Regional power grid new energy consumption method and system based on computing power-power space-time coordination

Abstract

The invention discloses a regional power grid new energy consumption method and system based on computing power-power space-time coordination. The method comprises the following steps: acquiring real-time operation state data of a regional power grid and to-be-processed computing power task data of a data center cluster; dividing the computing power task into flexible loads based on the time sensitivity of the task, and determining the energy consumption conversion relation of the flexible loads; on the basis of the real-time operation state and the energy consumption conversion relation, the cross-regional migration amount of the flexible load and the output state of the generator set serve as decision variables, and a cooperative scheduling instruction containing a computing power task cross-regional migration path and generator set power adjustment is generated; and in response to the instruction, the flexible load in the load center area is migrated to the data center for execution in a cross-area manner through the communication network, and the output of the generator set in the corresponding area is synchronously adjusted. The method is used for solving the problems that in the prior art, due to the fact that a physical power transmission channel is limited, new energy power abandoning is serious, and power loads and new energy resources are mismatched in space distribution.

Description

Technical Field

[0001] This invention relates to the field of power system operation and control technology, and more specifically, to a method and system for regional power grid renewable energy consumption based on computing power-power spatiotemporal coordination. Background Technology

[0002] Existing regional power grid renewable energy consumption technologies primarily rely on the physical grid's transmission capacity and the power source's regulation capabilities. Specifically, common technical solutions include: first, constructing ultra-high voltage (UHV) transmission channels to physically expand the capacity of energy-rich areas and transfer surplus power to load centers; second, configuring large-scale energy storage power stations (such as pumped-storage and electrochemical storage) or modifying thermal power units to improve flexibility, smoothing out random fluctuations in renewable energy output through "time shifting"; and third, implementing traditional demand-side response, guiding local users to adapt to grid operation by peak shaving and valley filling or load interruption. On the data center side, existing technologies mainly focus on reducing Power Usage Effectiveness (PUE), such as using liquid cooling technology or waste heat recovery, or conducting static site selection optimization during the planning phase.

[0003] However, the aforementioned existing technological solutions have significant limitations in addressing large-scale renewable energy consumption. First, physical transmission channels have long construction cycles, high investment costs, and physical constraints such as thermal stability limits. When congestion occurs, even if there is load demand at the receiving end, the sending end must curtail wind and solar power, leading to the physical bottleneck problem of "electricity not being able to be transmitted." Second, the configuration cost of energy storage resources is high, making it difficult to fully absorb large-scale surplus power. Third, traditional load regulation is mainly based on the logic of "local consumption" or "time transfer," lacking spatial flexibility. Existing data center scheduling systems are disconnected from grid scheduling systems, typically treating data centers as immovable rigid loads and failing to utilize the low-latency transmission characteristics of computing tasks across networks. This passive "electricity follows computing" model results in a serious waste of regulation resources and cannot effectively solve the renewable energy consumption bottleneck caused by geographical mismatch. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination. This method identifies flexible computing loads based on time sensitivity and uses communication networks to migrate them from load centers across regions to data centers for execution. Simultaneously, it coordinates and adjusts the output of generator units in the corresponding regions to solve the problem of renewable energy consumption difficulties caused by limited physical transmission channels and spatial mismatch between energy supply and demand.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A regional power grid renewable energy consumption method based on spatiotemporal coordination of computing power and power includes the following steps: acquiring real-time operating status data of the regional power grid and pending computing power task data of the data center cluster; based on the computing power task data, dividing the computing power tasks into flexible loads according to the time sensitivity of the tasks, and determining the energy consumption conversion relationship required for the execution of the computing power tasks; based on the real-time operating status data and the energy consumption conversion relationship, using the cross-regional migration amount of flexible loads and the output status of generator units as collaborative decision variables, generating a collaborative scheduling instruction that includes the cross-regional migration path of computing power tasks and the power adjustment of generator units; responding to the collaborative scheduling instruction, migrating the flexible loads located in the load center area across regions to the data center for execution through the communication network, and synchronously adjusting the output of the corresponding regional generator units.

[0006] In a preferred embodiment, determining the energy consumption conversion relationship required for executing the computing task includes: determining the basic idle power consumption of the data center under no-load conditions; determining the dynamic incremental power consumption required to process a unit of data volume according to the computing type of the computing task; and multiplying the basic idle power consumption and the dynamic incremental power consumption by the computing task load to obtain the real-time power load value required for the data center to absorb the computing task.

