Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Adaptive optimal AGC control method based on integral reinforcement learning

A technology of reinforcement learning and control methods, applied in the direction of reducing/preventing power oscillation, AC network circuits, electrical components, etc., can solve problems such as limited number of connection lines, inability to converge to the optimum, frequency deviation, etc.

Pending Publication Date: 2021-09-03
STATE GRID CHONGQING ELECTRIC POWER +1
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] The structure of today's power system is becoming more and more complex, and it continues to expand and extend to many remote areas. However, due to the limitation of distance and natural conditions, the cost of power transmission in remote areas is high, and the number of connection lines with other areas is limited or there is no connection line. When a fault occurs, the local power system tends to become a single-area system with island operation, so the AGC control strategy to maintain the stable operation of the single-area power grid is more important
At the same time, new energy power generation often accounts for a large proportion of the power system in these regions. Due to the instability of the output power of wind turbines, photovoltaic or tidal generator sets, the frequency response of the power grid is prone to fluctuations. The total inertia of the unit is small, it is difficult to adjust the random fluctuations of the generator end and the load end, resulting in a large frequency deviation
On the other hand, the system adjustment actions caused by frequent frequency fluctuations also accelerate the aging of generator set components such as governors, increasing operation and maintenance costs
The AGC control method based on the optimal control theory achieves the control purpose by minimizing the defined cost function related to the frequency deviation and unit output. However, from the current research situation, the existing optimal control method requires a systematic The complete dynamic information of , the optimal control strategy is difficult to solve, and is easily affected by parameter changes and disturbances
The adaptive optimal control method proposed by some scholars can solve the optimal control strategy through online learning, but it faces the problem that the learning speed is slow and cannot converge to the optimum, and it still needs the dynamic information of the system

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Adaptive optimal AGC control method based on integral reinforcement learning
  • Adaptive optimal AGC control method based on integral reinforcement learning
  • Adaptive optimal AGC control method based on integral reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0081] see Figure 1 to Figure 3 , an adaptive optimal AGC control method based on integral reinforcement learning, including the following steps:

[0082] 1) Establish a single-area power system frequency response model, and calculate the power system state space matrix;

[0083] The components of the power system include governors, turbines, generator rotors and loads.

[0084] The frequency response model of a single-area power system is described as follows:

[0085]

[0086] In the formula, ΔX g (t) is the change increment of the governor valve opening; Increment ΔX g Differential of (t); ΔP g (t) is the output variation of the generator; Increment ΔP g (t) differential; Δf(t) is the frequency error increment; is the differential of increment Δf(t); ΔI(t) is the integral increment of frequency error; is the differential of increment ΔI(t); ΔP d (t) is the load increment; T g , T t , T p are the time constants of governor, turbine and generator respecti...

Embodiment 2

[0150] An adaptive optimal AGC control method based on integral reinforcement learning, comprising the following steps:

[0151] 1) Establish a power system frequency response model

[0152] The invention mainly studies the frequency control of a single-area power system, in which typical devices include governors, turbines, generator rotors and loads, and their dynamic models can be approximated as first-order processes. The system state variable selects the change increment of the valve opening of the governor ΔX g (t), generator output variation ΔP g (t), frequency error increment Δf(t) and frequency error integral increment ΔI(t), the disturbance variable is load increment ΔP d (t), the differential equation of the system is summarized as follows:

[0153]

[0154] The system state space model is expressed as:

[0155]

[0156]

[0157]

[0158] 2) Strategy iteration of integral reinforcement learning

[0159] In the optimal control problem, a cost functio...

Embodiment 3

[0206] see Figure 4 with Figure 5 , an adaptive optimal AGC control method based on integral reinforcement learning, including the following steps:

[0207] 1) System parameter setting

[0208] The control object is figure 1 For the single zone power system shown, the governor time constant T g =0.08, turbine time constant T t =0.1, generator time constant T p =20, generator gain K p =120, governor speed drop rate R d =2.5, integral control gain K e =1.

[0209] Define the optimal control cost function as formula (6), where the state variable weight Q=I of the utility equation U(x,u), the control variable weight R=0.5, and the activation function χ(x) in the evaluation network is selected to contain 10 vector of quadratic elements The system state variable is initialized as x(0)=[0 0 0 0] T , the initial value of the evaluation network weight is The adaptive gain matrix Γ=10I, the adaptive forgetting factor β=1.2, and the sampling period of the integrated enh...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a self-adaptive optimal AGC control method based on integral reinforcement learning. The method comprises the steps: 1), building a single-region power system frequency response model, and calculating a power system state space matrix; 2) establishing an evaluator-executor neural network based on a strategy iterative algorithm in reinforcement learning, wherein the evaluator-executor neural network comprises an evaluator network and an executor network; and 3) inputting the state space matrix of the power system into the evaluator-executor neural network, and performing resolving to obtain an optimal control strategy. According to the method, the optimal cost function is learned by using the integral reinforcement learning strategy, so that the learning process can be carried out under the condition that a system dynamic model is unknown, and the learning speed and accuracy are improved from the perspective of weakening a continuous excitation condition.

Description

technical field [0001] The invention relates to the field of electric power system and automation thereof, in particular to an adaptive optimal AGC control method based on integral reinforcement learning. Background technique [0002] The structure of today's power system is becoming more and more complex, and it continues to expand and extend to many remote areas. However, due to the limitation of distance and natural conditions, the cost of power transmission in remote areas is high, and the number of connection lines with other areas is limited or there is no connection line. When a fault occurs, the local power system tends to become a single-area system with island operation, so the AGC control strategy to maintain the stable operation of the single-area power grid is more important. At the same time, new energy power generation often accounts for a large proportion of the power system in these regions. Due to the instability of the output power of wind turbines, photov...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): H02J3/46H02J3/24
CPCH02J3/466H02J3/24H02J2203/20H02J2300/40
Inventor 许懿欧睿胡润滋蒙永苹张明媚杨渝璐周宇晴熊伟廖新颖李德智甘潼临刘伟许洁李光杰李郅浩
Owner STATE GRID CHONGQING ELECTRIC POWER
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com