CN110007613A - Warming prediction method and system for heat storage type electric heater and storage medium - Google Patents
Warming prediction method and system for heat storage type electric heater and storage medium Download PDFInfo
- Publication number
- CN110007613A CN110007613A CN201910285191.3A CN201910285191A CN110007613A CN 110007613 A CN110007613 A CN 110007613A CN 201910285191 A CN201910285191 A CN 201910285191A CN 110007613 A CN110007613 A CN 110007613A
- Authority
- CN
- China
- Prior art keywords
- electric heater
- data
- heat quantity
- heat storage
- type electric
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000005338 heat storage Methods 0.000 title claims abstract description 34
- 238000010792 warming Methods 0.000 title abstract description 4
- 238000010438 heat treatment Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 28
- 238000009825 accumulation Methods 0.000 claims description 26
- 238000012417 linear regression Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 2
- 239000002699 waste material Substances 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 13
- 230000015654 memory Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000005485 electric heating Methods 0.000 description 5
- 230000001537 neural effect Effects 0.000 description 5
- 239000012141 concentrate Substances 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 230000007787 long-term memory Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24D—DOMESTIC- OR SPACE-HEATING SYSTEMS, e.g. CENTRAL HEATING SYSTEMS; DOMESTIC HOT-WATER SUPPLY SYSTEMS; ELEMENTS OR COMPONENTS THEREFOR
- F24D11/00—Central heating systems using heat accumulated in storage masses
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24D—DOMESTIC- OR SPACE-HEATING SYSTEMS, e.g. CENTRAL HEATING SYSTEMS; DOMESTIC HOT-WATER SUPPLY SYSTEMS; ELEMENTS OR COMPONENTS THEREFOR
- F24D13/00—Electric heating systems
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24D—DOMESTIC- OR SPACE-HEATING SYSTEMS, e.g. CENTRAL HEATING SYSTEMS; DOMESTIC HOT-WATER SUPPLY SYSTEMS; ELEMENTS OR COMPONENTS THEREFOR
- F24D19/00—Details
- F24D19/10—Arrangement or mounting of control or safety devices
- F24D19/1096—Arrangement or mounting of control or safety devices for electric heating systems
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2614—HVAC, heating, ventillation, climate control
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B10/00—Integration of renewable energy sources in buildings
- Y02B10/70—Hybrid systems, e.g. uninterruptible or back-up power supplies integrating renewable energies
Abstract
The embodiment of the invention provides a warming prediction method, a warming prediction system and a storage medium for a heat storage type electric heater, wherein the method comprises the steps of collecting meteorological data of a period of time and collecting related data of the electric heater of the period of time through a sensor arranged on the heat storage type electric heater; extracting characteristic data from the collected electric heater related data and meteorological data; estimating the heating consumption of the next day by using the extracted characteristic data and a pre-trained heating prediction model; and setting the heat storage duration of the heat storage type electric heater according to the estimated heating consumption of the next day. According to the technical scheme of the embodiment of the invention, the user heating demand and the heat storage time can be effectively predicted, and the heat storage time can be reasonably set, so that the situations of energy waste or insufficient heating are avoided as much as possible under the condition of ensuring the daily heating demand of the user.
Description
Technical field
The present invention relates to the use of smart home field, more particularly, to heat storage type electric heater to warm up prediction technique and system.
Background technique
Heat-storage electrical heater is a kind of novel electric heating equipment, and the low-price electricity that night electricity company provides is converted to thermal energy
And store, so that whole day is with warming up, to realize power grid power load peak load shifting, save the effect of user heating expense.
However, user can only empirically manually set the heat accumulation time at night, usually in existing heat-storage electrical heater use process
Energy waste caused by due to laying in heat and being more than user demand is also possible to because of the shortage that heats caused by laying in shortage of heat
Problem.This is heat-storage electrical heater industry urgent problem now.
Summary of the invention
Therefore, the defect for aiming to overcome that the above-mentioned prior art of the embodiment of the present invention provides a kind of for heat storage type
The use of electric heater warms up prediction technique, system and storage medium, common warm in the day for ensureing user by the way that the heat accumulation time is rationally arranged
The case where energy waste or chillout are avoided while demand as far as possible.
Above-mentioned purpose is achieved through the following technical solutions:
According to a first aspect of the embodiments of the present invention, a kind of warm prediction technique of the use for heat storage type electric heater is provided,
This method comprises: when the meteorological data of acquisition a period of time and one section of the acquisition of the sensor by being arranged on heat storage type electric heater
Between electric heater related data；Characteristic is extracted from electric heater related data collected and meteorological data；Using being mentioned
The characteristic taken estimates next day consumed heat quantity with warm prediction model with trained in advance；And according to estimated next
Its consumed heat quantity sets the heat accumulation duration of the heat storage type electric heater.
In some embodiments of the invention, meteorological data collected may include the time, outdoor temperature, outside humidity and
Wind-force size；Electric heater related data collected includes acquisition time, air outlet temperature, air outlet wind-force size, Indoor Temperature
Degree and indoor humidity.
In some embodiments of the invention, extracted characteristic may include date, interior from the data of acquisition
Temperature, indoor humidity, outdoor temperature, outside humidity, wind-force size, daily consumed heat quantity.
In some embodiments of the invention, trained in advance that two-way shot and long term memory mind can be used with warm prediction model
Through network model.
In some embodiments of the invention, for train the sample characteristics with warm prediction model may include the date,
Room temperature, indoor humidity, outdoor temperature, outside humidity, wind-force size, daily consumed heat quantity.
In some embodiments of the invention, for training the sample characteristics collection with warm prediction model can be by following
Step constructs:
The meteorological data and electric heater related data in preset a period of time are acquired, when acquired meteorological data includes
Between, outdoor temperature, outside humidity and wind-force size, electric heater related data collected include acquisition time, air outlet temperature,
Air outlet wind-force size, room temperature and indoor humidity；
Sample characteristics are constructed as unit of day based on data collected, wherein for room temperature, indoor humidity, room
Outer temperature, outside humidity and wind-force size calculate the average value in 24 hours as characteristic value；And for daily consumed heat quantity,
First calculate the temperature difference of air outlet temperature and room temperature in each data collection cycle, if the temperature difference be greater than 0, by the temperature difference with it is right
The product for the air outlet wind-force size answered represents the heating amount in this collection period, ignores if the temperature difference is less than 0；
Then the heating amount in each collection period in 24 hours is mutually added up and is obtained and as daily this sample characteristics of consumed heat quantity
Value.
In some embodiments of the invention, the step of building sample data set, which may also include, concentrates sample characteristics
Each sample characteristics be normalized so that each sample characteristics is fallen in [0,1].
In some embodiments of the invention, the next day consumed heat quantity estimated by sets the heat storing type electric
The heat accumulation duration of warmer can include: next day estimated consumed heat quantity is input to predetermined instruction day consumed heat quantity and day stores up
The unary linear regression equation of linear relationship between hot duration is come heat accumulation duration needed for next day is calculated；It is wherein described
The unary linear regression equation for indicating the linear relationship between day consumed heat quantity and day heat accumulation duration is by least square method and base
In in acquisition a period of time collected daily consumed heat quantity and user daily heat accumulation duration determine.
According to a second aspect of the embodiments of the present invention, it additionally provides and warms up forecasting system, packet for the use of heat storage type electric heater
Include data acquisition module, model training module, prediction module, control module and communication module.Wherein data acquisition module is used for
Acquisition meteorological data and the acquisition electric heater related data of the sensor by being arranged on heat storage type electric heater；Model training module
For being trained by data collecting module collected interior for a period of time meteorological data and electric heater related data with warm prediction mould
Type；Prediction module is used to extract characteristic from the electric heater related data and meteorological data of data collecting module collected, and
Next day use is estimated with warm prediction model using extracted characteristic and via model training module is trained in advance
Warm amount；Control module is used to set the heat accumulation duration of the heat storage type electric heater according to next day estimated consumed heat quantity.
According to a third aspect of the embodiments of the present invention, a kind of computer readable storage medium is provided, meter is stored thereon with
Calculation machine program, described program are performed the method realized as described in above-described embodiment first aspect.
The technical solution of the embodiment of the present invention can include the following benefits:
Using deep learning model based on device senses data in the past period and environmental data to heat storage type electric heating
The heating trend of device is predicted, accurate to estimate that user uses warm demand and heat accumulation duration, and heat accumulation duration can rationally be arranged, from
And while day for ensureing user common warm demand, the case where avoiding energy waste or chillout as far as possible.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
It can the limitation present invention.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention
Example, and be used to explain the principle of the present invention together with specification.It should be evident that the accompanying drawings in the following description is only the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.In the accompanying drawings:
Fig. 1 shows the process that the use according to an embodiment of the invention for heat storage type electric heater warms up prediction technique and shows
It is intended to.
Fig. 2 shows the training according to an embodiment of the invention flow diagrams of the method for warm prediction model.
Fig. 3 shows a typical LSTM neural unit structural schematic diagram.
The use that Fig. 4 shows another embodiment according to the present invention warms up prediction model training flow diagram.
Fig. 5 shows weight distribution process schematic according to an embodiment of the invention.
Fig. 6 shows the structure that the use according to an embodiment of the invention for heat storage type electric heater warms up forecasting system and shows
It is intended to.
Specific embodiment
In order to make the purpose of the present invention, technical solution and advantage are more clearly understood, and are passed through below in conjunction with attached drawing specific real
Applying example, the present invention is described in more detail.It should be appreciated that described embodiment is a part of the embodiments of the present invention, without
It is whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not in the case where making creative work
The every other embodiment obtained, shall fall within the protection scope of the present invention.
In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner
In example.In the following description, many details are provided to provide and fully understand to the embodiment of the present invention.However,
It will be appreciated by persons skilled in the art that technical solution of the present invention can be practiced without one or more in specific detail,
Or it can be using other methods, constituent element, device, step etc..In other cases, it is not shown in detail or describes known side
Method, device, realization or operation are to avoid fuzzy each aspect of the present invention.
Block diagram shown in the drawings is only functional entity, not necessarily must be corresponding with physically separate entity.
I.e., it is possible to realize these functional entitys using software form, or realized in one or more hardware modules or integrated circuit
These functional entitys, or these functional entitys are realized in heterogeneous networks and/or processor device and/or microcontroller device.
Flow chart shown in the drawings is merely illustrative, it is not necessary to including all content and operation/step,
It is not required to execute by described sequence.For example, some operation/steps can also decompose, and some operation/steps can close
And or part merge, therefore the sequence actually executed is possible to change according to the actual situation.
Fig. 1 gives a kind of use for heat storage type electric heater according to an embodiment of the invention and warms up prediction technique
Flow diagram.As shown in Figure 1, this method specifically includes that step S101) it acquires the meteorological data of a period of time and passes through setting
The electric heater related data of sensor acquisition a period of time on heat storage type electric heater；Step S102) from electric heating collected
Characteristic is extracted in device related data and meteorological data；Step S103) it trains using extracted characteristic and in advance
Next day consumed heat quantity is estimated with warm prediction model；Step S104) heat accumulation set according to next day consumed heat quantity of estimation
The heat accumulation duration of formula electric heater.
Wherein with warm prediction model for predicting the next day possible consumed heat quantity of user according to the collected data, so as to
The heat accumulation duration of electric heater is set according to the consumed heat quantity predicted.(Long Short-Term can be remembered using shot and long term
Memory, LSTM) neural network model or two-way LSTM neural network be used as with warm prediction model.The warm prediction model of training
Key be select suitably for training sample characteristics, select which sample characteristics be trained for predict it is accurate
Rate has direct influence.In one embodiment, sample characteristics may include the date, room temperature, indoor humidity, outdoor temperature,
Outside humidity, wind-force size, daily consumed heat quantity.Fig. 2 gives training according to an embodiment of the invention with warm prediction model
Method flow diagram.As shown in Fig. 2, being mainly trained by following step:
S21 the meteorological data and electric heater related data in preset a period of time) are acquired
Since the use of electric heater is concentrated mainly on heating season, current heating season number interior for a period of time can be acquired
According to, or acquire the related data of a upper heating season or preceding several heating seasons and carry out training pattern.It acquires in longer time section
Data favorably with the accuracy of model prediction.The data for needing to acquire include meteorological data and electric heater related data.Wherein gas
Image data can be obtained by network from external various source of meteorological data.Acquired meteorological data may include time, outdoor temp
Degree, outside humidity and wind-force size.Electric heater related data can by the sensor that is arranged on heat storage type electric heater come into
Row acquisition.Electric heater related data collected may include acquisition time, air outlet temperature, air outlet wind-force size, Indoor Temperature
Degree and indoor humidity.Wherein air outlet temperature and wind-force relevant data sources are in the temperature and wind-force that are deployed in electric heater air outlet
Sensor, room temperature and indoor humidity derive from the environmental sensor disposed in electric heater shell.The period of data acquisition can
To be configured according to actual needs, such as it is set as 1 minute, 5 minutes, 10 minutes, 30 minutes etc..
S22) sample characteristics collection is constructed based on data collected
Since what is predicted is that daily use warms up total amount, the interval of sample characteristics sampling can be set to one
It.For room temperature, indoor humidity, outdoor temperature, outside humidity and wind-force size these sample characteristics, value is one day
24 hours average value.For sample characteristics " daily consumed heat quantity ", air outlet temperature in each data collection cycle is calculated first
Indicate that electric heating equipment is heating, then by the temperature difference and corresponding air outlet if the temperature difference is greater than 0 with the temperature difference of room temperature
The product of wind-force size represents the heating amount in this collection period, indicates that electric heating equipment does not carry out if the temperature difference is less than 0
Heating can be neglected；Then the heating amount in each collection period in one day is mutually added up obtain the heating amount on the same day as
The value of sample characteristics " daily consumed heat quantity ".In this way, sample characteristics concentrate each sample by the date, room temperature, indoor humidity,
This seven characteristics of outdoor temperature, outside humidity, wind-force size and daily consumed heat quantity are constituted.
Furthermore, it is contemplated that each feature has different dimension and amplitude of variation in sample, neural network may be will affect
Learning effect can be normalized sample data In yet another embodiment, fall within each sample characteristics
[0,1] in.It is normalized for example, by using following formula:
Wherein S is raw sample data to be processed, Smax、SminRespectively original sample
The maximum value and minimum value of notebook data, SnormFor the sample data after normalized, make its value between 0 to 1.It is achieved in that
Sample characteristics collection actually contains multiple time series X1~XmThat is { X1(x1,x2,…xn)…Xm(x1,x2,…xn), wherein m
Indicate that sample characteristics concentrate sample size, n indicates the feature quantity that each sample includes, as mentioned above, the application's
It include 7 features, i.e. n=7 in embodiment.
S23 the warm prediction model of constructed sample characteristics collection training) is utilized
In one embodiment, warm prediction model is used using shot and long term Memory Neural Networks LSTM, using constructed
Sample characteristics collection learns and determines the parameters of the model.Sample characteristics collection is divided first, in accordance with 75%, 25% ratio
For training set and test set；Initial model is obtained by training set, then by test set correction model parameter, steps up mould
Type precision, using the revised model as the warm prediction model of use.The model includes input layer, hidden layer and output layer, wherein
Input layer includes 7 nodes, and corresponding above-mentioned seven sample characteristics, output layer includes 1 node, for export predicted it is daily
Consumed heat quantity.Hidden layer is made of multiple neural units, each neural unit by currently input and the output of last moment come
To current output.It include two states in hidden layer, i.e. hidden layer state h and location mode c, hidden layer state h is for short-term defeated
Enter very sensitive, and location mode c can be reserved for long-term state.These states pass through when transmitting between hidden layer each unit every time
Several controllable gates (forgeing door, input gate, candidate door, out gate), the memory of information and current information and something lost before can control
Forget degree.
Fig. 3 gives the example of a typical LSTM neural unit, includes that memory unit (is denoted as in the LSTM neuron
C), input gate it, forget door ftWith out gate ot, it is assumed that ctFor the state of the memory unit of moment t, then in moment t, it should
The calculating process of LSTM unit is as follows:
ft=σ (Wf*[ht-1, xt]+bf) (1)
it=σ (Wi*[ht-1, xt]+bi) (2)
ot=σ (wo*[ht-1, xt]+bo) (3)
ht=ot*tanh(ct) (6)
Wherein, W and b is respectively corresponding weight coefficient matrix and bias term；σ is sigmoid function, is referred to as mind
Threshold function table through network；Tanh is tanh activation primitive, can also be referred to as the activation primitive of neural network；Be by
The new candidate value vector of tanh layers of creation, it will be added with cell state；htFor the output valve of the neural unit.LSTM model instruction
Practice process usually using time-based back-propagation algorithm BPTT (Back Propagation Through Time), point
For four key steps:
A) output valve of LSTM unit is calculated according to above-mentioned propagated forward method ((1)-(6))；
B) error term of each LSTM unit is inversely calculated；
C) gradient of each weight is calculated according to corresponding error term；
D) weight is updated using the optimization algorithm based on gradient.
It can be according to the output of the information prediction subsequent time of previous moment using the LSTM model of above process training.So
And the sometimes output at current time is not only related with previous moment state, it is also related to future state.Such as continuously changing
Weather data in the case of, the weather data of moment v1 is not only related with historical weather data, also and in following a period of time
The trend correlation of weather data.That is, the output at current time is led to by previous moment input and subsequent input determination
More reasonable and accurate prediction result can be obtained in conjunction with previous moment state and future status information by crossing.In view of electric heater stores up
Dsc data also has the characteristics that such, it is preferable that it can be used as using two-way LSTM neural network model and use warm prediction model,
Middle forward direction LSTM network provides list entries by regular turn, i.e., from x1To xn, and input and forward direction hidden layer state (h → 1 ..., h
→n)；Backward LSTM network provides list entries by reversed order, i.e., from xnTo x1, and export reversed hidden layer state (h ← 1 ...,
h←n).Then each state final hidden layer state be by being obtained before concatenating to hidden layer state and reversed hidden layer state,
That is hi=[h → t；h←t].
The use that Fig. 4 shows another embodiment according to the present invention warms up prediction model training flow diagram, wherein the stream
Journey can be divided into four process layers, i.e., data analysis layer, weight distribution layer, model training layer and connect output layer entirely.Wherein count
Original temporal data (sample characteristics data i.e. described above) are normalized according to process layer, obtain normalized
Sample characteristics collection afterwards, i.e. { X1(x1,x2,…xn)…Xm(x1,x2,…xn), wherein m indicates that sample characteristics concentrate sample size,
N indicates the feature quantity that each sample includes, and includes 7 features, i.e. n=in embodiments herein as mentioned above
7.Weight distribution layer is increased in this embodiment, is used Automobile driving mechanism in weight distribution layer and is calculated previous hidden layer shape
For the attention weight of each feature under state, new feature is then obtained according to these weights.Fig. 5 gives weight distribution process
Signal, wherein for list entries from feature x1To feature xn, the calculation formula of weight distribution process is as follows:
Wherein,WUAnd QUIt is the parameter for needing to learn, xkIt is the k-th feature in list entries,Pay attention to
Power weight indicates xkThen feature influence power in timing node t ensures that all features are infused by softmax function
The sum of probability for power of anticipating is 1.Weight distribution in this way can select and the more relevant spy of prediction result from sample data
Levy xkLearning training is carried out, input isWhat expression extracted
With the stronger feature of prediction result correlation.In this way, in model training layer, based on the feature that weight distribution layer extracts, in the time
Neural network hidden layer state when node t is updated toWherein, f1It is LSTM unit, it can be by upper
Formula (1)-(6) are stated to be calculated,It is the feature vector after updating in t moment.By weight distribution layer before, so that mould
Type training layer can be concentrated on selectively and be predicted in more precisely relevant feature.Then, for example above in model training layer
Four step a)-d mentioned) it is trained, wherein use mean absolute error as loss function to calculate neural network
Reality output and desired output between difference.Full articulamentum includes a neuron, for export predicted it is often daily
Warm amount.
After the parameters with warm prediction model trained by above-mentioned training process, so that it may according to being acquired
Lower for user day consumed heat quantity of newest a period of time data predict.
Referring now to figure 1, more specifically, acquiring the meteorological data of nearest a period of time in step S101) and being existed by setting
Sensor on heat storage type electric heater acquires the electric heater related data of nearest a period of time.Wherein meteorological data can pass through network
It is obtained from external data source.The meteorological data for acquiring or obtaining may include current time, outdoor temperature, outside humidity and wind-force
Size.Electric heater related data collected may include acquisition time, air outlet temperature, air outlet wind-force size, room temperature
With indoor humidity.Wherein air outlet temperature and wind-force relevant data sources are in the temperature and wind-force that are deployed in electric heater air outlet library
Sensor, room temperature and indoor humidity derive from the environmental sensor disposed in electric heater shell.
