CN115441558B - Data center power supply architecture system - Google Patents
Data center power supply architecture system Download PDFInfo
- Publication number
- CN115441558B CN115441558B CN202211386089.0A CN202211386089A CN115441558B CN 115441558 B CN115441558 B CN 115441558B CN 202211386089 A CN202211386089 A CN 202211386089A CN 115441558 B CN115441558 B CN 115441558B
- Authority
- CN
- China
- Prior art keywords
- power supply
- discharge
- load
- task
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007599 discharging Methods 0.000 claims abstract description 50
- 238000011084 recovery Methods 0.000 claims abstract description 28
- 238000007689 inspection Methods 0.000 claims abstract description 27
- 230000002159 abnormal effect Effects 0.000 claims description 26
- 238000000605 extraction Methods 0.000 claims description 12
- 238000012544 monitoring process Methods 0.000 claims description 10
- 150000001875 compounds Chemical class 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 3
- 230000036541 health Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 239000002699 waste material Substances 0.000 abstract description 8
- 230000006870 function Effects 0.000 description 24
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005057 refrigeration Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000000034 method Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0063—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with circuits adapted for supplying loads from the battery
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/007—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources
- H02J3/0073—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources for providing alternative feeding paths between load and source when the main path fails, e.g. transformers, busbars
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0068—Battery or charger load switching, e.g. concurrent charging and load supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J9/00—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting
- H02J9/04—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source
- H02J9/06—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems
- H02J9/061—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems for DC powered loads
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J9/00—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting
- H02J9/04—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source
- H02J9/06—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems
- H02J9/08—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems requiring starting of a prime-mover
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention provides a data center power supply architecture system, which comprises power distribution equipment, UPS equipment, power management equipment and a data center load; the UPS equipment is provided with an electric storage battery pack; the power management equipment is connected with a data center load; the power management equipment is provided with a switching power supply module for switching the charging and discharging states of the storage battery pack; the data center load is provided with a main supply load, an inspection load and a disaster recovery load; the power storage battery pack is connected with the switching power supply module, and the switching power supply module is respectively connected with the inspection load and the disaster recovery load; the architecture system also includes a discharge management subsystem for controlling the switching of the charge-discharge states of the battery pack. Compared with the prior art, the invention takes the polling load and the disaster recovery load as the discharging load of the storage battery pack, and automatically controls the charging and discharging state of the storage battery pack through the discharging management subsystem, thereby improving the charging and discharging efficiency, and reducing the electric energy waste generated by discharging by fully utilizing the load.
Description
Technical Field
The invention relates to the field of power supply systems, in particular to a data center power supply architecture system.
Background
At present, the power supply of a data center is mainly divided into three parts: the UPS is connected with the mains for a long time, the storage battery can be in a floating charge state for a long time in the environment with high power supply quality and rarely occurring commercial power failure, the conversion activity of chemical energy and electric energy in the battery is reduced in the long-term past, the aging of the battery is accelerated, and the service life of the battery is shortened, so the storage battery needs to be charged and discharged periodically for maintenance, the storage battery is generally discharged completely every 2-3 months, the discharge time is determined according to the capacity of the storage battery and the size and the number of loads, and the storage battery needs to be charged for more than 8 hours after one full-load discharge. Because each charge and discharge needs to take a long time, when power failure occurs in the period, the storage battery cannot work well; in addition, the discharge of the storage battery is to discharge the useless load, which results in the waste of electric energy.
Disclosure of Invention
The invention aims to overcome at least one defect of the prior art, provides a power supply architecture system of a data center, and solves the problem of electric energy waste caused by charging and discharging of a storage battery in the data center.
The technical scheme of the invention is as follows:
the invention provides a data center power supply architecture system, which comprises power distribution equipment, UPS equipment, power management equipment and a data center load;
the distribution equipment is connected with the UPS equipment, and is externally connected with a commercial power and a generator;
the UPS equipment is connected with power management equipment, and an electric storage battery pack is arranged in the UPS equipment;
the power management equipment is connected with a data center load;
the power management equipment is provided with a switching power supply module, and the switching power supply module is used for switching the charging and discharging states of the storage battery pack;
the data center load is provided with a main supply load, an inspection load and a disaster recovery load;
the power storage battery pack is connected with a switching power supply module, and the switching power supply module is respectively connected with the inspection load and the disaster recovery load;
the framework system also comprises a discharge management subsystem, and the discharge management subsystem is used for calculating and controlling the charging and discharging state switching of the storage battery pack according to the power utilization state.
