A photovoltaic power storage integrated data center machine room micro-grid control method and system

Abstract

The application provides a photovoltaic and energy storage integrated data center machine room microgrid control method and system, which comprises obtaining photovoltaic output power, energy storage state of charge, real-time power consumption of machine room IT and air conditioner load, and public connection point voltage frequency; calculating air conditioner power consumption feedforward compensation amount according to a machine room thermodynamic model, ambient temperature and next period IT load prediction value, and adjusting energy storage power scheduling reference; calculating IT load interruptible level index according to photovoltaic power prediction and actual value deviation and energy storage state of charge, adjusting threshold value, and sending a load reduction request to an IT resource management system when the index is over the threshold value; calculating voltage reference floating amount, superimposing to obtain voltage control instruction, combining frequency adjustment instruction generated by frequency deviation and adjusted energy storage power scheduling reference to generate a control signal and send the control signal to a photovoltaic inverter and an energy storage converter.

Description

Technical Field

[0001] This application belongs to the field of control, and in particular relates to a control method and system for a data center microgrid integrating photovoltaic and energy storage. Background Technology

[0002] Microgrid systems enable local energy consumption and can provide power to data centers in both grid-connected and off-grid modes. However, photovoltaic power generation is intermittent and fluctuating, and data center loads exhibit complex characteristics. IT loads change in real time with computing tasks, while cooling system power consumption is related to IT load and ambient temperature, leading to difficulties in source-load matching within the microgrid. The superposition of these uncertainties can easily cause voltage and frequency fluctuations at the microgrid's common interconnection point, affecting the safe and stable operation of data center equipment.

[0003] A droop control strategy is used to initially regulate voltage and frequency, followed by real-time power scheduling through an energy management system. However, existing methods react passively to load fluctuations, particularly for air conditioning power consumption strongly correlated with IT loads, lacking predictive and compensation mechanisms, leading to lag in energy storage system responses. In emergency situations involving source-load mismatch, load shedding strategies based on fixed state of charge or power deviation thresholds cannot adapt to the real-time operating status of the microgrid, potentially causing unnecessary service interruptions. Therefore, a control method capable of coordinated prediction, adjustment, and multi-objective optimization is urgently needed to improve the stability and reliability of data center microgrids. Summary of the Invention

[0004] This invention proposes a photovoltaic-energy storage integrated microgrid control method for data center computer rooms, which addresses the problem that existing methods cannot achieve coordinated prediction, adjustment, and multi-objective optimization control, including: The system acquires the output power of photovoltaic units, the state of charge of energy storage units, the real-time power consumption of data center IT load and air conditioning load, and the voltage and frequency of common interconnection points within the microgrid. Based on the preset data center thermodynamic model, ambient temperature, and the predicted IT load for the next period, the system calculates the air conditioning power consumption feedforward compensation amount and adjusts the power scheduling benchmark of the energy storage units according to the air conditioning power consumption feedforward compensation amount. Based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit, the IT load interruptibility level index is calculated; and a preset threshold is adjusted according to the state of charge of the energy storage unit; when the IT load interruptibility level index exceeds the adjusted preset threshold, a load reduction request is sent to the IT resource management system. Based on the voltage deviation of the common correlation point, a basic voltage regulation command is generated. Combined with the state of charge of the energy storage unit and the interruptibility level index of the IT load, the voltage reference fluctuation is calculated. The voltage reference fluctuation is then superimposed with the basic voltage regulation command to form a voltage control command. Frequency adjustment commands are generated based on the frequency deviation of the common correlation point, and control signals are generated and sent to the photovoltaic inverter and energy storage converter based on the adjusted energy storage unit power scheduling benchmark, the voltage control command and the frequency adjustment command.

[0005] Furthermore, this invention also relates to a photovoltaic-energy storage integrated microgrid control system for data center computer rooms, comprising the following modules: The adjustment module is used to obtain the output power of photovoltaic units, the state of charge of energy storage units, the real-time power consumption of data center IT load and air conditioning load, and the voltage and frequency of common interconnection points in the microgrid; based on the preset data center thermodynamic model, ambient temperature and the predicted value of IT load in the next period, it calculates the air conditioning power consumption feedforward compensation amount, and adjusts the power scheduling benchmark of energy storage units according to the air conditioning power consumption feedforward compensation amount. The sending module is used to calculate the IT load interruptibility level index based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit; and to adjust a preset threshold according to the state of charge of the energy storage unit; when the IT load interruptibility level index exceeds the adjusted preset threshold, a load reduction request is sent to the IT resource management system. The superposition module is used to generate a basic voltage regulation command based on the voltage deviation of the common correlation point, and calculate the voltage reference fluctuation by combining the state of charge of the energy storage unit and the IT load interruptibility level index. The voltage reference fluctuation is superimposed with the basic voltage regulation command to form a voltage control command. The generation module is used to generate a frequency adjustment command based on the frequency deviation of the common correlation point, and generate a control signal based on the adjusted energy storage unit power scheduling benchmark, the voltage control command and the frequency adjustment command, and send it to the photovoltaic inverter and the energy storage converter.

[0006] This invention predicts air conditioning power consumption based on a thermodynamic model and provides feedforward compensation for energy storage, thereby mitigating power surges caused by air conditioning load fluctuations and enhancing the system's power balance capability. It determines the interruptibility level of IT loads by assessing photovoltaic prediction deviations and energy storage state of charge, and sets load reduction trigger conditions that change with energy storage state of charge, making load reduction decisions more accurate in emergency situations, avoiding unnecessary service interruptions, and ensuring system safety under extreme conditions. By utilizing energy storage state of charge and load-side regulation potential to handle voltage control and coordinating it with power and frequency control, it achieves synergistic optimization of voltage, frequency, and power, improving the overall operational stability of the microgrid. Attached Figure Description

[0007] Figure 1 A flowchart of the first embodiment; Figure 2 This is a diagram illustrating the load reduction threshold. Figure 3 A schematic diagram is needed to determine the power reduction for the load; Figure 4 This is a schematic diagram for calculating the interruptibility level index. Detailed Implementation

[0008] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0009] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0010] In the first embodiment, the present invention proposes a photovoltaic-energy storage integrated microgrid control method for data center computer rooms, such as... Figure 1 ,include: S1. Obtain the output power of photovoltaic units, state of charge of energy storage units, real-time power consumption of data center IT load and air conditioning load, and voltage and frequency of common interconnection points in the microgrid; calculate the air conditioning power consumption feedforward compensation amount based on the preset data center thermodynamic model, ambient temperature and IT load prediction value for the next period, and adjust the power scheduling benchmark of energy storage units according to the air conditioning power consumption feedforward compensation amount. By installing power metering modules or smart meters on the DC or AC side of the photovoltaic inverter, the AC side of the energy storage converter, and the power supply bus of the IT equipment and air conditioning system, the power consumption data of photovoltaic output power, IT load and air conditioning load are collected; the real-time state of charge (SOC) value is obtained by reading the internal register of the battery management system (BMS) of the energy storage system; a power quality monitoring device is deployed at the common connection point (PCC) between the microgrid and the main grid to measure and upload the three-phase voltage value and grid frequency at the PCC point in real time; all collected data are periodically sent to the central controller via industrial Ethernet or CAN bus using the Modbus communication protocol, with the acquisition period set to 100ms.

[0011] Establish an equivalent thermal parameter model for the data center. This model treats the entire server room as a centralized heat volume, and the rate of temperature change is related to the heat generation power of IT equipment, the cooling power of the air conditioning system, and the heat exchange power with the outside environment through walls and ventilation systems. For example, the established equivalent thermal parameter model for the server room is as follows: Where C represents the equivalent heat capacity of the computer room's air and overall structure, and its value can be estimated using physical parameters such as the computer room's design volume, air density, and specific heat capacity. Real-time temperature of the computer room. The target temperature for the computer room is [temperature], and the model control objective is to adjust [temperature] to achieve this. make Approaching ; This refers to the total heat dissipation power of the IT equipment. The heat exchange coefficient, which characterizes the intensity of heat exchange between the computer room and the external environment, can be calculated based on the materials, area, and insulation performance of the building envelope. For real-time monitoring of outdoor ambient temperature; This represents the actual cooling capacity provided by the air conditioning system. This model can predict... and the actual measurement Solving for the maintenance Stability required Predicted value.

[0012] The Long Short-Term Memory (LSTM) algorithm is used to predict the next control cycle based on historical IT load data. The LSTM prediction model is a network consisting of two LSTM layers, each with 128 neurons, and a fully connected output layer. Training data comes from the past year's IT load power time-series data of the data center, with a sampling interval of 15 minutes. After data standardization, 24 hours of continuous historical data are used as input to predict the load value for the next 15 minutes. The model uses mean squared error as the loss function and is trained using the Adam optimizer. After training, the model is deployed in the central controller and performs a rolling prediction every 15 minutes.

[0013] For example, IT load power in the next 15 minutes Using the real-time temperature collected by the outdoor ambient temperature sensor as input, and substituting it into the aforementioned thermodynamic model, the required air conditioning cooling power to maintain the computer room temperature within the preset target range is calculated. This is the feedforward compensation amount for air conditioning power consumption. This compensation amount is added to the original power dispatch benchmark of the energy storage unit. A new scheduling benchmark is obtained. This means instructing the energy storage unit to send out or absorb that portion of power in advance to cope with upcoming changes in air conditioning load.

[0014] In an optional embodiment, the calculation of the air conditioning power consumption feedforward compensation based on a preset data center thermodynamic model, ambient temperature, and predicted IT load for the next time period includes: The predicted IT load for the next period is input into the data center thermodynamic model to calculate the corresponding total heat generation of the servers. Combined with the current ambient temperature, the total cooling capacity required to maintain the target temperature of the data center is calculated. Based on the measured energy efficiency ratio of the air conditioning system, the required total cooling capacity is converted into the predicted power consumption of the air conditioning system for the next period, and the predicted power consumption is used as the feedforward compensation amount.

[0015] Specifically, the predicted IT load for the next control period, such as the next fifteen minutes, is obtained, for example, 500 kW. This predicted value is input into a pre-established data center thermodynamic model. This model can be a simplified equivalent thermal parameter model, representing the thermal characteristics of servers, racks, and the building envelope through physical equations. The model calculates the total heat generated by the servers based on the input, which is typically approximately equal to the power consumption, i.e., 500 kW. Simultaneously, the model combines the obtained outdoor ambient temperature (e.g., 30°C) with the target data center temperature (e.g., 25°C) to calculate the ambient heat transferred through walls and windows, for example, 50 kW. Therefore, the total cooling capacity required to maintain the target data center temperature is the sum of the server heat generation and the ambient heat, i.e., 550 kW. The measured energy efficiency ratio (EER) of the air conditioning system is obtained from the air conditioning monitoring unit, for example, 3.0. Dividing the required total cooling capacity by this EER yields the predicted power consumption of the air conditioning system for the next period, approximately 183.3 kW. This predicted power consumption value of 183.3 kW is the feedforward compensation.

[0016] S2, Based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit, calculate the IT load interruptibility level index; and adjust the preset threshold according to the state of charge of the energy storage unit; when the IT load interruptibility level index exceeds the adjusted preset threshold, send a load reduction request to the IT resource management system. In an optional embodiment, the IT load interruptibility index is calculated based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit, using the following formula:

[0017] in, The deviation factor is obtained by normalizing the absolute value of the deviation between the predicted and actual values ​​of photovoltaic output power. The energy factor is obtained by normalizing the complement of the state of charge of the energy storage unit. and These are the preset weighting coefficients.

[0018] Obtain the predicted photovoltaic (PV) output power at the current moment, for example, 80 kW, and the actual power output of the PV array, for example, 60 kW. Calculate the absolute value of the deviation between the two, which is 20 kW. To eliminate the influence of dimensions, normalize this deviation value. Assuming the maximum possible deviation based on historical data is 100 kW, the normalized deviation factor is... The value is 0.2. This factor reflects the degree of uncertainty in renewable energy output.

[0019] Read the current state of charge (SOC) of the energy storage unit, for example, 30%. Calculate the complement, which is 0.7. This complement value itself is in the range of 0 to 1 and can be used as a normalized energy factor. ,at this time The value is 0.7. This factor is related to the system's reserve capacity; a higher value indicates lower energy storage capacity and greater power supply risk. The two factors are then weighted and summed according to preset weighting coefficients. If the weight representing volatility... Set to 0.4, representing the weight of the energy storage state. If set to 0.6, then the IT load interruptibility index I is 0.5. Figure 2 .

[0020] In an optional embodiment, adjusting the preset threshold according to the state of charge of the energy storage unit includes: A base threshold is set. When the state of charge of the energy storage unit is lower than the first preset state of charge threshold, the adjusted preset threshold is reduced by an adjustment value proportional to the degree of state of charge decline. When the state of charge of the energy storage unit is higher than the second preset state of charge threshold, the adjusted preset threshold is increased by an adjustment value proportional to the degree of increase in state of charge.

[0021] Specifically, a base threshold of, for example, 0.6, serves as the baseline for determining whether load shedding needs to be initiated. Simultaneously, two state-of-charge (SOC) boundary points are defined: a first preset SOC threshold, i.e., the low-charge warning line, is set at 30%; and a second preset SOC threshold, i.e., the high-charge safety line, is set at 80%. The system becomes sensitive when the real-time SOC of the energy storage unit falls below the low-charge warning line. For example, when an SOC of 20% is detected, below 30%, the adjusted threshold will be reduced by an adjustment value from the base threshold. This adjustment value is proportional to the magnitude of the SOC falling below the warning line, calculated as a proportionality coefficient multiplied by the difference between the warning line and the real-time SOC. If the proportionality coefficient is 0.5, the adjustment value is 0.05. Therefore, the new threshold is 0.55, lowering the threshold for triggering load shedding.

[0022] Conversely, when the real-time SOC of the energy storage unit exceeds the high-capacity safety threshold, the system becomes conservative to avoid unnecessary IT load interruptions. For example, when the detected SOC is 90%, higher than 80%, the adjusted threshold will be increased by an adjustment value on the base threshold. This adjustment value is calculated similarly, as another scaling factor multiplied by the difference between the real-time SOC and the safety threshold. If this scaling factor is 0.2, the adjustment value is 0.02. Therefore, the new threshold is 0.62, raising the threshold for triggering load shedding. If the SOC is within the normal range of 30% to 80%, the threshold remains unchanged at the base threshold of 0.6. Figure 3 .

[0023] In an optional embodiment, sending a load reduction request to the IT resource management system when the IT load interruptibility level index exceeds the adjusted preset threshold includes: Calculate the difference between the IT load interruptibility level index and the adjusted preset threshold, determine the IT load power target value to be reduced based on the difference and a preset function, and encapsulate the IT load power target value in a load reduction request message and send it.

[0024] Once the calculated IT load interruptibility index I value exceeds the adjusted preset threshold, the reduction process is triggered. For example, if the current index I value is 0.7 and the threshold is 0.55, the difference between the two is calculated to be 0.15. This difference reflects the degree to which the current operational risk exceeds the safety threshold. Using a preset linear mapping function, the dimensionless difference is converted into a specific power value. This function can be in the form that the reduction power is equal to a scaling factor k multiplied by the difference.

[0025] For example, the proportionality coefficient k can be set to 1000 kilowatts based on the total IT load of the data center, meaning that for every 0.1 the index exceeds the threshold, 100 kilowatts of IT load needs to be reduced. Based on this mapping, the current target IT load power reduction is calculated to be 150 kilowatts. After calculating this target value, it is encapsulated into a standard-format load reduction request message, such as a JSON object containing the instruction type, power target value, and timestamp fields. This message is sent to a data center IT resource management system, such as OpenStack or VMware vCenter, via an internal network interface using an API call. The management system then performs specific operations such as shutting down low-priority virtual machines, limiting application performance, or migrating compute tasks. Figure 4 .

[0026] S3. Based on the voltage deviation of the common correlation point, generate a basic voltage regulation command, and combine the state of charge of the energy storage unit and the interruptibility level index of the IT load to calculate the voltage reference fluctuation. Then, superimpose the voltage reference fluctuation with the basic voltage regulation command to form a voltage control command. Real-time voltage at PCC point With rated voltage The voltage deviation ΔV is obtained through comparison. This deviation is then processed by a proportional-integral (PI) controller to generate a basic reactive current command. This is used for basic voltage regulation; a function, such as the hyperbolic tangent function, is planned to calculate the fluctuation of the voltage reference. Among them, A, B, , It is an adjustable parameter. This represents the median state of charge of the energy storage system. The above-calculated interruptibility level index; the float can slightly raise the voltage reference when energy storage is sufficient and stable, and appropriately lower the voltage reference when energy storage is insufficient and there is a risk; this float, Rated voltage reference Add them together to form a new voltage reference. This new benchmark is then used in the PI controller to generate reactive current commands. This instruction is part of the voltage control instruction.

[0027] In an optional embodiment, the generation of basic voltage regulation commands based on the voltage deviation at the common correlation point is implemented through a voltage-reactive power outer loop controller. Specifically, the actual voltage value at the common correlation point is measured in real time. Compare it with the system rated voltage reference value. By comparison, the voltage deviation is calculated. The deviation signal is input to a voltage outer-loop proportional-integral controller, whose transfer function is: The proportionality coefficient With integral coefficient The PI controller is tuned according to system impedance and dynamic response requirements. It outputs a basic reactive current reference value. This refers to the basic voltage regulation command.

[0028] In an optional embodiment, calculating the voltage reference fluctuation by combining the state of charge of the energy storage unit and the IT load interruptibility level index includes: The state of charge of the energy storage unit and the IT load interruptibility index are used as inputs, and the corresponding voltage reference fluctuation is obtained by looking up the corresponding voltage reference through a preset two-dimensional lookup table. The two-dimensional lookup table is pre-configured such that when the state of charge of the energy storage unit decreases from the first state of charge value to the second state of charge value and the IT load interruptibility level index increases from the first index value to the second index value, the absolute value of the corresponding voltage reference fluctuation increases monotonically from the first voltage value to the second voltage value.

[0029] A pre-defined two-dimensional lookup table maps two state variables—the energy storage unit's State of Charge (SOC) and the IT load interruptibility level index (I)—to a voltage reference fluctuation value. The lookup table is structured like a matrix, with row indices corresponding to SOC intervals (e.g., 0% to 20%, 20% to 50%, 50% to 80%, and 80% to 100%) and column indices corresponding to I intervals (e.g., 0 to 0.3, 0.3 to 0.6, and 0.6 to 1). Each cell in the table is pre-filled with a voltage fluctuation value optimized through simulation or operational experience.

[0030] The lookup process is as follows: When the real-time SOC is 30% and the exponent I is 0.7, the SOC value is mapped to its corresponding row range, i.e., 20% to 50%. The exponent I value is mapped to its corresponding column range, i.e., 0.6 to 1. By finding the cell at the intersection of these two indices, the corresponding voltage reference fluctuation can be obtained, for example, -0.04 volts. This lookup table follows the monotonicity principle, that is, as the SOC changes from a high range to a low range, or the exponent I changes from a low range to a high range, the absolute value of the obtained voltage fluctuation will monotonically increase from small to large. For example, when the SOC is in the highest range and I is in the lowest range, the fluctuation may be 0 or a very small positive value, while when the SOC is in the lowest range and I is in the highest range, the fluctuation will be a large negative value, such as -0.05 volts. Bilinear interpolation can also be performed between four adjacent data points to improve accuracy.

[0031] In an optional embodiment, the superposition of the voltage reference float and the basic voltage regulation command to form the voltage control command is achieved through an adder and a limiter. The voltage reference float is obtained by querying a preset two-dimensional lookup table based on the energy storage unit's state of charge (SOC) and the IT load interruptibility level index I. Multiply the floating amount by a preset voltage-reactive power conversion factor. This coefficient is determined based on the system short-circuit capacity and the converter's rated capacity, and is thus converted into an equivalent reactive current compensation value. The compensation value The basic voltage regulation instructions generated in the aforementioned embodiments Algebraic superposition is performed in the adder to obtain the preliminary reactive current command. Finally, The input is fed into a limiter, whose upper and lower limits are dynamically calculated based on the real-time available reactive power capacity of the energy storage converter, ensuring that the generated reactive current command is within acceptable limits. Always within the safe operating range of the converter, This refers to the voltage control command issued to the energy storage converter.

[0032] S4. A frequency adjustment command is generated based on the frequency deviation of the common correlation point, and a control signal is generated and sent to the photovoltaic inverter and the energy storage converter based on the adjusted energy storage unit power scheduling benchmark, the voltage control command and the frequency adjustment command.

[0033] Real-time frequency of PCC points and with the rated frequency For example, at 50Hz, a comparison is made to obtain the frequency deviation Δf; according to the droop control principle, a droop coefficient is used... Calculate the active power adjustment required to respond to the frequency deviation. This is the frequency regulation command; for the energy storage converter, the active power control target is set to... ,in The power reference and reactive power control target calculated above take into account air conditioning feedforward compensation. Then, based on the voltage control command generated above To determine; for photovoltaic inverters, active power, based on maximum power point tracking (MPPT), also undertakes a portion of frequency regulation, while reactive power targets... The active and reactive power are allocated according to the proportion of the rated reactive capacity of the photovoltaic inverter; the calculated reference values ​​for active and reactive power. As control signals, they are sent to the local controllers of the energy storage converter and photovoltaic inverter through the communication network, respectively. The internal current loop and phase-locked loop control circuits execute the signals to drive the IGBT power devices to perform switching actions, thereby realizing the control of the microgrid.

[0034] In an optional embodiment, the voltage control command and the frequency adjustment command generate control signals that are sent to the photovoltaic inverter and the energy storage converter, including: The adjusted energy storage unit power scheduling benchmark is used as the active power benchmark and superimposed with the frequency adjustment command to generate the active power command of the energy storage converter; and the reactive power command of the energy storage converter is generated based on the voltage control command.

[0035] In the active power channel, the energy storage unit power dispatch benchmark, calculated and adjusted by the upper-level energy management strategy, is used as the basic command, for example, set to discharge 100 kW. Simultaneously, a parallel frequency control loop monitors the microgrid frequency in real time. When the frequency falls below the nominal value, this loop generates a positive active power compensation command, for example, 10 kW, based on a preset droop coefficient, to provide frequency support. The active power command of the energy storage converter is the algebraic sum of the two parts, i.e., 110 kW. This command drives the converter to deliver 110 kW of active power to the grid.

[0036] In the reactive power channel, the control objective is to stabilize the microgrid voltage. The input is a voltage control command, which is the sum of the nominal voltage reference and a voltage fluctuation calculated based on risk. For example, a nominal voltage of 400 volts plus a fluctuation of -0.04 volts yields a voltage command of 399.96 volts. The voltage controller inside the energy storage converter typically employs a PI control algorithm, continuously comparing the commanded voltage with the actual measured bus voltage. Based on the voltage error, the controller generates and outputs a reactive power command. If the actual voltage is lower than the commanded value, the controller instructs the converter to output inductive reactive power to raise the voltage, and vice versa, until the voltage error converges to zero.

[0037] In the second embodiment, the present invention also proposes a photovoltaic-energy storage integrated data center microgrid control system, comprising the following modules: The adjustment module is used to obtain the output power of photovoltaic units, the state of charge of energy storage units, the real-time power consumption of data center IT load and air conditioning load, and the voltage and frequency of common interconnection points in the microgrid; based on the preset data center thermodynamic model, ambient temperature and the predicted value of IT load in the next period, it calculates the air conditioning power consumption feedforward compensation amount, and adjusts the power scheduling benchmark of energy storage units according to the air conditioning power consumption feedforward compensation amount. The sending module is used to calculate the IT load interruptibility level index based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit; and to adjust a preset threshold according to the state of charge of the energy storage unit; when the IT load interruptibility level index exceeds the adjusted preset threshold, a load reduction request is sent to the IT resource management system. The superposition module is used to generate a basic voltage regulation command based on the voltage deviation of the common correlation point, and calculate the voltage reference fluctuation by combining the state of charge of the energy storage unit and the IT load interruptibility level index. The voltage reference fluctuation is superimposed with the basic voltage regulation command to form a voltage control command. The generation module is used to generate a frequency adjustment command based on the frequency deviation of the common correlation point, and generate a control signal based on the adjusted energy storage unit power scheduling benchmark, the voltage control command and the frequency adjustment command, and send it to the photovoltaic inverter and the energy storage converter.

[0038] In an optional embodiment, the calculation of the air conditioning power consumption feedforward compensation based on a preset data center thermodynamic model, ambient temperature, and predicted IT load for the next time period includes: The predicted IT load for the next period is input into the data center thermodynamic model to calculate the corresponding total heat generation of the servers. Combined with the current ambient temperature, the total cooling capacity required to maintain the target temperature of the data center is calculated. Based on the measured energy efficiency ratio of the air conditioning system, the required total cooling capacity is converted into the predicted power consumption of the air conditioning system for the next period, and the predicted power consumption is used as the feedforward compensation amount.

[0039] In an optional embodiment, the IT load interruptibility index is calculated based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit, using the following formula:

[0040] in, The deviation factor is obtained by normalizing the absolute value of the deviation between the predicted and actual values ​​of photovoltaic output power. The energy factor is obtained by normalizing the complement of the state of charge of the energy storage unit. and These are the preset weighting coefficients.

[0041] In an optional embodiment, adjusting the preset threshold according to the state of charge of the energy storage unit includes: A base threshold is set. When the state of charge of the energy storage unit is lower than the first preset state of charge threshold, the adjusted preset threshold is reduced by an adjustment value proportional to the degree of state of charge decline. When the state of charge of the energy storage unit is higher than the second preset state of charge threshold, the adjusted preset threshold is increased by an adjustment value proportional to the degree of increase in state of charge.

[0042] In an optional embodiment, sending a load reduction request to the IT resource management system when the IT load interruptibility level index exceeds the adjusted preset threshold includes: Calculate the difference between the IT load interruptibility level index and the adjusted preset threshold, determine the IT load power target value to be reduced based on the difference and the preset linear mapping function, and encapsulate the IT load power target value in a load reduction request message and send it.

[0043] In an optional embodiment, calculating the voltage reference fluctuation by combining the state of charge of the energy storage unit and the IT load interruptibility level index includes: The state of charge of the energy storage unit and the IT load interruptibility index are used as inputs, and the corresponding voltage reference fluctuation is obtained by looking up the corresponding voltage reference through a preset two-dimensional lookup table. The two-dimensional lookup table is pre-configured such that when the state of charge of the energy storage unit decreases from the first state of charge value to the second state of charge value and the IT load interruptibility level index increases from the first index value to the second index value, the absolute value of the corresponding voltage reference fluctuation increases monotonically from the first voltage value to the second voltage value.

[0044] In an optional embodiment, the voltage control command and the frequency adjustment command generate control signals that are sent to the photovoltaic inverter and the energy storage converter, including: The adjusted energy storage unit power scheduling benchmark is used as the active power benchmark and superimposed with the frequency adjustment command to generate the active power command of the energy storage converter; and the reactive power command of the energy storage converter is generated based on the voltage control command.

[0045] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0046] The functional modules shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0047] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0048] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0049] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for controlling a microgrid in a data center computer room that integrates photovoltaic and energy storage, characterized in that, Includes the following steps: The system acquires the output power of photovoltaic units, the state of charge of energy storage units, the real-time power consumption of data center IT load and air conditioning load, and the voltage and frequency of common interconnection points within the microgrid. Based on the preset data center thermodynamic model, ambient temperature, and the predicted IT load for the next period, the system calculates the air conditioning power consumption feedforward compensation amount and adjusts the power scheduling benchmark of the energy storage units according to the air conditioning power consumption feedforward compensation amount. Based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit, the IT load interruptibility level index is calculated using the following formula: in, The deviation factor is obtained by normalizing the absolute value of the deviation between the predicted and actual values ​​of photovoltaic output power. The energy factor is obtained by normalizing the complement of the state of charge of the energy storage unit. and The preset weighting coefficient is used; and the preset threshold is adjusted according to the state of charge of the energy storage unit; when the IT load interruptibility level index exceeds the adjusted preset threshold, a load reduction request is sent to the IT resource management system. Based on the voltage deviation of the common correlation point, a basic voltage regulation command is generated. Combined with the state of charge of the energy storage unit and the interruptibility level index of the IT load, the voltage reference fluctuation is calculated. The voltage reference fluctuation is then superimposed with the basic voltage regulation command to form a voltage control command. Frequency adjustment commands are generated based on the frequency deviation of the common correlation point, and control signals are generated and sent to the photovoltaic inverter and energy storage converter based on the adjusted energy storage unit power scheduling benchmark, the voltage control command and the frequency adjustment command.

2. The method according to claim 1, characterized in that, The calculation of air conditioning power consumption feedforward compensation based on a preset data center thermodynamic model, ambient temperature, and predicted IT load for the next time period includes: The predicted IT load for the next period is input into the data center thermodynamic model to calculate the corresponding total heat generation of the servers. Combined with the current ambient temperature, the total cooling capacity required to maintain the target temperature of the data center is calculated. Based on the measured energy efficiency ratio of the air conditioning system, the required total cooling capacity is converted into the predicted power consumption of the air conditioning system for the next period, and the predicted power consumption is used as the feedforward compensation amount.

3. The method according to claim 1, characterized in that, The step of adjusting the preset threshold according to the state of charge of the energy storage unit includes: A base threshold is set. When the state of charge of the energy storage unit is lower than the first preset state of charge threshold, the adjusted preset threshold is reduced by an adjustment value proportional to the degree of state of charge decline. When the state of charge of the energy storage unit is higher than the second preset state of charge threshold, the adjusted preset threshold is increased by an adjustment value proportional to the degree of increase in state of charge.

4. The method according to claim 1, characterized in that, When the IT load interruptibility level index exceeds the adjusted preset threshold, a load reduction request is sent to the IT resource management system, including: Calculate the difference between the IT load interruptibility level index and the adjusted preset threshold, determine the IT load power target value to be reduced based on the difference and a preset function, and encapsulate the IT load power target value in a load reduction request message and send it.

5. The method according to claim 1, characterized in that, The calculation of voltage reference fluctuation by combining the state of charge of the energy storage unit and the IT load interruptibility level index includes: The state of charge of the energy storage unit and the IT load interruptibility index are used as inputs, and the corresponding voltage reference fluctuation is obtained by looking up the corresponding voltage reference through a preset two-dimensional lookup table. The two-dimensional lookup table is pre-configured such that when the state of charge of the energy storage unit decreases from the first state of charge value to the second state of charge value and the IT load interruptibility level index increases from the first index value to the second index value, the absolute value of the corresponding voltage reference fluctuation increases monotonically from the first voltage value to the second voltage value.

6. The method according to any one of claims 1-5, characterized in that, The voltage control command and the frequency adjustment command generate control signals that are sent to the photovoltaic inverter and the energy storage converter, including: The adjusted energy storage unit power scheduling benchmark is used as the active power benchmark and superimposed with the frequency adjustment command to generate the active power command of the energy storage converter; and the reactive power command of the energy storage converter is generated based on the voltage control command.

7. A photovoltaic-energy storage integrated microgrid control system for data center computer rooms, characterized in that, Includes the following modules: The adjustment module is used to obtain the output power of photovoltaic units, the state of charge of energy storage units, the real-time power consumption of data center IT load and air conditioning load, and the voltage and frequency of common interconnection points in the microgrid; based on the preset data center thermodynamic model, ambient temperature and the predicted value of IT load in the next period, it calculates the air conditioning power consumption feedforward compensation amount, and adjusts the power scheduling benchmark of energy storage units according to the air conditioning power consumption feedforward compensation amount. The transmitting module is used to calculate the IT load interruptibility level index based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit. The calculation of the IT load interruptibility level index based on the deviation between the predicted and actual values ​​of photovoltaic output power and the state of charge of the energy storage unit is performed using the following formula: in, The deviation factor is obtained by normalizing the absolute value of the deviation between the predicted and actual values ​​of photovoltaic output power. The energy factor is obtained by normalizing the complement of the state of charge of the energy storage unit. and The preset weighting coefficient is used; and the preset threshold is adjusted according to the state of charge of the energy storage unit; when the IT load interruptibility level index exceeds the adjusted preset threshold, a load reduction request is sent to the IT resource management system. The superposition module is used to generate a basic voltage regulation command based on the voltage deviation of the common correlation point, and calculate the voltage reference fluctuation by combining the state of charge of the energy storage unit and the IT load interruptibility level index. The voltage reference fluctuation is superimposed with the basic voltage regulation command to form a voltage control command. The generation module is used to generate a frequency adjustment command based on the frequency deviation of the common correlation point, and generate a control signal based on the adjusted energy storage unit power scheduling benchmark, the voltage control command and the frequency adjustment command, and send it to the photovoltaic inverter and the energy storage converter.

8. The system according to claim 7, characterized in that, The calculation of air conditioning power consumption feedforward compensation based on a preset data center thermodynamic model, ambient temperature, and predicted IT load for the next time period includes: The predicted IT load for the next period is input into the data center thermodynamic model to calculate the corresponding total heat generation of the servers. Combined with the current ambient temperature, the total cooling capacity required to maintain the target temperature of the data center is calculated. Based on the measured energy efficiency ratio of the air conditioning system, the required total cooling capacity is converted into the predicted power consumption of the air conditioning system for the next period, and the predicted power consumption is used as the feedforward compensation amount.

Images