CN114529207A - Energy storage battery distributed economic dispatching method based on differential privacy mechanism - Google Patents

Energy storage battery distributed economic dispatching method based on differential privacy mechanism Download PDF

Info

Publication number
CN114529207A
CN114529207A CN202210163557.1A CN202210163557A CN114529207A CN 114529207 A CN114529207 A CN 114529207A CN 202210163557 A CN202210163557 A CN 202210163557A CN 114529207 A CN114529207 A CN 114529207A
Authority
CN
China
Prior art keywords
energy storage
distributed
algorithm
privacy
storage 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
CN202210163557.1A
Other languages
Chinese (zh)
Other versions
CN114529207B (en
Inventor
蔡德福
王作维
周鲲鹏
陈霞
陈汝斯
刘海光
王文娜
万黎
王涛
张良一
孙冠群
王尔玺
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.)
Huazhong University of Science and Technology
State Grid Corp of China SGCC
Electric Power Research Institute of State Grid Hubei Electric Power Co Ltd
Original Assignee
Huazhong University of Science and Technology
State Grid Corp of China SGCC
Electric Power Research Institute of State Grid Hubei Electric Power 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 Huazhong University of Science and Technology, State Grid Corp of China SGCC, Electric Power Research Institute of State Grid Hubei Electric Power Co Ltd filed Critical Huazhong University of Science and Technology
Priority to CN202210163557.1A priority Critical patent/CN114529207B/en
Publication of CN114529207A publication Critical patent/CN114529207A/en
Application granted granted Critical
Publication of CN114529207B publication Critical patent/CN114529207B/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
    • 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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
    • 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
    • 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)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Bioethics (AREA)
  • Water Supply & Treatment (AREA)
  • Computer Security & Cryptography (AREA)
  • Databases & Information Systems (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention discloses an energy storage battery distributed economic dispatching method based on a differential privacy mechanism, and belongs to the technical field of electrical engineering. The invention solves the energy storage battery economic dispatching model by adopting a distributed method based on a consistency algorithm, and solves the output of each energy storage battery when the social benefit is maximized through iterative computation. The implementation of the distributed method relies on the communication between each energy storage and the power consumer, with the danger of stolen information. A differential privacy mechanism is introduced into the distributed method, noise masking is carried out on the transmitted information, the privacy information of the energy storage and power users is protected, and after the differential privacy algorithm is added, iterative convergence of the distributed method is not affected. After the differential privacy algorithm is added, the convergence precision and the privacy protection degree of the distributed method are related to the added noise signal, the convergence precision and the privacy protection degree are in negative correlation, and balance can be carried out according to the requirements of the convergence precision and the privacy protection degree in practical application.

Description

Energy storage battery distributed economic dispatching method based on differential privacy mechanism
Technical Field
The invention belongs to the technical field of electrical engineering, and particularly relates to an energy storage battery distributed economic dispatching method based on a differential privacy mechanism.
Background
In recent years, with the gradual increase of the proportion of renewable energy in a power system, energy storage is also widely applied to a power grid. The battery energy storage system has the characteristics of high response speed, wide adjustment range and easiness in bidirectional adjustment, can provide flexibility for a power grid, improves the operation management capability of the power grid, is a main power unit in an energy storage power station, has stable performance, flexibility in control and accurate and quick response, and has better response characteristic compared with the traditional thermal power unit.
The design of the energy storage battery economic dispatching method depends on communication technology, and can be divided into a centralized method and a distributed method according to different communication structures. Compared with a centralized method, the distributed method does not depend on a centralized controller, and has wide application in economic scheduling due to the advantages of expandability and flexibility. With the increasing of distributed power sources and power consumers, the real information content contained in data transmitted in a system communication network is significantly increased, and attention needs to be paid to the privacy protection problem.
Disclosure of Invention
Aiming at the problem that privacy protection in the distributed economic dispatching method of the energy storage battery cannot be guaranteed in the prior art, the invention provides the distributed economic dispatching method of the energy storage battery based on a differential privacy mechanism, and aims to solve the problem of energy storage economic dispatching by using a distributed method and carry out noise masking on communication information related to a distributed algorithm of economic dispatching; the method has the advantages that the balance analysis is carried out between the protection degree of the private information and the convergence precision of the algorithm, so that the protection of all the private information can be realized, and the convergence precision of the distributed economic dispatching algorithm are ensured.
In order to achieve the purpose, the invention adopts the following technical scheme:
a distributed economic dispatching method of an energy storage battery based on a differential privacy mechanism comprises the following steps:
step 1, establishing an energy storage battery economic dispatching model;
step 2, designing a distributed economic dispatching method according to a consistency algorithm on the basis of the established economic dispatching model;
and 3, introducing a differential privacy mechanism into the proposed distributed economic dispatching method, and iteratively solving the energy storage battery economic dispatching model according to an iterative formula after the differential privacy mechanism is introduced.
Further, said step 2 is to design the distributed economic dispatching method according to the consistency algorithm, and the method hasThe body content is as follows: selecting marginal cost lambda of each energy storage battery outputiAnd selecting the power of each energy storage battery as a feedback variable as a consistency variable, designing an adjacent matrix W of the communication network topology, designing an iteration rule of a distributed algorithm according to the consistency algorithm, and finally, iteratively solving an energy storage economic dispatching model.
Further, the adjacency matrix W ═ Wij}∈Rn×nThe connection number of each node in a communication link of the energy storage battery economic dispatching model is obtained, and the expression is as follows:
Figure BDA0003515698730000021
wherein N isiSet of neighboring nodes, n, for node ii、njThe number of connections for nodes i, j, respectively.
Further, the distributed algorithm 1 with coefficient iteration designed according to the consistency algorithm has the following iterative formula:
Figure BDA0003515698730000031
Figure BDA0003515698730000032
pi(t+1)＝φ(pvi(t+1))
si(t+1)＝pvi(t+1)-pi(t+1)
wherein λi(t)、λj(t) the marginal cost, w, of the energy storage cells i, j at time t, respectivelyijFor the elements of the adjacency matrix W corresponding to the communication link nodes i, j, μ1To control the gain, aiAnd biIs the benefit coefficient, p, in the energy storage economic dispatch modelvi(t) is an estimate of the optimum power command, pi(t) is the actual power command after clipping, si(t) is pvi(t) and pi(t) difference between sj(t) is pvj(t) and pj(t), wherein the clipping function is:
Figure BDA0003515698730000033
wherein
Figure BDA0003515698730000034
The maximum value and the minimum value of the stored energy charging and discharging power are respectively.
Further, the distributed algorithm 2 designed according to the consensus algorithm and not containing coefficient iteration has the following iterative formula:
Figure BDA0003515698730000035
Figure BDA0003515698730000036
pi(t+1)＝φ(pvi(t+1))
Figure BDA0003515698730000037
wherein λi(t)、λj(t) marginal cost of energy storage cells i, j at time t, miThe difference between the power supply and demand obtained for the energy storage cell i, characterizes the power balance of the system, wijFor the elements of the adjacency matrix W corresponding to the communication link nodes i, j, μ2To control the gain, aiAnd biIs the benefit coefficient, p, in the energy storage economic dispatch modelvi(t) is an estimate of the optimum power command, pi(t) is the actual power command after clipping, si(t) is pvi(t) and pi(t) difference between sj(t) is pvj(t) and pj(t) difference between (t).
Further, step 3 introduces a differential privacy mechanism to protect the privacy information, specifically, the iteration rule of the distributed algorithm is adjusted, and the iteration formula of the adjusted distributed algorithm 1 is as follows:
Figure BDA0003515698730000041
xj(t)＝λi(t)+ηj(t)
wherein etai(t) is an attenuated noise signal satisfying a Rayleigh distribution
Figure BDA0003515698730000042
qi∈(0，1)，ciFor the scaling factor, the corresponding probability density function is as follows:
Figure BDA0003515698730000043
at the same time, for pvi，piAnd siThe update rule of (2) is not changed.
Further, a differential privacy mechanism is introduced in the step 3 to protect the privacy information, specifically, the iteration rule of the distributed algorithm is adjusted, and the iteration formula of the adjusted distributed algorithm 2 is as follows:
Figure BDA0003515698730000044
Figure BDA0003515698730000045
xj(t)＝λj(t)+ηj(t)
yj(t)＝mj(t)+ηj(t)。
further, the convergence value of the distributed algorithm 1 finally falls within a range of an upper and lower deviation r based on the optimal value λ':
P{|λ′(∞)-λ′*|＜r}≥1-V(λ′∞)/r2
let r be the convergence accuracy range, where r is
Figure BDA0003515698730000056
V(λ′∞) Is lambda'∞The variance of (a) is determined,
Figure BDA0003515698730000051
further, the convergence value of the distributed algorithm 2 finally falls within a range of an upper and lower deviation r based on the optimal value λ':
P{|λ′(∞)-λ′*|＜r}≥1-V(λ′∞)/r2
let r be the convergence accuracy range, where r is
Figure BDA0003515698730000052
V(λ′∞) Is lambda'∞The variance of (a) is determined,
Figure BDA0003515698730000053
further, for delta-neighbor set x1And x2And observing the sequence O when the Alg algorithm is satisfied
P[Alg(x1(0))∈O]≤eεδP[Alg(x2(0))∈O]
Then, the Alg algorithm is called to meet epsilon-difference privacy, wherein epsilon is the privacy budget;
the privacy budget of any energy storage unit i in the distributed algorithm 1 is as follows:
Figure BDA0003515698730000054
the privacy budget of any energy storage unit i in the distributed algorithm 2 is as follows:
Figure BDA0003515698730000055
coefficient q of noise signaliAnd increasing the upper and lower ranges r of the convergence value, and reducing the privacy budget, which indicates that the privacy protection degree is increased, and the convergence precision is reduced, namely the algorithm realizes privacy protection by sacrificing the convergence precision.
Generally, by the above technical solution conceived by the present invention, the following beneficial effects can be obtained:
(1) aiming at the problem of solving the energy storage economic dispatching model, the distributed algorithm based on the differential privacy mechanism is introduced, so that the privacy information can be ensured to be protected while the energy storage economic dispatching model is subjected to iterative solving. The distributed economic dispatching method is designed based on a consistency algorithm, and marginal cost lambda is selectediAnd selecting the power of each energy storage battery as a feedback variable as a consistency variable. Distributed algorithms relying on communication networks may be eavesdropped on communication information, and an eavesdropper may further acquire important privacy information. When the differential privacy algorithm is used for communication between the energy storages, noise is added to information transmitted to the adjacent nodes to cover real information, and all privacy information is protected. Meanwhile, the iteration rules of the distributed algorithm after being adjusted by the differential privacy algorithm can still be converged to be consistent, namely, the added noise does not influence the solution of the energy storage economic dispatching model.
(2) Aiming at the evaluation calculation of the convergence precision and the privacy information protection degree of the distributed economic dispatching method of the energy storage battery after the addition of the differential privacy algorithm, the invention defines the convergence precision of the used distributed algorithm, namely the convergence value of the algorithm falls in the range of taking the optimal value lambda'. as the reference and taking the upper deviation and the lower deviation as r, and provides a formula of the convergence precision range r; meanwhile, in order to analyze the protection performance of the proposed algorithm, the protection degree of the algorithm on a certain energy storage unit i participating in communication is defined, namely the algorithm meets epsilon-difference privacy, and privacy prediction is givenAnd (4) calculating an epsilon formula. The formula of r and epsilon shows the coefficient q of the noise signal introduced when the privacy-differentiating algorithm is usediWhen the convergence precision range r is increased, the privacy budget epsilon is reduced, which shows that the differential privacy algorithm realizes privacy protection by sacrificing the convergence precision. In practical application, the noise coefficient q can be adjusted according to the balance of the system on the privacy protection degree and the convergence precisioni。
Drawings
Fig. 1 is a schematic flowchart of a distributed economic dispatching method for energy storage batteries based on a differential privacy mechanism according to the present invention;
fig. 2 is a schematic diagram of an energy storage battery network topology and communication links provided by an embodiment of the present invention;
FIG. 3 shows the result of iterative solution using the distributed algorithm 1 based on the differential privacy mechanism according to the embodiment of the present invention;
FIG. 4 shows coefficient pairs (a) of distributed algorithm 1 according to an embodiment of the present invention, which employs a differential privacy mechanism2，b2) The distribution of (c);
FIG. 5 shows the convergence accuracy and privacy budget with respect to the noise signal coefficient q when the distributed algorithm 1 based on the differential privacy mechanism is adopted in the embodiment of the present inventioniThe change curve of (2).
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
As shown in fig. 1, the present invention provides a system using a distributed energy storage economic dispatching method based on differential privacy, which includes 3 energy storage devices and 3 power consumers, and their electrical connections and communication links. The energy storage battery distributed economic dispatching method based on the differential privacy mechanism comprises the following steps:
firstly, an energy storage economic dispatching model of the system is established. The goal of economic dispatch is to maximize the load benefit while minimizing the overall cost of energy storage, i.e., maximizing social benefits. Can be converted into the following optimization problem:
Figure BDA0003515698730000071
Figure BDA0003515698730000072
SoCmin≤SoCi≤SoCmax，PESSmin，i≤PESS，i≤PESSmax，i，
PLmin，j≤PL，j≤PLmax，j
wherein P isESS，iIs the power generation (charge) amount of the ith energy storage unit, PL，jIs the power demand of the jth user, CiAnd UjRespectively representing the cost function of the energy storage unit i and the utility function, p, of the user jlossIs the active power loss of the system, PESSmin，iAnd PESSmax，iRespectively an upper limit and a lower limit of the generating (charging) capacity of the energy storage unit i, PLmin，jAnd PLmax，jRespectively, the upper and lower limits of the demand of load j.
The energy storage and power generation cost function generally adopts a quadratic model:
Figure BDA0003515698730000081
in the formula ai、bi、ciAre cost function coefficients.
The benefit function of the user can be approximated as follows:
Figure BDA0003515698730000082
in the formula aj、bjIs the coefficient of benefit。
Active loss p of systemlossCan be written as:
Figure BDA0003515698730000083
in the formula KiAnd KjThe power loss coefficients of the energy storage unit i and the user j are respectively.
From the above equations, the lagrangian function of the optimization problem can be listed:
Figure BDA0003515698730000084
wherein, for the energy storage unit, ξi＝1+Ki(ii) a To the user ξi＝1-Kj(ii) a λ is the lagrange multiplier. Order to
Figure BDA0003515698730000085
The obtained optimal solution lambda*Satisfies the following conditions:
Figure BDA0003515698730000086
for convenience of description, λ will be described hereinafteriξiAbbreviated as λi. And at this point, the energy storage economic dispatching model is established.
And solving the energy storage economic dispatching model by adopting a distributed method based on a consistency algorithm. At λiWhether the consistent iteration formula contains a cost/benefit function or not is distinguished, and the consistent iteration formula can be divided into a distributed algorithm 1 and a distributed algorithm 2. The former contains coefficient iterations and the latter does not.
The implementation of the consistency algorithm depends on the communication network, and the adjacency matrix W ═ W of the communication network needs to be designed firstij}∈Rn×n：
Figure BDA0003515698730000091
The iterative formula for distributed algorithm 1 is as follows:
Figure BDA0003515698730000092
Figure BDA0003515698730000093
pi(t+1)＝φ(pvi(t+1))
si(t+1)＝pvi(t+1)-pi(t+1)
wherein mu1To control the gain, pvi(t) is an estimate of the optimum power command, pi(t) is the actual power command after clipping, si(t) is pvi(t) and piDifference between (t), clipping function
Figure BDA0003515698730000094
The iterative formula for distributed algorithm 2 is as follows:
Figure BDA0003515698730000095
Figure BDA0003515698730000096
pi(t+1)＝φ(pvi(t+1))
Figure BDA0003515698730000101
wherein m isiCharacterizing the power balance of the system for the difference between the power supply and demand obtained at unit i, μ2To control the gain.
In distributed algorithms 1 and 2, a differential privacy algorithm is respectively introduced, namely, the communication information is subjected to noise masking. The iteration rule of the adjusted distributed algorithm 1 is as follows:
Figure BDA0003515698730000102
xj(t)＝λi(t)+ηj(t)
wherein etai(t) is an attenuated noise signal satisfying a Rayleigh distribution
Figure BDA0003515698730000103
qi∈(0，1)，ciIs a scaling factor. The corresponding probability density function is as follows:
Figure BDA0003515698730000104
at the same time, for pvi，piAnd siThe update rule of (2) is not changed.
The iteration rule of the adjusted distributed algorithm 2 is as follows:
Figure BDA0003515698730000105
Figure BDA0003515698730000106
xj(t)＝λj(t)+ηj(t)
yj(t)＝mj(t)+ηj(t)
so far, a distributed algorithm iterative formula based on difference privacy for solving the energy storage economic dispatching model is obtained. Due to the attenuation noise signal introduced by the differential privacy mechanism, a random component is added in the original deterministic convergence, namely, a randomness deviation exists between the convergence final value and the optimal value.
The convergence value of distributed Algorithm 1 eventually falls at the optimal value λ'*For reference, the upper and lower deviations are within the range of r. The convergence accuracy range r is:
Figure BDA0003515698730000111
the convergence value of distributed Algorithm 2 eventually falls at the optimal value λ'*The upper and lower deviations are within the range of r as a reference. The convergence accuracy range r is:
Figure BDA0003515698730000112
in order to conveniently analyze the protection degree of the differential privacy algorithm on the privacy information, the following definitions are made: for delta-neighbor set x1And x2And observing the sequence O when the Alg algorithm is satisfied
P[Alg(x1(0))∈O]≤eεδP[Alg(x2(0))∈O]
Then, the algorithm Alg is said to satisfy epsilon-differential privacy, where epsilon is the privacy budget. The smaller the privacy budget, the higher the degree of privacy protection.
The privacy budget of any energy storage unit i in the algorithm 1 is as follows:
Figure BDA0003515698730000113
the privacy budget of any energy storage unit i in the algorithm 2 is as follows:
Figure BDA0003515698730000114
coefficient q of noise signaliIncreasing the upper and lower range r of the convergence value increases and the privacy budget decreases. This indicates that the degree of privacy protection increases and the convergence accuracy decreases. Namely, the algorithm realizes privacy protection by sacrificing convergence precision. From the convergence accuracy range r and the privacy budget
Figure BDA0003515698730000115
ζiBoth of which are related to the noise signal coefficient qiIt is related. In practical application, the appropriate noise coefficient can be selected according to the specific requirements on convergence accuracy and privacy protection degree.
In the above distributed algorithm based on the difference privacy for energy storage economic scheduling, the energy storage economic scheduling model is only used for illustration, and in other embodiments, conditions such as a cost function, a benefit function, inequality constraints, and the like in the energy storage economic scheduling model may be modified as needed.
The following is a description of a specific embodiment:
there are 3 energy storage power stations and 3 power consumers in a system, and the electrical connections and communication links between them are shown in fig. 2. The values of the various parameters of the system are shown in table 1:
TABLE 1
Figure BDA0003515698730000121
The energy storage economic dispatching problem of the system is solved by adopting a distributed algorithm 1 based on differential privacy. Setting the control gain mu to 6, noise signal ci＝0.2，qiThe result of the iterative solution is shown in fig. 3, which is 0.3. Obviously, the marginal cost/benefit of all units converges to be consistent after 20 iterations, and simultaneously all power instructions also converge to the optimal value under the current situation, and the distributed algorithm with the introduced differential privacy mechanism can still ensure convergence. This demonstrates the effectiveness of the present invention in solving the energy storage economic dispatch model.
Due to the influence of noise, the system is in a suboptimal state at the moment, the convergence value is in certain probability distribution near the optimal value, and meanwhile, the protection degree of the system on the privacy information is also increased. To verify the protection of the system against private information, q is seti＝[0.3，0.4，0.5，0.6]At each different qiAnd repeating the simulation for 100 times, and observing the privacy information acquired by the attacker. Note that the privacy information includes cost function and effectGain function coefficient, generated energy, power consumption, sensitivity between power consumption and marginal cost and the like. According to the analysis process of privacy disclosure, an attacker firstly obtains a cost function and a benefit function coefficient a according to data of communication interactioni、biAnd further acquiring other privacy information on the basis thereof, so that the coefficient pair (a) is selected here2，b2) The results are shown in FIG. 4 as observed variables. The pairs of coefficients obtained by the attacker are randomly distributed, which means that the probability of deducing the true data is reduced. Meanwhile, the probability that an attacker observes the fitted distribution in a large amount to obtain the coefficients can be further reduced by actually changing the distribution condition of the noise. FIG. 4 also shows that qiThe larger the distribution, the more decentralized the distribution and the higher the degree of privacy. This demonstrates the effectiveness of the present invention for privacy information protection.
Increasing qiA higher degree of privacy protection can be achieved but also a certain convergence accuracy is sacrificed. FIG. 5 depicts the normalized system convergence accuracy, which obtains the true distribution from the actual values of 1000 simulations, and the privacy budget vs. when p is 0.1
Figure BDA0003515698730000131
Obtained according to the expression mentioned. It can be seen that with qiThe convergence radius is increased, and the privacy budget is reduced, which shows that the convergence precision is reduced and the privacy protection degree is increased, and proves the effectiveness of the method for analyzing the convergence precision and the privacy protection degree.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (10)

1. A distributed economic dispatching method of an energy storage battery based on a differential privacy mechanism is characterized by comprising the following steps:
step 1, establishing an energy storage battery economic dispatching model;
step 2, designing a distributed economic dispatching method according to a consistency algorithm on the basis of the established economic dispatching model;
and 3, introducing a differential privacy mechanism into the proposed distributed economic dispatching method, and iteratively solving the energy storage battery economic dispatching model according to an iterative formula after the differential privacy mechanism is introduced.
2. The distributed economic dispatching method of the energy storage battery based on the differential privacy as claimed in claim 1, wherein the step 2 is to design the distributed economic dispatching method according to a consistency algorithm, and the method comprises the following specific contents: selecting marginal cost lambda of each energy storage battery outputiAnd selecting the power of each energy storage battery as a consistency variable, designing an adjacent matrix W of the communication network topology, designing an iteration rule of a distributed algorithm according to a consistency algorithm, and finally using the iteration rule to iteratively solve an energy storage economic dispatching model.
3. The differential privacy-based energy storage battery distributed economic scheduling method according to claim 2, wherein the adjacency matrix W ═ Wij}∈Rn×nThe connection number of each node in a communication link of the energy storage battery economic dispatching model is obtained, and the expression is as follows:
Figure FDA0003515698720000011
wherein N isiSet of neighboring nodes, n, for node ii、njThe number of connections for nodes i, j, respectively.
4. The energy storage battery distributed economic scheduling method based on differential privacy as claimed in claim 3, wherein the distributed algorithm 1 with coefficient iteration designed according to the consistency algorithm has an iterative formula as follows:
Figure FDA0003515698720000021
Figure FDA0003515698720000022
pi(t+1)＝φ(pvi(t+1))
si(t+1)＝pvi(t+1)-pi(t+1)
wherein λi(t)、λj(t) marginal cost, w, of the energy storage cells i, j at time t, respectivelyijFor the elements of the adjacency matrix W corresponding to the communication link nodes i, j, μ1To control the gain, aiAnd biIs the benefit coefficient, p, in the energy storage economic dispatch modelvi(t) is an estimate of the optimum power command, pi(t) is the actual power command after clipping, si(t) is pvi(t) and piDifference between (t), sj(t) is pvj(t) and pj(t) wherein the clipping function is:
Figure FDA0003515698720000023
wherein
Figure FDA0003515698720000024
The maximum value and the minimum value of the stored energy charging and discharging power are respectively.
5. The distributed energy storage battery economic dispatching method based on the differential privacy as claimed in claim 4, characterized in that the distributed algorithm 2 without coefficient iteration designed according to the consistency algorithm has the following iterative formula:
Figure FDA0003515698720000025
Figure FDA0003515698720000026
pi(t+1)＝φ(pvi(t+1))
Figure FDA0003515698720000027
wherein λ isi(t)、λj(t) marginal cost of energy storage cells i, j at time t, respectively, miThe difference between the power supply and demand obtained for the energy storage cell i, characterizes the power balance of the system, wijFor the elements of the adjacency matrix W corresponding to the communication link nodes i, j, μ2To control the gain, aiAnd biIs the benefit coefficient, p, in the energy storage economic dispatch modelvi(t) is an estimate of the optimum power command, pi(t) is the actual power command after clipping, si(t) is pvi(t) and pi(t) difference between sj(t) is pvj(t) and pj(t) difference between (t).
6. The energy storage battery distributed economic scheduling method based on differential privacy as claimed in claim 4, wherein the step 3 introduces a differential privacy mechanism to realize privacy information protection, specifically, the iteration rule of the distributed algorithm is adjusted, and the adjusted distributed algorithm 1 has an iteration formula as follows:
Figure FDA0003515698720000031
xj(t)＝λi(t)+ηj(t)
wherein etai(t) is the attenuated noise signal satisfying the Laplace distribution
Figure FDA0003515698720000032
qi∈(0，1)，ciFor the scaling factor, the corresponding probability density function is as follows:
Figure FDA0003515698720000033
at the same time, for pvi，piAnd siThe update rule of (2) is not changed.
7. The energy storage battery distributed economic scheduling method based on differential privacy as claimed in claim 5, wherein step 3 introduces a differential privacy mechanism to achieve privacy information protection, specifically, the iteration rule of the distributed algorithm is adjusted, and the adjusted distributed algorithm 2 has an iteration formula as follows:
Figure FDA0003515698720000034
Figure FDA0003515698720000035
xj(t)＝λj(t)+ηj(t)
yj(t)＝mj(t)+ηj(t)。
8. the distributed economic scheduling method of energy storage batteries based on differential privacy as claimed in claim 6, characterized in that the convergence value of distributed algorithm 1 finally falls at the optimal value λ'*For reference, the upper and lower deviations are within the range of r:
P{|λ′(∞)-λ′*|＜r}≥1-V(λ′∞)/r2
let r be the convergence accuracy range, where r is
Figure FDA0003515698720000041
V(λ′∞) Is lambda'∞The variance of (a) is determined,
Figure FDA0003515698720000042
9. the distributed economic scheduling method of energy storage batteries based on differential privacy as claimed in claim 7, characterized in that the convergence value of the distributed algorithm 2 finally falls at the optimal value λ'*For reference, the upper and lower deviations are within the range of r:
P{|λ′(∞)-λ′*|＜r}≥1-V(λ′∞)/r2
let r be the convergence accuracy range, where r is
Figure FDA0003515698720000043
V(λ′∞) Is lambda'∞The variance of (a) is determined,
Figure FDA0003515698720000044
10. the differential privacy-based energy storage battery distributed economic scheduling method according to claim 8 or 9, characterized in that for delta-neighbor set x1And x2And observing the sequence O when the Alg algorithm is satisfied
P[Alg(x1(0))∈O]≤eεδP[Alg(x2(0))∈O]
Then, the Alg algorithm is called to meet epsilon-difference privacy, wherein epsilon is the privacy budget;
the privacy budget of any energy storage unit i in the distributed algorithm 1 is as follows:
Figure FDA0003515698720000051
the privacy budget of any energy storage unit i in the distributed algorithm 2 is as follows:
Figure FDA0003515698720000052
coefficient q of noise signaliAnd increasing the upper and lower ranges r of the convergence value, and reducing the privacy budget, which indicates that the privacy protection degree is increased, and the convergence precision is reduced, namely the algorithm realizes privacy protection by sacrificing the convergence precision.
CN202210163557.1A 2022-02-22 2022-02-22 Energy storage battery distributed economic dispatching method based on differential privacy mechanism Active CN114529207B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210163557.1A CN114529207B (en) 2022-02-22 2022-02-22 Energy storage battery distributed economic dispatching method based on differential privacy mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210163557.1A CN114529207B (en) 2022-02-22 2022-02-22 Energy storage battery distributed economic dispatching method based on differential privacy mechanism

Publications (2)

Publication Number Publication Date
CN114529207A true CN114529207A (en) 2022-05-24
CN114529207B CN114529207B (en) 2024-05-28

Family

ID=81624374

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210163557.1A Active CN114529207B (en) 2022-02-22 2022-02-22 Energy storage battery distributed economic dispatching method based on differential privacy mechanism

Country Status (1)

Country Link
CN (1) CN114529207B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116090014A (en) * 2023-04-07 2023-05-09 中国科学院数学与系统科学研究院 Differential privacy distributed random optimization method and system for smart grid

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150371059A1 (en) * 2014-06-18 2015-12-24 Palo Alto Research Center Incorporated Privacy-sensitive ranking of user data
CN112231749A (en) * 2020-10-14 2021-01-15 西安交通大学 Distributed single-dimensional time sequence data real-time privacy protection publishing method with consistency
CN112598211A (en) * 2020-10-30 2021-04-02 天津大学 Consistency-based distributed power grid economic dispatching injection attack mitigation method
CN112713612A (en) * 2020-12-29 2021-04-27 苏州科技大学 Multi-target scheduling privacy protection method for microgrid leader-following rapid consistency
US20210224708A1 (en) * 2018-01-02 2021-07-22 Shanghai Jiao Tong University Real-time economic dispatch method of power system
CN113839845A (en) * 2021-09-18 2021-12-24 南方电网科学研究院有限责任公司 Secret distributed optimal scheduling method, system, computer equipment and medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150371059A1 (en) * 2014-06-18 2015-12-24 Palo Alto Research Center Incorporated Privacy-sensitive ranking of user data
US20210224708A1 (en) * 2018-01-02 2021-07-22 Shanghai Jiao Tong University Real-time economic dispatch method of power system
CN112231749A (en) * 2020-10-14 2021-01-15 西安交通大学 Distributed single-dimensional time sequence data real-time privacy protection publishing method with consistency
CN112598211A (en) * 2020-10-30 2021-04-02 天津大学 Consistency-based distributed power grid economic dispatching injection attack mitigation method
CN112713612A (en) * 2020-12-29 2021-04-27 苏州科技大学 Multi-target scheduling privacy protection method for microgrid leader-following rapid consistency
CN113839845A (en) * 2021-09-18 2021-12-24 南方电网科学研究院有限责任公司 Secret distributed optimal scheduling method, system, computer equipment and medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘志坚;王旭辉;郑超铭;: "计及可再生能源发电成本的智能电网分布式经济调度", 电子测量技术, no. 02, 23 January 2020 (2020-01-23) *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116090014A (en) * 2023-04-07 2023-05-09 中国科学院数学与系统科学研究院 Differential privacy distributed random optimization method and system for smart grid
CN116090014B (en) * 2023-04-07 2023-10-10 中国科学院数学与系统科学研究院 Differential privacy distributed random optimization method and system for smart grid

Also Published As

Publication number Publication date
CN114529207B (en) 2024-05-28

Similar Documents

Publication Publication Date Title
Liang et al. Dynamic economic/emission dispatch including PEVs for peak shaving and valley filling
Zhang et al. Distributed energy management for multiuser mobile-edge computing systems with energy harvesting devices and QoS constraints
Zhao et al. Consensus-based energy management in smart grid with transmission losses and directed communication
Liu et al. Distributed optimal economic environmental dispatch for microgrids over time-varying directed communication graph
Lian et al. Game-theoretic multi-agent control and network cost allocation under communication constraints
CN107706921B (en) Micro-grid voltage regulation method and device based on Nash game
Li et al. Virtual-action-based coordinated reinforcement learning for distributed economic dispatch
Fathy et al. Recent coyote algorithm-based energy management strategy for enhancing fuel economy of hybrid FC/Battery/SC system
Wang et al. Distributed incremental cost consensus-based optimization algorithms for economic dispatch in a microgrid
Fang et al. Distributed deep reinforcement learning for renewable energy accommodation assessment with communication uncertainty in Internet of Energy
CN115000994A (en) Multi-energy storage unit grouping consistency power distribution method
CN115719113A (en) Intelligent power grid economic dispatching distributed accelerated optimization method based on directed imbalance topology
CN114529207A (en) Energy storage battery distributed economic dispatching method based on differential privacy mechanism
Chen et al. Multi-Objective Optimization Scheduling of Active Distribution Network Considering Large-Scale Electric Vehicles Based on NSGAII-NDAX Algorithm.
CN112713612B (en) Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method
CN109934476A (en) A kind of more tactful evolutionary Game Analysis methods of the micro-capacitance sensor source based on main body bounded rationality decision-storage joint planning
Li et al. Distributed tracking-ADMM approach for chance-constrained energy management with stochastic wind power in smart grid
CN112003279A (en) Method for evaluating new energy consumption capability of hierarchical micro-grid
CN116663798A (en) Energy system optimal scheduling method based on harmony search algorithm
Wu et al. Distributed hierarchical consensus algorithm for economic dispatch in smart grid
Chen et al. Battery energy storage system based on incremental cost consensus algorithm for the frequency control
Reddy et al. Optimal Power flow using particle swarm optimization
He et al. Distributed Optimal Power Scheduling for Microgrid System via Deep Reinforcement Learning with Privacy Preserving
Chiang et al. New approach with a genetic algorithm framework to multi‐objective generation dispatch problems
CN110729759A (en) Method and device for determining distributed power supply configuration scheme in micro-grid

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