Data center direct current power supply load storage cooperative control method, equipment and medium

Abstract

The invention discloses a data center direct current power supply load storage cooperative control method and device and a medium, and belongs to the technical field of direct current power supply control. Comprising the steps of extracting a charge state value at an event occurrence moment, a time integral value of a charge and discharge power absolute value within a preset time period before the event occurrence moment and a time constant of the discharge power rising from an initial value to a peak value after the event occurrence, and establishing a three-dimensional mapping relation; constructing an energy storage system transient response capability evaluation function according to the three-dimensional mapping relation, and predicting a time constant; substituting the predicted value of the time constant, the bus equivalent capacitance value and the preset maximum load step amplitude into a voltage dynamic response equation to solve and obtain a voltage drop depth predicted value; and when the voltage drop depth estimated value exceeds a voltage collapse threshold value, calculating a dynamic adjustment margin value, and calling a distributed power supply and load adjustment to enable the charge state value to move to a safe area. According to the invention, dynamic quantitative evaluation of the transient response capability of the energy storage system is realized, and the bus voltage collapse risk is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of direct current power supply control, and in particular relates to a data center direct current power supply source-load-storage collaborative control method, device and medium. BACKGROUND

[0002] As a high energy consumption infrastructure, the power supply reliability of a data center is crucial. The traditional data center adopts an alternating current power supply architecture, which has many energy conversion links and large energy loss. A direct current power supply architecture combined with an energy storage system can significantly improve efficiency, but faces the risk of voltage drop caused by load step. Dynamic migration of virtual machines leads to sudden changes in load, and the response delay of the energy storage system may cause voltage collapse, affecting business continuity. Existing technologies mainly rely on over-provisioning of energy storage capacity or setting conservative safety margins, resulting in low utilization of energy storage resources and inflexible regulation strategies. In addition, the characteristics of energy storage dynamically change with state of charge, historical operation trajectory, and aging degree, and a fixed model cannot accurately predict the response capability, so there is an urgent need for a data center direct current power supply source-load-storage collaborative control method. SUMMARY

[0003] In view of the above problems, the present application provides a data center direct current power supply source-load-storage collaborative control method, device and medium.

[0004] Therefore, the technical problem solved by the present application is that the existing data center direct current power supply system has the following problems: the instantaneous response capability of the energy storage system dynamically changes with the operating state, and the state of charge threshold alone cannot accurately reflect the real ability of the system to resist voltage collapse; there is a lack of quantitative description of the relationship between the response time constant of the energy storage system and the operating history and current state; the potential risk of insufficient response capability of the energy storage system cannot be identified in advance; there is a lack of source-load-storage collaborative regulation mechanism based on dynamic response capability, resulting in untimely or overly conservative voltage collapse prevention measures.

[0005] To solve the above technical problems, the present application provides the following technical solutions: a data center direct current power supply source-load-storage collaborative control method, comprising: collecting a direct current bus voltage sequence, an energy storage system charge and discharge power sequence and a state of charge sequence, identifying a voltage step drop event in the voltage sequence, extracting, for each voltage step drop event, a state of charge value at the event occurrence time, a time integral value of the absolute value of the charge and discharge power in a preset period before the event occurrence time, and a time constant of the discharge power rising from an initial value to a peak value after the event occurrence, and establishing a three-dimensional mapping relationship between the state of charge value at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in the preset period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence. According to the three-dimensional mapping relationship, a transient response capability evaluation function of the energy storage system is constructed, the state of charge value of the energy storage system at the current time and the time integral value of the absolute value of the charging and discharging power in the preset time period before the current time of the energy storage system are input into the transient response capability evaluation function of the energy storage system, and a predicted value of the time constant is obtained; The predicted value of the time constant, the equivalent capacitance value of the bus and the preset maximum load step amplitude value are substituted into the voltage dynamic response equation to obtain a voltage drop depth estimation value; When the voltage drop depth estimation value exceeds the voltage drop depth safety threshold, the target value of the time constant is obtained by inversely solving the voltage dynamic response equation, and the state of charge value required to reach the target value of the time constant is inversely searched according to the three-dimensional mapping relationship, and the difference between the inversely searched state of charge value and the current state of charge value is taken as a dynamic adjustment margin value; When the state of charge value of the energy storage system at the current time is less than the dynamic adjustment margin value from the state of charge boundary, the distributed power and the load are adjusted to move the state of charge value to a region that meets the requirement of the dynamic adjustment margin value; During the adjustment process, the actual value of the time constant is monitored, the deviation between the actual value of the time constant and the predicted value of the time constant is calculated, and the three-dimensional mapping relationship is updated according to the deviation by using the recursive least square method.

[0006] As a preferred scheme of the data center direct current power supply and storage collaborative control method, the method comprises the steps of: identifying a voltage step drop event in a voltage sequence, performing first-order time difference operation on the direct current bus voltage sequence to obtain a voltage change rate sequence; Identifying the time point when the absolute value of the voltage change rate in the voltage change rate sequence exceeds the preset change rate threshold, determining that the voltage change direction at the time point is negative, and taking the time point as a candidate voltage drop time point; Extracting a power reference value of the charging and discharging power sequence of the energy storage system before the candidate voltage drop time point, extracting a power peak value of the charging and discharging power sequence of the energy storage system within a preset verification time window after the candidate voltage drop time point, determining whether a step rise occurs with respect to the power reference value, and if the step rise occurs, confirming the candidate voltage drop time point as the voltage step drop event time point; According to the timestamp of the voltage step drop event time point, the state of charge value corresponding to the timestamp in the state of charge sequence is searched as the state of charge value at the voltage step drop event time point.

[0007] As a preferred scheme of the data center direct current power supply charge storage collaborative control method, the three-dimensional mapping relationship among the state of charge at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in the preset time period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence includes that the state of charge at the event occurrence time is taken as the first-dimensional coordinate, the time integral value of the absolute value of the charge and discharge power in the preset time period before the event occurrence time is taken as the second-dimensional coordinate, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence is taken as the third-dimensional coordinate, and a three-dimensional space coordinate system is established. The state of charge at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in the preset time period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence corresponding to a plurality of voltage step drop events are taken as sample point coordinate values in the three-dimensional space coordinate system. The Euclidean distance between any two sample points in the three-dimensional space coordinate system is calculated, sample point pairs with a Euclidean distance less than a preset distance threshold are identified, one sample point in the sample point pair is removed, and remaining sample points are obtained. The standard deviation of the remaining sample points in the first-dimensional coordinate direction, the standard deviation of the remaining sample points in the second-dimensional coordinate direction, and the standard deviation of the remaining sample points in the third-dimensional coordinate direction are calculated, the ratio of the maximum standard deviation to the minimum standard deviation among the three standard deviations is calculated, and the ratio is taken as a kernel width parameter of a radial basis function. The radial basis function surface fitting is performed on the remaining sample points by using the kernel width parameter, an explicit binary function of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence is obtained, the independent variables of the explicit binary function are the state of charge at the event occurrence time and the time integral value of the absolute value of the charge and discharge power in the preset time period before the event occurrence time, and the explicit binary function is taken as the three-dimensional mapping relationship.

[0008] As a preferred scheme of the data center direct current power supply charge storage collaborative control method, the three-dimensional mapping relationship among the state of charge at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in the preset time period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence includes that the state of charge at the event occurrence time is taken as the first-dimensional coordinate, the time integral value of the absolute value of the charge and discharge power in the preset time period before the event occurrence time is taken as the second-dimensional coordinate, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence is taken as the third-dimensional coordinate, and a three-dimensional space coordinate system is established. The three-dimensional space is divided into a high-density region and a low-density region according to the number density value, the high-density region is a region with a number density value greater than a preset density threshold, and the low-density region is a region with a number density value less than or equal to the preset density threshold. The high-density area is divided by a first grid step, and the low-density area is divided by a second grid step, and the first grid step is smaller than the second grid step; The first dimension coordinate value and the second dimension coordinate value of the center point of each spatial grid are substituted into the three-dimensional mapping relationship to obtain the time constant pre-stored value corresponding to the spatial grid; When the state of charge value of the energy storage system at the current time and the time integral value of the absolute value of the charge and discharge power in the preset time period before the current time of the energy storage system are received, the spatial grid to which the state of charge value and the time integral value belong is determined, the time constant pre-stored value corresponding to the spatial grid is obtained, and the time constant pre-stored value is output as the predicted value of the time constant.

[0009] As a preferred scheme of the data center direct current power supply and storage collaborative control method, the voltage dynamic response equation is expressed as, ; ; ; ; ; wherein, represents the equivalent capacitance value of the bus, represents the direct current bus voltage, represents the time from the response starting time of the energy storage system, represents the first derivative of the direct current bus voltage with respect to time, represents the energy storage system current, represents the load current, represents the peak current of the energy storage system, represents the dynamic response factor, represents the derating coefficient, represents the predicted value of the time constant, represents the rated voltage value of the direct current bus, represents the equivalent resistance of the load, represents a natural exponential function; The peak current of the energy storage system is determined according to the preset maximum load step amplitude and the rated voltage value of the direct current bus, the equivalent resistance of the load is determined according to the load power and the rated voltage value of the direct current bus, the equivalent capacitance value of the bus, the predicted value of the time constant, the peak current of the energy storage system, the equivalent resistance of the load, and the rated voltage value of the direct current bus are substituted into the voltage dynamic response equation, the rated voltage value of the direct current bus is used as the initial voltage value, and the fourth-order Runge-Kutta method is used to solve the voltage dynamic response equation to obtain the numerical solution sequence of the direct current bus voltage. The minimum value of the DC bus voltage is identified by traversing the numerical solution sequence of the DC bus voltage, and a difference between the rated voltage value of the DC bus and the minimum value of the DC bus voltage is calculated as a voltage drop depth estimation value.

[0010] As a preferred scheme of the data center direct current power supply and energy storage collaborative control method, the target value of the time constant is obtained by inversely solving the voltage dynamic response equation, the state of charge value required to reach the target value of the time constant is inversely searched according to the three-dimensional mapping relationship, and the difference between the state of charge value obtained by the inverse search and the current state of charge value is taken as a dynamic adjustment margin value. According to the three-dimensional mapping relationship, an isochrone where the time constant is equal to the target value of the time constant is identified in a two-dimensional space formed by the state of charge value and the time integral value of the absolute value of the charge and discharge power in a preset time period before the current time of the energy storage system. A point on the isochrone corresponding to the minimum Euclidean distance from a current state point formed by the state of charge value at the current time of the energy storage system and the time integral value of the absolute value of the charge and discharge power in a preset time period before the current time of the energy storage system is calculated as a target state point. The state of charge value coordinate of the target state point is extracted as a target state of charge value, and the difference between the target state of charge value and the state of charge value at the current time of the energy storage system is calculated as a dynamic adjustment margin value.

[0011] As a preferred scheme of the data center direct current power supply and energy storage collaborative control method, when the state of charge value at the current time of the energy storage system is less than the dynamic adjustment margin value, the distributed power supply and load adjustment are called to move the state of charge value to a region that meets the requirement of the dynamic adjustment margin value, and the state of charge value adjustment amount required to move the state of charge value at the current time of the energy storage system to a position equal to the dynamic adjustment margin value from the state of charge boundary is calculated. The charge and discharge power adjustment amount of the energy storage system is calculated according to the state of charge value adjustment amount and the capacity of the energy storage system. The sign of the charge and discharge power adjustment amount of the energy storage system is judged, and when the charge and discharge power adjustment amount of the energy storage system is positive, it is determined that the power flowing into the energy storage system needs to be increased, and when the charge and discharge power adjustment amount of the energy storage system is negative, it is determined that the power flowing into the energy storage system needs to be reduced. When it is needed to increase the power flowing into the energy storage system, the distributed power supply is called to increase the output power, and the load adjustment is called to reduce the load power of the data center; When it is needed to reduce the power flowing into the energy storage system, the distributed power supply is called to reduce the output power, and the load adjustment is called to increase the load power of the data center.

[0012] As a preferred scheme of the data center direct-current power supply load storage collaborative control method, in the execution of the adjustment process, the actual value of the time constant is monitored, the deviation of the actual value of the time constant from the predicted value of the time constant is calculated, and the recursive least square method is used to update the three-dimensional mapping relationship according to the deviation, including detecting the charging and discharging power response process of the energy storage system triggered by calling the distributed power supply and the load adjustment, and extracting the actual measured value of the time constant reflecting the response speed of the energy storage system from the charging and discharging power response process as the actual value of the time constant; The state of charge value corresponding to the response process of the energy storage system at the moment and the time integral value of the absolute value of the charging and discharging power in the preset time period before the moment are obtained, and the state of charge value and the time integral value are input into the three-dimensional mapping relationship to obtain the predicted value of the time constant; The difference between the actual value of the time constant and the predicted value of the time constant is calculated as a prediction deviation; The cumulative number of triggering the charging and discharging power response process of the energy storage system in a preset statistical time window is counted, the running strength of the energy storage system is determined according to the cumulative number, and the forgetting factor of the recursive least square method is adjusted according to the running strength, that is, the higher the running strength, the smaller the forgetting factor; The state of charge value and the time integral value are input into the radial basis function in the three-dimensional mapping relationship to obtain a radial basis function value vector, and the recursive least square method is used to update the weight coefficient vector of the radial basis function in the three-dimensional mapping relationship according to the radial basis function value vector, the prediction deviation and the forgetting factor, so that the three-dimensional mapping relationship adapts to the change of the characteristics of the energy storage system.

[0013] The application provides a computer device, including a memory and a processor, the memory stores a computer program, and the processor realizes the steps of the data center direct-current power supply load storage collaborative control method when executing the computer program.

[0014] The application provides a computer readable storage medium, which stores a computer program, and the computer program realizes the steps of the data center direct-current power supply load storage collaborative control method when executed by a processor.

[0015] The beneficial effects of the present application: the present application establishes a three-dimensional mapping relationship between the response time constant of the energy storage system and the state of charge, and the historical charge and discharge integral, realizes the dynamic quantitative evaluation of the instantaneous response capability of the energy storage system. Based on the mapping relationship, an evaluation function is constructed to predict the response speed of the energy storage system under the current state, and the voltage drop depth when the load suddenly increases is estimated through the voltage dynamic response equation, which overcomes the defect that the existing technology cannot accurately reflect the dynamic response characteristics of the system only according to the static state of charge threshold.

[0016] The concept of dynamic adjustment margin is introduced, and the operation boundary of the energy storage system is changed from the fixed state of charge threshold to the dynamic constraint based on the actual response capability, solving the problem of lagging or excessive conservative adjustment of the traditional method. When it is detected that the response capability of the energy storage system is insufficient to resist the preset load step, the distributed power or load is called in advance for coordinated adjustment, so that the state of charge of the energy storage system moves to the safe area that meets the dynamic response requirement, reducing the risk of bus voltage collapse.

[0017] The recursive least squares method is used to continuously correct the three-dimensional mapping relationship according to the actual monitored response time constant, so that the evaluation model tracks the characteristic drift of the energy storage system caused by aging and temperature change, and maintains the prediction accuracy. The present application improves the ability of the data center direct current power supply system to resist load impact and reduces the probability of voltage drop accidents. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 A general flowchart of a data center direct current power supply source-load storage collaborative control method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail in conjunction with the drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0021] Embodiment 1, refer to Figure 1 For an embodiment of the present application, the embodiment provides a data center direct current power supply source-load storage collaborative control method, which comprises: Step 1: Collect the DC bus voltage sequence, the energy storage system charge and discharge power sequence, and the state of charge sequence, identify the voltage step drop event in the voltage sequence, extract the state of charge value at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in a preset period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence for each voltage step drop event, and establish a three-dimensional mapping relationship among the state of charge value at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in a preset period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence; Step 2: Construct an energy storage system instantaneous response capability evaluation function according to the three-dimensional mapping relationship, input the state of charge value of the energy storage system at the current time and the time integral value of the absolute value of the charge and discharge power in a preset period before the current time of the energy storage system into the energy storage system instantaneous response capability evaluation function, and obtain the predicted value of the time constant; Step 3: Substitute the predicted value of the time constant, the bus equivalent capacitance value, and the preset maximum load step amplitude into the voltage dynamic response equation to solve and obtain the voltage drop depth estimate value; Step 4: When the voltage drop depth estimate value exceeds the voltage drop depth safety threshold, the target value of the time constant is obtained by reverse solving the voltage dynamic response equation, and the state of charge value required to reach the target value of the time constant is inversely searched according to the three-dimensional mapping relationship. The difference between the inversely searched state of charge value and the current state of charge value is taken as the dynamic adjustment margin value; Step 5: When the state of charge value of the energy storage system at the current time is less than the dynamic adjustment margin value from the state of charge boundary, the distributed power and load are adjusted to move the state of charge value to a region that meets the requirement of the dynamic adjustment margin value; Step 6: During the adjustment process, the actual value of the time constant is monitored, the deviation between the actual value of the time constant and the predicted value of the time constant is calculated, and the three-dimensional mapping relationship is updated according to the deviation using the recursive least squares method.

[0022] In step 1, identifying the voltage step drop event in the voltage sequence includes performing first-order time difference operation on the DC bus voltage sequence to obtain a voltage change rate sequence; Identifying the time when the absolute value of the voltage change rate in the voltage change rate sequence exceeds the preset change rate threshold, judging that the voltage change direction at the time is negative, and taking the time as a candidate voltage drop time; Extracting the power reference value of the energy storage system charge and discharge power sequence before the candidate voltage drop time, extracting the power peak value in a preset verification time window after the candidate voltage drop time, judging whether the power peak value appears a step rise relative to the power reference value, and if the step rise appears, confirming the candidate voltage drop time as the voltage step drop event time; According to the time stamp of the voltage step drop event time, the state of charge value corresponding to the time stamp is searched from the state of charge sequence as the state of charge value at the time when the voltage step drop event occurs.

[0023] It should be noted that the preset change rate threshold is determined according to the voltage fluctuation characteristics of the data center DC bus in normal operation. Specifically, the DC bus voltage sequence in the normal operation state of the data center is collected, a first-order time difference operation is performed on the voltage sequence to obtain a normal operation voltage change rate sequence, the standard deviation of the normal operation voltage change rate sequence is calculated, and the preset change rate threshold is set to 3 to 10 times of the standard deviation. In this embodiment, 5 times can be used according to experience. The threshold setting method is based on the 3σ principle of statistics, and can effectively distinguish between normal voltage fluctuation and abnormal voltage drop caused by virtual machine migration.

[0024] For example, in a certain data center test, the standard deviation of the normal operation voltage change rate is calculated to be 0.8V / ms, and the preset change rate threshold is set to 4.0V / ms (5 times the standard deviation). The preset change rate threshold can effectively detect the virtual machine migration event.

[0025] The judgment whether the power peak value appears a step rise with respect to the power reference value includes: calculating a power change amplitude, the power change amplitude being equal to a difference between the power peak value and the power reference value; calculating a power change rate, the power change rate being equal to the power change amplitude divided by a time interval from the candidate voltage drop time to the time when the power peak value appears; judging whether the power change amplitude exceeds a first threshold value and the power change rate exceeds a second threshold value, and if both conditions are met, it is determined that a step rise occurs.

[0026] The first threshold value is in the range of 10% to 30% of the rated power of the energy storage system, and in this embodiment, 20% can be used according to experience; the second threshold value is in the range of 50% to 200% per second of the rated power of the energy storage system, and in this embodiment, 100% per second can be used according to experience. The double condition judgment can effectively distinguish between step response and gradual response of slow load growth.

[0027] For example, for an energy storage system with a rated power of 500kW, the first threshold value is set to 100kW and the second threshold value is set to 500kW / s. In a certain event, the power rises from 50kW to 180kW, and the time interval is 150ms. The calculated power change amplitude is 130kW, which is greater than 100kW, and the power change rate is 867kW / s, which is greater than 500kW / s. It is determined that a step rise occurs.

[0028] Further, in step 1, the three-dimensional mapping relationship among the state of charge value at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in the preset period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence is established, including establishing a three-dimensional space coordinate system with the state of charge value at the event occurrence time as the first-dimensional coordinate, the time integral value of the absolute value of the charge and discharge power in the preset period before the event occurrence time as the second-dimensional coordinate, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence as the third-dimensional coordinate. The state of charge value at the event occurrence time, the time integral value of the absolute value of the charge and discharge power in the preset period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence corresponding to a plurality of voltage step drop events are taken as sample point coordinate values in the three-dimensional space coordinate system. The Euclidean distance between any two sample points in the three-dimensional space coordinate system is calculated, sample point pairs with a Euclidean distance less than a preset distance threshold are identified, one sample point in the sample point pair is removed, and the remaining sample points are obtained. The standard deviation of the remaining sample points in the first-dimensional coordinate direction, the standard deviation in the second-dimensional coordinate direction, and the standard deviation in the third-dimensional coordinate direction are calculated, and the ratio of the maximum standard deviation to the minimum standard deviation among the three standard deviations is taken as the kernel width parameter of the radial basis function. The radial basis function surface fitting is performed on the remaining sample points using the kernel width parameter to obtain an explicit binary function of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence, and the independent variable of the explicit binary function is the state of charge value at the event occurrence time and the time integral value of the absolute value of the charge and discharge power in the preset period before the event occurrence time. It should be noted that the preset period before the event occurrence time is based on the historical event time, and the preset period before the current time of the energy storage system is based on the current time, and the time length is the same, but the time reference point is different.

[0029] It should be noted that the value of the preset period before the event occurrence time is determined according to the thermal time constant of the battery monomer of the energy storage system, and reflects the influence of the temperature accumulation effect caused by the historical charge and discharge of the energy storage system on the instantaneous response capability. The value of the preset period is 0.5 to 2 times the thermal time constant of the battery monomer of the energy storage system, and in this embodiment, 1 times is used according to experience. The thermal time constant of the battery monomer of the energy storage system is obtained through thermal characteristic test, and the typical value is 300 seconds to 1800 seconds. For example, a certain data center energy storage system uses lithium iron phosphate battery, and the thermal time constant is 600 seconds, so the preset period is set to 600 seconds.

[0030] When calculating the time integral value of the absolute value of the charge and discharge power in the preset period before the event occurrence time, the trapezoidal integral method is used. ; wherein, represents the time integral value of the absolute value of the charge-discharge power in the preset period before the event occurrence time; represents the historical charge-discharge power sample point serial number; represents the sample point number in the preset period; represents the energy storage system charge-discharge power of the first sample point; represents the energy storage system charge-discharge power of the first sample point; represents the sampling time interval.

[0031] For example, the sampling interval of a certain data center energy storage system is 1 second, the preset period is 600 seconds, and the average absolute value of the charge-discharge power of the energy storage system in the 600 seconds before the event occurrence time is 80 kW. Then the time integral value of the absolute value of the charge-discharge power in the preset period before the event occurrence time is about .

[0032] The preset distance threshold is adaptively determined according to the spatial distribution of the sample points. The Euclidean distance between all pairs of sample points in a three-dimensional coordinate system is calculated to obtain a distance set, the distance set is sorted in ascending order, and the 5th percentile and the 95th percentile are calculated. The preset distance threshold is set as: ; wherein, is an adjustment coefficient, and the value range is 0.05 to 0.2. In the embodiment, 0.1 can be used according to experience; represents the 5th percentile; represents the 95th percentile.

[0033] The method is based on the physical nature of the response characteristics of the data center energy storage system: when the energy storage system responds multiple times under similar working conditions, its time constant should remain stable, so the sample points with close spatial distance actually reflect the same physical state. The degree of aggregation of the data center operating conditions is automatically identified by the percentile method, avoiding excessive retention of redundant data or incorrect deletion of effective data caused by manually setting a fixed threshold.

[0034] When calculating the Euclidean distance between any two sample points, since the state of charge at the event occurrence time, the time integral value of the absolute value of the charge-discharge power in the preset period before the event occurrence time, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence have different dimensions, the coordinates of each dimension need to be normalized first: ; wherein, represents the original coordinate value; denotes the normalized coordinate value; denotes the minimum value of the coordinate; denotes the maximum value of the coordinate.

[0035] The normalized Euclidean distance calculation formula is: ; wherein, denotes the Euclidean distance between the sample point and the sample point ; denotes the normalized state of charge value at the event occurrence time of the sample point ; denotes the time integral value of the absolute value of the charge and discharge power in a preset time period before the event occurrence time of the sample point ; denotes the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of the sample point ; denotes the normalized state of charge value at the event occurrence time of the sample point ; denotes the time integral value of the absolute value of the charge and discharge power in a preset time period before the event occurrence time of the sample point ; denotes the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of the sample point .

[0036] For example, a certain data center collects 50 voltage step-down event sample points, and calculates the distance set , , and sets , then the preset distance threshold is 1.17. After removing the redundant samples using the threshold, 42 sample points are retained.

[0037] The sample point pair with the Euclidean distance less than the preset distance threshold is identified, and one sample point in the sample point pair is removed. The absolute value of the numerical difference of the two sample points in the third coordinate direction is calculated, and the size relationship between the absolute value of the numerical difference and the preset difference threshold is compared. If , it is determined that the two sample points represent the same system state, the sample point with the smaller time constant of the discharge power rising from the initial value to the peak value after the event occurrence is retained, and the sample point with the larger time constant is removed; if , it is determined that the two sample points represent different system states, and neither of the two sample points is removed, wherein, denotes the absolute value of the difference of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of the two sample points. This indicates the preset difference threshold.

[0038] The physical basis for retaining sample points with smaller time constants is that, under the same state of charge and historical load conditions, a fast response speed (small time constant) in a data center energy storage system represents the optimal operating state, such as good consistency of individual battery cells, low internal resistance, and suitable temperature. Conversely, a slow response speed may be affected by measurement noise or transient disturbances. Retaining sample points of the optimal state can improve the accuracy of the three-dimensional mapping relationship in predicting the true capability of the energy storage system.

[0039] The preset difference threshold is determined based on the measurement accuracy of the time constant, and the value ranges from 5% to 15% of the median of the time constant. In this embodiment, 10% can be used based on experience.

[0040] For example, the median time constant of a data center energy storage system is 20ms, and a preset difference threshold of 2ms is set. Two sample points are detected where the state of charge (SBC) at the time of the event is 60%, and the time integral of the absolute value of the charging and discharging power within a preset time period before the event is 15kWh. However, the time constants for the discharge power to rise from its initial value to its peak value after the event are 18ms and 19.5ms respectively. The difference of 1.5ms is less than 2ms, so the sample point corresponding to 18ms is retained.

[0041] Calculate the standard deviation of the remaining sample points in the first dimension. Standard deviation in the second-dimensional coordinate direction Standard deviation in the third-dimensional coordinate direction : ; in, Indicates the first Standard deviation of the dimensional coordinate direction; Indicates the coordinate dimension index. , This indicates the traversal sequence number of the remaining sample points in the standard deviation calculation; Indicates the number of remaining sample points; Indicates the first The remaining sample points at the th th Normalized values ​​of 3D coordinates; This indicates that all remaining sample points are at the th... The average value of the normalized values ​​of the dimensional coordinates.

[0042] Calculate the ratio of the largest standard deviation to the smallest standard deviation among the three standard deviations: ; in, Indicates the ratio of standard deviations; This represents the function that takes the maximum value. This represents the function that takes the minimum value.

[0043] Boundary constraints are applied to the standard deviation ratio: when When, the kernel width parameter of the radial basis function ;when When, the kernel width parameter of the radial basis function ;when When, the kernel width parameter of the radial basis function ,in, This represents a preset lower limit value, ranging from 1.5 to 3.0. In this embodiment, based on experience, 2.0 can be used. This represents a preset upper limit value, ranging from 20 to 50. In this embodiment, based on experience, 30 can be used. The kernel width parameter represents the radial basis function.

[0044] This boundary treatment is based on the physical constraints of the dynamic response of the data center energy storage system: when the three-dimensional data is uniformly distributed (the ratio is close to 1), it indicates that the state of charge, historical load, and response speed are independent of each other. In this case, increasing the kernel width can improve the interpolation smoothness. When the data in a certain dimension is extremely dispersed (the ratio is too large), it indicates that there is strong nonlinearity or abnormal operating conditions in that dimension. Limiting the upper limit of the kernel width can prevent the fitting from diverging.

[0045] Constructing the radial basis function matrix , indicating that the dimension is The radial basis function matrix, The radial basis function matrix elements represent the number of remaining sample points: ; in, Represents the first element in the radial basis function matrix. The remaining sample points and the first Radial basis function values ​​between the remaining sample points; Indicates the index of the remaining sample point corresponding to the row of the matrix; This indicates the index of the remaining sample point corresponding to the column of the matrix; Indicates the first The normalized state of charge values ​​of the remaining sample points at the time of the event occurrence; Indicates the first The time integral of the absolute value of charging and discharging power within a preset time period before the event occurrence time, after normalization of the remaining sample points; Indicates the first The normalized state of charge values ​​of the remaining sample points at the time of the event occurrence; Indicates the first The time integral of the absolute value of charging and discharging power within a preset time period before the event occurrence time, after normalization of the remaining sample points; a kernel width parameter of a radial basis function; denotes a natural exponential function.

[0046] solving a weight coefficient vector: ; wherein, denotes a weight coefficient vector; denotes a transpose of a radial basis function matrix; denotes a radial basis function matrix; denotes a regularization coefficient, and the value range is to In the embodiment, the value of ; denotes a unit matrix with a dimension of ; denotes a time constant vector of the discharge power rising from an initial value to a peak value after the normalized event occurrence; denotes a remaining sample point number.

[0047] The regularization coefficient is determined according to the remaining sample point number: when the remaining sample point number is less than 20, the value of the regularization coefficient is , when the remaining sample point number is between 20 and 50, the value of the regularization coefficient is , and when the remaining sample point number is greater than 50, the value of the regularization coefficient is . The setting is based on the trade-off between the sparsity of the sample data in the data center and the matrix ill-conditioning: when the sample is small, the regularization is increased to prevent overfitting, and when the sample is large, the regularization is reduced to improve the fitting accuracy.

[0048] constructing an explicit binary function: ; ; ; wherein, denotes an explicit binary function; denotes a remaining sample point sequence number in the sum of the explicit binary function; denotes an input normalized state of charge value at the event occurrence time; denotes an input time integral value of the absolute value of the charge and discharge power in a preset time period before the normalized event occurrence time; denotes an element of the weight coefficient vector; denotes a remaining sample point number; denotes an element of the weight coefficient vector; denotes a normalized state of charge value at the event occurrence time of the th remaining sample point; denotes a normalized state of charge value at the event occurrence time of the a time integral value of the absolute value of the charge and discharge power in a preset time period before the normalized event occurrence time of the remaining sample point; represents a kernel width parameter of a radial basis function; represents a maximum value of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence in all the remaining sample points; represents a minimum value of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence in all the remaining sample points; represents a natural exponential function; RMSE represents a root mean square error; represents a verification set sample point sequence number; represents a verification set sample point number; represents a verification set sample point sequence number; represents a predicted value of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of the i-th verification set sample point; represents a predicted value of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of the i-th verification set sample point; represents a predicted value of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of the i-th verification set sample point; MAPE represents a mean absolute percentage error.

[0049] It is judged whether the RMSE is less than a preset RMSE threshold value and the MAPE is less than a preset MAPE threshold value, and if both conditions are met, the candidate explicit binary function is accepted as the three-dimensional mapping relationship. The preset RMSE threshold value is set to 10% to 20% of the median of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence, and in this embodiment, 15% is used according to experience. The preset MAPE threshold value is in the range of 8% to 15%, and in this embodiment, 12% is used according to experience.

[0050] The verification method is based on the diversity of operating conditions of the data center energy storage system: the leave-one-out method is used to ensure that the three-dimensional mapping relationship can still accurately predict the response capability under unseen operating conditions, and to avoid prediction failure in actual application due to overfitting of historical data.

[0051] For example, the median of the time constant of the discharge power rising from the initial value to the peak value after the event occurrence of a certain data center is 25 ms, the preset RMSE threshold value is set to 4 ms, and the preset MAPE threshold value is set to 12%. 7 are randomly selected from 30 remaining sample points as a verification set, and 23 are used as a training set. The verification set test obtains , , which meets the requirements and accepts the three-dimensional mapping relationship.

[0052] In step 2, constructing the energy storage system instantaneous response capability evaluation function according to the three-dimensional mapping relationship comprises: statistically distributing the state of charge values of event occurrence time and the time integral values of charge and discharge power absolute values in a preset time period before the event occurrence time corresponding to a plurality of voltage step drop events used when the three-dimensional mapping relationship is established in a three-dimensional space; and calculating the number density values of the voltage step drop events in each local region in the three-dimensional space. According to the number density values, the three-dimensional space is divided into a high-density region and a low-density region, the high-density region is a region with a number density value greater than a preset density threshold, and the low-density region is a region with a number density value less than or equal to the preset density threshold. The first grid step is used for spatial grid division for the high-density region, and the second grid step is used for spatial grid division for the low-density region, and the first grid step is less than the second grid step. For each spatial grid, the first-dimensional coordinate value and the second-dimensional coordinate value of the spatial grid center point are substituted into the three-dimensional mapping relationship to obtain a time constant preset value corresponding to the spatial grid. When receiving the state of charge value of the energy storage system at the current time and the time integral value of the charge and discharge power absolute value in a preset time period before the current time of the energy storage system, the spatial grid to which the state of charge value and the time integral value belong is determined, the time constant preset value corresponding to the spatial grid is obtained, and the time constant preset value is output as a predicted value of the time constant.

[0053] The number density values of the voltage step drop events in each local region in the three-dimensional space are calculated by using a kernel density estimation method. The three-dimensional space is preliminarily divided into regular grids, and the grid size is 5% to 10% of the range of each coordinate, and in this embodiment, 8% is used according to experience. The number of sample points in each grid is calculated, and the number density value is obtained by dividing the grid volume, which is represented as: , wherein denotes the number density value of the grid , denotes the number of sample points in the grid , and denotes the volume of the grid . This method reflects the running frequency of the data center energy storage system under different working conditions: the high-density region corresponds to common working conditions (such as medium state of charge and normal load change), and the low-density region corresponds to rare working conditions (such as extremely low state of charge and large load fluctuation). The number density values of all grids are counted, and the median of the number density values is calculated

[0054] , and the preset density threshold is set as . , wherein​​​ denotes a preset density threshold value; is a density coefficient, and the value range is 0.5 to 2.0, and in the embodiment, 1.0 can be used according to experience. The threshold value setting ensures that the high-density area contains about 50% of the sample points, and the low-density area contains about 50% of the sample points, so as to realize reasonable allocation of computing resources.

[0055] a first grid step is used for the high-density area spatial grid division is performed, and a second grid step is used for the low-density area spatial grid division is performed. The first grid step is determined according to the average nearest neighbor distance of the sample points in the high-density area: ; wherein, denotes the first grid step; is a grid coefficient, and the value range is 1.0 to 3.0, and in the embodiment, 2.0 can be used according to experience; denotes the average nearest neighbor distance of the sample points in the high-density area.

[0056] The second grid step is determined according to the spatial range of the low-density area: ; wherein, denotes the second grid step; is a step ratio coefficient, and the value range is 2 to 5, and in the embodiment, 3 can be used according to experience.

[0057] The hierarchical grid strategy is based on the real-time control requirements of the data center: common working conditions need to be finely predicted to ensure control accuracy, and rare working conditions allow rough prediction to save computing time. For example, the average nearest neighbor distance of a high-density area of a certain data center is 0.05 (normalized coordinates), and the density threshold value is set to 0.5, the grid coefficient is 2, and the step ratio coefficient is 3. , the first grid step is 0.1, and the second grid step is 0.3.

[0058] For each spatial grid, the spatial grid center point coordinates are calculated, and the time constant pre-stored value corresponding to the spatial grid is calculated by substituting the three-dimensional mapping relationship to obtain : ; wherein, denotes the time constant pre-stored value; denotes the state of charge value coordinate of the event occurrence time of the spatial grid center point; denotes the time integral value coordinate of the absolute value of the charge and discharge power in a preset period before the event occurrence time of the spatial grid center point; ​represents a three-dimensional mapping relationship.

[0059] The center point coordinates of all space grids and the corresponding time constant pre-stored values are stored as a lookup table. The lookup table is a two-dimensional array data structure, the first dimension index corresponds to the grid index in the direction of the state of charge value at the event occurrence time, the second dimension index corresponds to the grid index in the direction of the time integral value of the absolute value of the charge and discharge power in a preset period before the event occurrence time, and the array element stores the time constant pre-stored value corresponding to the center point of the space grid. The lookup table is pre-calculated in the initialization stage of the energy storage system instantaneous response capability evaluation function and stored in the memory, and at runtime, the time constant pre-stored value is obtained by directly accessing the corresponding array element through the grid index according to the input state of charge value and time integral value, without the need to repeatedly calculate the radial basis function. This pre-calculation strategy converts the online calculation of the radial basis function into a table lookup operation, reduces the execution time of the energy storage system instantaneous response capability evaluation function from milliseconds to microseconds, and meets the real-time control requirements of the data center.

[0060] For example, a certain data center high-density area is divided into a 20x15 space grid, and a low-density area is divided into a 10x8 space grid, and the lookup table contains a total of 380 pre-stored values. The lookup table data structure can be represented as wherein, and are the grid indexes in the direction of the state of charge value at the event occurrence time and the direction of the time integral value of the absolute value of the charge and discharge power in a preset period before the event occurrence time, respectively, and LUT represents the lookup table.

[0061] For example, the state of charge value of a certain data center energy storage system at the current time is 62%, and the time integral value of the absolute value of the charge and discharge power in a preset period before the current time of the energy storage system is 15 kWh, and the grid index is calculated as The time constant pre-stored value is directly obtained through , and the time consumption of a single table lookup operation is in the order of microseconds.

[0062] When receiving the state of charge value of the energy storage system at the current time and the time integral value of the absolute value of the charge and discharge power in a preset period before the current time of the energy storage system , the space grid index to which the input coordinates belong is determined according to the input coordinates: ; ; wherein, represents the grid index in the direction of the state of charge value at the event occurrence time; represents the grid index in the direction of the time integral value of the absolute value of the charge and discharge power in a preset period before the event occurrence time; a state of charge value of the energy storage system at a current time point; a time integral value of an absolute value of charge-discharge power in a preset time period before the event occurrence time point; a minimum value of the state of charge value at the event occurrence time point; a minimum value of the time integral value of the absolute value of the charge-discharge power in the preset time period before the event occurrence time point; a grid step in a direction of the state of charge value at the event occurrence time point; a grid step in a direction of the time integral value of the absolute value of the charge-discharge power in the preset time period before the event occurrence time point; a floor function.

[0063] obtaining a corresponding time constant preset value from the lookup table according to the grid index , and outputting the time constant preset value as a predicted value of the time constant. If the input coordinate is out of the three-dimensional space range, boundary extrapolation is adopted: projecting the input coordinate to the boundary, and using the time constant preset value of the boundary grid as the predicted value.

[0064] Specifically, the three-dimensional space range is determined according to the remaining sample points, the three-dimensional space range is determined according to the remaining sample points, the grid division covers the area from to and from to in a two-dimensional space composed of the first dimension coordinate and the second dimension coordinate, wherein is a maximum value of the state of charge value at the event occurrence time point in the remaining sample points, is a maximum value of the time integral value of the absolute value of the charge-discharge power in the preset time period before the event occurrence time point in the remaining sample points. The grid index range stored in the lookup table is: the grid index in the direction of the state of charge value at the event occurrence time point is from 0 to , and the grid index in the direction of the time integral value of the absolute value of the charge-discharge power in the preset time period before the event occurrence time point is from 0 to , wherein is a maximum value of the grid index in the direction of the state of charge value at the event occurrence time point, is a maximum value of the grid index in the direction of the time integral value of the absolute value of the charge-discharge power in the preset time period before the event occurrence time point. The operation of projecting the input coordinate to the boundary is: if determined according to the input coordinate is less than 0, then is corrected to 0, if is greater than , then is corrected to ; if is less than 0, then is corrected to 0, if is greater than then the is modified to The pre-stored value of the time constant corresponding to the modified grid index is obtained from the look-up table as the predicted value of the time constant.

[0065] In step 3, the voltage dynamic response equation is expressed as ; ; ; ; ; wherein, represents the equivalent bus capacitance value, represents the DC bus voltage, represents the time from the response time of the energy storage system, represents the first-order derivative of the DC bus voltage with respect to time, represents the energy storage system current, represents the load current, represents the peak current of the energy storage system, represents the dynamic response factor, represents the derating coefficient, represents the predicted value of the time constant, represents the rated DC bus voltage value, represents the equivalent load resistance, represents the natural exponential function; The derating coefficient describes the characteristic that the output current of the energy storage system decreases with the decrease of the DC bus voltage when the DC bus voltage is lower than the rated DC bus voltage value ; The peak current of the energy storage system is determined according to the preset maximum load step amplitude and the rated DC bus voltage value, the equivalent load resistance is determined according to the load power and the rated DC bus voltage value, and the equivalent bus capacitance value, the predicted value of the time constant, the peak current of the energy storage system, the equivalent load resistance, and the rated DC bus voltage value are substituted into the voltage dynamic response equation to obtain the numerical solution sequence of the DC bus voltage with the rated DC bus voltage value as the initial voltage value by using the fourth-order Runge-Kutta method. The minimum value of the DC bus voltage is identified by traversing the numerical solution sequence of the DC bus voltage, and the difference between the rated DC bus voltage value and the minimum value of the DC bus voltage is calculated as the voltage drop depth estimate.

[0066] Specifically, the preset maximum load step amplitude is determined according to data center virtual machine migration statistical data. The load power step amplitudes caused by all virtual machine migration events in the past 3 to 6 months are counted, the 95th percentile is calculated, and the preset maximum load step amplitude is set to: ;

[0067] wherein, represents the preset maximum load step amplitude; represents the 95th percentile of the load power step amplitude; is a load safety factor, and the value range is 1.2 to 1.5, and in the embodiment, 1.3 can be used according to experience.

[0068] The method is based on the statistical characteristics of the data center load step event: 95% of the virtual machine migration events have a power increment less than Reserving a 30% safety margin can cover extreme cases and consider future business growth. For example, a certain data center has a statistical result of , and sets .

[0069] In step 4, the target value of the time constant is obtained by inversely solving the voltage dynamic response equation, the state of charge value required to reach the target value of the time constant is inversely searched according to the three-dimensional mapping relationship, and the difference between the inversely searched state of charge value and the current state of charge value is taken as the dynamic adjustment margin value. The target value of the time constant is obtained by inversely solving the voltage dynamic response equation using the bisection method, the search interval of the time constant is initialized to the preset minimum time constant to the preset maximum time constant, and in each iteration, the midpoint value of the search interval is substituted into the voltage dynamic response equation to obtain the voltage drop depth calculation value. According to the size relationship between the voltage drop depth calculation value and the voltage drop depth safety threshold, the search interval is updated, and the iteration is performed until the absolute value of the difference between the voltage drop depth calculation value and the voltage drop depth safety threshold is less than the preset error threshold. According to the three-dimensional mapping relationship, the isochrones of the time constant equal to the target value of the time constant are identified in the two-dimensional space formed by the state of charge value and the time integral value of the charge and discharge power absolute value in the preset time period before the current time of the energy storage system. The point on the isochrone corresponding to the minimum Euclidean distance from the current state point formed by the state of charge value of the energy storage system at the current time and the time integral value of the charge and discharge power absolute value in the preset time period before the current time of the energy storage system is calculated as the target state point. The state of charge value coordinate of the target state point is extracted as the target state of charge value, and the difference between the target state of charge value and the state of charge value of the energy storage system at the current time is calculated as the dynamic adjustment margin value.

[0070] Specifically, the identifying the contour line of the target value of the time constant comprises: traversing each spatial grid by using the spatial grid established when the transient response capability evaluation function of the energy storage system is constructed in step 2; for each spatial grid, substituting the first-dimensional coordinate value and the second-dimensional coordinate value of the center point of the spatial grid into the explicit binary function to obtain a corresponding time constant calculation value; identifying a plurality of spatial grids with the minimum absolute difference value between the time constant calculation value and the target value of the time constant, and connecting the center points of the identified spatial grids according to the spatial positions to form the contour line.

[0071] Specifically, the preset minimum time constant and the preset maximum time constant define the search interval boundary of the time constant when the voltage dynamic response equation is inversely solved by the dichotomy method.

[0072] The preset minimum time constant is determined according to the physical limit of the response speed of the energy storage system converter: ; wherein, represents the preset minimum time constant; represents the output filter inductance value of the energy storage system converter; represents the equivalent resistance value of the active damping.

[0073] The time constant reflects the fastest rising speed of the current response of the energy storage system, which is constrained by the time constant of the hardware circuit of the converter. The typical value is 5 ms to 20 ms. For example, the output filter inductance of the converter of a certain data center energy storage system is 2 mH, and the equivalent resistance of the active damping is 0.2 Ω, and the preset minimum time constant is 10 ms.

[0074] The preset maximum time constant is determined according to the response speed of the energy storage system under the worst working condition: ; wherein, represents the preset maximum time constant; represents the nominal time constant of the energy storage system, which is the typical value of the time constant obtained by testing the energy storage system under the rated working condition; represents the time constant amplification coefficient caused by the low state of charge, and the value range is 0.2 to 0.5; represents the time constant amplification coefficient caused by the low temperature, and the value range is 0.3 to 0.8; represents the time constant amplification coefficient caused by the battery aging, and the value range is 0.2 to 0.6.

[0075] The time constant reflects the slowest response speed of the energy storage system under the superposition of multiple adverse factors such as low state of charge, low temperature environment, and battery aging. For example, the nominal time constant of an energy storage system in a certain data center is 25 ms, considering the comprehensive influence of state of charge, temperature, and aging, the preset maximum time constant is set to .

[0076] The preset error threshold represents the maximum deviation allowed between the voltage drop depth calculation value and the voltage drop depth safety threshold when the bisection iteration is terminated. The threshold is determined according to the DC bus voltage measurement accuracy: ; wherein, represents the preset error threshold; is an error coefficient, with a value range of 1 to 3, and in the embodiment, 2 can be used according to experience; represents the DC bus voltage measurement accuracy.

[0077] The DC bus voltage measurement accuracy is determined by the voltage sensor accuracy level, and the typical value is 0.1% to 0.5% of the rated voltage. For example, the rated voltage value of the DC bus in a certain data center is 500V, and the voltage sensor accuracy is 0.2%, so the DC bus voltage measurement accuracy is 1V, and the preset error threshold is set to 2V.

[0078] The threshold setting ensures that the bisection solution accuracy matches the measurement system accuracy: a too small preset error threshold leads to too many iterations but the accuracy improves without meaning, and a too large preset error threshold leads to insufficient solution accuracy affecting the control effect. Setting the preset error threshold to be 2 times the measurement accuracy can converge within 5 to 8 iterations, meeting the real-time control requirements.

[0079] The voltage drop depth safety threshold is determined according to the voltage tolerance of the data center IT equipment, and the IT equipment may restart or data loss when the voltage drop depth exceeds 15% of the rated voltage.

[0080] The voltage drop depth safety threshold is set as: ; wherein, represents the voltage drop depth safety threshold; is a voltage safety coefficient, with a value range of 0.08 to 0.12, and in the embodiment, 0.10 can be used according to experience; represents the DC bus rated voltage value.

[0081] The coefficient is set based on the data center power supply reliability requirements: controlling the voltage drop depth within 10% of the rated voltage can ensure that 99.99% of IT equipment is not affected, and reserving 5% margin should cope with device differences and measurement errors. For example, the DC bus voltage value of a certain data center is 500V, and the voltage drop depth safety threshold is set to 50V, that is, the DC bus voltage after voltage drop should not be lower than 450V.

[0082] Further, the voltage drop depth safety threshold can be set according to the data center business level: Tier IV level data center adopts , Tier III level data center adopts , Tier II level data center adopts .

[0083] The bisection iterative process is: initialize the search interval as , take the midpoint of the interval each time, substitute it into the voltage dynamic response equation to obtain the voltage drop depth calculation value , if , update the search interval as , if , update the search interval as , repeat the iteration until , wherein, represents the time constant value corresponding to the midpoint of the search interval; represents the voltage drop depth calculation value; represents the preset minimum time constant; represents the preset maximum time constant; represents the voltage drop depth safety threshold; represents the preset error threshold.

[0084] The final is the target value of the time constant.

[0085] In step 5, when the state of charge value of the energy storage system at the current time and the distance from the state of charge boundary are less than the dynamic adjustment margin value, the distributed power supply and load adjustment are called to move the state of charge value to the region that meets the requirement of the dynamic adjustment margin value, including calculating the state of charge value adjustment amount required for the state of charge value of the energy storage system at the current time to move to the position with a distance from the state of charge boundary equal to the dynamic adjustment margin value; According to the state of charge value adjustment amount and the capacity of the energy storage system, the charge and discharge power adjustment amount of the energy storage system is calculated; The sign of the charge and discharge power adjustment amount of the energy storage system is judged, when the charge and discharge power adjustment amount of the energy storage system is positive, it is determined that the power flowing into the energy storage system needs to be increased, when the charge and discharge power adjustment amount of the energy storage system is negative, it is determined that the power flowing into the energy storage system needs to be reduced; When the power flowing into the energy storage system needs to be increased, the distributed power is called to increase the output power, and the load adjustment is called to reduce the data center load power; When the power flowing into the energy storage system needs to be reduced, the distributed power is called to reduce the output power, and the load adjustment is called to increase the data center load power.

[0086] The state of charge boundary of the energy storage system includes an upper state of charge boundary and a lower state of charge boundary . The upper state of charge boundary is determined according to the charge cutoff threshold of the battery management system of the energy storage system, and a typical value is 90% to 95%. In this embodiment, 90% is used according to experience. The lower state of charge boundary is determined according to the discharge cutoff threshold of the battery management system of the energy storage system, and a typical value is 10% to 20%. In this embodiment, 20% is used according to experience.

[0087] When the distance between the current state of charge value of the energy storage system and the state of charge boundary is less than the dynamic adjustment margin value, the judgment logic is: If the current state of charge value of the energy storage system is , the distance from the upper state of charge boundary is calculated , and it is determined whether is true; if the current state of charge value of the energy storage system is , the distance from the lower state of charge boundary is calculated , and it is determined whether is true, wherein represents the current state of charge value of the energy storage system; represents the upper state of charge boundary; represents the lower state of charge boundary; represents the distance between the current state of charge value and the state of charge boundary; represents the dynamic adjustment margin value.

[0088] The judgment logic clearly defines the demand for two-way adjustment space of the data center energy storage system: when the state of charge approaches the upper boundary, a discharge margin needs to be reserved to cope with a sudden increase in load, and when the state of charge approaches the lower boundary, a charge margin needs to be reserved to cope with a sudden decrease in load. Take 50% as the dividing point to select the nearest boundary for judgment.

[0089] Specifically, the calling distributed power to increase output power and the calling load adjustment to reduce data center load power include: judging the size relationship between the absolute value of the energy storage system charge-discharge power adjustment amount and the distributed power that can be increased; when the absolute value of the energy storage system charge-discharge power adjustment amount is less than or equal to the distributed power that can be increased, calling the distributed power to increase the output power equal to the energy storage system charge-discharge power adjustment amount; when the absolute value of the energy storage system charge-discharge power adjustment amount is greater than the distributed power that can be increased, calling the distributed power to increase the output power equal to the distributed power that can be increased, calculating the remaining power demand equal to the absolute value of the energy storage system charge-discharge power adjustment amount minus the distributed power that can be increased, and calling the load adjustment to reduce the data center load power equal to the remaining power demand. When it is necessary to reduce the power flowing into the energy storage system, the distributed power is called to reduce the output power, and the load adjustment is called to increase the data center load power, and the same distribution principle is adopted.

[0090] In step 6, during the adjustment process, the actual value of the time constant is monitored, the deviation between the actual value of the time constant and the predicted value of the time constant is calculated, and the recursive least squares method is used to update the three-dimensional mapping relationship according to the deviation, including detecting the energy storage system charge-discharge power response process triggered by calling the distributed power and the load adjustment, extracting the actual measured value of the time constant reflecting the response speed of the energy storage system from the charge-discharge power response process as the actual value of the time constant; The state of charge value corresponding to the response process of the energy storage system at the moment and the time integral value of the absolute value of the charge-discharge power in the preset time period before the moment are obtained, and the state of charge value and the time integral value are input into the three-dimensional mapping relationship to calculate the predicted value of the time constant; The difference between the actual value of the time constant and the predicted value of the time constant is calculated as a prediction deviation; The cumulative number of times of triggering the energy storage system charge-discharge power response process in a preset statistical time window is counted, the running intensity of the energy storage system is determined according to the cumulative number, the forgetting factor of the recursive least squares method is adjusted according to the running intensity, and the higher the running intensity, the smaller the forgetting factor; The state of charge value and the time integral value are input into the radial basis function in the three-dimensional mapping relationship to calculate the radial basis function value vector, and the weight coefficient vector of the radial basis function in the three-dimensional mapping relationship is updated according to the radial basis function value vector, the prediction deviation and the forgetting factor using the recursive least squares method, so that the three-dimensional mapping relationship adapts to the change of the energy storage system characteristics.

[0091] Specifically, the time constant measured value reflecting the response speed of the energy storage system is extracted from the charge-discharge power response process, and an exponential fitting method is adopted. The time series of the charge-discharge power of the energy storage system from the start of the response to the steady-state value is collected, and the power value at each sampling time is recorded. The measured power response curve is fitted as a first-order exponential model: ; wherein, represents the charge-discharge power of the energy storage system at time t after the start of the response; represents the time from the start of the response of the energy storage system; represents the steady-state power value; represents the time constant measured value; represents the natural exponential function.

[0092] The nonlinear least squares method is used to solve the time constant measured value and the steady-state power value, and the objective function is minimized: ; wherein, represents the minimization operator; represents the time constant measured value; represents the steady-state power value; represents the sequence number of the power sampling point in the response process; represents the total number of sampling points; represents the charge-discharge power measurement value of the energy storage system at the i th sampling point; represents the relative time from the start of the response corresponding to the i th sampling point; represents the natural exponential function. For example, the charge-discharge power of the energy storage system is triggered to rise from 50 kW to 180 kW in a certain adjustment, the sampling interval is 50 ms, 100 data points are collected, and the time constant measured value of 22.8 ms is obtained by fitting, which is taken as the actual value of the time constant.

[0093] The preset statistical time window is determined according to the load change period of the data center, and is set to 0.5 to 2 times of the main period of load change. In this embodiment, 1 times can be used according to experience. For a data center with a daily period as the main period, the preset statistical time window is set to 12 to 48 hours, and in this embodiment, 24 hours can be used according to experience. Let the number of hours of the preset statistical time window be (unit: hours).

[0094]

[0095] ​​​The forgetting factor of the recursive least squares method is adjusted according to the cumulative number of triggering the energy storage system charging and discharging power response process within the preset statistical time window. The more the cumulative number is, the faster the energy storage system characteristics change, and the smaller the forgetting factor is to make the model adapt quickly; the less the cumulative number is, the slower the energy storage system characteristics change, and the larger the forgetting factor is to keep the model stable.

[0096] The forgetting factor of the recursive least squares method is determined according to the cumulative number: when the cumulative number of triggering the energy storage system charging and discharging power response process within the preset statistical time window is less than , the forgetting factor of the recursive least squares method is 0.98 to 0.99, and in the embodiment, 0.99 can be used according to experience; When the cumulative number of triggering the energy storage system charging and discharging power response process within the preset statistical time window is greater than or equal to and less than , the forgetting factor of the recursive least squares method is 0.95 to 0.97, and in the embodiment, 0.96 can be used according to experience; When the cumulative number of triggering the energy storage system charging and discharging power response process within the preset statistical time window is greater than or equal to , the forgetting factor of the recursive least squares method is 0.90 to 0.93, and in the embodiment, 0.92 can be used according to experience. Wherein, represents the number of hours of the preset statistical time window.

[0097] For example, the preset statistical time window is 24 hours, that is, , when the cumulative number is less than 48, the forgetting factor of the recursive least squares method is 0.99, when the cumulative number is 48 to 96, the forgetting factor of the recursive least squares method is 0.96, and when the cumulative number is greater than or equal to 96, the forgetting factor of the recursive least squares method is 0.92.

[0098] The weight coefficient vector of the radial basis function in the three-dimensional mapping relationship is updated by using the recursive least squares method, and the recursive formula is: ; ; ; Wherein, represents the updated weight coefficient vector; represents the weight coefficient vector before updating; represents the gain vector; represents the updated covariance matrix; represents the covariance matrix before updating; represents the radial basis function value vector; represents the forgetting factor of the recursive least squares method; Indicates prediction bias; This represents the transpose of the radial basis function value vector.

[0099] The radial basis function value vector is calculated by inputting the current state of charge and time integral values ​​into the radial basis function in the three-dimensional mapping relationship. The first radial basis function value vector is... The elements are: ; in, The first value of the radial basis function value vector represents the... One element; This represents the normalized state of charge value at the corresponding moment in the energy storage system's response process; This represents the time integral value of the normalized absolute value of the charging and discharging power within a preset time period before the corresponding moment of the energy storage system's response process. Indicates the first The normalized state of charge values ​​of the remaining sample points at the time of the event occurrence; Indicates the first The time integral of the absolute value of charging and discharging power within a preset time period before the event occurrence time, after normalization of the remaining sample points; The kernel width parameter represents the radial basis functions; This represents the natural exponential function.

[0100] Prediction bias is calculated as the difference between the normalized measured value and the normalized predicted value of the time constant: ; in, This represents the measured value of the normalized time constant. This represents the predicted value of the time constant after normalization.

[0101] The initial covariance matrix is ​​set to ,in The initialization coefficients range from 100 to 1000. In this embodiment, based on experience, 500 can be used. The dimension is The identity matrix; Indicates the number of remaining sample points; Let represent the initial covariance matrix.

[0102] Example 2: This example also provides an electronic device applicable to a data center DC power supply load-storage coordinated control method, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the data center DC power supply load-storage coordinated control method proposed in the above examples.

[0103] The embodiment further provides a storage medium, which stores a computer program, and the computer program is executed by a processor to implement the data center direct-current power supply storage collaborative control method proposed in the above embodiment.

[0104] The storage medium proposed in the embodiment and the data center direct-current power supply storage collaborative control method proposed in the above embodiment belong to the same inventive concept, and the technical details not described in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0105] Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary universal hardware, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part that contributes 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 a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk or an optical disk, etc., including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit the present application, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A method for coordinated control of DC power supply and energy storage in a data center, characterized in that: include, The system collects DC bus voltage sequences, energy storage system charge / discharge power sequences, and state of charge sequences. It identifies voltage step drop events in the voltage sequences and extracts the state of charge value at the time of the event, the time integral of the absolute value of charge / discharge power within a preset period before the event, and the time constant for the discharge power to rise from the initial value to the peak value after the event for each voltage step drop event. It then establishes a three-dimensional mapping relationship between the state of charge value at the time of the event, the time integral of the absolute value of charge / discharge power within a preset period before the event, and the time constant for the discharge power to rise from the initial value to the peak value after the event. Based on the three-dimensional mapping relationship, an instantaneous response capability evaluation function for the energy storage system is constructed. The current state of charge value of the energy storage system and the time integral value of the absolute value of the charging and discharging power within a preset period before the current time are input into the instantaneous response capability evaluation function to obtain the predicted value of the time constant. Substitute the predicted value of the time constant, the equivalent capacitance of the bus, and the preset maximum load step amplitude into the voltage dynamic response equation to obtain the estimated value of the voltage drop depth. When the estimated voltage drop depth exceeds the safe threshold for voltage drop depth, the voltage dynamic response equation is solved in reverse to obtain the target value of the time constant. Based on the three-dimensional mapping relationship, the state of charge value required to reach the target value of the time constant is looked up in reverse. The difference between the state of charge value obtained in reverse and the current state of charge value is used as the dynamic adjustment margin value. When the distance between the current state of charge (SOC) value of the energy storage system and the SOC boundary is less than the dynamic adjustment margin value, the distributed power source and load regulation are invoked to move the SOC value to the region that meets the requirements of the dynamic adjustment margin value. During the adjustment process, the actual value of the time constant is monitored, the deviation between the actual value and the predicted value of the time constant is calculated, and the three-dimensional mapping relationship is updated based on the deviation using the recursive least squares method.

2. The data center DC power supply load-storage coordinated control method as described in claim 1, characterized in that: The identification of voltage step drop events in the voltage sequence includes performing a first-order time difference operation on the DC bus voltage sequence to obtain a voltage change rate sequence. Identify the moment when the absolute value of the voltage change rate in the voltage change rate sequence exceeds a preset change rate threshold, determine that the voltage change direction at that moment is negative, and take that moment as a candidate voltage drop moment; Extract the power reference value of the energy storage system's charging and discharging power sequence before the candidate voltage drop moment, extract the power peak value of the energy storage system's charging and discharging power sequence within a preset verification time window after the candidate voltage drop moment, and determine whether the power peak value shows a step increase relative to the power reference value. If a step increase occurs, the candidate voltage drop moment is confirmed as the voltage step drop event moment. Based on the timestamp of the voltage step drop event, the state of charge value corresponding to the timestamp is found in the state of charge sequence as the state of charge value at the time of the voltage step drop event.

3. The data center DC power supply load-storage coordinated control method as described in claim 2, characterized in that: The establishment of a three-dimensional mapping relationship between the state of charge value at the moment of the event, the time integral value of the absolute value of the charging and discharging power within a preset period before the moment of the event, and the time constant of the discharge power rising from the initial value to the peak value after the event includes establishing a three-dimensional spatial coordinate system with the state of charge value at the moment of the event as the first dimension coordinate, the time integral value of the absolute value of the charging and discharging power within a preset period before the moment of the event as the second dimension coordinate, and the time constant of the discharge power rising from the initial value to the peak value after the event as the third dimension coordinate. The state of charge value at the time of occurrence of multiple voltage step drop events, the time integral value of the absolute value of charging and discharging power within a preset period before the event occurrence, and the time constant of the discharge power rising from the initial value to the peak value after the event occurrence are used as the coordinate values ​​of sample points in the three-dimensional spatial coordinate system. Calculate the Euclidean distance between any two sample points in the three-dimensional spatial coordinate system, identify sample point pairs whose Euclidean distance is less than a preset distance threshold, remove one sample point from the sample point pair, and obtain the remaining sample points; Calculate the standard deviation of the remaining sample points in the first-dimensional coordinate direction, the standard deviation in the second-dimensional coordinate direction, and the standard deviation in the third-dimensional coordinate direction. Calculate the ratio of the maximum standard deviation to the minimum standard deviation among the three standard deviations, and use the ratio as the kernel width parameter of the radial basis function. The remaining sample points are fitted with radial basis function surfaces using kernel width parameters to obtain an explicit binary function of the time constant for the discharge power to rise from the initial value to the peak value after the event. The independent variables of the explicit binary function are the state of charge value at the time of the event and the time integral value of the absolute value of the charge and discharge power within a preset period before the event. The explicit binary function is used as a three-dimensional mapping relationship.

4. The data center DC power supply load-storage coordinated control method as described in claim 3, characterized in that: The instantaneous response capability evaluation function of the energy storage system constructed based on the three-dimensional mapping relationship includes: statistically analyzing the distribution of the state of charge value at the time of occurrence of multiple voltage step drop events used in the establishment of the three-dimensional mapping relationship and the time integral value of the absolute value of charging and discharging power within a preset period before the time of occurrence of the event in the three-dimensional space; and calculating the number density value of voltage step drop events in each local region in the three-dimensional space. Based on the number density value, the three-dimensional space is divided into high-density regions and low-density regions. High-density regions are those with a number density value greater than a preset density threshold, while low-density regions are those with a number density value less than or equal to the preset density threshold. For high-density areas, a first grid step size is used for spatial grid division, and for low-density areas, a second grid step size is used for spatial grid division. The first grid step size is smaller than the second grid step size. For each spatial grid, the first and second dimension coordinates of the center point of the spatial grid are substituted into the three-dimensional mapping relationship to calculate the pre-stored value of the time constant corresponding to the spatial grid. When receiving the current state of charge (SOC) value of the energy storage system and the time integral value of the absolute value of the charging and discharging power within a preset time period prior to the current time, the spatial grid to which the SOC value and the time integral value belong are determined, the time constant pre-stored value corresponding to the spatial grid is obtained, and the time constant pre-stored value is output as the predicted value of the time constant.

5. The data center DC power supply load-storage coordinated control method as described in claim 4, characterized in that: The voltage dynamic response equation is expressed as follows: ； ； ； ； ； in, This represents the equivalent capacitance of the busbar. Indicates the DC bus voltage. This indicates the time from when the energy storage system begins to respond. This represents the first derivative of the DC bus voltage with respect to time. Indicates the current of the energy storage system. Indicates the load current. Indicates the peak current of the energy storage system. Represents the dynamic response factor. This represents the reduction coefficient. The predicted value representing the time constant. This indicates the rated voltage of the DC bus. Indicates the equivalent resistance of the load. Represents the natural exponential function; The peak current of the energy storage system is determined based on the preset maximum load step amplitude and the rated DC bus voltage. The equivalent resistance of the load is determined based on the load power and the rated DC bus voltage. The equivalent capacitance of the bus, the predicted value of the time constant, the peak current of the energy storage system, the equivalent resistance of the load, and the rated DC bus voltage are substituted into the voltage dynamic response equation. The rated DC bus voltage is used as the initial voltage value. The fourth-order Runge-Kutta method is used to solve the voltage dynamic response equation to obtain the numerical solution sequence of the DC bus voltage. The minimum DC bus voltage is identified by traversing the DC bus voltage numerical solution sequence. The difference between the rated DC bus voltage and the minimum DC bus voltage is calculated as the estimated voltage drop depth.

6. The data center DC power supply load-storage coordinated control method as described in claim 5, characterized in that: The reverse solution of the voltage dynamic response equation yields the target value of the time constant. Based on the three-dimensional mapping relationship, the state of charge value required to reach the target value of the time constant is retrieved. The difference between the retrieved state of charge value and the current state of charge value is used as the dynamic adjustment margin value. The reverse solution of the voltage dynamic response equation is performed using the bisection method to obtain the target value of the time constant. The time constant search interval is initialized by the bisection method to a preset minimum time constant to a preset maximum time constant. In each iteration, the midpoint value of the search interval is substituted into the voltage dynamic response equation to obtain the voltage drop depth calculation value. The search interval is updated according to the relationship between the voltage drop depth calculation value and the voltage drop depth safety threshold. The iteration continues until the absolute value of the difference between the voltage drop depth calculation value and the voltage drop depth safety threshold is less than a preset error threshold. Based on the three-dimensional mapping relationship, the contour lines of the target value whose time constant is equal to the time constant are identified in the two-dimensional space formed by the time integral of the absolute value of charging and discharging power during a preset period before the current time of the energy storage system. The target state point is the point on the contour line corresponding to the minimum Euclidean distance from the current state point, which is calculated by the time integral of the current state value of the energy storage system and the absolute value of the charging and discharging power during a preset period before the current state. Extract the coordinates of the state of charge (SOC) value of the target state point as the target SOC value, and calculate the difference between the target SOC value and the current SOC value of the energy storage system as the dynamic adjustment margin value.

7. The data center DC power supply load-storage coordinated control method as described in claim 6, characterized in that: The step of calling distributed power sources and load regulation to move the state of charge (SCC) value to a region that meets the requirements of the dynamic regulation margin when the distance between the current SCC value and the SCC boundary of the energy storage system is less than the dynamic regulation margin value includes calculating the amount of SCC value regulation required to move the current SCC value of the energy storage system to a position that is equal to the dynamic regulation margin value from the SCC boundary. Calculate the charging and discharging power adjustment of the energy storage system based on the state of charge adjustment amount and the energy storage system capacity. Determine the sign of the energy storage system's charge and discharge power adjustment amount. When the energy storage system's charge and discharge power adjustment amount is positive, it is determined that the power flowing into the energy storage system needs to be increased. When the energy storage system's charge and discharge power adjustment amount is negative, it is determined that the power flowing into the energy storage system needs to be reduced. When it is necessary to increase the power flowing into the energy storage system, the distributed power source is called to increase the output power, and the load regulation is called to reduce the data center load power. When it is necessary to reduce the power flowing into the energy storage system, distributed power sources are called to reduce output power, and load regulation is called to increase the load power of the data center.

8. The data center DC power supply load-storage coordinated control method as described in claim 7, characterized in that: During the adjustment process, the actual value of the time constant is monitored, the deviation between the actual value of the time constant and the predicted value of the time constant is calculated, and the three-dimensional mapping relationship is updated based on the deviation using the recursive least squares method. This includes detecting the charging and discharging power response process of the energy storage system triggered by the call to distributed power sources and load regulation, and extracting the measured value of the time constant reflecting the response speed of the energy storage system from the charging and discharging power response process as the actual value of the time constant. The state of charge (SOC) value at the corresponding moment of the energy storage system response process and the time integral value of the absolute value of the charging and discharging power within a preset period before the corresponding moment are obtained. The SOC value and the time integral value are input into a three-dimensional mapping relationship to calculate the predicted value of the time constant. The difference between the actual value and the predicted value of the time constant is calculated as the prediction bias; The cumulative number of times the energy storage system's charging and discharging power response process is triggered within a preset statistical time window is counted. The operating intensity of the energy storage system is determined based on the cumulative number of times. The forgetting factor of the recursive least squares method is adjusted based on the operating intensity. The higher the operating intensity, the smaller the forgetting factor. The radial basis function value vector is calculated by inputting the state of charge value and time integral value into the radial basis function in the three-dimensional mapping relationship. The weight coefficient vector of the radial basis function in the three-dimensional mapping relationship is updated by the recursive least squares method based on the radial basis function value vector, prediction bias and forgetting factor, so that the three-dimensional mapping relationship can adapt to the changes in the characteristics of the energy storage system.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the data center DC power supply load-storage coordinated control method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the data center DC power supply load-storage coordinated control method according to any one of claims 1 to 8.

Images