Hardware resource dynamic allocation method and device and electronic equipment

Abstract

The application provides a hardware resource dynamic allocation method, device and electronic equipment, and belongs to the technical field of power distribution networks. The method comprises the following steps: obtaining hardware resource requirements corresponding to a plurality of power distribution network applications and hardware resource state data of the power distribution network; constructing a target function based on the hardware resource requirements of the plurality of power distribution network applications and the hardware resource state data; the target function at least represents hardware resource utilization and load balancing of a resource allocation strategy corresponding to the plurality of power distribution network applications; solving the target function to obtain a target resource allocation strategy corresponding to the plurality of power distribution network applications. The application is used to solve the defect that the hardware device resource utilization is low in the existing scheme of the hardware resource allocation strategy for the power distribution network application, which mostly adopts the mode of 'one application one hardware'.

Description

Technical Field

[0001] This invention relates to the field of power distribution network technology, and more specifically to a method for dynamic allocation of hardware resources, a device for dynamic allocation of hardware resources, an electronic device, a machine-readable storage medium, and a computer program product. Background Technology

[0002] Against the backdrop of energy transition and the deep integration of digital technologies, smart distribution networks have become the core hub for the construction of new power systems, and the advanced application systems they support are constantly being enriched and improved. To achieve safe, efficient, and reliable power supply, distribution networks need to simultaneously support multi-dimensional application scenarios, including local intelligent protection and control, real-time monitoring of grid operation status, precise fault location and isolation, and non-intrusive load monitoring. These applications together constitute the core capabilities of smart distribution networks, driving the grid to upgrade from traditional operation and maintenance to proactive management and control.

[0003] Currently, most existing hardware resource allocation strategies for power distribution network applications adopt a "one application, one hardware" model, configuring dedicated devices for each advanced application, such as protection devices and monitoring terminals. This results in hardware redundancy, with some devices operating at low load for extended periods, leading to low resource utilization. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus, and electronic device for dynamic allocation of hardware resources, in order to solve the problem that most existing solutions for hardware resource allocation in power distribution network applications adopt a "one application, one hardware" model, resulting in low utilization of hardware resources.

[0005] To achieve the above objectives, embodiments of the present invention provide a method for dynamic allocation of hardware resources, including: Obtain hardware resource requirements and hardware resource status data of multiple distribution network applications; Based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications, an objective function is constructed; the objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategies corresponding to the multiple distribution network applications. Solving the objective function yields the target resource allocation strategy corresponding to the multiple power distribution network applications.

[0006] On the other hand, the present invention also provides a hardware resource dynamic allocation device, comprising: The first acquisition module is used to acquire the hardware resource requirements and hardware resource status data of the distribution network corresponding to multiple distribution network applications. The first construction module is used to construct an objective function based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications; the objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications. The first solution module is used to solve the objective function to obtain the target resource allocation strategy corresponding to the multiple distribution network applications.

[0007] On the other hand, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described dynamic allocation method for hardware resources.

[0008] On the other hand, the present invention also provides a machine-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described dynamic allocation method for hardware resources.

[0009] On the other hand, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described dynamic allocation method for hardware resources.

[0010] Through the above technical solution, this invention constructs an objective function that characterizes at least the hardware resource utilization and load balancing of resource allocation strategies corresponding to multiple distribution network applications, and solves the objective function to obtain the optimal target resource allocation strategy for hardware resource utilization and load balancing for multiple distribution network applications. Therefore, this invention can improve the hardware resource utilization and load balancing of distribution network hardware devices corresponding to multiple distribution network applications, reduce hardware redundancy, and solve the problem that most existing hardware resource allocation strategies for distribution network applications adopt a "one application, one hardware" model, resulting in low hardware resource utilization.

[0011] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0012] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the dynamic allocation method for hardware resources provided by the present invention; Figure 2 This is a schematic diagram of the hardware resource dynamic allocation device provided by the present invention; Figure 3 This is one of the structural schematic diagrams of the electronic device provided by the present invention; Figure 4 This is the second schematic diagram of the electronic device provided by the present invention. Detailed Implementation

[0013] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0014] Method Implementation Examples Please refer to Figure 1 This invention provides a method for dynamic allocation of hardware resources, including: Step 100: Obtain the hardware resource requirements and hardware resource status data of multiple distribution network applications.

[0015] Electronic devices at the distribution network end acquire hardware resource requirements and hardware resource status data corresponding to multiple distribution network applications. These devices include application adaptation interfaces and a hardware resource pool. The application adaptation interfaces provide standardized interfaces, encapsulating multiple distribution network applications, such as local protection applications, real-time monitoring applications, and fault location applications, into dynamically loadable software modules, thus decoupling applications from hardware resources. Furthermore, through the application adaptation interfaces in the system architecture, the electronic devices perform unified data collection and parsing for various applications, such as acquiring the hardware resource requirements corresponding to multiple distribution network applications, thereby ensuring the standardization and completeness of the requirement parameters. The hardware resource requirements corresponding to distribution network applications may include the number of real-time cores, the number of general-purpose cores, high-speed storage bandwidth requirements, and low-latency communication bandwidth requirements. For example, taking a local protection application as an example, in one embodiment, the hardware resource requirements for the local protection application are: 1 real-time core, 2GB memory, 10% communication bandwidth, and 30GB SSD storage.

[0016] The hardware resource status data of the power distribution network can be quantitative data of various hardware components of electronic devices (such as CPU, memory, storage devices, communication links, etc.). In one embodiment, electronic devices can collect real-time hardware resource status data according to resource type. For example, every 10ms, they can collect CPU core utilization rate and overall load; memory used capacity, remaining space, and cache usage; storage device used capacity, available capacity, and partition usage status; and communication link bandwidth utilization rate and actual transmission rate to obtain the occupied and remaining amounts of various resources and establish a resource ledger. Electronic devices can also synchronously collect CPU base frequency and real-time operating frequency; memory read / write rate; storage device input / output operations per second and read / write latency; and communication link packet loss rate, latency jitter, and other performance parameters.

[0017] In other embodiments, the electronic device can also monitor the resource health status data of its own hardware. For example, it can periodically collect and evaluate reliability indicators of the hardware, such as CPU temperature, write cycles and wear rate of storage media, fan speed, and power module output voltage. When the monitored data approaches or exceeds a threshold, the resource status is automatically marked as a warning or fault. The collected quantitative resource data is fused with the health status data to generate resource status markers, which are timestamped to ensure that every piece of status data is traceable, providing the decision-making level with real-time and comprehensive resource status.

[0018] Step 200: Based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications, construct an objective function; the objective function at least represents the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications.

[0019] The electronic device constructs an objective function based on the hardware resource requirements and hardware resource status data of multiple distribution network applications. The objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategies corresponding to the multiple distribution network applications. For example, in one embodiment, based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications, multiple resource allocation strategies can be formed for the multiple distribution network applications. Each resource allocation strategy has a hardware resource utilization index to evaluate hardware resource utilization and a load balancing index to evaluate load balancing. The hardware resource utilization index is used to improve overall resource utilization efficiency. The load balancing index is used to avoid overload of a single resource. The calculation formula for the hardware resource utilization index is as follows: ;Formula (1) in F utilization This represents the hardware resource utilization index, where k=1,2,3,4 correspond to four types of core hardware resources: CPU, memory, storage, and communication bandwidth, respectively. The value of the k-th type of hardware resources allocated to the i-th application; R kt Let n be the total capacity of the k-th type of hardware resources, and n be the total number of applications.

[0020] The formula for calculating load balancing metrics is as follows: ;Formula (2) in F balance This represents a load balancing metric. This represents the utilization rate of the j-th hardware resource unit within the k-th type of hardware resources. Let be the load balancing threshold for the k-th type of hardware resource; This represents the load difference among the units of the k-th type of hardware resource; the smaller the difference, the higher the load difference. The closer it is to 1, the more balanced the load.

[0021] The objective function of this invention aims to maximize at least the hardware resource utilization rate and load balancing rate of the resource allocation strategy. Guided by the above idea, the calculation formula of the objective function is as follows: ;Formula (3) in, F total This represents the overall optimization objective value. F utilization This represents a hardware resource utilization indicator. F balance This represents the load balancing metric. The optimization objective of the above objective function is to maximize the hardware resource utilization and load balancing performance of the resource allocation strategies corresponding to the multiple distribution network applications by maximizing the hardware resource utilization and load balancing metrics of the resource allocation strategies.

[0022] The objective function described above can be constrained by matching application requirements with resources and adjusting load balancing. Taking application requirements with resources as an example, suitable resource subsets can be selected from resource status markers based on the application's resource type requirements. For example, matching real-time CPU cores and high-speed cache memory to local intelligent protection applications; matching large-capacity hard drives or cost-effective SSDs to load monitoring applications; and matching low-packet-loss communication links to real-time communication applications. Performance verification is performed on the selected resource subsets to ensure that resource performance meets application requirements. Resources in the resource subset with warning or fault status are removed, and resources with good health status are prioritized for allocation to high-priority applications to reduce application operational risks. Taking load balancing as an example, load balancing thresholds are set for different resource types, such as a CPU single-core utilization threshold of 70%, an overall memory utilization threshold of 80%, and a storage partition utilization threshold of 85%. Based on resource status markers, resources with loads exceeding the thresholds are identified, and the application types and priorities running on these resources are analyzed. Low-to-medium priority application tasks on overloaded resources are migrated to less loaded, idle resources, while ensuring uninterrupted application services during the migration process. For example, some computing tasks of a load monitoring application can be migrated from high-load CPU cores to idle cores, reducing the load on the original cores to below 65%.

[0023] Step 300: Solve the objective function to obtain the target resource allocation strategy corresponding to the multiple power distribution network applications.

[0024] Electronic devices can use various multi-objective optimization decision-making algorithms to solve the above objective function. For example, electronic devices can use genetic algorithms, particle swarm optimization algorithms, and other multi-objective optimization decision-making algorithms to solve the objective function, and obtain the optimal target resource allocation strategy for the resource utilization index and load balancing index corresponding to the multiple distribution network applications. It should be noted that the embodiments of the present invention can set a fixed resource allocation refresh cycle, such as 100ms. Each cycle, the hardware resource dynamic allocation method of steps 100 to 300 of the embodiments of the present invention is started once, and the existing allocation scheme is evaluated in combination with the latest hardware resource requirement list and resource status flag of the distribution network application. In other aspects of the embodiments of the present invention, a real-time triggering mechanism can also be established. When an application state change or resource state change occurs, the hardware resource dynamic allocation method of the embodiments of the present invention is immediately triggered without waiting for the periodic refresh. For example, when a storage module suddenly fails, the decision-making process is immediately started to reallocate spare storage resources for applications that depend on the module.

[0025] This invention, through constructing an objective function that characterizes at least the hardware resource utilization and load balancing of resource allocation strategies corresponding to multiple distribution network applications, and solving the objective function, yields the optimal target resource allocation strategy for hardware resource utilization and load balancing for multiple distribution network applications. Therefore, this invention can improve the hardware resource utilization and load balancing of distribution network hardware devices corresponding to multiple distribution network applications, reduce hardware redundancy, and solve the problem that most existing hardware resource allocation strategies for distribution network applications adopt a "one application, one hardware" model, resulting in low hardware resource utilization.

[0026] In other aspects of the embodiments of the present invention, the step of constructing an objective function based on the hardware resource requirements of multiple distribution network applications and the hardware resource status data includes: determining a set of resource allocation strategies based on the hardware resource requirements of multiple distribution network applications and the hardware resource status data; the resource allocation strategies include multiple resource allocation strategies and hardware resource utilization indicators and load balancing indicators for each resource allocation strategy; constructing an objective function based on the hardware resource utilization indicators and the load balancing indicators; the hardware resource utilization indicators and the load balancing indicators are positively correlated with the total optimization objective value of the objective function.

[0027] Electronic devices can formulate multiple resource allocation strategies for multiple power distribution network applications based on their hardware resource requirements and status data. Each resource allocation strategy has a hardware resource utilization rate index to evaluate hardware resource utilization and a load balancing index to evaluate load balancing. The calculation formulas for the hardware resource utilization rate index and the load balancing index are described above. In one embodiment, a target function is constructed based on the hardware resource utilization rate index and the load balancing index, including: constructing the target function based on the hardware resource utilization rate index, a first weight corresponding to the hardware resource utilization rate index, the load balancing index, and a second weight corresponding to the load balancing index. In order to consider the different impacts of the hardware resource utilization rate index and the load balancing index on the overall optimization target value of the target function, the weights of the hardware resource utilization rate index and the load balancing index can be added to the target function. For example, the formula of the target function in another embodiment is as follows: ;Formula (4) in, F total This represents the overall optimization objective value. F utilization This represents a hardware resource utilization indicator. F balance This represents a load balancing metric. oh 1 indicates the first weight. oh 2. Second weight. Wherein... oh 1+ oh 2=1. If a hardware resource utilization metric that prioritizes resource allocation strategies is preferred, the first weight can be set to be greater than the second weight, for example... oh 1 = 0.6 oh 2 = 0.4. Conversely, if a load balancing metric that prioritizes resource allocation strategies is preferred, the second weight can be set to be greater than the first weight, for example... oh 1 = 0.4 oh 2 = 0.6. If considering the balance of hardware resource utilization and load balancing metrics in resource allocation strategies, it can be set to... oh 1 = 0.5 oh 2 = 0.5.

[0028] In other aspects of the embodiments of the present invention, the resource allocation strategy set further includes an application priority satisfaction index for each resource allocation strategy; the application priority satisfaction index characterizes the resource supply level of key applications in the resource allocation strategy; the step of constructing an objective function based on the hardware resource utilization index and the load balancing index includes: constructing an objective function based on the hardware resource utilization index, the load balancing index, and the application priority satisfaction index of the resource allocation strategy; the application priority satisfaction index, the hardware resource utilization index, and the load balancing index are positively correlated with the total optimization objective value of the objective function.

[0029] In existing dynamic allocation methods for hardware resources in power distribution network applications, each application hardware operates independently, lacking a unified resource scheduling mechanism. When multiple applications simultaneously initiate resource requests, resource conflicts easily occur, affecting the reliability of critical applications. Furthermore, fixed hardware resource allocation methods cannot adapt to dynamic changes in application requirements in real time. For example, when a power grid fault is detected, the fault location application needs to temporarily increase its computing power priority, but existing solutions struggle to quickly adjust resource allocation ratios, leading to response delays and impacting the real-time performance of critical applications. To achieve intelligent dynamic allocation of hardware resources and ensure the real-time performance and reliability of critical applications, this invention incorporates an application priority satisfaction index into the objective function. This index is used to guarantee the supply of resources for high-priority applications. In one embodiment, the application priority satisfaction index is calculated using the following formula: ;Formula (5) Where n is the total number of applications to be allocated hardware resources; oh i Let be the priority weight of the i-th application; R ia The total amount of hardware resources actually allocated to the i-th application; R ir Let be the total theoretical hardware resource requirement for the i-th application. F priorityTo ensure application priority meets the criteria, this invention can prioritize all applications based on their importance or criticality. For example, this invention categorizes all applications awaiting resource allocation into three levels: high, medium, and low. High-level applications are designated as P1-P2, medium-level as P3-P4, and low-level as P5, with assigned priority weights, such as P1 weight of 0.6, P3 weight of 0.3, and P5 weight of 0.1. For instance, the application priority for local intelligent protection is set to P1; the application priority for load monitoring is set to P3; and the application priority for system log backup is set to P5. In other embodiments, this invention further filters high-priority applications based on their time-sensitive requirements, marking applications with urgent needs, such as fault location and isolation applications during power grid failures, whose real-time requirements are ≤50ms. These applications enjoy the highest priority for resource allocation. For applications with urgent needs, at least 20% redundant resources are reserved to cope with sudden demands, ensuring that the resource supply for high-priority applications is not affected by low-priority applications.

[0030] In one embodiment, the objective function is calculated as follows: ;Formula (6) in, F total This represents the overall optimization objective value. F utilization This represents a hardware resource utilization indicator. F balance This represents a load balancing metric. F priority To meet application priority requirements, electronic devices can use multi-objective optimization decision-making algorithms such as genetic algorithms and particle swarm optimization algorithms to solve the objective function of formula (6), thereby obtaining a target resource allocation strategy that optimizes the application priority requirements, resource utilization rate, and load balancing indicators for the multiple distribution network applications. This embodiment of the invention utilizes a resource scheduling mechanism based on application priority and demand characteristics to accurately capture application demands and resource status, relying on intelligent algorithms to achieve optimal resource allocation and real-time adjustment. In other words, this embodiment of the invention, based on application priority and demand characteristics, combines application demand characteristics with the power grid operating status to achieve intelligent dynamic allocation of hardware resources, ensuring the real-time performance and reliability of critical applications, and solving the allocation optimization problem when multiple advanced applications share hardware resources in a distribution network.

[0031] In other aspects of the embodiments of the present invention, the construction of the objective function based on the hardware resource utilization index, load balancing index, and application priority satisfaction index of the resource allocation strategy includes: constructing the objective function based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, the second weight corresponding to the load balancing index, the application priority satisfaction index, and the third weight corresponding to the application priority satisfaction index.

[0032] ;Formula (7) in, F total This represents the overall optimization objective value. F utilization This represents a hardware resource utilization indicator. F balance This represents a load balancing metric. F priority Prioritize applications to meet performance metrics. oh 1 indicates the first weight. oh 2. Second weight, oh 3 indicates the third weight. Wherein... oh 1+ oh 2+ oh 3=1. Similarly, in order to consider the different impacts of application priority satisfaction, hardware resource utilization, and load balancing on the overall optimization objective value of the objective function, the embodiments of the present invention can add weights to the objective function for application priority satisfaction, hardware resource utilization, and load balancing. To better favor the application priority satisfaction indicator that considers resource allocation strategies, the third weight can be set to be greater than the first and second weights, for example, set to... oh 3 = 0.6 oh 1= oh 2 = 0.2. Therefore, this embodiment of the invention ensures the real-time performance and reliability of critical applications by increasing the weight of application priority to meet the indicators.

[0033] In other aspects of the embodiments of the present invention, the resource allocation strategy further includes an operational risk indicator for each resource allocation strategy; the construction of an objective function based on the hardware resource utilization indicator, load balancing indicator, and application priority satisfaction indicator of the resource allocation strategy includes: constructing an objective function based on the hardware resource utilization indicator, load balancing indicator, application priority satisfaction indicator, and operational risk indicator of the resource allocation strategy; the application priority satisfaction indicator, the hardware resource utilization indicator, and the load balancing indicator are positively correlated with the total optimization objective value of the objective function; the operational risk indicator is negatively correlated with the total optimization objective value of the objective function.

[0034] To consider the health status of hardware resources in the resource allocation strategy, embodiments of the present invention may also add a runtime risk indicator to the objective function. The runtime risk indicator describes the runtime risk of the resource allocation strategy. In one embodiment, the calculation formula for the runtime risk indicator is as follows: ;Formula (8) in s k,st The health status coefficient of the k-th type of hardware resource is 0 for normal status, 0.5 for warning status, and 1 for fault status. The percentage of hardware resources of type k allocated to the i-th application, where n is the total number of applications awaiting hardware resource allocation. F risk This indicates operational risk indicators. F risk The smaller the value, the lower the operational risk. In one embodiment, the objective function is calculated using the following formula after considering the operational risk index: ;Formula (9) in, F total This represents the overall optimization objective value. F utilization This represents a hardware resource utilization indicator. F balance This represents a load balancing metric. F priority To prioritize the application and meet the metrics, F risk This represents the operational risk indicator. Electronic devices can use multi-objective optimization decision-making algorithms such as genetic algorithms and particle swarm optimization algorithms to solve the objective function of formula (9), and obtain a target resource allocation strategy that comprehensively considers the supply of key application resources, resource utilization, load balancing and operational risks. This improves the real-time performance and reliability of key applications while enhancing the operational security of hardware resources, and solves the allocation optimization problem when multiple advanced applications share hardware resources in the distribution network.

[0035] In other aspects of the embodiments of the present invention, the construction of the objective function based on the hardware resource utilization index, load balancing index, application priority satisfaction index, and operational risk index of the resource allocation strategy includes: constructing the objective function based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, the second weight corresponding to the load balancing index, the application priority satisfaction index, the third weight corresponding to the application priority satisfaction index, the operational risk index, and the fourth weight corresponding to the operational risk index.

[0036] In one embodiment, the objective function is calculated as follows: ;Formula (10) in, F total This represents the overall optimization objective value. F utilization This represents a hardware resource utilization indicator. F balance This represents a load balancing metric. F priority To prioritize the application and meet the metrics, F risk This indicates operational risk indicators. oh 1 indicates the first weight. oh 2. Second weight, oh 3 indicates the third weight. oh 4 indicates the fourth weight. oh 1+ oh 2+ oh 3+ oh 4=1. Similarly, in order to consider the different impacts of application priority satisfaction indicators, hardware resource utilization indicators, load balancing indicators, and operational risk indicators on the overall optimization objective value of the objective function, the embodiments of the present invention can add weights to the objective function for application priority satisfaction indicators, hardware resource utilization indicators, load balancing indicators, and operational risk indicators. The resource scheduling mechanism based on application priority and demand characteristics in the embodiments of the present invention ensures the real-time performance of critical applications, reduces response latency by more than 50%, and avoids application failures caused by resource conflicts.

[0037] In other aspects of the embodiments of the present invention, the hardware resource requirements corresponding to each of the distribution network applications are determined by: obtaining operational requirement data corresponding to multiple distribution network applications; and determining the hardware resource requirements corresponding to the operational requirement data of each of the distribution network applications based on a resource requirement mapping model.

[0038] To establish a precise and dynamic correspondence between the inherent characteristics of multi-source data and hardware resource capabilities, and to achieve on-demand allocation, elastic scheduling, and collaborative optimization of resources, this invention establishes a resource demand mapping model. This model includes mapping rules between operational demand data of distribution network applications and hardware resource requirements. The mapping rules can be calculation formulas that map operational demand data to hardware resource requirements. Therefore, when operational demand data corresponding to multiple distribution network applications is obtained, the hardware resource requirements corresponding to the operational demand data of each distribution network application can be obtained. The operational demand data includes data real-time requirements, data transmission rate requirements, sampling rate requirements, and data integrity requirements; the hardware resource requirements include the number of real-time cores and the number of general-purpose cores, storage bandwidth requirements, and communication bandwidth requirements.

[0039] In other aspects of the embodiments of the present invention, the step of determining the hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model includes: determining the number of real-time cores required based on the real-time requirement, the data transmission rate requirement, the data integrity requirement, a set base number of cores, and feature weights; the real-time requirement, the data transmission rate requirement, the data integrity requirement, the set base number of cores, and the feature weights are positively correlated with the number of real-time cores required; determining the number of general cores required based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight; the data transmission rate requirement, the sampling rate requirement, the sampling rate requirement, the sampling rate requirement, the sampling rate requirement, the sampling rate requirement, the sampling rate requirement, and the data integrity requirement; the data transmission rate requirement, the sampling ... The data integrity requirement, the sampling rate weight, and the integrity weight are all positively correlated with the general core requirement. The storage bandwidth requirement is determined based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient. The data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient are all positively correlated with the storage bandwidth requirement. The communication bandwidth requirement is determined based on the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient. The data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient are all positively correlated with the communication bandwidth requirement.

[0040] The operational requirements data collected for power distribution network applications can include real-time data requirements, data transmission rate requirements, sampling rate requirements, and data integrity requirements. In other embodiments, operational requirements data may also include CPU computing power requirements and storage bandwidth requirements. CPU computing power requirements must specify the base frequency requirement and peak computing power. Storage bandwidth requirements include memory requirements and storage requirements. Memory requirements must specify peak usage and read / write rate requirements; storage requirements must distinguish between capacity requirements and read / write performance. It should be noted that, focusing on the application's time attributes, real-time requirements must be accurate to specific levels, such as fault handling at the millisecond level and data report generation at the second level; the runtime segment must be indicated as continuous operation or timed operation, and the specific time period parameters must be specified.

[0041] For steady-state scenarios, embodiments of the present invention can obtain the basic requirements for various resources through weighted calculations. In one embodiment, the calculation formula for the real-time core requirement is as follows: ;Formula (11) in, R CPU-RT To determine the real-time core demand, The feature weights can be calibrated experimentally. This sets the base number of cores and is the smallest scheduling unit. For the rounding operation, the number of cores is guaranteed to be a positive integer.f 1 is for real-time requirements. f 2. For data transmission rate requirements, f 4. Data integrity requirements. Data transmission rate requirements refer to the data volume and represent the data transmission rate. The formula for calculating the number of real-time cores required shows that the higher the real-time performance, data volume, and integrity requirements, the more real-time cores are needed. In one embodiment, the real-time performance requirement... f 1. Data integrity requirements f The formula for calculating 4 is as follows: ;Formula (12) ;Formula (13) In one embodiment, the formula for calculating the general core requirement is as follows: ;Formula (14) in, R CPU-G For general core requirements, for example Sampling rate weights This is the integrity weight. f 2 represents the data transmission rate requirement. f 3 represents the sampling rate requirement. f 4. Data integrity requirements. The formula for calculating the number of general-purpose cores shows that the higher the sampling rate and the larger the data volume, the greater the computational overhead of data preprocessing, requiring more general-purpose cores.

[0042] In one embodiment, the formula for calculating storage bandwidth requirements is as follows: ;Formula (15) in, R Storage-HS To meet storage bandwidth requirements, f 2 represents the data transmission rate requirement. f 3 represents the sampling rate requirement. f 4. Data integrity requirements. Sampling rate requirements represent the data temporal resolution requirements. The formula for calculating storage bandwidth requirements shows that the larger the data volume and the higher the sampling rate, the higher the storage bandwidth is needed to prevent data loss.

[0043] In one embodiment, the formula for calculating communication bandwidth requirements is as follows: ;Formula (16) in, R Comm-LD This is for communication bandwidth requirements. This is the communication bandwidth conversion factor; multiplying by 8 converts a byte to a bit. f 1 is for real-time requirements. f2 represents the sampling rate requirement. The formula for calculating communication bandwidth requirements shows that the higher the real-time performance and the larger the data volume, the higher the required low-latency communication bandwidth.

[0044] Through the above resource demand mapping model, the embodiments of the present invention achieve efficient feature differentiation adaptation and static resource allocation for different types of data.

[0045] In other aspects of the embodiments of the present invention, the operational demand data further includes a burst coefficient; the burst coefficient represents the ratio between the current data volume and the set steady-state data volume; the hardware resource demand further includes resource expansion demand; the step of determining the hardware resource demand corresponding to the operational demand data of each distribution network application based on the resource demand mapping model further includes: when the burst coefficient is greater than or equal to a first set threshold, determining the resource expansion demand based on the minimum value comparison result between the burst coefficient and the set maximum burst coefficient, the basic resource demand, and the reserve resource replenishment amount; the minimum value comparison result, the basic resource demand, and the reserve resource replenishment amount are positively correlated with the resource expansion demand; the reserve resource replenishment amount is calculated based on the burst coefficient and the basic resource demand, and the basic resource demand and the burst coefficient are positively correlated with the reserve resource replenishment amount.

[0046] For emergency scenarios, when the emergency coefficient is... ( When the sudden threshold, i.e. the first set threshold, is typically set to 3), is reached, this embodiment of the invention triggers resource expansion, and the expanded resource expansion demand... The calculation formula is: ;Formula (17) in The basic resource requirements are obtained from static mapping. The maximum suddenness coefficient, The backup resources can consist of cloud-based elastic computing power and local redundant resources. The utilization rate coefficient for backup resources is typically set to 0.8 to avoid redundancy during expansion. Therefore, this embodiment of the invention achieves elastic resource expansion in scenarios of sudden data surges by constructing a calculation formula for resource expansion requirements.

[0047] In summary, this embodiment of the invention first defines the characteristics of multi-source data, providing an input basis for subsequent resource requirement calculations. This involves defining data feature vectors. ,in For real-time requirements, The data volume represents the data transmission rate. The sampling rate represents the temporal resolution of the data. For data integrity requirements, For the suddenness coefficient. Define the resource demand vector. ,in This refers to the number of CPU cores required in real time. For general core requirements, To meet the high-speed storage bandwidth requirements, This invention addresses the bandwidth requirements for low-latency communication. Embodiments of the invention also define resource expansion requirements. Then, the hardware resource requirements corresponding to the operation requirement data of each distribution network application are determined through formulas (11), (14), (15), (16), and (17). This embodiment of the invention establishes a full-stack, autonomous, and controllable system architecture, designs a mapping model between multi-source data characteristics and resource requirements, deeply binds data processing requirements with resource allocation, and improves the overall utilization rate of hardware resources through hardware resource sharing and dynamic allocation.

[0048] In other aspects of this invention, after solving the objective function to obtain the target resource allocation strategy corresponding to the plurality of distribution network applications, the method further includes: forming an operation instruction based on the target resource allocation strategy, and sending the operation instruction to the hardware of the distribution network; when the hardware executes the target resource allocation strategy, acquiring the application operation status of the plurality of distribution network applications and the hardware resource operation status of the distribution network; determining the update hardware resource requirements corresponding to the plurality of distribution network applications and the update hardware resource status data of the distribution network based on the comparison result of the application operation status and a second set threshold, and the comparison result of the hardware resource operation status and a third set threshold; constructing an update objective function based on the update hardware resource requirements corresponding to the plurality of distribution network applications and the update hardware resource status data of the distribution network; the update objective function and the objective function have the same function expression; solving the update objective function to obtain the update resource allocation strategy corresponding to the plurality of distribution network applications.

[0049] This invention receives a target resource allocation strategy, parses it into hardware-recognizable operation instructions, and sends them to the corresponding hardware device through a standardized hardware driver interface. The device then completes the allocation and configuration of various resources according to the instructions. For example, it binds a specified application process to a specific CPU core, closes unnecessary processes on that core to ensure dedicated computing power; establishes a dedicated path mapping between the application and the storage device, sets storage permissions, and ensures the security and efficiency of data read and write; allocates a dedicated communication bandwidth channel to the application, limits the bandwidth usage of low-priority applications, and ensures the communication quality of high-priority applications. After resource configuration is completed, the configuration status of the hardware device is immediately collected and compared with the target in the resource adjustment instructions to confirm whether the execution was successful. If the execution fails, feedback is immediately sent to the decision-making unit to trigger a re-decision. This invention also establishes a two-way feedback channel, simultaneously collecting the application running status and the hardware resource running status: the application running status is collected every 50ms, including indicators such as application response time, task completion rate, and data transmission success rate; the hardware resource running status, such as the occupancy rate, remaining amount, and health status after resource allocation, is collected synchronously to verify whether resource utilization is reasonable. The collected data is compared with preset thresholds (a second threshold and a third threshold) to determine whether the resource allocation effect meets the standards. The evaluation results are pushed to the input layer and the core layer in real time to update the application requirement list and resource status markers, that is, to update the hardware resource requirements and the updated hardware resource status data of the distribution network. This provides an optimization basis for the next round of decision-making. Based on the updated hardware resource requirements and updated hardware resource status data of the multiple distribution network applications, this embodiment of the invention constructs an update objective function; the update objective function and the objective function have the same functional expression; solving the update objective function yields the updated resource allocation strategy corresponding to the multiple distribution network applications.

[0050] This invention's resource scheduling mechanism, based on application priority and demand characteristics, accurately captures application needs and resource status, relying on intelligent algorithms to achieve optimal resource allocation and real-time adjustment. The entire scheduling mechanism consists of three layers: an input layer, a core layer, and an execution layer, ensuring efficient and adaptable resource allocation.

[0051] The input layer primarily realizes the perception of application requirements and resource status, serving as the foundation of the resource scheduling mechanism. By comprehensively and accurately collecting core data from the application end and hardware resource end, it provides a reliable basis for subsequent decisions. This stage includes two processes: application requirement analysis and resource status perception (see step 100 for details). The core layer is the brain of the resource scheduling mechanism. Based on the application requirement list and resource status markers (hardware resource requirements corresponding to multiple distribution network applications and hardware resource status data of the distribution network) provided by the input layer, the above objective function (e.g., one of the above formulas (3), (4), (6), (7), (9), (10)) is constructed. The objective function is solved using multi-objective optimization decision-making algorithms such as improved genetic algorithm and particle swarm optimization algorithm. Under the premise of satisfying multiple constraints, the optimal resource allocation scheme is output. The execution layer is the key link in implementing the decision scheme. Through hardware-driven and status feedback mechanisms, it realizes closed-loop management of resource allocation, ensuring that the resource allocation effect meets expectations.

[0052] In one embodiment, the method for redetermining the resource allocation strategy based on runtime status feedback is as follows: (a) Hardware resource pool configuration CPU: 4-core industrial-grade processor, with 2 real-time cores for high-priority applications and 2 general-purpose cores for ordinary applications; Memory: 8GB DDR4, supports dynamic partition allocation; Storage: 128GB SSD for real-time data; and 2TB HDD for historical data storage; Communication interfaces: Ethernet, 4G / 5G, RS485, etc., supporting dynamic bandwidth allocation.

[0053] (II) Application Scenarios and Allocation Process Assume that the power distribution network operation scenario includes three types of core applications: P1 level local intelligent protection application (ensuring power grid safety, real-time requirement ≤50ms), P3 level load monitoring application (statistical power consumption data, real-time requirement ≤1s), and P5 level system log backup application (storing historical data, no strict real-time requirement).

[0054] Step 1: Initial resource allocation and operational status data collection 1. The initial resource allocation strategy is as follows: Local intelligent protection application: 1 real-time core, 2GB memory, 10% communication bandwidth, 30GB SSD storage; Load monitoring application: 1 general-purpose core, 3GB memory, 20% communication bandwidth, 50GB HDD storage; System log backup application: 1 general-purpose core, 2GB memory, 5% communication bandwidth, 100GB HDD storage.

[0055] 2. Running status data collection (execution layer data collection at 50ms intervals): Application running status: Local intelligent protection application: response time 35ms (meets the standard, ≤50ms), task completion rate 100% (meets the standard); Load monitoring application: response time 800ms (meets the standard, ≤1s), data transmission success rate 98% (does not meet the standard, preset threshold ≥99%). System log backup application: Task completion rate 95% (meets the standard, no strict threshold).

[0056] Resource running status: CPU: Real-time core utilization 60% (1 real-time core idle), General Core 1 (load monitoring) utilization 75%, General Core 2 (log backup) utilization 40%; Memory: 7GB used, 1GB remaining (the load monitoring application actually uses 3.2GB, exceeding the allocated 3GB); Communication interface: 35% of total bandwidth is used, 65% remaining (the load monitoring application actually uses 22%, exceeding the allocated 20%). Storage: SSD usage 30% (compliant), HDD usage 15% (compliant).

[0057] Step 2: Analysis of Evaluation Results 1. The load monitoring application has a mismatch between resource requirements and allocation: the actual memory usage is 3.2GB (0.2GB over the allocated amount) and the communication bandwidth usage is 22% (2% over the allocated amount), resulting in a data transmission success rate of 98% (not up to standard). The allocation of memory and communication bandwidth needs to be increased. 2. The system log backup application has low resource utilization: the general core utilization is only 40%, and the memory usage is 2GB, but the task pressure is low, and there is resource redundancy; 3. The hardware resource pool has adjustable capacity: 1GB of memory and 65% of communication bandwidth are idle, and redundant resources can be reclaimed from low-priority applications to supplement medium-priority applications.

[0058] Step 3: Update the application requirements list and resource status flags (update hardware resource requirements and update hardware resource status data). 1. The local intelligent protection application is operating as required, and the demand remains unchanged; 2. Insufficient memory / bandwidth in the load monitoring application resulted in a failure to meet transmission success rates. The requirements have been increased to 1 general-purpose core, 3.5GB of memory, and 25% of the communication bandwidth. 3. The system log backup application has low resource utilization. 0.5GB of memory can be freed up for high-priority applications and replaced with 1 general-purpose core, 1.5GB of memory, and 5% of communication bandwidth.

[0059] 4. The resource status marker update combines "current resource usage + post-assessed demand" to simultaneously mark the remaining resource quantity, health status, and adjustable priority.

[0060] Step 4: Update the results to support the next round of decision-making. The updated application requirement list and resource status flags (updated hardware resource requirements and updated hardware resource status data) are pushed to the input layer. Within a 100ms refresh cycle, the core layer initiates the decision-making algorithms from steps 100 to 300 above, optimizing the allocation scheme based on the new data. 1. Reclaim 0.5GB of memory from the system log backup application and add it to the load monitoring application to meet its 3.5GB memory requirement; 2. Increase the communication bandwidth from 20% to 25% for load monitoring applications to ensure a data transmission success rate of ≥99%; 3. Mark the utilization rate of General Core 1 (load monitoring) at 75% (close to the threshold of 70%). In the next cycle, focus on monitoring it. If it continues to rise, trigger load migration (such as migrating the log backup application to General Core 2 to reduce the load on General Core 1).

[0061] The significant differences in the technical characteristics of various applications lead to distinctly different hardware resource requirements. Local intelligent protection and control applications are typical real-time priority scenarios, with millisecond-level fault clearing requirements placing stringent demands on the CPU's real-time computing power and response speed. Real-time monitoring applications of power grid operation status rely on high-frequency data transmission and real-time analysis, placing high demands on the bandwidth and stability of communication interfaces. Accurate fault location and isolation applications rely on the collaborative processing of multi-source data, requiring a balanced allocation of CPU, memory, and communication resources. How to solve the challenges of task scheduling, computing resource allocation, and data synchronization when multiple advanced applications share the same hardware resources, achieving deep integration of "one set of hardware for multiple advanced applications," is one of the challenges currently facing the distribution network. This invention proposes a dynamic allocation method for hardware resources in distribution networks with multiple advanced applications. It establishes a full-stack, independently controllable system architecture based on domestically produced main control chips, storage, communication, and analog chips. It designs a mapping model between multi-source data characteristics and resource requirements, proposes a resource scheduling mechanism based on application priority and demand characteristics, and combines application demand characteristics with the power grid operating status to achieve intelligent dynamic allocation of hardware resources. This ensures the real-time performance and reliability of critical applications and solves the allocation optimization problem when multiple advanced applications in a distribution network share hardware resources.

[0062] Device Examples Please refer to Figure 2 On the other hand, embodiments of the present invention also provide a hardware resource dynamic allocation device, including: The first acquisition module 201 is used to acquire the hardware resource requirements and hardware resource status data of the distribution network corresponding to multiple distribution network applications. The first construction module 202 is used to construct an objective function based on the hardware resource requirements of multiple distribution network applications and the hardware resource status data; the objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications. The first solution module 203 is used to solve the objective function to obtain the target resource allocation strategy corresponding to the multiple power distribution network applications.

[0063] This invention, through constructing an objective function that characterizes at least the hardware resource utilization and load balancing of resource allocation strategies corresponding to multiple distribution network applications, and solving the objective function, yields the optimal target resource allocation strategy for hardware resource utilization and load balancing for multiple distribution network applications. Therefore, this invention can improve the hardware resource utilization and load balancing of distribution network hardware devices corresponding to multiple distribution network applications, reduce hardware redundancy, and solve the problem that most existing hardware resource allocation strategies for distribution network applications adopt a "one application, one hardware" model, resulting in low hardware resource utilization.

[0064] In other aspects of the embodiments of the present invention, the step of constructing an objective function based on the hardware resource requirements of multiple distribution network applications and the hardware resource status data includes: Based on the hardware resource requirements and hardware resource status data of multiple power distribution network applications, a set of resource allocation strategies is determined; the resource allocation strategies include multiple resource allocation strategies and hardware resource utilization indicators and load balancing indicators for each resource allocation strategy. Based on the hardware resource utilization index and the load balancing index, an objective function is constructed; the hardware resource utilization index and the load balancing index are positively correlated with the total optimization objective value of the objective function.

[0065] In other aspects of the embodiments of the present invention, the step of constructing an objective function based on the hardware resource utilization index and the load balancing index includes: Based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, and the second weight corresponding to the load balancing index, an objective function is constructed.

[0066] In other aspects of the embodiments of the present invention, the resource allocation strategy set further includes an application priority satisfaction index for each resource allocation strategy; the application priority satisfaction index characterizes the resource supply level of key applications in the resource allocation strategy; the construction of the objective function based on the hardware resource utilization index and the load balancing index includes: Based on the hardware resource utilization rate, load balancing rate, and application priority satisfaction rate of the resource allocation strategy, an objective function is constructed; the application priority satisfaction rate, the hardware resource utilization rate, and the load balancing rate are all positively correlated with the total optimization objective value of the objective function.

[0067] In other aspects of the embodiments of the present invention, the construction of the objective function based on the hardware resource utilization index, load balancing index, and application priority satisfaction index of the resource allocation strategy includes: Based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, the second weight corresponding to the load balancing index, the application priority satisfaction index, and the third weight corresponding to the application priority satisfaction index, an objective function is constructed.

[0068] In other aspects of the embodiments of the present invention, the resource allocation strategy further includes operational risk indicators for each resource allocation strategy; the objective function constructed based on the hardware resource utilization indicator, load balancing indicator, and application priority satisfaction indicator of the resource allocation strategy includes: Based on the hardware resource utilization rate, load balancing rate, application priority satisfaction rate, and operational risk rate of the resource allocation strategy, an objective function is constructed. The application priority satisfaction index, the hardware resource utilization index, and the load balancing index are all positively correlated with the overall optimization target value of the objective function; the operational risk index is negatively correlated with the overall optimization target value of the objective function.

[0069] In other aspects of the embodiments of the present invention, the construction of the objective function based on the hardware resource utilization index, load balancing index, application priority satisfaction index, and operational risk index of the resource allocation strategy includes: An objective function is constructed based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, the second weight corresponding to the load balancing index, the application priority satisfaction index, the third weight corresponding to the application priority satisfaction index, the operation risk index, and the fourth weight corresponding to the operation risk index.

[0070] In other aspects of the embodiments of the present invention, the hardware resource requirements corresponding to each of the power distribution network applications are determined in the following manner: Obtain operational requirement data for multiple power distribution network applications; The hardware resource requirements corresponding to the operation requirement data of each distribution network application are determined based on the resource requirement mapping model; the resource requirement mapping model includes the mapping rules between the operation requirement data of the distribution network application and the hardware resource requirements.

[0071] In other aspects of the embodiments of the present invention, the operational requirements data include data real-time requirements, data transmission rate requirements, sampling rate requirements, and data integrity requirements; the hardware resource requirements include the number of real-time cores required, the number of general-purpose cores required, storage bandwidth requirements, and communication bandwidth requirements; The determination of hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model includes: Based on the real-time requirements, data transmission rate requirements, data integrity requirements, a set base number of cores, and feature weights, the required number of real-time cores is determined; the real-time requirements, data transmission rate requirements, data integrity requirements, the set base number of cores, and the feature weights are all positively correlated with the required number of real-time cores. Based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight, the general core requirement number is determined; the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight are all positively correlated with the general core requirement number. Based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient, the storage bandwidth requirement is determined; the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient are all positively correlated with the storage bandwidth requirement. The communication bandwidth requirement is determined based on the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient; the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient are all positively correlated with the communication bandwidth requirement.

[0072] In other aspects of this invention, the operational requirement data further includes a burst coefficient; the burst coefficient represents the ratio between the current data volume and the set steady-state data volume; the hardware resource requirements also include resource expansion requirements; The step of determining the hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model also includes: When the burst coefficient is greater than or equal to a first set threshold, the resource expansion requirement is determined based on the minimum value comparison result between the burst coefficient and the set maximum burst coefficient, the basic resource requirement, and the reserve resource replenishment amount; the minimum value comparison result, the basic resource requirement, and the reserve resource replenishment amount are all positively correlated with the resource expansion requirement; the reserve resource replenishment amount is calculated based on the burst coefficient and the basic resource requirement, and the basic resource requirement and the burst coefficient are all positively correlated with the reserve resource replenishment amount.

[0073] In other aspects of embodiments of the present invention, the apparatus further includes: The sending module is used to generate operation instructions based on the target resource allocation strategy and to send the operation instructions to the distribution network. The second acquisition module is used to acquire the application operation status of multiple power distribution network applications and the hardware resource operation status of the power distribution network when the hardware executes the target resource allocation strategy. The determination module is used to determine the update hardware resource requirements corresponding to the multiple distribution network applications and the update hardware resource status data of the distribution network based on the comparison results of the application running status and the second set threshold, and the comparison results of the hardware resource running status and the third set threshold. The second construction module is used to construct an update objective function based on the update hardware resource requirements corresponding to the multiple distribution network applications and the update hardware resource status data of the distribution network; the update objective function and the objective function have the same function expression; The second solution module is used to solve the update objective function to obtain the update resource allocation strategy corresponding to the multiple distribution network applications.

[0074] The hardware resource dynamic allocation device includes a processor and a memory. The first acquisition module 201, the first construction module 202, and the first solution module 203 are all stored in the memory as program units. The processor executes the program units stored in the memory to realize the corresponding functions.

[0075] A processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured.

[0076] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0077] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a dynamic hardware resource allocation method. This method includes: acquiring hardware resource requirements and hardware resource status data of multiple distribution network applications; constructing an objective function based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications; the objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications; and solving the objective function to obtain the target resource allocation strategy corresponding to the multiple distribution network applications.

[0078] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0079] In other embodiments, the electronic device of the present invention employs a full-stack autonomous and controllable system architecture as follows: Figure 4 As shown, the architecture adopts a unified hardware resource pool and software application modules, which mainly includes three parts: a unified hardware resource pool (or hardware resource pool), a resource scheduling module, and an application adaptation interface.

[0080] The unified hardware resource pool integrates multi-core CPUs (distinguishing between real-time and general-purpose cores), scalable memory, distributed storage, and multi-protocol communication interfaces, serving as the hardware foundation shared by multiple applications. The resource scheduling module is the core execution unit, executing the dynamic allocation method for hardware resources described in the above method embodiments. It is responsible for the real-time perception, evaluation, allocation, and adjustment of resources, and includes an application requirement parsing unit, a resource status perception unit, a scheduling decision unit, and a resource allocation execution unit. The application adaptation interface provides standardized interfaces, encapsulating advanced applications such as local protection, real-time monitoring, and fault location into dynamically loadable software modules, thereby decoupling applications from hardware resources.

[0081] This invention establishes a fully independent and controllable system architecture based on domestically produced main control chips, storage, communication, and analog chips. It designs a mapping model between multi-source data characteristics and resource requirements, proposes a resource scheduling mechanism based on application priority and demand characteristics, and combines application demand characteristics with power grid operation status to achieve intelligent dynamic allocation of hardware resources, ensuring the real-time performance and reliability of critical applications, and solving the allocation optimization problem when multiple advanced applications in the distribution network share hardware resources.

[0082] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a machine-readable storage medium. When the computer program is executed by a processor, the computer can execute a dynamic hardware resource allocation method. The method includes: acquiring hardware resource requirements and hardware resource status data of multiple distribution network applications; constructing an objective function based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications; the objective function at least characterizing the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications; and solving the objective function to obtain the target resource allocation strategy corresponding to the multiple distribution network applications.

[0083] In another aspect, the present invention also provides a machine-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for dynamically allocating hardware resources. The method includes: acquiring hardware resource requirements and hardware resource status data of multiple distribution network applications; constructing an objective function based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications; the objective function at least characterizing the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications; and solving the objective function to obtain the target resource allocation strategy corresponding to the multiple distribution network applications.

[0084] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0085] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamically allocating hardware resources, characterized in that, include: Obtain hardware resource requirements and hardware resource status data of multiple distribution network applications; Based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications, an objective function is constructed; the objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategies corresponding to the multiple distribution network applications. Solving the objective function yields the target resource allocation strategy corresponding to the multiple power distribution network applications.

2. The method for dynamic allocation of hardware resources according to claim 1, characterized in that, The objective function is constructed based on the hardware resource requirements and hardware resource status data of multiple power distribution network applications, including: Based on the hardware resource requirements and hardware resource status data of multiple power distribution network applications, a set of resource allocation strategies is determined; the resource allocation strategies include multiple resource allocation strategies and hardware resource utilization indicators and load balancing indicators for each resource allocation strategy. Based on the hardware resource utilization index and the load balancing index, an objective function is constructed; the hardware resource utilization index and the load balancing index are positively correlated with the total optimization objective value of the objective function.

3. The method for dynamic allocation of hardware resources according to claim 2, characterized in that, The objective function is constructed based on the hardware resource utilization index and load balancing index, including: Based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, and the second weight corresponding to the load balancing index, an objective function is constructed.

4. The method for dynamic allocation of hardware resources according to claim 2, characterized in that, The resource allocation strategy set also includes an application priority satisfaction index for each resource allocation strategy; the application priority satisfaction index characterizes the resource supply level of key applications in the resource allocation strategy. The objective function is constructed based on the hardware resource utilization index and load balancing index, including: Based on the hardware resource utilization rate, load balancing rate, and application priority satisfaction rate of the resource allocation strategy, an objective function is constructed. The application priority satisfaction index, the hardware resource utilization index, and the load balancing index are all positively correlated with the total optimization objective value of the objective function.

5. The method for dynamic allocation of hardware resources according to claim 4, characterized in that, The objective function is constructed based on the hardware resource utilization index, load balancing index, and application priority satisfaction index according to the resource allocation strategy, including: Based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, the second weight corresponding to the load balancing index, the application priority satisfaction index, and the third weight corresponding to the application priority satisfaction index, an objective function is constructed.

6. The method for dynamic allocation of hardware resources according to claim 4, characterized in that, The resource allocation strategy also includes operational risk indicators for each resource allocation strategy; the objective function is constructed based on the hardware resource utilization rate indicator, load balancing indicator, and application priority satisfaction indicator of the resource allocation strategy, including: Based on the hardware resource utilization rate, load balancing rate, application priority satisfaction rate, and operational risk rate of the resource allocation strategy, an objective function is constructed. The application priority satisfaction index, the hardware resource utilization index, and the load balancing index are all positively correlated with the overall optimization target value of the objective function; the operational risk index is negatively correlated with the overall optimization target value of the objective function.

7. The method for dynamic allocation of hardware resources according to claim 6, characterized in that, The objective function is constructed based on the hardware resource utilization rate, load balancing rate, application priority satisfaction rate, and operational risk rate according to the resource allocation strategy, including: An objective function is constructed based on the hardware resource utilization index, the first weight corresponding to the hardware resource utilization index, the load balancing index, the second weight corresponding to the load balancing index, the application priority satisfaction index, the third weight corresponding to the application priority satisfaction index, the operation risk index, and the fourth weight corresponding to the operation risk index.

8. The method for dynamic allocation of hardware resources according to claim 1, characterized in that, The hardware resource requirements for each of the aforementioned power distribution network applications are determined in the following manner: Obtain operational requirement data for multiple power distribution network applications; The hardware resource requirements corresponding to the operation requirement data of each distribution network application are determined based on the resource requirement mapping model; the resource requirement mapping model includes the mapping rules between the operation requirement data of the distribution network application and the hardware resource requirements.

9. The method for dynamic allocation of hardware resources according to claim 8, characterized in that, The operational requirements include data real-time requirements, data transmission rate requirements, sampling rate requirements, and data integrity requirements; the hardware resource requirements include the number of real-time cores required, the number of general-purpose cores required, storage bandwidth requirements, and communication bandwidth requirements. The determination of hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model includes: Based on the real-time requirements, data transmission rate requirements, data integrity requirements, a set base number of cores, and feature weights, the required number of real-time cores is determined; the real-time requirements, data transmission rate requirements, data integrity requirements, the set base number of cores, and the feature weights are all positively correlated with the required number of real-time cores. Based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight, the general core requirement number is determined; the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight are all positively correlated with the general core requirement number. Based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient, the storage bandwidth requirement is determined; the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient are all positively correlated with the storage bandwidth requirement. The communication bandwidth requirement is determined based on the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient; the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient are all positively correlated with the communication bandwidth requirement.

10. The method for dynamic allocation of hardware resources according to claim 9, characterized in that, The operational requirements data also include a burst coefficient; the burst coefficient represents the ratio of the current data volume to the set steady-state data volume; The hardware resource requirements also include resource expansion requirements; The step of determining the hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model also includes: If the burst coefficient is greater than or equal to a first set threshold, the resource expansion requirement is determined based on the minimum value comparison result between the burst coefficient and the set maximum burst coefficient, the basic resource requirement, and the amount of backup resource replenishment. The minimum value comparison result, the basic resource requirement, and the reserve resource replenishment amount are all positively correlated with the resource expansion requirement; the reserve resource replenishment amount is calculated based on the burst coefficient and the basic resource requirement, and the basic resource requirement and the burst coefficient are all positively correlated with the reserve resource replenishment amount.

11. The method for dynamic allocation of hardware resources according to claim 1, characterized in that, After solving the objective function to obtain the target resource allocation strategies corresponding to the multiple distribution network applications, the method further includes: The hardware that generates operation instructions based on the target resource allocation strategy and sends the operation instructions to the distribution network; When the hardware executes the target resource allocation strategy, the application operation status of multiple power distribution network applications and the hardware resource operation status of the power distribution network are obtained. Based on the comparison results of the application running status and the second set threshold, and the comparison results of the hardware resource running status and the third set threshold, the update hardware resource requirements corresponding to the multiple distribution network applications and the update hardware resource status data of the distribution network are determined. Based on the update hardware resource requirements corresponding to the multiple distribution network applications and the update hardware resource status data of the distribution network, an update objective function is constructed; the update objective function and the objective function have the same function expression; Solving the update objective function yields the update resource allocation strategy corresponding to the multiple distribution network applications.

12. A hardware resource dynamic allocation device, characterized in that, include: The first acquisition module is used to acquire the hardware resource requirements and hardware resource status data of the distribution network corresponding to multiple distribution network applications. The first construction module is used to construct an objective function based on the hardware resource requirements and hardware resource status data of the multiple distribution network applications; the objective function at least characterizes the hardware resource utilization and load balancing of the resource allocation strategy corresponding to the multiple distribution network applications. The first solution module is used to solve the objective function to obtain the target resource allocation strategy corresponding to the multiple distribution network applications.

13. The hardware resource dynamic allocation device according to claim 12, characterized in that, The objective function is constructed based on the hardware resource requirements and hardware resource status data of multiple power distribution network applications, including: Based on the hardware resource requirements and hardware resource status data of multiple power distribution network applications, a set of resource allocation strategies is determined; the resource allocation strategies include multiple resource allocation strategies and hardware resource utilization indicators and load balancing indicators for each resource allocation strategy. Based on the hardware resource utilization index and the load balancing index, an objective function is constructed; the hardware resource utilization index and the load balancing index are positively correlated with the total optimization objective value of the objective function.

14. The hardware resource dynamic allocation device according to claim 13, characterized in that, The resource allocation strategy set also includes an application priority satisfaction index for each resource allocation strategy; the application priority satisfaction index characterizes the resource supply level of key applications in the resource allocation strategy. The objective function is constructed based on the hardware resource utilization index and load balancing index, including: Based on the hardware resource utilization rate, load balancing rate, and application priority satisfaction rate of the resource allocation strategy, an objective function is constructed. The application priority satisfaction index, the hardware resource utilization index, and the load balancing index are all positively correlated with the total optimization objective value of the objective function.

15. The hardware resource dynamic allocation device according to claim 14, characterized in that, The resource allocation strategy also includes operational risk indicators for each resource allocation strategy; the objective function is constructed based on the hardware resource utilization rate indicator, load balancing indicator, and application priority satisfaction indicator of the resource allocation strategy, including: Based on the hardware resource utilization rate, load balancing rate, application priority satisfaction rate, and operational risk rate of the resource allocation strategy, an objective function is constructed. The application priority satisfaction index, the hardware resource utilization index, and the load balancing index are all positively correlated with the overall optimization target value of the objective function; the operational risk index is negatively correlated with the overall optimization target value of the objective function.

16. The hardware resource dynamic allocation device according to claim 12, characterized in that, The hardware resource requirements for each of the aforementioned power distribution network applications are determined in the following manner: Obtain operational requirement data for multiple power distribution network applications; The hardware resource requirements corresponding to the operation requirement data of each distribution network application are determined based on the resource requirement mapping model; the resource requirement mapping model includes the mapping rules between the operation requirement data of the distribution network application and the hardware resource requirements.

17. The hardware resource dynamic allocation device according to claim 16, characterized in that, The operational requirements include data real-time requirements, data transmission rate requirements, sampling rate requirements, and data integrity requirements; the hardware resource requirements include the number of real-time cores required, the number of general-purpose cores required, storage bandwidth requirements, and communication bandwidth requirements. The determination of hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model includes: Based on the real-time requirements, data transmission rate requirements, data integrity requirements, a set base number of cores, and feature weights, the required number of real-time cores is determined; the real-time requirements, data transmission rate requirements, data integrity requirements, the set base number of cores, and the feature weights are all positively correlated with the required number of real-time cores. Based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight, the general core requirement number is determined; the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, the sampling rate weight, and the integrity weight are all positively correlated with the general core requirement number. Based on the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient, the storage bandwidth requirement is determined; the data transmission rate requirement, the sampling rate requirement, the data integrity requirement, and the storage adaptation coefficient are all positively correlated with the storage bandwidth requirement. The communication bandwidth requirement is determined based on the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient; the data real-time requirement, the data transmission rate requirement, and the communication bandwidth conversion coefficient are all positively correlated with the communication bandwidth requirement.

18. The hardware resource dynamic allocation device according to claim 17, characterized in that, The operational requirements data also include a burst coefficient; the burst coefficient represents the ratio of the current data volume to the set steady-state data volume; The hardware resource requirements also include resource expansion requirements; The step of determining the hardware resource requirements corresponding to the operational requirement data of each distribution network application based on the resource requirement mapping model also includes: If the burst coefficient is greater than or equal to a first set threshold, the resource expansion requirement is determined based on the minimum value comparison result between the burst coefficient and the set maximum burst coefficient, the basic resource requirement, and the amount of backup resource replenishment. The minimum value comparison result, the basic resource requirement, and the reserve resource replenishment amount are all positively correlated with the resource expansion requirement; the reserve resource replenishment amount is calculated based on the burst coefficient and the basic resource requirement, and the basic resource requirement and the burst coefficient are all positively correlated with the reserve resource replenishment amount.

19. The hardware resource dynamic allocation device according to claim 12, characterized in that, The device further includes: The sending module is used to generate operation instructions based on the target resource allocation strategy and to send the operation instructions to the distribution network. The second acquisition module is used to acquire the application operation status of multiple power distribution network applications and the hardware resource operation status of the power distribution network when the hardware executes the target resource allocation strategy. The determination module is used to determine the update hardware resource requirements corresponding to the multiple distribution network applications and the update hardware resource status data of the distribution network based on the comparison results of the application running status and the second set threshold, and the comparison results of the hardware resource running status and the third set threshold. The second construction module is used to construct an update objective function based on the update hardware resource requirements corresponding to the multiple distribution network applications and the update hardware resource status data of the distribution network; the update objective function and the objective function have the same function expression; The second solution module is used to solve the update objective function to obtain the update resource allocation strategy corresponding to the multiple distribution network applications.

20. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the hardware resource dynamic allocation method according to any one of claims 1 to 11.

21. A machine-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the hardware resource dynamic allocation method according to any one of claims 1 to 11.

22. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the hardware resource dynamic allocation method according to any one of claims 1 to 12.

Images