CN111833205A - Mobile charging pile group intelligent scheduling method in big data scene - Google Patents

Mobile charging pile group intelligent scheduling method in big data scene Download PDF

Info

Publication number
CN111833205A
CN111833205A CN202010583700.3A CN202010583700A CN111833205A CN 111833205 A CN111833205 A CN 111833205A CN 202010583700 A CN202010583700 A CN 202010583700A CN 111833205 A CN111833205 A CN 111833205A
Authority
CN
China
Prior art keywords
mobile charging
scheduling
charging pile
big data
battery
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
CN202010583700.3A
Other languages
Chinese (zh)
Other versions
CN111833205B (en
Inventor
彭纪昌
孟锦豪
刘海涛
刘凯龙
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.)
Nanjing Institute of Technology
Original Assignee
Nanjing Institute of Technology
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 Nanjing Institute of Technology filed Critical Nanjing Institute of Technology
Priority to CN202010583700.3A priority Critical patent/CN111833205B/en
Publication of CN111833205A publication Critical patent/CN111833205A/en
Application granted granted Critical
Publication of CN111833205B publication Critical patent/CN111833205B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Software Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Public Health (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Water Supply & Treatment (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a mobile charging pile group intelligent scheduling method in a big data scene, which comprises the following steps: establishing a scheduling model of the mobile charging pile cluster; solving the scheduling model through constraint conditions and a DQN algorithm to obtain an intelligent scheduling strategy; and scheduling the mobile charging pile according to the intelligent scheduling strategy. All mobile charging piles in the area are uniformly scheduled to participate in the electric power market auxiliary service through a scheduling model of the mobile charging pile cluster, a profit method of the mobile charging piles is increased, service is provided for a power grid, and the problems of frequency modulation of a regional power system, new energy consumption and the like are solved.

Description

Mobile charging pile group intelligent scheduling method in big data scene
Technical Field
The invention belongs to the field of charging device application, and relates to an intelligent mobile charging pile group scheduling method in a big data scene.
Background
Use the removal to fill electric pile and realize that electric automobile charges, possess high flexible, solved the fixed problem that the utilization rate of filling electric pile is low. In addition, with the explosive growth of new energy automobiles, large-scale retired power lithium batteries are generated. The mobile charging pile also becomes an important mode for exerting the residual value of the retired power lithium battery. And with the mode of the aggregator cloud platform, the idle mobile charging pile of operation participates in the electric power market auxiliary service, and the method has important significance for maximizing the value of the mobile charging pile.
To participate in the electric power market auxiliary service, a large number of mobile charging piles must be combined to form a certain scale, and then the service can be effectively provided for the regional power grid. At this time, the aggregator is required to collect the state information of each mobile charging pile, and all the charging piles are uniformly scheduled to participate in the power auxiliary service through cloud platform computing, so that benefit maximization is achieved. If a large number of mobile charging piles are contained in a specific area, due to various uncertain factors, a physical Model of a system is difficult to accurately establish, and an optimal scheduling method based on a deterministic Model, such as Model Predictive Control (MPC), is difficult to effectively implement system scheduling. The traditional reinforcement learning algorithms Q-learning, Sarsa and the like can not process continuous state variables, so that the accuracy of the scheduling strategy is restricted.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide an intelligent scheduling method for a mobile charging pile group in a big data scene so as to solve the problem that system scheduling is difficult in the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
a mobile charging pile group intelligent scheduling method in a big data scene comprises the following steps:
establishing a scheduling model of the mobile charging pile cluster;
solving the scheduling model through constraint conditions and a DQN algorithm to obtain an intelligent scheduling strategy;
and scheduling the mobile charging pile according to the intelligent scheduling strategy.
Further, the cost function of the scheduling model includes:
min J＝Bagg+EBat(7)
wherein J is a cost function of the cluster scheduling of the mobile charging piles, EBatFor the cost of lithium batteries in specific operating conditions, BaggIs the aggregate revenue for the aggregator.
Further, the aggregate revenue is as follows:
Bagg＝Bc·wc+Be·we(1)
in the formula ,BaggThe aggregate revenue for the aggregator; b iscBenefits obtained for aggregators actively participating in power market assistance services; w is acIs a division coefficient for the aggregator to actively participate in the electric power market auxiliary service; b iseProfit from energy arbitrage; w is aeThe division factor for the aggregator to profit in an energy arbitrage manner.
Further, the benefits obtained by the initiative participation of the aggregator in the electric power market assistance service are:
Figure BDA0002553072510000021
wherein ,rp and rvRespectively, the compensatory electricity prices, P, of the peaks and valleysp and PvRespectively, total power that the aggregator can provide at peak-to-valley;
the profit obtained by the energy arbitrage is as follows:
Figure BDA0002553072510000022
wherein ,Qi,tTo control the charge and discharge capacity provided in the period.
Further, the use cost of the lithium battery in the specific working condition is as follows:
Figure BDA0002553072510000031
wherein ,EBatFor the use cost of lithium battery in specific working condition，NBatTo the total number of lithium cells involved, EBat_iniFor the initial investment cost of the lithium battery,
Figure BDA0002553072510000032
is the percentage of the lithium battery decay.
Further, the percentage of the attenuation of the lithium battery is as follows:
Figure BDA0002553072510000033
wherein ,CBatAs capacity of the current battery, CEolThe corresponding capacity at the end of the battery life, CinitIs the initial capacity of the battery.
Further, the current battery capacity is:
CBat＝a·nc b+c (4)
wherein ,CBatIs the current capacity of the battery, ncThe number of cycles is a coefficient of a power function, b is the number of times of the power function, and c is an offset.
Further, the constraint condition includes a moving range constraint:
Figure BDA0002553072510000034
wherein ,
Figure BDA0002553072510000035
for the current movement of the distance of movement of the charging pile, LmaxThe maximum allowable moving distance of the mobile charging pile is obtained;
the number of the mobile charging piles is restricted:
Figure BDA0002553072510000036
wherein ,
Figure BDA0002553072510000037
unifying for participating aggregatorsNumber of mobile charging piles scheduled, NmaxThe maximum allowable number of the mobile charging piles which can participate in unified scheduling is determined;
power constraint of the mobile charging pile:
Figure BDA0002553072510000038
in the formula ,Pch and PdisThe rechargeable and discharge power is allowed for the battery energy storage unit respectively;
Figure BDA0002553072510000039
and
Figure BDA00025530725100000310