The inspection load and the disaster recovery load are configured to be in a power supply switching mode, namely, power can be supplied by using commercial power or can be switched to be supplied to the storage battery pack through the power supply switching module, when power is supplied to the storage battery pack, the inspection load and the disaster recovery load are used as discharging loads of the storage battery pack, the power supply switching module is respectively provided with independent discharging power supply loops for the inspection load and the disaster recovery load, the discharging power supply loops are coupled with the storage battery pack, the storage battery pack directly discharges the inspection load and the disaster recovery load through the discharging power supply loops, and the discharging management subsystem is used for managing the switching of the storage battery pack; because the discharging quality of the storage battery pack is poorer than that of a mains supply, the storage battery pack cannot be used for supplying power to loads such as a main power supply load directly related to a data center, and the requirements of the inspection load and the disaster-tolerant load on the power supply quality are not high, on one hand, the inspection function and the disaster-tolerant backup function corresponding to the inspection load and the disaster-tolerant load have self-repairing capability, respond and execute a task, if the power supply is insufficient and the task feedback is incomplete, data can be emptied and the task can be executed again, and great influence cannot be caused; on the other hand, the polling load and the disaster recovery load are timed tasks and can coordinate with the discharge of the storage battery pack, so that the discharge electric energy of the storage battery pack can be utilized to the maximum. Therefore, the circuit architecture can fully utilize the discharge electric energy of the storage battery, and meanwhile, the normal use of the data center is not influenced, so that the work is ensured, and the electric energy and the energy are saved. Meanwhile, the charging and discharging state switching of the electric storage battery is managed by using the discharging management subsystem, so that the charging and discharging switching process can be automatically carried out, the manual participation is reduced, and the working efficiency is improved.
Furthermore, the discharge management subsystem comprises a power supply prediction module, an electric power storage monitoring module and a load management module;
the power supply prediction module is used for acquiring power supply state information and calculating a power supply prediction waveform according to the power supply state information, wherein the power supply prediction waveform reflects the power supply quality of the next time period;
the electric power storage monitoring module is used for acquiring state information of the electric power storage battery pack and updating a discharge reliable value according to the state information of the electric power storage battery pack, and the discharge reliable value reflects a theoretical relation between the state of the electric power storage battery pack and time;
the load management module is used for acquiring load state information and generating an accumulated task value according to the load state information;
the discharge management subsystem is configured with a discharge triggering strategy, the discharge triggering strategy is used for determining discharge triggering time according to the power supply predicted waveform, the discharge reliable value, the accumulated task value and the discharge equilibrium value, a discharge triggering instruction is sent to the power supply management equipment at the discharge triggering time, and the power supply management equipment receives the discharge triggering instruction to control the switching power supply module to switch the charge-discharge state of the storage battery pack;
the discharge equilibrium value reflects a relationship between the degree of reliability of the time environment and the change of the discharge demand with time.
Further, the power supply prediction module calculates the power supply prediction waveform according to a power supply prediction algorithm, where the power supply prediction algorithm is:
in the formula (I), the compound is shown in the specification,
a waveform is predicted for the power supply in question,
the characteristic related weight corresponding to the nth power supply state information is
，
The feature similarity of the waveform feature of the previous time period and the waveform feature of the current time period of the nth power supply state information，
The weight of the time interval corresponding to the nth power supply state information is
，
The parameters are adjusted for a preset time period,
the time interval between the time corresponding to the nth power supply state information and the current time,
a load redundancy waveform corresponding to the nth power supply state information, wherein the load redundancy waveform reflects the relation between power supply quantity redundancy and time, and the
And providing a power supply quality waveform corresponding to the nth power supply state information, wherein the power supply quality waveform reflects the relation between the power supply quality and the time.
Further, the power supply prediction module is provided with a waveform feature extraction unit and a feature matching table;
the waveform feature extraction unit is used for extracting the waveform feature of the power supply waveform;
the characteristic matching table records a characteristic matching value corresponding to each waveform characteristic;
the feature similarity is the sum of feature matching values corresponding to the same waveform features in a period of time.
Furthermore, the power supply prediction module is also provided with a harmonic extraction unit and a power supply abnormity table;
an abnormal multiplier corresponding to each power supply abnormal information is recorded in the power supply abnormal table;
the harmonic extraction unit is used for extracting harmonic waveforms in the power supply waveforms, acquiring corresponding abnormal multipliers from the power supply abnormal table according to power supply abnormal information, and processing the harmonic waveforms through the abnormal multipliers to obtain corresponding power supply quality waveforms.
Through the power supply prediction waveform, the power supply situation in a future period of time can be predicted to decide a discharge strategy; because the discharge action can affect the UPS power supply, a time period with better power supply quality and less load quantity is selected for discharging; the power supply prediction waveform is obtained by a power supply prediction algorithm, the power supply prediction waveform is related to the historical power supply state, an abnormal multiplier is added on the basis of prediction through the historical power supply state, the power supply abnormal state is also brought into prediction, and the accuracy and the reliability of the prediction result are improved.
Further, the power storage monitoring module updates the discharge reliability value according to a discharge reliability function, where the discharge reliability function is:
in the formula (I), the compound is shown in the specification,
in order to be a reliable value for the discharge,
is a parameter of the state of health,
is a variable of the time, and is,
the time interval between the last complete discharge time and the current time,
for the total temporary discharge value, there are:
wherein
In order to average the time interval between discharges,
is a preset discharge reliability parameter and is,
the discharge capacity of the m-th temporary discharge.
By averaging the discharge time interval
The frequency of temporary discharge can be judged, and whether the floating condition exists or not can be judged according to the discharge amount of each discharge; meanwhile, the weight relation of each discharge electric quantity can be calculated according to the times of the discharge electric quantity, the nearest discharge can reflect the influence of the discharge on the floating condition most, the mth discharge is set as the earliest discharge, the 1 st discharge is the nearest discharge, and the calculation is carried out in a proportional weighting mode, so that a more accurate result can be obtained, and a discharge reliable value can be obtained
And time variable
And further predicting the discharge reliability under the time change.
Furthermore, the load management module is provided with a load task list and a task characteristic database;
the load task list is used for storing load tasks;
the task characteristic database is used for storing task priority levels, task demand levels and task power consumption levels corresponding to the load tasks;
the load management module extracts a task priority level, a task demand level and a task power consumption level corresponding to a load task of a load task list, and calculates an accumulated task sub-value of each load task according to a task value algorithm;
the task value algorithm is as follows:
in the formula (I), the compound is shown in the specification,
in order to accumulate the sub-values of the task,
in order to be the task timeout time,
in order to be the priority level of the task,
in order to be the level of the task requirements,
a power consumption value for the task;
and the accumulated task value is the sum of the accumulated task sub-values.
The demand degree of the current load task can be acquired through the accumulated task value, so that the power supply demand meeting all the load tasks is calculated, and a corresponding power supply strategy is formulated.
Further, the discharge triggering strategy specifically includes:
a1: acquiring a power supply predicted waveform, a discharge reliable value and an accumulated task value;
a2: configuring a task triggering condition, entering the step A3 when the accumulated task value meets the task triggering condition, and returning to the step A1 if the accumulated task value does not meet the task triggering condition;
a3: a discharge triggering condition is configured, when the discharge reliable value meets the discharge triggering condition, the step A4 is carried out, otherwise, a task execution instruction is generated and the step A1 is returned;
a4: configuring a preset discharge threshold value, and screening a discharge time interval with a discharge equilibrium value larger than the discharge threshold value
Taking the discharge triggering time as the discharge triggering time, and entering the step A5;
a5: matching the load task within each trigger undetermined time to ensure that the accumulated task value corresponding to the load task, determining the trigger undetermined time with the highest accumulated task value and the residual time greater than the preset reference residual power time as the discharge trigger time, and entering the step A6;
a6: and generating the discharge triggering instruction according to the combination of the corresponding matched load tasks.
Further, the task triggering condition is that the sum of the priority levels of the tasks in the load task list is greater than a preset priority triggering threshold;
the discharge triggering condition is as follows
And when the discharge reliability value is higher than the preset discharge triggering threshold value, the corresponding discharge reliability value is higher than the preset discharge triggering threshold value.
Further, the step A5 further includes:
a51: dividing the trigger undetermined time according to the task demand grade to obtain a discharge quality partition;
a52: sorting the load tasks from high to low according to the product of the task priority level of the load tasks and the task expiration time;
a53: sequentially matching the sequenced load tasks to the corresponding discharge quality subareas until the total power consumption value of the discharge quality subareas exceeds the discharge value, and matching the next discharge quality subarea;
a54: the step of a53 is repeated until all discharge quality sector matches are completed.
The discharge balance value is calculated by a discharge balance function, and the discharge balance function is as follows:
and the discharge balance function meets the constraint condition
In the formula (I), wherein,
in the form of a function of the period,
the value of the power supply at full load,
the average supply amount of power for the power supply prediction waveform is redundant,
in order to discharge the equilibrium value of the discharge,
to adjust the parameters; when the discharge interval with the screening balance discharge value larger than the preset discharge threshold value is selected
Step A5 is entered as the trigger undetermined time, and the relation between the reliability of the time environment and the discharge requirement changing along with the time is calculated by calculating an equalization function,
is a preset value.
The discharging state is limited by setting task triggering conditions and discharging triggering conditions, when the task triggering adjustment is not met, namely the accumulated task value is lower, the demand degree of load tasks is lower, the discharging efficiency of the loads is lower, and the waste of electric energy is easily caused; when the discharge trigger condition is not satisfied, that is, when t =0, the discharge reliability value is lower than the trigger threshold value, and the discharge reliability is low. After the task triggering condition and the discharging triggering condition are met, the discharging time interval with the discharging balance value larger than the discharging threshold value is selected as the discharging triggering time to distribute the discharging task, the discharging balance value in the discharging triggering time meets the requirement, each discharging interval can meet the basic discharging requirement of the load task through screening of the accumulated task value and the discharging reliable value, on the basis, the optimal load task is matched for each discharging interval area to discharge, normal discharging is ensured, meanwhile, the load task can work at the optimal efficiency, and the waste of electric energy is reduced.
Compared with the prior art, the invention has the beneficial effects that:
1. discharge power supply loops are respectively arranged between the patrol load and the disaster recovery load and the storage battery pack, and the patrol load and the disaster recovery load have low requirements on power supply quality, so that the patrol load and the disaster recovery load can be well matched with the storage battery pack with low power supply quality, and the waste of electric energy caused by the discharge of the storage battery pack by using useless loads is avoided;
2. the discharging management subsystem is used for controlling the charging and discharging states of the electric storage battery pack when the data center normally works according to the load tasks of the inspection load and the disaster recovery load and the states of the electric storage battery pack, long discharging time does not need to be additionally set, discharging and load power supply are combined, the discharging and charging states are automatically completed, and working efficiency is improved.
Drawings
FIG. 1 is a diagram of a data center power supply architecture of the present invention;
FIG. 2 is a block diagram of a discharge management subsystem of a data center power supply architecture in accordance with the present invention;
FIG. 3 is a diagram of a discharge balancing function of a data center power supply architecture according to the present invention;
the attached drawings are marked as follows: the system comprises a mains supply 101, a generator 102, a power distribution device 103, a UPS device 104, a storage battery pack 114, a power management device 105, a switching power supply module 115, a data center load 106, an inspection load 116, a disaster recovery load 126, a main power supply load 136, a power supply prediction module 210, a storage monitoring module 220 and a load management module 230.
Detailed Description
The drawings are only for purposes of illustration and are not to be construed as limiting the invention. For a better understanding of the following embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
Example 1
The present embodiment provides a data center power supply architecture system, as shown in fig. 1, the architecture system includes a power distribution device 103, a UPS device 104, a power management device 105, and a data center load 106;
the power distribution equipment 103 is connected with the UPS equipment 104, and the power distribution equipment 103 is also externally connected with a mains supply 101 and a generator 102;
the UPS device 104 is connected to the power management device 105, the UPS device 104 is provided with an electric storage battery pack 114, the electric storage battery pack 114 starts to work when power is cut off or lost, so as to ensure normal power supply of a power supply center, and in order to ensure normal and effective work of the electric storage battery pack 114, the electric storage battery pack 114 needs to be discharged periodically;
the power management device 105 is connected with a data center load 106;
a switching power supply module 115 is arranged in the power management device 105, and the switching power supply module 115 is used for switching the charging and discharging states of the storage battery pack;
the data center load 106 is provided with a main supply load 136, an inspection load 116 and a disaster recovery load 126, the inspection load 116 is mainly used as an inspection function, such as an inspection robot, an inspection sensor, an inspection terminal and other devices, and the disaster recovery load 126 is mainly used as a disaster recovery backup function of the data center, such as a disaster recovery backup server, a disaster recovery memory, a disaster recovery storage switching device and the like;
the electric storage battery pack 114 is connected with a switching power supply module 115, the switching power supply module 115 is respectively connected with the inspection load 116 and the disaster recovery load 126, specifically, the switching power supply module 115 is respectively provided with an independent discharging power supply loop for the inspection load 116 and the disaster recovery load 126, the inspection load 116 and the disaster recovery load 126 can be directly supplied with power by a UPS (uninterruptible power supply) by setting the switching power supply module 115, and can also be switched to the electric storage battery pack 114 for supplying power, and when the electric storage battery pack 114 is switched for supplying power, the electric storage battery pack 114 discharges through the inspection load 116 and the disaster recovery load 126. Due to the arrangement, the normal operation of the load can be ensured while the storage battery pack 114 is effectively discharged, because the power supply quality of the storage battery pack 114 is lower than that of the commercial power 101, the direct service of all the loads can cause the power loss or undervoltage of key modules of a data center to cause data damage or equipment damage, and the effect of complete discharge is difficult to achieve, while if the storage battery pack 114 is discharged by useless loads, the waste of electric energy is caused, and the functions of the inspection load 116 and the disaster-tolerant load 126 are described above, on one hand, the data of the inspection function and the disaster-tolerant backup function have the self-repairing capability, if a fault occurs, the data is fed back to be incomplete, the data is cleared for re-execution, and the requirement on the power supply quality is not high; on the other hand, the polling function and the disaster recovery backup function are timed tasks and do not need to be executed for a long time, so that the system can work by matching the discharge of the storage battery, and the utilization efficiency of electric energy is improved.
Specifically, the discharge power supply circuit includes a dc power supply circuit and an ac power supply circuit, the dc power supply circuit is coupled to a dc load through a transformer circuit, and the ac power supply circuit is coupled to an ac load through an inverter circuit and the transformer circuit. Because the discharge point power supply loop is directly arranged, the discharge point power supply loop can be directly configured according to equipment requirements, namely, for example, part of equipment is supplied with 220V alternating current or 380V alternating current, the discharge point power supply loop can be directly realized through an inverter circuit voltage boosting circuit, and if part of the load is supplied with 80V direct current or 15V direct current, the power supply loop can be directly supplied with power through a voltage boosting or reducing circuit, so that the loss caused by power conversion can be reduced, and meanwhile, each equipment can be independently switched to select whether to be connected into the discharge power supply loop for working, and on the other hand, the hardware structure support can be provided for the actions of different power supply equipment for realizing different power consumption qualities.
The architecture system in this embodiment further includes a discharge management subsystem as a software support part, where the discharge management subsystem is configured to calculate and control the charge/discharge state switching of the storage battery pack according to the power consumption state. As shown in fig. 2, the discharge management subsystem includes a power supply prediction module 210, an electrical storage monitoring module 220, and a load management module 230; since the discharge plan needs to take into account two parts, for the discharged battery, the service life can be guaranteed theoretically as long as the discharge is carried out in a similar time without strict specific time, so that the possibility of adjusting the discharge time selection is made.
