CN117039980A - Combined system optimal scheduling method based on deep reinforcement learning - Google Patents

Combined system optimal scheduling method based on deep reinforcement learning Download PDF

Info

Publication number
CN117039980A
CN117039980A CN202310783760.3A CN202310783760A CN117039980A CN 117039980 A CN117039980 A CN 117039980A CN 202310783760 A CN202310783760 A CN 202310783760A CN 117039980 A CN117039980 A CN 117039980A
Authority
CN
China
Prior art keywords
power
max
cost
generator set
constraint
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.)
Pending
Application number
CN202310783760.3A
Other languages
Chinese (zh)
Inventor
张一凡
马玉杰
周佳妮
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.)
China Three Gorges University CTGU
Original Assignee
China Three Gorges University CTGU
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 China Three Gorges University CTGU filed Critical China Three Gorges University CTGU
Priority to CN202310783760.3A priority Critical patent/CN117039980A/en
Publication of CN117039980A publication Critical patent/CN117039980A/en
Pending legal-status Critical Current

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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/092—Reinforcement learning
    • 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
    • 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
    • H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
    • 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
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • 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
    • H02J3/007—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources
    • H02J3/0075—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources for providing alternative feeding paths between load and source according to economic or energy efficiency considerations, e.g. economic dispatch
    • 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
    • H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
    • 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]
    • 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
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • 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
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/22—The renewable source being solar energy
    • H02J2300/24—The renewable source being solar energy of photovoltaic origin
    • 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
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/28—The renewable source being wind energy
    • 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
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/40—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation wherein a plurality of decentralised, dispersed or local energy generation technologies are operated simultaneously

Landscapes

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

Abstract

A joint system optimization scheduling method based on deep reinforcement learning comprises the following steps: s1: acquiring current power grid operation parameters and acquiring reservoir water level change, wind speed of a wind power plant, power generation load of a thermal power plant and illumination intensity and duration; s2: establishing an objective function and constraint conditions so as to obtain an optimal adjustment factor in a power grid system model; s3: solving a linear programming problem formed by an objective function and constraint conditions to obtain adjustment data; s4: and (3) adjusting the power generation plan according to the data obtained in the step (S3), and then carrying out grid connection. The invention optimizes the unit combination according to the characteristics of the hydroelectric and thermal power units, and ensures the optimized distribution among the hydroelectric and thermal power units on the basis of a scientific and reasonable start-stop plan, thereby reducing the influence on a power grid system when wind power and photoelectricity are connected.

Description

Combined system optimal scheduling method based on deep reinforcement learning
Technical Field
The invention belongs to the field of power dispatching, and particularly relates to a joint system optimization dispatching method based on deep reinforcement learning.
Background
How to combine and arrange start-stop plans in a certain operation period of an operable unit of an electric power system can meet load requirements and minimize the operation cost (fuel consumption) of the whole system, so that resources are saved as much as possible, which is an important aspect in the economic dispatch research of the electric power system, and the economic benefit brought by the operable unit is also considerable.
Because of the variability and uncertainty of wind power and photoelectricity, large-scale wind power and photoelectricity grid connection firstly cause serious difficulty to the economic operation scheduling of a power grid, and finally threaten the safe and stable operation of the power grid, so that the wind power plant and the photoelectricity plant cannot form scale effect due to the fact that the utilization time of the wind power plant and the photoelectricity plant is too low.
In order to solve the problem of safety and stability of large-scale wind power and photovoltaic grid-connected operation, optimization and coordination research of a whole-grid power supply including wind power grid-connected hydroelectric power, thermal power and thermal power units is required.
Disclosure of Invention
In view of the technical problems in the background art, the combined system optimization scheduling method based on deep reinforcement learning provided by the invention performs unit combination optimization according to the characteristics of hydroelectric and thermal power units, and ensures optimal allocation between the hydroelectric and thermal power units on the basis of a scientific and reasonable start-stop plan, thereby reducing the influence on a power grid system when wind power and photoelectricity are connected.
In order to solve the technical problems, the invention adopts the following technical scheme:
a joint system optimization scheduling method based on deep reinforcement learning comprises the following steps:
s1: acquiring current power grid operation parameters and acquiring reservoir water level change, wind speed of a wind power plant, power generation load of a thermal power plant and illumination intensity and duration;
s2: establishing an objective function and constraint conditions so as to obtain an optimal adjustment factor in a power grid system model;
s3: solving a linear programming problem formed by an objective function and constraint conditions to obtain adjustment data;
s4: and (3) adjusting the power generation plan according to the data obtained in the step (S3), and then carrying out grid connection.
Preferably, in step S2, the objective function established includes fuel cost, generator cost, and backup generator set cost;
fuel cost:
wherein F is the total running cost of each unit in each time period; n is the number of generator sets; m is a time interval; u (u) it For i states of the unit in the t time period: 1 is an operation state, and 0 is a stop state; p is p it The active output of the i machine set in the t time period is obtained; f (f) it (p it ) The running cost of the unit in the t time period is i;
the active output is:
wherein p is D For the system load in the t period, p L For active network loss in t time period, p it ,p it,max ,p it,min The active output of the unit in the t time period is i, and the upper limit and the lower limit of the active output are the active output of the unit in the t time period;
the cost of the generator set is as follows:
wherein N is the number of the generator sets; p (P) G，i，t The output of the generator set i in the scheduling period t is provided; a, a i ，b i ，c i Is a cost factor related to the characteristics of the generator set i;
cost of the standby generator set:
wherein: r is R U,i,t And R is D,i,t Reserve up and down capacities reserved for generators, c RU And c RD For its corresponding cost coefficient;
is the expected minimum of the random variable, where E (·) represents the expected value of the random variable; t is the scheduling period length; f (F) G,T ，F R,T The cost of the generator and the cost of the standby generator set in the dispatching period t are respectively;
the change of the reservoir water level is as follows:
wherein V is t For the remaining water quantity at time t (m 3 ) The method comprises the steps of carrying out a first treatment on the surface of the Δt is the optimization time interval (h); psi phi type t Inflow water flow (m) as tributary at time t 3 /s)。
Preferably, in step S2, the established constraint conditions include an output constraint condition, a generator set climbing rate constraint condition, a standby generator set constraint condition, a water level constraint condition, and an installation scale constraint condition;
force constraint conditions:
P G，i，t，min ≤P G，i，t ≤P G，i，t，max，
wherein P is G，i，t，max And P G，i，t，min The upper and lower limits of the output of the conventional unit i in the scheduling period t are respectively set;
and (3) a constraint condition of the climbing rate of the generator set:
wherein U is G，i And D G，i Respectively the maximum ascending and descending climbing power of the generator set i in a scheduling period;
constraint conditions of the standby generator set:
water level constraint conditions:
wherein,and->Constraint on upper and lower limits of the reservoir capacity; />And->Constraint on upper and lower limits of the leakage flow; />Andconstraint on upper and lower limits of force;
installation scale constraint conditions:
Z S,min ≤Z S ≤Z S,max
Z f,min ≤Z f ≤Z f,max
Z h,min ≤Z h ≤Z h,max
Z g,min ≤Z g ≤Z g,max
wherein Z is S ，Z f ，Z h ，Z g The installed scales of water power, wind power, thermal power and photoelectricity respectively, Z S,min ，Z f,min ，Z h,min ，Z g,min Lower limit of installation scale planning for hydropower, wind power, thermal power and photoelectricity of target year respectively, Z S,max ，Z f,max ，Z h,max ，Z g,max The upper limit of the installation scale planning of hydropower, wind power, thermal power and photoelectricity in the target year is respectively set;
in step S4, the grid-connected power is:
P t grid ＝P t pv +P t phs -P t load +P t page
P t pv the actual power of the photovoltaic at the moment t; p (P) t load The power is required for the load at the moment t; p (P) t phs For the power emitted by the hydropower station at time t, P t page The power is generated by the thermal power plant at the time t.
Preferably, the objective function and constraint are both obtained using a DDPG algorithm.
The invention has the following beneficial effects:
1. the invention optimizes the unit combination according to the characteristics of the hydroelectric and thermal power units, ensures the optimized distribution among the hydroelectric and thermal power units on the basis of a scientific and reasonable start-stop plan, thereby reducing the influence on a power grid system when wind power and photoelectricity are connected,
2. in the short-term operation scheduling of the power system, a certain power generation reserve capacity is reserved for the system in addition to arranging enough power generation capacity to meet the predicted load in consideration of uncertainty of wind power and load and possibility of power generation equipment failure. When wind power and photoelectric output suddenly drop, the stability of a power grid system is ensured through the rapid adjustment capability of a designated thermal power unit or a designated hydroelectric unit; when wind power and photoelectric output suddenly increase, the stability of the power grid system is ensured through the rapid adjustment capability of the designated thermal power generating unit or hydroelectric generating unit.
3. Under the condition that a large amount of wind power and photoelectricity are suddenly lost in the system, the method can provide enough output power to ensure the stability of a power grid, and meanwhile, the cost of an operation unit is minimized.
Drawings
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of the optimal allocation of the present invention;
FIG. 3 is a graph of wind power generation diagram of the present invention.
Detailed Description
Examples:
the preferred scheme is as shown in fig. 1 to 3, and the method for optimizing and scheduling the combined system based on deep reinforcement learning comprises the following steps:
s1: acquiring current power grid operation parameters and acquiring reservoir water level change, wind speed of a wind power plant, power generation load of a thermal power plant and illumination intensity and duration;
firstly, predicting wind power output according to wind speed forecast data, historical wind power data and historical wind speed data, giving a confidence level, and calculating a wind power output interval by adopting an empirical method; according to the wind power output interval, arranging the water power output process to ensure that the water power output provides sufficient flexibility as far as possible so as to cope with the fluctuation of wind power, and simultaneously providing more peak regulation electric quantity to ensure the stability of the water power output as far as possible;
s2: establishing an objective function and constraint conditions so as to obtain an optimal adjustment factor in a power grid system model;
and constructing a water and electricity-thermal power combined system optimization scheduling model according to the output model of the thermal power plant thermal power unit, the output of the hydroelectric power unit and the output of the wind power unit.
The output of the hydroelectric generating set is calculated according to the existing output model of any hydroelectric generating set, and the output of the wind generating set can be calculated according to the existing output model of any wind generating set.
And establishing an objective function of an optimal scheduling model of the combined water-fire machine set system by taking the minimum total energy consumption of the combined water-fire machine set system as the objective function.
The established objective functions include fuel cost, generator cost, and backup generator set cost.
Fuel cost:
wherein F is the total running cost of each unit in each time period; n is the number of generator sets; m is a time interval; u (u) it For i states of the unit in the t time period: 1 is an operation state, and 0 is a stop state; p is p it The active output of the i machine set in the t time period is obtained; f (f) it (p it ) And the running cost of the unit in the t time period is i.
The active output is:
wherein p is D For the system load in the t period, p L For active network loss in t time period, p it ,p it,max ,p it,min The active output of the i machine set in the t time period and the upper limit and the lower limit thereof are provided.
The cost of the generator set is as follows:
in order to ensure the safe operation of the system, a certain rotation reserve should be reserved to balance the randomness prediction error. The total cost of reserved spares for the scheduling period t can be expressed as:
wherein N is the number of the generator sets; p (P) G，i，t The output of the generator set i in the scheduling period t is provided; a, a i ，b i ，c i Is a cost factor associated with the characteristics of genset i.
Cost of the standby generator set:
wherein: r is R U,i,t And R is D,i,t Reserve up and down capacities reserved for generators, c RU And c RD For its corresponding cost factor.
Is the expected minimum of the random variable, where E (·) represents the expected value of the random variable; t is the scheduling period length; f (F) G,T ，F R,T The cost of the generator and the cost of the standby generator set in the scheduling period t are respectively.
The change of the reservoir water level is as follows:
hydropower stations should meet the water level remaining water volume and inflow water flow constraints.
Wherein V is t For the remaining water quantity at time t (m 3 ) The method comprises the steps of carrying out a first treatment on the surface of the Δt is the optimization time interval (h); psi phi type t Inflow water flow (m) as tributary at time t 3 /s)。
The established constraint conditions comprise an output constraint condition, a generator set climbing rate constraint condition, a standby generator set constraint condition, a water level constraint condition and a loading scale constraint condition.
Force constraint conditions:
because the starting and stopping time of the thermal power unit is longer, the starting and stopping of the thermal power unit in the day are not considered, namely the power generation state of the thermal power unit is once determined, and the power generation state of the thermal power unit cannot change in the day.
P G，i，t，min ≤P G，i，t ≤P G，i，t，max，
Wherein P is G，i，t，max And P G，i，t，min The upper and lower limits of the output of the conventional unit i in the scheduling period t are respectively set.
And (3) a constraint condition of the climbing rate of the generator set:
wherein U is G，i And D G，i The maximum ascending and descending climbing power of the generator set i in a scheduling period is respectively.
Constraint conditions of the standby generator set:
water level constraint conditions:
wherein,and->Constraint on upper and lower limits of the reservoir capacity; />And->Constraint on upper and lower limits of the leakage flow; />Andconstraint on upper and lower limits of force.
Installation scale constraint conditions:
Z S,min ≤ Z S ≤ Z S,max
Z f,min ≤ Z f ≤ Z f,max
Z h,min ≤ Z h ≤ Z h,max
Z g,min ≤ Z g ≤ Z g,max
wherein Z is S ，Z f ，Z h ，Z g The installed scales of water power, wind power, thermal power and photoelectricity respectively, Z S,min ，Z f,min ，Z h,min ，Z g,min Lower limit of installation scale planning for hydropower, wind power, thermal power and photoelectricity of target year respectively, Z S,max ，Z f,max ，Z h,max ，Z g,max The upper limit of the installation scale planning of hydropower, wind power, thermal power and photoelectricity in the target year is respectively set.
Grid-connected power:
P t grid ＝P t pv +P t phs -P t load +P t page
wherein P is t pv The actual power of the photovoltaic at the moment t; p (P) t load The power is required for the load at the moment t; p (P) t phs For the power emitted by the hydropower station at time t,P t page the power is generated by the thermal power plant at the time t.
S3: solving a linear programming problem formed by an objective function and constraint conditions to obtain adjustment data;
according to the hydropower station-thermal power plant combined optimization scheduling method, the influence on wind power consumption of the hydropower station-thermal power plant combined system is considered by a hydropower-thermal power plant combined system optimization scheduling model; the optimal scheduling model of the water-electricity-thermal power combined system aims at the minimum total energy consumption, so that reasonable allocation of resources is achieved; the hydrothermal and electric combined system is a multidimensional, complex and nonlinear optimization problem, the calculation difficulty is high by applying a traditional optimization algorithm, the calculation complexity of the system is reduced by adopting a DDPG algorithm, the system can be quickly converged to the optimal value of the system, and the hydrothermal and electric combined system can be used for the optimization problem of a large-scale large system in actual engineering.
S4: and (3) adjusting the power generation plan according to the data obtained in the step (S3), and then carrying out grid connection.
The objective function and constraint are obtained using a DDPG algorithm.
The above embodiments are only preferred embodiments of the present invention, and should not be construed as limiting the present invention, and the scope of the present invention should be defined by the claims, including the equivalents of the technical features in the claims. I.e., equivalent replacement modifications within the scope of this invention are also within the scope of the invention.

Claims (5)

1. A joint system optimization scheduling method based on deep reinforcement learning is characterized in that: the method comprises the following steps:
s1: acquiring current power grid operation parameters and acquiring reservoir water level change, wind speed of a wind power plant, power generation load of a thermal power plant and illumination intensity and duration;
s2: establishing an objective function and constraint conditions so as to obtain an optimal adjustment factor in a power grid system model;
s3: solving a linear programming problem formed by an objective function and constraint conditions to obtain adjustment data;
s4: and (3) adjusting the power generation plan according to the data obtained in the step (S3), and then carrying out grid connection.
2. The joint system optimization scheduling method based on deep reinforcement learning according to claim 1, wherein: in step S2, the objective function includes fuel cost, generator cost, and backup generator set cost;
fuel cost:
wherein F is the total running cost of each unit in each time period; n is the number of generator sets; m is a time interval; u (u) it For i states of the unit in the t time period: 1 is an operation state, and 0 is a stop state; p is p it The active output of the i machine set in the t time period is obtained; f (f) it (p it ) The running cost of the unit in the t time period is i;
the active output is:
wherein p is D For the system load in the t period, p L For active network loss in t time period, p it ,p it,max ,p it,min The active output of the unit in the t time period is i, and the upper limit and the lower limit of the active output are the active output of the unit in the t time period;
the cost of the generator set is as follows:
wherein N is the number of the generator sets; p (P) G，i，t The output of the generator set i in the scheduling period t is provided; a, a i ，b i ，c i Is a cost factor related to the characteristics of the generator set i;
cost of the standby generator set:
wherein: r is R U,i,t And R is D,i,t Reserve up and down capacities reserved for generators, c RU And c RD For its corresponding cost coefficient;
is the expected minimum of the random variable, where E (·) represents the expected value of the random variable; t is the scheduling period length; f (F) G,T ，F R,T The cost of the generator and the cost of the standby generator set in the dispatching period t are respectively;
the change of the reservoir water level is as follows:
wherein V is t For the remaining water quantity at time t (m 3 ) The method comprises the steps of carrying out a first treatment on the surface of the Δt is the optimization time interval (h); psi phi type t Inflow water flow (m) as tributary at time t 3 /s)。
3. The joint system optimization scheduling method based on deep reinforcement learning according to claim 1, wherein: in step S2, the established constraint conditions include an output constraint condition, a generator set climbing rate constraint condition, a standby generator set constraint condition, a water level constraint condition and an installation scale constraint condition;
force constraint conditions:
P G，i，t，min ≤P G，i，t ≤P G，i，t，max，
wherein P is G，i，t，max And P G，i，t，min The upper and lower limits of the output of the conventional unit i in the scheduling period t are respectively set;
and (3) a constraint condition of the climbing rate of the generator set:
wherein U is G，i And D G，i Respectively the maximum ascending and descending climbing power of the generator set i in a scheduling period;
constraint conditions of the standby generator set:
water level constraint conditions:
wherein,and->Constraint on upper and lower limits of the reservoir capacity; />And->Constraint on upper and lower limits of the leakage flow; />And->Constraint on upper and lower limits of force;
installation scale constraint conditions:
Z S,min ≤Z S ≤Z S,max
Z f,min ≤Z f ≤Z f,max
Z h,min ≤Z h ≤Z h,max
Z g,min ≤Z g ≤Z g,max
wherein Z is S ，Z f ，Z h ，Z g The installed scales of water power, wind power, thermal power and photoelectricity respectively, Z S,min ，Z f,min ，Z h,min ，Z g,min Lower limit of installation scale planning for hydropower, wind power, thermal power and photoelectricity of target year respectively, Z S,max ，Z f,max ，Z h,max ，Z g,max The upper limit of the installation scale planning of hydropower, wind power, thermal power and photoelectricity in the target year is respectively set.
4. The joint system optimization scheduling method based on deep reinforcement learning according to claim 1, wherein: in step S4, the grid-connected power is:
P t grid ＝P t pv +P t phs -P t load +P t page
P t pv the actual power of the photovoltaic at the moment t; p (P) t load The power is required for the load at the moment t; p (P) t phs For the power emitted by the hydropower station at time t, P t page The power is generated by the thermal power plant at the time t.
5. The joint system optimization scheduling method based on deep reinforcement learning according to claim 1, wherein: the objective function and the constraint condition are obtained by adopting a DDPG algorithm.
CN202310783760.3A 2023-06-29 2023-06-29 Combined system optimal scheduling method based on deep reinforcement learning Pending CN117039980A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310783760.3A CN117039980A (en) 2023-06-29 2023-06-29 Combined system optimal scheduling method based on deep reinforcement learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310783760.3A CN117039980A (en) 2023-06-29 2023-06-29 Combined system optimal scheduling method based on deep reinforcement learning

Publications (1)

Publication Number Publication Date
CN117039980A true CN117039980A (en) 2023-11-10

Family

ID=88621546

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310783760.3A Pending CN117039980A (en) 2023-06-29 2023-06-29 Combined system optimal scheduling method based on deep reinforcement learning

Country Status (1)

Country Link
CN (1) CN117039980A (en)

Similar Documents

Publication Publication Date Title
CN108599269B (en) Large power grid self-adaptive robust standby optimization method considering risk cost
CN106786799B (en) Power stepped power generation plan optimization method for direct current connecting line
CN108092324B (en) AGC control system and control method for wind power participating in peak shaving frequency modulation
CN109936170A (en) Consider the honourable extreme misery complementation coordination optimization dispatching method of power supply flexibility nargin
CN103219751B (en) Control method of active power of clustered wind power plants
CN107104462B (en) A method of it is dispatched for wind power plant energy storage
CN112886645B (en) New energy power system operation simulation method based on ultrahigh hydrogen energy proportion
CN116667395B (en) Capacity allocation method for water-wind-solar-energy-storage complementary pump station based on cascade hydropower transformation
CN110829408A (en) Multi-domain scheduling method considering energy storage power system based on power generation cost constraint
CN110661301A (en) Capacity allocation optimization method for water-light-storage multi-energy complementary power generation system
CN110571868A (en) Optimal configuration method of microgrid
Ran et al. Economic dispatch of off-grid photovoltaic generation system with hybrid energy storage
Wang et al. Multi-objective optimal dispatch of wind-integrated power system based on distributed energy storage
CN115842375A (en) Collaborative scheduling optimization method considering energy Internet wind, light, water, fire and electricity storage system
CN115566731A (en) Flexible power supply planning method based on system multi-energy complementation
CN117039980A (en) Combined system optimal scheduling method based on deep reinforcement learning
CN115765034A (en) Photo-thermal-photovoltaic-thermal power combined cooperative control method and system
Huang et al. A multi-time scale scheduling method for wind-PV-pumped storage-electrochemical energy storage integrated generation system
Ma et al. Two-stage optimal dispatching based on wind-photovoltaic-pumped storage-thermal power combined power generation system
CN114465226A (en) Method for establishing multi-level standby acquisition joint optimization model of power system
CN211183436U (en) Light storage micro-grid system based on energy prediction
CN110417002B (en) Optimization method of island micro-grid energy model
CN112884191A (en) Thermoelectric day-ahead scheduling model based on network source coordination and calculation method
Ma et al. Multi-objective optimal scheduling of power system considering the coordinated operation of photovoltaic-wind-pumped storage hybrid power
CN113839412B (en) Hydrogen production peak regulation calculation method and device for photovoltaic power station

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