CN115325666A - Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism - Google Patents

Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism Download PDF

Info

Publication number
CN115325666A
CN115325666A CN202210842189.3A CN202210842189A CN115325666A CN 115325666 A CN115325666 A CN 115325666A CN 202210842189 A CN202210842189 A CN 202210842189A CN 115325666 A CN115325666 A CN 115325666A
Authority
CN
China
Prior art keywords
air conditioner
load
lstm
heat storage
model
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
CN202210842189.3A
Other languages
Chinese (zh)
Inventor
刘盼盼
章锐
周吉
钱俊良
邰伟
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.)
Nanjing Dongbo Intelligent Energy Research Institute Co ltd
Liyang Research Institute of Southeast University
Original Assignee
Nanjing Dongbo Intelligent Energy Research Institute Co ltd
Liyang Research Institute of Southeast University
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 Nanjing Dongbo Intelligent Energy Research Institute Co ltd, Liyang Research Institute of Southeast University filed Critical Nanjing Dongbo Intelligent Energy Research Institute Co ltd
Priority to CN202210842189.3A priority Critical patent/CN115325666A/en
Publication of CN115325666A publication Critical patent/CN115325666A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
    • F24F11/46—Improving electric energy efficiency or saving
    • F24F11/47—Responding to energy costs
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/61—Control or safety arrangements characterised by user interfaces or communication using timers
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Combustion & Propulsion (AREA)
  • Chemical & Material Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Biophysics (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Human Computer Interaction (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The invention relates to the technical field of power systems, and discloses an air conditioner response capacity evaluation method and device based on an LSTM and a heat storage mechanism, wherein the technical scheme is characterized by comprising the following steps of: according to the heat storage mechanism of the air conditioner, an air conditioner heat energy conversion efficiency model is constructed, and the conversion efficiency between the heat stored in the heat exchange process of the air conditioner and the consumed electric energy is obtained; constructing an average power range evaluation model; according to the historical controllable air conditioner quantity data and the LSTM, constructing a controllable air conditioner quantity prediction model based on the LSTM; according to the average maximum power and the average minimum of a single air conditioner in one operation periodPower, predicted next time
Figure 512596DEST_PATH_IMAGE001
The number of controllable air conditioners capable of participating in load adjustment and a load aggregation quotient are used for constructing an air conditioner load maximum and minimum response capability evaluation model based on the load aggregation quotient, and the air conditioner load maximum and minimum response capability under the control of the load aggregation quotient is evaluated and obtained.