[0007] In a preferred embodiment, generating a collaborative scheduling instruction that includes cross-regional migration paths for computing tasks and adjustments to generator power includes: constructing a collaborative scheduling model aimed at minimizing overall operating costs and maximizing renewable energy consumption; the collaborative scheduling model includes: calculating the total cost of computing power and power dispatch, the total cost including the operating fuel cost of conventional generators, the cost of wind and solar curtailment penalties due to the inability to consume renewable energy, and the network bandwidth rental cost and latency penalty cost incurred from transmitting computing tasks across regions; and using a multivariate constraint optimization algorithm to solve the collaborative scheduling model within a solution space that satisfies power system operating constraints and computing network constraints, to obtain the computing task allocation scheme and generator power output scheme that minimizes the total cost.

[0008] In a preferred embodiment, the power system operation constraints are used to ensure the safe operation of the power network within physical transmission limits, including node power balance constraints, line transmission capacity constraints, and unit operating status constraints; the computing power network constraints are used to ensure that the amount of tasks migrated into the data center does not exceed the upper limit of the data center's outbound network bandwidth, and that the sum of the total processing time of the tasks and the network transmission time does not exceed the maximum allowable delay time of the tasks.

[0009] In a preferred embodiment, classifying computing tasks into flexible loads based on their time sensitivity includes: obtaining the deadline requirements and estimated processing time of the computing tasks; calculating the latency tolerance of the tasks, wherein the latency tolerance characterizes the ratio of the longest allowed waiting time to the estimated processing time; and marking tasks with latency tolerance higher than a preset threshold as flexible loads that can be migrated across regions.

[0010] In a preferred embodiment, generating a collaborative scheduling instruction that includes cross-regional migration paths for computing tasks and generator power adjustments includes: when it is detected that there is renewable energy curtailment in the first region and flexible load demand in the second region, comparing the remaining capacity of the physical transmission channel with the available bandwidth of the computing power transmission channel; if the remaining capacity of the physical transmission channel is insufficient to transmit and absorb the power required by the renewable energy, then a routing instruction is preferentially generated to send the flexible load data of the second region to the data center of the first region for calculation via the optical fiber network.

[0011] In a preferred embodiment, the regional power grid renewable energy consumption method is implemented in two phases: day-ahead planning and intraday real-time. The day-ahead planning phase involves formulating start-up and shutdown plans for conventional generating units based on the next day's weather forecast and historical task data, and reserving redundant computing resources for the data center. In the intraday real-time phase, based on ultra-short-term renewable energy output forecasts, the coordinated scheduling instructions are updated on a rolling basis at preset time intervals, and the migration amount of flexible loads is dynamically adjusted according to the instantaneous deviation of renewable energy output.

[0012] In a preferred embodiment, the step of migrating flexible loads located in the load center area to the data center across regions via a communication network includes: detecting the congestion status and transmission delay of the inter-regional communication network before performing the cross-regional migration of computing tasks; during the task migration process, if the transmission delay of the communication network is detected to exceed a safety threshold, a rollback mechanism is triggered to stop the cross-regional migration of subsequent tasks, switch the unmigrated tasks back to the local data center for processing, and simultaneously trigger the backup regulation resources of the local power grid.

[0013] This invention provides a regional power grid renewable energy consumption system based on spatiotemporal coordination of computing power and power, comprising: a data acquisition module for acquiring real-time operating status data of the regional power grid and pending computing power task data of the data center cluster; a task analysis module for dividing computing power tasks into flexible loads based on the computing power task data and the time sensitivity of the tasks, and determining the energy consumption conversion relationship required for the execution of the computing power tasks; a strategy decision module for generating a coordinated scheduling instruction that includes the cross-regional migration path of computing power tasks and the power adjustment of generator units, based on the real-time operating status data and the energy consumption conversion relationship, taking the cross-regional migration amount of flexible loads and the output status of generator units as coordinated decision variables; and a coordinated execution module for responding to the coordinated scheduling instruction to migrate flexible loads located in the load center area across regions to the data center for execution through the communication network, and synchronously adjust the output of generator units in the corresponding area.

[0014] A regional power grid renewable energy consumption device based on computing power-electricity spatiotemporal coordination includes a memory and a processor: the memory is used to store a program; the processor is used to execute the program to implement the various steps of the regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination.

