CN114186469A - Air conditioning system energy efficiency prediction method and air conditioning system - Google Patents
Air conditioning system energy efficiency prediction method and air conditioning system Download PDFInfo
- Publication number
- CN114186469A CN114186469A CN202010959889.1A CN202010959889A CN114186469A CN 114186469 A CN114186469 A CN 114186469A CN 202010959889 A CN202010959889 A CN 202010959889A CN 114186469 A CN114186469 A CN 114186469A
- Authority
- CN
- China
- Prior art keywords
- air conditioning
- conditioning system
- energy efficiency
- data
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
-
- 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"
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Signal Processing (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Computer Hardware Design (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Fuzzy Systems (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention provides an energy efficiency prediction method of an air conditioning system, which comprises the following steps: acquiring historical data of an air conditioning system to form a data set, and preprocessing the data set; establishing an air conditioner energy efficiency prediction model based on the comprehensive CNN-LSTM based on the processed data set; acquiring real-time operation data of an air conditioning system, preprocessing the data, and inputting the processed real-time data into an established air conditioning energy efficiency prediction model to obtain an air conditioning energy efficiency prediction value at the t +1 moment; by combining a convolutional neural network CNN and a long-short term memory network LSTM, a prediction model combining a high-dimensional vector space and a time sequence is established, the energy efficiency of an air-conditioning system is predicted more accurately, the energy consumption is saved, and the air-conditioning system also comprises: a parameter sensor; and the controller controls the parameter sensor to measure the parameter value of the air conditioning system, can execute the energy efficiency prediction method of the air conditioning system, and can accurately control the air conditioning system according to the prediction result, thereby improving the user experience and saving the energy consumption.
Description
Technical Field
The invention belongs to the technical field of air conditioners, and particularly relates to an air conditioning system energy efficiency prediction method and an air conditioning system.
Background
Along with the improvement of the living standard of people, the air conditioner is more and more one of the electrical appliances essential to the life of people, and meanwhile, the intelligent level of the air conditioner is also required to be higher. The intelligence level of the air conditioner is embodied not only in the aspect of control, but also in the aspect of control of energy consumption of the air conditioner. Manufacturers strive to improve the energy efficiency ratio of the air conditioner so that the air conditioner can operate at higher energy efficiency and meet the requirement of energy conservation. The other aspect is to predict the air conditioner energy efficiency ratio which is used as a feedback input of an air conditioner control system to complete more accurate control of the air conditioner. At present, various methods are used for predicting the air conditioner energy efficiency, such as linear regression, Bayesian estimation algorithm, genetic algorithm and the like, but the air conditioner energy efficiency is influenced by various factors and parameters, and is a very complex nonlinear system, and the accurate prediction effect is difficult to obtain by the traditional method. With the development of artificial intelligence and deep learning technology, many technologies are applied to the prediction of air conditioner energy efficiency, such as a recurrent neural network and its variant long-short term memory network LSTM, wherein the LSTM algorithm obtains better engineering effect, because its algorithm logic and the operation mechanism of the air conditioner have great similarity, but its disadvantage is that it can only predict from time series, neglect some characteristics of the reference variable itself.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art. For this purpose,
according to an embodiment of the disclosure, an energy efficiency prediction method for an air conditioning system is provided, which includes:
s1: acquiring historical data of an air conditioning system to form a data set, and preprocessing the data set;
s2: establishing an air conditioner energy efficiency prediction model based on the comprehensive CNN-LSTM based on the processed data set;
s3: and acquiring real-time operation data of the air conditioning system, preprocessing the data, and inputting the processed real-time data into the established air conditioning energy efficiency prediction model to obtain an air conditioning energy efficiency prediction value at the t +1 moment.
The convolutional neural network CNN and the long-short term memory network LSTM in deep learning are combined, a prediction model combining a high-dimensional vector space and a time sequence is established, the energy efficiency of an air conditioning system is predicted more accurately, and timely control is adopted to save energy consumption on the premise of meeting the comfort.
According to the embodiment of the disclosure, in step S2, based on the processed data set, the specific steps of establishing the comprehensive CNN-LSTM-based air conditioner energy efficiency prediction model are as follows:
s21: inputting the preprocessed data set into a CNN network to obtain a time sequence with high-dimensional characteristic information;
s22: and inputting the time sequence with the high-dimensional characteristic information into an LSTM network for training to obtain a CNN-LSTM model.
According to the embodiment of the present disclosure, in step S21, the specific step of obtaining the time series with the high-dimensional feature information is:
selecting samples from the preprocessed data set, inputting the samples into a convolution functionExtracting high-dimensional characteristic information; extracting important features through a pooling layer, flattening input into a one-dimensional vector through a Flatten layer, and obtaining a time sequence with high-dimensional feature information.
According to the embodiment of the present disclosure, the LSTM network includes a memory unit, and the memory unit includes a forgetting gate, an input gate, and an output gate, and in step S22, the input time sequence is calculated as follows in the LSTM network:
forget the door: f. oft=σ(Wf·[ht-1,xt]+bf);
An input gate: i.e. it=σ(Wi·[ht-1,xt]+bi);
An output gate: ot=σ(Wo·[ht-1,xt]+bo);
the node output value is: h ist=ot·tanh(ct);
Wf,Wi,Wo,WcAs a weight value, bf,bi,bo,bcTo be offset, ht-1Is the output value, x, of the node at time t-1tIs the input value of the node at the time t.
According to the embodiment of the disclosure, a loss function in the training process selects a mean square error loss function, and an optimizer selects an Adam optimizer.
According to the embodiment of the disclosure, in step S1, a data set formed by collecting historical data of the air conditioning system is X,
X=[X1,X2,……,Xt]wherein t represents a time point of data acquisition;
Xt=[x1,x2,……,xn,y]wherein x is1,x2,……,xnAnd y is the energy efficiency ratio at the time point, wherein the n parameters affect the energy efficiency ratio of the air conditioner.
According to an embodiment of the present disclosure, in the step S1, preprocessing the data set includes:
performing one-hot encoding on the data of each independent variable;
then, carrying out normalization processing on the data set: converting each data in the data set into a range interval of [0,1], removing unit limitation of the data, and converting the unit limitation of the data into a dimensionless numerical value, wherein the conversion function is as follows:
x*=(x-xmin)/(xmax-xmin),
where x is the value before conversion, xmaxIs the maximum value, x, in each column of dataminIs the minimum value in each column of data, and x is the value after the transition.
According to an embodiment of the present disclosure, the normalized data sets are classified according to a: the proportion of B is divided into a training set and a test set.
According to the disclosed embodiment, the prediction accuracy of the model is evaluated by using the root mean square error RSME,
wherein n is the total number of evaluation samples, yi objIs the actual value of the i-th sample, yi modelIs the model prediction value of the ith sample.
According to an embodiment of the present disclosure, there is also provided an air conditioning system including:
a parameter sensor;
and the controller is used for controlling the parameter sensor to measure the parameter value of the air conditioning system and can execute any one of the air conditioning system energy efficiency prediction methods.
The energy efficiency prediction method for the air conditioning system can be operated by setting the controller of the air conditioning system, so that the air conditioning system can adopt accurate control according to the prediction result, the user experience is improved, and the energy consumption is saved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of an air conditioning system energy efficiency prediction method according to an embodiment of the present disclosure;
FIG. 2 is a partial flow diagram of an air conditioning system energy efficiency prediction method according to an embodiment of the present disclosure;
FIG. 3 is a graph of the loss function of the CNN-LSTM model according to an embodiment of the present disclosure;
FIG. 4 is a graph of the loss function according to the prior art LSTM model;
FIG. 5 is a graph comparing predicted values and test values of CNN-LSTM model energy consumption ratios according to an embodiment of the disclosure;
FIG. 6 is a graph comparing predicted values and test values according to the prior LSTM model energy consumption ratio;
FIG. 7 is a graph comparing predicted values and test values of CNN-LSTM model energy consumption ratios according to an embodiment of the disclosure;
FIG. 8 is a graph comparing predicted values and test values of energy consumption ratios according to the prior LSTM model.
Detailed Description
The invention is described in detail below by way of exemplary embodiments. It should be understood, however, that elements, structures and features of one embodiment may be beneficially incorporated in other embodiments without further recitation.
The air conditioning system energy efficiency prediction method and the air conditioning system of the present invention will be described below with reference to fig. 1 to 8, and fig. 1 and 2 are flowcharts of the air conditioning system energy efficiency prediction method according to the embodiment of the present disclosure.
The invention provides an energy efficiency prediction method of an air conditioning system, which comprises the following steps:
s1: acquiring historical data of an air conditioning system, forming a data set, and preprocessing the data set;
s2: establishing an air conditioner energy efficiency prediction model based on the comprehensive CNN-LSTM;
s3: and acquiring real-time data of the air conditioning system, preprocessing the data, and inputting the processed real-time data into the established air conditioning energy efficiency prediction model to obtain an air conditioning energy efficiency prediction value at the t +1 moment.
The convolutional neural network CNN and the long-short term memory network LSTM in deep learning are combined, a prediction model combining a high-dimensional vector space and a time sequence is established, the energy efficiency of an air conditioning system is predicted more accurately, and timely control is adopted to save energy consumption on the premise of meeting the comfort.
In step S1, when acquiring historical data of the air conditioning system, parameters affecting the energy efficiency ratio of the air conditioning system need to be determined, and when selecting the parameters, parameters having a large impact on the energy efficiency ratio are mainly selected, where the parameters include evaporator water outlet Temperature (TEI), evaporator water outlet Temperature (TEO), condenser water inlet Temperature (TCI), condenser water outlet Temperature (TCO), heat exchanger water inlet Temperature (TSI), condenser loop temperature (TS0), heat exchanger water outlet Temperature (TBI), evaporator loop Temperature (TBO), building water inlet temperature (Cond Tons), building water outlet temperature (Cooling Tons), steam heating amount (kW), evaporator flow (TEA), condenser flow (TCA), condenser temperature difference representation value (TRE), and refrigerant Temperature (TRC), and all 15 parameter values may be obtained by using a sensor.
Parameter value and energy efficiency ratio on one time point of air conditioning system are collected to form parameter vector Xi=[x1,x2,……,xn,y]Wherein x is1,x2,……,xnIn order to obtain parameter values corresponding to n parameters influencing the energy efficiency ratio of the air conditioner, wherein n represents that n parameters are collected totally, y is the energy efficiency ratio at the time point, and a historical data set X of the air conditioner system can be obtained by collecting the parameter values and the energy efficiency ratio of the air conditioner system at different time points, wherein X is [ X ═ X1,X2,……,Xt]And t represents the time point of data acquisition, the data acquisition is carried out at certain time intervals, and the acquisition forms a data set with certain time characteristics. Specifically, the parameter values of the 15 parameters affecting the energy efficiency ratio, COP, which is the cooling capacity/loss power, X, can be obtained by calculationi=[TEI,TEO,TCI,TCO,TSI,TSO,TBI,TBO,Cond Tons,Cooling Tons,Kw,TEA,TCA,TRE,TRC,COP]The vector dimension is 16 dimensions.
In step S1, preprocessing the data set includes:
performing one-hot encoding on the data of each independent variable;
then, carrying out normalization processing on the data set: converting each data in the data set into a range interval of [0,1], removing unit limitation of the data, and converting the unit limitation of the data into a dimensionless numerical value, wherein the conversion function is as follows:
x*=(x-xmin)/(xmax-xmin),
where x is the value before conversion, xmaxIs the maximum value, x, in each column of dataminIs the minimum value in each column of data, and x is the value after the transition.
And processing each column of parameters in the data set during normalization processing.
Dividing the data set after normalization into a training set and a testing set, wherein the data set can be divided into a training set and a testing set according to the following steps: and B, splitting according to the proportion of 8:2, specifically, if the acquired data are 5000 vectors, 4000 vectors are training sets, and 1000 vectors are testing sets.
In step S2, the specific steps of establishing the comprehensive CNN-LSTM-based air conditioner energy efficiency prediction model based on the processed data set are as follows:
s21: inputting the preprocessed data set into a CNN network to obtain a time sequence with high-dimensional characteristic information;
s22: and inputting the time sequence with the high-dimensional characteristic information into an LSTM network for training to obtain a CNN-LSTM model.
In step S21, the specific steps of obtaining the time series with the high-dimensional feature information are:
selecting samples in the preprocessed training set, inputting the samples into a convolution functionExtracting high-dimensional characteristic information; extracting important features through a pooling layer, flattening input into a one-dimensional vector through a Flatten layer, and obtaining a time sequence with high-dimensional feature information.
Specifically, the principle of data prediction and sample selection is to predict the last COP data of the N +1 th row by using the data of the first N rows, and so on, sliding is performed according to a time window, the sliding step is 1, and the N rows are equivalent to the width of the sliding window being N.
The LSTM network includes a memory unit, the memory unit includes a forgetting gate, an input gate and an output gate, and the memory unit can selectively memorize the correction parameters of the feedback loss function decreasing with the gradient, in step S22, the time sequence with high-dimensional feature information is input into the LSTM network for training, and the input time sequence is calculated in the LSTM network as follows:
forget the door: f. oft=σ(Wf·[ht-1,xt]+bf);
ftOutput value of forgetting gate, WfTo forget the weight of the door neural network, ht-1Is the output of the node at time t-1, xtAs input to the node at time t, bfTo forget the biasing of the door neural network.
An input gate: i.e. it=σ(Wi·[ht-1,xt]+bi);
itFor inputting the output value of the gate, WiIs the weight of the input gate neural network, ht-1Is the output of the node at time t-1, xtAs input to the node at time t, biIs the bias of the input gated neural network.
An output gate: ot=σ(Wo·[ht-1,xt]+bo);
otTo output the output value of the gate, WoIs the weight of the output gate neural network, ht-1Is the output of the node at time t-1, xtAs input to the node at time t, boIs the bias of the output gated neural network.
Wcis the weight of the cell input, ht-1Is the output of the node at time t-1, xtAs input to the node at time t, bcIs the bias of the cell input.
the element-by-element multiplication operation representing a matrix.
The node output value is: h ist=ot·tanh(ct);
htThe output of the node at time t represents the element-by-element multiplication of the matrix.
Fusion activation function in CNN-LSTM model:
in the training process, the LSTM network is converted into a corresponding output value through a full connection layer Dense, a mean square error loss function is selected as a loss function, and an Adam optimizer is selected as an optimizer.
After a CNN-LSTM model is established, selecting samples in a test set to test the precision of the CNN-LSTM model, calculating the error of the CNN-LSTM model, if the error is larger than a set value, increasing the training times to obtain a new CNN-LSTM model, if the error is smaller than or equal to the set value, evaluating the precision of the CNN-LSTM model to meet the conditions, specifically, evaluating the prediction precision of the model by adopting the root mean square error RSME,
wherein n is the total number of evaluation samples, yi objIs the actual value of the i-th sample, yi modelFor the model prediction value of the ith sample, the RMSE model predicts the root mean square error of the data from the true data.
The principle of sample selection in the data test is to predict the last COP data in the N +1 th row by using the data in the first N rows, and the sliding is performed according to the time window in the same way, wherein the sliding step length is 1, and the width of the sliding window in the N rows is N.
After the CNN-LSTM model is determined, real-time data of the air conditioning system are collected and preprocessed, the processed real-time data are input into the CNN-LSTM model, and the COP of the air conditioner at the t +1 moment can be obtained, wherein the preprocessing of the real-time data is the same as a method for processing historical data.
Specifically, 5000 sets of data are collected for the air conditioning system to obtain an initial data set, and part of samples of the collected data set are shown in table 1.
TABLE 1
The deep learning network based on the CNN and the LSTM is composed of an input layer, a 1D-CNN layer, a pooling layer, an LSTM layer, a full-connection layer and an output layer, wherein a mean square error loss function is selected as a loss function, and an Adam optimizer is selected as an optimizer. The deep learning network model based on 1D-CNN and LSTM is shown in Table 2.
TABLE 2
Layer name | Output tensor shape | Number of parameters |
dense_1(Dense) | (None,5,128) | 2176 |
conv1d_1(Conv1D) | (None,5,80) | 10320 |
max_pooling1d_1 | (None,2,80) | 0 |
conv1d_2(Conv1D) | (None,2,48) | 3888 |
max_pooling1d_2 | (None,2,48) | 0 |
dropout_1(Dropout) | (None,2,48) | 0 |
lstm_1(LSTM) | (None,2,32) | 10368 |
lstm_2(LSTM) | (None,16) | 3136 |
dense_2(Dense) | (None,32) | 544 |
dense_3(Dense) | (None,1) | 33 |
FIG. 3 is a graph of the loss function of the CNN-LSTM model according to the embodiment of the present disclosure, and FIG. 4 is a graph of the loss function of the conventional LSTM model, which shows that the CNN-LSTM model converges faster and has smaller loss value, which indicates that the CNN-LSTM model predicts more accurately than the LSTM model.
Fig. 5 is a comparison graph of the predicted value and the test value of the CNN-LSTM model energy consumption ratio according to the embodiment of the disclosure, and fig. 6 is a comparison graph of the predicted value and the test value of the CNN-LSTM model energy consumption ratio according to the prior LSTM model energy consumption ratio, wherein the comparison is performed on 1000 time series of the test set, and as can be seen from fig. 5 and 6, the predicted value and the test value of the CNN-LSTM model are closer, which shows that the CNN-LSTM model predicts more accurately compared with the LSTM model. Fig. 7 is a comparison graph of the predicted value and the test value of the CNN-LSTM model energy consumption ratio according to the embodiment of the disclosure, and fig. 8 is a comparison graph of the predicted value and the test value of the CNN-LSTM model energy consumption ratio according to the conventional LSTM model energy consumption ratio, wherein the comparison graph is performed on 5000 time series of the total data set, as can be seen from fig. 7 and 8, the predicted value and the test value of the CNN-LSTM model are closer, which indicates that the CNN-LSTM model predicts more accurately compared with the LSTM model. Among them, the COP Value in fig. 5 to 8 is a normal COP Value after being reduced by 100 times.
The invention also provides an air conditioning system, which comprises a parameter sensor and a controller, wherein the controller controls the parameter sensor to measure the parameter value of the air conditioning system, and can execute the energy efficiency prediction method of the air conditioning system.
The energy efficiency prediction method for the air conditioning system can be operated by setting the controller of the air conditioning system, so that the air conditioning system can adopt accurate control according to the prediction result, the user experience is improved, and the energy consumption is saved.
The air conditioning system also comprises a compressor, a condenser, an expansion valve, an evaporator and other structures, and can complete the refrigeration and/or heating functions of the air conditioning system.
In the description of the present invention, it is to be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention.
The terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; they may be connected directly or indirectly through intervening media, or they 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.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (10)
1. An air conditioning system energy efficiency prediction method is characterized by comprising the following steps:
s1: acquiring historical data of an air conditioning system to form a data set, and preprocessing the data set;
s2: establishing an air conditioner energy efficiency prediction model based on the comprehensive CNN-LSTM based on the processed data set;
s3: and acquiring real-time operation data of the air conditioning system, preprocessing the data, and inputting the processed real-time data into the established air conditioning energy efficiency prediction model to obtain an air conditioning energy efficiency prediction value at the t +1 moment.
2. The energy efficiency prediction method of the air conditioning system according to claim 1, wherein in step S2, the specific steps of establishing the comprehensive CNN-LSTM-based air conditioning energy efficiency prediction model based on the processed data set are as follows:
s21: inputting the preprocessed data set into a CNN network to obtain a time sequence with high-dimensional characteristic information;
s22: and inputting the time sequence with the high-dimensional characteristic information into an LSTM network for training to obtain a CNN-LSTM model.
3. The method for predicting the energy efficiency of an air conditioning system according to claim 2, wherein the step S21 of obtaining the time series having the high-dimensional characteristic information includes:
selecting samples from the preprocessed data set, inputting the samples into a convolution functionExtracting high-dimensional characteristic information; extracting important features through a pooling layer, flattening input into a one-dimensional vector through a Flatten layer, and obtaining a time sequence with high-dimensional feature information.
4. The method for predicting the energy efficiency of an air conditioning system according to claim 2, wherein the LSTM network includes a memory unit, the memory unit includes a forgetting gate, an input gate and an output gate, and in step S22, the input time sequence is calculated in the LSTM network as follows:
forget the door: f. oft=σ(Wf·[ht-1,xt]+bf);
An input gate: i.e. it=σ(Wi·[ht-1,xt]+bi);
An output gate: ot=σ(Wo·[ht-1,xt]+bo);
the node output value is: h ist=ot·tanh(ct);
Wf,Wi,Wo,WcAs a weight value, bf,bi,bo,bcTo be offset, ht-1Is the output value, x, of the node at time t-1tIs the input value of the node at the time t.
5. The energy efficiency prediction method of the air conditioning system according to claim 2, characterized in that the loss function during the training process is a mean square error loss function, and the optimizer is an Adam optimizer.
6. The energy efficiency prediction method of an air conditioning system according to claim 1, wherein in step S1, a data set formed by collecting historical data of the air conditioning system is X,
X=[X1,X2,……,Xt]wherein t represents a time point of data acquisition;
Xt=[x1,x2,……,xn,y]wherein x is1,x2,……,xnAnd y is the energy efficiency ratio at the time point, wherein the n parameters affect the energy efficiency ratio of the air conditioner.
7. The air conditioning system energy efficiency prediction method according to claim 6, wherein in the step S1, the preprocessing the data set comprises:
performing one-hot encoding on the data of each independent variable;
then, carrying out normalization processing on the data set: converting each data in the data set into a range interval of [0,1], removing unit limitation of the data, and converting the unit limitation of the data into a dimensionless numerical value, wherein the conversion function is as follows:
x*=(x-xmin)/(xmax-xmin),
where x is the value before conversion, xmaxIs the maximum value, x, in each column of dataminIs the minimum value in each column of data, and x is the value after the transition.
8. The air conditioning system energy efficiency prediction method according to claim 7, characterized in that the normalized data sets are classified according to A: the proportion of B is divided into a training set and a test set.
9. The energy efficiency prediction method of an air conditioning system according to claim 8, characterized in that the model prediction accuracy is evaluated using a root mean square error RSME,
wherein n is the total number of evaluation samples, yi objIs the actual value of the i-th sample, yi modelIs the model prediction value of the ith sample.
10. An air conditioning system, comprising:
a parameter sensor;
a controller controlling the parameter sensor to measure a value of an air conditioning system parameter and capable of performing the air conditioning system energy efficiency prediction method according to any one of claims 1 to 9.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010959889.1A CN114186469A (en) | 2020-09-14 | 2020-09-14 | Air conditioning system energy efficiency prediction method and air conditioning system |
PCT/CN2021/100427 WO2022052547A1 (en) | 2020-09-14 | 2021-06-16 | Method for predicting energy efficiency of air-conditioning system, and air-conditioning system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010959889.1A CN114186469A (en) | 2020-09-14 | 2020-09-14 | Air conditioning system energy efficiency prediction method and air conditioning system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114186469A true CN114186469A (en) | 2022-03-15 |
Family
ID=80538959
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010959889.1A Pending CN114186469A (en) | 2020-09-14 | 2020-09-14 | Air conditioning system energy efficiency prediction method and air conditioning system |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN114186469A (en) |
WO (1) | WO2022052547A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117234080A (en) * | 2023-09-21 | 2023-12-15 | 江苏亚奥科技股份有限公司 | Room temperature intelligent regulation and control method for large-scale moving ring monitoring scene |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113780405B (en) * | 2021-09-07 | 2023-07-25 | 科希曼电器有限公司 | Air conditioner parameter regression optimization method based on deep neural network |
CN114781744A (en) * | 2022-05-07 | 2022-07-22 | 东南大学 | Deep learning multi-step long radiance prediction method based on codec |
CN115062847A (en) * | 2022-06-16 | 2022-09-16 | 北京嘉洁能科技股份有限公司 | Carbon fiber heating energy consumption prediction method and system based on logistic regression algorithm |
CN115081342A (en) * | 2022-07-26 | 2022-09-20 | 国网安徽省电力有限公司电力科学研究院 | LSTNet-based 'carbon consumption-energy consumption-output' prediction method for key emission control enterprises |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106765932A (en) * | 2016-12-14 | 2017-05-31 | 深圳达实智能股份有限公司 | The Energy Efficiency Ratio Forecasting Methodology and device of central air conditioner system refrigeration host computer |
WO2018184102A1 (en) * | 2017-04-03 | 2018-10-11 | Royal Bank Of Canada | Systems and methods for malicious code detection |
CN109961177A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of general water cooled central air conditioner energy consumption prediction technique based on shot and long term memory Recognition with Recurrent Neural Network |
CN110443725A (en) * | 2019-08-14 | 2019-11-12 | 国网甘肃省电力公司信息通信公司 | Wind power prediction model based on CNN+LSTM |
CN110648014B (en) * | 2019-08-28 | 2022-04-15 | 山东大学 | Regional wind power prediction method and system based on space-time quantile regression |
-
2020
- 2020-09-14 CN CN202010959889.1A patent/CN114186469A/en active Pending
-
2021
- 2021-06-16 WO PCT/CN2021/100427 patent/WO2022052547A1/en active Application Filing
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117234080A (en) * | 2023-09-21 | 2023-12-15 | 江苏亚奥科技股份有限公司 | Room temperature intelligent regulation and control method for large-scale moving ring monitoring scene |
CN117234080B (en) * | 2023-09-21 | 2024-05-28 | 江苏亚奥科技股份有限公司 | Room temperature intelligent regulation and control method for large-scale moving ring monitoring scene |
Also Published As
Publication number | Publication date |
---|---|
WO2022052547A1 (en) | 2022-03-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114186469A (en) | Air conditioning system energy efficiency prediction method and air conditioning system | |
CN103912966B (en) | A kind of earth source heat pump refrigeration system optimal control method | |
CN106920006B (en) | Subway station air conditioning system energy consumption prediction method based on ISOA-LSSVM | |
CN113326651B (en) | Refrigerating station load and energy efficiency ratio dynamic modeling method based on T-S fuzzy model | |
Afram et al. | Review of modeling methods for HVAC systems | |
CN109959123A (en) | A kind of energy-saving method for air conditioner based on genetic algorithm and shot and long term memory Recognition with Recurrent Neural Network | |
CN113112077B (en) | HVAC control system based on multi-step prediction deep reinforcement learning algorithm | |
CN109961177A (en) | A kind of general water cooled central air conditioner energy consumption prediction technique based on shot and long term memory Recognition with Recurrent Neural Network | |
CN113240184B (en) | Building space unit cold load prediction method and system based on federal learning | |
CN112884012A (en) | Building energy consumption prediction method based on support vector machine principle | |
KR20180138371A (en) | Method for evaluating data based models and conducting predictive control of capsule type ice thermal storage system using the same | |
CN112984616A (en) | Method for identifying regulation and control rule of heating power station of heating system | |
Yang et al. | Research on energy-saving optimization of commercial central air-conditioning based on data mining algorithm | |
Deng et al. | Optimal control of chilled water system with ensemble learning and cloud edge terminal implementation | |
CN113432247B (en) | Water chilling unit energy consumption prediction method and system based on graph neural network and storage medium | |
CN113028610B (en) | Method and device for global optimization and energy-saving control of dynamic load of central air conditioner | |
CN112268350B (en) | Air conditioner side load prediction method based on system delay | |
CN117455725A (en) | Building energy consumption management method and system based on BIM (building information modeling) building | |
CN112180733B (en) | Fuzzy logic-based building energy consumption system prediction control parameter setting method | |
CN112747418A (en) | Air conditioner and cloud server | |
CN116045461B (en) | Energy-saving control method and device for air-cooled air conditioner based on water supply and return temperature adjustment | |
CN111291465A (en) | Ship air conditioner thermal comfort temperature prediction method based on genetic algorithm and neural network | |
CN111189201A (en) | Air conditioner prediction control method based on machine vision | |
CN116241991A (en) | Central air conditioner control method and system based on deep reinforcement learning | |
CN112728737A (en) | Air conditioner and cloud server |
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 |