Specifically, the power supply prediction module 210 is configured to obtain power supply state information, and calculate a power supply prediction waveform according to the power supply state information, where the power supply prediction waveform reflects power supply quality in a next time period; the power supply quality in a future period of time is predicted through the power supply prediction waveform, and then the period of time with better power supply quality and less load demand is selected for discharging, so that the discharging demand of the storage battery pack 114 can be met, and the load can work normally. The power supply prediction module 210 calculates a power supply prediction waveform through a power supply prediction algorithm, which is:
in the formula (I), the compound is shown in the specification,
a waveform is predicted for the power supply in question,
the characteristic related weight corresponding to the nth power supply state information is
，
The feature similarity of the waveform feature of the previous time period of the nth power supply state information and the waveform feature of the current time period reflects the waveform of the previous time period and the waveform of the current time periodThe similarity of the previously acquired waveforms aims to set different similarity weights for each waveform, and if n waveforms exist, the power supply quality and the power supply load of each waveform may be different, and if direct superposition prediction is performed, a large deviation occurs, and the reason is as follows: 1. the time interval of the historical waveform is long, the system is maintained for a long time, the change is large, and the referential is low; 2. the waveform characteristics of the previous period and the next period of each waveform may be related, for example, a load with a larger power supply is connected, or the refrigeration equipment operates at a high power, so that a more accurate result can be obtained by analyzing the relationship between the waveform characteristics of the current period and the waveform characteristics of the previous period of each historical waveform.
According to each parameter in the power supply prediction algorithm, a waveform feature extraction unit and a feature matching table are arranged in the power supply prediction module 210, and the waveform feature extraction unit is used for extracting the waveform feature of the power supply waveform; the feature matching table records a feature matching value corresponding to each waveform feature, and the feature similarity is a sum of feature matching values corresponding to the same waveform feature in a time period.
The above-mentioned
And providing a power supply quality waveform corresponding to the nth power supply state information, wherein the power supply quality waveform reflects the relation between the power supply quality and the time. The power supply quality waveform is obtained by analyzing the abnormal condition of the waveform and performing multiplier operation on the harmonic waveform, so a harmonic extraction unit and a power supply abnormal table are also arranged in the power supply prediction module 210, an abnormal multiplier corresponding to each piece of power supply abnormal information is pre-recorded in the power supply abnormal table, the harmonic extraction unit is used for extracting the harmonic waveform in the power supply waveform and inquiring the power supply abnormal table according to the harmonic waveform, if the harmonic waveform is abnormal, the corresponding abnormal multiplier can be extracted from the power supply abnormal table, and then the power supply quality waveform can be obtained by performing multiplier operation on the harmonic waveform and the abnormal multiplier, and then the power supply prediction waveform is calculated.
Specifically, the storage monitoring module 220 is configured to monitor the storage battery pack 114, if the storage battery is frequently used, the discharge interval period may be appropriately increased, if the storage battery is not frequently used, the discharge interval period may be decreased, and on the other hand, the discharge interval period may be related to the actual health status of the storage battery and the full discharge time at the current time, and the storage monitoring module 220 is further configured to obtain the status of the storage battery pack 114 and update the discharge reliability value according to the status of the storage battery pack 114 by using a discharge reliability function:
in the formula (I), the compound is shown in the specification,
in order to be a reliable value for the discharge,
is a parameter of the state of health,
is a variable of the time, and is,
the time interval between the last full discharge time and the current time,
for the total value of temporary discharge, there are
Wherein
In order to average the time interval between discharges,
is a preset discharge reliability parameter and is a preset discharge reliability parameter,
the discharge capacity of the m-th temporary discharge.
By the arrangement, on one hand, the frequency of temporary discharge can be judged according to the average discharge interval; on the other hand, whether the floating condition exists can be judged according to the electric quantity discharged every time. The weight relation of each discharge electric quantity is calculated through the discharge times, the influence of the discharge on the floating condition can be reflected most by the discharge of the nearest time, namely the mth discharge is the earliest discharge, the weight relation is lower, the 1 st temporary discharge is the discharge of the nearest time, the weight relation is higher, and the discharge time is the discharge time of the first discharge, and the discharge time is the discharge time of the second discharge, so that the discharge time is the discharge time of the first discharge, and the discharge time is the discharge time of the second dischargeThen, operation is carried out in a proportion weighting mode, and a relatively accurate result can be obtained; finally get about
And
thereby predicting the discharge reliability under the time change.
Specifically, the load management module 230 is configured to obtain load state information, and generate an accumulated task value according to the load state information; the load management module 230 is provided with a load task list and a task feature database; the load task list is used for storing load tasks; the task characteristic database is used for storing task priority levels, task demand levels and task power consumption levels corresponding to the load tasks; the load management module 230 extracts a task priority level, a task demand level and a task power consumption level corresponding to a load task in a load task list, and calculates an accumulated task sub-value of each load task according to a task value algorithm;
the task value algorithm is as follows:
in the formula (I), the compound is shown in the specification,
in order to accumulate the sub-values of the task,
in order to be the task timeout time,
in order to be the priority level of the task,
in order to be the task demand level,
a power consumption value for the task; and the accumulated task value is the sum of the accumulated task sub-values.
Specifically, the polling load tasks include periodically counting temperature change conditions, analyzing temperature control response efficiency, or periodically judging response and execution efficiency of the main server to a duty data packet, periodically detecting temperature feedback efficiency, periodically calling error reporting logs of a data center, and the like, and then calculating corresponding accumulated task sub-values for each task according to assigned priorities, demand grades and power consumption values to complete calculation of the demand degree of the corresponding load tasks.
The demand degree of the current load task can be acquired through the accumulated task value, so that the power supply demand meeting all the load tasks is calculated, and a corresponding power supply strategy is formulated.
Specifically, a discharge triggering strategy is configured in the discharge management subsystem to control the switching of the discharge and charge states of the storage battery pack 114, and the discharge triggering strategy specifically includes:
a1: acquiring a power supply predicted waveform, a discharge reliable value and an accumulated task value;
a2: configuring a task triggering condition, entering the step A3 when the accumulated task value meets the task triggering condition, and otherwise, returning to the step A1; the task triggering condition is that the sum of the task priority levels in the load task list is greater than a preset priority triggering threshold, and the accumulated task value is obtained by calculation according to the task priority levels, so that judgment can be performed by the accumulated task value, at this time, the corresponding load task needs to be executed, and the next judgment is performed.
A3: configuring a discharge triggering condition, entering the step A4 when the discharge reliable value meets the discharge triggering condition, otherwise generating a task execution instruction and returning to the step A1; the discharge triggering condition is as follows
If the corresponding discharge reliability value is lower than the preset discharge triggering threshold value, that is, if the discharge reliability value is higher, the load task is executed by discharging storage battery pack 114, otherwise, by UPAnd S, supplying power to execute a load task.
A4: configuring a preset discharge threshold value, and screening a discharge time interval with a discharge equilibrium value larger than the discharge threshold value
Step A5 is entered as the discharge triggering time; specifically, the discharge equalization value is calculated by a discharge equalization function, which is:
and the discharge balance function meets the constraint condition
Wherein, in the step (A),
in order to be a function of the period,
the value of the power supply at the full load,
the average supply amount of power for the power supply prediction waveform is redundant,
in order to discharge the equilibrium value of the discharge,
to adjust the parameters; when a discharging interval with the equilibrium discharging value larger than a preset discharging threshold value exists
Selecting the interval
Entering step A5 as the trigger undetermined time, and calculating the time by calculating the balance functionThe environmental reliability and the discharge demand over time,
to a preset value, by adjusting
The corresponding weight proportion and function sensitivity can be adjusted, constraint conditions of functions are met at the same time, namely a figure taking range is selected according to the proportion of the actual total time, namely the greater the actual average redundancy is, the greater the figure taking range is, and in the figure taking range serving as a time interval, the area formed by the surrounding arrangement of the two corresponding functions is calculated so as to complete the calculation of the discharge equilibrium value, as shown in fig. 3, then the interval of the discharge equilibrium value is used as the trigger undetermined time, a plurality of intervals can be obtained, and it is required to be ensured that no intersection exists between the intervals or the interval between the intervals does not exceed a preset value.
A5: matching the load task within each trigger undetermined time to ensure that the accumulated task value corresponding to the load task, determining the trigger undetermined time with the highest accumulated task value and the residual time greater than the preset reference residual power time as the discharge trigger time, and entering the step A6;
the specific step A5 further includes:
a51: dividing the trigger undetermined time according to the task demand grade to obtain a discharge quality partition;
a52: sorting the load tasks from high to low according to the product of the task priority level of the load tasks and the task expiration time;
a53: sequentially matching the sequenced load tasks to the corresponding discharge quality subareas until the total power consumption value of the discharge quality subareas exceeds a discharge value, and matching the next discharge quality subarea;
a54: the step of a53 is repeated until all discharge quality sector matches are completed.
Through the previous screening, the load tasks in the step A5 are all the load tasks meeting the conditions, but further distribution is needed to better remove matching, the trigger undetermined time is divided by the series of the task demand grades to obtain the discharge quality partitions, and the corresponding load tasks are sequentially matched to the corresponding discharge quality partitions by taking the product of the task priority grade and the task timeout time as a sequence until the total task power consumption value of the discharge quality partition exceeds the total discharge value, so that all discharge intervals can be utilized most reasonably, the normal operation of discharge is ensured, the load tasks are executed with the optimal efficiency, and the waste of electric energy is reduced.
A6: and generating the discharge triggering instruction according to the combination of the corresponding matched load tasks.
It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the technical solutions of the present invention, and are not intended to limit the specific embodiments of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention claims should be included in the protection scope of the present invention claims.
Claims (9)
1. A data center power supply architecture system comprises power distribution equipment, UPS equipment, power management equipment and a data center load;
the power distribution equipment is connected with the UPS equipment, and is externally connected with a commercial power and a generator;
the UPS equipment is connected with power management equipment, and an electric storage battery pack is arranged in the UPS equipment;
the power management equipment is connected with a data center load;
the power management device is characterized in that a switching power supply module is arranged in the power management device and used for switching the charging and discharging states of a storage battery pack;
the data center load is provided with a main supply load, an inspection load and a disaster recovery load;
the power storage battery pack is connected with a switching power supply module, and the switching power supply module is respectively connected with the inspection load and the disaster recovery load;
the architecture system also comprises a discharge management subsystem, wherein the discharge management subsystem is used for calculating and controlling the charging and discharging state switching of the storage battery pack according to the power utilization state;
the discharge management subsystem comprises a power supply prediction module, an electric power storage monitoring module and a load management module;
the power supply prediction module is used for acquiring power supply state information and calculating a power supply prediction waveform according to the power supply state information, wherein the power supply prediction waveform reflects the power supply quality of the next time period;
the electric power storage monitoring module is used for acquiring state information of the electric power storage battery pack and updating a discharge reliable value according to the state information of the electric power storage battery pack, and the discharge reliable value reflects a theoretical relation between the state of the electric power storage battery pack and time;
the load management module is used for acquiring load state information and generating an accumulated task value according to the load state information;
the discharge management subsystem is configured with a discharge triggering strategy, the discharge triggering strategy is used for determining discharge triggering time according to the power supply predicted waveform, the discharge reliable value, the accumulated task value and the discharge equilibrium value, a discharge triggering instruction is sent to the power supply management equipment at the discharge triggering time, and the power supply management equipment receives the discharge triggering instruction to control the switching power supply module to switch the charge-discharge state of the storage battery pack;
the discharge equilibrium value reflects a relationship between the degree of reliability of the time environment and the change of the discharge demand with time.
2. The data center power supply architecture system of claim 1, wherein the power supply prediction module calculates the power supply prediction waveform according to a power supply prediction algorithm, the power supply prediction algorithm being:
in the formula (I), the compound is shown in the specification,
a waveform is predicted for the power supply in question,
the characteristic related weight corresponding to the nth power supply state information comprises
，
For the feature similarity of the waveform feature of the previous time period of the nth power supply state information and the waveform feature of the current time period,
the weight of the time interval corresponding to the nth power supply state information is
，
The parameters are adjusted for a pre-set time,
the time interval between the time corresponding to the nth power supply state information and the current time,
a load redundancy waveform corresponding to the nth power supply state information, wherein the load redundancy waveform reflects the relation between power supply quantity redundancy and time, and the
And providing a power supply quality waveform corresponding to the nth power supply state information, wherein the power supply quality waveform reflects the relation between the power supply quality and the time.
3. The data center power supply architecture system according to claim 2, wherein the power supply prediction module is provided with a waveform feature extraction unit and a feature matching table;
the waveform feature extraction unit is used for extracting the waveform feature of the power supply waveform;
the characteristic matching table records a characteristic matching value corresponding to each waveform characteristic;
the feature similarity is the sum of feature matching values corresponding to the same waveform features in a period of time.
4. The data center power supply architecture system according to claim 2, wherein the power supply prediction module is further provided with a harmonic extraction unit and a power supply abnormity table;
an abnormal multiplier corresponding to each power supply abnormal information is recorded in the power supply abnormal table;
the harmonic extraction unit is used for extracting harmonic waveforms in the power supply waveforms, acquiring corresponding abnormal multipliers from the power supply abnormal table according to power supply abnormal information, and processing the harmonic waveforms through the abnormal multipliers to obtain corresponding power supply quality waveforms.
5. A data center power supply architecture system as claimed in claim 1, wherein the electrical storage monitoring module updates the discharge reliability value according to a discharge reliability function, the discharge reliability function being:
in the formula (I), the compound is shown in the specification,
in order to be a reliable value for the discharge,
is a parameter of the state of health,
is a variable of the time, and is,
the time interval between the last complete discharge time and the current time,
for the total temporary discharge value, there are:
wherein
In order to average the time interval between discharges,
is a preset discharge reliability parameter and is,
the discharge capacity of the m-th temporary discharge.
6. The data center power supply architecture system according to claim 1, wherein the load management module is provided with a load task list and a task feature database;
the load task list is used for storing load tasks;
the task characteristic database is used for storing task priority levels, task demand levels and task power consumption levels corresponding to the load tasks;
the load management module extracts a task priority level, a task demand level and a task power consumption level corresponding to a load task of a load task list, and calculates an accumulated task sub-value of each load task according to a task value algorithm;
the task value algorithm is as follows:
in the formula (I), the compound is shown in the specification,
in order to accumulate the sub-values of the task,
in order to be the task timeout time,
in order to be the priority level of the task,
in order to be the level of the task requirements,
a power consumption value for the task;
and the accumulated task value is the sum of the accumulated task sub-values.
