CN117575113A - Edge collaborative task processing method, device and equipment based on Markov chain - Google Patents

Edge collaborative task processing method, device and equipment based on Markov chain Download PDF

Info

Publication number
CN117575113A
CN117575113A CN202410065715.9A CN202410065715A CN117575113A CN 117575113 A CN117575113 A CN 117575113A CN 202410065715 A CN202410065715 A CN 202410065715A CN 117575113 A CN117575113 A CN 117575113A
Authority
CN
China
Prior art keywords
edge computing
processed
computing terminal
target
power equipment
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202410065715.9A
Other languages
Chinese (zh)
Other versions
CN117575113B (en
Inventor
刘德宏
习伟
蔡田田
陈军健
杨英杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southern Power Grid Digital Grid Research Institute Co Ltd
Original Assignee
Southern Power Grid Digital Grid Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southern Power Grid Digital Grid Research Institute Co Ltd filed Critical Southern Power Grid Digital Grid Research Institute Co Ltd
Priority to CN202410065715.9A priority Critical patent/CN117575113B/en
Publication of CN117575113A publication Critical patent/CN117575113A/en
Application granted granted Critical
Publication of CN117575113B publication Critical patent/CN117575113B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Educational Administration (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • General Engineering & Computer Science (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application relates to a method, a device and equipment for processing edge collaborative tasks based on a Markov chain. The method comprises the steps of inputting historical operation information of power equipment corresponding to an edge computing terminal to be processed into an operation state model to obtain a state probability distribution matrix, inputting service description information of the power equipment and the state probability distribution matrix into a load model to obtain a predicted computing load corresponding to the power equipment, and uploading a target task to be processed of target power equipment corresponding to the edge computing terminal to be processed to the target edge computing terminal for processing when the predicted computing load is larger than the residual computing resources of the edge computing terminal to be processed. Compared with the traditional method for intensively processing the business of the power equipment through the cloud, the method and the device have the advantages that the calculation load is predicted, and when the load is too high, the business of the power equipment related to the plurality of edge calculation terminals is flexibly distributed to the corresponding edge calculation terminals for processing, so that the task processing efficiency is improved.

Description

Edge collaborative task processing method, device and equipment based on Markov chain
Technical Field
The present application relates to the field of electric power technology, and in particular, to a markov chain-based edge collaborative task processing method, apparatus, computer device, storage medium, and computer program product.
Background
Along with the continuous promotion of the construction process of a novel electric power system, more and more terminal devices are connected into a power grid, such as an intelligent meter, a charging pile, a photovoltaic device, wind power devices, energy storage devices and the like, so that mass data generated by the terminal devices are in urgent need of strong calculation support and optimization of a calculation network, and efficient processing of the mass data is achieved. At present, task processing modes of all devices in a micro-grid taking new energy as a theme are generally carried out through uploading cloud processing. However, by means of centralized management and control of the cloud, the calculation load of the cloud is larger, and further the efficiency of task processing is reduced.
Therefore, the current task processing method for the devices in the micro-grid has the defect of low processing efficiency.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a markov chain-based edge collaborative task processing method, apparatus, computer device, computer readable storage medium, and computer program product that can improve task processing efficiency.
In a first aspect, the present application provides a method for processing a side collaborative task based on a markov chain, where the method includes:
acquiring historical operation information of power equipment corresponding to an edge computing terminal to be processed;
Inputting the historical operation information into an operation state model based on a Markov chain, and acquiring a state probability distribution matrix output by the operation state model based on the historical operation information; the state probability distribution matrix comprises a plurality of predicted running states of the power equipment at a plurality of moments;
acquiring service description information corresponding to the power equipment, inputting the state probability distribution matrix and the service description information into a load model, and acquiring a predicted calculation load corresponding to the power equipment, which is output by the load model, based on a plurality of predicted running states in the state probability distribution matrix and the service description information;
if the predicted computing load is greater than the residual computing resources corresponding to the edge computing terminal to be processed, uploading a target task to be processed of target power equipment corresponding to the edge computing terminal to the target edge computing terminal, and processing the target task to be processed by the target edge computing terminal; the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal; and the delay of the target edge computing terminal for processing the target task to be processed of the target power equipment is smaller than a preset threshold value.
In one embodiment, the obtaining the historical operation information of the power device corresponding to the edge computing terminal to be processed includes:
and acquiring at least one of historical electrical quantity information, historical environment quantity information and historical state quantity information of the power equipment corresponding to the edge computing terminal to be processed, and obtaining operation information of the power equipment.
In one embodiment, the inputting the historical operating information into a markov chain based operating state model comprises:
inputting the historical operation information into an operation state model based on a Markov chain, and obtaining an initial state probability matrix by the operation state model based on a plurality of preset equipment states corresponding to the power equipment and the historical operation information;
obtaining a state transition probability matrix according to the state transition probabilities among a plurality of groups of operation states in the initial state probability matrix;
and predicting a plurality of predicted running states of the power equipment at a plurality of moments corresponding to the power equipment according to the initial state probability matrix and the state transition probability matrix, and outputting a state probability distribution matrix containing the plurality of predicted running states at the plurality of moments.
In one embodiment, the obtaining the service description information corresponding to the power device includes:
and acquiring the inherent task calculated amount and application function configuration information corresponding to the power equipment to obtain the service description information.
In one embodiment, the inputting the state probability distribution matrix and the service description information into a load model includes:
inputting the state probability distribution matrix and the service description information into a load model, obtaining event probability of a root event corresponding to each running state in the state probability distribution matrix by the load model according to the service description information and the state probability distribution matrix, and obtaining an event probability distribution matrix according to a plurality of event probabilities corresponding to a plurality of running states;
and acquiring the service corresponding to each root event, acquiring the association relation between the root event corresponding to each running state and the task corresponding to each root event according to each running state, and outputting the predicted calculation load corresponding to the power equipment according to the event probability distribution matrix and the association relation.
In one embodiment, the method further comprises:
Acquiring at least one other edge computing terminal associated with the target power device;
aiming at each other edge computing terminal in the at least one other edge computing terminal, acquiring communication delay and computing delay corresponding to the other edge computing terminal;
inputting the communication delay and the calculation delay into a preset optimization function, and obtaining the total delay corresponding to the other edge calculation terminals according to the minimum value of the function value corresponding to the preset optimization function;
and obtaining the target edge computing terminal according to the minimum value in the total delay of the plurality of other edge computing terminals.
In one embodiment, the obtaining the communication delay and the computation delay corresponding to the other edge computing terminals includes:
according to the data volume of the target task to be processed, the transmission rate, the transmission bandwidth, the transmission power, the channel gain and the Gaussian white noise power spectral density corresponding to the other edge computing terminals, the communication delay corresponding to the other edge computing terminals is obtained;
and acquiring the calculation load of the target task to be processed, and acquiring the calculation delay corresponding to the other edge calculation terminals according to the calculation load and the calculation rate of the other edge calculation terminals.
In a second aspect, the present application provides an edge collaborative task processing device based on a markov chain, where the device includes:
the acquisition module is used for acquiring historical operation information of the power equipment corresponding to the edge computing terminal to be processed;
the first input module is used for inputting the historical operation information into an operation state model based on a Markov chain and acquiring a state probability distribution matrix output by the operation state model based on the historical operation information; the state probability distribution matrix comprises a plurality of predicted running states of the power equipment at a plurality of moments;
the second input module is used for acquiring service description information corresponding to the power equipment, inputting the state probability distribution matrix and the service description information into a load model, acquiring a plurality of predicted running states and the service description information in the load model based on the state probability distribution matrix, and outputting a predicted calculation load corresponding to the power equipment;
the processing module is used for uploading a target to-be-processed task of target power equipment corresponding to the to-be-processed edge computing terminal to the target edge computing terminal and processing the target to-be-processed task by the target edge computing terminal if the predicted computing load is larger than the residual computing resources corresponding to the to-be-processed edge computing terminal; the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal; and the delay of the target edge computing terminal for processing the target task to be processed of the target power equipment is smaller than a preset threshold value.
In a third aspect, the present application provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when the processor executes the computer program.
In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method described above.
In a fifth aspect, the present application provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method described above.
According to the edge collaborative task processing method, the device, the computer equipment, the storage medium and the computer program product based on the Markov chain, the historical operation information of the power equipment corresponding to the edge computing terminal to be processed is input into the operation state model to obtain the state probability distribution matrix, the service description information of the power equipment and the state probability distribution matrix are input into the load model to obtain the prediction computing load corresponding to the power equipment, and when the prediction computing load is larger than the residual computing resources of the edge computing terminal to be processed, the target task to be processed of the target power equipment corresponding to the edge computing terminal to be processed is uploaded to the target edge computing terminal to be processed. Compared with the traditional method for intensively processing the business of the power equipment through the cloud, the method and the device have the advantages that the calculation load is predicted, and when the load is too high, the business of the power equipment related to the plurality of edge calculation terminals is flexibly distributed to the corresponding edge calculation terminals for processing, so that the task processing efficiency is improved.
Drawings
FIG. 1 is an application environment diagram of a markov chain-based edge collaborative task processing approach in one embodiment;
FIG. 2 is a flow diagram of a method of edge collaborative task processing based on Markov chains in one embodiment;
FIG. 3 is a schematic diagram of a markov chain-based edge collaborative task processing system in one embodiment;
FIG. 4 is a flow chart of a method of edge collaborative task processing based on Markov chains in another embodiment;
FIG. 5 is a block diagram of an edge collaborative task handling device based on Markov chains in one embodiment;
fig. 6 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The edge collaborative task processing method based on the Markov chain, provided by the embodiment of the application, can be applied to an application environment shown in fig. 1. Wherein, the management end communicates with the micro-grid through a network. The micro-grid can comprise a plurality of edge computing terminals and power equipment corresponding to the edge computing terminals, the management end can manage the micro-grid, for example, by acquiring the running state of the power equipment, predicting the computing load based on the running state, and when the predicted computing load is overlarge, adjusting the computing terminals of the target tasks to be processed of the target power equipment in the micro-grid, so that the load is balanced, and the task processing efficiency is improved. The terminal may be, but not limited to, various personal computers, notebook computers, and tablet computers.
In one embodiment, as shown in fig. 2, a method for processing edge collaborative tasks based on markov chains is provided, and the method is applied to the management end in fig. 1 for illustration, and includes the following steps:
step S202, acquiring historical operation information of the power equipment corresponding to the edge computing terminal to be processed.
The edge computing terminal to be processed may be an edge computing terminal in a micro-grid. The edge computing terminal to be processed is associated with a plurality of power devices, and needs to process tasks of the respective power devices. The management end can predict the calculation load of the tasks of the electric equipment corresponding to the edge calculation terminal to be processed, so that when the calculation load of the edge calculation terminal to be processed is overloaded, the tasks of the electric equipment are distributed to other edge calculation terminals, load balancing processing is realized, and the efficiency of processing the tasks of the electric equipment is improved.
The management end can determine the calculation load of the edge calculation terminal to be processed at each moment in a preset time period through prediction. For example to predict the computational load at various times of the day. The management end can firstly acquire the historical operation information of the power equipment corresponding to the edge computing terminal to be processed. The power equipment corresponding to the edge computing terminal to be processed can be multiple. And the power equipment corresponding to the edge computing terminal to be processed represents that the power equipment is in the control range of the edge computing terminal to be processed. One power device may be within the control range of one or more edge computing terminals.
The above-described history operation information may include various types of information. For example, in one embodiment, the management end may obtain at least one of historical electrical quantity information, historical environmental quantity information and historical state quantity information of the electrical device corresponding to the edge computing terminal to be processed, so as to obtain operation information of the electrical device. The power equipment in the micro-grid can comprise photovoltaic equipment, wind power equipment, energy storage equipment and the like, and in order to realize management and control of a new energy source, an energy storage device, an intelligent meter, acquisition and sensing equipment and the like in the new energy micro-grid, different unit areas are divided in the new energy micro-grid, and each unit area is provided with a relevant information acquisition terminal for acquiring various operation information of the unit area of the new energy micro-grid. For example, the acquisition terminal acquires environmental quantities such as temperature and humidity, electric quantities such as current and voltage, state quantities describing equipment state information, description information of data processing tasks requested by equipment, and the like.
Wherein, the new energy power supply has larger randomness and volatility as the operation is greatly influenced by the environmentThe change rule of the new energy micro-grid in the operation process is well mastered, and different types of data need to be analyzed on a long time scale. In order to obtain the change rule of the running state of the new energy micro-grid, the running data of the new energy micro-grid needs to be uniformly expressed, and the expression for obtaining the historical running data set at the c-th day d moment is as follows: . Wherein A is d,c The method is a set of environmental quantities such as temperature, humidity and the like at the time of the c day d; b (B) d,c The method is an electric quantity set such as current and voltage; e (E) d,c To describe a set of state quantities of device state information. Construction yc= { Y 0,c ，Y 1,c ，…，Y n,c }，Y c The method is characterized in that a new energy micro-grid operation data set subset is c day, n is the total acquisition point number of a terminal within 24 hours, Y d,c ∈Y c . Therefore, the management end can acquire the historical operation information of the power equipment corresponding to the edge computing terminal to be processed based on the data.
Step S204, inputting the historical operation information into an operation state model based on a Markov chain, and acquiring a state probability distribution matrix of the operation state model output based on the historical operation information; the state probability distribution matrix includes a plurality of predicted operating states of the power device at a plurality of times.
Wherein the terminal may pre-train a markov chain based operational state model. The operational state model may be used to predict the operational state of the electrical device. Such as normal operation, overload, abnormal, etc. The management end can input the obtained historical operation information into an operation state model based on a Markov chain, and obtain a state probability distribution matrix which is output by the operation state model based on the historical operation information. The state probability distribution matrix may include a plurality of predicted operation states corresponding to the power device, where the plurality of predicted operation states respectively correspond to a plurality of moments, for example, a plurality of moments within 24 hours a day. In some embodiments, the state probability distribution matrix may further include probabilities of each predicted operation state and the predicted operation state transition probabilities between different moments.
Step S206, acquiring service description information corresponding to the power equipment, inputting the state probability distribution matrix and the service description information into a load model, and acquiring a predicted calculation load corresponding to the power equipment, which is output by the load model, based on a plurality of predicted running states and the service description information in the state probability distribution matrix.
The service description information may be preset information corresponding to the power equipment, and the service description information may be determined from a power grid experience knowledge base based on descriptions of various services of the power equipment. The service description information may include a plurality of types. For example, in one embodiment, the management end may obtain the intrinsic task computation amount and the application function configuration information corresponding to the power device, so as to obtain the service description information according to the intrinsic task computation amount and the application function configuration information. The inherent task calculation amount can be determined according to the description of each service in the power grid experience knowledge base.
The management end can input the state probability distribution matrix and the service description information into a load model, and the load model obtains the predicted calculation load corresponding to the power equipment based on a plurality of predicted running states and the service description information in the state probability distribution matrix. Therefore, the management end can obtain the predicted calculation load corresponding to the power equipment output by the load model. The output predicted calculation load may include the predicted calculation loads at a plurality of times. Such as a predicted computational load at various times of the day, for example, 24 hours.
Step S208, if the predicted computing load is greater than the residual computing resources corresponding to the edge computing terminal to be processed, uploading the target task to be processed of the target power equipment corresponding to the edge computing terminal to the target edge computing terminal, and processing the target task to be processed by the target edge computing terminal; the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal; the delay of the target edge computing terminal for processing the target task to be processed of the target power equipment is smaller than a preset threshold value.
After obtaining the predicted computing load, the management end can compare the predicted computing load with the residual computing resources corresponding to the edge computing terminal to be processed. I.e. the management end may first obtain the remaining computing resources of the edge computing terminal to be processed. If the management end detects that the predicted computing load is smaller than or equal to the residual computing resources corresponding to the edge computing terminal to be processed, the management end can determine that the edge computing terminal to be processed still has residual force to process the tasks of the electric equipment, and the management end determines to upload the tasks to be processed of the electric equipment to the edge computing terminal to be processed for processing.
If the management end detects that the predicted computing load is greater than the residual computing resources corresponding to the edge computing terminal to be processed, the management end determines that the computing resources of the edge computing terminal to be processed are insufficient for supporting task processing of the power equipment. The management end can acquire the target power equipment in the plurality of power equipment corresponding to the edge computing terminal to be processed. Wherein the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal. I.e. the target power device may be in an overlapping region of the regulatory domains of at least two edge computing terminals. The management end may acquire the delay of the target task to be processed of the target power device by using the at least one other edge computing terminal, and determine, from the at least one other edge computing terminal, a target edge computing terminal that the delay of the target task to be processed of the target power device is less than a preset threshold. Therefore, the management end can upload the target task to be processed of the target power equipment to the target edge computing terminal, so that the target task to be processed can be processed by the target edge computing terminal, and the efficiency of processing the task of the power equipment in the micro-grid is improved through cooperative processing.
In the edge collaborative task processing method based on the Markov chain, the historical operation information of the power equipment corresponding to the edge computing terminal to be processed is input into the operation state model to obtain the state probability distribution matrix, the service description information and the state probability distribution matrix of the power equipment are input into the load model to obtain the prediction computing load corresponding to the power equipment, and when the prediction computing load is larger than the residual computing resources of the edge computing terminal to be processed, the target task to be processed of the target power equipment corresponding to the edge computing terminal to be processed is uploaded to the target edge computing terminal to be processed. Compared with the traditional method for intensively processing the business of the power equipment through the cloud, the method and the device have the advantages that the calculation load is predicted, and when the load is too high, the business of the power equipment related to the plurality of edge calculation terminals is flexibly distributed to the corresponding edge calculation terminals for processing, so that the task processing efficiency is improved.
In one embodiment, inputting historical operating information into a Markov chain-based operating state model includes: inputting historical operation information into an operation state model based on a Markov chain, and obtaining an initial state probability matrix by the operation state model based on a plurality of preset equipment states and historical operation information corresponding to the power equipment; obtaining a state transition probability matrix according to state transition probabilities among a plurality of groups of running states in the initial state probability matrix; and predicting a plurality of predicted running states of the power equipment at a plurality of moments according to the initial state probability matrix and the state transition probability matrix, and outputting a state probability distribution matrix containing the plurality of predicted running states at the plurality of moments.
In this embodiment, the management end may determine the state probability distribution matrix corresponding to the power device through an operation state model based on a markov chain. The management end can input historical operation information into an operation state model based on a Markov chain, and the operation state model obtains an initial state probability matrix based on a plurality of preset device states and the historical operation information corresponding to the power device. The initial state probability matrix may include a plurality of groups of operation states, and the operation state model may obtain a state transition probability matrix according to state transition probabilities among the plurality of groups of operation states in the initial state probability matrix. The operation state model may predict a plurality of predicted operation states of the power device at a plurality of times according to the initial state probability matrix and the state transition probability matrix, and output a state probability distribution matrix including the plurality of predicted operation states at the plurality of times. Wherein the plurality of time instants may be a plurality of time instants of 24 hours in a day.
In particular, a markov chain means that the events are discrete in both time and state, i.e. in a discrete state space s= { S 1 ,s 2 In …, there is a random process { X } on the discrete time parameters t |X t E S, t is greater than or equal to 0, and t has different states for X at different discrete times, and t k+1 The state at the moment is only t k The state determination at the time is specifically as follows:
P[X k+1 =s(t k+1 )|X 1 =s(t 1 ),X 2 =s(t 2 ), …,X k =s(t k )]=P[X k+1 =s(t k+1 )| X k =s(t k )]. Wherein P is conditional probability, t k The corresponding state variable at the moment is X k The corresponding specific running state is s (t k )。
If the state space consists of m states, at t k At time, the state transition probability matrix can be expressed as:
. Wherein the state transition probability matrix satisfies the following properties: />. Wherein t is k-1 Time state s i Conversion to t k Time state s i+1 Elements for transition probabilities of (a)And (3) representing.
The step of managing the output of the state probability distribution matrix based on the new energy microgrid operational state model of the markov chain may comprise: the management end obtains all states of the new energy micro-grid to form a state set S of the Markov chain, wherein the state set S comprises m states. The management end can acquire the running information, the equipment state information and the running state conditions of the time scale and the space scale through the terminalObtaining an initial state probability matrix Z in a Markov chain model 0 . The management end obtains a state transition probability matrix through the state transition probability of sampling points in each sampling period, wherein the total of the state transition probability matrix is n sampling points, and the obtained state transition probability matrix is P= { P t1 ,P t2 , …P tn }. Through the steps, the management end can obtain the state probability distribution at each moment to form a state probability distribution matrix. Thus if the management end knows t k-1 State Z of (2) k-1 After passing Deltat, t k The state probability of a moment in time can be expressed as: z (k) =Z (k-1) P k 。
According to the embodiment, the management end can be combined with the operation state model based on the Markov chain to obtain the state probability distribution matrix containing a plurality of prediction operation states at a plurality of moments, so that the management end can schedule tasks of the edge computing terminal based on the state probability distribution matrix, and the efficiency of processing the tasks of the micro-grid is improved.
In one embodiment, inputting the state probability distribution matrix and the traffic description information into the load model includes: inputting the state probability distribution matrix and the service description information into a load model, obtaining event probability of a root event corresponding to each running state in the state probability distribution matrix by the load model according to the service description information and the state probability distribution matrix, and obtaining an event probability distribution matrix according to a plurality of event probabilities corresponding to a plurality of running states; and acquiring the service corresponding to each root event, acquiring the association relation between the root event corresponding to each running state and the task corresponding to each root event according to each running state, and outputting the predicted calculation load corresponding to the power equipment according to the event probability distribution matrix and the association relation.
In this embodiment, the management end may determine the calculation load of the task of the above-mentioned power device based on the coincidence model. For example, the management end may input the state probability distribution matrix and the service description information into the load model, and obtain, by using the load model, event probabilities of source events corresponding to each operation state in the state probability distribution matrix according to the service description information and the state distribution probability matrix, so that the load model may obtain the event probability distribution matrix according to a plurality of event probabilities corresponding to a plurality of operation states. Wherein the root event represents an event corresponding to a root cause of an impact on the operating state in the microgrid. The load model can acquire the service corresponding to each root event, and can acquire the association relation between the root event corresponding to the running state and the task corresponding to each root event according to each running state, wherein the association relation indicates whether the root event is related to the task, and for different association relations, the load model can determine different values. Therefore, the load model can output the predicted calculation load corresponding to the power equipment according to the event probability distribution matrix and the association relation.
In particular, the load model may be a load model that considers state-event-application associations. In the power grid accessed by a large amount of new energy, wind power, photovoltaic and the like are deeply influenced by factors such as environment and the like, and the same service is applied to different computing task amounts and the like at different moments. Thus, in order to obtain the computational load amounts of different scenarios for different applications, a computational load model is built that takes into account the state-event-application associations as follows. The load model described above can be expressed as:。
wherein J is r ={J r (t 0 ),J r (t 1 ),…J r (t n ) The (R) is the calculated quantity distribution of the (r) new energy service, and the initial calculated quantity is the inherent calculated quantity of the service，J r (t k ) At t for the r new energy service k Calculated amount of time. The inherent calculation amount is described according to the power grid experience knowledge base, and the operation is kept unchanged at each moment of each corresponding day and is changed at different moments of the day, so that at different t k Different values at the moment.
The management end is according to the state-event association relation and t k The moment state probability distribution Z (k) can analyze the root causeThe piece probability distribution is expressed as:
。
wherein f i,j [z i (k)](k=1,2,3,…n；i=1,2,3…m；j=1,2…N i ) For t in n samples k Probability of occurrence of jth root event under state condition of moment, N i Is the total root event. The management end can obtain a probability distribution matrix according to the probability distribution of the root event:. And the management end can use W to represent the relation between the new energy micro-grid event and the application according to the application-event association relation, and the relation is specifically expressed as follows: w= [ W ] 1 ,w 2 ,…w N ]. Wherein a is ij (t k ) Indicated at t k State s at moment of time i Probability of occurrence of the next jth root cause event. w (w) j (j=1, 2,3 … N) represents a set of tasks to be processed in the event of j, and the management end can determine the calculation load information of each new energy task by using the calculation amount and the service arrival frequency.
The management end can obtain the predicted calculated amount and the arrival frequency of the new energy service by using a Markov chain according to the state-event-application association relation, and the predicted calculated amount and the arrival frequency are specifically expressed as follows:。
wherein,representing the calculated amount of the class r task obtained by modifying the state-event-application association relation, namely the predicted calculation load; gamma ray ijr Is 1 in state s i The jth root event is related to the type-r task, otherwise, the jth root event is 0; gamma ray ijr 0 represents irrelevant; η (eta) jr 1 when w is j The application of the class r task is contained, otherwise, the application is 0.
According to the embodiment, the management end can determine the predicted calculation load of the power equipment task through the load model, so that the management end can schedule the task of the edge calculation terminal based on the predicted calculation load, and the efficiency of processing the micro-grid task is improved.
In one embodiment, further comprising: acquiring at least one other edge computing terminal associated with a target power device; aiming at each other edge computing terminal in at least one other edge computing terminal, acquiring communication delay and computing delay corresponding to the other edge computing terminal; inputting the communication delay and the calculation delay into a preset optimization function, and obtaining the total delay corresponding to the other edge calculation terminals according to the minimum value of the function value corresponding to the preset optimization function; and obtaining the target edge computing terminal according to the minimum value in the total delay of the plurality of other edge computing terminals.
In this embodiment, the target power device is associated with at least two edge computing terminals, and there is a difference in task processing delay between the target power device and each associated edge computing terminal, and the management end may determine the target edge computing terminal from the at least one other edge computing terminal. For example, for each of the at least one other edge computing terminals. The management end can acquire the communication delay and the calculation delay corresponding to the other edge calculation terminals, input the communication delay and the calculation delay into a preset optimization function, and acquire the total delay corresponding to the other edge calculation terminals according to the minimum value of the function value corresponding to the preset optimization function. Therefore, the management end can obtain the target edge computing terminal according to the minimum value in the total delay of the plurality of other edge computing terminals.
Specifically, as shown in fig. 3, fig. 3 is a schematic structural diagram of an edge collaborative task processing system based on a markov chain in one embodiment. In a plurality of unit areas divided by the new energy micro-grid, the data interaction mode between the edges and the ends is shown in fig. 3. The unit area is divided into an overlapping area and a non-overlapping area; edge computing terminals are deployed in each unit area, and the terminals in the non-overlapping areas are controlled by local edge computing terminals; terminals in the overlapping area are controlled by a plurality of edge computing terminals, namely, the area commonly controlled by the plurality of edge computing terminals is called an overlapping area; the terminal calculation tasks in the overlapping area can be flexibly uploaded to any managed edge calculation terminal according to requirements, and the minimum communication and calculation delay is considered for the selection of the data uploading mode of the terminal. The new energy terminal shown in fig. 3 may be the above-mentioned power device.
The delay between the target power device and the edge computing terminal may include a communication delay and a computation delay, where the communication delay represents a delay when data transmission is performed between the target power device and the edge computing terminal, and the computation delay represents a delay when the edge computing terminal processes a task of the target power device. The management end can determine the communication delay and the calculation delay in different modes. For example, in one embodiment, the management end may obtain the communication delay corresponding to the other edge computing terminal according to the data volume of the target task to be processed, the transmission rate, the transmission bandwidth, the transmission power, the channel gain, and the gaussian white noise power spectrum density corresponding to the other edge computing terminal. And the management end can also acquire the calculation load of the target task to be processed, and obtain the calculation delay corresponding to the other edge calculation terminals according to the calculation load and the calculation rate of the other edge calculation terminals.
Specifically, the communication delay between the above-mentioned power device and the edge computing terminal may be expressed as:；R i =B ij log 2 (1+P ij h ij /N 0 B ij ) I+.j. Wherein l i For the size of the amount of data transferred; r is R i The rate at which the traffic data is transmitted for the edge channel; b (B) ij Broadband communication from an initial point i to an end point j; p (P) ij The transmission power between i and j; h is a ij Gain for the side channel; n (N) 0 Is the gaussian white noise power spectral density.
In addition, the calculation time of the tasks of the power equipment is related to the calculation task quantity of the tasks and the calculation speed of the edge calculation terminal, and the calculation delay can be expressed asThe method is shown as follows:. Wherein Q is i Calculating load quantity for the task; r is (r) i The computation rate of the traffic is processed for the ith edge.
In the case of concurrent service, in order to improve the processing efficiency of the service, the task processing delay of the terminal should be avoided from exceeding the specified delay tolerance due to insufficient resource of the edge computing terminal. Therefore, when the management end detects that the local to-be-processed edge computing terminal resources are insufficient, the processing tasks of the terminal in the overlapped area, namely the target power equipment, can be offloaded to other edge computing terminals, the selection of the offloaded edge computing terminal aims at minimizing the total delay, and the task of the equipment terminal is specifically offloaded to one of the edge computing terminals in the polygon, for example, the target edge computing terminal. The total delay can be expressed specifically as: 。
Through the embodiment, the management end can upload the target task to be processed to the target edge computing terminal when the predicted computing load of the task is overlarge, so that the computing load of the edge computing terminal to be processed is reduced, and the task processing efficiency is improved.
In one embodiment, as shown in fig. 4, fig. 4 is a flow chart of a method for processing edge collaborative tasks based on markov chains in another embodiment. The method can comprise the following steps:
step one: the edge calculation terminal in the new energy micro-grid collects the operation information such as the electrical quantity information, the environmental quantity information and the equipment health state of the equipment in the management and control area;
step two: the management end obtains the time distribution of the system running state according to the new energy micro-grid running state model based on the Markov chain, and obtains a state probability distribution matrix at the corresponding moment;
step three: the management end inputs a state probability distribution matrix of the corresponding moment obtained by the running state time distribution of the system and service description information such as the inherent task calculated amount and the application function configuration of equipment in a management and control area, and the service processing calculated load of the new energy equipment in the management and control area at the next moment is obtained through prediction based on a state-event-application associated load model, namely the predicted calculated load;
Step four: comparing the calculated load of the service at the next moment of the predicted new energy terminal equipment with the calculated resource of the task which can be processed by the edge calculation terminal;
step five: if the calculation load of the predicted new energy terminal equipment service at the next moment is smaller than the calculation resource of the task which can be processed by the edge calculation terminal, the edge calculation terminal processes the new energy terminal equipment service in the management and control area; otherwise, the edge computing terminal processes the task of the new energy equipment terminal only controlled by the edge computing terminal preferentially according to the left computing resource; and the new energy terminal equipment controlled by the edge computing terminals together uploads the target with minimum service computing delay and communication delay to the corresponding edge computing terminal, namely the target edge computing terminal.
According to the method and the device for processing the tasks, the management end predicts the computing load, and flexibly distributes the service of the power equipment associated with the edge computing terminals to the corresponding edge computing terminals for processing when the load is too high, so that the task processing efficiency is improved. In addition, the request of the power equipment for the resources of the edge computing terminal can be effectively predicted, and the request is beneficial to the local edge terminal to reserve the resources to the adjacent edge computing terminal. Moreover, by setting the overlapping area in the control range of the edge computing terminal, the terminal in the overlapping area is controlled by the adjacent edge computing terminal, so that the terminal in the overlapping area can flexibly select to unload the data to be processed to the edge computing terminal for controlling the overlapping area, the terminal in the overlapping area can more flexibly select to upload the data to the edge computing terminal according to the resource occupation condition of the edge computing terminal, the resources of the adjacent edge computing terminals are reserved, and the reliability of the data processing of the terminal in the overlapping area is greatly improved.
It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a markov chain-based edge collaborative task processing device for implementing the markov chain-based edge collaborative task processing method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of one or more edge collaborative task processing devices based on markov chains provided below may be referred to above for the limitation of the edge collaborative task processing method based on markov chains, which is not described herein.
In one embodiment, as shown in fig. 5, there is provided an edge collaborative task processing device based on a markov chain, including: an acquisition module 500, a first input module 502, a second input module 504, and a processing module 506, wherein:
the obtaining module 500 is configured to obtain historical operation information of the power device corresponding to the edge computing terminal to be processed.
The first input module 502 is configured to input historical operation information into an operation state model based on a markov chain, and obtain a state probability distribution matrix output by the operation state model based on the historical operation information; the state probability distribution matrix includes a plurality of predicted operating states of the power device at a plurality of times.
The second input module 504 is configured to obtain service description information corresponding to the power device, input the state probability distribution matrix and the service description information into the load model, obtain a predicted calculation load corresponding to the power device based on a plurality of predicted operation states and the service description information in the state probability distribution matrix by the load model.
The processing module 506 is configured to upload, if the predicted computing load is greater than the remaining computing resources corresponding to the edge computing terminal to be processed, a target task to be processed of the target power device corresponding to the edge computing terminal to be processed to the edge computing terminal, where the target task to be processed is processed by the edge computing terminal to be processed; the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal; the delay of the target edge computing terminal for processing the target task to be processed of the target power equipment is smaller than a preset threshold value.
In one embodiment, the obtaining module 500 is specifically configured to obtain at least one of historical electrical quantity information, historical environmental quantity information, and historical state quantity information of the electrical device corresponding to the edge computing terminal to be processed, so as to obtain operation information of the electrical device.
In one embodiment, the first input module 502 is specifically configured to input historical operation information into an operation state model based on a markov chain, where the operation state model obtains an initial state probability matrix based on a plurality of preset device states and historical operation information corresponding to the power device; obtaining a state transition probability matrix according to state transition probabilities among a plurality of groups of running states in the initial state probability matrix; and predicting a plurality of predicted running states of the power equipment at a plurality of moments according to the initial state probability matrix and the state transition probability matrix, and outputting a state probability distribution matrix containing the plurality of predicted running states at the plurality of moments.
In one embodiment, the second input module 504 is specifically configured to obtain the intrinsic task computation amount and the application function configuration information corresponding to the power device, so as to obtain service description information.
In one embodiment, the second input module 504 is specifically configured to input a state probability distribution matrix and service description information into a load model, obtain, by the load model, event probabilities of a source event corresponding to each operation state in the state probability distribution matrix according to the service description information and the state probability distribution matrix, and obtain an event probability distribution matrix according to a plurality of event probabilities corresponding to a plurality of operation states; and acquiring the service corresponding to each root event, acquiring the association relation between the root event corresponding to each running state and the task corresponding to each root event according to each running state, and outputting the predicted calculation load corresponding to the power equipment according to the event probability distribution matrix and the association relation.
In one embodiment, the apparatus further comprises: a determining module for acquiring at least one other edge computing terminal associated with the target power device; aiming at each other edge computing terminal in at least one other edge computing terminal, acquiring communication delay and computing delay corresponding to the other edge computing terminal; inputting the communication delay and the calculation delay into a preset optimization function, and obtaining the total delay corresponding to the other edge calculation terminals according to the minimum value of the function value corresponding to the preset optimization function; and obtaining the target edge computing terminal according to the minimum value in the total delay of the plurality of other edge computing terminals.
In one embodiment, the determining module is specifically configured to obtain the communication delay corresponding to the other edge computing terminal according to the data volume of the target task to be processed, the transmission rate, the transmission bandwidth, the transmission power, the channel gain, and the gaussian white noise power spectrum density corresponding to the other edge computing terminal; and acquiring the calculation load of the target task to be processed, and acquiring the calculation delay corresponding to the other edge calculation terminals according to the calculation load and the calculation rate of the other edge calculation terminals.
The above-mentioned markov chain-based edge collaborative task processing device may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a management terminal, and its internal structure may be as shown in fig. 6. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program, when executed by a processor, implements a markov chain-based edge collaborative task processing method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in fig. 6 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, including a memory and a processor, where the memory stores a computer program, and the processor implements the markov chain-based edge collaborative task processing method described above when the processor executes the computer program.
In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the markov chain-based edge collaborative task processing method described above.
In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the markov chain-based edge collaborative task processing method described above.
It should be noted that, user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims (10)

1. An edge collaborative task processing method based on a Markov chain, which is characterized by comprising the following steps:
acquiring historical operation information of power equipment corresponding to an edge computing terminal to be processed;
inputting the historical operation information into an operation state model based on a Markov chain, and acquiring a state probability distribution matrix output by the operation state model based on the historical operation information; the state probability distribution matrix comprises a plurality of predicted running states of the power equipment at a plurality of moments;
Acquiring service description information corresponding to the power equipment, inputting the state probability distribution matrix and the service description information into a load model, and acquiring a predicted calculation load corresponding to the power equipment, which is output by the load model, based on a plurality of predicted running states in the state probability distribution matrix and the service description information;
if the predicted computing load is greater than the residual computing resources corresponding to the edge computing terminal to be processed, uploading a target task to be processed of target power equipment corresponding to the edge computing terminal to the target edge computing terminal, and processing the target task to be processed by the target edge computing terminal; the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal; and the delay of the target edge computing terminal for processing the target task to be processed of the target power equipment is smaller than a preset threshold value.
2. The method of claim 1, wherein the obtaining historical operating information of the power device corresponding to the edge computing terminal to be processed includes:
and acquiring at least one of historical electrical quantity information, historical environment quantity information and historical state quantity information of the power equipment corresponding to the edge computing terminal to be processed, and obtaining operation information of the power equipment.
3. The method of claim 1, wherein said inputting the historical operating information into a markov chain based operating state model comprises:
inputting the historical operation information into an operation state model based on a Markov chain, and obtaining an initial state probability matrix by the operation state model based on a plurality of preset equipment states corresponding to the power equipment and the historical operation information;
obtaining a state transition probability matrix according to the state transition probabilities among a plurality of groups of operation states in the initial state probability matrix;
and predicting a plurality of predicted running states of the power equipment at a plurality of moments corresponding to the power equipment according to the initial state probability matrix and the state transition probability matrix, and outputting a state probability distribution matrix containing the plurality of predicted running states at the plurality of moments.
4. The method of claim 1, wherein the obtaining service description information corresponding to the power device includes:
and acquiring the inherent task calculated amount and application function configuration information corresponding to the power equipment to obtain the service description information.
5. The method of claim 1, wherein said inputting the state probability distribution matrix and the traffic description information into a load model comprises:
Inputting the state probability distribution matrix and the service description information into a load model, obtaining event probability of a root event corresponding to each running state in the state probability distribution matrix by the load model according to the service description information and the state probability distribution matrix, and obtaining an event probability distribution matrix according to a plurality of event probabilities corresponding to a plurality of running states;
and acquiring the service corresponding to each root event, acquiring the association relation between the root event corresponding to each running state and the task corresponding to each root event according to each running state, and outputting the predicted calculation load corresponding to the power equipment according to the event probability distribution matrix and the association relation.
6. The method according to claim 1, wherein the method further comprises:
acquiring at least one other edge computing terminal associated with the target power device;
aiming at each other edge computing terminal in the at least one other edge computing terminal, acquiring communication delay and computing delay corresponding to the other edge computing terminal;
inputting the communication delay and the calculation delay into a preset optimization function, and obtaining the total delay corresponding to the other edge calculation terminals according to the minimum value of the function value corresponding to the preset optimization function;
And obtaining the target edge computing terminal according to the minimum value in the total delay of the plurality of other edge computing terminals.
7. The method of claim 6, wherein the obtaining the communication delay and the computation delay corresponding to the other edge computing terminal includes:
according to the data volume of the target task to be processed, the transmission rate, the transmission bandwidth, the transmission power, the channel gain and the Gaussian white noise power spectral density corresponding to the other edge computing terminals, the communication delay corresponding to the other edge computing terminals is obtained;
and acquiring the calculation load of the target task to be processed, and acquiring the calculation delay corresponding to the other edge calculation terminals according to the calculation load and the calculation rate of the other edge calculation terminals.
8. An edge collaborative task processing device based on a markov chain, the device comprising:
the acquisition module is used for acquiring historical operation information of the power equipment corresponding to the edge computing terminal to be processed;
the first input module is used for inputting the historical operation information into an operation state model based on a Markov chain and acquiring a state probability distribution matrix output by the operation state model based on the historical operation information; the state probability distribution matrix comprises a plurality of predicted running states of the power equipment at a plurality of moments;
The second input module is used for acquiring service description information corresponding to the power equipment, inputting the state probability distribution matrix and the service description information into a load model, acquiring a plurality of predicted running states and the service description information in the load model based on the state probability distribution matrix, and outputting a predicted calculation load corresponding to the power equipment;
the processing module is used for uploading a target to-be-processed task of target power equipment corresponding to the to-be-processed edge computing terminal to the target edge computing terminal and processing the target to-be-processed task by the target edge computing terminal if the predicted computing load is larger than the residual computing resources corresponding to the to-be-processed edge computing terminal; the target power device represents a power device associated with the edge computing terminal to be processed and at least one other edge computing terminal; and the delay of the target edge computing terminal for processing the target task to be processed of the target power equipment is smaller than a preset threshold value.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
CN202410065715.9A 2024-01-17 2024-01-17 Edge collaborative task processing method, device and equipment based on Markov chain Active CN117575113B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410065715.9A CN117575113B (en) 2024-01-17 2024-01-17 Edge collaborative task processing method, device and equipment based on Markov chain

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410065715.9A CN117575113B (en) 2024-01-17 2024-01-17 Edge collaborative task processing method, device and equipment based on Markov chain

Publications (2)

Publication Number Publication Date
CN117575113A true CN117575113A (en) 2024-02-20
CN117575113B CN117575113B (en) 2024-05-03

Family

ID=89892278

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410065715.9A Active CN117575113B (en) 2024-01-17 2024-01-17 Edge collaborative task processing method, device and equipment based on Markov chain

Country Status (1)

Country Link
CN (1) CN117575113B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113301169A (en) * 2021-07-23 2021-08-24 杭州雅观科技有限公司 Edge network switching method based on dynamic mobile device behavior prediction
KR20210133434A (en) * 2020-04-29 2021-11-08 한양대학교 산학협력단 Forming method of generated power prediction model, energy storage system capacity calculation method and energy storage system operation method
CN113660303A (en) * 2021-07-02 2021-11-16 山东师范大学 Task unloading method and system based on end side network cloud cooperation
CN115103313A (en) * 2022-05-18 2022-09-23 北京邮电大学 Intelligent road task cooperative processing method and system based on position prediction
CN115413044A (en) * 2022-08-31 2022-11-29 中国科学院沈阳自动化研究所 Computing and communication resource joint distribution method for industrial wireless network
WO2023124947A1 (en) * 2021-12-29 2023-07-06 苏州浪潮智能科技有限公司 Task processing method and apparatus, and related device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210133434A (en) * 2020-04-29 2021-11-08 한양대학교 산학협력단 Forming method of generated power prediction model, energy storage system capacity calculation method and energy storage system operation method
CN113660303A (en) * 2021-07-02 2021-11-16 山东师范大学 Task unloading method and system based on end side network cloud cooperation
CN113301169A (en) * 2021-07-23 2021-08-24 杭州雅观科技有限公司 Edge network switching method based on dynamic mobile device behavior prediction
WO2023124947A1 (en) * 2021-12-29 2023-07-06 苏州浪潮智能科技有限公司 Task processing method and apparatus, and related device
CN115103313A (en) * 2022-05-18 2022-09-23 北京邮电大学 Intelligent road task cooperative processing method and system based on position prediction
CN115413044A (en) * 2022-08-31 2022-11-29 中国科学院沈阳自动化研究所 Computing and communication resource joint distribution method for industrial wireless network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
陈思颖;郭跃;胡欢波;: "移动云计算随机任务序列的执行调度", 电脑知识与技术, no. 21, 25 July 2018 (2018-07-25), pages 42 - 45 *

Also Published As

Publication number Publication date
CN117575113B (en) 2024-05-03

Similar Documents

Publication Publication Date Title
CA3032409C (en) Distributed resource electrical demand forecasting system and method
WO2023007704A1 (en) Prediction device, prediction method, and prediction program
CN112817721B (en) Task scheduling method and device based on artificial intelligence, computer equipment and medium
CN116154768B (en) Power interval prediction method adopting point prediction error empirical distribution inverse transformation
CN117575113B (en) Edge collaborative task processing method, device and equipment based on Markov chain
CN111091138A (en) Irradiation forecast processing method and stacked generalization model training method and device
CN115456425A (en) Voltage adjustment method, device, equipment, storage medium and computer program product
CN115277570A (en) Flow distribution method and device, computer equipment and storage medium
CN114862615A (en) Aggregation method and device of distributed power resources and computer equipment
Abdellah et al. RAP-G: Reliability-aware service placement using genetic algorithm for deep edge computing
CN115079306B (en) Lightning prediction method, device, apparatus, storage medium and computer program product
CN114637606A (en) Task unloading method, device and equipment based on distribution substation area distribution transformation gateway
CN118095714A (en) Scheduling method and device for power service, computer equipment and storage medium
US20230352939A1 (en) Cognitive framework for improving responsivity in demand response programs
WO2022057427A1 (en) Method, apparatus and device for power prediction
CN116681203A (en) Enterprise management consultation method and system based on big data analysis
CN118034906A (en) Cloud server cluster processing method and device, computer equipment and storage medium
CN117610865A (en) Converter station calculation task processing method, device, computer equipment and storage medium
CN116366451A (en) Bandwidth adjustment method, device, computer equipment and readable storage medium
CN117293805A (en) Method, device, equipment and storage medium for determining capacity of optical storage system
CN116505516A (en) Power control method, apparatus, device, storage medium, and computer program product
CN117713048A (en) Power supply method, device and equipment for power utilization terminal based on rural power network power utilization integration
CN117077028A (en) Method, device, computer equipment and storage medium for constructing risk prediction model
CN117950833A (en) Task scheduling method, device, computer equipment and storage medium
CN117453248A (en) Gray scale issuing method, device, computer equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant