Calculation center direct-current voltage multi-source coordination control method considering inertia lifting

Abstract

Aiming at the problems of frequent fluctuation of direct current bus voltage and insufficient system inertia caused by violent load fluctuation of a computing power center and high-proportion new energy access, the invention provides a voltage dynamic stability control strategy combining multi-source coordination and virtual inertia improvement. According to the strategy, firstly, the difference between the output characteristics of an actual direct current motor model and the output characteristics of a DC / DC converter controlled based on a virtual direct current motor (VDCM) is analyzed, and on this basis, a consistency algorithm is utilized to construct a computing center direct current voltage multi-source coordination control strategy considering inertia lifting. Through the control of the DC / DC converter, the virtual inertia self-adaptive distribution and multi-source cooperative response are realized, so that the DC voltage fluctuation and unbalanced voltage of the computing power center are effectively inhibited, and the dynamic stability, power supply reliability and electric energy quality of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of DC voltage control and stabilization technology in computing centers, specifically to a multi-source coordinated control method for DC voltage in computing centers that takes into account inertia enhancement. Background Technology

[0002] With the rapid development of computing infrastructure, high-load operations such as artificial intelligence training, data analysis, and cloud computing place higher demands on power quality and supply stability. Computing centers typically connect to multiple types of power sources, such as photovoltaic, energy storage, grid, and DC bus power supply modules. Due to the high dynamics of the load and frequent power fluctuations, coupled with the significant impact of environmental factors on the output of new energy sources, the DC bus voltage is prone to fluctuations, and the problem of insufficient system inertia is becoming increasingly prominent.

[0003] Existing computing center power supply systems mostly employ traditional droop control or static power distribution methods to achieve voltage and power balance. However, when there are sudden load changes or uneven power distribution among multiple sources, these control methods cannot provide sufficient dynamic support, resulting in decreased system voltage stability and potential control coupling and response delay issues between multiple sources.

[0004] To address the aforementioned issues, control methods combining multi-source coordination and virtual inertia enhancement have been proposed in recent years. This method introduces virtual inertia elements into multi-power DC-DC converters to simulate the dynamic energy regulation characteristics of motors, thereby enhancing the system's inertia support capability. However, existing multi-source coordinated control strategies often fail to adequately consider the inertia distribution and response coordination among power sources, resulting in voltage deviation and power unevenness even under severe load fluctuations. To address this problem, this invention proposes a multi-source coordinated control strategy for the DC voltage of computing centers that considers inertia enhancement. By constructing a multi-source consensus algorithm and introducing an inductor current feedback adjustment mechanism, adaptive inertia distribution and multi-source coordinated support are achieved, thereby improving the dynamic stability and operational reliability of the DC voltage in computing centers. Summary of the Invention

[0005] This invention provides a multi-source coordinated control method for DC voltage in computing centers that takes inertia enhancement into account, addressing issues such as insufficient multi-source coordination, limited inertia support capability, and poor voltage stability in existing DC power supply systems for computing centers. By establishing a multi-source DC power supply system model and combining a consensus algorithm with a virtual inertia adjustment mechanism, collaborative operation and adaptive inertia allocation among multiple power sources are achieved, thereby enhancing the system's dynamic response capability and steady-state support performance to voltage disturbances.

[0006] According to a first aspect, one embodiment provides a multi-source coordinated control method for DC voltage in a computing center that takes inertia enhancement into account, the method comprising: Construct a topology model of the power supply system for the computing center, and determine all nodes on the positive and negative DC buses used to connect loads or power sources; When a DC bus voltage fluctuation is detected at a certain node, the voltage fluctuation is suppressed by using the established VDCM-based computing center DC voltage multi-source coordinated control strategy. The VDCM-based DC voltage multi-source coordinated control strategy for computing centers includes a multi-source voltage consistency control part, a voltage closed-loop feedback control part, a VDCM part, and a current closed-loop feedback control part. The multi-source voltage consistency control section is used to output the voltage correction amount of the current node based on the voltage consistency control algorithm of each node on the positive or negative DC bus. The voltage closed-loop feedback control section is used to calculate the virtual electromagnetic torque and output the virtual speed reference value based on the detected DC bus voltage of the current node, the voltage correction amount of the current node, and the rated voltage, combined with PI control. The VDCM section is used to simulate the characteristics of an actual DC motor, calculate the virtual electromotive force based on the input virtual speed reference value, and output the current loop reference value. The current closed-loop feedback control section is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in combination with PI control, so as to achieve stable adjustment of the final DC bus voltage.

[0007] Furthermore, the multi-source voltage consistency control section is used to output the current node voltage correction amount based on the voltage consistency control algorithm of each node on the positive or negative DC bus, specifically including: For node i, the formula for calculating its discrete-time uniform average voltage is:

[0008] In the formula, Let be the DC bus voltage of node i at the k-th sampling time. and Let be the average voltages of nodes i and j at the k-th sampling time, respectively. Let be the average voltage of node i at the (k+1)th sampling time; is the communication weighting coefficient between node i and node j, and N is the number of nodes other than node i on the positive or negative bus. To suppress voltage deviations at various nodes on the DC bus, voltage consistency control utilizes the error between the average voltage and the reference voltage for adjustment, and its output is expressed as:

[0009] in, Let be the voltage correction amount for node i. It is the voltage correction amount at node i. and These are the proportional and integral coefficients for voltage consistency PI control, respectively, and s is the Laplace operator. This is the reference voltage.

[0010] Furthermore, the voltage closed-loop feedback control section is used to calculate the virtual electromagnetic torque based on the detected DC bus voltage of the current node, the current node voltage correction amount, and the rated voltage, combined with PI control, and output a virtual speed reference value, specifically including: First, calculate the virtual electromagnetic torque of the VDCM mechanical part. :

[0011] in , and These are the rated voltage, the voltage correction at node i, and the DC bus voltage at node i, respectively. and These are the proportional and integral coefficients of the voltage loop PI control, respectively, and s is the Laplace operator; Based on the obtained virtual electromagnetic torque The virtual speed reference value is calculated based on the virtual motor torque-speed equation. .

[0012] Furthermore, the VDCM section is used to simulate the characteristics of an actual DC motor, calculates the virtual electromotive force based on the input virtual speed reference value, and outputs a current loop reference value, specifically including: Virtual electromotive force The calculation is as follows:

[0013] in, The torque coefficient, For magnetic flux, This is a virtual speed reference value. This refers to the rotor's rated speed. Based on the calculated virtual electromotive force The reference value of the current loop is obtained. :

[0014] in, This is a virtual impedance.

[0015] Furthermore, the current closed-loop feedback control section is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in conjunction with PI control, so as to achieve stable adjustment of the final DC bus voltage, specifically including:

[0016] in, It is the duty cycle of the DC / DC converter. and These are the current loop reference value and the inductor current, respectively. and These are the proportional and integral coefficients for the current loop PI control, respectively.

[0017] According to a second aspect, one embodiment provides a multi-source coordinated control system for DC voltage in a computing center that takes into account inertia enhancement, the system comprising: The multi-source voltage consistency control module is used to output the voltage correction amount of the current node based on the voltage consistency control algorithm of each node on the positive or negative DC bus. The voltage closed-loop feedback control module is used to calculate the virtual electromagnetic torque and output the virtual speed reference value based on the detected DC bus voltage of the current node, the voltage correction amount of the current node, and the rated voltage, combined with PI control. The VDCM module is used to simulate the characteristics of an actual DC motor, calculate the virtual electromotive force based on the input virtual speed reference value, and output the current loop reference value. The current closed-loop feedback control module is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in combination with PI control, so as to achieve stable adjustment of the final DC bus voltage.

[0018] Furthermore, the multi-source voltage consistency control module is specifically used for: For node i, the formula for calculating its discrete-time uniform average voltage is:

[0019] In the formula, Let be the DC bus voltage of node i at the k-th sampling time. and Let be the average voltages of nodes i and j at the k-th sampling time, respectively. Let be the average voltage of node i at the (k+1)th sampling time; is the communication weighting coefficient between node i and node j, and N is the number of nodes other than node i on the positive or negative bus. To suppress voltage deviations at various nodes on the DC bus, voltage consistency control utilizes the error between the average voltage and the reference voltage for adjustment, and its output is expressed as:

[0020] in, Let be the voltage correction amount for node i. It is the voltage correction amount at node i. and These are the proportional and integral coefficients for voltage consistency PI control, respectively, and s is the Laplace operator. This is the reference voltage.

[0021] Furthermore, the voltage closed-loop feedback control module is specifically used for: First, calculate the virtual electromagnetic torque of the VDCM mechanical part. :

[0022] in , and These are the rated voltage, the voltage correction at node i, and the DC bus voltage at node i, respectively. and These are the proportional and integral coefficients of the voltage loop PI control, respectively, and s is the Laplace operator; Based on the obtained virtual electromagnetic torque The virtual speed reference value is calculated based on the virtual motor torque-speed equation. .

[0023] Furthermore, the VDCM module is specifically used for: Virtual electromotive force The calculation is as follows:

[0024] in, The torque coefficient, For magnetic flux, This is a virtual speed reference value. This refers to the rotor's rated speed. Based on the calculated virtual electromotive force The reference value of the current loop is obtained. :

[0025] in, This is a virtual impedance.

[0026] Furthermore, the current closed-loop feedback control module is specifically used for:

[0027] in, It is the duty cycle of the DC / DC converter. and These are the current loop reference value and the inductor current, respectively. and These are the proportional and integral coefficients for the current loop PI control, respectively.

[0028] This invention provides a multi-source coordinated control method for DC voltage in computing centers that takes inertia enhancement into account, which has the following beneficial effects: 1. This invention proposes a multi-source coordinated control strategy for DC voltage in computing centers that considers inertia enhancement. By introducing a multi-source consensus algorithm and a virtual inertia collaborative control mechanism, it achieves inertia sharing and voltage coordinated adjustment among distributed power sources. This method can quickly provide dynamic support when the computing center load fluctuates drastically, effectively suppressing bus voltage deviation and oscillation, and improving the system's voltage stability and dynamic response performance.

[0029] 2. The multi-source adaptive inertia adjustment model established in this invention enables the inertia allocation of each power source to be optimized in real time according to power changes and topology status, thereby ensuring system stability while also taking into account rapid response and energy balance. Compared with the traditional single-source fixed inertia control method, this invention can achieve adaptive inertia enhancement and multi-source collaborative support under different operating conditions, significantly improving the power supply reliability of the DC power supply system of the computing center. Attached Figure Description

[0030] Figure 1 A topology diagram of the power supply system of a computing center in a multi-source coordinated control method for DC voltage of a computing center that takes inertia enhancement into account, provided in an embodiment of the present invention; Figure 2 This is a control block diagram based on VDCM in a multi-source coordinated control method for DC voltage in a computing center that takes inertia enhancement into account, provided in one embodiment of the present invention. Figure 3 This invention provides a model of a separately excited DC motor system in a multi-source coordinated control method for DC voltage in a computing center that takes inertia enhancement into account, as an embodiment of the present invention. Figure 4 A flowchart of a multi-source coordinated control method for DC voltage in a computing center that takes inertia enhancement into account, provided as an embodiment of the present invention; Figure 5 A control block diagram of a multi-source coordinated control method for DC voltage in a computing center that takes inertia enhancement into account, provided as an embodiment of the present invention; Figure 6 This is a block diagram of the consensus algorithm control in a multi-source coordinated control method for DC voltage of a computing center that takes inertia enhancement into account, provided as an embodiment of the present invention. Detailed Implementation

[0031] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0032] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0033] The first embodiment of this invention provides a multi-source coordinated control method for DC voltage in a computing center that takes into account inertia enhancement. This will be described in detail below.

[0034] (1) Construct a topology model of the power supply system of the computing center to provide a theoretical basis for subsequent virtual inertia modeling and voltage support strategy design.

[0035] First, a topology model of the computing center power supply system is constructed, including key components such as photovoltaic systems, energy storage, DC buses, computing loads, AC / DC converters, and DC / DC converters. Figure 1 As shown in the diagram, the AC / DC converter receives three-phase AC input and outputs bipolar DC, thus connecting to a bipolar DC bus. The energy storage unit, photovoltaic unit, and computing center are all unipolar structures, connected to either the positive or negative DC bus. By establishing this topology, the power flow and control logic between nodes can be clearly described, providing a theoretical basis for subsequent virtual inertia modeling and voltage support strategy design.

[0036] (2) Compare the output characteristics of the actual DC motor model and the DC converter based on virtual DC motor (VDCM) control, and analyze the inertial response law in the energy absorption and release process.

[0037] To utilize the virtual inertia of the VDCM-controlled DC / DC converter in the computing center to simulate the rotational inertia of an actual DC motor model, the control strategy for the DC / DC converter connected to the energy storage unit adopts the VDCM control strategy. The control block diagram is as follows: Figure 2 As shown, this strategy is based on the principle that a DC motor provides inertia in a DC system. Figure 2 middle, R a For virtual impedance, J and D These are virtual inertia and damping, respectively; ω and ω r These represent the virtual speed and rated virtual speed of the VDCM control strategy, respectively. E a This is a virtual electromotive force. The VDCM control strategy suppresses voltage fluctuations by controlling the output characteristics of the DC / DC converter to mimic the output characteristics of a DC motor absorbing or releasing stored energy.

[0038] In this control strategy, the torque equation of the VDCM is as follows: (1) (2) In the formula, T e Electromagnetic torque, and armature current i a And excitation flux ψ f Proportional, that is T e = K t ψ f i a ,in K t The torque constant; T m For mechanical torque; J and D These are the coefficient of moment of inertia and the damping coefficient, respectively. D It characterizes the frictional loss of the rotor during rotation; ω This is the actual rotor speed. ω r This refers to the rotor's rated speed. v o and V These represent the converter's output voltage and rated output voltage, respectively. P e To output electromagnetic power, P mThis refers to the input power.

[0039] Figure 3 This is an equivalent circuit model of a typical separately excited DC motor. In this model, the excitation circuit and the armature circuit constitute independent control systems. In the excitation circuit, v f , i f and R f These represent the excitation voltage, excitation current, and excitation resistance, respectively; in the armature circuit, i a , R a and L a These represent armature current, armature resistance, and armature inductance, respectively. E a It is the back electromotive force, the value of which is related to the motor speed. ω and excitation flux ψ f Proportional, that is E a = K e ψ f ω, where K e The back electromotive force coefficient; v a This is the terminal voltage applied to the armature circuit.

[0040] For a DC generator, its mechanical system torque-speed equation (rotor dynamic equation) can be expressed as: (3) In the electrical section, the electromotive force balance equation for the armature circuit of a DC motor is: (4) in, C T This is the torque coefficient, which is related to the machine's structural parameters; φ m Magnetic flux, determined by the excitation current i f The equations determine the relationship between voltage, current, and rotational speed, and these factors are proportional within the linear region. From an electrical perspective, this set of equations describes the dynamic characteristics of the armature circuit, reflecting the interrelationship between voltage, current, and rotational speed.

[0041] (3) Construct a multi-source consistency algorithm for the computing center system to realize voltage coordination control among distributed power sources and ensure the dynamic consistency and stable operation of the multi-source system.

[0042] Consensus algorithms are mainly used to solve the consensus problem in multi-agent systems composed of continuous-time integrators. Their mathematical model can be expressed as: (5) in, x i ( t ) indicates the first i An intelligent agent at time t The state variables are used to characterize their coordinateable physical quantities, such as voltage or power. u i ( t ) is the corresponding control input signal, used to adjust the rate of change of the agent's state, so as to achieve dynamic control and consistent convergence of the target variable.

[0043] For the above system, the following linear consistency control protocol is adopted: (6) in, N i Indicates the connection between the DC system of the computing center and the first i A set of neighboring nodes that a distributed power node directly communicates with; a ij For nodes i With nodes j The topological connection weighting coefficients between the two nodes reflect the coupling relationship between the voltage or power information of the two nodes. x i and x j They represent the first i The and the first j The state variables of each node can correspond to the node's DC voltage or power output; x i x j ) represents the state deviation between nodes.

[0044] A multi-source system can constitute a multi-agent system, and its communication topology can be described as a weighted undirected connected graph. G =( V , E , W ),in V ={1, 2, , n} is the set of vertices. n The total number of vertices. E V × V Let be the set of edges.W =[ w i,j ]∈ R n ×n This is a weighted adjacency matrix. If the graph... G A graph is a collection of graphs in which any two nodes are connected by a path along an edge. G It is a strong connection. W It is a symmetric matrix. When node i can obtain information from node j, then... W Matrix elements w i,j >0, otherwise w i,j =0. w i,j The size of this value represents the communication weight of the communication link. Further, the Laplace matrix of the graph is obtained. L n = [ l ij ]satisfy: (7) Based on the above analysis, a multi-source coordinated control strategy for DC voltage in computing centers that considers inertia enhancement is proposed. Building upon the consensus algorithm, this strategy effectively suppresses bus voltage fluctuations and improves system inertia and voltage stability through virtual inertia adaptive adjustment and multi-source collaborative response. The flowchart is shown below. Figure 4 As shown, the control block diagram is as follows: Figure 5 As shown.

[0045] The control strategy mainly consists of four parts: multi-source voltage consistency control, voltage closed-loop feedback control, VDCM, and current closed-loop feedback control.

[0046] (1) The multi-source voltage consistency control section is used for voltage consistency control algorithms based on nodes on the positive or negative DC bus, and outputs the voltage correction amount of the current node.

[0047] Under directed graph network topology, average consistency can also be achieved using a linear consistency control protocol, thereby ensuring the voltage and power coordination control effect of the multi-source system in the computing center. The control block diagram of the voltage consistency algorithm is as follows: Figure 6 As shown. Specifically: For node i, the formula for calculating its discrete-time uniform average voltage is: (8) In the formula, Let be the DC bus voltage of node i at the k-th sampling time. and Let be the average voltages of nodes i and j at the k-th sampling time, respectively. Let be the average voltage of node i at the (k+1)th sampling time; is the communication weighting coefficient between node i and node j, and N is the number of nodes other than node i on the positive or negative bus. To suppress voltage deviations at various nodes on the DC bus, voltage consistency control utilizes the error between the average voltage and the reference voltage for adjustment, and its output is expressed as: (9) in, Let be the voltage correction amount for node i. It is the voltage correction amount at node i. and These are the proportional and integral coefficients for voltage consistency PI control, respectively, and s is the Laplace operator. This is the reference voltage.

[0048] The voltage consistency algorithm is shown in formulas (8) and (9). The nodes on the positive and negative DC buses are calculated according to formulas (8) and (9), respectively. When node i is a node on the positive DC bus, node j only needs to select the other nodes on the positive DC bus except for node i. Similarly, when node i is a node on the negative DC bus, node j only needs to select the other nodes on the negative DC bus except for node i. Thus, the positive and negative voltage correction amounts are obtained respectively. and .

[0049] (2) The voltage closed-loop feedback control section is used to calculate the virtual electromagnetic torque based on the detected DC bus voltage of the current node, the voltage correction amount of the current node, and the rated voltage, combined with PI control, and output the virtual speed reference value, as follows: First, calculate the virtual electromagnetic torque of the VDCM mechanical part. : (10) in , and These are the rated voltage, the voltage correction at node i, and the DC bus voltage at node i, respectively. and These are the proportional and integral coefficients of the voltage loop PI control, respectively, and s is the Laplace operator; Based on the obtained virtual electromagnetic torque The virtual speed reference value is calculated based on the virtual motor torque-speed equation (refer to formula (3)). In formula (3) ωThe actual rotor speed of the real DC motor is used, but it is replaced with a virtual speed in this calculation. The virtual motor torque-speed equation can then be obtained.

[0050] (3) The VDCM section is used to simulate the characteristics of an actual DC motor. It calculates the virtual electromotive force based on the input virtual speed reference value and outputs the current loop reference value, as follows: Virtual electromotive force The calculation is as follows: (11) in, The torque coefficient, For magnetic flux, This is a virtual speed reference value. This refers to the rotor's rated speed. Based on the calculated virtual electromotive force The reference value of the current loop is obtained. : (12) in, This is a virtual impedance.

[0051] (4) The current closed-loop feedback control section is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in combination with PI control, so as to achieve stable adjustment of the final DC bus voltage, as follows: (13) Where d is the duty cycle of the DC / DC converter. and These are the current loop reference value and the inductor current, respectively. and These are the proportional and integral coefficients for the current loop PI control, respectively.

[0052] From a control principle perspective, when the load of the computing center changes abruptly or operating conditions change, the current and voltage of the multi-source system will experience disturbances. By adaptively adjusting the virtual inertia, the inertia support of each power source output can dynamically track power fluctuations, thereby balancing system energy exchange and suppressing bus voltage deviation. This method can significantly improve the dynamic response speed and steady-state voltage accuracy of the computing center's DC system, ensuring the stability and reliability of multi-source collaborative operation.

[0053] Corresponding to the aforementioned disclosed method for multi-source coordinated control of DC voltage in computing centers considering inertia enhancement, this invention also discloses a multi-source coordinated control system for DC voltage in computing centers considering inertia enhancement, which specifically includes: The multi-source voltage consistency control module is used to output the voltage correction amount of the current node based on the voltage consistency control algorithm of each node on the positive or negative DC bus. The voltage closed-loop feedback control module is used to calculate the virtual electromagnetic torque and output the virtual speed reference value based on the detected DC bus voltage of the current node, the voltage correction amount of the current node, and the rated voltage, combined with PI control. The VDCM module is used to simulate the characteristics of an actual DC motor, calculate the virtual electromotive force based on the input virtual speed reference value, and output the current loop reference value. The current closed-loop feedback control module is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in combination with PI control, so as to achieve stable adjustment of the final DC bus voltage.

[0054] Furthermore, the multi-source voltage consistency control module is specifically used for: For node i, the formula for calculating its discrete-time uniform average voltage is:

[0055] In the formula, Let be the DC bus voltage of node i at the k-th sampling time. and Let be the average voltages of nodes i and j at the k-th sampling time, respectively. Let be the average voltage of node i at the (k+1)th sampling time; is the communication weighting coefficient between node i and node j, and N is the number of nodes other than node i on the positive or negative bus. To suppress voltage deviations at various nodes on the DC bus, voltage consistency control utilizes the error between the average voltage and the reference voltage for adjustment, and its output is expressed as:

[0056] in, Let be the voltage correction amount for node i. It is the voltage correction amount at node i. and These are the proportional and integral coefficients for voltage consistency PI control, respectively, and s is the Laplace operator. This is the reference voltage.

[0057] Furthermore, the voltage closed-loop feedback control module is specifically used for: First, calculate the virtual electromagnetic torque of the VDCM mechanical part. :

[0058] in , and These are the rated voltage, the voltage correction at node i, and the DC bus voltage at node i, respectively. and These are the proportional and integral coefficients of the voltage loop PI control, respectively, and s is the Laplace operator; Based on the obtained virtual electromagnetic torque The virtual speed reference value is calculated based on the virtual motor torque-speed equation. .

[0059] Furthermore, the VDCM module is specifically used for: Virtual electromotive force The calculation is as follows:

[0060] in, The torque coefficient, For magnetic flux, This is a virtual speed reference value. This refers to the rotor's rated speed. Based on the calculated virtual electromotive force The reference value of the current loop is obtained. :

[0061] in, This is a virtual impedance.

[0062] Furthermore, the current closed-loop feedback control module is specifically used for:

[0063] in, It is the duty cycle of the DC / DC converter. and These are the current loop reference value and the inductor current, respectively. and These are the proportional and integral coefficients for the current loop PI control, respectively.

[0064] It should be noted that for a detailed description of the multi-source coordinated control system for DC voltage of computing centers that takes inertia enhancement into account, provided in the embodiments of the present invention, please refer to the relevant description of the multi-source coordinated control method for DC voltage of computing centers that takes inertia enhancement into account, provided in the embodiments of the present invention, which will not be repeated here.

[0065] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A multi-source coordinated control method for DC voltage in a computing center, taking into account inertia enhancement, characterized in that, The method includes: Construct a topology model of the power supply system for the computing center, and determine all nodes on the positive and negative DC buses used to connect loads or power sources; When a DC bus voltage fluctuation is detected at a certain node, the voltage fluctuation is suppressed by using the established VDCM-based computing center DC voltage multi-source coordinated control strategy. The VDCM-based DC voltage multi-source coordinated control strategy for computing centers includes a multi-source voltage consistency control part, a voltage closed-loop feedback control part, a VDCM part, and a current closed-loop feedback control part. The multi-source voltage consistency control section is used to output the voltage correction amount of the current node based on the voltage consistency control algorithm of each node on the positive or negative DC bus. The voltage closed-loop feedback control section is used to calculate the virtual electromagnetic torque and output the virtual speed reference value based on the detected DC bus voltage of the current node, the voltage correction amount of the current node, and the rated voltage, combined with PI control. The VDCM section is used to simulate the characteristics of an actual DC motor, calculate the virtual electromotive force based on the input virtual speed reference value, and output the current loop reference value. The current closed-loop feedback control section is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in combination with PI control, so as to achieve stable adjustment of the final DC bus voltage.

2. The multi-source coordinated control method for DC voltage in a computing center considering inertia enhancement as described in claim 1, characterized in that, The multi-source voltage consistency control section is used to output the current node voltage correction amount based on the voltage consistency control algorithm of each node on the positive or negative DC bus, specifically including: For node i, the formula for calculating its discrete-time uniform average voltage is: In the formula, Let be the DC bus voltage of node i at the k-th sampling time. and Let be the average voltages of nodes i and j at the k-th sampling time, respectively. Let be the average voltage of node i at the (k+1)th sampling time; is the communication weighting coefficient between node i and node j, and N is the number of nodes other than node i on the positive or negative bus. To suppress voltage deviations at various nodes on the DC bus, voltage consistency control utilizes the error between the average voltage and the reference voltage for adjustment, and its output is expressed as: in, Let be the voltage correction amount for node i. It is the voltage correction amount at node i. and These are the proportional and integral coefficients for voltage consistency PI control, respectively, and s is the Laplace operator. This is the reference voltage.

3. The multi-source coordinated control method for DC voltage in a computing center considering inertia enhancement as described in claim 1, characterized in that, The voltage closed-loop feedback control section is used to calculate the virtual electromagnetic torque and output a virtual speed reference value based on the detected DC bus voltage of the current node, the current node voltage correction amount, and the rated voltage, combined with PI control. Specifically, it includes: First, calculate the virtual electromagnetic torque of the VDCM mechanical part. : in , and These are the rated voltage, the voltage correction at node i, and the DC bus voltage at node i, respectively. and These are the proportional and integral coefficients of the voltage loop PI control, respectively, and s is the Laplace operator; Based on the obtained virtual electromagnetic torque The virtual speed reference value is calculated based on the virtual motor torque-speed equation. .

4. The multi-source coordinated control method for DC voltage in a computing center considering inertia enhancement as described in claim 3, characterized in that, The VDCM section is used to simulate the characteristics of an actual DC motor. It calculates the virtual electromotive force based on the input virtual speed reference value and outputs a current loop reference value, specifically including: Virtual electromotive force The calculation is as follows: in, The torque coefficient, For magnetic flux, This is a virtual speed reference value. This refers to the rotor's rated speed. Based on the calculated virtual electromotive force The reference value of the current loop is obtained. : in, This is a virtual impedance.

5. The multi-source coordinated control method for DC voltage in a computing center considering inertia enhancement as described in claim 4, characterized in that, The current closed-loop feedback control section is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in conjunction with PI control, so as to achieve stable adjustment of the final DC bus voltage. Specifically, it includes: in, It is the duty cycle of the DC / DC converter. and These are the current loop reference value and the inductor current, respectively. and These are the proportional and integral coefficients for the current loop PI control, respectively.

6. A multi-source coordinated control system for DC voltage in a computing center, taking into account inertia enhancement, characterized in that, The system includes: The multi-source voltage consistency control module is used to output the voltage correction amount of the current node based on the voltage consistency control algorithm of each node on the positive or negative DC bus. The voltage closed-loop feedback control module is used to calculate the virtual electromagnetic torque and output the virtual speed reference value based on the detected DC bus voltage of the current node, the voltage correction amount of the current node, and the rated voltage, combined with PI control. The VDCM module is used to simulate the characteristics of an actual DC motor, calculate the virtual electromotive force based on the input virtual speed reference value, and output the current loop reference value. The current closed-loop feedback control module is used to output the duty cycle of the DC / DC converter based on the input current loop reference value and in combination with PI control, so as to achieve stable adjustment of the final DC bus voltage.

7. A multi-source coordinated control system for DC voltage in a computing center, taking into account inertia enhancement, as described in claim 6, characterized in that, The multi-source voltage consistency control module is specifically used for: For node i, the formula for calculating its discrete-time uniform average voltage is: In the formula, Let be the DC bus voltage of node i at the k-th sampling time. and Let be the average voltages of nodes i and j at the k-th sampling time, respectively. Let be the average voltage of node i at the (k+1)th sampling time; is the communication weighting coefficient between node i and node j, and N is the number of nodes other than node i on the positive or negative bus. To suppress voltage deviations at various nodes on the DC bus, voltage consistency control utilizes the error between the average voltage and the reference voltage for adjustment, and its output is expressed as: in, Let be the voltage correction amount for node i. It is the voltage correction amount at node i. and These are the proportional and integral coefficients for voltage consistency PI control, respectively, and s is the Laplace operator. This is the reference voltage.

8. The multi-source coordinated control system for DC voltage in a computing center, taking into account inertia enhancement, as described in claim 7, is characterized in that... The voltage closed-loop feedback control module is specifically used for: First, calculate the virtual electromagnetic torque of the VDCM mechanical part. : in , and These are the rated voltage, the voltage correction at node i, and the DC bus voltage at node i, respectively. and These are the proportional and integral coefficients of the voltage loop PI control, respectively, and s is the Laplace operator; Based on the obtained virtual electromagnetic torque The virtual speed reference value is calculated based on the virtual motor torque-speed equation. .

9. A multi-source coordinated control system for DC voltage in a computing center, taking into account inertia enhancement, as described in claim 8, characterized in that, The VDCM module is specifically used for: Virtual electromotive force The calculation is as follows: in, The torque coefficient, For magnetic flux, This is a virtual speed reference value. This refers to the rotor's rated speed. Based on the calculated virtual electromotive force The reference value of the current loop is obtained. : in, This is a virtual impedance.

10. A multi-source coordinated control system for DC voltage in a computing center, taking into account inertia enhancement, as described in claim 9, characterized in that, The current closed-loop feedback control module is specifically used for: in, It is the duty cycle of the DC / DC converter. and These are the current loop reference value and the inductor current, respectively. and These are the proportional and integral coefficients for the current loop PI control, respectively.

Images