7. The data center power supply architecture system according to claim 1, wherein the discharge triggering strategy is specifically:
a1: acquiring a power supply predicted waveform, a discharge reliable value and an accumulated task value;
a2: configuring a task triggering condition, entering the step A3 when the accumulated task value meets the task triggering condition, and returning to the step A1 if the accumulated task value does not meet the task triggering condition;
a3: a discharge triggering condition is configured, when the discharge reliable value meets the discharge triggering condition, the step A4 is carried out, otherwise, a task execution instruction is generated and the step A1 is returned;
a4: configuring a preset discharge threshold value, and screening a discharge time interval with a discharge equilibrium value larger than the discharge threshold value
Taking the discharge triggering time as the discharge triggering time, and entering the step A5;
a5: matching the load task within each trigger undetermined time to ensure that the accumulated task value corresponding to the load task, determining the trigger undetermined time with the highest accumulated task value and the residual time greater than the preset reference residual power time as the discharge trigger time, and entering the step A6;
a6: and generating the discharge triggering instruction according to the combination of the corresponding matched load tasks.
8. The data center power supply architecture system according to claim 7, wherein the task trigger condition is that a sum of priority levels of tasks in the load task list is greater than a preset priority trigger threshold;
the discharge triggering condition is as follows
And when the discharge reliability value is higher than the preset discharge triggering threshold value, the corresponding discharge reliability value is higher than the preset discharge triggering threshold value.
9. The data center power architecture system of claim 7, wherein the step A5 further comprises:
a51: dividing the trigger undetermined time according to the task demand grade to obtain a discharge quality partition;
a52: sorting the tasks from high to low according to the product of the task priority level of the load task and the task expiration time;
a53: sequentially matching the sequenced load tasks to the corresponding discharge quality subareas until the total power consumption value of the discharge quality subareas exceeds the discharge value, and matching the next discharge quality subarea;
a54: the step of a53 is repeated until all discharge quality sector matches are completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211386089.0A CN115441558B (en) | 2022-11-07 | 2022-11-07 | Data center power supply architecture system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211386089.0A CN115441558B (en) | 2022-11-07 | 2022-11-07 | Data center power supply architecture system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115441558A CN115441558A (en) | 2022-12-06 |
| CN115441558B true CN115441558B (en) | 2023-01-17 |
Family
ID=84252409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211386089.0A Active CN115441558B (en) | 2022-11-07 | 2022-11-07 | Data center power supply architecture system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115441558B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101276944B (en) * | 2008-05-12 | 2010-06-16 | 绿源投资控股集团有限公司 | Accumulator repairing system and method with function of recovering electric energy |
| CN102549871A (en) * | 2010-08-23 | 2012-07-04 | 三洋电机株式会社 | Power management system |
| US9337683B2 (en) * | 2012-12-20 | 2016-05-10 | Powergenix Systems, Inc. | Controlling battery states of charge in systems having separate power sources |
| CN106100050A (en) * | 2016-07-19 | 2016-11-09 | 西安交通大学 | A kind of data center based on layering mixed tensor storage system and consumption control method |
| CN115000992B (en) * | 2022-07-18 | 2022-11-01 | 联通(广东)产业互联网有限公司 | Green standby power supply regulation and control system for data center |
-
2022
- 2022-11-07 CN CN202211386089.0A patent/CN115441558B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115441558A (en) | 2022-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110707804B (en) | Energy storage and standby power device with online test function and test method | |
| JP5801706B2 (en) | IT equipment and storage battery linkage control system and linkage control method | |
| US11165264B2 (en) | Battery management systems having battery failure detection and related methods and uninterruptible power supplies (UPSs) | |
| EP4131706A1 (en) | Charging and discharging management method for parallel battery packs, electronic device, and electrical system | |
| US20120176095A1 (en) | Electric power management system | |
| JP3892812B2 (en) | Power supply | |
| US11949273B2 (en) | Method for managing charging and discharging of parallel-connected battery pack, electronic device, and electrical system | |
| KR102268033B1 (en) | An Intergrated Solution System of PV-ESS with Reinforced Safety | |
| CN105703475A (en) | Method and system of uninterrupted intelligent power supply of ATM | |
| WO2017135069A1 (en) | Management device and power storage system | |
| CN116601509A (en) | Control method of energy storage system and energy storage system | |
| CN113300403B (en) | Solar micro-grid energy storage capacity configuration optimization processing system and processing method | |
| CN115378015B (en) | Operation control method, system, equipment and medium of micro-grid | |
| CN115441558B (en) | Data center power supply architecture system | |
| CN116191626B (en) | Battery management method, device and system for household energy storage system | |
| KR101729273B1 (en) | Method and energy management system for controlling maximum demand power of energy storage system comprising plurality of uninterruptible power supply | |
| WO2023027137A1 (en) | Power control device, power control system, and power control method | |
| CN108767883B (en) | Response processing method of demand side | |
| US20100312411A1 (en) | Ac consumption controller, method of managing ac power consumption and a battery plant employing the same | |
| CN115102187A (en) | Island microgrid management system and energy management method thereof | |
| Liu et al. | A Reliability Analysis and Comparison of Battery Energy Storage Systems | |
| CN111913110B (en) | Commercial power battery configuration evaluation method and system | |
| CN113013980A (en) | UPS system, and control method and device of UPS system | |
| WO2004082052A1 (en) | Power supply system | |
| US20180069400A1 (en) | Power source system and method of resetting power-assist start point for the system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |