Cloud computing-based computing resource scheduling method and system

Abstract

The invention relates to the technical field of cloud computing, and discloses a computing resource scheduling method and system based on cloud computing, and the method comprises the steps: carrying out the preprocessing of power data collected by each node of a power grid system in real time, improving the data transmission capability, and carrying out the resource demand prediction of each cloud computing node, and cloud computing resource allocation of each cloud computing node is optimized by using a resource demand prediction result and taking minimization of the waste amount of computing resources and minimization of system response time as optimization targets, so that resource demand changes can be sensed and responded in real time, and efficient utilization of resources is ensured. The utilization efficiency and the response speed of dynamic allocation computing resources are improved, and the stability and the reliability of system operation are ensured.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to a computing resource scheduling method and system based on cloud computing. Background Art

[0002] With the continuous growth of global energy demand and the rapid development of power technology, the scale of the power system is expanding at an unprecedented rate, covering an increasingly wide area, and the types and number of equipment involved are increasing dramatically. This trend has accelerated the transformation of the power grid towards intelligence and automation. Among them, intelligent collection devices, as the core of the perception layer, are spread throughout all links of power generation, transmission, transformation, distribution and consumption. They can collect various types of data in the power system in real time and accurately, such as voltage, current, power factor and other key parameters. At the same time, the data processing device, as the brain of intelligent analysis, is responsible for converting massive raw data into valuable information, providing decision support for operation monitoring, fault warning, and optimized scheduling of the power system.

[0003] However, with the large-scale deployment of intelligent data collection devices and data processing devices, and the continuous expansion of data transmission lines, the amount of data information generated in the power grid has shown an explosive growth trend, and the traditional data processing model can no longer meet the needs of high efficiency, real-time and accuracy. In this context, cloud computing technology, with its powerful computing power, flexible resource allocation and high scalability, has gradually become a key way to solve the huge data processing problems of the power system.

[0004] It is not easy to effectively apply cloud computing technology to power systems. First, the servers in the power system are widely distributed, not only covering every corner of cities and villages, but also across different geographical regions and time zones. Second, the operating characteristics of the power system determine that its demand for computing resources is highly dynamic and uncertain. For example, during peak hours, the demand for computing resources will rise sharply, while during low hours, it may drop sharply. However, the current cloud computing scheduling algorithm is difficult to perceive and respond to such changes in real time, and it is difficult to ensure the efficient use of resources, resulting in low utilization efficiency of the dynamic allocation of computing resources by the cloud computing scheduling algorithm, and it is difficult to ensure the stability and reliability of the system operation. Summary of the invention

[0005] In view of this, the present invention provides a computing resource scheduling method and system based on cloud computing, which solves the technical problems that the current cloud computing scheduling algorithm is difficult to perceive and respond to such changes in real time, and difficult to ensure efficient utilization of resources, resulting in low utilization efficiency of dynamically allocated computing resources in the cloud computing scheduling algorithm, and difficulty in ensuring the stability and reliability of system operation.

[0006] A first aspect of the present invention provides a computing resource scheduling method based on cloud computing, comprising:

[0007] Upload the power data collected in real time by each node of the power grid system to the edge computing node, and pre-process the power data through the edge computing node;

[0008] By uploading the pre-processed power data to each cloud computing node in the cloud platform;

[0009] By predicting the resource demand of each cloud computing node, and using the resource demand prediction results to optimize the cloud computing resource allocation of each cloud computing node with the optimization goals of minimizing the waste of computing resources and minimizing the system response time;

[0010] The cloud computing resources are allocated and optimized to execute computing tasks on the power data by using each cloud computing node.

[0011] Preferably, the preprocessing includes data denoising and data compression.

[0012] Preferably, the step of uploading the pre-processed power data to each cloud computing node in the cloud platform includes:

[0013] Encrypting the preprocessed power data to obtain encrypted power data;

[0014] The encrypted power data is uploaded to each cloud computing node in the cloud platform through a multi-level cache strategy.

[0015] Preferably, the process of uploading the encrypted power data to the cloud platform adopts the transmission control protocol to upload data.

[0016] Preferably, the step of performing resource demand prediction on each cloud computing node and optimizing the cloud computing resource allocation of each cloud computing node by using the resource demand prediction result with minimizing the waste of computing resources and minimizing the system response time as the optimization goal includes:

[0017] The historical load data of the power system is used to predict the load data at future times through the LSTM neural network model, and the resource demand prediction results are determined based on the load data prediction results at future times;

[0018] Based on the resource demand prediction results, a multi-objective optimization model is constructed with minimizing the waste of computing resources and minimizing the system response time as the optimization goals. The multi-objective optimization model is:

[0019]

[0020] In the formula, and are the control weight coefficients of resource waste and response time respectively; is the computing resources actually allocated to the ith cloud computing node at time t; N is the number of nodes; Predict results for load data; is the time interval; t is the current time period; represents the response time of the i-th cloud computing node;

[0021] The multi-objective optimization model is optimized and solved to obtain an optimization solution for the computing resources allocated to each cloud computing node.

[0022] Preferably, the method further comprises:

[0023] Real-time monitoring of the actual resource demand data of each cloud computing node;

[0024] For each cloud computing node, determining a resource demand prediction error according to the current actual resource demand data and the resource demand prediction result;

[0025] Generate computing resource adjustment data through a PID controller according to the resource demand prediction error;

[0026] The computing resources of the cloud computing nodes are dynamically adjusted according to the computing resource adjustment data.

[0027] Preferably, after the step of dynamically adjusting the computing resources of the cloud computing node according to the computing resource adjustment data, the step further includes:

[0028] Determine, according to the computing resource adjustment data and the resource demand prediction error, an adaptive adjustment model with minimization of the sum of load error and resource consumption as an optimization goal;

[0029] The computing resource adjustment data is optimized according to the adaptive adjustment model.

[0030] In a second aspect, the present invention further provides a computing resource scheduling system based on cloud computing, comprising:

[0031] A data preprocessing module is used to upload the power data collected in real time by each node of the power grid system to the edge computing node, and preprocess the power data through the edge computing node;

[0032] A data upload module is used to upload the pre-processed power data to each cloud computing node in the cloud platform;

[0033] A resource allocation module is used to predict the resource demand of each cloud computing node and optimize the cloud computing resource allocation of each cloud computing node by using the resource demand prediction results to minimize the waste of computing resources and minimize the system response time as the optimization goals;

[0034] The computing execution module is used to perform computing tasks on the power data using each cloud computing node after optimization of cloud computing resource allocation.

[0035] In a third aspect, the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the computing resource scheduling method based on cloud computing as described in the first aspect.

[0036] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of the computing resource scheduling method based on cloud computing as described in the first aspect are implemented.

[0037] It can be seen from the above technical solutions that the present invention improves the data transmission capacity by pre-processing the power data collected in real time by each node of the power grid system, and predicts the resource demand of each cloud computing node. The resource demand prediction results are used to minimize the waste of computing resources and minimize the system response time. The cloud computing resource allocation of each cloud computing node is optimized, thereby being able to perceive and respond to changes in resource demand in real time, ensuring efficient use of resources, improving the utilization efficiency and response speed of dynamically allocated computing resources, and ensuring the stability and reliability of system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 An application environment for a computing resource scheduling method based on cloud computing provided in an embodiment of the present invention;

[0039] Figure 2 A flowchart of a computing resource scheduling method based on cloud computing provided by an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of the structure of a computing resource scheduling system based on cloud computing provided by an embodiment of the present invention;

[0041] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] The computing resource scheduling method based on cloud computing provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the intelligent collection device of each node of the power grid system communicates with the server 102 through the network. The data storage system can store the data that the server 102 needs to process. The data storage system can be integrated on the server 102, or it can be placed on the cloud or other network servers. The server 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0044] The nodes of the power grid system include but are not limited to nodes in the power generation, transmission, distribution and consumption links.

[0045] like Figure 2 As shown, the embodiment of the present application provides a computing resource scheduling method based on cloud computing, and the method is applied to Figure 1 The server 102 in the example is used as an example to illustrate the method, which includes the following steps S1 to S4. Among them:

[0046] Step S1: upload the power data collected in real time by each node of the power grid system to the edge computing node, and pre-process the power data through the edge computing node.

[0047] Among them, a large number of data collection devices need to be deployed at each node of the power grid, which not only covers the different links of power generation, transmission, distribution and consumption, but also can capture the power data of each node in the operation of the power grid in real time. The power data includes but is not limited to voltage, current and power factor.

[0048] In order to improve data transmission efficiency and reduce the computing pressure of the central cloud platform, it is first necessary to pre-process these real-time collected power data through edge computing nodes to perform preliminary processing and filtering on the power data.

[0049] Among them, preprocessing includes data denoising and data compression.

[0050] For data denoising, in order to ensure the accuracy and reliability of collected data, common data denoising methods include algorithms based on wavelet transform or Fourier transform. These algorithms can identify and separate noise components by converting signals from the time domain to the frequency domain, and then remove unnecessary noise interference through technical means such as threshold filtering. Ultimately, the denoised data will be more accurate and can effectively reflect the actual operating status of the power system, providing a basis for subsequent data analysis and intelligent scheduling.

[0051] Data compression is the use of distributed edge computing technology. Each edge computing node has a certain computing power and can process part of the data on-site at the source of data collection to avoid transmitting a large amount of unprocessed raw data directly to the cloud. The application of edge computing mode can effectively reduce the bandwidth usage in data transmission and reduce latency. In addition, by introducing compression algorithms such as differential coding or predictive coding in edge computing nodes, the power data can be compressed through compression algorithms, which can further reduce the data volume and optimize transmission efficiency.

[0052] Step S2: uploading the pre-processed power data to each cloud computing node in the cloud platform.

[0053] In some embodiments, when uploading power data to the cloud platform, the preprocessed power system data needs to be transmitted in a secure and reliable manner to ensure that it is not interfered with by the outside world during transmission and can reach the cloud efficiently and with low latency for storage and further processing.

[0054] To this end, in step S2, the process of uploading the pre-processed power data to each cloud computing node in the cloud platform includes:

[0055] Step S201: encrypt the preprocessed power data to obtain encrypted power data.

[0056] Among them, the embodiment of the present application encrypts the pre-processed power data by using encryption technologies such as SSL (Secure Sockets Layer) / TLS (Transport Layer Security) to ensure the security of data in the transmission link and prevent any unauthorized access or tampering. These encryption protocols can not only provide reliable security in data transmission, but also ensure the integrity of the data, thereby eliminating the risk of data loss caused by security issues during network transmission.

[0057] Step S202: Upload the encrypted power data to each cloud computing node in the cloud platform through a multi-level cache strategy.

[0058] It is understandable that, considering the huge amount of data in the power system and the possible instability of the network environment, the embodiment of the present application introduces a multi-level caching strategy to cope with potential network fluctuations. During the transmission process, a multi-level caching strategy is used to cope with network fluctuations to ensure that data can be transmitted to cloud storage reliably and with low latency.

[0059] Specifically, the multi-level caching strategy is implemented through multiple levels of caching mechanisms. Starting from the edge computing node cache, the power data is first temporarily stored in the edge node cache close to the collection source, and then gradually uploaded to the regional cache, where it is again stored for a short period of time and finally transmitted to the central cloud platform cache.

[0060] It is understandable that the multi-level cache strategy can ensure that even if there is a temporary interruption or congestion in the network transmission path, the data can be continuously and stably uploaded to the cloud through the support of each layer of cache. At the same time, the introduction of the multi-level cache mechanism can also significantly reduce the data loss or delay caused by sudden network jitter, ensuring the smoothness and efficiency of the entire transmission process.

[0061] In addition, in order to improve the efficiency of data transmission and reduce delays, especially in a complex and dynamically changing network environment such as the power system, the embodiment of the present application introduces the Transmission Control Protocol (TCP), that is, the process of uploading encrypted power data to the cloud platform uses the Transmission Control Protocol for data uploading.

[0062] Specifically, the transmission control protocol can use algorithms such as Fast Retransmit. Under the traditional TCP protocol, when data packets are lost or delayed, the entire transmission process is often affected. With the fast retransmit mechanism, when some data packets are detected to be lost, the system will quickly resend these lost data packets, avoiding the long delay caused by waiting timeout in conventional transmission. This optimization process can effectively improve the reliability of transmission and greatly shorten the response time of data transmission.

[0063] Step S3: Predicting resource requirements for each cloud computing node and optimizing the cloud computing resource allocation of each cloud computing node by using the resource demand prediction results to minimize the waste of computing resources and minimize the system response time as the optimization goals.

[0064] It should be noted that due to the highly dynamic load of the power system, cloud computing resources need to be intelligently allocated according to the actual load conditions to ensure rapid response under high load and effective saving of computing resources under low load.

[0065] To this end, the embodiments of the present application aim at the highly dynamic and unpredictable nature of the load of the power system, and reasonably allocate cloud computing resources through accurate load prediction and resource scheduling to avoid resource waste and ensure the timeliness of system response. To this end, the present invention adopts a resource dynamic scheduling algorithm based on load prediction and multi-objective optimization to achieve this process.

[0066] Specifically, step S3 includes:

[0067] Step S301: Use the historical load data of the power system to predict the load data at future times through the LSTM neural network model, and determine the resource demand prediction result based on the load data prediction result at future times.

[0068] Among them, the embodiment of the present application adopts a time series analysis algorithm to predict historical load data and estimate resource demand in a certain period of time in the future.

[0069] First, load prediction is the basis of the entire dynamic resource allocation. In order to accurately predict future computing resource requirements, the following algorithm is used:

[0070] Assume that the historical load data of the power system at time t is ,in Represents the load value of the i-th node at time t. This algorithm predicts the future time by learning this set of historical data Resource requirements ,Right now:

[0071]

[0072] In the formula, Indicates that the i-th node will be in the future Forecast demand, It is the network parameter of the LSTM neural network model. Through continuous updating and training, the LSTM neural network model can provide more accurate load prediction.

[0073] Among them, this method uses a load prediction model based on long short-term memory (LSTM), which can accurately predict future load changes, so that the cloud platform can respond quickly under high load and allocate computing resources in time, and reasonably reduce resource allocation under low load, thereby greatly improving resource utilization and reducing the operating cost of the power system.

[0074] Step S302: Based on the resource demand prediction result, a multi-objective optimization model is constructed with minimizing the waste of computing resources and minimizing the system response time as the optimization goals. The multi-objective optimization model is:

[0075]

[0076] In the formula, and are the control weight coefficients of resource waste and response time respectively; is the computing resources actually allocated to the ith cloud computing node at time t; N is the number of nodes; Predict results for load data; is the time interval; t is the current time period; Represents the response time of the i-th cloud computing node.

[0077] in, and It can be set based on empirical values. The embodiment of the present application dynamically allocates cloud computing resources based on the results of load prediction through a multi-objective optimization algorithm. The main goal is to minimize resource waste and maximize system response speed, that is, to minimize response time.

[0078] Step S303: searching and solving the multi-objective optimization model to obtain an optimization solution for computing resources allocated to each cloud computing node.

[0079] Among them, to solve the multi-objective optimization problem, a solution method based on genetic algorithm (GA) or particle swarm optimization algorithm (PSO) is adopted. Through iterative optimization, the best computing resource allocation solution C can be obtained. i (t).

[0080] Step S4: Utilize the cloud computing resources to allocate the optimized cloud computing nodes to perform computing tasks on the power data.

[0081] It should be noted that the embodiment of the present application improves the data transmission capacity by pre-processing the power data collected in real time by each node of the power grid system, and predicts the resource demand of each cloud computing node. The resource demand prediction results are used to minimize the waste of computing resources and minimize the system response time to optimize the cloud computing resource allocation of each cloud computing node. This can sense and respond to changes in resource demand in real time, ensure efficient use of resources, improve the utilization efficiency and response speed of dynamically allocated computing resources, and ensure the stability and reliability of system operation.

[0082] In some embodiments, the cloud platform needs to monitor the entire cloud platform system in real time based on the computing resource scheduling results, and continuously adjust the resource allocation strategy through a feedback mechanism to ensure efficient scheduling and smooth operation of the power system. To this end, the embodiments of the present application dynamically optimize computing resources and task scheduling by monitoring the real-time operation data of the system to improve the overall performance of the system.

[0083] Specifically, the method also includes:

[0084] Step S501: monitor the current actual resource demand data of each cloud computing node in real time.

[0085] Among them, the entire feedback control process can be regarded as a closed-loop control system, the core of which is to dynamically adjust resource allocation and task scheduling based on real-time monitoring data, and the current actual resource demand data of each cloud computing node can be obtained by collecting data from each cloud computing node in real time.

[0086] Step S502: for each cloud computing node, determine the resource demand prediction error according to the current actual resource demand data and the resource demand prediction result.

[0087] Assume that at time t, the actual resource demand of the cloud computing node is L(t), the resource allocation state of the system is C(t), and the expected resource demand is the predicted value obtained by the aforementioned load prediction model. , the feedback control mechanism will adjust the resource allocation strategy according to the error e(t), and the error formula is:

[0088]

[0089] Among them, e(t) represents the difference between the predicted load and the actual load. In order to ensure system stability and reasonable resource scheduling, the system needs to adjust the resource allocation strategy C(t) in real time according to the changes in e(t) to minimize the error.

[0090] Step S503: Generate computing resource adjustment data through a PID controller according to the resource demand prediction error.

[0091] Among them, a proportional integral differential controller (PID controller) is used to dynamically adjust resource allocation. The PID controller generates an adjustment signal based on the current error, historical error integral, and error change rate. , and applied to resource allocation at the next moment, the output of the PID controller can be expressed as:

[0092]

[0093] In the formula, K p is the proportionality coefficient, reflecting the impact of the current error on the resource allocation adjustment;

[0094] K i is the integral coefficient, reflecting the impact of historical error accumulation on system adjustment;

[0095] K d is the differential coefficient, which reflects the influence of the error change rate on the system response.

[0096] Step S504: dynamically adjust the computing resources of the cloud computing nodes according to the computing resource adjustment data.

[0097] Among them, through the real-time monitoring system Dynamically adjust resource allocation so that the system gradually approaches the optimal resource allocation state, thereby reducing the error e(t). As the system load changes, this adaptive adjustment can ensure that the power system responds quickly under high load and saves resources under low load.

[0098] In some embodiments, after step S504, the method further includes:

[0099] Step S505: Determine an adaptive adjustment model with minimizing the sum of load error and resource consumption as an optimization goal according to the computing resource adjustment data and the resource demand prediction error.

[0100] Among them, through the adaptive adjustment process, the goal of the feedback control system is to minimize the following objective function:

[0101]

[0102] In the formula, It represents the square of load prediction error, reflecting the deviation of system load; Indicates the energy consumption of system resource allocation to ensure that no resources are wasted during the scheduling process; The weight coefficient that represents the balance between load error and resource consumption.

[0103] Step S506: Optimize the computing resource adjustment data according to the adaptive adjustment model.

[0104] Among them, the computing resource adjustment data is optimized according to the adaptive adjustment model, that is, by continuously optimizing the energy consumption of system resource allocation, it is ensured that the system can reasonably adjust resource allocation through the adaptive feedback mechanism when the load fluctuates, so as to achieve the dual goals of load balancing and resource conservation.

[0105] It can be understood that by real-time monitoring of the system's load and resource usage, using the PID controller to adjust the resource allocation strategy, and minimizing errors and optimizing resources, adaptive load adjustment is achieved, thereby ensuring the efficiency and flexibility of the power grid cloud platform's dispatching strategy under various operating conditions.

[0106] It should be noted that this application aims to solve the resource scheduling problem caused by the explosive growth of data volume and the highly dynamic load in the power system. With the advancement of intelligent and automated power grids, the deployment of large-scale intelligent acquisition devices and data processing equipment has resulted in the inability of traditional data processing methods to meet real-time and high efficiency requirements. Through this method, the cloud platform can improve resource utilization efficiency, reduce system latency, and ensure the stability and reliability of system operation in a complex and dynamic power system environment.

[0107] Based on the same inventive concept, an embodiment of the present application also provides a cloud computing-based computing resource scheduling system for implementing the cloud computing-based computing resource scheduling method involved above.

[0108] The implementation solution for solving the problem provided by the system is similar to the implementation solution recorded in the above method. Therefore, the specific limitations in one or more cloud computing-based computing resource scheduling system embodiments provided below can refer to the limitations on the cloud computing-based computing resource scheduling method above, and will not be repeated here.

[0109] like Figure 3 As shown, the embodiment of the present application also provides a computing resource scheduling system based on cloud computing, including:

[0110] The data preprocessing module 100 is used to upload the power data collected in real time by each node of the power grid system to the edge computing node, and preprocess the power data through the edge computing node;

[0111] The data uploading module 200 is used to upload the pre-processed power data to each cloud computing node in the cloud platform;

[0112] The resource allocation module 300 is used to predict the resource demand of each cloud computing node and optimize the cloud computing resource allocation of each cloud computing node by minimizing the waste of computing resources and minimizing the system response time based on the resource demand prediction results;

[0113] The computing execution module 400 is used to execute computing tasks on power data using each cloud computing node after optimization of cloud computing resource allocation.

[0114] In some embodiments, preprocessing includes data denoising and data compression.

[0115] In some embodiments, the data upload module 200 includes:

[0116] An encryption module, used for encrypting the pre-processed power data to obtain encrypted power data;

[0117] The cache upload module is used to upload the encrypted power data to each cloud computing node in the cloud platform through a multi-level cache strategy.

[0118] In some embodiments, the data upload module 200 further includes:

[0119] The transmission module is used to encrypt the power data and upload it to the cloud platform using the transmission control protocol for data upload.

[0120] In some embodiments, the resource allocation module 300 includes:

[0121] The load prediction module is used to use the historical load data of the power system to predict the load data at future times through the LSTM neural network model, and determine the resource demand prediction result according to the load data prediction result at future times;

[0122] The model building module is used to build a multi-objective optimization model based on the resource demand prediction results, with the optimization objectives of minimizing the waste of computing resources and minimizing the system response time. The multi-objective optimization model is:

[0123]

[0124] In the formula, and are the control weight coefficients of resource waste and response time respectively; is the computing resources actually allocated to the ith cloud computing node at time t; N is the number of nodes; Predict results for load data; is the time interval; t is the current time period; represents the response time of the i-th cloud computing node;

[0125] The optimization module is used to find the optimal solution for the multi-objective optimization model and obtain the optimal solution for the computing resources allocated to each cloud computing node.

[0126] In some embodiments, the system further comprises:

[0127] The demand monitoring module is used to monitor the actual resource demand data of each cloud computing node in real time;

[0128] An error module is used to determine the resource demand prediction error for each cloud computing node based on the current actual resource demand data and the resource demand prediction result;

[0129] A resource adjustment generation module is used to generate computing resource adjustment data through a PID controller according to resource demand prediction errors;

[0130] The resource adjustment module is used to dynamically adjust the computing resources of the cloud computing nodes according to the computing resource adjustment data.

[0131] In some embodiments, the system further comprises:

[0132] An adaptive module, used for determining an adaptive adjustment model with minimization of the sum of load error and resource consumption as an optimization target according to computing resource adjustment data and resource demand prediction error;

[0133] The adaptive optimization module is used to optimize the computing resource adjustment data according to the adaptive adjustment model.

[0134] like Figure 4 As shown, an embodiment of the present application also provides an electronic device, the electronic device 10 includes a memory 20 and a processor 30, the memory 20 stores a computer program, and when the computer program is executed by the processor 30, the processor 30 executes the steps of the cloud computing-based computing resource scheduling method in any of the above embodiments.

[0135] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the steps of the computing resource scheduling method based on cloud computing in any of the above embodiments are implemented.

[0136] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, electronic device and computer storage medium can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0137] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0138] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a part of a module, a program segment or a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.

[0139] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0140] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0141] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0143] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A computing resource scheduling method based on cloud computing, characterized in that: include: Upload the power data collected in real time by each node of the power grid system to the edge computing node, and pre-process the power data through the edge computing node; By uploading the pre-processed power data to each cloud computing node in the cloud platform; By predicting the resource demand of each cloud computing node, and using the resource demand prediction results to optimize the cloud computing resource allocation of each cloud computing node with the optimization goals of minimizing the waste of computing resources and minimizing the system response time; The cloud computing resources are allocated and optimized to execute computing tasks on the power data by using each cloud computing node.

2. The computing resource scheduling method based on cloud computing according to claim 1, characterized in that: The preprocessing includes data denoising and data compression.

3. The computing resource scheduling method based on cloud computing according to claim 1, characterized in that: The step of uploading the pre-processed power data to each cloud computing node in the cloud platform includes: Encrypting the preprocessed power data to obtain encrypted power data; The encrypted power data is uploaded to each cloud computing node in the cloud platform through a multi-level cache strategy.

4. The computing resource scheduling method based on cloud computing according to claim 3, characterized in that: The process of uploading the encrypted power data to the cloud platform adopts the transmission control protocol to upload the data.

5. The computing resource scheduling method based on cloud computing according to claim 1, characterized in that: The step of predicting resource requirements for each cloud computing node and optimizing the cloud computing resource allocation of each cloud computing node by using the resource demand prediction results to minimize the waste of computing resources and minimize the system response time as the optimization goals includes: The historical load data of the power system is used to predict the load data at future times through the LSTM neural network model, and the resource demand prediction results are determined based on the load data prediction results at future times; Based on the resource demand prediction results, a multi-objective optimization model is constructed with minimizing the waste of computing resources and minimizing the system response time as the optimization goals. The multi-objective optimization model is: ， In the formula, and are the control weight coefficients of resource waste and response time respectively; is the computing resources actually allocated to the ith cloud computing node at time t; N is the number of nodes; Predict results for load data; is the time interval; t is the current time period; represents the response time of the i-th cloud computing node; The multi-objective optimization model is optimized and solved to obtain an optimization solution for the computing resources allocated to each cloud computing node.

6. The computing resource scheduling method based on cloud computing according to claim 1, characterized in that: Also includes: Real-time monitoring of the actual resource demand data of each cloud computing node; For each cloud computing node, determining a resource demand prediction error according to the current actual resource demand data and the resource demand prediction result; Generate computing resource adjustment data through a PID controller according to the resource demand prediction error; The computing resources of the cloud computing nodes are dynamically adjusted according to the computing resource adjustment data.

7. The computing resource scheduling method based on cloud computing according to claim 6, characterized in that: After the step of dynamically adjusting the computing resources of the cloud computing node according to the computing resource adjustment data, the method further includes: Determine, according to the computing resource adjustment data and the resource demand prediction error, an adaptive adjustment model with minimization of the sum of load error and resource consumption as an optimization goal; The computing resource adjustment data is optimized according to the adaptive adjustment model.

8. A computing resource scheduling system based on cloud computing, characterized in that: include: A data preprocessing module is used to upload the power data collected in real time by each node of the power grid system to the edge computing node, and preprocess the power data through the edge computing node; A data upload module is used to upload the pre-processed power data to each cloud computing node in the cloud platform; A resource allocation module is used to predict the resource demand of each cloud computing node and optimize the cloud computing resource allocation of each cloud computing node by using the resource demand prediction results to minimize the waste of computing resources and minimize the system response time as the optimization goals; The computing execution module is used to perform computing tasks on the power data using each cloud computing node after optimization of cloud computing resource allocation.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the cloud computing-based computing resource scheduling method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the computing resource scheduling method based on cloud computing as described in any one of claims 1 to 7 are implemented.

Images