Description

Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism
Technical Field
The invention relates to the technical field of power systems, in particular to an air conditioner response capacity evaluation method and device based on an LSTM and a heat storage mechanism.
Background
Along with the development of low-carbon energy, the power grid is changed to a novel power system taking new energy as a main body, the novel power system faces the impact of variable operation modes of a power supply side and multiple energy sources and loads inside a load side in a normalized mode, the peak-valley difference or the peak-valley number of the net demand of the novel power system is increased possibly, indexes such as the peak-valley regulation amplitude, the frequency and the span of the power grid are increased rapidly, and higher requirements are provided for the dispatching capacity of the power grid.
The power grid can regulate resource shortage, and needs to excavate load side resources to participate in power grid dispatching and support power grid day-to-day operation plan correction. In recent years, with the change of global climate and the improvement of the requirement of people on comfort level, the air conditioning load is rapidly increased, the urban air conditioning load is rapidly increased, in summer, part of the urban air conditioning load is as high as 40% of the power consumption of the whole society, the essence of air conditioning operation is the process of converting electric energy into heat energy, the heat energy has thermal inertia, a human body has a certain range on the comfort level and a certain range on the demand of the heat energy, the consumption of the electric energy can be adjusted by utilizing the thermal inertia of the heat energy, such as early refrigeration and late refrigeration, and the heat energy is kept within a certain range, so that the adjustment of the electric energy is realized, and the air conditioning load has a larger adjusting space. The day-to-day adjustment response capability of the air conditioner is evaluated, and the method has important significance for supporting power grid dispatching and correcting day-to-day operation plans.
Disclosure of Invention
The invention aims to provide an air conditioner response capacity evaluation method and device based on an LSTM and a heat storage mechanism, which can effectively evaluate the air conditioner load response capacity of a load aggregator and master the upper and lower limit values of the air conditioner load power of the load aggregator.
The technical purpose of the invention is realized by the following technical scheme: the air conditioner response capacity evaluation method based on the LSTM and the heat storage mechanism comprises the following steps:
according to the heat storage mechanism of the air conditioner, an air conditioner heat energy conversion efficiency model is constructed, and the conversion efficiency between the heat stored in the heat exchange process of the air conditioner and the consumed electric energy is obtained;
according to the conversion efficiency between the heat stored and the consumed electric energy in the heat exchange process of the air conditioner, an average power range evaluation model is constructed to obtain the average maximum power and the average minimum power of a single air conditioner in one operation period;
according to the historical controllable air conditioner quantity data and the LSTM, an LSTM-based controllable air conditioner quantity prediction model is constructed, and the current time is used
Figure 38776DEST_PATH_IMAGE001
The data is substituted into the LSTM-based controllable air conditioner quantity prediction model, and the next moment is obtained through prediction
Figure 274585DEST_PATH_IMAGE002
The number of controllable air conditioners which can participate in load adjustment;
predicting the next moment according to the average maximum power and the average minimum power of the single air conditioner in one operation period
Figure 363895DEST_PATH_IMAGE003
The method comprises the steps of setting up a load aggregation quotient based air conditioner load maximum and minimum response capability evaluation model, and evaluating to obtain the air conditioner load maximum under the control of the load aggregation quotientMaximum minimum response capability.
As a preferred technical scheme of the method, different methods are set
Figure 659747DEST_PATH_IMAGE004
And circularly iterating the energy-storage capacity estimation model to an air conditioner heat energy conversion efficiency model, an average power range estimation model, an LSTM-based controllable air conditioner quantity prediction model and a load aggregator-based air conditioner load maximum and minimum response capacity estimation model to obtain the energy-storage capacity estimation model for the future
Figure 224720DEST_PATH_IMAGE005
Maximum and minimum response capability of air conditioner load under the control of load aggregators in time.
As a preferred technical scheme of the method, the construction process of the air conditioner heat energy conversion efficiency model comprises the following steps:
the heat storage mechanism of the air conditioner is determined, namely the air conditioner operates in a heat storage process of converting electric energy into heat energy;
based on the heat storage mechanism of the air conditioner, the air conditioner is arranged
Figure 504699DEST_PATH_IMAGE006
The indoor temperature is controlled by
Figure 104308DEST_PATH_IMAGE007
Is reduced to
Figure 836640DEST_PATH_IMAGE008
The heat stored in the process is
Figure 233118DEST_PATH_IMAGE009
Air conditioner
Figure 217254DEST_PATH_IMAGE006
The indoor temperature is controlled by
Figure 530424DEST_PATH_IMAGE008
Is reduced to
Figure 417346DEST_PATH_IMAGE010
The heat stored in the process is
Figure 691333DEST_PATH_IMAGE011
Then, the formula of the heat energy conversion efficiency of the air conditioner is as follows:
Figure 72635DEST_PATH_IMAGE012
；
Figure 381257DEST_PATH_IMAGE013
(ii) a Wherein the content of the first and second substances,
Figure 940545DEST_PATH_IMAGE014
for air-conditioning
Figure 92041DEST_PATH_IMAGE006
The outdoor temperature of (a) is set,
Figure 886822DEST_PATH_IMAGE008
、
Figure 692360DEST_PATH_IMAGE010
are air conditioners respectively
Figure 812763DEST_PATH_IMAGE006
The set indoor maximum temperature and minimum temperature,
Figure 185975DEST_PATH_IMAGE015
、
Figure 50026DEST_PATH_IMAGE016
are air conditioners respectively
Figure 208606DEST_PATH_IMAGE006
The on/off operation cycle of the switch is started,
Figure 358965DEST_PATH_IMAGE017
for the load aggregators
Figure 94840DEST_PATH_IMAGE018
Controlled air conditioneriThe power of the operation is that of the power,
Figure 136483DEST_PATH_IMAGE019
as a load aggregatorAControlled air conditioner
Figure 274203DEST_PATH_IMAGE006
The heat energy conversion efficiency of (2).
As a preferred technical scheme of the method, the process of constructing the average power range evaluation model comprises the following steps:
obtaining the air conditioner according to the model of the heat energy conversion efficiency of the air conditioner
Figure 595463DEST_PATH_IMAGE006
During the whole operation period, the minimum heat storage energy is
Figure 818634DEST_PATH_IMAGE020
Maximum stored heat energy of
Figure 165433DEST_PATH_IMAGE021
Then can obtain
Figure 157659DEST_PATH_IMAGE022
Wherein, in the process,
Figure 915400DEST_PATH_IMAGE023
and
Figure 625867DEST_PATH_IMAGE024
are respectively the load aggregatorsAControlled air conditioner
Figure 746663DEST_PATH_IMAGE006
Average minimum power and average maximum power over a run period.
As a preferred technical scheme of the method, the process of constructing the controllable air conditioner quantity prediction model based on the LSTM comprises the following steps:
acquiring historical controllable air conditioner quantity data, and constructing an LSTM model which takes the controllable air conditioner quantity data at the current moment as input and takes the controllable air conditioner quantity data at the next moment as output; training the model by taking the data of the number of the historical controllable air conditioners as sample data;
the LSTM model is:
Figure 718030DEST_PATH_IMAGE025
(ii) a Wherein, the first and the second end of the pipe are connected with each other,
Figure 787617DEST_PATH_IMAGE026
in order to be an LSTM model,
Figure 595167DEST_PATH_IMAGE027
as a load aggregatorAThe number of the air conditioners can be controlled historically,
Figure 408403DEST_PATH_IMAGE028
after time t, to be predicted
Figure 968697DEST_PATH_IMAGE029
The number of air conditioners can be controlled at any time;
the formula used in the training process is:
Figure 209185DEST_PATH_IMAGE030
wherein
Figure 2567DEST_PATH_IMAGE031
Is an activation function;
Figure 353914DEST_PATH_IMAGE032
in order to forget the output of the gate,
Figure 34294DEST_PATH_IMAGE033
、
Figure 321050DEST_PATH_IMAGE034
is a corresponding forgetting gate matrix;
Figure 227826DEST_PATH_IMAGE035
is the output of the input gate or gates,
Figure 507498DEST_PATH_IMAGE036
、
Figure 183330DEST_PATH_IMAGE037
a weight matrix for the corresponding input gate;
Figure 142452DEST_PATH_IMAGE038
is the information of the state of old cells,
Figure 270945DEST_PATH_IMAGE039
in order to select the addition of the candidate state information,
Figure 88728DEST_PATH_IMAGE040
in order to update the information on the cells,
Figure 884646DEST_PATH_IMAGE041
、
Figure 247625DEST_PATH_IMAGE042
is a corresponding neuron matrix;
Figure 988048DEST_PATH_IMAGE043
is the output of the output gate or gates,
Figure 484888DEST_PATH_IMAGE044
、
Figure 509214DEST_PATH_IMAGE045
is a corresponding matrix of output gates;
Figure 433308DEST_PATH_IMAGE046
in order to output the result of the process,
Figure 395447DEST_PATH_IMAGE047
for the samples at the time t, the samples are,
Figure 430399DEST_PATH_IMAGE048
is composed of
Figure 76276DEST_PATH_IMAGE049
Time hidden layer input quantity。
As a preferred technical scheme of the method, the working process of the air conditioner load maximum and minimum response capability evaluation model based on the load aggregators is as follows:
obtaining the average maximum power and the average minimum power of a single air conditioner in one operation period, and obtaining the next time obtained by prediction
Figure 171270DEST_PATH_IMAGE050
The number of the controllable air conditioners which can participate in load adjustment is obtained, and the controlled air conditioners of the load aggregators are obtained;
substituting into a model formula:
Figure 620706DEST_PATH_IMAGE051
(ii) a Wherein the content of the first and second substances,
Figure 847899DEST_PATH_IMAGE052
、
Figure 472916DEST_PATH_IMAGE053
respectively future time
Figure 863446DEST_PATH_IMAGE054
The load aggregator a controls the maximum and minimum response power of the cluster air conditioners,
Figure 675544DEST_PATH_IMAGE055
the number of air conditioners may be controlled for the load aggregator history,
Figure 927665DEST_PATH_IMAGE056
after time t, to be predicted
Figure 407188DEST_PATH_IMAGE057
The number of air conditioners can be controlled at any time.
Air conditioner response ability evaluation device based on LSTM and heat storage mechanism includes: the system comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor executes the computer program to realize the air conditioner response capacity evaluation method based on the LSTM and the heat storage mechanism.
In conclusion, the invention has the following beneficial effects: the air conditioner load response capacity of the load aggregator can be evaluated, the upper limit value and the lower limit value of the air conditioner load power can be mastered, support is provided for plan correction and power grid dispatching of a power grid, the power grid regulation and control level is improved, and the stable operation level of the power grid is improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of the heat storage mechanism of the air conditioner of the present invention;
FIG. 3 is a schematic diagram of the LSTM model training of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The invention provides an air conditioner response capability assessment method based on LSTM and a heat storage mechanism, which comprises the following steps as shown in figure 1:
s1, constructing an air conditioner heat energy conversion efficiency model according to a heat storage mechanism of an air conditioner, and obtaining the conversion efficiency between the heat stored in the heat exchange process of the air conditioner and the consumed electric energy.
The construction process of the air conditioner heat energy conversion efficiency model comprises the following steps:
s11, determining a heat storage mechanism of the air conditioner, namely, the air conditioner runs in a heat storage process of converting electric energy into heat energy;
s12, based on the heat storage mechanism of the air conditioner, as shown in figure 2, the air conditioner is arranged
Figure 968619DEST_PATH_IMAGE006
The indoor temperature is controlled by
Figure 533593DEST_PATH_IMAGE058
Is reduced to
Figure 822360DEST_PATH_IMAGE059
The heat stored in the process is
Figure 281023DEST_PATH_IMAGE060
After entering the comfortable area, the air conditioner
Figure 888722DEST_PATH_IMAGE006
The indoor temperature is controlled by
Figure 816358DEST_PATH_IMAGE061
Is reduced to
Figure 800495DEST_PATH_IMAGE062
The heat stored in the process is
Figure 848085DEST_PATH_IMAGE063
The comfort zone is the set maximum temperature and minimum temperature in the air-conditioning room, and then the air-conditioning heat energy conversion efficiency formula is:
Figure 626685DEST_PATH_IMAGE012
；
Figure 543082DEST_PATH_IMAGE013
(ii) a Wherein the content of the first and second substances,
Figure 65330DEST_PATH_IMAGE064
for air-conditioning
Figure 498586DEST_PATH_IMAGE006
The outdoor temperature of (a) is set,
Figure 57874DEST_PATH_IMAGE065
、
Figure 84736DEST_PATH_IMAGE066
are air conditioners respectively
Figure 269730DEST_PATH_IMAGE006
The set maximum temperature and the minimum temperature in the room,
Figure 432858DEST_PATH_IMAGE067
、
Figure 927162DEST_PATH_IMAGE068
are air conditioners respectively
Figure 175741DEST_PATH_IMAGE006
The switch of (2) is turned on and off for a running period,
Figure 164425DEST_PATH_IMAGE069
as a load aggregator
Figure 447639DEST_PATH_IMAGE018
Controlled air conditioneriThe power of the operation is controlled by the power controller,
Figure 348730DEST_PATH_IMAGE070
as a load aggregatorAControlled air conditioner
Figure 209239DEST_PATH_IMAGE006
The efficiency of thermal energy conversion, in figure 2,
Figure 876980DEST_PATH_IMAGE071
、
Figure 391532DEST_PATH_IMAGE072
respectively represent the load aggregators
Figure 853737DEST_PATH_IMAGE018
Air conditioner
Figure 201542DEST_PATH_IMAGE006
In thattThe time is in idle state and running state.
S2, according to the conversion efficiency between the heat stored in the heat exchange process of the air conditioner and the consumed electric energy, an average power range evaluation model is constructed, and the average maximum power and the average minimum power of the single air conditioner in one operation period are obtained.
Obtaining the air conditioner according to the model of the heat energy conversion efficiency of the air conditioner
Figure 141816DEST_PATH_IMAGE006
During the whole operation period, the minimum heat storage energy is
Figure 9409DEST_PATH_IMAGE073
I.e. the air conditioner will control the temperature
Figure 908095DEST_PATH_IMAGE074
Is reduced to
Figure 743196DEST_PATH_IMAGE075
After entering the comfort zone, the indoor temperature is maintained at the higher comfort zone temperature. Maximum heat storage energy of
Figure 487161DEST_PATH_IMAGE076
I.e. the air conditioner will control the temperature
Figure 707795DEST_PATH_IMAGE077
Is reduced to
Figure 636437DEST_PATH_IMAGE078
After entering the comfort zone, the indoor temperature is maintained at the lower comfort zone temperature, and the load aggregation quotient can be calculatedAControlled air conditioner
Figure 834200DEST_PATH_IMAGE006
Average minimum power over a run period
Figure 257222DEST_PATH_IMAGE079
And average maximum power
Figure 958462DEST_PATH_IMAGE080
It behaves as:
Figure 323584DEST_PATH_IMAGE081
。
s3, according to the historical quantity data of the controllable air conditioners and the LSTM, constructing a controllable air conditioner quantity prediction model based on the LSTM, and enabling the current time to be the same
Figure 743064DEST_PATH_IMAGE082
Substituting the data into a controllable air conditioner quantity prediction model based on the LSTM, and predicting to obtain the next moment
Figure 471242DEST_PATH_IMAGE083
The number of controllable air conditioners which can participate in load adjustment.
As shown in fig. 3, the process of constructing the LSTM-based controlled air-conditioning quantity prediction model includes:
acquiring historical controllable air conditioner quantity data, and constructing an LSTM model which takes the controllable air conditioner quantity data at the current moment as input and takes the controllable air conditioner quantity data at the next moment as output; training the model by taking the data of the number of the historical controllable air conditioners as sample data;
the LSTM model is:
Figure 292568DEST_PATH_IMAGE084
(ii) a Wherein the content of the first and second substances,
Figure 563012DEST_PATH_IMAGE085
in order to be an LSTM model,
Figure 345155DEST_PATH_IMAGE086
for the load aggregatorsAThe number of air conditioners can be controlled historically,
Figure 234613DEST_PATH_IMAGE087
after time t, to be predicted
Figure 300658DEST_PATH_IMAGE088
The number of air conditioners can be controlled at any time;
the formula used in the training process is:
Figure 882949DEST_PATH_IMAGE089
wherein
Figure 385344DEST_PATH_IMAGE090
Is an activation function;
Figure 78493DEST_PATH_IMAGE032
in order to forget the output of the gate,
Figure 733465DEST_PATH_IMAGE033
、
Figure 486658DEST_PATH_IMAGE034
is a corresponding forgetting gate matrix;
Figure 243392DEST_PATH_IMAGE035
is the output of the input gate or gates,
Figure 740233DEST_PATH_IMAGE036
、
Figure 249711DEST_PATH_IMAGE037
a weight matrix for the corresponding input gate;
Figure 273338DEST_PATH_IMAGE038
as information on the state of the old cells,
Figure 642003DEST_PATH_IMAGE039
in order to select the addition of the candidate state information,
Figure 801589DEST_PATH_IMAGE040
in order to update the information on the cells,
Figure 572098DEST_PATH_IMAGE041
、
Figure 11301DEST_PATH_IMAGE042
is a corresponding neuron matrix;
Figure 867262DEST_PATH_IMAGE043
in order to output the output of the gate,
Figure 830539DEST_PATH_IMAGE044
、
Figure 189976DEST_PATH_IMAGE045
is a corresponding output gate matrix;
Figure 829773DEST_PATH_IMAGE046
in order to output the result of the process,
Figure 907451DEST_PATH_IMAGE047
for the samples at the time t,
Figure 408839DEST_PATH_IMAGE048
is composed of
Figure 763728DEST_PATH_IMAGE049
The time instant implies the layer input quantity.
S4, predicting the next moment according to the average maximum power and the average minimum power of the single air conditioner in one operation period
Figure 200526DEST_PATH_IMAGE091
The number of controllable air conditioners capable of participating in load adjustment and a load aggregation quotient are used for constructing an air conditioner load maximum and minimum response capability evaluation model based on the load aggregation quotient, and the air conditioner load maximum and minimum response capability under the control of the load aggregation quotient is evaluated and obtained.
The working process of the air conditioner load maximum and minimum response capacity evaluation model based on the load aggregators is as follows:
obtaining the average maximum power and the average minimum power of a single air conditioner in an operation period, and obtaining the next predicted time
Figure 624554DEST_PATH_IMAGE092
The number of the controllable air conditioners which can participate in load adjustment is obtained, and the controlled air conditioners of the load aggregators are obtained;
substituting into a model formula:
Figure 805000DEST_PATH_IMAGE093
(ii) a Wherein, the first and the second end of the pipe are connected with each other,
Figure 515860DEST_PATH_IMAGE094
、
Figure 123559DEST_PATH_IMAGE095
respectively the future time
Figure 300462DEST_PATH_IMAGE096
Load aggregation business A controlled cluster air conditionerThe large and the minimum response power are set,
Figure 19020DEST_PATH_IMAGE097
the number of air conditioners can be controlled for the history of the load aggregator,
Figure 82922DEST_PATH_IMAGE098
after time t, to be predicted
Figure 986156DEST_PATH_IMAGE099
The number of air conditioners can be controlled at any time.
S5, setting up differently
Figure 525721DEST_PATH_IMAGE100
The method comprises the steps of circularly iterating the method to an air conditioner heat energy conversion efficiency model, an average power range evaluation model, an LSTM-based controllable air conditioner quantity prediction model and a load aggregator-based air conditioner load maximum and minimum response capacity evaluation model to obtain a future-to-future air conditioner load maximum and minimum response capacity evaluation model
Figure 421871DEST_PATH_IMAGE101
Maximum and minimum response capacity of air conditioning load under control of load aggregator over time. E.g. arranged differently according to predicted demand
Figure 730493DEST_PATH_IMAGE102
And the effective prediction and evaluation of the air conditioner load response capacity in the future time of 15min, 1h, 2h and the like can be realized.
Corresponding to the method, the invention also provides an air conditioner response capability evaluation device based on the LSTM and the heat storage mechanism, which comprises the following steps: the system comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor executes the computer program to realize the air conditioner response capacity evaluation method based on the LSTM and the heat storage mechanism.
By the evaluation method and the evaluation device, the air conditioner load response capacity of the load aggregator can be evaluated, the upper limit value and the lower limit value of the air conditioner load power can be mastered, support is provided for plan correction and power grid dispatching of a power grid, the power grid regulation and control level is improved, and the stable operation level of the power grid is improved.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.

Claims (7)

1. An air conditioner response capability assessment method based on LSTM and heat storage mechanism is characterized in that: the method comprises the following steps:
according to the heat storage mechanism of the air conditioner, an air conditioner heat energy conversion efficiency model is constructed, and the conversion efficiency between the heat stored in the heat exchange process of the air conditioner and the consumed electric energy is obtained;
according to the conversion efficiency between the heat stored and the consumed electric energy in the heat exchange process of the air conditioner, an average power range evaluation model is constructed to obtain the average maximum power and the average minimum power of a single air conditioner in one operation period;
according to the historical controllable air conditioner quantity data and the LSTM, an LSTM-based controllable air conditioner quantity prediction model is constructed, and the current time is used
Figure 170470DEST_PATH_IMAGE001
The data is substituted into the LSTM-based controllable air conditioner quantity prediction model, and the next moment is obtained through prediction
Figure 761245DEST_PATH_IMAGE002
The number of controllable air conditioners which can participate in load adjustment;
predicting the next moment according to the average maximum power and the average minimum power of a single air conditioner in an operation period
Figure 322676DEST_PATH_IMAGE003
Controllable air conditioner quantity and load aggregators capable of participating in load adjustment, and load-based constructionAnd evaluating the maximum and minimum response capability of the air conditioner load of the aggregation provider to obtain the maximum and minimum response capability of the air conditioner load under the control of the load aggregation provider.
2. The method of claim 1 for evaluating responsiveness of an air conditioner based on LSTM and heat storage mechanisms, wherein: by setting up differently
Figure 887650DEST_PATH_IMAGE004
The method comprises the steps of circularly iterating the method to an air conditioner heat energy conversion efficiency model, an average power range evaluation model, an LSTM-based controllable air conditioner quantity prediction model and a load aggregator-based air conditioner load maximum and minimum response capacity evaluation model to obtain a future-to-future air conditioner load maximum and minimum response capacity evaluation model
Figure 677882DEST_PATH_IMAGE004
Maximum and minimum response capacity of air conditioning load under control of load aggregator over time.
3. The method of claim 2 for evaluating responsiveness of an air conditioner based on LSTM and heat storage mechanisms, wherein: the construction process of the air conditioner heat energy conversion efficiency model comprises the following steps:
the heat storage mechanism of the air conditioner is determined, namely the air conditioner operates in a heat storage process of converting electric energy into heat energy;
based on the heat storage mechanism of the air conditioner, the air conditioner is arranged
Figure 277491DEST_PATH_IMAGE005
The indoor temperature is controlled by
Figure DEST_PATH_IMAGE006
Is reduced to
Figure 213086DEST_PATH_IMAGE007
The heat stored in the process is
Figure DEST_PATH_IMAGE008
Air conditioner
Figure 967153DEST_PATH_IMAGE005
The indoor temperature is controlled by
Figure 685710DEST_PATH_IMAGE009
Is reduced to
Figure DEST_PATH_IMAGE010
The heat stored in the process is
Figure 484033DEST_PATH_IMAGE011
Then, the formula of the heat energy conversion efficiency of the air conditioner is as follows:
Figure 121688DEST_PATH_IMAGE012
；
Figure 661254DEST_PATH_IMAGE013
；
wherein, the first and the second end of the pipe are connected with each other,
Figure 560333DEST_PATH_IMAGE014
for air-conditioning
Figure 868955DEST_PATH_IMAGE005
The outdoor temperature of the air conditioner is set,
Figure 677511DEST_PATH_IMAGE015
、
Figure 314159DEST_PATH_IMAGE016
are air conditioners respectively
Figure 640099DEST_PATH_IMAGE005
The set indoor maximum temperature and minimum temperature,
Figure 927860DEST_PATH_IMAGE017
、
Figure 782684DEST_PATH_IMAGE018
are air conditioners respectively
Figure 670743DEST_PATH_IMAGE005
The switch of (2) is turned on and off for a running period,
Figure 534794DEST_PATH_IMAGE019
as a load aggregator
Figure 411483DEST_PATH_IMAGE020
Controlled air conditioneriThe power of the operation is controlled by the power controller,
Figure 843733DEST_PATH_IMAGE021
、
Figure 579608DEST_PATH_IMAGE022
respectively represent the load aggregators
Figure 840825DEST_PATH_IMAGE020
Air conditioner
Figure 978545DEST_PATH_IMAGE005
In thattThe time is in an idle state and a running state,
Figure 71442DEST_PATH_IMAGE023
for the load aggregatorsAControlled air conditioner
Figure 29033DEST_PATH_IMAGE005
The heat energy conversion efficiency of (2).
4. The method of claim 3 for evaluating responsiveness of an air conditioner based on LSTM and heat storage mechanisms, wherein: the process of constructing the average power range evaluation model comprises the following steps:
according to the model of the heat energy conversion efficiency of the air conditioner,get in the air conditioner
Figure 359521DEST_PATH_IMAGE005
During the whole operation period, the minimum heat storage energy is
Figure 227114DEST_PATH_IMAGE024
Maximum heat storage energy of
Figure 860220DEST_PATH_IMAGE025
Then can obtain
Figure 695321DEST_PATH_IMAGE026
Wherein, in the step (A),
Figure 439286DEST_PATH_IMAGE027
and
Figure 659921DEST_PATH_IMAGE028
are respectively the load aggregatorsAControlled air conditioner
Figure 729508DEST_PATH_IMAGE005
Average minimum power and average maximum power over a run period.
5. The method of claim 4 for evaluating responsiveness of an air conditioner based on LSTM and heat storage mechanisms, wherein: the process of constructing the controllable air conditioner quantity prediction model based on the LSTM comprises the following steps:
acquiring historical controllable air conditioner quantity data, and constructing an LSTM model which takes the controllable air conditioner quantity data at the current moment as input and takes the controllable air conditioner quantity data at the next moment as output; training the model by taking the data of the number of the historical controllable air conditioners as sample data;
the LSTM model is:
Figure 520747DEST_PATH_IMAGE029
(ii) a Wherein the content of the first and second substances,
Figure 209348DEST_PATH_IMAGE030
in order to be the model of the LSTM,
Figure 176167DEST_PATH_IMAGE031
as a load aggregatorAThe number of the air conditioners can be controlled historically,
Figure 275710DEST_PATH_IMAGE032
after time t, to be predicted
Figure 695190DEST_PATH_IMAGE033
The number of air conditioners can be controlled at any time;
the formula used in the training process is:
Figure 423368DEST_PATH_IMAGE034
in which
Figure 244693DEST_PATH_IMAGE035
Is an activation function;
Figure 515138DEST_PATH_IMAGE036
in order to forget the output of the gate,
Figure 766122DEST_PATH_IMAGE037
、
Figure DEST_PATH_IMAGE038
is a corresponding forgetting gate matrix;
Figure 514635DEST_PATH_IMAGE039
is the output of the input gate or gates,
Figure 190467DEST_PATH_IMAGE040
、
Figure 881080DEST_PATH_IMAGE041
a weight matrix for the corresponding input gate;
Figure DEST_PATH_IMAGE042
as information on the state of the old cells,
Figure 868628DEST_PATH_IMAGE043
in order to select the addition of the candidate state information,
Figure DEST_PATH_IMAGE044
in order to update the information on the cells,
Figure 905985DEST_PATH_IMAGE045
、
Figure DEST_PATH_IMAGE046
is a corresponding neuron matrix;
Figure 29799DEST_PATH_IMAGE047
is the output of the output gate or gates,
Figure DEST_PATH_IMAGE048
、
Figure 628664DEST_PATH_IMAGE049
is a corresponding output gate matrix;
Figure DEST_PATH_IMAGE050
in order to output the result of the process,
Figure 837928DEST_PATH_IMAGE051
for the samples at the time t,
Figure 944556DEST_PATH_IMAGE052
is composed of
Figure 719613DEST_PATH_IMAGE053
The time instant implies the layer input quantity.
6. The method of claim 5 for evaluating responsiveness of an air conditioner based on LSTM and heat storage mechanisms, wherein: the working process of the air conditioner load maximum and minimum response capability evaluation model based on the load aggregators is as follows:
obtaining the average maximum power and the average minimum power of a single air conditioner in an operation period, and obtaining the next predicted time
Figure 643707DEST_PATH_IMAGE054
The number of the controllable air conditioners which can participate in load adjustment is obtained, and the controlled air conditioners of the load aggregators are obtained;
substituting into a model formula:
Figure 386273DEST_PATH_IMAGE055
(ii) a Wherein the content of the first and second substances,
Figure 421225DEST_PATH_IMAGE056
、
Figure 50790DEST_PATH_IMAGE057
respectively future time
Figure 145784DEST_PATH_IMAGE059
The load aggregator a controls the maximum and minimum response power of the cluster air conditioners,
Figure 345953DEST_PATH_IMAGE060
the number of air conditioners may be controlled for the load aggregator history,
Figure 184596DEST_PATH_IMAGE061
after time t, to be predicted
Figure 934246DEST_PATH_IMAGE062
The number of air conditioners can be controlled at any time.
7. Air conditioner response ability evaluation device based on LSTM and heat storage mechanism, characterized by: the method comprises the following steps: a processor and a memory, the memory storing a computer program executable by the processor, the processor implementing the LSTM and heat storage mechanism based air conditioner responsiveness evaluation method of any of claims 1-6 when executing the computer program.
CN202210842189.3A 2022-07-18 2022-07-18 Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism Pending CN115325666A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210842189.3A CN115325666A (en) 2022-07-18 2022-07-18 Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210842189.3A CN115325666A (en) 2022-07-18 2022-07-18 Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism

Publications (1)

Publication Number Publication Date
CN115325666A true CN115325666A (en) 2022-11-11

Family

ID=83918524

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210842189.3A Pending CN115325666A (en) 2022-07-18 2022-07-18 Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism

Country Status (1)

Country Link
CN (1) CN115325666A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107101322A (en) * 2017-04-13 2017-08-29 东南大学 The convertible frequency air-conditioner group potential evaluation method of unified maximum reduction plans duration
CN111271839A (en) * 2020-02-14 2020-06-12 河海大学 Method for adjusting short-term power of fixed-frequency air conditioner
CN111380160A (en) * 2018-12-27 2020-07-07 江苏方天电力技术有限公司 Method for mining user comfort level heating ventilation air conditioner load demand response potential
CN111555304A (en) * 2019-04-11 2020-08-18 东南大学 Air conditioner load virtual energy storage scheduling method for power grid frequency modulation service
WO2022127652A1 (en) * 2020-12-17 2022-06-23 东南大学 Air conditioning aggregation control method, system, and regulation apparatus

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107101322A (en) * 2017-04-13 2017-08-29 东南大学 The convertible frequency air-conditioner group potential evaluation method of unified maximum reduction plans duration
CN111380160A (en) * 2018-12-27 2020-07-07 江苏方天电力技术有限公司 Method for mining user comfort level heating ventilation air conditioner load demand response potential
CN111555304A (en) * 2019-04-11 2020-08-18 东南大学 Air conditioner load virtual energy storage scheduling method for power grid frequency modulation service
CN111271839A (en) * 2020-02-14 2020-06-12 河海大学 Method for adjusting short-term power of fixed-frequency air conditioner
WO2022127652A1 (en) * 2020-12-17 2022-06-23 东南大学 Air conditioning aggregation control method, system, and regulation apparatus

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
唐早;刘俊勇;刘友波;李婷;胥威汀;苟竞;: "空调聚合商参与下的负荷控制与配电网重构" *
戴晖 *

Similar Documents

Publication Publication Date Title
Satrio et al. Optimization of HVAC system energy consumption in a building using artificial neural network and multi-objective genetic algorithm
Liu et al. Energy flexibility of a nearly zero-energy building with weather predictive control on a convective building energy system and evaluated with different metrics
Xu et al. Performance analysis and comparison on energy storage devices for smart building energy management
CN105320118A (en) Method for electric power demand response control of air conditioning systems based on cloud platform
Cole et al. Use of model predictive control to enhance the flexibility of thermal energy storage cooling systems
Merabti et al. A review of control systems for energy and comfort management in buildings
CN104049716A (en) Computer energy-saving method and system combined with temperature sensing
CN107576015A (en) A kind of building air conditioning model predictive control method and device for realizing Demand Side Response
CN110661253B (en) Multi-type electric power elastic load combined adjusting method in building group
CN107726555A (en) A kind of building air conditioning model predictive control method and device
CN114784811A (en) Power demand response method and device
Cui et al. Application potential analysis of different control strategies for radiant floor cooling systems in office buildings in different climate zones of China
CN113610330B (en) User experience-based user-side flexible resource energy utilization behavior optimization method
Yuan et al. Two-level collaborative demand-side management for regional distributed energy system considering carbon emission quotas
TW201027014A (en) Method for managing air conditioning power consumption
CN112906988B (en) Robust double-layer coordinated scheduling method for multi-energy building system
Bayer et al. Enhancing the performance of multi-agent reinforcement learning for controlling HVAC systems
CN115325666A (en) Air conditioner response capability assessment method and device based on LSTM and heat storage mechanism
CN108805445B (en) Grouping sequence scheduling method for providing rotary standby for air conditioner load group
CN114936529A (en) Temperature control load group aggregation model, modeling method and temperature control load group adjustable potential evaluation method
CN115051374A (en) Control method and device for electric heating equipment to participate in electric power peak regulation and storage medium
Michailidis et al. Improving energy savings and thermal comfort in large-scale buildings via adaptive optimization
Teo et al. Energy management controls for chiller system: A review
CN114234370B (en) Multi-split air conditioner control method and device and multi-split air conditioner
Weiss Energy flexibility and shiftable heating power of building components and technologies

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20221111