CN108964019A - A kind of polynary regulation method of power grid - Google Patents
A kind of polynary regulation method of power grid Download PDFInfo
- Publication number
- CN108964019A CN108964019A CN201810615099.4A CN201810615099A CN108964019A CN 108964019 A CN108964019 A CN 108964019A CN 201810615099 A CN201810615099 A CN 201810615099A CN 108964019 A CN108964019 A CN 108964019A
- Authority
- CN
- China
- Prior art keywords
- power grid
- power
- generator unit
- grid
- day
- 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
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
-
- 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
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A kind of polynary regulation method of power grid, system constantly adjust system regulation coefficient by successive ignition by repetition learning the previous day same period historical data in a period of time, determine that the node needs each generator unit power output numerical value of power grid.Then it is added to contributing needed for nodes all in power grid, obtains each generator unit plan power output situation of the period power grid one day after.Then this step is repeated in subsequent period, each generator unit power output of all period power grids of final adjustment whole day makes to plan a few days ago more accurate.For generating error, real-time monitoring can be carried out by AGC, to guarantee that mains frequency is stablized.
Description
Technical field
The invention belongs to intelligent power grid technology field, and in particular to a kind of adaptively to be regulated and controled based on multi-model neural network
Polynary electric network coordination control method.
Background technique
In recent years, new round energy revolution just promotes in the world, and China's energy industry structural adjustment is accelerated, with light
Volt, wind-powered electricity generation are that the clean energy resource grid connection capacity of representative is continuously increased, and the utilization of the energy gradually forms high efficiency, diversification, intelligence
The characteristics of change, it is desirable that power grid accelerates to improve, to adapt to the energy distributing rationally in China.In face of global environment deteriorate with
Resource scarcity pressure, energy-saving and emission-reduction, raising energy efficiency of energy utilization have become the requirement that the following power grid must satisfy.At present
Existing power grid regulation mainly uses two kinds of scheduling methods of operation plan a few days ago and Automatic Generation Control (AGC).For dispatching a few days ago
Plan, due to there is the sufficient time to formulate scheduling scheme, it is possible on the basis learnt to previous a large amount of historical datas
On, make the operation plan of following 24 periods.But it is big due to planning manpower participation factor a few days ago, coverage is wide, department
More, often error is larger with each node real-time load of power grid for each unit output；Again due to wind-power electricity generation, photovoltaic power generation it is uncontrollable
Property and it is uncertain the disadvantages of, increase the difficulty of power grid regulation.
Summary of the invention
Goal of the invention:
The present invention is disclosed one kind and is adaptively adjusted based on multi-model neural network for deficiency existing for currently existing technology
The theoretical smart grid of control regulates and controls method.It is the problems of previous the purpose is to solve.
Technical solution:
To achieve the goals above, the present invention takes following technical scheme to realize:
To each single grid nodes, system passes through repetition learning the previous day same period historical data in a period of time,
By successive ignition, system regulation coefficient is constantly adjusted, determines that the node needs each generator unit power output numerical value of power grid.Then right
Power output needed for all nodes is added in power grid, obtains each generator unit plan power output situation of the period power grid one day after.Then exist
Subsequent period repeats this step, and each generator unit power output of all period power grids of final adjustment whole day makes to plan a few days ago more accurate.
For generating error, real-time monitoring can be carried out by AGC, to guarantee that mains frequency is stablized.
Specific step is as follows:
Step 1: data acquisition:
Each node load x of different periods power grid in power grid in P daysiAnd each fired power generating unit power output al,p, each thermoelectricity unit goes out
Power bj,p, each Wind turbines contribute ck,p, photovoltaic power generation contribute dp, energy-storage battery contribute sp, and each power grid section of synchronization daily
Point prediction loadEach generating set plans generated energy a few days ago (l=1,2, L L), (j
=1,2, L J), (k=1,2, L K), L, J, K are respectively the number of thermoelectricity, thermoelectricity, Wind turbines；(i=1,2, L 24), i is
When number of segment.(p=1,2, L P), P are total observation number of days.Such as al,pIndicate first of fired power generating unit pth day power output.
Step 2: calculating single node regulator control system autonomous learning model.
Step 2.1: definition is directed to m-th of node control system call interception model:
Wherein (m=1,2, L M), M are total node number in power grid.
Define planning model a few days ago:
Step 2.2: calculate pth day day part system call interception error:
Step 2.3: defining regulator control system autonomous learning model.
Power output scheme of the P+1 days each units to m-th of node:
Wherein WPTo regulate and control coefficient, A after iterationPTo regulate and control matrix, αP、βPFor the correction factor of the P+1 times iteration.
Step 3: the regulator control system adaptive control model for being directed to certain single node is determined by iteration
Step 3.1: calculating correction factor
Pth time iterated revision coefficient:
Step 3.2: calculating regulation matrix
Pth time iteration regulates and controls matrix:
Step 3.3: P iteration being carried out to primary data, determines that each power generation needed for the P+1 days m-th grid nodes is single
Member power output.
Step 4: all generator unit gross capabilities are obtained to the power output superposition of each generator unit needed for each node:
Beneficial effect
By previous experiences it is found that each node day injecting power variation of power grid has stronger similar features, pass through acquisition electricity
Net panorama operation information records daily synchronization grid nodes load variations and each machine of same time thermoelectricity, thermoelectricity, wind-powered electricity generation
Group and photovoltaic power generation, energy-storage system power output situation, and by the previous data of power grid regulation systematic learning and self adaptive control, it builds
Vertical progressive learning type power grid regulation strategy.This method can be generated electricity a few days ago with accurate calculation power grid, balance each node load of power grid with
Grid generation amount, electric power grid frequency stabilization.
That is, the method for the present invention establishes system self-adaption control mould by power grid regulation systematic learning historical data
Precision is planned in type, the power output of comprehensive regulation thermoelectricity, thermoelectricity, wind-powered electricity generation, photovoltaic, the multiple generator units of energy storage, raising a few days ago.Separately
Outside, relative to conventional scheduling method, the method for the present invention is less by artifical influence factor, improves dispatching efficiency.
Detailed description of the invention
Fig. 1 is that grid control system adaptively regulates and controls flow chart.
Fig. 2 is drawing self-adaptive controlled for the multi-model neural network of single node.
Fig. 3 is the point self-adapted control plan power generation curve of certain partial electric grid 16-20 one day, former Plan Curve and reality a few days ago
Load curve.
Specific embodiment
To a kind of above-mentioned power grid universe Multi-value coordination control method adaptively regulated and controled based on multi-model neural network, with certain
The practical operation situation of power grid implements this method for calculating:
Certain partial electric grid totally 5 nodes, daily i-th of period load in 10 days are as follows:
| Number of days | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Load (kVA) | 499.5 | 502.3 | 511.1 | 496.2 | 490.9 | 503.7 | 508.9 | 500.1 | 492.8 | 503.3 |
Each grid nodes period load in 10 days are as follows:
The period in the power grid 10 days, each unit output situation are as follows:
It is as follows to plan power output situation in the daytime for each generating set in the power grid 10 days:
In addition, α0=α1=1, β0=β1=1,Estimated value
1. being directed to this partial electric grid No.1 node, first day, the second day data are brought into correction factors formula:
Calculate to obtain α1,2=1.032, β1,2=0.984,
2. the correction factor of first time iteration is substituting to model:
Solve W1,1=0.4897,
3. repetition previous step P-1 (P=10) is secondary, W is obtained1,11=0.5011,I.e.
It is No.1 node to each generating set power output demand.
4. repeating first two steps, continue to calculate by iteration
5. pair each node does and the plan a few days ago as adjusted by system self-adaption each generator unit power output demand.
The point self-adapted control plan power generation curve of certain power grid 16-20 one day, former Plan Curve and realized load curve a few days ago
As shown in Figure 3.
As seen in Figure 3, power grid self adaptive control generation schedule in 16-20 point time range obviously closer in
Actual power generation.
In conclusion the present invention adaptively regulates and controls theory based on multi-model neural network, mentions for previous tactful drawback
A kind of artificial intelligence power grid regulation strategy is gone out.In existing power system network, by learning previous data, and monitor real-time
Network load, adjustment are planned a few days ago.Since the northern area of China mainly uses thermoelectricity, thermoelectricity, and a small amount of wind-powered electricity generation and photovoltaic hair
Electricity, so we only discuss the dispatching of power netwoks situation for containing only thermoelectricity, wind-powered electricity generation, hot spot, photovoltaic power generation and energy-storage system.
Claims (4)
1. a kind of polynary regulation method of power grid, it is characterised in that: this method is to each single grid nodes, when system passes through one section
Same period historical data on the day before interior repetition learning constantly adjusts system regulation coefficient, determines the section by successive ignition
Point needs each generator unit power output numerical value of power grid；Then to needed for nodes all in power grid contribute be added, obtain one day after this when
Each generator unit plan power output situation of section power grid；Then this step, all period electricity of final adjustment whole day are repeated in subsequent period
Each generator unit power output is netted, makes to plan a few days ago more accurate；For generating error, real-time monitoring is carried out by AGC.
2. a kind of polynary regulation method of power grid according to claim 1, it is characterised in that:
Specific step is as follows for the program:
Step 1: data acquisition:
Each node load x of different periods power grid in power grid in P daysiAnd each fired power generating unit power output al,p, each thermoelectricity unit output bj,p, each wind
Electric unit output ck,p, photovoltaic power generation contribute dp, energy-storage battery contribute sp, and each grid nodes of synchronization predict load dailyRespectively
Generating set plans generated energy a few days ago
L, J, K are respectively the number of thermoelectricity, thermoelectricity, Wind turbines；(i=1,2, L 24), number of segment when i is；(p=1,2, L P), P is
Total observation number of days, such as al,pIndicate first of fired power generating unit pth day power output；
Step 2: calculate single node regulator control system autonomous learning model:
Step 3: the regulator control system adaptive control model for being directed to certain single node is determined by iteration:
Step 4: all generator unit gross capabilities are obtained to the power output superposition of each generator unit needed for each node:
3. a kind of polynary regulation method of power grid according to claim 2, it is characterised in that:
In step (2):
Step 2.1: definition is directed to m-th of node control system call interception model:
Wherein (m=1,2, L M), M are total node number in power grid；
Define planning model a few days ago:
Step 2.2: calculate pth day day part system call interception error:
Step 2.3: defining regulator control system autonomous learning model；
Power output scheme of the P+1 days each units to m-th of node:
Wherein WPTo regulate and control coefficient, A after iterationPTo regulate and control matrix, αP、βPFor the correction factor of the P+1 times iteration.
4. a kind of polynary regulation method of power grid according to claim 2, it is characterised in that: in step (3):
Step 3.1: calculating correction factor
Pth time iterated revision coefficient:
Step 3.2: calculating regulation matrix
Pth time iteration regulates and controls matrix:
Step 3.3: P iteration being carried out to primary data, each generator unit needed for determining the P+1 days m-th grid nodes goes out
Power.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810615099.4A CN108964019B (en) | 2018-06-14 | 2018-06-14 | Power grid multi-element regulation and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810615099.4A CN108964019B (en) | 2018-06-14 | 2018-06-14 | Power grid multi-element regulation and control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108964019A true CN108964019A (en) | 2018-12-07 |
| CN108964019B CN108964019B (en) | 2022-05-27 |
Family
ID=64489105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810615099.4A Active CN108964019B (en) | 2018-06-14 | 2018-06-14 | Power grid multi-element regulation and control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108964019B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104881706A (en) * | 2014-12-31 | 2015-09-02 | 天津弘源慧能科技有限公司 | Electrical power system short-term load forecasting method based on big data technology |
| CN105576699A (en) * | 2016-01-12 | 2016-05-11 | 四川大学 | Independent micro-grid energy storage margin detection method |
| US20160156186A1 (en) * | 2010-07-02 | 2016-06-02 | General Electric Technology Gmbh | Multi-interval dispatch method for enabling dispatchers in power grid control centers to manage changes background |
-
2018
- 2018-06-14 CN CN201810615099.4A patent/CN108964019B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160156186A1 (en) * | 2010-07-02 | 2016-06-02 | General Electric Technology Gmbh | Multi-interval dispatch method for enabling dispatchers in power grid control centers to manage changes background |
| CN104881706A (en) * | 2014-12-31 | 2015-09-02 | 天津弘源慧能科技有限公司 | Electrical power system short-term load forecasting method based on big data technology |
| CN105576699A (en) * | 2016-01-12 | 2016-05-11 | 四川大学 | Independent micro-grid energy storage margin detection method |
Non-Patent Citations (2)
| Title |
|---|
| DAUD MUSTAFA MINHAS ET AL: "Load control for supply-demand balancing under Renewable Energy forecasting", 《2017 IEEE SECOND INTERNATIONAL CONFERENCE ON DC MICROGRIDS (ICDCM)》 * |
| 陶旭嫣等: "一种未来电网调控策略研究", 《山东电力技术》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108964019B (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107301472B (en) | Distributed photovoltaic planning method based on scene analysis method and voltage regulation strategy | |
| CN105846461B (en) | Control method and system for large-scale energy storage power station self-adaptive dynamic planning | |
| CN109492861B (en) | Method for decomposing medium-term electricity quantity trading plan of cascade hydropower station group | |
| CN104616069B (en) | A kind of annual generation schedule rolls decomposition optimization | |
| CN107958300A (en) | A kind of more microgrid interconnected operation coordinated scheduling optimization methods for considering interactive response | |
| CN104063808B (en) | Trans-provincial power transmission cascade hydropower station group peak-shaving dispatching two-phase search method | |
| CN104915737A (en) | Coordinating economic dispatching method for power system containing large-scale wind power based on risk control | |
| CN108039737B (en) | Source-grid-load coordinated operation simulation system | |
| CN109066805A (en) | A kind of transregional interconnected network generating and transmitting system dynamic dispatching optimization method | |
| CN104065168B (en) | A kind of dynamic frequency-conversion collecting method of wind-solar-storage joint generating state monitoring | |
| CN106026084B (en) | A kind of AGC power dynamic allocation methods based on virtual power generation clan | |
| CN104239967A (en) | Multi-target economic dispatch method for power system with wind farm | |
| CN104333047B (en) | Real-time rolling planning method applied to wind power integration of power system | |
| CN104037761B (en) | AGC power multi-objective random optimization distribution method | |
| CN103683337A (en) | Interconnected power system CPS instruction dynamic allocation and optimization method | |
| CN109149648A (en) | A kind of adaptive width Dynamic Programming intelligent power generation control method | |
| CN103050998B (en) | Thermal power system dynamic scheduling method of wind power integration | |
| CN107612045A (en) | A kind of Power Plant generated energy intelligent dispensing system and method | |
| CN108964121B (en) | Wind, light and water real-time control method considering water and power planning and power target in day before water and power | |
| CN112186802A (en) | Multi-time scale rolling scheduling method and system for dynamic economic scheduling | |
| US20240006891A1 (en) | Two-stage self-organizing optimized aggregation method and system for distributed resources of virtual power plant (vpp) | |
| CN111864811B (en) | Photovoltaic capacity configuration method of water-light complementary system considering power grid output demand | |
| CN108667077A (en) | A kind of wind storage association system Optimization Scheduling | |
| CN113241780B (en) | Power grid secondary frequency modulation control method and system based on bald eagle search algorithm | |
| CN114336592A (en) | Wind power plant AGC control method based on model predictive control |
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 |