Characteristic is extracted from electric heater related data collected and meteorological data in step S102).From being acquired
Data in extract the date, room temperature, indoor humidity, outdoor temperature, outside humidity, wind-force size and daily consumed heat quantity are made
For the characteristic for being predicted.
Next day is estimated with warm prediction model with trained in advance using extracted characteristic in step S103)
Consumed heat quantity.The characteristic obtained through step S102 input is trained in warm prediction model, and the output of the model is pre-
The next day consumed heat quantity surveyed.
The storage of the heat storage type electric heater is set according to the next day consumed heat quantity predicted with warm prediction model in step S104)
Hot duration.In this embodiment, relationship can be indicated by equation of linear regression between day consumed heat quantity and day heat accumulation duration, so
Daily consumed heat quantity and the daily heat accumulation duration of user in a period of time are acquired afterwards, are gone through by least square method using collected these
The daily heat accumulation duration of daily consumed heat quantity and user in history data determines regression equation coefficient, thus obtained day consumed heat quantity with
Linear relationship between day heat accumulation duration.The next day consumed heat quantity predicted can be converted into according to such linear relationship
Electric heater heat accumulation duration needed for next day.
Fig. 6 is the structural representation that forecasting system is warmed up for the use of heat storage type electric heater according to one embodiment of the invention
Figure.As shown in fig. 6, the system includes data acquisition module, model training module, memory module, prediction module, control module,
Communication module.Although the block diagram describes component in functionally separated mode, such description is exclusively for the purposes of illustration.
Component shown in figure can arbitrarily be combined or be divided into independent software, firmware and/or hardware component.Moreover, nothing
How to be combined or divided by such component, they can hold on same computing device or multiple computing devices
Row, plurality of computing device can be to be connected to the network by one or more.
In system shown in Fig. 6, data acquisition module acquires meteorological number from outside weather data source by communication module
According to, and electric heater related data is obtained from the sensor being mounted on electric heater, data collected are saved in memory module
In.Training module obtains data collected from memory module, and as explained above, is based on data structure collected
It builds sample characteristics collection and the warm prediction model of training, training result is also reside in memory module.When control module receives
When predictions request, notice prediction module starts to be predicted, prediction module is obtained according to memory module uses each of warm prediction model
Parameter, and meteorological data and electric heater related data are obtained by data acquisition module, feature is extracted from acquired data
Data are input to trained in advance use and warm up in prediction model, by the output of the model as the daily consumed heat quantity predicted.In advance
It surveys module and the daily consumed heat quantity predicted is returned into control module, control module as explained above turns daily consumed heat quantity
It is changed to heat accumulation duration, and is back to the electric heater for sending predictions request, to carry out subsequent setting.
In yet another embodiment of the present invention, a kind of computer readable storage medium is additionally provided, meter is stored thereon with
Calculation machine program or executable instruction, when the computer program or executable instruction are performed realization such as institute in previous embodiment
The technical solution stated, realization principle is similar, and details are not described herein again.In an embodiment of the present invention, computer-readable storage medium
Matter can be it is any can storing data and can by computing device read tangible medium.The reality of computer readable storage medium
Example include hard disk drive, network attached storage (NAS), read-only memory, random access memory, CD-ROM, CD-R,
CD-RW, tape and other optics or non-optical data storage device.Computer readable storage medium also may include being distributed in
Computer-readable medium in network coupled computer system, so as to store and execute computer program in a distributed manner or refer to
It enables.
For the ginseng of " each embodiment ", " some embodiments ", " one embodiment " or " embodiment " etc. in this specification
Examine reference is that the special characteristic in conjunction with described in the embodiment, structure or property are included at least one embodiment.Cause
This, phrase " in various embodiments ", " in some embodiments ", " in one embodiment " or " in embodiment " etc. exists
The appearance of each place not necessarily refers to identical embodiment in the whole instruction.In addition, special characteristic, structure or property can
To combine in any way as suitable in one or more embodiments.Therefore, in conjunction with shown in one embodiment or description
Special characteristic, structure or property can wholly or partly with the feature, structure or property of one or more other embodiments
It unlimitedly combines, as long as the combination is not non-logicality or cannot work.
The term of " comprising " and " having " and similar meaning is expressed in this specification, it is intended that covers non-exclusive packet
Contain, such as contains the process, method, system, product or equipment of a series of steps or units and be not limited to listed step
Rapid or unit, but optionally further comprising the step of not listing or unit, or optionally further comprising for these processes, side
Other intrinsic step or units of method, product or equipment."a" or "an" is also not excluded for multiple situations.In addition, the application
Each element in attached drawing is not necessarily drawn to scale just to schematically illustrate.
Although the present invention is described through the foregoing embodiment, the present invention is not limited to described here
Embodiment, without departing from the present invention further include made various changes and variation.
Claims (10)
1. a kind of use for heat storage type electric heater warms up prediction technique, comprising:
It acquires the meteorological data of a period of time and the sensor by being arranged on heat storage type electric heater acquires the electricity of a period of time
Warmer related data；
Characteristic is extracted from electric heater related data collected and meteorological data；
Next day consumed heat quantity is estimated with warm prediction model with trained in advance using extracted characteristic；
The heat accumulation duration of the heat storage type electric heater is set according to next day estimated consumed heat quantity.
2. according to the method described in claim 1, wherein meteorological data collected includes time, outdoor temperature, outside humidity
And wind-force size；Electric heater related data collected includes acquisition time, air outlet temperature, air outlet wind-force size, interior
Temperature and indoor humidity.
3. according to the method described in claim 1, wherein extracted characteristic includes date, interior from the data of acquisition
Temperature, indoor humidity, outdoor temperature, outside humidity, wind-force size, daily consumed heat quantity.
4. according to the method described in claim 1, wherein trained in advance remembered with warm prediction model using two-way shot and long term
Neural network model.
5. according to the method described in claim 1, wherein for train the sample characteristics with warm prediction model include the date,
Room temperature, indoor humidity, outdoor temperature, outside humidity, wind-force size, daily consumed heat quantity.
6. according to the method described in claim 5, being wherein to pass through for training the sample characteristics collection with warm prediction model
The following steps building:
The meteorological data and electric heater related data in preset a period of time are acquired, acquired meteorological data includes time, room
Outer temperature, outside humidity and wind-force size, electric heater related data collected include acquisition time, air outlet temperature, outlet air
One's intention as revealed in what one says power size, room temperature and indoor humidity；
Sample characteristics are constructed as unit of day based on data collected, wherein for room temperature, indoor humidity, outdoor temp
Degree, outside humidity and wind-force size calculate the average value in 24 hours as characteristic value；
For daily consumed heat quantity, the temperature difference of air outlet temperature and room temperature in each data collection cycle is first calculated, if warm
Difference is greater than 0, then the temperature difference is represented to the heating amount in this collection period with the product of corresponding air outlet wind-force size, if
The temperature difference is ignored less than 0；Then the heating amount in each collection period in 24 hours is mutually added up and is obtained and as daily
The value of this sample characteristics of consumed heat quantity.
7. according to the method described in claim 6, the step of building sample data set further includes concentrating to sample characteristics
Each sample characteristics is normalized, so that each sample characteristics is fallen in [0,1].
8. according to the method described in claim 1, the next day consumed heat quantity estimated by sets the heat storing type electric
The heat accumulation duration of warmer includes: that next day estimated consumed heat quantity is input to predetermined instruction day consumed heat quantity and day heat accumulation
The unary linear regression equation of linear relationship between duration is come heat accumulation duration needed for next day is calculated；
Wherein the unary linear regression equation of the linear relationship between the instruction day consumed heat quantity and day heat accumulation duration is by most
Small square law and based in acquisition a period of time collected daily consumed heat quantity and user daily heat accumulation duration determine.
9. a kind of use for heat storage type electric heater warms up forecasting system, comprising:
Data acquisition module, for acquiring meteorological data and the acquisition electric heater of the sensor by being arranged on heat storage type electric heater
Related data；
Model training module, for the meteorological data and electric heater related data interior for a period of time by data collecting module collected
To train with warm prediction model；
Prediction module, for extracting characteristic from the electric heater related data and meteorological data of data collecting module collected,
And next day is estimated with warm prediction model using extracted characteristic and via model training module is trained in advance
Consumed heat quantity；
Control module, for setting the heat accumulation duration of the heat storage type electric heater according to next day estimated consumed heat quantity.
10. a kind of computer readable storage medium, is stored thereon with computer program, described program is performed realization such as right
It is required that method described in 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910285191.3A CN110007613B (en) | 2019-04-10 | 2019-04-10 | Warming prediction method and system for heat storage type electric heater and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910285191.3A CN110007613B (en) | 2019-04-10 | 2019-04-10 | Warming prediction method and system for heat storage type electric heater and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110007613A true CN110007613A (en) | 2019-07-12 |
| CN110007613B CN110007613B (en) | 2022-04-29 |
Family
ID=67170799
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910285191.3A Active CN110007613B (en) | 2019-04-10 | 2019-04-10 | Warming prediction method and system for heat storage type electric heater and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110007613B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111023254A (en) * | 2019-12-23 | 2020-04-17 | 北京华远意通热力科技股份有限公司 | Refined control method and system for water temperature of heating system |
| CN113983532A (en) * | 2021-12-27 | 2022-01-28 | 中熵科技(北京)有限公司 | Photovoltaic intelligent heat supply system and heat supply method |
| CN114136021A (en) * | 2021-11-30 | 2022-03-04 | 中国电力工程顾问集团西北电力设计院有限公司 | Solar energy-ground source heat pump system control method and system combined with resource prediction |
| CN114498702A (en) * | 2022-01-26 | 2022-05-13 | 南京清电物联科技有限公司 | Micro-grid energy management method and system based on energy router |
| CN114879770A (en) * | 2022-04-29 | 2022-08-09 | 广东迅扬科技股份有限公司 | Constant temperature control method based on linear regression prediction |
| CN115560381A (en) * | 2022-12-06 | 2023-01-03 | 建科环能科技有限公司 | Intelligent group control electric heating system, method and equipment based on edge calculation |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4213032A (en) * | 1975-09-03 | 1980-07-15 | Danfoss A/S | Control system for charging and discharging an electric storage heater |
| JPH0674674A (en) * | 1992-08-26 | 1994-03-18 | Kyushu Electric Power Co Inc | Storage heat heating system |
| CN101021914A (en) * | 2006-03-22 | 2007-08-22 | 侯春海 | Heating ventilating and air conditioner load predicting method and system |
| CN101957011A (en) * | 2009-07-15 | 2011-01-26 | 大荣E&B株式会社 | Heat storage type heating device with external data collecting function and control method thereof |
| EP2702657A1 (en) * | 2011-04-27 | 2014-03-05 | Steffes Corporation | Energy storage device control |
| CN204629671U (en) * | 2015-05-07 | 2015-09-09 | 北京华源恒科蓄热电暖气有限公司 | Accumulated electric heater |
| CN106197763A (en) * | 2016-07-28 | 2016-12-07 | 国网北京市电力公司 | The determination method and apparatus of thermal store quantity of heat storage |
| CN206222440U (en) * | 2016-11-29 | 2017-06-06 | 任丘市富强采暖设备有限公司 | Energy-storage type electric heater |
| CN107239859A (en) * | 2017-06-05 | 2017-10-10 | 国网山东省电力公司电力科学研究院 | The heating load forecasting method of Recognition with Recurrent Neural Network is remembered based on series connection shot and long term |
| CN107749645A (en) * | 2017-09-26 | 2018-03-02 | 国网辽宁省电力有限公司 | A kind of method for controlling high-voltage large-capacity thermal storage heating device |
| CN107909220A (en) * | 2017-12-08 | 2018-04-13 | 天津天大求实电力新技术股份有限公司 | Electric heating load prediction method |
| CN108240679A (en) * | 2018-02-22 | 2018-07-03 | 烟台科创捷能机电工程有限公司 | A kind of heat supply method based on building heating load prediction, device and system |
| CN108564230A (en) * | 2018-04-28 | 2018-09-21 | 湖南红太阳新能源科技有限公司 | A kind of family distributed energy management method and system |
| CN108736515A (en) * | 2018-05-30 | 2018-11-02 | 国网电力科学研究院(武汉)能效测评有限公司 | Wind electricity digestion phase-change thermal storage station load prediction system and method based on neural network |
| CN108758765A (en) * | 2018-04-17 | 2018-11-06 | 天津欣顺科技有限公司 | A kind of the accumulated electric heater system and charging method of wireless charging |
| CN109028278A (en) * | 2018-07-17 | 2018-12-18 | 哈尔滨工业大学 | A kind of the area operation system and scheduling strategy of wind power heating |
| CN109214624A (en) * | 2017-07-01 | 2019-01-15 | 杭州慧橙科技有限公司 | A kind of energy storage capacity optimization method based on Monte Carlo method, apparatus and system |
| CN109377060A (en) * | 2018-10-29 | 2019-02-22 | 东北电力大学 | BP neural network electric heating equipment regulating power appraisal procedure based on similarity |
-
2019
- 2019-04-10 CN CN201910285191.3A patent/CN110007613B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4213032A (en) * | 1975-09-03 | 1980-07-15 | Danfoss A/S | Control system for charging and discharging an electric storage heater |
| JPH0674674A (en) * | 1992-08-26 | 1994-03-18 | Kyushu Electric Power Co Inc | Storage heat heating system |
| CN101021914A (en) * | 2006-03-22 | 2007-08-22 | 侯春海 | Heating ventilating and air conditioner load predicting method and system |
| CN101957011A (en) * | 2009-07-15 | 2011-01-26 | 大荣E&B株式会社 | Heat storage type heating device with external data collecting function and control method thereof |
| EP2702657A1 (en) * | 2011-04-27 | 2014-03-05 | Steffes Corporation | Energy storage device control |
| CN204629671U (en) * | 2015-05-07 | 2015-09-09 | 北京华源恒科蓄热电暖气有限公司 | Accumulated electric heater |
| CN106197763A (en) * | 2016-07-28 | 2016-12-07 | 国网北京市电力公司 | The determination method and apparatus of thermal store quantity of heat storage |
| CN206222440U (en) * | 2016-11-29 | 2017-06-06 | 任丘市富强采暖设备有限公司 | Energy-storage type electric heater |
| CN107239859A (en) * | 2017-06-05 | 2017-10-10 | 国网山东省电力公司电力科学研究院 | The heating load forecasting method of Recognition with Recurrent Neural Network is remembered based on series connection shot and long term |
| CN109214624A (en) * | 2017-07-01 | 2019-01-15 | 杭州慧橙科技有限公司 | A kind of energy storage capacity optimization method based on Monte Carlo method, apparatus and system |
| CN107749645A (en) * | 2017-09-26 | 2018-03-02 | 国网辽宁省电力有限公司 | A kind of method for controlling high-voltage large-capacity thermal storage heating device |
| CN107909220A (en) * | 2017-12-08 | 2018-04-13 | 天津天大求实电力新技术股份有限公司 | Electric heating load prediction method |
| CN108240679A (en) * | 2018-02-22 | 2018-07-03 | 烟台科创捷能机电工程有限公司 | A kind of heat supply method based on building heating load prediction, device and system |
| CN108758765A (en) * | 2018-04-17 | 2018-11-06 | 天津欣顺科技有限公司 | A kind of the accumulated electric heater system and charging method of wireless charging |
| CN108564230A (en) * | 2018-04-28 | 2018-09-21 | 湖南红太阳新能源科技有限公司 | A kind of family distributed energy management method and system |
| CN108736515A (en) * | 2018-05-30 | 2018-11-02 | 国网电力科学研究院(武汉)能效测评有限公司 | Wind electricity digestion phase-change thermal storage station load prediction system and method based on neural network |
| CN109028278A (en) * | 2018-07-17 | 2018-12-18 | 哈尔滨工业大学 | A kind of the area operation system and scheduling strategy of wind power heating |
| CN109377060A (en) * | 2018-10-29 | 2019-02-22 | 东北电力大学 | BP neural network electric heating equipment regulating power appraisal procedure based on similarity |
Non-Patent Citations (3)
| Title |
|---|
| C.K. JOTSHI: "Heat transfer characteristics of a high temperature sensible heat storage water heater using cast iron as a storage material", 《IECEC 96. PROCEEDINGS OF THE 31ST INTERSOCIETY ENERGY CONVERSION ENGINEERING CONFERENCE》 * |
| QU XIAOYUN: "Short-Term Prediction of Wind Power Based on Deep Long Short-Term Memory", 《 2016 IEEE PES ASIA-PACIFIC POWER AND ENERGY ENGINEERING CONFERENCE (APPEEC) 》 * |
| 杜赛: "高压大功率电能转换固体储能供暖技术研究", 《中国优秀硕士学位论文全文数据库 工程科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111023254A (en) * | 2019-12-23 | 2020-04-17 | 北京华远意通热力科技股份有限公司 | Refined control method and system for water temperature of heating system |
| CN111023254B (en) * | 2019-12-23 | 2020-10-30 | 北京华远意通热力科技股份有限公司 | Refined control method and system for water temperature of heating system |
| CN114136021A (en) * | 2021-11-30 | 2022-03-04 | 中国电力工程顾问集团西北电力设计院有限公司 | Solar energy-ground source heat pump system control method and system combined with resource prediction |
| CN114136021B (en) * | 2021-11-30 | 2023-08-22 | 中国电力工程顾问集团西北电力设计院有限公司 | Solar energy-ground source heat pump system control method and system combined with resource prediction |
| CN113983532A (en) * | 2021-12-27 | 2022-01-28 | 中熵科技(北京)有限公司 | Photovoltaic intelligent heat supply system and heat supply method |
| CN114498702A (en) * | 2022-01-26 | 2022-05-13 | 南京清电物联科技有限公司 | Micro-grid energy management method and system based on energy router |
| CN114879770A (en) * | 2022-04-29 | 2022-08-09 | 广东迅扬科技股份有限公司 | Constant temperature control method based on linear regression prediction |
| CN114879770B (en) * | 2022-04-29 | 2024-02-20 | 广东迅扬科技股份有限公司 | Constant temperature control method based on linear regression prediction |
| CN115560381A (en) * | 2022-12-06 | 2023-01-03 | 建科环能科技有限公司 | Intelligent group control electric heating system, method and equipment based on edge calculation |
| CN115560381B (en) * | 2022-12-06 | 2023-03-10 | 建科环能科技有限公司 | Intelligent group control electric heating system, method and equipment based on edge calculation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110007613B (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110007613A (en) | Warming prediction method and system for heat storage type electric heater and storage medium | |
| KR102212663B1 (en) | An apparatus for hvac system input power control based on target temperature and method thereof | |
| CN107120782B (en) | A kind of HVAC system control method based on multi-user's hot comfort data | |
| Korkas et al. | Grid-connected microgrids: Demand management via distributed control and human-in-the-loop optimization | |
| Moon et al. | Development of a thermal control algorithm using artificial neural network models for improved thermal comfort and energy efficiency in accommodation buildings | |
| Wang et al. | Building load forecasting using deep neural network with efficient feature fusion | |
| Dobbs et al. | Model predictive HVAC control with online occupancy model | |
| US20180225585A1 (en) | Systems and methods for prediction of occupancy in buildings | |
| WO2018161723A1 (en) | Power load forecasting system based on long short-term memory neural network | |
| CN104484715A (en) | Neural network and particle swarm optimization algorithm-based building energy consumption predicting method | |
| Houimli et al. | Short-term electric load forecasting in Tunisia using artificial neural networks | |
| CN107430388A (en) | Method and apparatus for controlling the environmental management system in building | |
| KR102194002B1 (en) | energy operation management system by using optimized physical learning model and machine learning methods | |
| Ozoh et al. | A comparative analysis of techniques for forecasting electricity consumption | |
| CN113902582A (en) | Building comprehensive energy load prediction method and system | |
| CN104823120B (en) | For the intelligent controller of environmental control system | |
| CN111582588B (en) | Building energy consumption prediction method based on triple convolution fusion GRU | |
| CN107330518A (en) | Energy management control method and system based on temperature adjustment load prediction | |
| JP3160693B2 (en) | Heat load prediction device | |
| Godahewa et al. | A generative deep learning framework across time series to optimize the energy consumption of air conditioning systems | |
| CN111898856A (en) | Extreme learning machine-based physical-data fusion building analysis method | |
| CN116989442A (en) | Central air conditioner load prediction method and system | |
| Palak et al. | Smart building energy management using deep learning based predictions | |
| KR101057430B1 (en) | How to predict the heat demand of district heating | |
| CN116362471A (en) | Flexible deep reinforcement learning building load demand response method considering energy storage participation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |