CN108320016A - A kind of building energy consumption short term prediction method - Google Patents
A kind of building energy consumption short term prediction method Download PDFInfo
- Publication number
- CN108320016A CN108320016A CN201810191237.0A CN201810191237A CN108320016A CN 108320016 A CN108320016 A CN 108320016A CN 201810191237 A CN201810191237 A CN 201810191237A CN 108320016 A CN108320016 A CN 108320016A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- data
- building
- model
- training
- 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
- 238000005265 energy consumption Methods 0.000 title claims abstract description 248
- 238000000034 method Methods 0.000 title claims abstract description 76
- 238000013135 deep learning Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 63
- 238000005286 illumination Methods 0.000 claims description 44
- 238000012360 testing method Methods 0.000 claims description 22
- 230000000694 effects Effects 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 14
- 210000002569 neuron Anatomy 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 6
- 230000002457 bidirectional effect Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 230000001174 ascending effect Effects 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012731 temporal analysis Methods 0.000 claims description 3
- 238000000700 time series analysis Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims 1
- 239000002994 raw material Substances 0.000 claims 1
- 238000004378 air conditioning Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000007177 brain activity Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a kind of building energy consumption short term prediction methods, including acquisition building subitem energy consumption historical data, determine and acquire the major influence factors historical data for influencing building energy consumption subitem prediction；It analyzes and determines and lighting energy consumption prediction model is built based on time series autoregression model；Build the energy consumption prediction model based on deep learning DBN networks；Subitem prediction air conditioning energy consumption, power energy consumption, special energy consumption.Beneficial effects of the present invention：A kind of building energy consumption short term prediction method provided by the invention can more accurately and efficiently predict each subitem energy consumption in building energy consumption.
Description
Technical Field
The invention relates to the technical field of building energy consumption prediction, in particular to a building energy consumption prediction method based on a time series AR model and a deep confidence network.
Background
With the annual increase of power consumption of buildings in recent years, the energy consumption of buildings has become a main object for supervision and modification of building energy conservation. The current main prediction methods are: multivariate linear regression method, artificial neural network method, Bayes theory, grey theory method, etc. The prediction methods mostly belong to shallow structure algorithms, the learning effect of complex nonlinear relations in high-dimensional data samples is poor, deep learning is an algorithm for simulating human brain activities to analyze, learning is performed step by step from shallow to deep, the network topological structure is continuously optimized, and ideal learning parameters are selected, so that the problem that the training effect of the shallow structure algorithms in a multi-hidden-layer network is not ideal is effectively solved. Unlike many other machine learning methods, deep learning can automatically learn data effective characteristics from a large amount of non-identification historical data, and has strong data classification recognition and data prediction capabilities.
Deep Belief Network (DBN) is a learning model with the most extensive application in deep learning, and the combination of time-series Autoregressive (AR) model and DBN network to predict the subentry energy consumption of large buildings is not yet involved.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The invention is provided in view of the problems of the existing building energy consumption short-term prediction method.
Therefore, the invention aims to provide a method for predicting the energy consumption of each item in the building more accurately and effectively.
In order to solve the technical problems, the invention provides the following technical scheme: a short-term building energy consumption prediction method comprises the steps of collecting historical building energy consumption data, determining and collecting historical data of main influence factors influencing building energy consumption prediction, wherein the main influence factors comprise historical building illumination energy consumption data;
dividing collected historical building illumination energy consumption data into input data and verification data, analyzing and determining an illumination energy consumption prediction model constructed on the basis of a time series autoregressive model, taking the input data as input parameters of the prediction model, predicting illumination energy consumption in a short term and verifying results through the verification data;
comprehensively analyzing according to the historical building subentry energy consumption data, establishing a building subentry energy consumption database, then carrying out normalization processing on the building subentry energy consumption database and the collected main influence factors of the building subentry energy consumption, finally dividing the preprocessed data into training data and testing data, and constructing an energy consumption prediction model based on a deep learning DBN network through training and testing by utilizing the training data and the testing data;
and (3) the lighting energy consumption predicted by the time series model and the actually monitored main influence factors are taken as input parameters of the DBN model after training, and the air conditioner energy consumption, the power energy consumption and the special energy consumption are predicted in terms.
As a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the building energy consumption items comprise air conditioner energy consumption, power energy consumption and special energy consumption, and the main influence factors comprise seven characteristics of lighting energy consumption, outdoor average temperature, outdoor average humidity, weather characteristic value, holidays, average wind speed and 24 integral points in a day as main influence factors of building energy consumption item prediction.
As a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the illumination energy consumption prediction model predicts illumination energy consumption based on a time series analysis method, and the illumination energy consumption prediction model comprises definitions of illumination energy consumption autocorrelation coefficients and partial autocorrelation coefficients;
autocorrelation coefficient: given a 24 hour day value of energy consumption for illumination, the 7 th order difference value is [ df ]1,df2,L,df24]，dfi+k and dfiThe degree of linear dependence between is defined as:
wherein r (k) ═ Cov (df)i,dfi+k) As auto-covariance, Var (df)i) Is the variance, the k lag number is indicated. Since the variance of stationary time series is equal, so
Var(dfi+k)＝Var(dfi)＝L＝Var(0) (2)
Further, the variance defines that Var (0) ═ r (0) ═ σ2So the recursion of equation (1) is:
thus, p (k) can be calculated using recursion (3);
partial autocorrelation coefficient: given the middle k-1 random difference variables [ dfi,dfi+1Λ,dfi-k+1]Then df isi+kTo dfi-1The degree of correlation of (d) is defined as:
wherein ,
as a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the DBN network model comprises a DBN network model combined by an unsupervised Restricted Boltzmann Machine (RBM) and a supervised Back Propagation (BP) neural network to predict the energy consumption of the building subentry, wherein:
the RBM energy function, also called the expert product system, is defined as:
wherein n represents the number of visual elements and m represents the number of implicit elements;
under the energy function represented by equation (5), the probability that hidden layer neuron hj is activated is:
because the RBM is a bidirectional connection, the neurons in the visible layer can be activated by the neurons in the hidden layer, and the probability is as follows:
wherein S (x) is a tanh activation function, the output is between [ -1,1], which is equivalent to adjusting the average value of the input to 0 for subsequent processing, and the expression thereof
The neurons in the same layer have independence, so the probability density satisfies the independence, and the following formula is obtained:
and training the RBM in a layer-by-layer iteration mode to obtain a value of a learning parameter theta { W, b, c }, wherein bi is the bias of a visible node i, cj is the bias of a hidden layer node j, Wij is a connection matrix between the visible node i and the hidden layer node j, and the given training data is fitted through the learning parameter.
Training the training data by adopting a random gradient ascending method to maximize a log-likelihood function, wherein the updating formula of the final weight is as follows:
VWij＝η(<vihj>data-<vihj>recon) (10)
Vbi＝η(<vi>data-<vi>recon) (11)
Vcj＝η(<hj>data-<hj>recon) (12)
wherein, η is the learning rate,<～>datain order to train the distribution defined for the sample,<～>reconthe distribution defined for the sample after model reconstruction.
As a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the normalization process includes the steps of,
and (4) converting the building energy consumption data (including meteorological data) into the range of [ -1,1] for normalization processing. The normalized formula is:
wherein x' is normalized data, x is original input data of the sample, and xmin and xmax are minimum and maximum values of the original data.
As a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the deep belief network structure further includes the training step of the restricted boltzmann machine:
inputting the history data after the normalization preprocessing into a first RBM to start the unsupervised training, determining the weight and the bias of the history data, outputting the first RBM as the input of a second RBM, training the second RBM, repeating the training of three RBMs in sequence, repeating the training for multiple times, and realizing the initialization of model parameters; secondly, a traditional BP neural network supervision type learning mode is used, errors are transmitted to each layer of RBM from top to bottom in a back propagation mode, model parameters of all RBMs are adjusted, and the DBN can learn the intrinsic rule of complex data and is used for building a DBN model; and finally, predicting data by using the trained network.
As a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the deep belief network further comprises the training steps of:
data acquisition and preprocessing: collecting influence factors related to the building subentry energy consumption, wherein the influence factors comprise artificial activity factors and non-artificial activity factors, the artificial activity factors comprise the time of going to work and getting out of work, holidays, illumination and the like, and the non-artificial activity factors comprise the temperature, the humidity, the wind speed and the like;
comprehensively analyzing according to the historical building subentry energy consumption data, establishing a building subentry energy consumption database, then carrying out normalization processing together with the collected influence factors of the building subentry energy consumption, and finally dividing the preprocessed data into training data and test data;
constructing a DBN (direct bus network) prediction model, and optimally setting parameters of the DBN prediction model, wherein the DBN prediction model comprises the number of layers of hidden layers of an RBM (radial basis function) model, the number of nodes of the hidden layers of the BP network and the like;
training a DBN model by using training data, training by using a contrast divergence algorithm, and calculating errors of actual output and target output;
and inputting the test data into the trained DBN model for testing to obtain the prediction results of air conditioner energy consumption, power energy consumption and special energy consumption.
As a preferable scheme of the short-term prediction method for building energy consumption, the method comprises the following steps: the method also comprises the step of adopting the average absolute error (MAE) and the Root Mean Square Error (RMSE) as evaluation measurement values of the performance of the prediction model, wherein the MAE better reflects the actual situation of the error of the predicted value, and the RMSE reflects the dispersion degree of the error distribution. If the RMSE is smaller, the error is smaller, and the prediction effect is better. The expression is as follows:
where n denotes the number of samples, i denotes the prediction sequence, yiThe actual value of the ith sequence is represented,indicates the predicted value of the ith sequence.
The invention has the beneficial effects that: the short-term prediction method for the building energy consumption can more accurately and effectively predict each subentry energy consumption in the building energy consumption.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
FIG. 1 is a building energy consumption item prediction DBN network flow topology diagram of a building energy consumption short-term prediction method according to a first embodiment of the invention;
fig. 2 is a high-order difference diagram of illumination energy consumption of a short-term building energy consumption prediction method according to a first embodiment of the present invention;
fig. 3 is a view illustrating an RBM network structure of a short-term prediction method of building energy consumption according to a first embodiment of the present invention;
fig. 4 is a diagram of a deep confidence network structure of a short-term prediction method for building energy consumption according to a first embodiment of the present invention;
FIG. 5 is a diagram illustrating an autocorrelation coefficient analysis of a short-term method for predicting building energy consumption according to a second embodiment of the present invention;
FIG. 6 is a graph illustrating a partial autocorrelation coefficient analysis of a short-term building energy consumption prediction method according to a second embodiment of the present invention;
fig. 7 is a result of predicting lighting energy consumption by a short-term building energy consumption prediction method according to a second embodiment of the present invention;
fig. 8 is a comparison diagram of air conditioner energy consumption prediction of the short-term building energy consumption prediction method according to the second embodiment of the invention;
FIG. 9 is a comparison diagram of power energy consumption prediction of a short-term building energy consumption prediction method according to a second embodiment of the present invention;
FIG. 10 is a comparison diagram of specific energy consumption prediction of the short-term building energy consumption prediction method according to the second embodiment of the present invention;
fig. 11 is a comparison graph of absolute error of air conditioner energy consumption in a short-term prediction method of building energy consumption according to a second embodiment of the present invention;
FIG. 12 is a comparison graph of absolute errors of power consumption in a short-term prediction method of building energy consumption according to a second embodiment of the present invention;
fig. 13 is a diagram illustrating absolute error comparison of specific energy consumption for a short-term prediction method of building energy consumption according to a second embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
Furthermore, the present invention is described in detail with reference to the drawings, and in the detailed description of the embodiments of the present invention, the cross-sectional view illustrating the structure of the device is not enlarged partially according to the general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
In the embodiment, aiming at the problems that the consumption destination of the building energy consumption cannot be accurately distinguished and the prediction precision is low in the existing total energy consumption prediction method, the total energy consumption is divided into four main items according to the energy consumption application, and a novel method for the building energy consumption item-by-item short-term prediction is provided. Firstly, constructing a lighting energy consumption prediction method based on a time series Autoregressive (AR) model, and performing short-term prediction on lighting energy consumption of a building; secondly, an energy consumption prediction model based on a deep learning DBN is built, the air conditioner energy consumption, the power energy consumption (an elevator, pressurized water supply and the like) and the special energy consumption (a computer room, a kitchen and the like) of the building are predicted in terms, and the energy consumption prediction results are compared with the energy consumption prediction results of other network models. The result shows that the prediction method provided by the invention can more accurately and effectively predict each subentry energy consumption in the building energy consumption. Specifically, referring to fig. 1, the method includes the following steps:
acquiring historical building item energy consumption data, and determining and acquiring historical data of main influence factors influencing the prediction of the building energy consumption items, wherein the main influence factors comprise historical building illumination energy consumption data;
dividing collected historical building illumination energy consumption data into input data and verification data, analyzing and determining an illumination energy consumption prediction model constructed on the basis of a time series autoregressive model, taking the input data as input parameters of the prediction model, predicting illumination energy consumption in a short term and verifying results through the verification data;
comprehensively analyzing according to the historical building subentry energy consumption data, establishing a building subentry energy consumption database, then carrying out normalization processing on the building subentry energy consumption database and the collected main influence factors of the building subentry energy consumption, finally dividing the preprocessed data into training data and testing data, and constructing an energy consumption prediction model based on a deep learning DBN network through training and testing by utilizing the training data and the testing data;
and (3) the lighting energy consumption predicted by the time series model and the actually monitored main influence factors are taken as input parameters of the DBN model after training, and the air conditioner energy consumption, the power energy consumption and the special energy consumption are predicted in terms.
Here, it should be noted that: the illumination energy consumption reflects the number of office workers in a building, and the number of office workers directly influences the energy consumption of an air conditioner, the energy consumption of power (used by an elevator and used by secondary water supply) and the special energy consumption (such as electricity used by a machine room and electricity used by a kitchen). The outdoor average temperature is closely related to the energy consumption of the air conditioner, the 24 integral time is closely related to the number of office workers (on-duty time), and meanwhile, the weather characteristic value, holiday days and average wind speed can influence the energy consumption of the air conditioner, so that in conclusion of theoretical analysis, the 7 characteristics of illumination energy consumption, outdoor average temperature, outdoor average humidity, weather characteristic value, holiday days, average wind speed and the 24 integral time in one day are selected as main influence factors of the building energy consumption subentry prediction.
Because the illumination energy consumption is closely related to the number of office workers in the building and reflects the number of people, the number of office workers cannot be counted and is reflected only by the illumination energy consumption, and therefore the illumination energy consumption is predicted by using a time series autoregressive model. Therefore, the lighting energy consumption is used as an influence factor which is predicted by the input parameter. Therefore, the data required to be collected in the invention are the building energy consumption items including air conditioner energy consumption, power energy consumption and special energy consumption, and the main influence factors include seven characteristics of lighting energy consumption, outdoor average temperature, outdoor average humidity, weather characteristic value, holidays, average wind speed and 24 integral points in one day as the main influence factors of the building energy consumption item prediction.
Further, in the embodiment, the lighting energy consumption often shows a random fluctuation on the basis of the past energy consumption in a short time, and belongs to a non-stationary time sequence, but the high-order difference of the lighting energy consumption is stationary. If the value of the illumination energy consumption is [ x ] after 24 hours of collection for a certain day1,x2Λ,x24]＝[49.8,51.6,Λ,218.2](unit: Kwh) where xiRepresenting the value of the illumination energy consumption at the ith time, as shown in fig. 2, the 7 th order difference of the illumination energy consumption is displayed to be stable, and then the illumination energy consumption can be predicted by adopting a time series analysis method:
the invention uses a P-order Autoregressive (AR) model to construct a prediction model of the illumination energy consumption, and the definitions of the autocorrelation coefficient and the partial autocorrelation coefficient of the illumination energy consumption are given below so as to determine the value of the order P.
Definition 1: autocorrelation of the illumination energy consumption, given a 24-hour illumination energy consumption value of 7 orders of a dayDifferential value of [ df1,df2,L,df24]，dfi+k and dfiThe degree of linear dependence between is defined as:
wherein r (k) ═ Cov (df)i,dfi+k) As auto-covariance, Var (df)i) Is the variance, the k lag number is indicated. Since the variance of stationary time series is equal, so
Var(dfi+k)＝Var(dfi)＝L＝Var(0) (2)
Further, the variance defines that Var (0) ═ r (0) ═ σ2So the recursion of equation (1) is:
therefore, p (k) can be calculated by recursive formula (3).
Definition 2: partial autocorrelation coefficient of illumination energy consumption, given intermediate k-1 random difference variables [ dfi,dfi+1Λ,dfi-k+1]Then df isi+kTo dfi-1The degree of correlation of (d) is defined as:
wherein ,
in the above steps, the lighting energy consumption of the building is predicted by using the time series autoregressive model, the prediction model of the lighting energy consumption is constructed by using the P-order Autoregressive (AR) model, the collected input data is used as the input parameters of the prediction model, the lighting energy consumption is predicted in a short period, and the result is verified by verifying the data.
Based on the collected historical energy consumption data and the predicted illumination energy consumption value, an energy consumption prediction model based on a deep learning DBN network is constructed, and the data is used for training, testing and predicting, and the method specifically comprises the following steps:
the method comprises the following steps: deep belief network selection
In the embodiment, a DBN network model combining 3 unsupervised Restricted Boltzmann Machines (RBMs) and 1 supervised Back Propagation (BP) neural network is adopted to predict the energy consumption of the building sub-items.
As shown in FIG. 3, the RBM acts as a neural sensor. The method comprises a hidden layer h and a visible layer v, and full bidirectional connection is realized between the hidden layer and the visible layer by taking a connection matrix w as a connection weight.
The RBM energy function, also called the expert product system, can be defined as:
where n represents the number of visual elements and m represents the number of implicit elements.
Under the energy function represented by equation (5), the probability that hidden layer neuron hj is activated is:
because the RBM is a bidirectional connection, the neurons in the visible layer can be activated by the neurons in the hidden layer, and the probability is as follows:
wherein S (x) is a tanh activation function with an output of [ -1,1 [ ]]The mean value of the input is adjusted to 0 for subsequent processing, and the expression is
The neurons in the same layer have independence, so the probability density satisfies the independence, and the following formula is obtained:
and training the RBM in a layer-by-layer iteration mode to obtain a value of a learning parameter theta which is { W, b, c }, wherein bi is the bias of a visible node i, cj is the bias of an implicit layer node j, and Wij is a connection matrix between the visible node i and the implicit layer node j. Given training data is fitted by learning parameters.
Training the training data by adopting a random gradient ascending method to maximize a log-likelihood function, wherein the updating formula of the final weight is as follows:
VWij＝η(<vihj>data-<vihj>recon) (10)
Vbi＝η(<vi>data-<vi>recon) (11)
Vcj＝η(<hj>data-<hj>recon) (12)
wherein, η is the learning rate,<～>datain order to train the distribution defined for the sample,<～>reconthe distribution defined for the sample after model reconstruction.
Step two: preprocessing, i.e. normalisation, of data
In order to enable data to be in a uniform scale and prevent modeling ill-condition problems caused by different orders of magnitude, the invention converts all building energy consumption data (including meteorological data) into the range of [ -1,1] to carry out normalization processing, and preprocessed original data comprise 7 influence factors and one-year-history energy consumption values of air conditioners, power and special energy consumption. The normalized formula is:
wherein x' is normalized data, x is original input data of the sample, and xmin and xmax are minimum and maximum values of the original data.
Step two: deep belief network architecture
The step is the training of the RBM (limited boltzmann machine), specifically, as shown in fig. 4, firstly inputting the preprocessed original data into a first RBM to start the unsupervised training, determining the weight and the bias of the first RBM, outputting the first RBM as the input of a second RBM, training the second RBM, repeating the training for 3 RBMs in sequence, repeating the training for many times, and realizing the initialization of the model parameters; secondly, a traditional BP neural network supervision type learning mode is used, errors are transmitted to each layer of RBM from top to bottom in a back propagation mode, model parameters of all RBMs are adjusted, and the DBN can learn the intrinsic rule of complex data and is used for building a DBN model; and finally, predicting data by using the trained network. It is easy to see that the DBN training comprises two 2 processes of unsupervised pre-training and supervised fine tuning, change characteristics among input data are automatically extracted based on strong nonlinear mapping capacity of the DBN, and test data are input into a DBN model to obtain a predicted value of the DBN.
Step three: deep confidence network training method
This step is the training of the DBN model, and referring again to fig. 1, the deep belief network training step is as follows:
data acquisition and preprocessing: collecting influence factors related to the building subentry energy consumption, wherein the influence factors comprise artificial activity factors and non-artificial activity factors, the artificial activity factors comprise the time of going to work and getting out of work, holidays, illumination and the like, and the non-artificial activity factors comprise the temperature, the humidity, the wind speed and the like;
comprehensively analyzing according to the historical building subentry energy consumption data, establishing a building subentry energy consumption database, then carrying out normalization processing together with the collected influence factors of the building subentry energy consumption, and finally dividing the preprocessed data into training data and test data;
constructing a DBN (direct bus network) prediction model, and optimally setting parameters of the DBN prediction model, wherein the parameters comprise the number of layers of hidden layers of the RBM model, the number of nodes of the hidden layers of the BP network and the like;
training a DBN model by using training data, training by using a contrast divergence algorithm in order to accelerate a training process, and calculating errors of actual output and target output;
and inputting the test data into the trained DBN model for testing to obtain the prediction results of air conditioner energy consumption, power energy consumption and special energy consumption.
Step four: evaluation metric of model performance
The Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE) are used as evaluation measurement values of model performance, the MAE better reflects the actual situation of predicted value errors, and the RMSE reflects the dispersion degree of error distribution. If the RMSE is smaller, the error is smaller, and the prediction effect is better. The expression is as follows:
wherein n represents the number of samples, and i represents the predicted orderColumn, yiThe actual value of the ith sequence is represented,indicates the predicted value of the ith sequence.
Example 2
In the embodiment, based on the first embodiment, a lighting energy consumption prediction method based on a time series Autoregressive (AR) model is constructed, and short-term prediction is performed on lighting energy consumption of a building; secondly, an energy consumption prediction model based on a deep learning DBN is built, air conditioner energy consumption, power energy consumption (elevators, pressurized water supply and the like) and special energy consumption (computer rooms, kitchens and the like) of a building are predicted in a itemized mode, and the built model is used for prediction and is compared with energy consumption prediction results of other network models. The result shows that the prediction method provided by the invention can more accurately and effectively predict each subentry energy consumption in the building energy consumption. Specifically, the method comprises the following steps:
the method comprises the following steps: short term prediction of lighting energy consumption
The method comprises the steps of collecting the hourly illumination energy consumption values of a whole year from 2016 to 8 to 6 to 2017 of a certain building, predicting by using a time series AR model, and selecting the illumination energy consumption values at the whole time of 168 whole points from 7 to 31 to 6 to 7 to 8.24 of 2017 to 7.
As can be seen from fig. 5, the process of the value attenuation of the autocorrelation coefficient to small value fluctuation is slow and continuous, and a long tail is dragged, so that the tailing is met; as can be seen from fig. 6, the values of the partial autocorrelation coefficients are damped, suddenly converge to the critical value 0 when the hysteresis order is 6, and the values of the partial autocorrelation coefficients are all 0 after 6 orders, which is consistent with the truncation, so that the order P of the autoregressive model AR (P) is 6, and it is further verified that the lighting energy consumption predictor is consistent with the autoregressive model AR (6).
As shown in fig. 7, the curve of the predicted value and the actual value of the lighting energy consumption at the time of the whole hour from 7/31/8/6/2017 is high in fitting degree with the curve of the actual value, the average absolute error MAE is 9.9591, and the root mean square error RMSE is 13.5182, so that the expected prediction effect is achieved.
Step two: short term prediction of air conditioning, power and specific energy consumption
The experimental sample data is derived from the itemized monitoring data of air conditioner energy consumption, power energy consumption and special energy consumption of a whole year from 2016, 8 and 7 days to 2017, 8 and 6 days of a certain building. Wherein 7296 group data of 10 months per hour from 2016, 8/7/6/2017, 6/2017 is used as training data, and 1296 group data of 2017, 6/7/30/2017 is used as test data.
And (3) establishing a DBN model to predict air-conditioning energy consumption, power energy consumption and special energy consumption from 31 days in 7 and 31 days in 2017 to 6 days in 8 and 6 days in 2017 by taking the illumination energy consumption predicted by the time series model and the actually monitored outdoor average temperature, outdoor average humidity, weather characteristic value, holidays, average wind speed and 24 integral points in one day as input parameters.
For the DBN network model, the larger the number of hidden layer and hidden layer neurons, the better the performance of the algorithm, but the computational complexity also increases. Therefore, under the comprehensive consideration of algorithm performance and calculated amount, the number of RBM layers is set to be 3, the number of neurons in hidden layers of each layer is 400, 200 and 100, the number of iterations is 200, and the learning rate is 0.001; when the back propagation fine tuning is carried out, the learning rate is changed to 0.1, and the iteration number is 500.
Step three: comparison of predicted results
Based on the same historical data, the data predicted by the building energy consumption prediction model provided by the invention and the existing 2 building energy consumption prediction models are as follows: and comparing the data predicted based on the BP neural network and the iPSO-BP neural network. The predicted and actual values of energy consumption for one week are selected and compared with the 3 models, for example, as shown in fig. 8, 9 and 10.
As can be seen from fig. 8, 9 and 10, the predicted value curve of the DBN model has better fitting degree with the actual value curve than the predicted value curve of the iPSO-BP model and the BP model. To further verify the predicted results, 12 sets of data were randomly selected from fig. 8, 9 and 10, and the predicted values and actual values of the three models were compared, respectively, as shown in table 1. The prediction precision of the DBN model for the building energy consumption subentry prediction is higher than that of the iPSO-BP model and the BP model.
TABLE 1 comparison of predicted and actual values of energy consumption for DBN model, iPSO-BP model and BP model
Step four: air conditioner, power and special energy consumption prediction result analysis
The absolute error pairs of the DBN model, the iPSO-BP model and the BP neural network model for the air conditioner energy consumption, the power energy consumption and the special energy consumption prediction are shown in the figures 11, 12 and 13.
It can be seen from fig. 11, 12 and 13 that the prediction errors of the DBN model, the iPSO-BP model and the BP model fluctuate around the value of 0, but the prediction error of the DBN model fluctuates less than those of the iPSO-BP model and the BP model, and is relatively stable. Therefore, the accuracy and stability of the DBN model in the training and learning process are higher.
According to the performance indexes of the formula (14) and the formula (15), the test data are analyzed and compared respectively, and the comparison results of the 3 models to the three large energy consumption errors are shown in table 2. The average absolute errors of the BP model for air conditioner, power and special energy consumption prediction are 36.2153, 9.0127 and 6.6296 respectively, the average absolute errors of the iPSO-BP model for air conditioner, power and special energy consumption prediction are 10.0822, 3.3848 and 2.7347 respectively, and the average absolute errors of the DBN model for air conditioner, power and special energy consumption prediction are 5.4707, 1.7916 and 1.6075 respectively; the root mean square errors of the BP model for air conditioner, power and special energy consumption prediction are 42.7680, 12.3035 and 8.3477 respectively, the root mean square errors of the iPSO-BP model for air conditioner, power and special energy consumption prediction are 14.2698, 5.1325 and 2.9536 respectively, and the average relative errors of the DBN model for air conditioner, power and special energy consumption prediction are 8.6575, 2.9536 and 2.2178 respectively. It can be seen that the average absolute error and the root mean square error of the DBN model energy consumption prediction are smaller than those of the iPSO-BP model prediction and are much smaller than those of the BP model, which shows that the DBN prediction model provided by the invention has better effect on building energy consumption subentry prediction and improves the accuracy of building energy consumption prediction.
TABLE 2 comparison of performance indexes for energy consumption prediction of DBN model, iPSO-BP model and BP model
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (8)
1. A short-term prediction method for building energy consumption is characterized by comprising the following steps: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
acquiring historical building item energy consumption data, and determining and acquiring historical data of main influence factors influencing the prediction of the building energy consumption items, wherein the main influence factors comprise historical building illumination energy consumption data;
dividing collected historical building illumination energy consumption data into input data and verification data, analyzing and determining an illumination energy consumption prediction model constructed on the basis of a time series autoregressive model, taking the input data as input parameters of the prediction model, predicting illumination energy consumption in a short term and verifying results through the verification data;
comprehensively analyzing according to the historical building subentry energy consumption data, establishing a building subentry energy consumption database, then carrying out normalization processing on the building subentry energy consumption database and the collected main influence factors of the building subentry energy consumption, finally dividing the preprocessed data into training data and testing data, and constructing an energy consumption prediction model based on a deep learning DBN network through training and testing by utilizing the training data and the testing data;
and (3) the lighting energy consumption predicted by the time series model and the actually monitored main influence factors are taken as input parameters of the DBN model after training, and the air conditioner energy consumption, the power energy consumption and the special energy consumption are predicted in terms.
2. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the building energy consumption items comprise air conditioner energy consumption, power energy consumption and special energy consumption, and the main influence factors comprise seven characteristics of lighting energy consumption, outdoor average temperature, outdoor average humidity, weather characteristic value, holidays, average wind speed and 24 integral points in a day as main influence factors of building energy consumption item prediction.
3. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the illumination energy consumption prediction model predicts illumination energy consumption based on a time series analysis method, and the illumination energy consumption prediction model comprises definitions of illumination energy consumption autocorrelation coefficients and partial autocorrelation coefficients;
autocorrelation coefficient: given a 24 hour day value of energy consumption for illumination, the 7 th order difference value is [ df ]1,df2,L,df24]，dfi+k and dfiThe degree of linear dependence between is defined as:
wherein r (k) ═ Cov (df)i,dfi+k) As auto-covariance, Var (df)i) K represents the number of lags for variance. Due to the stabilityThe variance of the time series is equal, so
Var(dfi+k)＝Var(dfi)＝L＝Var(0) (2)
Further, the variance defines that Var (0) ═ r (0) ═ σ2So the recursion of equation (1) is:
thus, p (k) can be calculated using recursion (3);
partial autocorrelation coefficient: given the middle k-1 random difference variables [ dfi,dfi+1Λ,dfi-k+1]Then df isi+kTo dfi-1The degree of correlation of (d) is defined as:
wherein , is the partial autocorrelation coefficient, D (k) is the variance with lag order k, D is the total variance, and ρkIs the autocorrelation coefficient with lag order k.
4. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the DBN network model comprises a DBN network model combined by an unsupervised Restricted Boltzmann Machine (RBM) and a supervised Back Propagation (BP) neural network to predict the energy consumption of the building subentry, wherein:
the RBM energy function, also called the expert product system, is defined as:
wherein n represents the number of visual elements and m represents the number of implicit elements;
under the energy function represented by equation (5), the probability that hidden layer neuron hj is activated is:
because the RBM is a bidirectional connection, the neurons in the visible layer can be activated by the neurons in the hidden layer, and the probability is as follows:
wherein S (x) is a tanh activation function, the output is between [ -1,1], which is equivalent to adjusting the average value of the input to 0 for subsequent processing, and the expression thereof
The neurons in the same layer have independence, so the probability density satisfies the independence, and the following formula is obtained:
and training the RBM in a layer-by-layer iteration mode to obtain a value of a learning parameter theta { W, b, c }, wherein bi is the bias of a visible node i, cj is the bias of a hidden layer node j, Wij is a connection matrix between the visible node i and the hidden layer node j, and the given training data is fitted through the learning parameter.
Training the training data by adopting a random gradient ascending method to maximize a log-likelihood function, wherein the updating formula of the final weight is as follows:
VWij＝η(<vihj>data-<vihj>recon) (10)
Vbi＝η(<vi>data-<vi>recon) (11)
Vcj＝η(<hj>data-<hj>recon) (12)
wherein, η is the learning rate,<～>datain order to train the distribution defined for the sample,<～>reconthe distribution defined for the sample after model reconstruction.
5. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the normalization process includes the steps of,
and (4) converting the building energy consumption data (including meteorological data) into the range of [ -1,1] for normalization processing. The normalized formula is:
wherein x' is normalized data, x is original input data of the sample, and xmin and xmax are minimum and maximum values of the original data.
6. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the deep belief network structure further includes the training step of the restricted boltzmann machine:
inputting the history data after the normalization preprocessing into a first RBM to start the unsupervised training, determining the weight and the bias of the history data, outputting the first RBM as the input of a second RBM, training the second RBM, repeating the training of three RBMs in sequence, repeating the training for multiple times, and realizing the initialization of model parameters; secondly, a traditional BP neural network supervision type learning mode is used, errors are transmitted to each layer of RBM from top to bottom in a back propagation mode, model parameters of all RBMs are adjusted, and the DBN can learn the intrinsic rule of complex data and is used for building a DBN model; and finally, predicting data by using the trained network.
7. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the deep belief network further comprises the following DBN network training steps:
data acquisition and preprocessing: collecting influence factors related to the building subentry energy consumption, wherein the influence factors comprise artificial activity factors and non-artificial activity factors, the artificial activity factors comprise the time of going to work and getting out of work, holidays, illumination and the like, and the non-artificial activity factors comprise the temperature, the humidity, the wind speed and the like;
comprehensively analyzing according to the historical building subentry energy consumption data, establishing a building subentry energy consumption database, then carrying out normalization processing together with the collected influence factors of the building subentry energy consumption, and finally dividing the preprocessed data into training data and test data;
constructing a DBN (direct bus network) prediction model, and optimally setting parameters of the DBN prediction model, wherein the DBN prediction model comprises the number of layers of hidden layers of an RBM (radial basis function) model, the number of nodes of the hidden layers of the BP network and the like;
training a DBN model by using training data, training by using a contrast divergence algorithm, and calculating errors of actual output and target output;
and inputting the test data into the trained DBN model for testing to obtain the prediction results of air conditioner energy consumption, power energy consumption and special energy consumption.
8. The short-term prediction method of building energy consumption as claimed in claim 1, characterized in that: the method also comprises the step of adopting the average absolute error (MAE) and the Root Mean Square Error (RMSE) as evaluation measurement values of the performance of the prediction model, wherein the MAE better reflects the actual situation of the error of the predicted value, and the RMSE reflects the dispersion degree of the error distribution. If the RMSE is smaller, the error is smaller, and the prediction effect is better. The expression is as follows:
whereinN denotes the number of samples, i denotes the prediction sequence, yiThe actual value of the ith sequence is represented,indicates the predicted value of the ith sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810191237.0A CN108320016B (en) | 2018-03-08 | 2018-03-08 | Short-term prediction method for building energy consumption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810191237.0A CN108320016B (en) | 2018-03-08 | 2018-03-08 | Short-term prediction method for building energy consumption |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108320016A true CN108320016A (en) | 2018-07-24 |
| CN108320016B CN108320016B (en) | 2023-09-19 |
Family
ID=62900506
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810191237.0A Active CN108320016B (en) | 2018-03-08 | 2018-03-08 | Short-term prediction method for building energy consumption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108320016B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118020A (en) * | 2018-09-06 | 2019-01-01 | 南京工业职业技术学院 | A kind of subway station energy consumption short term prediction method and its forecasting system |
| CN109214572A (en) * | 2018-09-06 | 2019-01-15 | 南京工业职业技术学院 | One kind is based on EPSO-BP to commercial building lamp decoration energy consumption prediction technique |
| CN109255472A (en) * | 2018-08-20 | 2019-01-22 | 天津科技大学 | A kind of Office Building Energy Consumption prediction technique based on Multivariate adaptive regression splines batten |
| CN109376939A (en) * | 2018-11-01 | 2019-02-22 | 三峡大学 | A kind of grid stability real-time predicting method based on adaptive neural network |
| CN109582912A (en) * | 2018-11-19 | 2019-04-05 | 北京理工大学 | A kind of public building power consumption monitoring data method for parameter estimation |
| CN109695944A (en) * | 2018-11-29 | 2019-04-30 | 中国汽车工业工程有限公司 | A kind of control method of the coating fresh air conditioner based on multi-model deep learning |
| CN109934422A (en) * | 2019-04-25 | 2019-06-25 | 中国人民解放军国防科技大学 | Neural network wind speed prediction method based on time series data analysis |
| CN110009156A (en) * | 2019-04-08 | 2019-07-12 | 北京博锐尚格节能技术股份有限公司 | Building energy consumption budget management method and storage medium |
| CN110309560A (en) * | 2019-06-13 | 2019-10-08 | 清华大学 | Multi-layer Migration Simulation method and apparatus |
| CN110400018A (en) * | 2019-07-29 | 2019-11-01 | 上海电力大学 | Progress control method, system and device for coal-fired firepower electrical plant pulverized coal preparation system |
| CN110435009A (en) * | 2019-07-15 | 2019-11-12 | 中国科学院重庆绿色智能技术研究院 | A kind of Intelligentized design method of concrete production match ratio |
| CN110533243A (en) * | 2019-08-27 | 2019-12-03 | 西安建筑科技大学 | A kind of building energy consumption prediction technique based on fractal theory |
| CN110942228A (en) * | 2019-10-25 | 2020-03-31 | 万翼科技有限公司 | Energy consumption control method and device, computer equipment and storage medium |
| CN111199305A (en) * | 2018-11-20 | 2020-05-26 | 上海烟草集团有限责任公司 | Neural network-based production energy consumption prediction method and system, electronic terminal and storage medium |
| CN111753470A (en) * | 2020-06-30 | 2020-10-09 | 桂林电子科技大学 | Building energy consumption prediction method based on rough set and deep belief neural network |
| CN112712213A (en) * | 2021-01-15 | 2021-04-27 | 上海交通大学 | Method and system for predicting energy consumption of deep migration learning of centralized air-conditioning house |
| CN113344192A (en) * | 2021-05-31 | 2021-09-03 | 中国标准化研究院 | Enterprise-level motor system energy-saving optimization automatic control method and system |
| CN117077854A (en) * | 2023-08-15 | 2023-11-17 | 广州视声智能科技有限公司 | Building energy consumption monitoring method and system based on sensor network |
| CN117371624A (en) * | 2023-12-07 | 2024-01-09 | 邢台职业技术学院 | Building energy-saving transformation optimization management method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106874581A (en) * | 2016-12-30 | 2017-06-20 | 浙江大学 | A kind of energy consumption of air conditioning system in buildings Forecasting Methodology based on BP neural network model |
| CN106991504A (en) * | 2017-05-09 | 2017-07-28 | 南京工业大学 | Building energy consumption Forecasting Methodology, system and building based on metering separate time series |
-
2018
- 2018-03-08 CN CN201810191237.0A patent/CN108320016B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106874581A (en) * | 2016-12-30 | 2017-06-20 | 浙江大学 | A kind of energy consumption of air conditioning system in buildings Forecasting Methodology based on BP neural network model |
| CN106991504A (en) * | 2017-05-09 | 2017-07-28 | 南京工业大学 | Building energy consumption Forecasting Methodology, system and building based on metering separate time series |
Non-Patent Citations (1)
| Title |
|---|
| 史佳琪等: "基于深度结构多任务学习的园区型综合能源系统多元负荷预测", 《电网技术》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255472A (en) * | 2018-08-20 | 2019-01-22 | 天津科技大学 | A kind of Office Building Energy Consumption prediction technique based on Multivariate adaptive regression splines batten |
| CN109118020A (en) * | 2018-09-06 | 2019-01-01 | 南京工业职业技术学院 | A kind of subway station energy consumption short term prediction method and its forecasting system |
| CN109214572A (en) * | 2018-09-06 | 2019-01-15 | 南京工业职业技术学院 | One kind is based on EPSO-BP to commercial building lamp decoration energy consumption prediction technique |
| CN109376939A (en) * | 2018-11-01 | 2019-02-22 | 三峡大学 | A kind of grid stability real-time predicting method based on adaptive neural network |
| CN109582912A (en) * | 2018-11-19 | 2019-04-05 | 北京理工大学 | A kind of public building power consumption monitoring data method for parameter estimation |
| CN109582912B (en) * | 2018-11-19 | 2021-02-02 | 北京理工大学 | Public building power consumption monitoring data parameter estimation method |
| CN111199305A (en) * | 2018-11-20 | 2020-05-26 | 上海烟草集团有限责任公司 | Neural network-based production energy consumption prediction method and system, electronic terminal and storage medium |
| CN109695944A (en) * | 2018-11-29 | 2019-04-30 | 中国汽车工业工程有限公司 | A kind of control method of the coating fresh air conditioner based on multi-model deep learning |
| CN110009156A (en) * | 2019-04-08 | 2019-07-12 | 北京博锐尚格节能技术股份有限公司 | Building energy consumption budget management method and storage medium |
| CN109934422A (en) * | 2019-04-25 | 2019-06-25 | 中国人民解放军国防科技大学 | Neural network wind speed prediction method based on time series data analysis |
| CN110309560A (en) * | 2019-06-13 | 2019-10-08 | 清华大学 | Multi-layer Migration Simulation method and apparatus |
| CN110435009A (en) * | 2019-07-15 | 2019-11-12 | 中国科学院重庆绿色智能技术研究院 | A kind of Intelligentized design method of concrete production match ratio |
| CN110435009B (en) * | 2019-07-15 | 2020-11-10 | 中国科学院重庆绿色智能技术研究院 | Intelligent design method for concrete production mix proportion |
| CN110400018A (en) * | 2019-07-29 | 2019-11-01 | 上海电力大学 | Progress control method, system and device for coal-fired firepower electrical plant pulverized coal preparation system |
| CN110400018B (en) * | 2019-07-29 | 2023-06-27 | 上海电力大学 | Operation control method, system and device for coal-fired power plant pulverizing system |
| CN110533243A (en) * | 2019-08-27 | 2019-12-03 | 西安建筑科技大学 | A kind of building energy consumption prediction technique based on fractal theory |
| CN110942228A (en) * | 2019-10-25 | 2020-03-31 | 万翼科技有限公司 | Energy consumption control method and device, computer equipment and storage medium |
| CN111753470A (en) * | 2020-06-30 | 2020-10-09 | 桂林电子科技大学 | Building energy consumption prediction method based on rough set and deep belief neural network |
| CN112712213A (en) * | 2021-01-15 | 2021-04-27 | 上海交通大学 | Method and system for predicting energy consumption of deep migration learning of centralized air-conditioning house |
| CN112712213B (en) * | 2021-01-15 | 2023-07-04 | 上海交通大学 | Method and system for predicting deep migration learning energy consumption of concentrated air conditioning house |
| CN113344192A (en) * | 2021-05-31 | 2021-09-03 | 中国标准化研究院 | Enterprise-level motor system energy-saving optimization automatic control method and system |
| CN113344192B (en) * | 2021-05-31 | 2022-01-11 | 中国标准化研究院 | Enterprise-level motor system energy-saving optimization automatic control method and system |
| CN117077854A (en) * | 2023-08-15 | 2023-11-17 | 广州视声智能科技有限公司 | Building energy consumption monitoring method and system based on sensor network |
| CN117077854B (en) * | 2023-08-15 | 2024-04-16 | 广州视声智能科技有限公司 | Building energy consumption monitoring method and system based on sensor network |
| CN117371624A (en) * | 2023-12-07 | 2024-01-09 | 邢台职业技术学院 | Building energy-saving transformation optimization management method |
| CN117371624B (en) * | 2023-12-07 | 2024-02-20 | 邢台职业技术学院 | Building energy-saving transformation optimization management method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108320016B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108320016A (en) | A kind of building energy consumption short term prediction method | |
| CN106951611A (en) | A kind of severe cold area energy-saving design in construction optimization method based on user's behavior | |
| CN105868853B (en) | Method for predicting short-term wind power combination probability | |
| CN110046743A (en) | Energy Consumption of Public Buildings prediction technique and system based on GA-ANN | |
| CN111639823B (en) | Building cold and heat load prediction method constructed based on feature set | |
| CN116579506B (en) | Building energy consumption data intelligent management method and system based on big data | |
| CN110766200A (en) | Method for predicting generating power of wind turbine generator based on K-means mean clustering | |
| CN113554466B (en) | Short-term electricity consumption prediction model construction method, prediction method and device | |
| CN112818604A (en) | Wind turbine generator risk degree assessment method based on wind power prediction | |
| CN108399470B (en) | Indoor PM2.5 prediction method based on multi-example genetic neural network | |
| CN106778838A (en) | A kind of method for predicting air quality | |
| CN112329990A (en) | User power load prediction method based on LSTM-BP neural network | |
| CN116316599A (en) | Intelligent electricity load prediction method | |
| CN111815039A (en) | Weekly scale wind power probability prediction method and system based on weather classification | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| CN112365056A (en) | Electrical load joint prediction method and device, terminal and storage medium | |
| CN114819102A (en) | GRU-based air conditioning equipment fault diagnosis method | |
| CN108647817B (en) | Energy consumption load prediction method and system | |
| CN111242266A (en) | Operation data management system | |
| CN115496264A (en) | Method for predicting generated power of wind turbine generator | |
| CN115759343A (en) | E-LSTM-based user electric quantity prediction method and device | |
| Balabanova et al. | Voice control and management in smart home system by artificial intelligence | |
| CN115293406A (en) | Photovoltaic power generation power prediction method based on Catboost and Radam-LSTM | |
| CN114444763A (en) | Wind power prediction method based on AFSA-GNN | |
| Ramsami et al. | Neural Network Frameworks for Electricity Forecasting in Mauritius and Rodrigues Islands |
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 |