[0015] The technical effects and advantages of this invention's regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination are as follows: This invention integrates real-time power grid operation status and computing task data to accurately identify flexible computing loads that can be migrated across regions based on the time sensitivity of the tasks, and establishes a quantitative relationship between these loads and energy consumption. Utilizing a communication network as a "virtual transmission channel," the flexible loads are dynamically migrated from the load center to the data center for execution, while simultaneously adjusting generator output. Its significant advantages are: breaking the limitations of fixed geographical locations for traditional power loads, using cross-domain data flow transmission instead of physical current transmission, achieving flexible spatial transfer of power loads; thus, even when physical power grid transmission channels are limited, electricity demand can still be accurately dispatched to renewable energy-rich areas for local consumption, effectively solving the spatial mismatch problem between renewable energy production and consumption, and significantly improving the overall renewable energy consumption capacity and operational flexibility of the entire network. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the process for a regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination, provided in an embodiment of the present invention. Figure 2 A schematic diagram illustrating the collaborative response of physical channel blocking triggering cross-regional migration of computing power, provided in an embodiment of the present invention. Figure 3 A comparison chart of day-ahead planning and intraday real-time rolling scheduling curves provided for embodiments of the present invention; Figure 4A schematic diagram of the fallback mechanism for network congestion-triggered task migration circuit breaking provided in an embodiment of the present invention; Figure 5 This is a block diagram of a regional power grid renewable energy consumption system based on computing power-electricity spatiotemporal coordination, provided in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, Figure 1 This invention presents a method for regional power grid renewable energy consumption based on computing power-electricity spatiotemporal coordination, comprising the following steps: S1 acquires real-time operating status data of the regional power grid and pending computing power task data of the data center cluster.

[0019] It should be noted that this step aims to establish a panoramic data perception foundation for the power grid physical system and the computing power information system, specifically through the construction of a two-layer heterogeneous data acquisition interface. On the regional power grid side, the system communicates with the Energy Management System (EMS) and Wide Area Measurement System (WAMS) via the power dispatch data network, utilizing the IEC 60870-5-104 or C37.118 communication protocol to frequently acquire data on the power grid topology, real-time power flow at key transmission sections, real-time output status of conventional thermal power units, and real-time output and ultra-short-term forecast values ​​of wind and solar power plants. This data is then integrated into a set of power grid operating statuses. On the data center side, the system connects to the cluster resource schedulers (such as Kubernetes API Server or the Slurm scheduling system) of each regional data center via a fiber optic broadband network, retrieving metadata from the queue of tasks to be processed in real time. This metadata includes task submission time, the amount of data required for the task, task priority, and the expected completion time set by the task publisher. This data is then integrated into a set of computing power task statuses. To eliminate data latency discrepancies between the two systems, the system performs timestamp alignment and outlier cleanup on the collected raw data, forming a unified time-section state matrix. The system at time... The obtained comprehensive operational status set It is expressed as follows: (1) In the formula, for The set of comprehensive system operating states at any given moment; It is the real-time active power output vector of each conventional generator unit in the regional power grid, including the current power generation and ramp status of each unit; This is the real-time output vector of the new energy power station, which includes the actual power generation of wind power and photovoltaic power and the predicted power for the next dispatch cycle; This provides the real-time active power flow vectors for each transmission line and tie line, which are used to subsequently assess the congestion status of physical channels. This is a vector of pending computing tasks in a data center cluster, where each element contains a unique identifier (ID) for a single computing task, the amount of input data, and a deadline timestamp.

[0020] To visually illustrate the data structure acquired in this step, Table 1 is provided, which displays a real-time snapshot of the status of regional power grid nodes and data center nodes at a given moment. This status matrix allows the system to simultaneously assess the physical surplus / deficit of power supply and the load demand of computing power.

[0021] Table 1

[0022] S2, based on the computing power task data, the computing power tasks are divided into flexible loads according to the time sensitivity of the tasks, and the energy consumption conversion relationship required for the execution of the computing power tasks is determined.

[0023] It's important to note that traditional power dispatching schemes typically treat data centers as a single, rigid load node, focusing only on fluctuations in their total power output while ignoring the significant variability and adjustment potential of the tasks running within the data center. This "black box" management approach makes it impossible to distinguish which energy consumption must be met immediately and which can be transferred over time and space, resulting in a significant waste of dispatching resources. This step, through in-depth analysis of the temporal attributes of computing tasks, breaks through the limitations of the traditional perspective, accurately identifying the "flexible" components that can cooperate with the power grid for cross-regional dispatching. This transforms the data center from a passive large power consumer into a flexible resource that can actively participate in grid interaction, greatly enhancing the system's degree of adjustment freedom.

[0024] The method of dividing computing tasks into flexible loads based on the time sensitivity of the tasks includes: To obtain the deadline requirements and estimated processing time for computing tasks, the system first reads the Service Level Agreement (SLA) parameters from the task metadata and extracts the... The absolute deadline timestamp for each computing task and the estimated processing time based on statistics of similar historical tasks; Next, the latency tolerance of the task is calculated. The latency tolerance represents the ratio of the longest allowable waiting time to the estimated processing time. Specifically, it is the ratio of the maximum floating time window of the task under the premise of not defaulting to the actual execution time. The calculation formula is as follows: (2) In the formula, For the first Latency tolerance for each computing task; This is the deadline timestamp for the task; This is the current system decision-making moment; The estimated processing time required for this task; This represents the total time remaining until the deadline for the task.

[0025] Subsequently, tasks with latency tolerance exceeding a preset threshold are marked as flexible workloads that can be migrated across regions, specifically by setting a time sensitivity threshold. ,when If the latency is sufficient to cover the additional delay caused by cross-regional network transmission, the task is marked as a "flexible load" and added to the migration queue; otherwise, it is marked as a "rigid load" and must be processed locally immediately.

[0026] The energy consumption conversion relationship required to execute the computing task includes: Determine the basic no-load power consumption of the data center. This power consumption does not fluctuate instantaneously with changes in computing power and workload. It mainly includes the minimum power required to keep the server powered on, the basic operating power of the cooling system, and the energy consumption of lighting and security. Then, based on the computation type of the computing task, the dynamic incremental power consumption required to process a unit of data is determined. Different types of tasks (such as floating-point intensive AI training and I / O intensive database queries) have different energy consumption requirements for CPU / GPU. The system matches the corresponding energy efficiency coefficient according to the task type. Finally, the sum of the basic idle power consumption and the dynamic incremental power consumption multiplied by the computing power task load is used as the real-time power load value required for the data center to absorb the computing power tasks. This conversion relationship realizes the quantization mapping from "bit stream" to "watt stream". Data centers in Real-time power load at any given moment The calculation formula is as follows: (3) In the formula, For data centers Total power; This represents the base idle power consumption of the data center. This is the power efficiency coefficient for the data center. For the current allocation in the data center The set of tasks to be executed; Dynamic energy consumption coefficient for processing a unit of data volume for a specific type of task (unit: kWh / GB or kWh / TFLOPS). For the first Data load of each task.

[0027] This step, by establishing a precise task elasticity assessment model and energy consumption conversion model, transforms the originally abstract computer task into a virtual power load that the power grid dispatching system can perceive, quantify, and control, laying the physical foundation for subsequent "calculation follows power" collaborative dispatching.

[0028] S3. Based on the real-time operating status data and the energy consumption conversion relationship, the cross-regional migration amount of flexible loads and the output status of generator sets are used as collaborative decision variables to generate collaborative scheduling instructions that include the cross-regional migration path of computing tasks and the power adjustment of generator sets.

[0029] It should be noted that traditional power dispatching schemes, when faced with congestion in inter-regional transmission channels, are often constrained by the topological limitations of the physical power grid. They are forced to either cut off surplus renewable energy from the sending-end grid (wind and solar curtailment) or restrict load demand from the receiving-end grid (orderly power consumption). This passive "grid-centric" adjustment approach not only results in a huge waste of clean energy but also increases the overall operating cost of the system. The method in this step introduces "computing power flow" as a virtual energy carrier that can cross the boundaries of the physical power grid. Utilizing the extremely high transmission bandwidth and extremely low transmission energy consumption of fiber optic networks, it "transports" the power demand that would otherwise have to be consumed at the load center to renewable energy-rich areas for local consumption. Thus, without increasing investment in physical transmission lines, it effectively breaks through the bottleneck of the power grid's transmission capacity and achieves optimal allocation of network resources over a larger spatiotemporal range.

[0030] The generation of collaborative scheduling instructions, which includes cross-regional migration paths for computing tasks and adjustments to generator power, includes: A collaborative scheduling model is constructed with the goal of minimizing overall operating costs and maximizing renewable energy consumption. This model includes the total cost of computing power and power dispatch, encompassing the operating fuel costs of conventional generating units, the cost of wind and solar curtailment penalties due to the inability to absorb renewable energy, and the network bandwidth rental costs and latency penalties incurred from transmitting computing tasks across regions. The objective function for this total cost is... The calculation formula is as follows: (4) In the formula, The scheduling period; This refers to the number of conventional generator sets; For the unit exist Contributing effort at all times; For the unit exist Start-stop status variables at any time (1 indicates power on, 0 indicates power off); This refers to the unit's consumption characteristic coefficient; The cost of starting the unit once, Used to identify whether the unit has started up; The number of new energy power stations; The unit of power curtailment penalty coefficient; For new energy power stations The amount of abandoned power; This refers to the number of data center nodes. for From the source data center at all times Migrate to target data center The amount of data required for computing power tasks; Cost of network bandwidth transmission per unit of data volume; This is the penalty coefficient for delay time; For nodes arrive Network transmission latency.

[0031] A multivariate constrained optimization algorithm is employed to solve the cooperative scheduling model within a solution space that satisfies power system operation constraints and computing network constraints. This yields the computing power task allocation scheme and unit output scheme that minimizes the total cost. Specifically, a mixed-integer linear programming (MILP) algorithm or a particle swarm optimization (PSO) algorithm is used to reduce the computing power migration... and unit output Iterative optimization is performed using these variables as joint optimization variables.

[0032] The power system operation constraints are used to ensure the safe operation of the power network within physical transmission limits, specifically covering node power balance constraints, line transmission capacity constraints, and unit operating status constraints, namely: Establish nodal power balance equations based on a DC power flow model. For any node in the power grid... The injected power (generation power and virtual increment) and outflow power (load and line power flow) must be balanced in real time. The formula for the node power balance constraint is as follows: (5) In the formula, To connect at node A collection of generating units or stations; Contribute to new energy forecasting; This refers to the amount of abandoned electricity. For nodes The basic rigid load; This is the equivalent injected power (negative for migration in, positive for migration out) caused by the migration of computing power tasks to the data center power of this node. and They are nodes and adjacent nodes The voltage phase angle; This refers to the line reactance.

[0033] It should be noted that, This is the equivalent value of the change in net injected power of the node caused by the cross-region migration of computing power tasks, and it is related to the computing power migration variable. The relationship is defined as follows: (6) That is, when computing power tasks are generated from nodes When migrating out, the local load decreases, which is equivalent to injecting positive power into the power grid (migrating out is positive); when computing tasks migrate into the node... When the local load increases, it is equivalent to injecting negative power into the power grid (migration is negative). This is the energy consumption conversion factor.

[0034] The formula for the line transmission capacity constraint is as follows: (7) In the formula, For the line Thermal stability limit of transmission capacity.

[0035] The unit operating status constraint formula is as follows: (8) (9) In the formula, For the unit start-up and shutdown state variables (0 or 1), this constraint guarantees that when (When shutting down), the unit output must be 0; when When powered on, the output is at the minimum technical output. With maximum output between; This is a limit on the unit's ramp-up rate.

[0036] The computing power network constraints are used to ensure that the amount of tasks migrating into the data center does not exceed the upper limit of the data center's outbound network bandwidth, and that the sum of the total processing time and network transmission time of the tasks does not exceed the maximum allowed latency time of the tasks. The specific constraint formula is as follows: (10) (11) In the formula, For target data center Maximum receiving bandwidth; The length of the scheduling period; The data processing rate of the target data center; This refers to the link transmission rate; For the task The maximum allowable delay.

[0037] Based on the solution logic of the above optimization model, this embodiment does not employ simple conditional judgments in its specific space scheduling strategy, but rather a routing optimization mechanism guided by shadow price. Specific implementation details are as follows: During the solution process, the system calculates the dual variable or Lagrange multiplier of the capacity constraints of each transmission line in the power system operation constraints in real time. When the physical transmission channel from the first region (new energy rich area) to the second region (load center) has not reached its limit, the dual variable is zero, and the power transmission cost is only the network loss; when the physical transmission channel reaches the thermal stability limit (i.e., blockage occurs), the dual variable of the constraint rises rapidly, which means that the marginal cost of transmitting an additional unit of electricity through the physical line tends to infinity.

[0038] At this point, the optimization algorithm will automatically compare the computing power transmission cost (i.e. ) and the marginal cost of congestion. As long as the generalized cost of migrating computing power tasks is lower than the cost of power-wasting penalties caused by physical congestion ( The algorithm will then transfer computing power. (From the load center) To the rich area () is a non-zero preferred solution.

[0039] In practice, the system generates routing instructions based on the following logic: 1) Read the physical line at the current time Trend value and limit value ; 2) If Prioritize increasing physical delivery and maintain local execution of computing tasks. ); 3) If Lock the physical transmission power and start the computing power flow calculation; 4) Calculate the unit load from the area Migrate to Comprehensive cost and the area's power curtailment penalty contrast; 5) If Then let Its value is determined by the objective function. The gradient descent direction determines the path until the bandwidth limit is reached. Or it can absorb all the surplus electricity.

[0040] Among them, the comprehensive cost This refers to the specific quantification of the aforementioned computing power transmission cost, representing the migration cost per unit of data volume, and its calculation formula is as follows: ; The benefits of the power abandonment penalty The formula for calculating the marginal loss incurred when the corresponding amount of electricity cannot be consumed locally and is therefore forcibly abandoned is as follows: .

[0041] To verify the effectiveness of the aforementioned shadow price-based routing mechanism, this embodiment simulates a typical congestion mitigation process, and the results are as follows: Figure 2 As shown in the figure, the upper curve represents the real-time power flow changes of the physical power transmission channel (West-to-East Power Transmission), and the lower curve represents the cross-regional migration rate of the generated computing power tasks.

[0042] from Figure 2 It can be seen that, At a certain moment, the power flow in the physical channel reaches its thermal stability limit, at which point the congestion shadow price surges. The optimization algorithm responds rapidly, with the computing power migration rate below increasing from 0 to 50Gbps, indicating that the system "bypasses" some of the energy demand through the fiber optic network. As the computing power load migrates out, the power demand at the load center decreases, and the power flow pressure in the physical channel... The moment was alleviated, proving the decoupling effect of coordinated scheduling.

[0043] In addition, in order to cope with the output of new energy sources Due to the random fluctuations, the generation of the aforementioned coordinated scheduling instructions is not a one-time static process, but rather a phased execution relying on a multi-timescale rolling coordination mechanism. The objective function of the aforementioned coordinated scheduling model... The solution is broken down into two phases: day-ahead planning and intraday scheduling, and is solved progressively. The specific implementation steps are as follows: 1) During the day-ahead planning phase, the system, based on the renewable energy power prediction curve for the next 24 hours and historical computing power task data, calculates the total cost for the entire day. With minimization as the objective, a long-term solution is performed on the cooperative scheduling model. The core purpose of this solution is to determine the start-up and shutdown state variables of conventional generator units. (A binary variable of 0 / 1), which determines which generating units will remain operational the following day to provide inertia support. Simultaneously, the system, based on the predicted distribution of renewable energy surplus, performs an objective function... Under the constraints, the computing resources and network bandwidth of the data center in the new energy rich area are reserved in advance to reserve physical channels for the "computing power migration" the next day and form the basic scheduling plan for the next day. 2) During the intraday scheduling phase, the system introduces Model Predictive Control (MPC) technology, based on ultra-short-term renewable energy forecast data with higher time resolution (e.g., 15-minute level), to determine the unit start-up and shutdown status during the day-ahead phase. Given a fixed premise, the collaborative scheduling model is solved using high-frequency rolling iterations. Specifically, the system opens a finite prediction time window (e.g., the next 4 hours) that extends backward every preset time period (e.g., 15 minutes), and minimizes the objective function again within this window. The focus is on the real-time active power output of the generator sets. and the amount of flexible loads migrating across regions These two continuous decision variables are optimized in a refined manner. After the solution is completed, the system only issues the coordinated scheduling instruction for the first time segment to the executor, and uses the optimization results of subsequent time segments as the initial state input for the next round of calculation. Through this rolling update mechanism, when the output of new energy sources is monitored in real time... When there is a deviation from the predicted value, the optimization algorithm can adjust the variables in real time. The numerical value is dynamically adjusted to increase or decrease the computing power data flow to the new energy rich area, thereby mitigating the impact of prediction errors on the system in the time dimension.

[0044] The effect of multi-timescale rolling mechanism on the correction of scheduling instructions is as follows: Figure 3 As shown in the figure, the dashed line represents the baseline power consumption plan for the western data center formulated during the planning phase, while the solid line represents the actual execution curve during the intraday rolling phase. It can be seen that around 14:00, due to a sudden increase in the actual output of western wind power compared to the predicted value (a positive deviation), the intraday rolling mechanism dynamically increased the flexible computing load relocated to the region (the solid line is higher than the dashed line), thereby achieving "instant absorption" of the sudden surplus wind power and preventing wind curtailment.

[0045] This step expands traditional single-dimensional power scheduling into multi-dimensional collaborative scheduling of computing power and power by constructing an optimized scheduling model that deeply couples computing power and power, and combining spatial dynamic routing strategies and temporal rolling correction mechanisms. Under the premise of ensuring grid security and computing power service quality, it maximizes the utilization of the entire network's transmission channel resources and new energy power generation resources.

[0046] S4, in response to the coordinated scheduling command, the flexible load located in the load center area is migrated across regions to the data center for execution through the communication network, and the output of the generator units in the corresponding area is adjusted synchronously.

[0047] It should be noted that this step is the closed-loop execution link of the "computing power-power" collaborative scheduling. The system first decouples the collaborative scheduling instructions generated by S3, decomposing them into computing power scheduling sub-instructions and power scheduling sub-instructions. For the computing power side, the system issues routing policies through the software-defined networking (SDN) controller and cloud resource management platform, establishes a data transmission tunnel from the load center (such as the eastern node) to the renewable energy rich area (such as the western node), and initiates virtual machine migration or container image pull operations. For the power side, the system issues new active power base points to relevant generator units through the automatic generation control (AGC) system. Specifically, at the sending end (renewable energy rich area), due to the receipt of additional computing power load, the system will order an increase in the output limit of wind and solar power stations in the area or a reduction in the voltage drop of thermal power units to achieve local consumption of renewable energy; at the receiving end (load center), due to the relocation of some computing power load, the system will order a reduction in the output of thermal power units in the area, thereby reducing fossil fuel consumption.

[0048] The process of migrating flexible loads located in the load center area to the data center across regions via a communication network includes: Before performing cross-regional migration of computing power tasks, the congestion status and transmission latency of the inter-regional communication network are detected. Specifically, a network probe is used to send detection data packets in real time to measure the round-trip time (RTT) and packet loss rate of the link, and to query the port bandwidth utilization of the backbone network router. Data transmission channels are only opened when the link status meets the quality of service (QoS) requirements.

[0049] During task migration, if the transmission latency of the communication network exceeds a safety threshold, a rollback mechanism is triggered to stop the cross-regional migration of subsequent tasks, switch unmigrated tasks back to the local data center for processing, and simultaneously activate the backup regulation resources of the local power grid. Specifically, the system sets a network latency safety threshold. Continuously monitor real-time latency during transmission. .when If the network link is congested or fails after a certain period of time, the system will immediately perform a "circuit breaker" operation: First, it will interrupt the sending of new data packets to remote locations and push the remaining pending tasks back into the local data center's job queue to ensure that no tasks are lost; Second, as the return of tasks causes the local power load to rebound, the power dispatching system will quickly call upon the local spinning reserve or fast start-stop units to fill the power gap and ensure the stability of the local power grid frequency.

[0050] The triggering process of communication link status awareness and fallback mechanism is as follows: Figure 4 As shown in the figure, the real-time relationship between network transmission latency and task migration throughput is illustrated. At any moment, network latency suddenly increased and exceeded the safety threshold. (50ms) The system's circuit breaker protection logic is triggered, and the task migration throughput automatically drops to zero within milliseconds, effectively preventing task accumulation and timeout failure in congested networks.

[0051] This step addresses the risk of computing power services being highly dependent on network communication quality by establishing a dual guarantee mechanism of "pre-event detection" and "in-event circuit breaking." It ensures that when communication network fluctuations occur, the system can smoothly revert from "cross-domain collaborative mode" to "local autonomous mode," thus guaranteeing the continuity of data services and the safe and stable operation of the power system.

[0052] Example 2, Figure 5 A regional power grid renewable energy consumption system based on computing power-electricity spatiotemporal coordination is presented, including: The data acquisition module is used to acquire real-time operating status data of the regional power grid and pending computing power task data of the data center cluster; The task analysis module is used to divide the computing power tasks into flexible loads based on the computing power task data and the time sensitivity of the tasks, and to determine the energy consumption conversion relationship required for the execution of the computing power tasks. The strategy decision module is used to generate a collaborative scheduling instruction that includes the cross-regional migration path of computing tasks and the power adjustment of generator sets, based on the real-time operating status data and the energy consumption conversion relationship, taking the cross-regional migration amount of flexible loads and the output status of generator sets as collaborative decision variables. The collaborative execution module is used to respond to the collaborative scheduling command, migrate flexible loads located in the load center area across regions to the data center for execution through the communication network, and synchronously adjust the output of generator sets in the corresponding areas.

[0053] Example 3: A regional power grid renewable energy consumption device based on computing power-electricity spatiotemporal coordination, comprising a memory and a processor: the memory is used to store a program; the processor is used to execute the program to implement any of the embodiments in Example 1.

[0054] Since the regional power grid renewable energy consumption device based on computing power-electricity spatiotemporal coordination described in this embodiment is the device used to implement the method in Embodiment 1 of this invention, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the method described in Embodiment 1 of this application. Therefore, how the electronic device implements the method in this application embodiment will not be described in detail here. Any device used by those skilled in the art to implement the method in this application embodiment falls within the scope of protection of this application.

[0055] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0056] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0057] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0058] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0059] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0060] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for regional power grid renewable energy consumption based on computing power-electricity spatiotemporal coordination, characterized in that, Includes the following steps: Acquire real-time operational status data of the regional power grid and pending computing power task data of the data center cluster; Based on the time sensitivity of computing task data, computing tasks are divided into flexible loads, and the energy consumption conversion relationship required for the execution of computing tasks is determined. Using the cross-regional migration amount of flexible loads and the output status of generator sets as collaborative decision variables, and based on real-time operating status data and energy consumption conversion relationships, collaborative scheduling instructions are generated that include cross-regional migration paths of computing tasks and generator set power adjustments. The system executes coordinated scheduling commands to migrate flexible loads in the load center area to the data center for processing across regions via the communication network, and simultaneously adjusts the output of generator units in the corresponding areas.

2. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 1, characterized in that, The determination of the energy consumption conversion relationship required for the execution of the computing task includes: Determine the baseline no-load power consumption of the data center; Determine the dynamic incremental power consumption required to process a unit of data based on the type of computation of the computing task. The sum of the base idle power consumption and the dynamic incremental power consumption multiplied by the computing power task load is used as the real-time power load value required by the data center to absorb computing power tasks.

3. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 1, characterized in that, The generation of collaborative scheduling instructions, which includes cross-regional migration paths for computing tasks and adjustments to generator power, includes: A collaborative scheduling model is constructed with the goal of minimizing the overall operating cost and maximizing the absorption of new energy sources. The overall operating cost includes the fuel cost of the power system, the penalty cost for abandoning new energy sources, and the cost of cross-domain migration of computing tasks. Under the constraints of power system operation and computing network, the collaborative scheduling model is solved to obtain the optimal cross-regional allocation scheme of computing tasks and generator output scheme.

4. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 3, characterized in that, The power system operation constraints are used to ensure the safe operation of the power network within the physical transmission limits, including power balance constraints, line transmission capacity constraints, and unit operating status constraints. The computing power network constraints are used to ensure that the amount of tasks migrated into the data center does not exceed the upper limit of the data center's outbound network bandwidth, and that the sum of the processing latency and network transmission latency of each computing power task does not exceed the maximum allowed latency time for that task.

5. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 1, characterized in that, The time sensitivity of computing power task data is used to classify computing power tasks into flexible loads, including: Obtain the deadline requirements and estimated processing time for computing power tasks; Calculate the latency tolerance of the task, which represents the ratio of the longest allowed waiting time to the estimated processing time; Tasks with latency tolerance exceeding a preset threshold are marked as flexible loads that can be migrated across regions.

6. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 1, characterized in that, The generation of collaborative scheduling instructions, which includes cross-regional migration paths for computing tasks and adjustments to generator power, includes: When it is detected that there is curtailment of new energy in the first region and there is flexible load demand in the second region, the remaining capacity of the physical transmission channel is compared with the available bandwidth of the computing power transmission channel. If the remaining capacity of the physical power transmission channel is insufficient to transmit and absorb the power required by new energy sources, then routing instructions will be generated first to send the flexible load data of the second region to the data center of the first region for calculation via the fiber optic network.

7. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 1, characterized in that, The regional power grid renewable energy consumption method also includes two phases: day-ahead planning and intraday real-time implementation. Specific implementation methods include: During the current planning phase, based on future cycles of new energy output forecasts and computing power demand forecasts, generator start-up and shutdown plans and data center resource reservation plans are formulated. During the intraday scheduling phase, based on ultra-short-term renewable energy output forecasts, the coordinated scheduling instructions are periodically updated and the cross-regional migration of flexible loads is dynamically adjusted according to the real-time deviation of renewable energy output.

8. The regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination according to claim 1, characterized in that, The process of migrating flexible loads from the load center area to the data center for processing across regions via a communication network includes: Before and during the cross-regional migration of computing power tasks, monitor the transmission latency of the inter-regional communication network; When the transmission delay is detected to exceed a preset safety threshold, a rollback mechanism is triggered. The rollback mechanism includes: suspending the cross-regional migration of subsequent tasks, switching unmigrated tasks to the local data center for processing, and triggering the deployment of backup regulation resources of the local power grid.

9. A system using the regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination as described in any one of claims 1-8, characterized in that, include: The data acquisition module is used to acquire real-time operating status data of the regional power grid and pending computing power task data of the data center cluster; The task analysis module is used to divide computing tasks into flexible loads based on the time sensitivity of computing task data, and determine the energy consumption conversion relationship required for the execution of computing tasks. The strategy decision module is used to generate collaborative scheduling instructions that include the cross-regional migration path of computing tasks and the power output status of generator sets, based on real-time operating status data and energy consumption conversion relationships, using the cross-regional migration amount of flexible loads and the output status of generator sets as collaborative decision variables. The collaborative execution module is used to execute collaborative scheduling instructions, which migrate flexible loads in the load center area to the data center for processing across regions through the communication network, and synchronously adjust the output of generator sets in the corresponding areas.

10. A computer device, characterized in that, The device includes a processor and a memory: The memory is used to store computer programs and send the instructions of the computer programs to the processor; The processor executes, according to the instructions of the computer program, a regional power grid renewable energy consumption method based on computing power-electricity spatiotemporal coordination as described in any one of claims 1-8.

Images