CN112713612B - Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method - Google Patents

Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method Download PDF

Info

Publication number
CN112713612B
CN112713612B CN202011597825.8A CN202011597825A CN112713612B CN 112713612 B CN112713612 B CN 112713612B CN 202011597825 A CN202011597825 A CN 202011597825A CN 112713612 B CN112713612 B CN 112713612B
Authority
CN
China
Prior art keywords
node
consistency
noise
power
privacy
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.)
Active
Application number
CN202011597825.8A
Other languages
Chinese (zh)
Other versions
CN112713612A (en
Inventor
陈珍萍
邵雪莲
吴征天
付保川
许馨尹
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.)
Suzhou University of Science and Technology
Original Assignee
Suzhou University of Science and 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 Suzhou University of Science and Technology filed Critical Suzhou University of Science and Technology
Priority to CN202011597825.8A priority Critical patent/CN112713612B/en
Publication of CN112713612A publication Critical patent/CN112713612A/en
Application granted granted Critical
Publication of CN112713612B publication Critical patent/CN112713612B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • 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
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • 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
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/28—Arrangements for balancing of the load in a network by storage of energy
    • H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • 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

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

Abstract

The invention discloses a micro-grid leader-follower rapid consistency multi-target scheduling privacy protection method, which solves the problems that privacy leakage possibly occurs in existing multi-target scheduling and privacy protection on leader-follower consistency is relatively lacking. The invention is based on a fast consistency algorithm, firstly, a multi-target scheduling model is established, and an optimal solution is obtained. Then, the situation that power information is possibly leaked in the consistency communication process, particularly the situation containing the leader node, is analyzed. And thirdly, calculating a global power difference for the leader node by adopting random numbers added with noise to the consistency variable and added with zero to the node as power values to protect the privacy information. And utilizeData-privacy analyses the degree of protection of the initial privacy of the proposed policy.

Description

Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method
Technical Field
The invention relates to the field of micro-grid information privacy protection methods, in particular to a micro-grid leading-following rapid consistency multi-target scheduling privacy protection method.
Background
Economic dispatch based on consistency algorithms in micro-grids has received much research attention. In general, these algorithms are iterative and incremental costs are used as consistency variables. These studies consider scheduling problems on the power generation side. In the micro-grid, the distributed energy management problem is enabled to be more practical and significant by considering the power generation side and the user demand side.
Considering energy scheduling on both sides, rahbari et al (literature: n.rahbafi-Asr, u.ojha, Z.Zhang, M.Chow.Incremental welfare consensus algorithm for cooperative distributed genertion/demand response in smart grid [ J ]. IEEE Transactions on Smart Grid,2014.5 (6): 2836-2845.) propose a distributed energy management algorithm based on consistency, but these algorithms are only adapted to undirected communication networks. To overcome these limitations, scholars have created a non-convex social benefit maximization problem and converted the non-convex optimization problem into a convex optimization problem, proposing a consistency-based distributed energy management algorithm. In addition, the multi-objective scheduling problem, which is a compromise between the power generation cost, the power consumption cost, the environmental protection and the like, has gradually formed a research hotspot, and related researches have been conducted by the students aiming at the problem. However, most existing studies ignore the privacy disclosure problem of the participants.
The micro-grid is generally provided with an intelligent ammeter, an intelligent electric appliance, a distributed energy source and the like, so that distributed control and energy management are realized. During the distributed consistency algorithm negotiation process, each agent broadcasts its true state to neighboring agents. However, this method of information transmission provides the possibility for a malicious attacker to listen to the detection to infer the privacy information of the participants. Disclosure of electricity consumption data may lead to disclosure of privacy of a user's life, because an attacker may use fine-grained power consumption data of a smart meter to reveal private information about a consumer's daily work or in-house equipment. In this regard, a relevant study has been conducted by a learner.
Mandal (literature: A.Mandal. "Privacy Preserving Consensus-Based Economic Dispatch in Smart Grid Systems", in International Conference on Future Network Systems and Security, springer International Publishing,98-110,2016.) discloses disclosure of genset privacy information, i.e., final power generation and cost function parameters of power generation in a consistency-based economic dispatch process, and a privacy protection strategy is proposed on the basis of this, but is only applicable to undirected communication topology.
In addition, great efforts have been made to respond to privacy protection requirements by encrypting or aggregating communication designs (literature: H.Li, X.Lin, H.Yang, X.Liang, R.Lu, and X.shen, "EPPDR: an efficient privacy-preserving demand response scheme with adaptive key evolution in smart grid", IEEE Transactions on Parallel and Distributed Systems, vol.25, no.8,2053-2064,2014.). However, in consistency-based distributed energy management, these strategies can lead to significant computational and communication burden due to progressive convergence.
Much research attention has also been paid to the average consistency based on differential privacy, but its final solution converges to a random value instead of an average value that converges exactly to an initial value. Mo (document Y.Mo, R.M.Murray.Privacy preserving average consensus [ J ]. IEEE Transactions on Automatic control.2017,62 (2): 753-765.) proposes an average consistency algorithm with privacy protection, adding zero and exponentially decaying noise per node to a local state. It is worth noting that existing privacy protection based on consistency mostly only focuses on initial value privacy information of consistency variables, but power scheduling based on consistency is not only required to protect initial value privacy, but also required to protect power consumption privacy information and convergence final value power privacy information in an iterative communication process. And there is a lack of research on how the leader node obtains the global power difference in the leader-follower consistency algorithm without the node power being compromised.
Disclosure of Invention
The invention aims to provide a micro-grid leader-follower rapid consistency multi-target scheduling privacy protection method, which aims to solve the problems that privacy leakage possibly occurs in the existing multi-target scheduling and privacy protection for leader-follower consistency is relatively lacking.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
the micro-grid leading-following rapid consistency multi-target scheduling privacy protection method is characterized by comprising the following steps of: the method comprises the following steps:
(1) Establishing a fast consistency algorithm with feedback gain:
the communication relation between the generator and the flexible load node in the micro-grid is represented by using a directional strong communication graph G, and an adjacency matrix A= (a) pq ) n×n For describing a communication relationship between nodes p and q, where n represents the number of nodes, a when node p is able to obtain state information of node q pq =1; when node p is unavailableTaking the information of node q or when p=q, a pq ＝0；
Defining a neighbor set of node p as N p The method comprises the steps of carrying out a first treatment on the surface of the The degree matrix of the directed strong communication graph G is d=diag (D pp ) Wherein
The fast consistency model for the economic dispatch of the micro-grid is established as follows:
wherein lambda is p (t) is the state of the p-th agent node, t is the iterative sequence, m pq The (p, q) th element of the row random matrix;
in order to increase the convergence speed, a fast consistency model with feedback gain is built based on the formula (1) as follows:
wherein: k (k) p Feedback gain, k, for agent node p p ＞0，p＝1,2,…,n，n p For the number of neighbor agents of agent node p, i.e. the degree of ingress of agent node p, there is n p ＝d pp ；
(2) Establishing a multi-objective model of comprehensive environment and economy:
taking economic and environmental factors into consideration, taking the minimum running cost and pollutant emission of the micro-grid as comprehensive overall targets, firstly establishing an environmental scheduling model as follows:
wherein alpha is ～ ,β ～ ,γ ～ Is the emission coefficient of the generator set; p (P) G The output power of the generator set; e (P) G ) The pollution gas emission amount of the generator set;
and establishing an economic dispatch objective equation containing flexible load as follows:
min∑F B (P B )+∑F G (P G )-∑F D (P D )，
wherein:
wherein F is B 、F G 、F D The definition of the energy storage battery is the cost of the energy storage battery, the active power generation cost of the traditional fuel generator and the electricity utilization benefit of the flexible load, and alpha, beta, gamma and zeta are the cost parameters of the power generation and energy storage units respectively; a, b and c are parameters of the electricity utilization benefit of the flexible load; p (P) D And P B Power for flexible load and stored energy;
in order to simultaneously process two targets of environment and economy, the multi-target problem is converted into a single-target problem, the power balance is considered, and a multi-target model is established as follows:
wherein Δp= Σp D -∑P G -P REG +P d +P B Representing the active power difference of the system, P REG Representing renewable energy unit power; p (P) d Representing the total power of the pure load of the system; omega 1 、ω 2 Is a linear weighting factor;
(3) Converting the multi-objective model obtained in the step (2) to obtain a leader-follower rapid consistency scheduling model:
converting the formula (5) into the following unconstrained problem by using a Lagrangian multiplier method, and obtaining a scheduling model as follows:
wherein λ represents the lagrange multiplier;
for variable P in equation (8) D 、P G 、P B And respectively solving the partial derivatives with lambda to obtain the optimal solution of the scheduling model as follows:
(4) Noise-adding the consistency variable in the leader-follower fast consistency scheduling model:
to improve the privacy protection of the initial value, a continuous encryption function F is introduced for noise added by the initial iteration pq (z pq ) The node q is the outbound neighbor node of the node p in the strong communication graph G;
setting each encryption function F pq (z pq ) Can only be known by the neighboring nodes p and q, while each node randomly generates z for all (p, q) pq As a parameter of the encryption function, letRepresenting the noise fused by the encryption function, and obtaining:
wherein,only once calculated in the first iteration, i.e. t=1;
ν p (0) Representing added noise at t=0；
ν p (t) represents all iterations, i.e. t > =0, v comprising t=0 p (0) Added noise, and v p (t) obeys the same independent distribution in each iteration, but each value is random;
the noise function of the consistency variable may be expressed by:
wherein μ is a constant and 0 < μ < 1, t is the number of iterations, μ t-1 Is the power series of mu t-1 times, theta p (t) is noise added by node p at the t-th iteration, whose upper absolute value is decaying exponentially with t;
when the iteration is carried out for the t time, the node p adds the noise consistency variableBroadcasting to neighbor nodes, and updating the consistency state of the node p by adopting a leader-follower fast consistency algorithm according to the formula (2) in the step (1) and considering the power balance of the system, wherein the consistency state is as follows:
in the method, in the process of the invention,is a power balance adjustment coefficient; />Representing the consistency state of the noise added by the node p; />Adding noise consistency variables to the neighbor node q of the node p;
according to formulas (3), (4), (5), (6) and (9) of consistency variables and power, and considering upper and lower limit constraints of active power of a generator, a flexible load and an energy storage battery in an actual micro-grid, establishing a node real power updating rule as follows:
wherein P is G，max/min Upper and lower limits of the output of the generator, P D,max/min Upper and lower limits of force, P, for flexible load B,max/min The upper limit and the lower limit of the power of the energy storage battery are set;
(5) The privacy of the leader-follower fast consistency scheduling model is improved by adding random values to the nodes, and the process is as follows:
(5.1) initializing:
calculation of P from node parameters p (0) And lambda (lambda) p (0) The leading node also needs to calculate global delta P (t) according to the random number power added by each node and the sum of which is zero;
parameter z of node broadcast encryption function pq And calculates an encryption function F pq (z pq ) Node is 0 from mean and sigma from variance 2 Is uniformly distributed in (a)Is generated by medium random v p (0) Wherein σ > 0, and θ p (0)＝ν p (0) Then calculate according to formula (10) to get +.>At the same time node calculates +.>And will noise the consistency variable +.>Broadcasting to neighbor nodes;
(5.2) iterating:
if |ΔP (t) | is not less than 0.001 orEach node updates its own ++according to formulas (12), (13), (14), (15)>And P p (t) the leader node also needs to calculate global Δp (t) from the random number powers added by each node to sum to zero;
v of each node p (t) fromRandom generation in the computer and updating the noise +.>The node then adds this noise to the consistency variable lambda p In (t), the consistency variable after noise addition is +.>Broadcasting to neighbor nodes;
(5.3) outputP p (t), ΔP (t), the addition of the random number is completed.
The micro-grid leading-following rapid consistency multi-target scheduling privacy protection method is characterized by comprising the following steps of: in the step (2), the energy storage battery cost, the active power generation cost of the traditional fuel generator and the electricity utilization benefit of the flexible load related to the economic dispatch are combined to obtain an economic dispatch objective equation of the flexible load.
Said micro-grid lead-follow fast consistencyThe multi-target scheduling privacy protection method is characterized by comprising the following steps of: in the step (2), a linear weighting method is adopted to convert the multi-objective problem into a single-objective problem, so that omega 1 +ω 2 =1 and 0 < ω 1 ,ω 2 ＜1。
The micro-grid leading-following rapid consistency multi-target scheduling privacy protection method is characterized by comprising the following steps of: in step (4), θ for all nodes in all iterations p The sum of (t) converges to zero and all θ p The sum of absolute values of (t) is bounded, i.eH is a constant, thereby ensuring that the system consistency variation converges and the optimality of the final value is solved.
The invention considers the electricity utilization benefit of the electricity utilization unit, the electricity utilization cost of the electricity generation unit and the environmental protection cost, and designs a privacy protection multi-target scheduling algorithm based on a leader-follower consistency algorithm considering the directed communication of both sides of electricity generation and demand. Solving the optimal solution of the established multi-objective model and simultaneously analyzing the adopted leader-follow rapid consistency algorithm, wherein privacy leakage possibility exists; adding a noise function to the consistency variable to protect privacy in the iterative process; the mode that the node power is added with a random value and then communicated ensures that the transmitted power is false information and has no influence on the global power difference acquired by the leading node; the above privacy preserving method ensures that the final result of the algorithm is the optimal value of the system.
Aiming at the problem of multi-objective scheduling privacy protection of the independent micro-grid of the leader-follower consistency algorithm, the invention provides a scheduling strategy for protecting user privacy and improving convergence speed, and the invention protects the user privacy information while rapidly realizing economic and environmental comprehensive benefits. Firstly, establishing a micro-grid multi-objective scheduling model which comprises a multi-objective optimization function and constraint conditions, and solving an optimal solution of the problem; then, zero and decaying exponential noise are added to the consistency variable, a new method of adding random values is adopted for calculating the global power difference, and the added noise can achieve convergence and optimality of the system.
The noise-added leader-follower rapid consistency algorithm can realize multi-target scheduling of the micro-grid; the convergence speed is faster than that of the traditional consistency algorithm under the same parameters; the problem of consistency privacy disclosure of the leader-containing nodes is solved; the requirement of system privacy protection is met.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The invention will be further described with reference to the drawings and examples.
As shown in fig. 1, the micro-grid leader-follower fast consistent multi-objective scheduling privacy protection method comprises the following steps:
(1) Establishing a fast consistency algorithm with feedback gain:
the communication relation between the generator and the flexible load node in the micro-grid is represented by using a directional strong communication graph G, and an adjacency matrix A= (a) pq ) n×n For describing a communication relationship between nodes p and q, where n represents the number of nodes, a when node p is able to obtain state information of node q pq =1; a when node p cannot acquire information of node q or when p=q pq ＝0；
Defining a neighbor set of node p as N p The method comprises the steps of carrying out a first treatment on the surface of the The degree matrix of the directed strong communication graph G is d=diag (D pp ) Wherein
The fast consistency model for the economic dispatch of the micro-grid is established as follows:
wherein lambda is p (t) is the state of the p-th agent node, t is the iterative sequence, m pq The (p, q) th element of the row random matrix;
in order to increase the convergence speed, a fast consistency model with feedback gain is built based on the formula (1) as follows:
wherein: k (k) p Feedback gain, k, for agent node p p ＞0，p＝1,2,…,n，n p For the number of neighbor agents of agent node p, i.e. the degree of ingress of agent node p, there is n p ＝d pp ；
(2) Establishing a multi-objective model of comprehensive environment and economy:
taking economic and environmental factors into consideration, taking the minimum running cost and pollutant emission of the micro-grid as comprehensive overall targets, firstly establishing an environmental scheduling model as follows:
wherein alpha is ～ ,β ～ ,γ ～ Is the emission coefficient of the generator set; p (P) G The output power of the generator set; e (P) G ) The pollution gas emission amount of the generator set;
and establishing an economic dispatch objective equation containing flexible load as follows:
min∑F B (P B )+∑F G (P G )-∑F D (P D )，
wherein:
wherein F is B 、F G 、F D Definition of (2)The cost of the energy storage battery, the active power generation cost of the traditional fuel generator and the electricity utilization benefit of the flexible load are respectively the cost parameters of the power generation and energy storage unit; a, b and c are parameters of the electricity utilization benefit of the flexible load; p (P) D And P B Power for flexible load and stored energy;
in order to simultaneously process two targets of environment and economy, the multi-target problem is converted into a single-target problem, the power balance is considered, and a multi-target model is established as follows:
wherein Δp= Σp D -∑P G -P REG +P d +P B Representing the active power difference of the system, P REG Representing renewable energy unit power; p (P) d Representing the total power of the pure load of the system; omega 1 、ω 2 Is a linear weighting factor;
(3) Converting the multi-objective model obtained in the step (2) to obtain a leader-follower rapid consistency scheduling model:
converting the formula (5) into the following unconstrained problem by using a Lagrangian multiplier method, and obtaining a scheduling model as follows:
wherein λ represents the lagrange multiplier;
for variable P in equation (8) D 、P G 、P B And respectively solving the partial derivatives with lambda to obtain the optimal solution of the scheduling model as follows:
the privacy analysis in steps (1) and (2) is as follows:
the invention mainly focuses on the power privacy information of the power generation/consumption of the controllable unit in the micro-grid. For example, if the listening node steals the power data of the user load for a long time, the listening node can predict future power information and behavior habits of the user. Revealing private information today in big data will put the system in a dangerous state with information exposed.
As can be seen by the fast consistency algorithm, each node broadcasts a consistency variable to its neighbor nodes without broadcasting its private information power information. According to this algorithm, private information is not directly broadcast, but broadcast information is a function of private information in the iterative rule (see formula (9)), so private information can still be inferred from broadcast information. It is worth noting that, due to the adoption of the leader-follower consistency algorithm, the leader node needs to collect and calculate global power information, that is, each node needs to transmit the power value to the leader node, so that the possibility of privacy information leakage is greatly increased. Therefore, ensuring privacy protection while ensuring convergence and optimal solutions is of interest.
Existing average consistency algorithms with privacy protection (document: J.He, L.Cai, C.Zhao, R Cheng. Privacy-preserving average consensus: privacy analysis and optimal algorithm design [ J ]. ArXiv preprint arXiv:1609.06368, 2016.) make algorithms with privacy protection possible, since the algorithms are based on consistency. Unlike the average consistency privacy protection, which only protects the initial value of power, the present invention also requires protection of the values in the iterative process and the final convergence value.
(4) Noise-adding the consistency variable in the leader-follower fast consistency scheduling model:
under the inspired by the literature (J.He, L.Cai, C.Zhao, R Cheng. Privacy-preserving average consensus: privacy analysis and optimal algorithm design [ J ]. ArXiv preprint arXiv:1609.06368, 2016.), the invention provides a privacy protection algorithm containing a leader node and based on encryption functions, which realizes the protection of user privacy and can ensure the high efficiency of rapid consistent multi-objective scheduling.
Since the broadcast information is hidden, the simplest way to protect privacy is to add noise to the broadcast information. In order to ensure the convergence of the algorithm after noise addition and the optimality of the convergence final value, the invention designs a noise adding process aiming at the consistency variable.
To improve the privacy protection of the initial value, a continuous encryption function F is introduced for noise added by the initial iteration pq (z pq ) The node q is the outbound neighbor node of the node p in the strong communication graph G;
setting each encryption function F pq (z pq ) Can only be known by the neighboring nodes p and q, while each node randomly generates z for all (p, q) pq As a parameter of the encryption function, letRepresenting the noise fused by the encryption function, and obtaining:
wherein,only once calculated in the first iteration, i.e. t=1;
v p (0) Represents added noise at t=0;
v p (t) represents all iterations, i.e. t > =0, v comprising t=0 p (0) Added noise, and v p (t) obeys the same independent distribution in each iteration, but each value is random;
the noise function of the consistency variable may be expressed by:
wherein μ is a constant and 0 < μ < 1, t is the number of iterations, μ t-1 Is the power series of mu t-1 times, theta p (t) is noise added by node p at the t-th iteration, whose upper absolute value is decaying exponentially with t;
in all iterations, θ of node p p The sum of (t) converges to zero and all θ p The sum of absolute values of (t) is bounded, i.eH is a constant, thereby ensuring that the system consistency variation converges and the optimality of the final value is solved.
When the iteration is carried out for the t time, the node p adds the noise consistency variableBroadcasting to neighbor nodes, and updating the consistency state of the node p by adopting a leader-follower fast consistency algorithm according to the formula (2) in the step (1) and considering the power balance of the system, wherein the consistency state is as follows:
in the method, in the process of the invention,is a power balance adjustment coefficient; />Representing the consistency state of the noise added by the node p; />Adding noise consistency variables to the neighbor node q of the node p;
according to formulas (3), (4), (5), (6) and (9) of consistency variables and power, and considering upper and lower limit constraints of active power of a generator, a flexible load and an energy storage battery in an actual micro-grid, establishing a node real power updating rule as follows:
wherein P is G，max/min Upper and lower limits of the output of the generator, P D,max/min Upper and lower limits of force, P, for flexible load B,max/min The upper limit and the lower limit of the power of the energy storage battery are set;
(5) Improving the privacy of the leader-follower fast consistency scheduling model by adding random values to the nodes:
because the leader node exists in the consistency scheduling, the leader node needs to collect and calculate global power information. Once the listening node invades the leader node, the power information of each participating consistency scheduling node in the system is at risk of being compromised. And the global power balance is adjusted by taking account that the leader node only performs addition calculation on the acquired power information, and the specific power value of each node is not required to be known. Therefore, each node adds a random value before locally calculating the real power information and sending the real power information to the leader node in each iteration, and the sum of the random values added by each node is zero, so that the calculated value added by the leader node is not influenced, the transmitted power information is false information, and privacy cannot be revealed. By adding the means that the sum of the random values is zero, node characteristic information in the original data is destroyed, power transmission/utilization data information of specific nodes can be hidden, and the privacy of a consistency scheduling algorithm containing leading nodes is improved.
The specific process is as follows
(5.1) initializing:
calculation of P from node parameters p (0) And lambda (lambda) p (0) The leading node also needs to calculate global delta P (t) according to the random number power added by each node and the sum of which is zero;
parameter z of node broadcast encryption function pq And is combined withCalculating encryption function F pq (z pq ) Node is 0 from mean and sigma from variance 2 Is uniformly distributed in (a)Random generation v p (0) Wherein σ > 0, and θ p (0)＝ν p (0) Then calculate according to formula (10) to get +.>At the same time node calculates +.>And will noise the consistency variable +.>Broadcasting to neighbor nodes;
(5.2) iterating:
if |ΔP (t) | is not less than 0.001 orEach node updates its own ++according to formulas (12), (13), (14), (15)>And P p (t) the leader node also needs to calculate global Δp (t) from the random number powers added by each node to sum to zero;
v of each node p (t) fromRandom generation in the computer and updating the noise +.>The node then adds this noise to the consistency variable lambda p In (t), the consistency variable after noise addition is +.>Broadcast to neighbor nodes；
(5.3) outputP p (t), ΔP (t), the addition of the random number is completed.
The privacy preserving effect of the present invention is analyzed as follows.
1. Broadcast value privacy preserving analysis:
the consistency algorithm can know that the node needs to broadcast power information to the leader node and consistency variables to the neighbor nodes. For broadcast power information, the power information broadcast by the node is the power added with the random value, not the real power, so that the power information can be protected.
For broadcast consistency variable information. Since the node performs all received information and numerical calculations locally. Therefore, even if the listening node steals the consistent variable value or the long-time change condition of the consistent variable value, even if the iteration rule of the consistent variable is deduced, the real power and the change rule thereof cannot be fitted through the function curve because the power information of the eavesdropping is random and irregular.
2. Initial privacy preserving analysis:
at the t iteration time, defining the noise adding state of the node pAnd an initial state lambda p (0) The noise between isNote node j is a listening node which is noise +.>The estimated value of (2) is +.>Node j is based on information broadcasted by node p>And estimated noise->Deducing initial state lambda of node p according to (16) p (0)。
Then node j estimates the error for the initial state of the input neighbor node pAnd noise estimation error->The method meets the following conditions:
wherein the method comprises the steps ofAnd->Respectively indicate->And->And ε.gtoreq.0 is the allowable state estimation error.
Thus, node p (ε, δ) p ) The degree of privacy protection can be defined as the initial value λ of all neighbor nodes according to the estimated error ε at all iteration instants t p (0) The maximum probability of estimation is specifically:
now consider that node j estimates the initial value lambda of node p p (0) This is the case. Assume that node p is one of the following nodes, and that the consistency variable iteration rule of node p is global information. Node pCan be written in the following form: />
At theta p (t) in the case of the design according to the formula (11), due to
Therefore, even if the neighbor node q of the node p is the input neighbor node of the node j, as long as F pq (z pq )-F qp (z qp ) The value is not acquired by the node j, and the increase of the iteration times t does not increase the initial value lambda of the node p p (0) Is a leakage probability of (1). Thus, the node p initial value leakage lambda can be inferred p (0) The probability of (2) satisfies:
according toDefinition of->And in the formula (11), there is θ p (0)＝ν p (0). Thus there is
Wherein the method comprises the steps ofIs v p (0) Is a value space of->Is v p (0) Probability density function of (a). V (v) p (0) In the algorithm herein, the obeying mean is zero and the variance is sigma 2 Is uniformly distributed, i.e.)>Thus, the (ε, δ) of the node p can be obtained p ) -privacy protection degree of:
that is, in the case where the estimation error ε is constant, the larger the noise variance σ of the node p, i.e., the greater the added noise intensity, the node p initial value λ p (0) The smaller the estimated probability, the higher the privacy level of the node p.
The embodiments of the present invention are merely described in terms of preferred embodiments of the present invention, and are not intended to limit the spirit and scope of the present invention, and various modifications and improvements made by those skilled in the art to the technical solutions of the present invention should fall within the protection scope of the present invention, and the technical content of the present invention as claimed is fully described in the claims.

Claims (4)

1. The micro-grid leading-following rapid consistency multi-target scheduling privacy protection method is characterized by comprising the following steps of: the method comprises the following steps:
(1) Establishing a fast consistency algorithm with feedback gain:
the communication relation between the generator and the flexible load node in the micro-grid is represented by using a directional strong communication graph G, and an adjacency matrix A= (a) pq ) n×n For describing a communication relationship between nodes p and q, where n represents the number of nodes, a when node p is able to obtain state information of node q pq =1; a when node p cannot acquire information of node q or when p=q pq ＝0；
Defining a neighbor set of node p as N p The method comprises the steps of carrying out a first treatment on the surface of the The degree matrix of the directed strong communication graph G is d=diag (D pp ) Wherein
The fast consistency model for the economic dispatch of the micro-grid is established as follows:
wherein lambda is p (t+1) and lambda q (t) states of the p and q th agent nodes, respectively, t+1 and t being an iterative sequence, m pq The (p, q) th element of the row random matrix;
in order to increase the convergence speed, a fast consistency model with feedback gain is built based on the formula (1) as follows:
wherein: k (k) p Feedback gain, k, for agent node p p ＞0，p＝1,2,…,n，n p For the number of neighbor agents of agent node p, i.e. the degree of ingress of agent node p, there is n p ＝d pp ；
(2) Establishing a multi-objective model of comprehensive environment and economy:
taking economic and environmental factors into consideration, taking the minimum running cost and pollutant emission of the micro-grid as comprehensive overall targets, firstly establishing an environmental scheduling model as follows:
wherein alpha is ～ ,β ～ ,γ ～ Is the emission coefficient of the generator set; p (P) G The output power of the generator set; e (P) G ) The pollution gas emission amount of the generator set;
and establishing an economic dispatch objective equation containing flexible load as follows:
min∑F B (P B )+∑F G (P G )-∑F D (P D )，
wherein:
wherein F is B 、F G 、F D The definition of the energy storage battery is the cost of the energy storage battery, the active power generation cost of the traditional fuel generator and the electricity utilization benefit of the flexible load, and alpha, beta, gamma and zeta are the cost parameters of the power generation and energy storage units respectively; a, b and c are parameters of the electricity utilization benefit of the flexible load; p (P) D And P B Power for flexible load and stored energy;
in order to simultaneously process two targets of environment and economy, the multi-target problem is converted into a single-target problem, the power balance is considered, and a multi-target model is established as follows:
wherein Δp= Σp D -∑P G -P REG +P d +P B Representing the active power difference of the system, P REG Representing renewable energy unit power; p (P) d Representing the total power of the pure load of the system; omega 1 、ω 2 Is a linear weighting factor;
(3) Converting the multi-objective model obtained in the step (2) to obtain a leader-follower rapid consistency scheduling model:
converting the formula (7) into the following unconstrained problem by using a Lagrangian multiplier method, and obtaining a scheduling model as follows:
wherein λ represents the lagrange multiplier;
for variable P in equation (8) D 、P G 、P B And respectively solving the partial derivatives with lambda to obtain the optimal solution of the scheduling model as follows:
(4) Noise-adding the consistency variable in the leader-follower fast consistency scheduling model:
to improve the privacy protection of the initial value, a continuous encryption function F is introduced for noise added by the initial iteration pq (z pq ) The node q is the outbound neighbor node of the node p in the strong communication graph G;
setting each encryption function F pq (z pq ) Can only be known by the neighboring nodes p and q, while each node randomly generates z for all (p, q) pq As a parameter of the encryption function, letRepresenting the noise fused by the encryption function, and obtaining:
wherein,only once calculated in the first iteration, i.e. t=1;
ν p (0) Represents added noise at t=0;
ν p (t) represents all iterations, i.e., t>=0, v containing t=0 p (0) Added noise, and v p (t) obeys the same independent distribution in each iteration, but each value is random;
the noise function of the consistency variable may be expressed by:
wherein μ is a constant and 0 < μ < 1, t is the number of iterations, μ t-1 Is the power series of mu t-1 times, theta p (t) is noise added by node p at the t-th iteration, whose upper absolute value is decaying exponentially with t;
when the iteration is carried out for the t time, the node p adds the noise consistency variableBroadcasting to neighbor nodes, and updating the consistency state of the node p by adopting a leader-follower fast consistency algorithm according to the formula (2) in the step (1) and considering the power balance of the system, wherein the consistency state is as follows:
in the method, in the process of the invention,is a power balance adjustment coefficient; />Representing the consistency state of the noise added by the node p; />Adding noise consistency variables to the neighbor node q of the node p;
according to formulas (3), (4), (5), (6) and (9) of consistency variables and power, and considering upper and lower limit constraints of active power of a generator, a flexible load and an energy storage battery in an actual micro-grid, establishing a node real power updating rule as follows:
wherein P is G，max/min Upper and lower limits of the output of the generator, P D,max/min Upper and lower limits of force, P, for flexible load B,max/min The upper limit and the lower limit of the power of the energy storage battery are set;
(5) The privacy of the leader-follower fast consistency scheduling model is improved by adding random values to the nodes, and the process is as follows:
(5.1) initializing:
calculation of P from node parameters p (0) And lambda (lambda) p (0) The leading node also needs to calculate global delta P (t) according to the random number power added by each node and the sum of which is zero;
parameter z of node broadcast encryption function pq And calculates an encryption function F pq (z pq ) Node is 0 from mean and sigma from variance 2 Is uniformly distributed in (a)Is generated by medium random v p (0) Wherein σ > 0, and θ p (0)＝ν p (0) Then calculate according to formula (10) to get +.>At the same time node calculates +.>And will noise the consistency variable +.>Broadcasting to neighbor nodes;
(5.2) iterating:
if |ΔP (t) | is not less than 0.001 orEach node updates its own ++according to formulas (12), (13), (14), (15)>And P p (t) the leader node also needs to calculate global Δp (t) from the random number powers added by each node to sum to zero;
v of each node p (t) fromRandom generation in the computer and updating the noise +.>The node then adds this noise to the consistency variable lambda p In (t), the consistency variable after noise addition is +.>Broadcasting to neighbor nodes;
(5.3) outputP p (t), ΔP (t), the addition of the random number is completed.
2. The micro-grid lead-following fast consistent multi-objective scheduling privacy protection method of claim 1, wherein: in the step (2), the energy storage battery cost, the active power generation cost of the traditional fuel generator and the electricity utilization benefit of the flexible load related to the economic dispatch are combined to obtain an economic dispatch objective equation of the flexible load.
3. The micro-grid lead-following fast consistent multi-objective scheduling privacy protection method of claim 1, wherein: in the step (2), a linear weighting method is adopted to convert the multi-objective problem into a single-objective problem, so that omega 1 +ω 2 =1 and 0 < ω 1 ,ω 2 ＜1。
4. The micro-grid lead-following fast consistent multi-objective scheduling privacy protection method of claim 1, wherein: in step (4), θ for all nodes in all iterations p The sum of (t) converges to zero and all θ p The sum of absolute values of (t) is bounded, i.eH is a constant, thereby ensuring that the system consistency variation converges and the optimality of the final value is solved.
CN202011597825.8A 2020-12-29 2020-12-29 Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method Active CN112713612B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011597825.8A CN112713612B (en) 2020-12-29 2020-12-29 Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011597825.8A CN112713612B (en) 2020-12-29 2020-12-29 Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method

Publications (2)

Publication Number Publication Date
CN112713612A CN112713612A (en) 2021-04-27
CN112713612B true CN112713612B (en) 2024-03-29

Family

ID=75546546

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011597825.8A Active CN112713612B (en) 2020-12-29 2020-12-29 Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method

Country Status (1)

Country Link
CN (1) CN112713612B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114529207B (en) * 2022-02-22 2024-05-28 国网湖北省电力有限公司电力科学研究院 Energy storage battery distributed economic dispatching method based on differential privacy mechanism

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109066812A (en) * 2018-10-29 2018-12-21 东北大学 A kind of electric system Multiobjective Optimal Operation method based on the double-deck consistency algorithm
CN109149568A (en) * 2018-09-10 2019-01-04 上海交通大学 A kind of interconnection micro-capacitance sensor and scheduling Price optimization method based on distributed agent
CN110602129A (en) * 2019-09-24 2019-12-20 苏州科技大学 Privacy protection optimization method based on average consistency of utility mechanism

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109149568A (en) * 2018-09-10 2019-01-04 上海交通大学 A kind of interconnection micro-capacitance sensor and scheduling Price optimization method based on distributed agent
CN109066812A (en) * 2018-10-29 2018-12-21 东北大学 A kind of electric system Multiobjective Optimal Operation method based on the double-deck consistency algorithm
CN110602129A (en) * 2019-09-24 2019-12-20 苏州科技大学 Privacy protection optimization method based on average consistency of utility mechanism

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于一致性的微网分布式能量管理调度策略;阮博等;《电力系统保护与控制》;20180801;第46卷(第15期);23-27 *

Also Published As

Publication number Publication date
CN112713612A (en) 2021-04-27

Similar Documents

Publication Publication Date Title
Chen et al. Mobile edge computing based task offloading and resource allocation in 5G ultra-dense networks
Anjana et al. A review on the features and technologies for energy efficiency of smart grid
Zhang et al. Distributed power allocation for coordinated multipoint transmissions in distributed antenna systems
Wang et al. Node energy consumption balanced multi-hop transmission for underwater acoustic sensor networks based on clustering algorithm
Yang et al. Distributed interference-aware power control in ultra-dense small cell networks: A robust mean field game
Chen et al. Multi-population coevolutionary dynamic multi-objective particle swarm optimization algorithm for power control based on improved crowding distance archive management in CRNs
Badr et al. Privacy-preserving and communication-efficient energy prediction scheme based on federated learning for smart grids
Pan et al. Differential privacy and IRS empowered intelligent energy harvesting for 6G Internet of Things
Gao et al. Dynamic access point and service selection in backscatter-assisted RF-powered cognitive networks
Huang et al. Renewable powered cellular networks: Energy field modeling and network coverage
Liu et al. Robust optimisation of power control for femtocell networks
Zhang et al. Power versus spectrum 2-D sensing in energy harvesting cognitive radio networks
CN112713612B (en) Micro-grid leading-following rapid consistency multi-target scheduling privacy protection method
Mahmud et al. Adaptive reinforcement learning framework for NOMA-UAV networks
Li et al. Reinforcement learning-based intelligent reflecting surface assisted communications against smart attackers
Tsuo et al. Energy-aware transmission control for wireless sensor networks powered by ambient energy harvesting: A game-theoretic approach
Cheng et al. Small-cell sleeping and association for energy-harvesting-aided cellular IoT with full-duplex self-backhauls: A game-theoretic approach
CN114529207B (en) Energy storage battery distributed economic dispatching method based on differential privacy mechanism
Sharma et al. Secrecy rate maximization in THz-aided heterogeneous networks: A deep reinforcement learning approach
Dang et al. Suboptimal energy efficient power allocation scheme for distributed antenna system with antenna selection
Zhou et al. Binary quantum elite particle swarm optimization algorithm for spectrum allocation in cognitive wireless medical sensor network
Fasoulakis et al. A new theoretical evaluation framework for satisfaction equilibria in wireless networks
Lan et al. UAV-Assisted Computation Offloading Towards Energy-Efficient Blockchain Operations in Internet of Things
Soleimanpour‐moghadam et al. Jointly optimal rate control and total transmission power for cooperative cognitive radio system
Yang et al. An energy‐efficient scheme for multirelay cooperative networks with energy harvesting

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