CN106487011A - A kind of based on the family of Q study microgrid energy optimization method - Google Patents
A kind of based on the family of Q study microgrid energy optimization method Download PDFInfo
- Publication number
- CN106487011A CN106487011A CN201611067159.0A CN201611067159A CN106487011A CN 106487011 A CN106487011 A CN 106487011A CN 201611067159 A CN201611067159 A CN 201611067159A CN 106487011 A CN106487011 A CN 106487011A
- Authority
- CN
- China
- Prior art keywords
- air
- family
- energy
- power
- conditioning
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000004378 air conditioning Methods 0.000 claims abstract description 63
- 238000004146 energy storage Methods 0.000 claims abstract description 62
- 230000008859 change Effects 0.000 claims abstract description 17
- 238000012937 correction Methods 0.000 claims abstract description 9
- 230000002068 genetic effect Effects 0.000 claims abstract description 5
- 230000009471 action Effects 0.000 claims description 29
- 230000005611 electricity Effects 0.000 claims description 28
- 239000011159 matrix material Substances 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 9
- 230000003993 interaction Effects 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 5
- 238000010438 heat treatment Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 240000002853 Nelumbo nucifera Species 0.000 claims description 3
- 235000006508 Nelumbo nucifera Nutrition 0.000 claims description 3
- 235000006510 Nelumbo pentapetala Nutrition 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 abstract description 9
- 230000006978 adaptation Effects 0.000 abstract description 4
- 230000003044 adaptive effect Effects 0.000 abstract description 4
- 230000008569 process Effects 0.000 abstract description 4
- 238000007726 management method Methods 0.000 description 27
- 230000000694 effects Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000005406 washing Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000004659 sterilization and disinfection Methods 0.000 description 2
- 239000005955 Ferric phosphate Substances 0.000 description 1
- 244000131316 Panax pseudoginseng Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000002355 dual-layer Substances 0.000 description 1
- 239000000975 dye Substances 0.000 description 1
- 239000003792 electrolyte Substances 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 229940032958 ferric phosphate Drugs 0.000 description 1
- 239000005357 flat glass Substances 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- WBJZTOZJJYAKHQ-UHFFFAOYSA-K iron(3+) phosphate Chemical compound [Fe+3].[O-]P([O-])([O-])=O WBJZTOZJJYAKHQ-UHFFFAOYSA-K 0.000 description 1
- 229910000399 iron(III) phosphate Inorganic materials 0.000 description 1
- 229910052744 lithium Inorganic materials 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
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
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- 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)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Power Engineering (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses a kind of based on the family of Q study microgrid energy optimization method, with air conditioner load as research emphasis, in conjunction with other controllable burdens and energy storage device in the micro-capacitance sensor of family, with electric cost and users'comfort as target, invented a kind of based on the family of Q study microgrid energy optimization method.The method can be according to room thermal capacity, thermal resistance, the difference of air-conditioning model, the thermodynamical model of current scene is obtained in last stage day self adaptation based on genetic algorithm algorithm, and can be according to the dynamic change of current scene environment in application process, in the in a few days stage, the setting temperature of adaptive correction air-conditioning is learnt by Q, improve the accuracy of energy management further.Take into full account user intention, user can select corresponding energy management modes in conjunction with self-demand, realizes the personal management to family micro-capacitance sensor simultaneously.The method can improve accuracy and the practicality of family micro-capacitance sensor scheduling, and it is commonly used that promotion family microgrid energy manages.
Description
Technical field
The present invention relates to a kind of based on the family of Q study microgrid energy optimization method, belong to microgrid energy management neck
Domain.
Background technology
Distributed power generation has the advantages that investment is little, generation mode is flexible, loss is low, be beneficial to environmental protection, with peak period electric power
Load centrally connected power supply is compared, and it is more economical, effectively.And family micro-capacitance sensor is then to be integrated with distributed power source, energy storage device, bear
Units such as lotus and possess the autonomous networkses of independent control ability.With micro-capacitance sensor, energy management is carried out to family, can use meeting as far as possible
On the premise of the comfort level of family, give full play to the peak regulation potentiality of load, guide user's rational utilization of electricity, improve the economy of family micro-capacitance sensor
Property and operation of power networks efficiency.Meanwhile, the raising of the rapid growth with national economy and living standards of the people, air conditioner load exists
Ratio in customer charge is gradually increasing, in developed area ratio more than 40%.And air-conditioning is same as important load
When, also there is certain hot storage capacity, there are in terms of Load Regulation huge potentiality.Controlled by rational air conditioner load
Means, not only can alleviate the contradiction of peak period power supply and demand, can also be in the case of not affecting users'comfort as far as possible, fall
The operating cost of humble electrical network.How air-conditioning is accurately modeled, given full play to it as the demand response of temperature control load
Performance, improves the comfort level of user simultaneously, realizes the precise control to temperature, it has also become family microgrid energy management research
Emphasis.
The research being directed to air-conditioning modeling and optimization at present both at home and abroad is a lot, and existing main models have to be based on directly to be born
The scheduling of air conditioner load dual-layer optimization and Controlling model, the bilinearity PDE model of polymerization air conditioner load, base that lotus controls
In distributing air conditioner load equivalent heat parameter model etc..But existing research generally existing following point：Most models by
Load aggregation business passes through the Optimized Operation controlling the start and stop of air-conditioning to realize to colony's air-conditioning, but the method is not suitable for controlling
Air-conditioning arranges the single resident of temperature；Only with the air-conditioning setting research as target for the temperature generally approximate for air-conditioning setting temperature
It is equivalent to actual indoor temperature, have ignored the impact to room temperature for the environment dynamic change；Most researchs are all adopted to air-conditioning
With simplify hot equivalent parameterss model modeling, but lack in practical application to wherein room parameter and air-conditioning Energy Efficiency Ratio coefficient
The detailed description obtaining.And the research to air-conditioning modeling in residents micro-capacitance sensor is also less, how room conditioning is carried out accurately
Modeling, give full play to it as the demand response performance of temperature control load, improve the comfort level of user simultaneously, realize to temperature
Precise control, it has also become the family emphasis of microgrid energy management research.
Three above problem makes the energy management strategies at present air conditioner load formulated excessively idealize, if cannot solve
It can be made can not to be used widely.Therefore, invention is a kind of is a tool based on the family microgrid energy optimization method of Q study
There is the problem of great Research Significance.
Content of the invention
Goal of the invention：In order to overcome the deficiencies in the prior art, the present invention provides a kind of
Technical scheme：
A kind of based on the family of Q study microgrid energy optimization method, including step：
Step (1)：It is equipped with photovoltaic generating system and energy storage device in the micro-grid system of described family；Described family is used
In micro-capacitance sensor load be divided into translatable load, can reduction plans and uncontrollable load three class；
Step (2)：To described family indoor and outdoor temperature during micro-capacitance sensor current scene air-conditioning work and power historical data
Carry out real-time sampling, by genetic algorithm, historical data matching is obtained with the thermodynamical model being suitable for current building, simultaneously right
Historical data carries out off-line training and passes through pre- study acquisition initial Q matrix；
Step (3)：In the air conditioner heat mechanical model that obtained according to step (2), step (1) the energy storage model of energy storage device with
And the translatable load model of load last stage day by user select need energy management modes, with electric cost with comfortable
Spend for target, using power-balance constraint with interaction point Power Limitation as constraints, calculate the work of indoor temperature, energy storage
Instruct and translatable load optimal result；
Step (4)：Q matrix is according to the change of outdoor temperature and house internal staff, environment, ceaselessly on-line study reality
Shi Gengxin；Issue the work order of translatable load and energy storage according to step (3), the Indoor Temperature that air-conditioning obtains according to step (3)
Degree optimum results issue setting temperature, thus realizing family micro-capacitance sensor after the in a few days stage is according to the Q matrix correction of online updating
Energy-optimised.
The thermodynamical model that described step (2) obtains is specific as follows：
When air-conditioning affiliated building thermodynamical model equivalent heat parameter model is to freeze at present：
Wherein, TIn, tRepresent t indoor temperature, Tout,tRepresent t outdoor temperature, Δ t is time interval, C represents
The thermal capacity in room, R represents room thermal resistance, Qair,tRepresent the heating capacity of t air-conditioning, be represented by：
Qair,t=COPair,t·Pair,t
Wherein, COPair,tFor air-conditioning Energy Efficiency Ratio, i.e. quantitative relationship between air-conditioning heating amount and power, to fixed frequency air conditioner,
COPair,tFor fixed constant；To convertible frequency air-conditioner, COPair,tChange with frequency of air condition compressor change；
For fixed frequency air conditioner, obtaining object function is：
Wherein, TIn, tRepresent t indoor temperature, Tout,tRepresent t outdoor temperature, Δ t is time interval, C represents
The thermal capacity in room, R represents room thermal resistance, Pair,tRepresent air-conditioning power, historical data number is n；
For convertible frequency air-conditioner, obtaining object function is：
Described energy storage model comprises operating cost model and constrains two parts with discharge and recharge；
Operating cost model：Calculating the operating cost in the t period for the energy storage is：
Wherein：PcmaxAnd PdmaxBe respectively energy storage charge, electric discharge peak power, be on the occasion of；PBT () is t time period energy storage
Charge-discharge electric power, on the occasion of representing electric discharge, negative value represents charging；For energy-optimised, think within a dispatching cycle and set
Standby power is constant, and performance number takes its mean power within this cycle；
Discharge and recharge constrains：
Wherein, SOCmaxAnd SOCminIt is respectively energy-storage units state-of-charge upper limit value and lower limit value；ΔSOCmax(t) and Δ SOCmin
T () is respectively t time period energy-storage units state-of-charge variable quantity upper lower limit value；PcmaxAnd PdmaxIt is respectively energy storage to charge, discharge
High-power, be on the occasion of；PBT () is the charge-discharge electric power of t time period energy storage, on the occasion of representing electric discharge, negative value represents charging.
Described translatable load model is：
The real work power P of translatable load isliT () is：
Psli(t)=xsli(t)PNsli
Wherein, PNsliRepresent the rated power of translatable load i, xsliT () represents the working condition of translatable load i, its
It is worth and represents translatable load operation for 1, be the translatable load synthesis of 0 expression；
Translatable load needs meet the constraint condition：
Wherein, Tistart、Tifinish、TsliRepresent respectively translatable load i Earliest Starting Time, at the latest dwell time and
Continuous running time, this constraint representation translatable load operation duration satisfaction requires and works to interrupt.
Described optimization aim is specially：
Wherein, F represents system whole day electric cost；N is to divide a time hop count in；Fss (t) is that the purchase of t time period is sold
The electricity charge are used；α, β are respectively the weight coefficient of electric cost and users'comfort；
Described purchase sale of electricity expense carries out expense or the income producing when power interacts for micro-capacitance sensor and higher level's electrical network：
FSS(t)=c (t) Pcc(t)Δt
Wherein, PCCT () is t time period dominant eigenvalues, on the occasion of representing from electrical network power purchase, negative value represents to electrical network sale of electricity；
Δ t is the duration of a dispatching cycle；C (t), sell_price (t), buy_price (t) be respectively the t time period purchase sale of electricity valency,
Sale of electricity price, power purchase price.
Described user selects the energy management modes needing to include electric cost Optimizing Mode, users'comfort Optimizing Mode
With complex optimum pattern；Corresponding energy management modes are selected according to self-demand by user, determines that in optimization aim, electricity consumption becomes
This proportion with users'comfort, show that the family micro-capacitance sensor adapting to different user demands is planned a few days ago.
Described Q learning algorithm is specially：
Assume that state set and behavior aggregate are respectively divided into M and N number of discrete segment, then by the evaluation of each state action pair
(s, a) is established as the Q matrix of a M*N rank to value Q, and its formula is as follows：
In formula, α is learning rate, and a' is executable everything under state s；(s, value a) is from the execution of state s to Q
The accumulative return value obtaining after action a；
In each moment t, the maximum action a of corresponding Q-value is selected according to ambient condition s, and observes instantaneous award r and new shape
State s ', and update Q-value, its primitive form：
In formula, s is current state, and s' is subsequent time ambient condition, Q*(s a) represents that execution action a obtains under state s
The return summation obtaining, P (s, a, s') is transformed into the probability of s ' for state after execution action a from s, and (s, s' are a) from the choosing of s state to R
It is transformed into the award that s ' obtains afterwards, γ is discount factor, S is ambient condition collection, A is controller action collection after selecting action a；
Using indoor temperature and target temperature TgoalDeviation delta T as Q study input ambient condition variable, wherein
TgoalDrawn according to energy management optimization a few days ago；Indoor temperature deviation delta T is divided into series of discrete interval { Δ T1,Δ
T2,…ΔTm, corresponding ambient condition collection；The control targe of indoor temperature is set to Tgoal± 0.5 DEG C, state set Δ T is set
It is set to：(- ∞, -3], (and -3, -2], (- 2, -1], (- 1,0.5], (- 0.5,0], (0,0.5], (0.5,1], (1,2], (2,3],
(3,+∞)}；
Reward function is defined as：
When indoor temperature deviation during 0.5 DEG C of | Δ T | ＞, according to different deviation sizes, study will obtain different degrees of punishing
Penalize, deviation is bigger, the punishment being subject to is bigger, then the Q-value obtaining after iteration is less, hereafter selects the probability of this action to get over
Little.
Beneficial effect：The present invention compared with prior art, has advantages below：
The present invention is directed to the problem that air-conditioning Energy Efficiency Ratio coefficient, room preset parameter are difficult to obtain, by historical data something lost
Propagation algorithm matching obtains the room thermal resistance of current scene, thermal capacity, the parameter such as air-conditioning Energy Efficiency Ratio coefficient, sets up more accurately empty
Adjust thermodynamical model, improve the accuracy to air-conditioning power prediction.
The present invention is directed to the problem of the dynamic change such as indoor occupant activity, other cold and hot type loading effects it is considered to environment becomes
Change the impact to air-conditioning work, in the in a few days stage, the setting temperature of on-line amending air-conditioning is learnt by Q, be adapted to application process
The dynamic change of middle current scene environment, improves the accuracy of energy management further.
The present invention considers the different demand of user in the micro-capacitance sensor of family, selects energy management modes by user, determines and optimizes
In target, electric cost and the proportion of users'comfort are it can be deduced that the family micro-capacitance sensor adapting to different user demands is counted a few days ago
Draw, realize the personal management to family micro-capacitance sensor.
The present invention is carried strategy and can be realized the accurate regulation to air-conditioner temperature, and it is accurate that raising family micro-capacitance sensor is dispatched
Property, it is adapted to the demand of different application scenarios and different user, it is universal that promotion family microgrid energy manages simultaneously
Application.
Brief description
Fig. 1 is family micro-capacitance sensor structure chart；
Fig. 2 is family microgrid energy management strategy figure；
Fig. 3 is the family microgrid energy Optimizing Flow figure based on Q study.
Specific embodiment
Below in conjunction with the accompanying drawings the present invention is further described.
(1) set up each unit model in the micro-capacitance sensor of family
Typical family micro-capacitance sensor is made up of photovoltaic cell, energy-storage system, all kinds of household loads, and electric pressure is single phase ac
220V, its structural representation is as shown in Figure 1.
1) photovoltaic generating system
User is equipped with low profile photovoltaic battery and can make full use of solar energy resources, changes user unidirectional passive from electrical network power purchase
Power mode, and by unnecessary back electrical energy to electrical network.With the development of new forms of energy, all kinds of low profile photovoltaic batteries have started to occupying
It is applied in the people, the dye-sensitized cell being such as made up of transparent conducting glass and dyestuff and electrolyte, it is used as window glass
Glass, not only printing opacity but also can work as battery use.
2) family load
In the micro-grid system of family, load is mainly household loads, such as air-conditioning, washing machine, electromagnetic oven etc..User is using
There is certain custom during each type load electrical equipment, the degree according to user and system operation interaction wish and load electrical characteristics, will
Household loads be divided into translatable load, can reduction plans (air-conditioning) and uncontrollable load three class.
Uncontrollable load refers to：Electricity consumption is characterized as rigidity, is set by the user the work plan of load, is not involved in the mutual of system
Dynamic load；As illumination, electromagnetic oven etc..Translatable load refers to：Daily power consumption is certain, and electricity consumption curve can translate in one day,
But not interruptable load；As washing machine, disinfection cabinet etc..
3) small-sized energy-storage system
Energy-storage units are to ensureing the power supply continuity of family micro-capacitance sensor, stabilize intermittent power supply and go out fluctuation and bent to load
Line peak load shifting has vital effect.It is equipped with energy storage device for family micro-capacitance sensor, system can be made to pass through storage when grid-connected
Energy peak load shifting, improves the economy of system；Power for load in isolated island, maintain system stable operation.Family micro-capacitance sensor can
From life-span length, free of contamination ferric phosphate lithium cell.
(2) with historical datas such as temperature during micro-capacitance sensor current scene air-conditioning work, power, real-time sampling is carried out to family, from
And matching obtains being suitable for the thermodynamical model of current building, obtain initial Q matrix simultaneously.
(3) according to air conditioner heat mechanical model, energy storage model and translatable load model, selected by user in last stage day
The energy management modes needing, with electric cost and comfort level as target, be given translatable load, the work order of energy storage and
Indoor temperature optimum results.
(4) directly issue the work order of translatable load, energy storage, air-conditioning setting temperature then in the in a few days stage according to online
Issue, thus realizing the energy-optimised of family micro-capacitance sensor after the Q matrix correction updating.
Microgrid energy optimization mainly has Multiple Time Scales, the rolling optimization based on Model Predictive Control, robust optimization etc.
Several primitive forms, compared with conventional micro-capacitance sensor, family micro-grid system small scale, energy management strategies should not be excessively complicated.
But family microgrid energy management region be directly facing resident, need to take into full account the wish that user accepts energy management, to
Family electricity consumption behavior carries out the analysis that becomes more meticulous, and with exempt from customs examination, user brings discomfort.
Therefore the present invention sets up, to each type load in the micro-capacitance sensor of family and energy storage, the model that becomes more meticulous, with electric cost and user
Comfort level is target, using by the way of plan combines indoor correction a few days ago it is proposed that count and air-conditioning model adaptation correction family
With microgrid energy optimisation strategy it is contemplated that the accuracy of energy management and the inadaptability that may bring to user, optimize
Minimum time yardstick elect 15min as, general frame is as shown in Fig. 2 obtain model base according to system mode and information of forecasting
This parameter, sets up energy storage model, purchase sale of electricity model and all kinds of load model, wherein air conditioner load as key object, by ginseng
The mode of number matching sets up more accurate thermodynamical model.With minimum electric cost as target, provide storage through optimizing a few days ago
The optimization temperature of energy discharge and recharge plan, translatable load electricity consumption plan and air-conditioning.Energy storage discharge and recharge plan, translatable load electricity consumption
Plan directly issue, air-conditioning design temperature according to environment dynamic change by air-conditioning adaptive model carry out on-line amending again under
Send out.Fig. 3 is the family microgrid energy Optimizing Flow figure of meter and air-conditioning model adaptation, comprises the following steps that：
(1) with historical datas such as temperature during micro-capacitance sensor current scene air-conditioning work, power, real-time sampling is carried out to family, from
And matching obtains being suitable for the thermodynamical model of current building, obtain initial Q matrix simultaneously.
Historical data is to control the data that (MGCC) software collection PLC, PCS records to obtain by micro-capacitance sensor central authorities, data
To gather in real time and be shown in interface, regularly be stored in historical data base.Historical data includes photovoltaic, load (indoor and outdoor temperature
Degrees of data, power data, on off state etc.), energy storage state etc..Indoor and outdoor temperature data and work(are only used in step (2)
Rate data.
The learning target of Q study is that more excellent action or optimum are chosen in study under dynamic environment according to external evaluation signal
Action, is substantially the process of a dynamic decision.Its thought be learn each state action pair evaluation of estimate Q (s, a),
(s, value a) is the accumulative return value obtaining after state s execution action a to Q, in each moment t, is selected according to ambient condition s
The maximum action a of corresponding Q-value, and observe instantaneous award r and new state s ', and update Q-value, shown in its primitive form such as formula (1)：
In formula, s is current state, and s' is subsequent time ambient condition, Q*(s a) represents that execution action a obtains under state s
The return summation obtaining, P (s, a, s') is transformed into the probability of s ' for state after execution action a from s, and (s, s' are a) from the choosing of s state to R
It is transformed into the award that s ' obtains afterwards, γ is discount factor, S is ambient condition collection, A is controller action collection after selecting action a.
Assume that state set and behavior aggregate are respectively divided into M and N number of discrete segment, then can (s a) be established as a M*N by Q
The Q matrix of rank, Q study is realized by iterative approach optimal solution according to after learning strategy selection action, and its formula is as follows：
In formula, α is learning rate, and a' is executable everything under state s, and as a rule Q study only needs to choose
Greedy strategy can ensure convergence, select the action executing of highest q value every time.
In order to meet under users'comfort demand condition, the operating power of adjustment air-conditioning makes air-conditioning participate in whole system
Optimized Operation, need to set up the thermodynamical model of the affiliated building of air-conditioning.This model include building equivalent specific heat hold, equivalent
The parameters such as thermal resistance, air-conditioning Energy Efficiency Ratio, and these parameters hardly result in accurate measurement in reality and calculate, and have a strong impact on
The accuracy of model.The present invention by current building under operation of air conditioner operating mode indoor temperature, outdoor temperature, air-conditioning work(
The data of rate measures and analyzes, and uses for reference the form of conventional thermal mechanical model, with genetic algorithm, historical data is fitted,
It is suitable for the thermodynamical model of current building after drawing corrected parameter.
At present commonly use air-conditioning affiliated building thermodynamical model equivalent heat parameter model, when freezing as a example, as formula (1) institute
Show：
Wherein, TIn, tRepresent t indoor temperature, Tout,tRepresent t outdoor temperature, Δ t is time interval, C represents
The thermal capacity in room, R represents room thermal resistance, Qair,tRepresent t air-conditioning heats (cold) amount, is represented by：
Qair,t=COPair,t·Pair,t(2)
Wherein, COPair,tFor air-conditioning Energy Efficiency Ratio, that is, the quantitative relationship between (cold) amount of air-conditioning heating and power, empty to determining frequency
Adjust, COPair,tFor fixed constant；To convertible frequency air-conditioner, COPair,tChange with frequency of air condition compressor change.
1) fixed frequency air conditioner
Fixed frequency air conditioner Energy Efficiency Ratio is constant, i.e. COPair,t=COP, only need to be fitted to wherein R, C, COP, if history number
It is n according to number, the object function of matching is：
Wherein, TIn, tRepresent t indoor temperature, Tout,tRepresent t outdoor temperature, Δ t is time interval, C represents
The thermal capacity in room, R represents room thermal resistance, Pair,tRepresent air-conditioning power.
2) convertible frequency air-conditioner
Convertible frequency air-conditioner power Pair,t, compressor frequency fair,t, refrigerating capacity Qair,tWith Energy Efficiency Ratio COPair,tCan be approximately as follows
Relation：
Wherein, n, m, a, b, c are constant coefficient, substitute into formula (2) and eliminate variable f, can be derived from COPair,tWith Pair,tRelational expression
As follows：
Using genetic algorithm to R, C, k1、k2、k3It is fitted, object function is：
(2) energy storage and other load models are set up
1) energy storage model
Energy storage model comprises operating cost model and constrains two parts with discharge and recharge.The operating cost of energy storage and the behaviour of energy storage
Work, maintenance cost are relevant, also contains amortization charge simultaneously, convert wherein the acquisition expenses of energy storage.Energy storage is in t
Section operating cost be：
Wherein：PcmaxAnd PdmaxBe respectively energy storage charge, electric discharge peak power, be on the occasion of；PBT () is t time period energy storage
Charge-discharge electric power, on the occasion of representing electric discharge, negative value represents charging.For energy-optimised, think within a dispatching cycle and set
Standby power is constant, and performance number takes its mean power within this cycle.
Energy storage discharge and recharge also includes state-of-charge SOC constraint except the constraint of peak power.SOC is energy storage charge and discharge protecting
An important decision variable, reflect the ratio that energy storage residual capacity accounts for total capacity, the SOC of current time and previous moment
The discharge and recharge of residual capacity and previous period is relevant, and computing formula is as follows：
Wherein, ηc、ηdIt is respectively efficiency for charge-discharge；Q is rated capacity.By carrying out to variation delta SOC of energy storage SOC
Dynamic constrained, it is to avoid super-charge super-discharge causes damage to battery life.
Wherein, SOCmaxAnd SOCminIt is respectively energy-storage units state-of-charge upper limit value and lower limit value；ΔSOCmax(t) and Δ SOCmin
T () is respectively t time period energy-storage units state-of-charge variable quantity upper lower limit value.
To sum up, energy storage charging and recharging model comprises following 3 constraintss：
2) other load models
In the micro-capacitance sensor of family, except can also there is uncontrollable load and translatable load in addition to reduction plans.Uncontrollable load
Refer to：Electricity consumption is characterized as rigidity, is set by the user the work plan of load, is not involved in the load of the interaction of system；As illumination, electricity
Magnetic stove etc..Translatable load refers to：Daily power consumption is certain, and electricity consumption curve can translate in one day, but not interruptable load；As
Washing machine, disinfection cabinet etc..The model of translatable load is as follows：
Use xsliT () represents the working condition of translatable load i, its value is the 1 translatable load operation of expression, is that 0 expression can
Translation load synthesis.
The real work power P of translatable load isliT () is：
Psli(t)=xsli(t)PNsli(11)
Wherein, PNsliRepresent the rated power of translatable load i.
Translatable load needs meet the constraint condition：
Wherein, Tistart、Tifinish、TsliRepresent respectively translatable load i Earliest Starting Time, at the latest dwell time and
Continuous running time, this constraint representation translatable load operation duration satisfaction requires and works to interrupt.
(3) according to air conditioner heat mechanical model, energy storage model and translatable load model, in last stage day with electric cost
It is target with comfort level, provide translatable load, the work order of energy storage and indoor temperature optimum results.
1) optimization aim
Optimized Operation target considers electric cost and users'comfort simultaneously.For the day operation of micro-grid system, clearly
Clean energy energy consumption cost and operation management cost can almost see zero as, thus micro-grid system day electric cost includes and higher level
The interactive expense of electrical network and the operating cost of energy storage.Users'comfort is mainly reflected in the inclined of indoor temperature and user's ideal temperature
Difference, therefore optimization aim is represented by：
Wherein, F represents system whole day electric cost；N is to divide a time hop count in；Fss (t) is that the purchase of t time period is sold
The electricity charge are used；α, β are respectively the weight coefficient of electric cost and users'comfort.Three kinds of energy management modes are set herein, respectively
For electric cost optimization (Mode A), users'comfort optimization (Mode B) and complex optimum (pattern C), by user according to itself need
Seek the corresponding energy management modes of selection, determine electric cost and the proportion of users'comfort in optimization aim, draw adaptation not
Family micro-capacitance sensor with user's request is planned a few days ago, and the value of corresponding α, β can be as shown in table 1.
Table 1 energy management modes
Purchase sale of electricity expense is mainly expense or the income that micro-capacitance sensor and higher level's electrical network carry out producing when power interacts：
FSS(t)=c (t) Pcc(t)Δt (14)
Wherein, PCCT () is t time period dominant eigenvalues, on the occasion of representing from electrical network power purchase, negative value represents to electrical network sale of electricity；
Δ t is the duration of a dispatching cycle；C (t), sell_price (t), buy_price (t) be respectively the t time period purchase sale of electricity valency,
Sale of electricity price, power purchase price.
2) constraints
A. power-balance constraint
It is necessary that each time period energy storage charge-discharge electric power interacts power, photovoltaic generation power, total load power with higher level's electrical network
Meet power-balance constraint：
PB(t)+PCC(t)+PPV(t)=Pload(t) (16)
Wherein, PPVT () represents t time period photovoltaic generation power；PloadT () represents t time period total load.
B. interaction point Power Limitation
Each time period is interacted power and should meet the restriction to interaction power for higher level's electrical network with electrical network：
-Psmax≤PCC≤Pbmax(17)
Wherein, PsmaxFor user's superior electrical network sale of electricity power upper limit；PbmaxFor user from higher level's electrical network power purchase power
Limit.
(4) directly issue the work order of translatable load, energy storage, air-conditioning setting temperature then in the in a few days stage according to online
Issue, thus realizing the energy-optimised of family micro-capacitance sensor after the Q matrix correction updating.
Air-conditioning design temperature is generally directly disposed as required for user or energy management system to the optimization of air-conditioning by tradition
System optimizes the indoor temperature that obtains, and is actually affected due to being changed by indoor environment, actual indoor temperature often with air-conditioning
Design temperature has certain error.For solving this problem, become for indoor occupant activity, other cold and hot type loading effects etc. are dynamic
The problem changed, the present invention is based on Q study and sets up air-conditioner temperature adaptive model, according to actual indoor temperature and optimization ideal room
The deviation of interior temperature, goes out to need the temperature of setting during air-conditioning real time execution in building, improves using Q learning algorithm dynamic calculation
The accuracy of energy management.Concrete grammar is：
Using indoor temperature and target temperature TgoalDeviation delta T as Q study input ambient condition variable, wherein
TgoalDrawn according to energy management optimization a few days ago.Indoor temperature deviation delta T is divided into series of discrete interval { Δ T1,Δ
T2,…ΔTm, corresponding ambient condition collection.The control targe of indoor temperature is set to Tgoal± 0.5 DEG C, state set Δ T is set
It is set to：(- ∞, -3], (and -3, -2], (- 2, -1], (- 1,0.5], (- 0.5,0], (0,0.5], (0.5,1], (1,2], (2,3],
(3,+∞)}.
Reward function is defined as：
When indoor temperature deviation during 0.5 DEG C of | Δ T | ＞, according to different deviation sizes, study will obtain different degrees of punishing
Penalize, deviation is bigger, the punishment being subject to is bigger, then the Q-value obtaining after iteration is less, hereafter selects the probability of this action to get over
Little.From indoor temperature deviation as state variable, its benefit is in equivalent environment, when target temperature changes it is only necessary to
Change Tgoal, the state set being pre-designed need not be adjusted.
Air-conditioning is arranged temperature TsetAs the output action variable of controller, TsetlFor air-conditioning, lowest temperature, T are setseth
Temperature upper limit is set for air-conditioning, control action collection A is { Tsetl,Tsetl+1,...,Tseth-1,Tseth}.
By the off-line training of historical data, the Q matrix after pre- study can be obtained.As family micro-capacitance sensor air conditioning system
The Q matrix of initial launch, in micro-capacitance sensor actual moving process, due to the change of outdoor temperature and house internal staff, environment,
Needs carry out ceaselessly on-line study, real-time update Q matrix, do not stop to revise the temperature of air-conditioning setting, to realize air-conditioning setting temperature
The adaptive correction to environmental change for the degree.
The present invention is adapted to different application scene and the change of environment, improves the accuracy of energy management and reality should
The property used.And consider the power load that in the micro-capacitance sensor of family, other have regulation and control potentiality and energy storage device, with electric cost
It is target with users'comfort, realize the effective management to family micro-capacitance sensor.
Using plan a few days ago with a few days revise and combine by the way of.Because energy storage, translatable load etc. are subject to environmental change shadow
Ring less, Planning Directive before the direct execution day, and air conditioner load is subject to environment Dynamic Effect larger, in a few days through ceaselessly
Send instructions under correction.
The above be only the preferred embodiment of the present invention it should be pointed out that：Ordinary skill people for the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (7)
1. a kind of family based on Q study with microgrid energy optimization method it is characterised in that：Including step：
Step (1)：It is equipped with photovoltaic generating system and energy storage device in the micro-grid system of described family；By the micro- electricity in described family
In net load be divided into translatable load, can reduction plans and uncontrollable load three class；
Step (2)：Described family is carried out with indoor and outdoor temperature during micro-capacitance sensor current scene air-conditioning work and power historical data
Real-time sampling, obtains the thermodynamical model being suitable for current building, simultaneously to history by genetic algorithm to historical data matching
Data carries out off-line training and passes through pre- study acquisition initial Q matrix；
Step (3)：The energy storage model of energy storage device and negative in the air conditioner heat mechanical model that obtained according to step (2), step (1)
The translatable load model of lotus is selected the energy management modes needing in last stage day by user, with electric cost with comfort level is
Target, using power-balance constraint with interaction point Power Limitation as constraints, calculates indoor temperature, the work of energy storage refers to
Order and translatable load optimal result；
Step (4)：Q matrix according to the change of outdoor temperature and house internal staff, environment, ceaselessly on-line study in real time more
Newly；Issue the work order of translatable load and energy storage according to step (3), the indoor temperature that air-conditioning obtains according to step (3) is excellent
Change result and issue setting temperature after the in a few days stage is according to the Q matrix correction of online updating, thus realizing the energy of family micro-capacitance sensor
Amount optimizes.
2. family according to claim 1 with microgrid energy optimization method it is characterised in that：Described step (2) obtains
Thermodynamical model is specific as follows：
When air-conditioning affiliated building thermodynamical model equivalent heat parameter model is to freeze at present：
Wherein, TIn, tRepresent t indoor temperature, Tout,tRepresent t outdoor temperature, Δ t is time interval, C represents room
Thermal capacity, R represents room thermal resistance, Qair,tRepresent the heating capacity of t air-conditioning, be represented by：
Qair,t=COPair,t·Pair,t
Wherein, COPair,tFor air-conditioning Energy Efficiency Ratio, i.e. quantitative relationship between air-conditioning heating amount and power, to fixed frequency air conditioner,
COPair,tFor fixed constant；To convertible frequency air-conditioner, COPair,tChange with frequency of air condition compressor change；
For fixed frequency air conditioner, obtaining object function is：
Wherein, TIn, tRepresent t indoor temperature, Tout,tRepresent t outdoor temperature, Δ t is time interval, C represents room
Thermal capacity, R represents room thermal resistance, Pair,tRepresent air-conditioning power, historical data number is n；
For convertible frequency air-conditioner, obtaining object function is：
3. family according to claim 1 with microgrid energy optimization method it is characterised in that：Described energy storage model comprises to transport
Row cost model constrains two parts with discharge and recharge；
Operating cost model：Calculating the operating cost in the t period for the energy storage is：
Wherein：PcmaxAnd PdmaxBe respectively energy storage charge, electric discharge peak power, be on the occasion of；PBT () is filling of t time period energy storage
Discharge power, on the occasion of representing electric discharge, negative value represents charging；For energy-optimised, within a dispatching cycle, think equipment
Power is constant, and performance number takes its mean power within this cycle；
Discharge and recharge constrains：
Wherein, SOCmaxAnd SOCminIt is respectively energy-storage units state-of-charge upper limit value and lower limit value；ΔSOCmax(t) and Δ SOCmin(t) point
Wei not t time period energy-storage units state-of-charge variable quantity upper lower limit value；PcmaxAnd PdmaxIt is respectively energy storage charging, electric discharge maximum work
Rate, be on the occasion of；PBT () is the charge-discharge electric power of t time period energy storage, on the occasion of representing electric discharge, negative value represents charging.
4. family according to claim 1 with microgrid energy optimization method it is characterised in that：Described translatable load model
For：
The real work power P of translatable load isliT () is：
Psli(t)=xsli(t)PNsli
Wherein, PNsliRepresent the rated power of translatable load i, xsliT () represents the working condition of translatable load i, its value is 1
Represent translatable load operation, be the translatable load synthesis of 0 expression；
Translatable load needs meet the constraint condition：
Wherein, Tistart、Tifinish、TsliRepresent Earliest Starting Time, the dwell time and continuous at the latest of translatable load i respectively
Run duration, this constraint representation translatable load operation duration satisfaction requires and work to interrupt.
5. family according to claim 1 with microgrid energy optimization method it is characterised in that：Described optimization aim is concrete
For：
Wherein, F represents system whole day electric cost；N is to divide a time hop count in；Fss (t) is the purchase sale of electricity expense of t time period
With；α, β are respectively the weight coefficient of electric cost and users'comfort；
Described purchase sale of electricity expense carries out expense or the income producing when power interacts for micro-capacitance sensor and higher level's electrical network：
FSS(t)=c (t) Pcc(t)Δt
Wherein, PCCT () is t time period dominant eigenvalues, on the occasion of representing from electrical network power purchase, negative value represents to electrical network sale of electricity；Δ t is
The duration of one dispatching cycle；C (t), sell_price (t), buy_price (t) the respectively t time period purchases sale of electricity valency, sale of electricity
Price, power purchase price.
6. family according to claim 1 with microgrid energy optimization method it is characterised in that：Described user selects needs
Energy management modes include electric cost Optimizing Mode, users'comfort Optimizing Mode and complex optimum pattern；By user according to
Self-demand selects corresponding energy management modes, determines electric cost and the proportion of users'comfort in optimization aim, draws
The family micro-capacitance sensor adapting to different user demands is planned a few days ago.
7. family according to claim 1 with microgrid energy optimization method it is characterised in that：Described Q learning algorithm is concrete
For：
Assume that state set and behavior aggregate are respectively divided into M and N number of discrete segment, then by evaluation of estimate Q of each state action pair
(s, a) is established as the Q matrix of a M*N rank, and its formula is as follows：
In formula, α is learning rate, and a' is executable everything under state s；(s, value a) is from state s execution action a to Q
The accumulative return value obtaining afterwards；
In each moment t, the maximum action a of corresponding Q-value is selected according to ambient condition s, and observes instantaneous award r and new state
S ', and update Q-value, its primitive form：
In formula, s is current state, and s' is subsequent time ambient condition, Q*(s a) represents execution action a acquisition under state s
Return summation, P (s, a, s') is transformed into the probability of s ' for state after execution action a from s, and (s, s' are a) to move from s condition selecting to R
It is transformed into the award that s ' obtains afterwards, γ is discount factor, S is ambient condition collection, A is controller action collection after making a；
Using indoor temperature and target temperature TgoalDeviation delta T as Q study input ambient condition variable, wherein TgoalRoot
Draw according to energy management optimization a few days ago；Indoor temperature deviation delta T is divided into series of discrete interval { Δ T1,ΔT2,…Δ
Tm, corresponding ambient condition collection；The control targe of indoor temperature is set to Tgoal± 0.5 DEG C, state set Δ T is set as：{(-
∞, -3], (- 3, -2], (- 2, -1], (- 1,0.5], (- 0.5,0], (0,0.5], (0.5,1], (1,2], (2,3], (3 ,+
∞)}；
Reward function is defined as：
When indoor temperature deviation during 0.5 DEG C of | Δ T | ＞, according to different deviation sizes, study will obtain different degrees of punishment, partially
Difference is bigger, and the punishment being subject to is bigger, then the Q-value obtaining after iteration is less, hereafter selects the probability of this action less.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611067159.0A CN106487011B (en) | 2016-11-28 | 2016-11-28 | A kind of family microgrid energy optimization method based on Q study |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611067159.0A CN106487011B (en) | 2016-11-28 | 2016-11-28 | A kind of family microgrid energy optimization method based on Q study |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106487011A true CN106487011A (en) | 2017-03-08 |
| CN106487011B CN106487011B (en) | 2019-06-25 |
Family
ID=58274545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611067159.0A Active CN106487011B (en) | 2016-11-28 | 2016-11-28 | A kind of family microgrid energy optimization method based on Q study |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106487011B (en) |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392465A (en) * | 2017-07-19 | 2017-11-24 | 北京上格云技术有限公司 | Build the operation management method and server of electromechanical equipment |
| CN107545326A (en) * | 2017-08-22 | 2018-01-05 | 国网能源研究院 | A kind of household energy scheduling method based on information physical emerging system |
| CN107785888A (en) * | 2017-10-13 | 2018-03-09 | 陆炜 | A kind of intelligent electric power load control system and control method |
| CN108092804A (en) * | 2017-12-08 | 2018-05-29 | 国网安徽省电力有限公司信息通信分公司 | Power telecom network maximization of utility resource allocation policy generation method based on Q-learning |
| CN108321793A (en) * | 2018-01-17 | 2018-07-24 | 东北电力大学 | The active distribution network of integrated intelligent building flexible load models and Optimization Scheduling |
| CN108490791A (en) * | 2018-05-10 | 2018-09-04 | 燕山大学 | Temperature control load Cost Controlling Policy |
| CN108629470A (en) * | 2017-03-17 | 2018-10-09 | 华北电力大学 | System capacity management of providing multiple forms of energy to complement each other based on non-cooperative game is run with optimization |
| CN109035812A (en) * | 2018-09-05 | 2018-12-18 | 平安科技（深圳）有限公司 | Control method, device, computer equipment and the storage medium of traffic lights |
| CN109066739A (en) * | 2018-07-26 | 2018-12-21 | 西南交通大学 | A kind of tractive power supply system regenerating braking energy energy-accumulating medium power and capacity collocation method |
| CN109063925A (en) * | 2018-08-16 | 2018-12-21 | 合肥工业大学 | It is a kind of meter and Load aggregation quotient regional complex energy resource system optimizing operation method |
| CN109103936A (en) * | 2018-09-30 | 2018-12-28 | 南京铭越创信电气有限公司 | Optimal unit starting order calculation method after a kind of electric system is had a power failure on a large scale |
| CN109190988A (en) * | 2018-09-11 | 2019-01-11 | 浙江大学 | A kind of Demand Side Response game method for realizing the optimal collaboration of temperature control load |
| CN109347149A (en) * | 2018-09-20 | 2019-02-15 | 国网河南省电力公司电力科学研究院 | Micro-capacitance sensor energy storage dispatching method and device based on depth Q value network intensified learning |
| CN109685396A (en) * | 2019-01-31 | 2019-04-26 | 河海大学 | It is a kind of meter and public building demand response resource power distribution network energy management method |
| CN109812927A (en) * | 2019-01-24 | 2019-05-28 | 新奥数能科技有限公司 | Family microgrid energy optimization method, device, readable medium and electronic equipment |
| WO2019165702A1 (en) * | 2018-02-28 | 2019-09-06 | 东南大学 | Double-layer coordinated robust optimized scheduling method for multi-microgrids system |
| CN110212533A (en) * | 2019-07-10 | 2019-09-06 | 南方电网科学研究院有限责任公司 | Method and system for determining power of either person from birth or death |
| CN110414725A (en) * | 2019-07-11 | 2019-11-05 | 山东大学 | The integrated wind power plant energy-storage system dispatching method of forecast and decision and device |
| CN110958680A (en) * | 2019-12-09 | 2020-04-03 | 长江师范学院 | Energy efficiency-oriented unmanned aerial vehicle cluster multi-agent deep reinforcement learning optimization method |
| CN111324167A (en) * | 2020-02-27 | 2020-06-23 | 上海电力大学 | Photovoltaic power generation maximum power point tracking control method and device |
| CN111404146A (en) * | 2020-03-19 | 2020-07-10 | 南方电网科学研究院有限责任公司 | Power distribution method, system and terminal based on user load transfer comfort level |
| CN111431216A (en) * | 2020-03-18 | 2020-07-17 | 国网浙江嘉善县供电有限公司 | High-proportion photovoltaic microgrid reactive power sharing control method adopting Q learning |
| WO2020199648A1 (en) * | 2019-04-01 | 2020-10-08 | 珠海格力电器股份有限公司 | Control method and device for air conditioner |
| CN112117760A (en) * | 2020-08-13 | 2020-12-22 | 国网浙江省电力有限公司台州供电公司 | Micro-grid energy scheduling method based on double-Q-value network deep reinforcement learning |
| CN112465212A (en) * | 2020-11-24 | 2021-03-09 | 国网河北省电力有限公司经济技术研究院 | Building cluster demand side energy management method and system |
| CN112594872A (en) * | 2020-12-15 | 2021-04-02 | 深圳供电局有限公司 | Multi-agent consistency control method considering air conditioner response cost |
| CN112671033A (en) * | 2020-12-14 | 2021-04-16 | 广西电网有限责任公司电力科学研究院 | Priority-level-considered microgrid active power scheduling control method and system |
| CN113011101A (en) * | 2021-03-29 | 2021-06-22 | 广东电网有限责任公司电力调度控制中心 | Control method and system for energy storage participating in frequency modulation auxiliary service optimization |
| CN113219930A (en) * | 2021-05-21 | 2021-08-06 | 上海交通大学 | Variable frequency air conditioner second-order equivalent thermal parameter model online identification method based on particle swarm optimization |
| CN113381635A (en) * | 2021-06-29 | 2021-09-10 | 中南大学 | Traction converter thermal performance dynamic optimization control method and system |
| CN113420413A (en) * | 2021-05-27 | 2021-09-21 | 国网上海市电力公司电力科学研究院 | Flexible load adjustability quantification method and system based on load plasticity |
| CN113609102A (en) * | 2021-08-11 | 2021-11-05 | 佛山仙湖实验室 | Construction method of energy management database of hybrid drive mining truck |
| CN113822493A (en) * | 2021-10-13 | 2021-12-21 | 国网天津市电力公司电力科学研究院 | Commercial building energy management method based on demand response |
| CN113890057A (en) * | 2021-09-18 | 2022-01-04 | 国网河北省电力有限公司经济技术研究院 | Control method and device based on multi-microgrid collaborative optimization and storage medium |
| CN115268536A (en) * | 2022-08-02 | 2022-11-01 | 阳光电源股份有限公司 | Temperature control method of energy storage system and related device |
| CN116628413A (en) * | 2023-07-24 | 2023-08-22 | 国网山西电力勘测设计研究院有限公司 | Method for calculating capacity of user side energy storage device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104699051A (en) * | 2015-02-12 | 2015-06-10 | 天津大学 | Demand response control method of temperature control device |
| CN104952001A (en) * | 2015-07-02 | 2015-09-30 | 华侨大学 | Method for performing power optimized scheduling on controllable loads comprising air conditioning loads |
| CN105322550A (en) * | 2015-08-28 | 2016-02-10 | 南方电网科学研究院有限责任公司 | Method for optimizing operation of household micro-grid |
| CN106094521A (en) * | 2016-06-30 | 2016-11-09 | 中国南方电网有限责任公司电网技术研究中心 | Flexible load energy efficiency power plant scheduling control method and system |
| CN106096790A (en) * | 2016-06-22 | 2016-11-09 | 东南大学 | Based on convertible frequency air-conditioner virtual robot arm modeling virtual plant a few days ago with Real-time markets Optimization Scheduling |
-
2016
- 2016-11-28 CN CN201611067159.0A patent/CN106487011B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104699051A (en) * | 2015-02-12 | 2015-06-10 | 天津大学 | Demand response control method of temperature control device |
| CN104952001A (en) * | 2015-07-02 | 2015-09-30 | 华侨大学 | Method for performing power optimized scheduling on controllable loads comprising air conditioning loads |
| CN105322550A (en) * | 2015-08-28 | 2016-02-10 | 南方电网科学研究院有限责任公司 | Method for optimizing operation of household micro-grid |
| CN106096790A (en) * | 2016-06-22 | 2016-11-09 | 东南大学 | Based on convertible frequency air-conditioner virtual robot arm modeling virtual plant a few days ago with Real-time markets Optimization Scheduling |
| CN106094521A (en) * | 2016-06-30 | 2016-11-09 | 中国南方电网有限责任公司电网技术研究中心 | Flexible load energy efficiency power plant scheduling control method and system |
Cited By (55)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108629470A (en) * | 2017-03-17 | 2018-10-09 | 华北电力大学 | System capacity management of providing multiple forms of energy to complement each other based on non-cooperative game is run with optimization |
| CN107392465A (en) * | 2017-07-19 | 2017-11-24 | 北京上格云技术有限公司 | Build the operation management method and server of electromechanical equipment |
| CN107545326B (en) * | 2017-08-22 | 2021-02-26 | 国网能源研究院 | Household energy scheduling method based on cyber-physical system |
| CN107545326A (en) * | 2017-08-22 | 2018-01-05 | 国网能源研究院 | A kind of household energy scheduling method based on information physical emerging system |
| CN107785888A (en) * | 2017-10-13 | 2018-03-09 | 陆炜 | A kind of intelligent electric power load control system and control method |
| CN107785888B (en) * | 2017-10-13 | 2019-07-23 | 陆炜 | A kind of intelligent electric power load control system and control method |
| CN108092804A (en) * | 2017-12-08 | 2018-05-29 | 国网安徽省电力有限公司信息通信分公司 | Power telecom network maximization of utility resource allocation policy generation method based on Q-learning |
| CN108321793A (en) * | 2018-01-17 | 2018-07-24 | 东北电力大学 | The active distribution network of integrated intelligent building flexible load models and Optimization Scheduling |
| CN108321793B (en) * | 2018-01-17 | 2020-11-27 | 东北电力大学 | Active power distribution network modeling and optimal scheduling method integrating flexible loads of intelligent building |
| WO2019165702A1 (en) * | 2018-02-28 | 2019-09-06 | 东南大学 | Double-layer coordinated robust optimized scheduling method for multi-microgrids system |
| CN108490791A (en) * | 2018-05-10 | 2018-09-04 | 燕山大学 | Temperature control load Cost Controlling Policy |
| CN108490791B (en) * | 2018-05-10 | 2020-05-12 | 燕山大学 | Temperature controlled load cost control strategy |
| CN109066739A (en) * | 2018-07-26 | 2018-12-21 | 西南交通大学 | A kind of tractive power supply system regenerating braking energy energy-accumulating medium power and capacity collocation method |
| CN109063925A (en) * | 2018-08-16 | 2018-12-21 | 合肥工业大学 | It is a kind of meter and Load aggregation quotient regional complex energy resource system optimizing operation method |
| CN109063925B (en) * | 2018-08-16 | 2021-08-17 | 合肥工业大学 | Optimized operation method for regional comprehensive energy system considering load aggregators |
| CN109035812A (en) * | 2018-09-05 | 2018-12-18 | 平安科技（深圳）有限公司 | Control method, device, computer equipment and the storage medium of traffic lights |
| CN109035812B (en) * | 2018-09-05 | 2021-07-27 | 平安科技（深圳）有限公司 | Traffic signal lamp control method and device, computer equipment and storage medium |
| CN109190988A (en) * | 2018-09-11 | 2019-01-11 | 浙江大学 | A kind of Demand Side Response game method for realizing the optimal collaboration of temperature control load |
| CN109347149A (en) * | 2018-09-20 | 2019-02-15 | 国网河南省电力公司电力科学研究院 | Micro-capacitance sensor energy storage dispatching method and device based on depth Q value network intensified learning |
| CN109347149B (en) * | 2018-09-20 | 2022-04-22 | 国网河南省电力公司电力科学研究院 | Micro-grid energy storage scheduling method and device based on deep Q-value network reinforcement learning |
| CN109103936A (en) * | 2018-09-30 | 2018-12-28 | 南京铭越创信电气有限公司 | Optimal unit starting order calculation method after a kind of electric system is had a power failure on a large scale |
| CN109812927A (en) * | 2019-01-24 | 2019-05-28 | 新奥数能科技有限公司 | Family microgrid energy optimization method, device, readable medium and electronic equipment |
| CN109685396B (en) * | 2019-01-31 | 2021-10-19 | 河海大学 | Power distribution network energy management method considering public building demand response resources |
| CN109685396A (en) * | 2019-01-31 | 2019-04-26 | 河海大学 | It is a kind of meter and public building demand response resource power distribution network energy management method |
| US11965666B2 (en) | 2019-04-01 | 2024-04-23 | Gree Electric Appliances, Inc. Of Zhuhai | Control method for air conditioner, and device for air conditioner and storage medium |
| WO2020199648A1 (en) * | 2019-04-01 | 2020-10-08 | 珠海格力电器股份有限公司 | Control method and device for air conditioner |
| CN110212533A (en) * | 2019-07-10 | 2019-09-06 | 南方电网科学研究院有限责任公司 | Method and system for determining power of either person from birth or death |
| CN110414725B (en) * | 2019-07-11 | 2021-02-19 | 山东大学 | Wind power plant energy storage system scheduling method and device integrating prediction and decision |
| CN110414725A (en) * | 2019-07-11 | 2019-11-05 | 山东大学 | The integrated wind power plant energy-storage system dispatching method of forecast and decision and device |
| CN110958680B (en) * | 2019-12-09 | 2022-09-13 | 长江师范学院 | Energy efficiency-oriented unmanned aerial vehicle cluster multi-agent deep reinforcement learning optimization method |
| CN110958680A (en) * | 2019-12-09 | 2020-04-03 | 长江师范学院 | Energy efficiency-oriented unmanned aerial vehicle cluster multi-agent deep reinforcement learning optimization method |
| CN111324167A (en) * | 2020-02-27 | 2020-06-23 | 上海电力大学 | Photovoltaic power generation maximum power point tracking control method and device |
| CN111324167B (en) * | 2020-02-27 | 2022-07-01 | 上海电力大学 | Photovoltaic power generation maximum power point tracking control method |
| CN111431216A (en) * | 2020-03-18 | 2020-07-17 | 国网浙江嘉善县供电有限公司 | High-proportion photovoltaic microgrid reactive power sharing control method adopting Q learning |
| CN111431216B (en) * | 2020-03-18 | 2024-06-04 | 国网浙江省电力有限公司嘉善县供电公司 | Reactive power equipartition control method for high-proportion photovoltaic micro-grid by adopting Q learning |
| CN111404146B (en) * | 2020-03-19 | 2021-08-13 | 南方电网科学研究院有限责任公司 | Power distribution method, system, terminal and medium based on user load transfer comfort |
| CN111404146A (en) * | 2020-03-19 | 2020-07-10 | 南方电网科学研究院有限责任公司 | Power distribution method, system and terminal based on user load transfer comfort level |
| CN112117760A (en) * | 2020-08-13 | 2020-12-22 | 国网浙江省电力有限公司台州供电公司 | Micro-grid energy scheduling method based on double-Q-value network deep reinforcement learning |
| CN112465212A (en) * | 2020-11-24 | 2021-03-09 | 国网河北省电力有限公司经济技术研究院 | Building cluster demand side energy management method and system |
| CN112671033B (en) * | 2020-12-14 | 2022-12-23 | 广西电网有限责任公司电力科学研究院 | Priority-level-considered microgrid active scheduling control method and system |
| CN112671033A (en) * | 2020-12-14 | 2021-04-16 | 广西电网有限责任公司电力科学研究院 | Priority-level-considered microgrid active power scheduling control method and system |
| CN112594872A (en) * | 2020-12-15 | 2021-04-02 | 深圳供电局有限公司 | Multi-agent consistency control method considering air conditioner response cost |
| CN113011101A (en) * | 2021-03-29 | 2021-06-22 | 广东电网有限责任公司电力调度控制中心 | Control method and system for energy storage participating in frequency modulation auxiliary service optimization |
| CN113011101B (en) * | 2021-03-29 | 2024-01-23 | 广东电网有限责任公司电力调度控制中心 | Control method and system for energy storage to participate in frequency modulation auxiliary service optimization |
| CN113219930A (en) * | 2021-05-21 | 2021-08-06 | 上海交通大学 | Variable frequency air conditioner second-order equivalent thermal parameter model online identification method based on particle swarm optimization |
| CN113420413A (en) * | 2021-05-27 | 2021-09-21 | 国网上海市电力公司电力科学研究院 | Flexible load adjustability quantification method and system based on load plasticity |
| CN113381635A (en) * | 2021-06-29 | 2021-09-10 | 中南大学 | Traction converter thermal performance dynamic optimization control method and system |
| CN113609102A (en) * | 2021-08-11 | 2021-11-05 | 佛山仙湖实验室 | Construction method of energy management database of hybrid drive mining truck |
| CN113609102B (en) * | 2021-08-11 | 2024-03-19 | 佛山仙湖实验室 | Construction method of energy management database of hybrid drive mining truck |
| CN113890057A (en) * | 2021-09-18 | 2022-01-04 | 国网河北省电力有限公司经济技术研究院 | Control method and device based on multi-microgrid collaborative optimization and storage medium |
| CN113822493A (en) * | 2021-10-13 | 2021-12-21 | 国网天津市电力公司电力科学研究院 | Commercial building energy management method based on demand response |
| CN115268536A (en) * | 2022-08-02 | 2022-11-01 | 阳光电源股份有限公司 | Temperature control method of energy storage system and related device |
| CN115268536B (en) * | 2022-08-02 | 2024-05-14 | 阳光电源股份有限公司 | Temperature control method and related device of energy storage system |
| CN116628413A (en) * | 2023-07-24 | 2023-08-22 | 国网山西电力勘测设计研究院有限公司 | Method for calculating capacity of user side energy storage device |
| CN116628413B (en) * | 2023-07-24 | 2023-12-08 | 国网山西电力勘测设计研究院有限公司 | Method for calculating capacity of user side energy storage device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106487011B (en) | 2019-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106487011B (en) | A kind of family microgrid energy optimization method based on Q study | |
| Wu et al. | Deep learning adaptive dynamic programming for real time energy management and control strategy of micro-grid | |
| Li et al. | Reinforcement learning of room temperature set-point of thermal storage air-conditioning system with demand response | |
| CN105931136A (en) | Building micro-grid optimization scheduling method with demand side virtual energy storage system being fused | |
| CN106228258A (en) | A kind of meter and the home energy source LAN energy optimal control method of dsm | |
| CN110350523A (en) | Multi-energy complementation Optimization Scheduling based on demand response | |
| CN107706932B (en) | A kind of energy method for optimizing scheduling based on dynamic self-adapting fuzzy logic controller | |
| CN108429288A (en) | A kind of off-network type micro-capacitance sensor energy storage Optimal Configuration Method considering demand response | |
| CN110474370B (en) | Cooperative control system and method for air conditioner controllable load and photovoltaic energy storage system | |
| CN107453356B (en) | User side flexible load dispatching method based on adaptive Dynamic Programming | |
| CN106096790A (en) | Based on convertible frequency air-conditioner virtual robot arm modeling virtual plant a few days ago with Real-time markets Optimization Scheduling | |
| CN110991773A (en) | Two-stage source load-storage optimization scheduling method for wind power consumption | |
| CN116663820A (en) | Comprehensive energy system energy management method under demand response | |
| CN107451931A (en) | The Optimization Scheduling of home intelligent power equipment | |
| CN111047097A (en) | Day-to-day rolling optimization method for comprehensive energy system | |
| CN109884888A (en) | A kind of more building microgrid model predictions regulation method based on non-cooperative game | |
| Yu et al. | Optimal dispatching method for integrated energy system based on robust economic model predictive control considering source–load power interval prediction | |
| CN114997715A (en) | Improved fuzzy C-means clustering-based combined cooling, heating and power system configuration method | |
| Harrold et al. | Battery control in a smart energy network using double dueling deep q-networks | |
| CN113435042B (en) | Reinforced learning modeling method for demand response of building air conditioning system | |
| Wang et al. | Data-driven real-time pricing strategy and coordinated optimization of economic load dispatch in electricity market | |
| CN117833316A (en) | Method for dynamically optimizing operation of energy storage at user side | |
| CN108197412A (en) | A kind of multiple-energy-source coupling Energy Management System and optimization method | |
| CN116780627A (en) | Micro-grid regulation and control method in building park | |
| CN115619431A (en) | Scheduling method, device, terminal and storage medium of microgrid |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |