the first change edge;

[0038] During the power compensation process for the first change edge, the energy storage margin required in the opposite direction for the second change edge is maintained within the bidirectional voltage range.

[0039] After the first change is completed along the power compensation, the energy storage margin of the second change is locked in the opposite direction as required.

[0040] Preferably, when the second change occurs, the matching margin lock is released and power compensation is performed in the opposite direction to obtain the event compensation amount, including:

[0041] Based on the credible paired events, determine the occurrence state of the second change edge and the opposite power compensation direction;

[0042] When the occurrence state of the second changing edge meets the compensation condition, the matching margin lock of the second changing edge is released;

[0043] Based on the bidirectional voltage range and the preset voltage value, determine the energy storage margin of the second change along the available opposite direction;

[0044] According to the opposite direction of power compensation, the opposite direction of energy storage margin is called to perform opposite direction power compensation on the second change edge;

[0045] The event compensation amount is obtained based on the power compensation process in the opposite direction of the second change.

[0046] Preferably, the event compensation amount is used to restore the capacitor state and generate energy storage control results, including:

[0047] The direction and degree of deviation of the capacitor state relative to the bidirectional voltage range are determined based on the event compensation amount.

[0048] Based on the capacitor state results and the preset voltage value, determine the recovery direction of the capacitor state recovery;

[0049] After the trusted pairing event is completed, the capacitor state is restored according to the recovery direction, so that the capacitor state is adjusted to the preset voltage value;

[0050] During the capacitor state recovery process, the recovered capacitor state is compared with the bidirectional voltage range;

[0051] When the restored capacitor state meets the bidirectional voltage range or reaches the preset voltage value, the capacitor state recovery stops and the energy storage control result is generated.

[0052] Preferably, calculating the required energy storage margin for the first and second transition edges based on the trusted paired events includes:

[0053] Based on the trusted paired events, the occurrence time, direction of change, and magnitude of change of the first and second change edges are read respectively;

[0054] Based on the direction of change of the first change edge, determine the charging margin or discharging margin that matches the first change edge.

[0055] Based on the direction of change of the second change edge, determine the discharge margin or charging margin that matches the second change edge.

[0056] The change amplitude of the first change edge is accumulated and calculated within the matching time to obtain the energy storage margin required for the first change edge.

[0057] The amplitude of the second change edge is accumulated and calculated within the matching time to obtain the energy storage margin required for the second change edge.

[0058] Preferably, power compensation in the opposite direction along the second change includes:

[0059] Based on the credible paired events, determine the direction and magnitude of the second change along the path of change;

[0060] Based on the direction of change of the second change edge, determine the power compensation direction that is opposite to the power compensation direction of the first change edge;

[0061] The power compensation amount in the opposite direction is determined based on the magnitude of the change along the second change and the available energy storage margin in the opposite direction along the second change.

[0062] According to the opposite power compensation direction and the opposite power compensation amount, the second change edge is subjected to opposite power compensation;

[0063] During the power compensation process at the second changing edge, the compensated capacitor state is maintained within the bidirectional voltage range.

[0064] In a second aspect, the present invention provides a supercapacitor energy storage control system, comprising:

[0065] The data acquisition module is used to acquire computing power stage data, electrical operation data, and capacitor status data;

[0066] The state identification module is used to identify the capacitor state data and obtain the capacitor state result.

[0067] The event extraction module is used to extract the change edges of the computing power stage data to obtain candidate paired events containing the first change edge and the second change edge.

[0068] The event verification module is used to perform consistency verification on the candidate paired events based on the electrical operation data to obtain reliable paired events;

[0069] The interval calculation module is used to perform margin interval calculation based on the capacitor state results and reliable paired events to obtain the bidirectional voltage interval.

[0070] A voltage preset module is used to preset the capacitor voltage in the bidirectional voltage range to obtain a preset voltage value.

[0071] The first compensation module is used to perform power compensation on the first change edge based on the trusted paired events, bidirectional voltage range and preset voltage value, and lock the required energy storage margin in the opposite direction of the second change edge.

[0072] The second compensation module is used to release the matching margin lock and perform power compensation in the opposite direction when the second change occurs, so as to obtain the event compensation amount.

[0073] The state recovery module is used to restore the capacitor state of the event compensation amount and generate energy storage control results.

[0074] Thirdly, the present invention provides an electronic 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 the method in the first aspect.

[0075] Fourthly, the present invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in the first aspect.

[0076] The present invention has the following beneficial effects:

[0077] 1. This invention adopts a pairwise event recognition and consistency verification technology to achieve the technical effect of identifying continuous opposite load changes in advance, realize the unified reservation of energy storage margin, and solve the shortcomings of control actions relying on the already occurred bus deviation or single-step power change.

[0078] 2. The present invention adopts a bidirectional voltage range preset and opposite direction energy storage margin locking technology to achieve the technical effect of retaining the compensation capability of the next changing edge, realize the coordinated compensation between the preceding and following changing edges, and solve the shortcomings of fixed return center occupying the opposite direction margin in advance and introducing power disturbance. Attached Figure Description

[0079] Figure 1 This is a schematic flowchart of the capacitor energy storage control method provided in an embodiment of the present invention;

[0080] Figure 2 This is a framework diagram of a capacitor energy storage control system provided in an embodiment of the present invention. Detailed Implementation

[0081] To enable those skilled in the art to understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0082] The present invention will now be described in detail with reference to the accompanying drawings:

[0083] A specific embodiment of the present invention discloses a supercapacitor energy storage control method, such as... Figure 1 This method is applicable to multiple... Racks, computing server clusters, Training cabinets, high-performance computing nodes, DC bus power supply systems, and energy storage power supply scenarios using supercapacitors for transient power support.

[0084] In this embodiment, the capacitor energy storage control method is used to identify, verify, reserve margins, preset voltages, perform segmented compensation, and restore the state of paired power changes formed by load fluctuations after a sharp drop and subsequent rise, and load fluctuations after a rapid rise, during the switching of computing load events. The method first identifies a first and second transition edge with a sequential correlation based on computing load stage data, verifies the credibility of candidate paired events by combining electrical operation data, and then calculates a bidirectional voltage range based on the capacitor state results. This enables the capacitor to compensate for the power of the first transition edge before or during its occurrence, while simultaneously retaining the energy storage margin required for the second transition edge in the opposite direction.

[0085] In this embodiment, computing power stage data refers to data that can characterize... A collection of data that changes during the operation of a rack or computing node, including task phase identifiers, Power status, Utilization changes, computing power task scheduling status, task batch switching status, training phase switching status, inference request aggregation status, and GPU memory access status. Power consumption trend. Computing power stage data is used to determine whether load power changes originate from the switching process of the same task stage.

[0086] In this embodiment, electrical operating data refers to a set of data that characterizes the operating status of the DC power supply system, including bus voltage, bus current, load power, load power change rate, bus voltage change rate, upstream power output power, and bidirectional power. Converter operating status and Side power change information. Electrical operation data is used to verify the consistency of candidate paired events extracted from computing power stage data, avoiding misjudgment of energy storage compensation demand based solely on computing power scheduling signals.

[0087] In this embodiment, the capacitor status data refers to a set of data that can characterize the current available energy storage capacity and safety boundary of the capacitor, including capacitor terminal voltage, charging and discharging current, module temperature, historical charging and discharging response, current capacitor voltage change, current charging and discharging direction, and historical compensation process.

[0088] In this embodiment, the capacitor state result refers to the set of states obtained by capacitor state data through capacitor state identification. The capacitor state result includes at least the effective capacitance, equivalent series resistance, safe voltage boundary, current capacitor voltage, rechargeable directional margin, dischargeable directional margin, temperature-corrected state, and historical response-corrected state. The capacitor state result is derived from the capacitor state data and is generated by jointly identifying the terminal voltage, current, temperature, and historical response. Its purpose is to determine the capacitor's current ability to absorb or release energy.

[0089] In this embodiment, a trusted paired event refers to the event result formed after candidate paired events are verified by electrical operation data. A trusted paired event includes a confirmed first changing edge, a confirmed second changing edge, occurrence time, changing direction, changing amplitude, power compensation direction, opposite power compensation direction, and event completion status. Trusted paired events originate from candidate paired events and electrical operation data. The generation method involves comparing the direction, amplitude, and occurrence time of the changing edge with the electrical changing direction and electrical changing amplitude in the electrical operation data to eliminate untrusted events.

[0090] In this embodiment, the bidirectional voltage range refers to the voltage range that simultaneously satisfies the energy storage margin required for the capacitor to perform power compensation on the first changing edge and power compensation in the opposite direction on the second changing edge. The bidirectional voltage range is derived from the capacitor state results and reliable paired events. It is generated by calculating based on the energy storage margin required for the first and second changing edges, combined with the voltage range that the capacitor can use for charging and discharging. Its function is to limit the selectable range of preset voltage values ​​and to serve as a constraint for margin locking and state recovery during the power compensation process.

[0091] In this embodiment, the preset voltage value refers to the target voltage determined based on the bidirectional voltage range and used to perform capacitor voltage preset. The preset voltage value is derived from the bidirectional voltage range and is generated by selecting a voltage value within this range that satisfies the first change edge compensation and retains the energy storage margin in the opposite direction of the second change edge. Its function is to ensure that the capacitor is in a bidirectional compensable state before the occurrence of paired events.

[0092] In this embodiment, the event compensation amount refers to the compensation process quantity formed after the second change occurs and power compensation is performed in the opposite direction. The event compensation amount includes the compensation direction of the second change, the compensation power, the compensation duration, the compensation energy, the change in capacitor state, and the degree of deviation of the capacitor state after compensation. The event compensation amount originates from the power compensation process in the opposite direction of the second change and serves to provide a basis for the recovery direction and degree of capacitor state recovery.

[0093] The supercapacitor energy storage control method provided in this embodiment of the invention includes the following steps:

[0094] Acquire computing power stage data, electrical operation data, and capacitor status data.

[0095] In this step, the control system consists of a computing power management platform, Node power management unit, DC bus monitoring unit, bidirectional The converter control unit and capacitor monitoring unit acquire computing power stage data, electrical operation data, and capacitor status data. The acquisition process is aligned according to a unified time base, enabling changes in computing power stage, electrical response, and capacitor status to be analyzed on the same time axis.

[0096] Specifically, computing power stage data may include task stage identifiers, Utilization rate Power status, task batch switching indicator, Power change information and computing power stage timestamps. Electrical operation data may include bus voltage, bus current, load power, bus voltage change rate, load power change rate, and upstream power supply output power. Capacitor status data may include supercapacitor terminal voltage, charging and discharging current, module temperature, and historical response data.

[0097] In one implementation, the collected data is constructed into a unified data sequence based on timestamps: ,

[0098] in, This represents the control input data at the same time. This represents data from the computing power stage. Indicates electrical operating data, This indicates the capacitor status data. Indicates the sampling time.

[0099] The capacitor state data is used to identify the capacitor state, and the capacitor state result is obtained.

[0100] In this step, based on the supercapacitor terminal voltage, charging and discharging current, module temperature, and historical response data, the effective capacitance, equivalent series resistance, safe voltage boundary, and current available energy state of the supercapacitor are identified to obtain the capacitance state results.

[0101] Specifically, outlier values ​​are first removed and time-aligned from the capacitor status data to ensure that the terminal voltage and charging / discharging current are from the same sampling time. Then, the effective capacitor is identified based on the change in terminal voltage and the integral of the charging / discharging current, the equivalent series resistance is identified based on the voltage jump and the current change, and the safety voltage boundary is corrected based on the module temperature and historical response.

[0102] In one implementation, the effective capacitance can be identified by the following relationship: ,

[0103] in, Indicates the currently effective capacitance. Indicates the supercapacitor at time [time]. The charging and discharging current, This represents the voltage across the supercapacitor at the initial moment. This represents the terminal voltage of the supercapacitor at the end time. and Indicates the start and end times used for identification.

[0104] The equivalent series resistance can be estimated using the current step response: ,

[0105] in, This represents the equivalent series resistance. It represents the instantaneous change in terminal voltage caused by a change in current. It represents the change in charging and discharging current.

[0106] The safe voltage boundary is adjusted based on temperature and historical response: , ,

[0107] in, Indicates the lower boundary of the safe voltage. Indicates the upper boundary of the safe voltage. Indicates the lower limit of the rated voltage. Indicates the upper limit of the rated voltage. This indicates the deviation of the module temperature from the reference temperature. This indicates the historical response correction amount. This represents the correction factor.

[0108] Based on the effective capacitance and safe voltage boundary, the current releaseable energy and absorbable energy can be calculated: , ,

[0109] in, Indicates the energy that can be released at present. Indicates the current absorbable energy. This indicates the current terminal voltage of the supercapacitor.

[0110] The capacitor state results are generated from capacitor state data, which includes at least the current terminal voltage, effective capacitance, equivalent series resistance, upper boundary of safe voltage, lower boundary of safe voltage, releaseable energy, and absorbable energy.

[0111] The computing power stage data is subjected to edge extraction to obtain candidate paired events containing the first edge and the second edge.

[0112] In this step, the task stage switching position is identified from the computing power stage data, and the change edge is extracted based on the direction of computing power change between adjacent stages.

[0113] Specifically, the computing power stage data is first analyzed to identify stage changes and determine the stage transition locations. Stage change identification refers to identifying the task stage identifiers in the computing power stage data... Power State and The utilization rate change is used for joint judgment to determine whether the computing load has entered a changing phase from a stable phase. If the computing load phase identifier changes within adjacent time periods, or The power status shows a continuous changing trend, and the corresponding time position is determined as the stage switching position.

[0114] In one implementation, the computing power stage intensity can be expressed as: ,

[0115] in, Indicates the intensity of computing power at each stage. express Utilization rate express Power status, Represents the state variables of a task phase. This represents the weighting coefficient.

[0116] The change in a given stage can be expressed as: ,

[0117] in, This represents the change in computing power between the current time and the previous sampling time. Indicates the sampling interval. If... The change is consistently positive, with the corresponding change edge being an upward change edge; if It continues to show a negative change, with the corresponding change edge being a descending change edge.

[0118] The direction of computing power change between adjacent stages is determined based on the stage switching location, resulting in the change edge. The change edge includes the time of occurrence, direction, and magnitude. The direction of change characterizes whether the computing power load increases or decreases; the magnitude of change characterizes the degree of load change caused by the computing power stage switching.

[0119] The changing edges are paired according to time sequence. If the changing edges occur consecutively and in opposite directions, the changing edge that occurs first in time is identified as the first changing edge, and the changing edge that occurs later in time is identified as the second changing edge. This pairing is used to identify paired events such as a sudden drop in load followed by a rebound or a rapid rise in load followed by a drop, enabling the control system to treat two power changes in opposite directions as a switching process of the same task phase.

[0120] The first and second changing edges are combined to obtain candidate paired events. The candidate paired events include the occurrence time, direction, and magnitude of the first changing edge, the occurrence time, direction, and magnitude of the second changing edge, and the temporal order relationship between the first and second changing edges.

[0121] Based on the electrical operation data, the candidate paired events are subjected to consistency verification to obtain reliable paired events.

[0122] In this step, consistency verification is performed on the first and second changing edges of the candidate paired events based on electrical operation data. This consistency verification is used to confirm whether the changing edges identified by the computing power stage data produce corresponding power changes or bus response changes on the electrical side, so as to avoid miscompensation caused by inconsistencies between computing power scheduling information and actual electrical load.

[0123] Specifically, the occurrence time, direction, and magnitude of the first and second changing edges are first determined based on candidate paired events. Then, the bus voltage change rate, load power change rate, and [other parameters] that match the occurrence time are extracted from the electrical operation data. Power change information. Next, the direction and magnitude of electrical change are identified from the electrical operating data to obtain the electrical change direction and magnitude.

[0124] In one implementation, the load power change rate can be expressed as: ,

[0125] in, Indicates the rate of change of load power. Indicates the current load power. This represents the load power at the previous sampling time.

[0126] The rate of change of bus voltage can be expressed as: ,

[0127] in, This represents the rate of change of bus voltage. Indicates the current bus voltage. This represents the bus voltage at the previous sampling time.

[0128] The direction of electrical change can be determined by combining the load power change rate and the bus voltage change rate. When the load power change rate shows an increase in load and the bus voltage change rate shows a decrease in bus voltage, the direction of electrical change can be determined as the direction of increasing power demand; when the load power change rate shows a decrease in load and the bus voltage change rate shows an increase in bus voltage, the direction of electrical change can be determined as the direction of decreasing power demand.

[0129] Consistency comparison includes directional consistency comparison and amplitude consistency comparison. Directional consistency comparison is used to determine whether the directions of change of the first and second edges are consistent with the electrical change direction, respectively. Amplitude consistency comparison is used to determine whether the change amplitude matches the electrical change amplitude.

[0130] In one implementation, the consistency score can be expressed as: ,

[0131] in, Indicates the consistency score. This indicates the directional consistency evaluation value. This indicates the magnitude consistency evaluation value. This indicates the time-matching evaluation value. This represents the weighting coefficient.

[0132] The directional consistency evaluation value can be expressed as: ,

[0133] in, Indicates the direction of change in candidate pairwise events. It indicates the direction of electrical change in the electrical operation data.

[0134] The amplitude consistency evaluation value can be expressed as: ,

[0135] in, Indicates the change along the candidate pair event The range of change, This indicates the magnitude of electrical changes in the electrical operating data. This represents a correction amount to prevent the denominator from being zero.

[0136] The time-matching evaluation value can be expressed as: ,

[0137] in, Indicates the change along the candidate pair event The time of occurrence, This indicates the time when the corresponding change occurred in the electrical operation data. Indicates the time matching scale.

[0138] When both the first and second edges of the transition satisfy the conditions of consistent direction and amplitude, the candidate paired events are determined as reliable paired events. Reliable paired events are generated jointly from candidate paired events and electrical operating data, including the first edge of the transition, the second edge of the transition, the occurrence time, the direction of change, the amplitude of change, and the reliable state. These reliable paired events are used for margin interval calculation, capacitor voltage preset, and power compensation control.

[0139] Based on the capacitance state results and reliable paired events, a margin interval is calculated to obtain the bidirectional voltage interval.

[0140] In this step, based on the capacitor state results and reliable paired events, a bidirectional voltage range is calculated that simultaneously satisfies the compensation requirements for the first changing edge and the compensation requirements for the second changing edge in the opposite direction. This bidirectional voltage range is used to prevent the supercapacitor from immediately returning to a fixed intermediate voltage after compensation for the first changing edge, thus occupying the energy storage margin required for the second changing edge in the opposite direction.

[0141] Specifically, the power compensation direction along the first change and the opposite direction of the second change are first determined based on the reliable paired events. If the first change corresponds to a rapid increase in the load, the supercapacitor needs to discharge for compensation, and it needs to absorb energy when the load falls back down along the second change. If the first change corresponds to a sharp drop in the load, the supercapacitor needs to absorb energy, and it needs to discharge for compensation when the load rises back up along the second change.

[0142] The required energy storage margin for the first and second edges of the change are calculated based on trusted paired events. The energy storage margin can be calculated cumulatively based on the change amplitude and duration. For each edge... The required energy storage margin can be expressed as: ,

[0143] in, Indicates change along Required energy storage capacity Indicates change along The corresponding power change This indicates that the change occurred along the start time. This indicates the change occurred along the end time.

[0144] The required energy storage margin for the first change and the required energy storage margin for the second change can be expressed as follows: ,

[0145] in, This indicates the required energy storage margin along the first change. This indicates the required energy storage margin along the second change. This indicates the amount of change in power along the first change.

[0146] This indicates the amount of change in power along the second change. These represent the start and end times of the first and second transition edges, respectively.

[0147] The voltage range within which the capacitor can be used for charging and discharging is determined based on the capacitor's state. The capacitor's discharge capability is determined by the energy level above the lower boundary of the safe voltage from the current voltage, while its charge capability is determined by the energy level above the upper boundary of the safe voltage from the current voltage.

[0148] When calculating the bidirectional voltage range, if the capacitor is at voltage... The energy that can be released is: ,

[0149] The absorbable energy is: ,

[0150] in, This indicates the voltage of the capacitor to be evaluated. Indicates the effective capacitance. Indicates the lower boundary of the safe voltage. This indicates the upper boundary of the safe voltage.

[0151] Using the required energy storage margins at the first and second voltage transition edges as constraints, calculate the voltage range that simultaneously satisfies both charging and discharging margins within the voltage range. If discharging is required at the first voltage transition edge and charging is required at the second voltage transition edge, the bidirectional voltage range must satisfy: ,

[0152] If the first voltage change requires charging and the second voltage change requires discharging, the bidirectional voltage range must satisfy the following: ,

[0153] The set of voltages that satisfy the above constraints is defined as a bidirectional voltage range: ,

[0154] in, Indicates a bidirectional voltage range. This indicates that the first change is along the margin constraint. This indicates that the second change is along the margin constraint.

[0155] The bidirectional voltage range is generated jointly by the capacitor state results and reliable paired events, and includes the upper boundary of the range, the lower boundary of the range, the releasable margin, the absorbable margin, and the energy storage margin constraint along the second change in the opposite direction. The bidirectional voltage range is used for subsequent capacitor voltage preset, power compensation along the first change, margin locking along the second change, and state recovery.

[0156] The capacitor voltage is preset in the bidirectional voltage range to obtain the preset voltage value.

[0157] In this step, the control system determines the preset voltage value based on the bidirectional voltage range, and through bidirectional... The converter controls the adjustment of the supercapacitor terminal voltage toward a preset voltage value. The preset voltage value is not a fixed intermediate voltage, but a dynamic target voltage determined based on reliable paired events and bidirectional energy storage margin requirements.

[0158] Specifically, the preset voltage value is selected within the bidirectional voltage range. The selection principle is to simultaneously reduce the voltage deviation before the start of the first change along the compensation, retain the energy storage margin in the opposite direction of the second change, and avoid causing additional disturbance to the bus during the preset process.

[0159] In one implementation, the preset voltage value can be determined by an objective function: ,

[0160] in, Indicates the preset voltage value. Indicates a bidirectional voltage range. This indicates the first change along the compensation preparation deviation. This indicates that the second change retains the deviation along the margin. This indicates a pre-set process disturbance evaluation. This represents the weighting coefficient.

[0161] The evaluation of pre-set process disturbances can be expressed as: ,

[0162] in, This indicates the supercapacitor terminal voltage at the preset start time. This evaluation is used to reduce unnecessary large preset actions.

[0163] The control system generates bidirectional voltage based on the preset voltage value. Converter control command. If the current capacitor voltage is lower than the preset voltage value, bidirectional... The converter controls the charging of the supercapacitor; if the current capacitor voltage is higher than the preset voltage value, it controls the supercapacitor to discharge; if the current capacitor voltage is within the allowable deviation range of the preset voltage value, it maintains the current voltage.

[0164] The preset voltage value is generated from a bidirectional voltage range, including the target voltage value and the preset direction. This preset voltage value is used for subsequent compensation along the first change and in opposite directions along the second change, enabling the supercapacitor to be in a bidirectionally adjustable state when paired events occur.

[0165] Based on the trusted paired events, bidirectional voltage range, and preset voltage value, power compensation is performed on the first change edge, and the required energy storage margin in the opposite direction is locked for the second change edge.

[0166] In this step, when the first changing edge occurs, the control system determines the direction of the first changing edge based on reliable paired events, and uses the available power compensation margin for the first changing edge to perform compensation according to the corresponding power compensation direction. At the same time, it locks the energy storage margin required in the opposite direction for the second changing edge to prevent the first changing edge compensation or subsequent return-to-center operation from prematurely consuming this margin.

[0167] Specifically, the direction of change of the first change edge and the opposite direction of change of the second change edge are first determined based on the reliable paired events. If the first change edge is a rapid increase in load, the compensation direction of the first change edge is the discharge of the supercapacitor, and the energy storage margin in the opposite direction of the second change edge is the charging margin; if the first change edge is a rapid decrease in load, the compensation direction of the first change edge is the charging of the supercapacitor, and the energy storage margin in the opposite direction of the second change edge is the discharging margin.

[0168] Based on the preset voltage value and the bidirectional voltage range, the available power compensation margin for the first changing edge and the required energy storage margin in the opposite direction for the second changing edge are determined. The available power compensation margin for the first changing edge is used for the current compensation action, while the energy storage margin in the opposite direction for the second changing edge remains unavailable during the compensation process of the first changing edge.

[0169] In one implementation, the first change edge can be represented by the compensated energy as follows: ,

[0170] in, This indicates that the first change is along the available compensating energy. This indicates the energy available along the compensation direction at the first change under the preset voltage value.

[0171] The second change along the locking margin can be expressed as: ,

[0172] in, This indicates the energy storage margin in the opposite direction of the second change. This indicates the second change along the required energy storage margin.

[0173] When the first change edge occurs, the control system uses the power compensation margin to perform power compensation according to the direction of the change edge. The compensation power can be determined based on the bus voltage deviation, load power change rate, and the amplitude of the first change edge in the electrical operating data.

[0174] In one implementation, the first change along the compensation power can be expressed as: ,

[0175] in, Indicates the first change along the compensation power. This indicates the first change along the power change. Indicates the bus reference voltage. Indicates the current bus voltage. Indicates the rate of change of load power. This represents the compensation coefficient.

[0176] During the compensation process at the first changing edge, the control system continuously calculates the remaining absorbable and releaseable energy and determines whether the energy storage margin required in the opposite direction for the second changing edge remains within the bidirectional voltage range. If the compensation action would cause the locking margin for the second changing edge to be compromised, the compensation power at the first changing edge is limited to ensure that the energy storage margin in the opposite direction for the second changing edge remains available.

[0177] Locking constraints can be expressed as: ,

[0178] in, This indicates the available energy storage margin along the compensation direction at the current capacitor voltage along the second change. This indicates the second change along the locked energy storage margin.

[0179] After the power compensation at the first changing edge is completed, the supercapacitor is not immediately restored to the fixed intermediate voltage. Instead, the required energy storage margin in the opposite direction for the second changing edge is locked. This locked state is maintained until the second changing edge occurs, a reliable paired event is completed, or the waiting condition ends. In this way, the fixed return-to-center action after compensation at the previous changing edge is prevented from prematurely occupying the margin required for the next changing edge.

[0180] When the second change occurs, the matching margin lock is released and power compensation is performed in the opposite direction to obtain the event compensation amount.

[0181] In this step, after the control system confirms the occurrence of the second changing edge based on the reliable paired event, it releases the margin lock of the second changing edge matching and performs power compensation on the second changing edge in the opposite power compensation direction to obtain the event compensation amount.

[0182] Specifically, the occurrence state of the second change edge and the opposite power compensation direction are first determined based on trusted paired events. The occurrence state can be confirmed by both computing power stage data and electrical operation data; the opposite power compensation direction is determined by the change direction of the second change edge and is opposite to the power compensation direction of the first change edge.

[0183] When the compensation conditions are met during the occurrence of the second changing edge, the matching margin lock for the second changing edge is released. The compensation conditions include that the second changing edge has occurred, the direction of the second changing edge and the reliable paired event record are consistent, and the current capacitor state still has the corresponding compensation capability. After the lock is released, the energy storage margin in the opposite direction that was originally reserved for the second changing edge can be called up.

[0184] Based on the bidirectional voltage range and the preset voltage value, determine the available energy storage margin in the opposite direction along the second change path. If the second change path requires supercapacitor discharge, the release margin is invoked; if the second change path requires supercapacitor charging, the absorption margin is invoked.

[0185] In one implementation, the second change in power compensation along the opposite direction can be expressed as: ,

[0186] in, This indicates that the second change compensates for the power in the opposite direction. This indicates the second change along the power variation. This indicates the remaining energy storage capacity in the opposite direction of the second change. This represents the compensation coefficient.

[0187] The second change, along the available opposite direction, can be expressed as: ,

[0188] in, This indicates the remaining energy storage capacity in the opposite direction of the second change. This indicates the available energy storage margin along the compensation direction at the current capacitor voltage along the second change. This indicates the energy already used for compensation along the second change.

[0189] Power compensation is performed on the second changing edge according to the opposite power compensation direction and the opposite power compensation amount. During the compensation process, the control system monitors the bus voltage, capacitor terminal voltage, charging and discharging current, and compensation power to ensure that the compensated capacitor state remains within the bidirectional voltage range or within the allowable range of the safe voltage boundary.

[0190] The event compensation amount is obtained from the power compensation process in the opposite direction of the second change, and can be expressed as: ,

[0191] in, Indicates the amount of compensation for the event. This indicates that the second change compensates for the power in the opposite direction. This indicates the time when the second change begins along the compensation process. This indicates the second change along the compensation end time.

[0192] The event compensation parameters include the compensation direction, compensation power, compensation energy, compensation duration, and the capacitor state after compensation. These parameters are used for subsequent capacitor state recovery, determining the recovery direction, recovery extent, and recovery termination conditions.

[0193] The capacitor state is restored based on the event compensation amount, and the energy storage control result is generated.

[0194] In this step, the control system determines the direction and degree of deviation of the capacitor state relative to the bidirectional voltage range based on the event compensation amount, and performs a constrained recovery of the capacitor state after the completion of the credible paired events, generating the energy storage control result. This recovery is not immediately fixed back to the center after the first changing edge compensation, but is performed after the paired events are completed, based on the current event compensation amount, the bidirectional voltage range, and the preset voltage value.

[0195] Specifically, the direction and degree of deviation of the capacitor state relative to the bidirectional voltage range are first determined based on the event compensation amount. If the capacitor voltage after compensation is higher than the bidirectional voltage range after the second change, the recovery direction is the discharge direction; if the capacitor voltage is lower than the bidirectional voltage range, the recovery direction is the charging direction; if the capacitor voltage is within the bidirectional voltage range, it is maintained or slowly adjusted towards the preset voltage value.

[0196] The degree of deviation can be expressed as: ,

[0197] in, Indicates the degree of deviation of the capacitor's state. Indicates the lower boundary of the bidirectional voltage range. Indicates the upper boundary of the bidirectional voltage range. This indicates the current terminal voltage of the supercapacitor.

[0198] The recovery direction of the capacitor state is determined based on the capacitor status results and the preset voltage value. The recovery process should meet the bus operation constraints and avoid introducing significant bus power disturbances during the recovery action. The recovery power can be limited based on the bus's remaining power capacity, the current capacitor state, and the recovery direction.

[0199] In one implementation, the recovery power can be expressed as: ,

[0200] in, Indicates the recovery power. Represents the coefficient of restitution. Indicates the preset voltage value. This indicates the current terminal voltage of the supercapacitor. Indicates the lower limit of recovery power. Indicates the upper limit of recovery power. This indicates that the image is being limited.

[0201] After a trusted paired event is completed, the control system restores the capacitor state according to the recovery direction, adjusting the capacitor state towards a preset voltage value. During the recovery process, the system continuously compares the restored capacitor state with the bidirectional voltage range. When the restored capacitor state meets the bidirectional voltage range or reaches the preset voltage value, the capacitor state recovery stops.

[0202] The energy storage control results are generated jointly by event compensation amounts, capacitor state results, bidirectional voltage ranges, and preset voltage values. They include at least paired event identification states, reliable verification states, first-edge compensation records, second-edge compensation records, margin lock records, capacitor state recovery records, and the final capacitor state. These results guide subsequent supercapacitor energy storage control, compensation parameter adjustments, and control threshold optimization.

[0203] Phase change identification refers to extracting the locations where changes occur in task or load phases from computing power phase data. The control system first smooths the computing power phase data to reduce the impact of single-point fluctuations on change phase identification; it then calculates the change in computing power phase intensity within adjacent time windows; when the change consistently maintains the same direction and exceeds preset identification conditions, the corresponding location is determined as the phase switch location. The phase switch location is generated from the computing power phase data, including the switch time, the phase state before the switch, and the phase state after the switch. This phase switch location is used to determine the direction of computing power change between adjacent phases.

[0204] When determining the direction of computing power change between adjacent stages based on the stage switching location, the control system compares the computing power stage intensity before and after the switch. If the computing power stage intensity after the switch is higher than before, the direction of change is upward; if the computing power stage intensity after the switch is lower than before, the direction of change is downward. The change path is generated by the stage switching location and the direction of change, including the occurrence time, direction of change, and magnitude of change.

[0205] When pairing transition edges, the control system scans the transition edge sequence in chronological order. When two consecutive transition edges appear in opposite directions, the earlier one is designated as the first transition edge, and the later one as the second. Consecutive occurrence means that no independent transition edges unrelated to the task phase are inserted between them, and both belong to the same computing power phase switching link. The first and second transition edges are combined to form a candidate pair event, which is used for subsequent electrical consistency verification.

[0206] Electrical change direction identification refers to determining the actual direction of load change from electrical operating data. The control system can determine this based on the load power change rate, bus voltage change rate, and... Power change information is used for joint judgment. If the load power increases and the bus voltage shows a downward trend, the direction of electrical change matches the increase in load; if the load power decreases and the bus voltage shows an upward trend, the direction of electrical change matches the decrease in load. This direction of electrical change is generated from electrical operation data and is used for consistency comparison with the direction of change in candidate paired events.

[0207] Electrical change amplitude identification refers to calculating the degree of electrical power change that matches the occurrence times of the first and second change edges. The control system can select a matching time window near the occurrence time to calculate the load power change, the peak value of the bus voltage change rate, or... Power variation amplitude. The electrical variation amplitude is generated from electrical operating data and is used for consistency comparison with the variation amplitudes in candidate paired events.

[0208] Consistency matching includes time matching, direction matching, and amplitude matching. Time matching confirms that the electrical change occurs near the time corresponding to the candidate change edge; direction matching confirms that the change direction on the computing side and the electrical side are consistent; amplitude matching confirms that the change amplitude on the computing side and the electrical side are consistent. When both the first and second change edges satisfy the conditions of consistent direction and consistent amplitude, the candidate paired events are determined as reliable paired events. Reliable paired events are used for subsequent margin interval calculations to avoid unreliable events triggering unnecessary energy storage actions.

[0209] The power compensation direction is determined based on the direction of change in credible paired events. When the load increases, the supercapacitor needs to discharge to compensate for insufficient response from the upstream power source; when the load decreases, the supercapacitor needs to charge to absorb excess energy and suppress the rise in bus voltage. The first change along the matched power compensation direction and the second change along the opposite direction of the matched power compensation direction together determine the calculation constraints of the bidirectional voltage range.

[0210] The energy storage margin required for the first and second change edges is calculated based on the occurrence time, direction, and amplitude of the reliable paired events. The control system accumulates the amplitude over the corresponding occurrence time to obtain the energy required for the corresponding change edge. If the change edge corresponds to a load increase, this energy storage margin corresponds to a discharge margin; if the change edge corresponds to a load decrease, this energy storage margin corresponds to a charging margin. This energy storage margin is used to constrain the bidirectional voltage range, ensuring that any selectable voltage within the range can meet the compensation requirements before and after the paired events.

[0211] The voltage range within which a capacitor can be used for charging and discharging is determined by the capacitor state results. These results provide the effective capacitance, the safe voltage boundary, and the current terminal voltage. The control system calculates the absorbable and release energy within the safe voltage boundary, using the required energy storage margin at the first and second voltage transition edges as constraints, to obtain the voltage range that simultaneously satisfies both charging and discharging margins. This voltage range is the bidirectional voltage interval.

[0212] For the first change-edge power compensation and the second change-edge margin locking, the control system first determines the change direction of the first change-edge and the opposite change direction of the second change-edge based on reliable paired events. Then, based on the preset voltage value and the bidirectional voltage range, it calculates the currently available power compensation margin for the first change-edge and the required energy storage margin in the opposite direction for the second change-edge. The available power compensation margin for the first change-edge is used to respond to the current load change, while the required energy storage margin in the opposite direction for the second change-edge remains locked.

[0213] At the first change edge, the control system allocates power compensation margin according to the direction of change. If the first change edge corresponds to a load increase, the supercapacitor discharges to the bus; if the first change edge corresponds to a load decrease, the supercapacitor absorbs energy from the bus. During compensation, the control system continuously monitors whether the capacitor terminal voltage still meets the lock-in margin requirement for the second change edge. Continuing compensation would encroach on the energy storage margin required in the opposite direction for the second change edge, limiting the compensation power for the first change edge or shortening the compensation duration.

[0214] After the power compensation for the first changing edge is completed, the required energy storage margin for the second changing edge in the opposite direction is locked. This locked state prevents the supercapacitor from immediately returning to a fixed intermediate voltage, causing additional bus power disturbance, and also prevents premature consumption of the required charging or discharging margin for the second changing edge.

[0215] For power compensation along the second change edge in the opposite direction, the control system first determines whether the second change edge has occurred based on reliable paired events. The occurrence of the second change edge can be confirmed by the stage switching in the computing power stage data and the direction change in the electrical operation data. When the occurrence of the second change edge meets the compensation conditions, the control system releases the corresponding matching margin lock, making the previously reserved energy storage margin in the opposite direction available for use.

[0216] The control system determines the available energy storage margin in the opposite direction for the second voltage change based on the bidirectional voltage range and the preset voltage value. If the second voltage change occurs during load recovery, the supercapacitor utilizes its discharge margin; if the second voltage change occurs during load decline, it utilizes its charging margin. The control system utilizes this energy storage margin in the opposite direction of power compensation to perform power compensation for the second voltage change.

[0217] The event compensation amount is obtained from the power compensation process in the opposite direction of the second change edge. This event compensation amount includes the compensation energy, compensation power, compensation time, and the degree of deviation of the capacitor state after compensation. The event compensation amount is used for subsequent capacitor state recovery to ensure that the recovery action is performed based on the actual compensation result, rather than restoring indiscriminately according to a fixed intermediate voltage.

[0218] For capacitor state recovery, the control system determines the direction and degree of deviation of the capacitor state relative to the bidirectional voltage range based on the event compensation amount. If the event compensation amount results in a higher capacitor voltage, it indicates that the capacitor has absorbed more energy, and the recovery direction is the discharge direction; if the event compensation amount results in a lower capacitor voltage, it indicates that the capacitor has released more energy, and the recovery direction is the charging direction. The recovery direction is jointly determined by the event compensation amount, the capacitor state result, and the preset voltage value.

[0219] After a trusted paired event occurs, the control system restores the capacitor state according to the recovery direction. During recovery, instead of directly and rapidly returning to a fixed intermediate voltage, the capacitor state is adjusted towards a preset voltage value within the bidirectional voltage range and under the allowable conditions of the bus operation. The control system continuously compares the restored capacitor state with the bidirectional voltage range, and stops recovery when the restored capacitor state satisfies the bidirectional voltage range or reaches the preset voltage value. This constrained recovery method avoids introducing additional power disturbances to the bus during the recovery process.

[0220] For calculating the energy storage margin, the control system reads the occurrence time, direction, and amplitude of the first and second changing edges based on reliable paired events. The direction of change is used to determine whether the corresponding changing edge requires a charging margin or a discharging margin, while the amplitude and occurrence time are used to calculate the required energy storage margin. If the direction of change of the first changing edge corresponds to a decrease in load, the first changing edge is matched with a charging margin; if the direction of change of the first changing edge corresponds to an increase in load, the first changing edge is matched with a discharging margin.

[0221] When calculating the cumulative change in magnitude over the time of matching, the control system can use discrete sampled data for calculation: ,

[0222] in, Indicates change along Required energy storage capacity Indicates sampling point The corresponding power change Indicates the sampling interval. and This indicates the start and end sampling points within the matching time. In this way, the energy storage margins of both the first and second transition edges are derived from the time, direction, and amplitude of the reliable paired events, and the calculation results are used to construct the bidirectional voltage range.

[0223] For power compensation along the second change edge in the opposite direction, the control system determines the direction and magnitude of the second change edge based on reliable paired events. Since the direction of the second change edge is opposite to that of the first change edge, the power compensation direction is also opposite to that of the first change edge. If the first change edge uses discharge compensation, the second change edge uses charge compensation; conversely, if the first change edge uses charge compensation, the second change edge uses discharge compensation.

[0224] The control system determines the power compensation amount in the opposite direction based on the amplitude of the second voltage change and the available energy storage margin in the opposite direction at the second voltage change. If the required compensation amount for the second voltage change is less than the available energy storage margin in the opposite direction, compensation is made according to actual needs; if the required compensation amount for the second voltage change is greater than the available energy storage margin in the opposite direction, compensation is limited according to the available energy storage margin in the opposite direction. During the compensation process, the control system continuously monitors the capacitor state after compensation to keep it within the bidirectional voltage range or the safe voltage boundary, avoiding the capacitor state from going out of bounds due to compensation at the second voltage change.

[0225] This embodiment provides a supercapacitor energy storage control system, such as Figure 2 The system includes a data acquisition module, a state identification module, an event extraction module, an event verification module, an interval calculation module, a voltage preset module, a first compensation module, a second compensation module, and a state recovery module.

[0226] The data acquisition module is used to acquire computing power stage data, electrical operation data, and capacitor status data. This module can be integrated with the computing power scheduling platform, Power monitoring unit, bus sampling unit, bidirectional The converter control unit and the supercapacitor monitoring unit are communicatively connected. The data acquisition module performs time alignment and field organization on the collected data, providing input for the status identification module, event extraction module, and event verification module.

[0227] The state identification module is used to identify the capacitor state data and obtain the capacitor state results. Based on the supercapacitor terminal voltage, charging / discharging current, module temperature, and historical response data, the state identification module identifies the effective capacitance, equivalent series resistance, and safe voltage boundary, and calculates the absorbable and release energy. The capacitor state results output by the state identification module are used by the interval calculation module, voltage preset module, second compensation module, and state recovery module.

[0228] The event extraction module is used to extract transition edges from the computing power stage data, obtaining candidate paired events containing a first transition edge and a second transition edge. The event extraction module performs stage transition identification, stage switch location determination, change direction judgment, transition edge generation, and transition edge pairing. Candidate paired events are output by the event extraction module and invoked by the event verification module.

[0229] The event verification module is used to verify the consistency of candidate paired events based on electrical operation data, obtaining reliable paired events. The module extracts the occurrence time, direction of change, and magnitude of change from the candidate paired events, and then extracts the matching electrical change direction and magnitude from the electrical operation data, comparing direction and magnitude consistency. The event verification module outputs reliable paired events, which are then invoked by the interval calculation module, the first compensation module, and the second compensation module.

[0230] The interval calculation module is used to calculate the margin interval based on the capacitor state results and reliable paired events, obtaining a bidirectional voltage interval. The interval calculation module calculates the required energy storage margin for the first and second changing edges, and, combined with the voltage range that the capacitor can use for charging and discharging, determines the bidirectional voltage interval that simultaneously satisfies the charging and discharging margins. The bidirectional voltage interval is invoked by the voltage preset module, the first compensation module, the second compensation module, and the state recovery module.

[0231] The voltage preset module is used to preset the capacitor voltage within the bidirectional voltage range to obtain a preset voltage value. The voltage preset module determines the preset voltage value within the bidirectional voltage range and controls the bidirectional... The converter charges or discharges the supercapacitor, adjusting the capacitor terminal voltage to near a preset voltage value. This preset voltage value is then used by the first compensation module, the second compensation module, and the state recovery module.

[0232] The first compensation module performs power compensation on the first voltage change edge based on trusted paired events, bidirectional voltage ranges, and preset voltage values, while locking the required energy storage margin in the opposite direction for the second voltage change edge. The first compensation module utilizes the available compensation margin when the first voltage change edge occurs, while simultaneously ensuring that the energy storage margin in the opposite direction for the second voltage change edge is not occupied. This module addresses the problem of insufficient margin in the second voltage change edge caused by immediately fixing the voltage back to the center after compensation for the first voltage change edge.

[0233] The second compensation module is used to unlock the matching margin lock and perform power compensation in the opposite direction when the second changing edge occurs, obtaining the event compensation amount. After confirming the occurrence of the second changing edge, the second compensation module converts the previously locked opposite-direction energy storage margin into a usable margin and performs power compensation according to the compensation direction of the second changing edge. The second compensation module outputs the event compensation amount, which is then called by the state recovery module.

[0234] The state recovery module is used to restore the capacitor state based on the event compensation amount and generate energy storage control results. The state recovery module determines the direction and degree of capacitor state deviation based on the event compensation amount, and performs limited recovery after the completion of a reliable paired event, adjusting the capacitor state towards a preset voltage value or a bidirectional voltage range. The state recovery module outputs the energy storage control results for subsequent control recording, parameter correction, and operational optimization.

[0235] The energy storage control results in this embodiment include trusted paired event identification information, bidirectional voltage range, preset voltage value, first change-edge compensation process, second change-edge margin locking process, second change-edge opposite direction compensation process, event compensation amount, and capacitor state recovery process. These energy storage control results are derived step-by-step from computing power stage data, electrical operation data, capacitor state data, capacitor state results, candidate paired events, trusted paired events, bidirectional voltage range, preset voltage value, and event compensation amount. These energy storage control results can be used to guide the operation of supercapacitors. Real-time charge / discharge control, bus voltage stabilization control, and bidirectional control in the rack power supply system The converter control parameters are optimized and subsequent event predictions are corrected. This energy storage control solution addresses the shortcomings of existing control actions that rely on pre-existing bus deviations or single-step power changes, making it difficult to identify paired load events. It also addresses the issue that immediately fixing the return to center after the first changing edge compensation occupies the required reverse-direction margin for the second changing edge. This energy storage control solution demonstrates a comprehensive technical effect: early identification of paired events, unified reservation of bidirectional energy storage margin, coordinated compensation of the first and second changing edges, reduced bus voltage disturbances, and improved supercapacitor energy storage utilization efficiency.

[0236] Embodiments of the present invention have been presented and described. It will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0237] Those skilled in the art should understand that the embodiments of the present invention can be implemented using a pure hardware architecture, a pure software architecture, or an integrated hardware and software architecture. The present invention can be prepared as a computer program product, which can be stored in various non-volatile computer-readable storage media, including but not limited to solid-state drives, flash memory chips, mobile storage devices, optical discs, cloud storage servers, and other standardized storage media, and is not limited to traditional storage media.

[0238] The above description describes embodiments of the present invention. Without departing from the embodiments and broad aspects of the present invention, those skilled in the art can make data modifications and method changes based on this description in specific operations. The appended claims are intended to include all data modifications and method changes that do not depart from the embodiments of the present invention.

[0239] Based on the foregoing description in conjunction with the accompanying drawings, those skilled in the art will understand that the embodiments of this application can be implemented by software programs. Therefore, this application also provides a computer-readable storage medium. This computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the method described above in conjunction with the accompanying drawings.

[0240] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, or by hardware. The technical solutions described above, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as... Disks, optical disks, etc., including a number of instructions to cause a computer device to perform the methods described in the various embodiments or some parts of the embodiments.

[0241] It should be understood that when the terms "first," "second," "third," and "fourth," etc., are used in the claims, specification, and drawings of this application, they are used to distinguish different objects, not to describe a specific order. The terms "comprising" and "including" as used in the specification and claims of this application indicate the presence of the described features, units, steps, operations, elements, and / or components, but do not exclude the presence or addition of other features, units, steps, operations, elements, components, and / or sets.

[0242] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments and is not intended to limit the application. As used in this specification and claims, the singular forms “a” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should be further understood that the term “and / or” as used in this specification and claims refers to any combination of one or more of the associated listed items, and all possible combinations thereof.

[0243] The embodiments of this application are as described above, but the content described is merely an example for the purpose of understanding this application and is not intended to limit the scope or application scenarios of this application. Any person skilled in the art described in this application may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application, but the scope of patent protection of this application shall still be determined by the scope defined in the appended claims.

Claims

1. A method for controlling energy storage in a supercapacitor, characterized in that, include: Acquire computing power phase data, electrical operation data, and capacitor status data; The capacitor state data is used to identify the capacitor state, and the capacitor state result is obtained. The computing power stage data is subjected to edge extraction to obtain candidate paired events containing the first edge and the second edge; Based on the electrical operation data, the candidate paired events are subjected to consistency verification to obtain reliable paired events; Based on the capacitor state results and reliable paired events, a margin interval is calculated to obtain the bidirectional voltage interval. The capacitor voltage is preset in the bidirectional voltage range to obtain the preset voltage value; Based on the trusted paired events, bidirectional voltage range and preset voltage value, power compensation is performed on the first change edge, and the required energy storage margin in the opposite direction of the second change edge is locked. When the second change occurs, the matching margin lock is released and power compensation is performed in the opposite direction to obtain the event compensation amount; The event compensation amount is used to restore the capacitor state and generate energy storage control results; Based on the electrical operation data, consistency verification is performed on the candidate paired events to obtain trusted paired events, including: The occurrence time, direction of change, and magnitude of change of the first and second change edges are determined based on the candidate paired events. The electrical operation data is analyzed to identify the direction of change, thereby obtaining the electrical change direction that matches the occurrence time. The electrical operating data is analyzed to identify the magnitude of the change, and the electrical change magnitude that matches the occurrence time is obtained. The change directions of the first change edge and the second change edge are compared with the electrical change direction for consistency, and the change amplitude is compared with the electrical change amplitude for consistency. When both the first and second change edges satisfy the conditions of having the same direction and the same amplitude, the candidate paired events are determined as credible paired events; Based on the capacitance state results and reliable paired events, a margin interval is calculated to obtain a bidirectional voltage interval, including: Based on the trusted paired events, determine the power compensation direction of the first change along the matched power compensation direction and the power compensation direction of the second change along the opposite direction of the matched power compensation direction. Calculate the required energy storage margin for the first and second transition edges based on the trusted paired events; The voltage range within which the capacitor can be used for charging and discharging is determined based on the capacitor status results. Using the energy storage margin required for the first and second change edges as constraints, calculate the voltage range that simultaneously satisfies the charging margin and the discharging margin within the voltage range. The voltage range that simultaneously satisfies both charging and discharging margins is defined as the bidirectional voltage range.

2. The capacitor energy storage control method according to claim 1, characterized in that, The computational power stage data is subjected to edge extraction to obtain candidate paired events containing a first edge and a second edge, including: The computing power stage data is used to identify stage changes and obtain stage switching positions; The direction of computing power change between adjacent stages is determined based on the stage switching position, and the change path is obtained; According to the time sequence, the changing edges are paired up, and two changing edges that appear in opposite directions and consecutively are respectively identified as the first changing edge and the second changing edge; The first and second changing edges are combined to obtain candidate paired events.

3. The capacitor energy storage control method according to claim 1, characterized in that, Performing power compensation on the first change edge and locking the required energy storage margin in the opposite direction for the second change edge includes: Based on the trusted paired events, determine the direction of change of the first change edge and the opposite direction of change of the second change edge; Based on the preset voltage value and the bidirectional voltage range, determine the available power compensation margin for the first change and the required energy storage margin for the second change in the opposite direction. When the first change edge occurs, power compensation is performed by calling the power compensation margin according to the change direction of the first change edge; During the power compensation process for the first change edge, the energy storage margin required in the opposite direction for the second change edge is maintained within the bidirectional voltage range. After the first change is completed along the power compensation, the energy storage margin of the second change is locked in the opposite direction as required.

4. The capacitor energy storage control method according to claim 1, characterized in that, When the second change occurs, the matching margin lock is released and power compensation is performed in the opposite direction to obtain the event compensation amount, including: Based on the credible paired events, determine the occurrence state of the second change edge and the opposite power compensation direction; When the occurrence state of the second changing edge meets the compensation condition, the matching margin lock of the second changing edge is released; Based on the bidirectional voltage range and the preset voltage value, determine the energy storage margin of the second change along the available opposite direction; According to the opposite direction of power compensation, the opposite direction of energy storage margin is called to perform opposite direction power compensation on the second change edge; The event compensation amount is obtained based on the power compensation process in the opposite direction of the second change.

5. The capacitor energy storage control method according to claim 1, characterized in that, The event compensation amount is used to restore the capacitor state and generate energy storage control results, including: The direction and degree of deviation of the capacitor state relative to the bidirectional voltage range are determined based on the event compensation amount. Based on the capacitor state results and the preset voltage value, determine the recovery direction of the capacitor state recovery; After the trusted pairing event is completed, the capacitor state is restored according to the recovery direction, so that the capacitor state is adjusted to the preset voltage value; During the capacitor state recovery process, the recovered capacitor state is compared with the bidirectional voltage range; When the restored capacitor state meets the bidirectional voltage range or reaches the preset voltage value, the capacitor state recovery stops and the energy storage control result is generated.

6. The capacitor energy storage control method according to claim 1, characterized in that, Calculate the required energy storage margin for the first and second transition edges based on the trusted paired events, including: Based on the trusted paired events, the occurrence time, direction of change, and magnitude of change of the first and second change edges are read respectively; Based on the direction of change of the first change edge, determine the charging margin or discharging margin that matches the first change edge. Based on the direction of change of the second change edge, determine the discharge margin or charging margin that matches the second change edge. The change amplitude of the first change edge is accumulated and calculated within the matching time to obtain the energy storage margin required for the first change edge. The amplitude of the second change edge is accumulated and calculated within the matching time to obtain the energy storage margin required for the second change edge.

7. The capacitor energy storage control method according to claim 4, characterized in that, Performing power compensation in the opposite direction along the second change includes: Based on the credible paired events, determine the direction and magnitude of the second change along the path of change; Based on the direction of change of the second change edge, determine the power compensation direction that is opposite to the power compensation direction of the first change edge; The power compensation amount in the opposite direction is determined based on the magnitude of the change along the second change and the available energy storage margin in the opposite direction along the second change. According to the opposite power compensation direction and the opposite power compensation amount, the second change edge is subjected to opposite power compensation; During the power compensation process at the second changing edge, the compensated capacitor state is maintained within the bidirectional voltage range.

8. A supercapacitor energy storage control system, characterized in that, include: The data acquisition module is used to acquire computing power stage data, electrical operation data, and capacitor status data; The state identification module is used to identify the capacitor state data and obtain the capacitor state result. The event extraction module is used to extract the change edges of the computing power stage data to obtain candidate paired events containing the first change edge and the second change edge. The event verification module is used to extract the occurrence time, change direction, and change amplitude from the candidate paired events. It extracts the electrical change direction and electrical change amplitude that match the occurrence time from the electrical operation data, compares the change direction with the electrical change direction for direction consistency, and compares the change amplitude with the electrical change amplitude for amplitude consistency. Based on the comparison results, it obtains the reliable paired events. The interval calculation module is used to calculate the required energy storage margin for the first change edge and the required energy storage margin for the second change edge according to the trusted paired events, determine the voltage range in which the capacitor can be used for charging and discharging according to the capacitor state results, and determine a bidirectional voltage interval that simultaneously satisfies the charging margin and the discharging margin in the voltage range by combining the required energy storage margin for the first change edge and the second change edge. A voltage preset module is used to preset the capacitor voltage in the bidirectional voltage range to obtain a preset voltage value. The first compensation module is used to perform power compensation on the first change edge based on the trusted paired events, bidirectional voltage range and preset voltage value, and lock the required energy storage margin in the opposite direction of the second change edge. The second compensation module is used to release the matching margin lock and perform power compensation in the opposite direction when the second change occurs, so as to obtain the event compensation amount. The state recovery module is used to restore the capacitor state of the event compensation amount and generate energy storage control results.

Images