the maximum allowable charging and discharging power of the battery energy storage unit is respectively the maximum allowable charging and discharging power of the battery energy storage unit;
capacity constraint of the mobile charging pile:
Figure BDA0002553072510000041
wherein ,
Figure BDA0002553072510000042
and
Figure BDA0002553072510000043
respectively the upper and lower limit values of the state of charge of the battery energy storage unit.
Further, the DQN algorithm includes:
initializing the observation Q(s)t,at)、
Figure BDA0002553072510000044
And a discount factor;
selecting a scheduling policy a for probabilitytViewing the profit r of the systemtAnd state st+1；
Storing(s)t,at,rt,st+1) To the playback memory unit D;
randomly extracting from DTaking the appropriate amount of learning experience(s)t,at,rt,st+1) Training a target neural network;
training a current neural network by a minimum loss function by adopting a gradient descent method;
copying the current neural network parameters to a target neural network every N time windows;
repeating the above steps until the state stTo the target expectation
Figure BDA0002553072510000045
And finishing the algorithm.
Further, the output result of the neural network training is:
Figure BDA0002553072510000046
the minimization loss function is:
(yj-Q(sj,aj|θ))2，
wherein ,rjIs j reward, gamma is discount factor, Q is observed value, theta is parameter of neural network, sjIs the j-th state, ajIs the j-th action.
A mobile charging pile group intelligent scheduling system under big data scene, the system includes:
a scheduling model module: the scheduling model is used for establishing a mobile charging pile cluster;
a solving module: the system comprises a scheduling model, a constraint condition and a DQN algorithm, wherein the scheduling model is used for solving through the constraint condition and the DQN algorithm to obtain an intelligent scheduling strategy;
a scheduling module: and the mobile charging pile is scheduled according to the intelligent scheduling strategy.
A mobile charging pile group intelligent scheduling system under a big data scene comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method described above.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method described above.
Compared with the prior art, the invention has the following beneficial effects:
all mobile charging piles in the area are uniformly scheduled to participate in the electric power market auxiliary service through a scheduling model of the mobile charging pile cluster, so that a profit method of the mobile charging piles is increased, the power grid is also served, the problems of frequency modulation of a regional power system, new energy consumption and the like are solved, and the method is a win-win operation mode; the invention provides a group intelligent scheduling strategy for applying the reinforcement learning DQN to the mobile charging pile in a big data application scene, and can help a aggregator cloud platform to make a real-time scheduling decision to obtain the maximum benefit.
Drawings
Fig. 1 is an aggregator-based mobile charging pile population scheduling cloud platform system architecture in an embodiment of the invention;
fig. 2 is an overall framework of the reinforcement learning DQN algorithm in an embodiment of the invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
Firstly, establishing a model for scheduling a mobile charging pile group in an aggregator mode, and designing corresponding constraint conditions; on the basis, a cost model for dispatching the mobile charging piles is established; and finally, a DQN algorithm applied to the cluster scheduling of the mobile charging piles is designed, and an intelligent scheduling strategy of the mobile charging pile cluster is obtained.
Fig. 1 shows a system structure for scheduling mobile charging piles in a certain area to participate in an electric power market auxiliary service in an aggregator manner. The method mainly comprises the following steps: the system comprises important components such as a mobile charging pile, a convergence business cloud platform and a regional power grid. The removal fills electric pile contains: the energy storage unit is used for storing and releasing electric energy, and the autonomous mobile unit can move to the target vehicle to provide charging service for the target vehicle, and can also follow the scheduling instruction of the cloud platform to participate in the electric power market auxiliary service. In addition, the mobile charging post must also include a wireless communication device for data exchange with the cloud platform. The mobile charging pile can work in a service state and an idle state, the service state is used for charging the electric automobile, and the rest time is in the idle state and can be used for exchanging energy with a power grid to provide auxiliary service for a regional power grid. The electric pile is filled in each removal of aggregator with the help of the unified scheduling of cloud platform, and the cloud platform needs to collect the real-time status that each removal was filled electric pile, includes: state of charge, power state, idle state, scheduling cost, etc. And then, obtaining an optimized scheduling strategy by using the DQN according to the electricity price information of the regional power grid and the like, and realizing maximization of the group benefit of the mobile charging piles. And the regional power grid comprehensively judges according to the local load and the new energy consumption condition to formulate regional electricity prices.
1) And aggregating revenue calculation of the merchant cloud platform. The benefits of the aggregator mainly include benefits participating in auxiliary services of the power market and energy arbitrage realization in a high generation and low storage mode according to the real-time electricity price of the regional power grid. Thus, the revenue for an aggregator can be expressed as:
Bagg＝Bc·wc+Be·we(1)
in the formula ,BaggThe aggregate revenue for the aggregator; b iscBenefits obtained for aggregators actively participating in power market assistance services; w is acIs a division coefficient for the aggregator to actively participate in the electric power market auxiliary service; b iseProfit from energy arbitrage; w is aeThe division factor for the aggregator to profit in an energy arbitrage manner.
Respectively defining B according to the profit modes of the aggregation business cloud platformc and BeThe following were used:
Figure BDA0002553072510000071
Figure BDA0002553072510000072
wherein ,rp and rvRespectively, the compensatory electricity prices, P, of the peaks and valleysp and PvTotal power, Q, respectively, available to the aggregator at peak-to-valleyi,tTo control the charge and discharge capacity provided in the period.
2) The use cost of electric pile is filled in the removal. The type of the battery selected by the mobile charging pile is determined, and the cyclic accelerated aging test is performed through a plurality of lithium batteries of the same type. According to the embodiment of the invention, the actual working condition of the auxiliary service participating in the power market is taken as the cycle working condition, and the relationship between the battery capacity decline and the cycle times is obtained by collecting data through long-period aging tests. The invention obtains the cycle life empirical model of the lithium battery through superposition of a power function and a linear function and nonlinear least square fitting, and the cycle life empirical model is specifically shown as the following formula:
CBat＝a·nc b+c (4)
wherein ,CBatIs the current capacity of the battery, ncAnd a, b and c are parameters to be fitted.
In actual operation, the circulation times can be obtained by counting by a rain flow method. The rules of the rain flow counting method include: (1) setting the initial value as a maximum value; (2) the rain flow flows downwards from the inner side of each peak in turn and falls at the next peak until there is a peak opposite which is larger than its starting point; (3) when the rain flow encounters a rain flow under the upper layer flow, it stops immediately. The cycle times corresponding to the specific working conditions can be calculated through the steps.
According to the calculated current battery capacity CBatThe percentage of lithium battery decay can be calculated from the following equation:
Figure BDA0002553072510000073
in the formula ,CBatAs capacity of the current battery, CEolThe corresponding capacity at the end of the battery life, CinitIs the initial capacity of the battery.
On the basis of obtaining the attenuation percentage of the lithium battery, the initial test investment cost of the battery can be spread to each cycle working condition, and the scheduling cost of the battery is obtained, and the scheduling cost is specifically shown as the following formula:
Figure BDA0002553072510000081
in the formula ,EBatFor the cost of use of lithium batteries in specific operating conditions, NBatTo the total number of lithium cells involved, EBat_iniThe initial investment cost of the lithium battery is reduced.
3) And (5) a scheduling model of the mobile charging pile cluster. By analyzing the system structure and the income condition, the following objective function can be set for scheduling the mobile charging pile cluster.
min J＝Bagg+EBat(7)
The benefit condition of the convergence merchant cloud platform and the scheduling cost of the mobile charging pile energy storage unit are comprehensively considered in the above formula. In addition, the unified scheduling mobile charging pile also needs to consider the following constraints.
And limiting the operation range of the mobile charging pile. For each mobile charging pile, the mobile charging pile cannot move in a wireless large range, and the general moving range of the mobile charging pile is limited to a certain range, namely the unified scheduling needs to meet the following conditions:
Figure BDA0002553072510000082
wherein ,
Figure BDA0002553072510000083
for the current movement of the distance of movement of the charging pile, LmaxThe maximum allowable moving distance of the mobile charging pile.
And limiting the quantity of the mobile charging piles participating in unified scheduling. Due to the restriction of various factors such as fields and cable power in practical application, the number of the mobile charging piles participating in unified scheduling has an upper limit. The number of the mobile charging piles which can participate in the unified scheduling of the aggregator meets the following constraint:
Figure BDA0002553072510000084
wherein ,
Figure BDA0002553072510000085
number of mobile charging piles for participating in uniform scheduling of aggregators, NmaxThe maximum allowed number of mobile charging piles which can participate in unified scheduling.
In addition, the power limit of each mobile charging pile can be expressed as follows:
Figure BDA0002553072510000091
in the formula ,
Figure BDA0002553072510000092
and
Figure BDA0002553072510000093
the maximum allowable charging and discharging power of the battery energy storage unit is respectively.
The capacity limit of the mobile charging pile can be expressed as:
Figure BDA0002553072510000094
wherein ,
Figure BDA0002553072510000095
and
Figure BDA0002553072510000096
respectively the upper and lower limit values of the state of charge of the battery energy storage unit.
4) And intelligently scheduling the mobile charging pile cluster. According to the method, the DQN is adopted, and the intelligent scheduling of the mobile charging pile group under the big data scene is realized in a mode of combining reinforcement learning and deep learning. Firstly, solving an optimization problem with constraints by utilizing quadratic programming through an established system model to obtain coarse time scale optimization of day-ahead scheduling. In this embodiment, to implement long-time-scale day-ahead scheduling, dynamic electricity prices issued by electricity-selling companies are combined, and the equations (7) are used as objective functions, and the equations (8) to (11) are selected as constraint conditions to perform long-time-scale mobile charging pile cluster scheduling.
On the basis, the DQN method is selected to enable the whole scheduling strategy to gradually tend to be optimized through reinforcement learning, and 1 hour is taken as a time interval to perform finer scheduling. The main structure of the DQN algorithm is shown in fig. 2, and includes: context, action a, observation Q, reward r, etc. Reinforcement learning establishes a link between the current state and the action to be taken in a data-driven manner by taking action, interacting with the environment, observing results and obtaining rewards. Compared with traditional reinforcement learning Q-learning and Sarsa, the DQN used by the method can realize the processing of continuous states due to the use of the neural network.
Defining the reward r of the reinforcement learning DQN as an objective function (7), defining the state s as the charging and discharging capacity required by the auxiliary service of the power market in the planning in the day ahead, defining the action a as the specific strategy used in each scheduling, and using a DQN algorithm to complete the unified scheduling steps of the short-time-scale charging piles as follows:
step1. Arbitrary initialization Q Table Q(s)t,at)、
Figure BDA0002553072510000101
And the value of the discount factor gamma is between 0 and 1.
Step2. selection of scheduling policy a for probabilitytViewing the profit r of the systemtAnd state st+1；
Step3. store(s)t,at,rt,st+1) To the playback memory unit D;
random extraction of a few learning experiences from D(s)t,at,rt,st+1) The method is used for training the target neural network, and defines the following target neural network training output results:
Figure BDA0002553072510000102
step5. minimizing the loss function (y) by gradient descent methodj-Q(sj,aj|θ))2Training a current neural network, wherein theta is a parameter of the neural network;
and step6, copying the current neural network parameters to the target neural network every N time windows.
Step7, repeating the steps from Step2 to Step6 until the state stTo the target expectation
Figure BDA0002553072510000103
And finishing the algorithm.
Through the training process of the DQN reinforcement learning, the mobile charging pile cluster intelligent scheduling strategy meeting the benefit maximization of the aggregator cloud platform can be obtained, and the optimal scheduling of the mobile charging pile cluster within 1 hour time scale is achieved.
In the face of a big data application scene after future popularization of the mobile charging piles, Deep learning and reinforcement learning are combined by Deep Qnetwork (DQN), a large amount of data are directly used for driving, and the intelligent mobile charging pile group scheduling strategy based on DQN is obtained through continuous enhancement of training under the big data scene, so that idle mobile charging piles actively participate in electric power market auxiliary service to obtain profits, and economic benefits of operators are effectively improved.
In order to realize the intelligent dispatching of the mobile charging pile cluster in a certain scale, the invention needs an optimized dispatching algorithm capable of processing large data information containing different state quantities. Therefore, the DQN-based mobile charging pile group intelligent scheduling method is provided, deep learning and reinforcement learning can be combined, big data scenes can be responded, continuous state variables are processed, and a unified intelligent scheduling strategy of a large number of mobile charging piles in a aggregator cloud platform mode is obtained.
A mobile charging pile group intelligent scheduling system under big data scene, the system includes:
a scheduling model module: the scheduling model is used for establishing a mobile charging pile cluster;
a solving module: the system comprises a scheduling model, a constraint condition and a DQN algorithm, wherein the scheduling model is used for solving through the constraint condition and the DQN algorithm to obtain an intelligent scheduling strategy;
a scheduling module: and the mobile charging pile is scheduled according to the intelligent scheduling strategy.
A mobile charging pile group intelligent scheduling system under a big data scene comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method described above.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method described above.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The present invention is not limited to the above embodiments, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are included in the scope of the claims of the present invention which are filed as the application.

Claims (10)

1. A mobile charging pile group intelligent scheduling method in a big data scene is characterized by comprising the following steps:
establishing a scheduling model of the mobile charging pile cluster;
solving the scheduling model through constraint conditions and a DQN algorithm to obtain an intelligent scheduling strategy;
and scheduling the mobile charging pile according to the intelligent scheduling strategy.
2. The method for intelligent scheduling of the mobile charging pile group in the big data scene according to claim 1, wherein the cost function of the scheduling model comprises:
min J＝Bagg+EBat(7)
wherein J is a cost function of the cluster scheduling of the mobile charging piles, EBatFor the cost of lithium batteries in specific operating conditions, BaggIs the aggregate revenue for the aggregator.
3. The method for intelligent scheduling of the mobile charging pile group in the big data scene according to claim 2, wherein the aggregate revenue is as follows:
Bagg＝Bc·wc+Be·we(1)
in the formula ,BaggThe aggregate revenue for the aggregator; b iscBenefits obtained for aggregators actively participating in power market assistance services; w is acIs a division coefficient for the aggregator to actively participate in the electric power market auxiliary service; b iseProfit from energy arbitrage; w is aeThe division factor for the aggregator to profit in an energy arbitrage manner.
4. The method for intelligent scheduling of the mobile charging pile group in the big data scene according to claim 3, wherein the benefits obtained by the aggregator actively participating in the electric power market auxiliary service are as follows:
Figure FDA0002553072500000011
wherein ,rp and rvRespectively, the compensatory electricity prices, P, of the peaks and valleysp and PvRespectively, total power that the aggregator can provide at peak-to-valley;
the profit obtained by the energy arbitrage is as follows:
Figure FDA0002553072500000021
wherein ,Qi,tTo control the charge and discharge capacity provided in the period.
5. The method for intelligently scheduling the mobile charging pile group in the big data scene according to claim 2, wherein the use cost of the lithium battery in the specific working condition is as follows:
Figure FDA0002553072500000022
wherein ,EBatFor the cost of use of lithium batteries in specific operating conditions, NBatTo the total number of lithium cells involved, EBat_iniFor the initial investment cost of the lithium battery,
Figure FDA0002553072500000023
is the percentage of the lithium battery decay.
6. The method for intelligently scheduling the mobile charging pile group in the big data scene according to claim 5, wherein the percentage of the attenuation of the lithium battery is as follows:
Figure FDA0002553072500000024
wherein ,CBatAs capacity of the current battery, CEolThe corresponding capacity at the end of the battery life, CinitIs the initial capacity of the battery.
7. The method for intelligently scheduling the mobile charging pile group in the big data scene according to claim 6, wherein the current battery capacity is:
CBat＝a·nc b+c (4)
wherein ,CBatIs the current capacity of the battery, ncThe number of cycles is a coefficient of a power function, b is the number of times of the power function, and c is an offset.
8. The method for intelligent scheduling of the mobile charging pile group in the big data scene according to claim 1, wherein the constraint condition includes a moving range constraint:
Figure FDA0002553072500000025
wherein ,
Figure FDA0002553072500000026
for the current movement of the distance of movement of the charging pile, LmaxThe maximum allowable moving distance of the mobile charging pile is obtained;
the number of the mobile charging piles is restricted:
Figure FDA0002553072500000031
wherein ,
Figure FDA0002553072500000032
number of mobile charging piles for participating in uniform scheduling of aggregators, NmaxThe maximum allowable number of the mobile charging piles which can participate in unified scheduling is determined;
power constraint of the mobile charging pile:
Figure FDA0002553072500000033
in the formula ,Pch and PdisThe rechargeable and discharge power is allowed for the battery energy storage unit respectively;
Figure FDA0002553072500000034
and
Figure FDA0002553072500000035
the maximum allowable charging and discharging power of the battery energy storage unit is respectively the maximum allowable charging and discharging power of the battery energy storage unit;
capacity constraint of the mobile charging pile:
Figure FDA0002553072500000036
wherein ,
Figure FDA0002553072500000037
and
Figure FDA0002553072500000038
respectively the upper and lower limit values of the state of charge of the battery energy storage unit.
9. The method of claim 1, wherein the DQN algorithm comprises:
initializing the observation Q(s)t,at)、
Figure FDA0002553072500000039
And a discount factor gamma;
selecting a scheduling policy a for probabilitytViewing the profit r of the systemtAnd state st+1；
Storing(s)t,at,rt,st+1) To the playback memory unit D;
randomly extracting an appropriate amount of learning experience(s) from Dt,at,rt,st+1) Training a target neural network;
training a current neural network by a minimum loss function by adopting a gradient descent method;
copying the current neural network parameters to a target neural network every N time windows;
repeating the above steps until the state stTo the target expectation
Figure FDA00025530725000000310
And finishing the algorithm.
10. The method for intelligent scheduling of the mobile charging pile group in the big data scene according to claim 9, wherein the neural network training output result is:
Figure FDA00025530725000000311
the minimization loss function is:
(yj-Q(sj,aj|θ))2，
wherein ,rjIs j reward, gamma is discount factor, Q is observed value, theta is parameter of neural network, sjIs the j-th state, ajIs the j-th action.
CN202010583700.3A 2020-06-23 2020-06-23 Intelligent scheduling method for mobile charging pile group under big data scene Active CN111833205B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010583700.3A CN111833205B (en) 2020-06-23 2020-06-23 Intelligent scheduling method for mobile charging pile group under big data scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010583700.3A CN111833205B (en) 2020-06-23 2020-06-23 Intelligent scheduling method for mobile charging pile group under big data scene

Publications (2)

Publication Number Publication Date
CN111833205A true CN111833205A (en) 2020-10-27
CN111833205B CN111833205B (en) 2023-09-22

Family

ID=72899398

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010583700.3A Active CN111833205B (en) 2020-06-23 2020-06-23 Intelligent scheduling method for mobile charging pile group under big data scene

Country Status (1)

Country Link
CN (1) CN111833205B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112632380A (en) * 2020-12-24 2021-04-09 北京百度网讯科技有限公司 Training method of interest point recommendation model and interest point recommendation method
CN112668874A (en) * 2020-12-25 2021-04-16 天津大学 Electric vehicle cluster charging cooperative scheduling method participating in power grid peak shaving frequency modulation
CN113131584A (en) * 2021-04-26 2021-07-16 国家电网有限公司信息通信分公司 Data center battery charging and discharging optimization control method and device

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205911764U (en) * 2016-05-27 2017-01-25 易电通（北京）储能科技有限公司 Intelligent city network system that charges
CN106557872A (en) * 2016-11-10 2017-04-05 浙江工业大学 Many parking stall intelligent three-phase charging group charging systems and method
CN106849348A (en) * 2016-11-11 2017-06-13 江苏中科瀚星数据科技有限公司 A kind of intelligent power distribution equipment and intelligent power distribution method for platform area charging pile
CN107640049A (en) * 2017-09-20 2018-01-30 东北大学 The mobile terminal system that a kind of electric automobile for charging station charges in order
CN207459802U (en) * 2017-12-07 2018-06-05 永州市键特科技有限公司 A kind of power-supply controller of electric for charging pile
CN108183473A (en) * 2017-12-13 2018-06-19 国网上海市电力公司 A kind of cluster electric vehicle participates in the optimization Bidding system of assisted hatching
CN108197765A (en) * 2018-03-23 2018-06-22 华北电力大学 The parking lot charging schedule method and computing device distributed towards battery loss single-candidate
CN109687530A (en) * 2019-01-08 2019-04-26 南京工程学院 A kind of power grid mixing rolling scheduling method considering obstruction and energy storage tou power price
TW202005838A (en) * 2018-07-12 2020-02-01 岳鼎股份有限公司 Control method of charging pile system for effectively improving defects of manually turning on/off of charging pile
CN110866636A (en) * 2019-11-06 2020-03-06 南京工程学院 Microgrid planning method comprehensively considering electric vehicle charging station and distributed energy

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205911764U (en) * 2016-05-27 2017-01-25 易电通（北京）储能科技有限公司 Intelligent city network system that charges
CN106557872A (en) * 2016-11-10 2017-04-05 浙江工业大学 Many parking stall intelligent three-phase charging group charging systems and method
CN106849348A (en) * 2016-11-11 2017-06-13 江苏中科瀚星数据科技有限公司 A kind of intelligent power distribution equipment and intelligent power distribution method for platform area charging pile
CN107640049A (en) * 2017-09-20 2018-01-30 东北大学 The mobile terminal system that a kind of electric automobile for charging station charges in order
CN207459802U (en) * 2017-12-07 2018-06-05 永州市键特科技有限公司 A kind of power-supply controller of electric for charging pile
CN108183473A (en) * 2017-12-13 2018-06-19 国网上海市电力公司 A kind of cluster electric vehicle participates in the optimization Bidding system of assisted hatching
CN108197765A (en) * 2018-03-23 2018-06-22 华北电力大学 The parking lot charging schedule method and computing device distributed towards battery loss single-candidate
TW202005838A (en) * 2018-07-12 2020-02-01 岳鼎股份有限公司 Control method of charging pile system for effectively improving defects of manually turning on/off of charging pile
CN109687530A (en) * 2019-01-08 2019-04-26 南京工程学院 A kind of power grid mixing rolling scheduling method considering obstruction and energy storage tou power price
CN110866636A (en) * 2019-11-06 2020-03-06 南京工程学院 Microgrid planning method comprehensively considering electric vehicle charging station and distributed energy

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
孟高军;余海涛;胡敏强;刘海涛;酒晨霄;: "一种基于非线性反馈重复控制策略的磁通切换直线电机推力波动抑制方法", 电工技术学报, no. 08 *
陈静鹏;艾芊;肖斐;: "基于集群响应的规模化电动汽车充电优化调度", 电力系统自动化, no. 22 *
黄宇;杨健维;何正友;: "基于双层离散粒子群优化的智能小区车辆与家庭互动调度策略", 电网技术, no. 10 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112632380A (en) * 2020-12-24 2021-04-09 北京百度网讯科技有限公司 Training method of interest point recommendation model and interest point recommendation method
CN112668874A (en) * 2020-12-25 2021-04-16 天津大学 Electric vehicle cluster charging cooperative scheduling method participating in power grid peak shaving frequency modulation
CN112668874B (en) * 2020-12-25 2022-08-26 天津大学 Electric vehicle cluster charging cooperative scheduling method participating in power grid peak shaving frequency modulation
CN113131584A (en) * 2021-04-26 2021-07-16 国家电网有限公司信息通信分公司 Data center battery charging and discharging optimization control method and device

Also Published As

Publication number Publication date
CN111833205B (en) 2023-09-22

Similar Documents

Publication Publication Date Title
Ahmad et al. A cost-efficient energy management system for battery swapping station
CN103969585B (en) Assess method and apparatus, related system and the vehicle of the behaviour in service of battery
Wang et al. Hybrid centralized-decentralized (HCD) charging control of electric vehicles
CN111833205A (en) Mobile charging pile group intelligent scheduling method in big data scene
Alfaverh et al. Optimal vehicle-to-grid control for supplementary frequency regulation using deep reinforcement learning
CN114997631B (en) Electric vehicle charging scheduling method, device, equipment and medium
Vayá et al. Optimal bidding of plug-in electric vehicles in a market-based control setup
CN111738611B (en) Intelligent scheduling method for mobile charging pile group based on Sarsa algorithm
CN113794199A (en) Maximum profit optimization method of wind power energy storage system considering electric power market fluctuation
CN116896086A (en) Virtual power plant adjustable resource regulation and control system and method considering demand response
Ren et al. Study on optimal V2G pricing strategy under multi-aggregator competition based on game theory
CN106873552B (en) The electric car charging monitoring system and method for group's intarconnected cotrol
Rahman et al. On efficient operation of a V2G-enabled virtual power plant: when solar power meets bidirectional electric vehicle charging
Zhang et al. Transfer deep reinforcement learning-based large-scale V2G continuous charging coordination with renewable energy sources
CN117057547A (en) Method, device and storage medium for constructing multi-form load resource scheduling model of intelligent energy service platform
CN111244938A (en) Source network load storage coordination control method, device and system applied to power grid
CN110341537A (en) A kind of vehicle-mounted bidirectional charger charge control strategy based on Model Predictive Control
Yamashita et al. Hierarchical model predictive control to coordinate a vehicle-to-grid system coupled to building microgrids
Avdevicius et al. Bus charging management based on AI prediction and MILP optimization
Tajeddini et al. Decentralized charging coordination of plug-in electric vehicles based on reverse stackelberg game
Wang et al. Predictive management of electric vehicles in a community microgrid
CN112109580A (en) Micro-grid electric automobile charge and discharge control system with electric quantity self-distribution function
Scarabaggio et al. On Controlling Battery Degradation in Vehicle-to-Grid Energy Markets
CN117081059B (en) Optimal control method, device, equipment and medium for charging and replacing power station cluster
Bjurek et al. Vehicle-to-Everything Optimization Considering Battery Degradation

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