CN113970170A - Central air-conditioning system energy consumption prediction method and device and computing equipment - Google Patents
Central air-conditioning system energy consumption prediction method and device and computing equipment Download PDFInfo
- Publication number
- CN113970170A CN113970170A CN202010725321.3A CN202010725321A CN113970170A CN 113970170 A CN113970170 A CN 113970170A CN 202010725321 A CN202010725321 A CN 202010725321A CN 113970170 A CN113970170 A CN 113970170A
- Authority
- CN
- China
- Prior art keywords
- layer
- historical
- energy consumption
- neural network
- network model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 186
- 238000004378 air conditioning Methods 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000003062 neural network model Methods 0.000 claims abstract description 146
- 238000010606 normalization Methods 0.000 claims abstract description 79
- 238000012549 training Methods 0.000 claims abstract description 63
- 230000004913 activation Effects 0.000 claims abstract description 51
- 238000012545 processing Methods 0.000 claims description 30
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 25
- 238000004891 communication Methods 0.000 claims description 15
- 239000000498 cooling water Substances 0.000 claims description 15
- 238000001816 cooling Methods 0.000 claims description 12
- 230000017525 heat dissipation Effects 0.000 claims description 10
- 230000007613 environmental effect Effects 0.000 claims description 9
- 230000003213 activating effect Effects 0.000 claims description 8
- 230000005855 radiation Effects 0.000 claims description 8
- 238000001994 activation Methods 0.000 claims 6
- 238000010586 diagram Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000012141 concentrate Substances 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000005057 refrigeration Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000008400 supply water Substances 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
- F24F11/46—Improving electric energy efficiency or saving
- F24F11/47—Responding to energy costs
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—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
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2140/00—Control inputs relating to system states
- F24F2140/60—Energy consumption
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The embodiment of the invention relates to the technical field of infrastructure, and discloses a method, a device and a computing device for predicting energy consumption of a central air-conditioning system, wherein the method comprises the following steps: acquiring historical influence factors related to energy consumption of a central air-conditioning system and corresponding historical energy consumption data; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; and predicting the energy consumption of the central air-conditioning system by adopting the neural network model. Through the mode, the embodiment of the invention can improve the training efficiency of the neural network model and improve the prediction accuracy and generalization capability of the neural network model.
Description
Technical Field
The embodiment of the invention relates to the technical field of infrastructure, in particular to a method and a device for predicting energy consumption of a central air-conditioning system and computing equipment.
Background
An air conditioner is a common refrigeration device and is applied to various industries. Taking a data center as an example, the air conditioning system provides an adaptive temperature, humidity and cleanliness environment for the server. But as data center auxiliary assembly, the air conditioning system energy consumption accounts for more than 30% of total energy consumption, has huge energy-conserving potentiality. To adopt accurate and effective energy-saving optimization adjustment of the air conditioning system, an accurate energy consumption prediction model of the air conditioning system must be established. Due to the reasons of complex air-conditioning system architecture, various equipment types, different equipment performances and the like, the difficulty in establishing the energy consumption model of the central air-conditioning system is very high.
At present, a thermophysical theory formula or an empirical formula is often adopted to perform energy consumption modeling in a design stage. On one hand, the correlation performance coefficients of different devices in the formula of the method are difficult to determine; on the other hand, the air conditioning system equipment can generate certain operations such as performance attenuation, equipment replacement and the like during the use process. The accuracy and the timeliness of the prediction model obtained by applying the formula modeling method are low.
With the development of artificial intelligence technology, the air conditioning system can be greatly reduced in energy consumption modeling difficulty by adopting the neural network algorithm and taking the operation data as a training set, and the problem that the air conditioning system is low in accuracy and timeliness can be solved to a certain extent thanks to the universal approximation theorem of the neural network algorithm. The energy consumption prediction algorithm of the neural network model is gradually developed, and the conventional neural network model still has the problems of a large amount of characteristic engineering, low model training efficiency, poor generalization capability and the like.
Disclosure of Invention
In view of the above problems, embodiments of the present invention provide a method, an apparatus, and a computing device for predicting energy consumption of a central air conditioning system, which overcome or at least partially solve the above problems.
According to an aspect of an embodiment of the present invention, there is provided a central air conditioning system energy consumption prediction method, including: acquiring historical influence factors related to energy consumption of a central air-conditioning system and corresponding historical energy consumption data; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; and predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
In an alternative mode, the obtaining historical influencing factors related to energy consumption of the central air conditioning system and corresponding historical energy consumption data includes: acquiring historical setting parameters, historical outdoor environment and historical indoor demand factors of the central air-conditioning system; selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors; and acquiring the historical energy consumption data corresponding to the historical influence factors.
In an optional manner, the obtaining of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors of the central air conditioning system includes: acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower; acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation; acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
In an optional manner, the training the neural network model according to the historical influence factors and outputting corresponding predicted energy consumption data includes: inputting the historical influencing factors into the input layer of a neural network model; carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer; normalizing the output data of the full connection layer through the first normalization layer of the hidden layer; performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer; and outputting the corresponding predicted energy consumption data through the output layer.
In an alternative form, the inputting the historical influence factors into the input layer of a neural network model further comprises: adding a second normalization layer after the input layer, and performing normalization processing on the input historical influence factors through the second normalization layer; before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer.
In an optional manner, the normalizing, by the first normalization layer of the hidden layer, the output data of the fully-connected layer includes: normalizing the output data by applying the following relational expression, and concentrating most of the output data in a preset range: y ═ x-xmean) V (delta + epsilon), where y is the result after normalization, xmeanIs the average value of the historical influencing factors input, delta is the standard deviation of the historical influencing factors input, and epsilon is a very small value.
In an optional manner, the activating, by the active layer of the hidden layer, the normalized output data includes: activating the output data in the preset range through the activation layer; and regarding the output data outside the preset range as error data.
According to another aspect of the embodiments of the present invention, there is provided an energy consumption prediction apparatus for a central air conditioning system, the apparatus including: the training data acquisition unit is used for acquiring historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data; the training unit is used for training a neural network model according to the historical influence factors and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a normalization layer and an activation layer, and an output layer; the model output unit is used for adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; and the energy consumption prediction unit is used for predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
According to another aspect of embodiments of the present invention, there is provided a computing device including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the steps of the central air-conditioning system energy consumption prediction method.
According to another aspect of the embodiments of the present invention, there is provided a computer storage medium having at least one executable instruction stored therein, the executable instruction causing the processor to execute the steps of the central air conditioning system energy consumption prediction method.
According to the embodiment of the invention, historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; the neural network model is adopted to predict the energy consumption of the central air-conditioning system, so that the training efficiency of the neural network model can be improved, and the prediction accuracy and generalization capability of the neural network model can be improved.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and the embodiments of the present invention can be implemented according to the content of the description in order to make the technical means of the embodiments of the present invention more clearly understood, and the detailed description of the present invention is provided below in order to make the foregoing and other objects, features, and advantages of the embodiments of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a schematic flow chart illustrating a method for predicting energy consumption of a central air-conditioning system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating an influence factor of energy consumption of an air conditioning system in an energy consumption prediction method for a central air conditioning system according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a neural network model of a central air-conditioning system energy consumption prediction method according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a neural network model of a central air-conditioning system energy consumption prediction method according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating an example of the accuracy of a method for predicting energy consumption of a central air conditioning system according to an embodiment of the present invention;
FIG. 6 is a generalized exemplary diagram illustrating a method for predicting energy consumption of a central air conditioning system according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram illustrating an energy consumption prediction apparatus of a central air conditioning system according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a computing device provided in an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
Fig. 1 is a schematic flow chart illustrating a method for predicting energy consumption of a central air-conditioning system according to an embodiment of the present invention. As shown in fig. 1, the method for predicting energy consumption of a central air-conditioning system includes:
step S11: historical influencing factors related to energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained.
The air conditioning system generally comprises five major devices, namely an air conditioning host, a chilled water pump, an air conditioning tail end, a cooling water pump and a cooling tower. The energy consumption of the air conditioning system is influenced by the setting parameters of internal equipment of the air conditioning system, and the outdoor environment and the indoor heat dissipation capacity are important influencing factors.
Taking a water-cooling chilled water air-conditioning system of a data center as an example, a Building Automation (BA) system couples operation states of five devices of the air-conditioning system together. In order to ensure the safe and stable operation of the air conditioning system, the BA system hides the direct influence factor of the energy consumption of each device, namely the operation frequency, at the rear end for indirect control, and the setting value exposed at the front end can be adjusted and set by an operator. As shown in fig. 2, the factors affecting the energy consumption of the air conditioning system are classified into three major categories: the air-conditioning system comprises an air-conditioning system internal equipment operation parameter setting value X, an outdoor environment factor Y and an indoor demand factor Z. Among the three variables, X is a free controllable variable and is determined by manual setting; y is an environment uncontrollable variable and is determined by local climate conditions; z is a semi-controllable variable meeting the demand, is determined by the indoor business demand, and can be partially adjusted.
The operation parameters X of the air conditioning system mainly comprise the following five aspects:
setting parameter X of air conditioner hostAGenerally comprising: the temperature of the outlet water of the chilled water and other settable values;
setting parameter X of freezing water pumpBGenerally comprising: the pressure difference setting value of the supply water and the return water of the chilled water, the minimum running frequency of the chilled water pump and other settable values;
setting parameter X of air conditioner terminalCGenerally comprising: air supply/return temperature, air supply/return relative humidity, valve opening, and other settable values;
setting parameter X of cooling water pumpDGenerally comprising: a cooling water supply and return water differential pressure set value, a cooling water supply and return water differential temperature set value, a minimum operating frequency of a cooling water pump and other settable values;
setting parameter X of cooling towerEGenerally comprising: cooling water outlet temperature, minimum fan operating frequency and other settable values.
The outdoor influencing factors Y generally include: outdoor temperature, outdoor relative humidity, wind speed, solar radiation, other environmental factors.
Indoor demand factors Z generally include: indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
For different BA systems, the monitoring parameters and the equipment controllable parameters are different, and different 'X + Y + Z' parameters are selected as input layers of the neural network model according to different air-conditioning system architectures and BA system designs. The relationship for the total energy consumption of the air conditioning system can be determined as: p ═ f (X, Y, Z).
In step S11, the history setting parameters, the history outdoor environment, and the history indoor demand factors of the central air conditioning system are acquired. Specifically, historical monitoring data of the BA system are collected, variable decoupling is carried out on factors influencing air conditioner energy consumption, and the factors are divided into historical setting parameters, historical outdoor environment and historical indoor demand factors. Acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower; acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation; acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
Then, selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors; and acquiring the historical energy consumption data corresponding to the historical influence factors. For example, 5 parameters of the outlet water temperature of chilled water of a host, the wet bulb temperature deviation of a cooling tower, the supply and return water pressure difference of the chilled water, the air supply/return air temperature at the tail end of an air conditioner and the supply and return water pressure difference of a cooling water pump are selected from historical setting parameters to be used as controllable variables in historical influence factors, 2 parameters of the temperature and the relative humidity are selected from historical outdoor environment factors to be used as environment variables in the historical influence factors, 1 parameter of the heating value of an indoor IT load is selected from historical indoor demand factors to be used as a demand variable in the historical influence factors, and historical energy consumption data corresponding to the selected parameters are further obtained.
After the historical influence factors used for training the neural network model are obtained, the historical influence factors are divided into a training set and a testing set, the training set is used for training the neural network model, and the testing set is used for verifying and testing the trained neural network model.
Step S12: and training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer.
The structure of the neural network model is as shown in fig. 3, and in step S12, the historical influence factors are input to the input layer of the neural network model; carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer; normalizing the output data of the full connection layer through the first normalization layer of the hidden layer; performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer; and outputting the corresponding predicted energy consumption data through the output layer.
In the embodiment of the present invention, before the historical influencing factors are input to the input layer of the neural network model, the historical influencing factors need to be preprocessed, for example, normalized. The first normalization layer of the hidden layer applies the following relational expression to normalize the output data, and concentrates most of the output data in a preset range:
y＝(x–xmean)/(δ+ε)，
wherein y is the result after standardization treatment, and xmeanIs the average value of the historical influencing factors input, delta is the standard deviation of the historical influencing factors input, and epsilon is a very small value. The preset range is located in an activation region of the activation function.
The activation layer in the embodiment of the invention adopts a tanh activation function, and after output data is subjected to standardization processing by the first standardization layer, the distribution of the output data of the full connection layer is concentrated between (-3,3) and is always positioned in an activation region of the tanh function. And subsequently, activating the output data in the preset range through the activation layer. And regarding the output data outside the preset range as error data. That is, for data outside the range of (-3,3), the data is considered as error data in the training process, and the activation function tanh does not activate the error data, so that the learning effect of the data is reduced.
In the training process, for each batch of training data, the data distribution of the training data before the activation layer can generate the condition of scattered distribution, and the more the number of hidden layers is, the more easily the problem of gradient disappearance is generated, and effective training cannot be carried out. The data distribution is concentrated in the activation range of the tanh activation function through the first normalization layer in the improved hidden layer, so that the training efficiency can be improved; for error data, the error data are distributed outside the activation range of the tanh function, the influence degree on the training of the neural network model is low, overfitting can be prevented, and the generalization capability is improved.
In an embodiment of the present invention, a second normalization layer is added after the input layer, as shown in FIG. 4. In the neural network model, the input layer includes a fully-connected layer (not shown), and the input layer transmits the input historical influencing factors to the hidden layer through the fully-connected layer. And adding a second standardization layer after the fully-connected layer of the input layer and before the hidden layer, and carrying out standardization processing on the input historical influence factors through the second standardization layer. The addition of the second normalization layer eliminates the need for pre-processing of historical influences prior to their input into the input layer of the neural network model. Therefore, by adding the second normalization layer behind the input layer, the normalization step in the early-stage feature engineering can be reduced and the workload of the feature engineering can be reduced.
With continued reference to FIG. 4, a third normalization layer may also be added before the output layer. Before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer. Specifically, in the neural network model, the output layer also includes a fully connected layer (not shown), and the third normalization layer is added after the fully connected layer in the output layer for normalization, so as to further improve the training efficiency of the neural network model. The normalization processing methods of the second normalization layer and the third normalization layer are the same as the processing method of the first normalization layer, and are not described herein again.
Step S13: and adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model.
In the embodiment of the invention, the purpose of the neural network model training is to make the predicted energy consumption data as identical as possible to the actual energy consumption data, i.e. the corresponding historical energy consumption data. During training, a gradient descent optimization algorithm is adopted for training, the predicted energy consumption data is compared with the historical energy consumption data, the error between the predicted energy consumption data and the historical energy consumption data is calculated, and whether the error is larger than a preset threshold value or not is judged. If the error is larger than the preset threshold value, updating the model parameters of the neural network model by adopting a reverse gradient method, and then training the neural network model by adopting the historical energy consumption data in the training set again. If the error is smaller than a preset threshold value, testing the neural network model by using a test set, and outputting model parameters after the neural network model is trained when the error between the predicted energy consumption data and the historical energy consumption data of the neural network model meets the preset precision.
Step S14: and predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
The real-time monitoring data of the BA system are collected, variable decoupling is carried out on factors influencing the energy consumption of the air conditioner, corresponding influencing factors relevant to the energy consumption of the central air conditioning system are obtained, and the trained neural network model is applied to predict the energy consumption of the central air conditioning system according to the influencing factors.
Taking a certain data center air conditioning system as an example, the controllable variables of the air conditioning system are selected as 5 parameters such as the outlet water temperature of the chilled water of a host, the wet bulb temperature deviation of a cooling tower, the supply and return water pressure difference of a cooling water pump and the like, the environment variables are selected as 2 parameters of temperature and relative humidity, the demand variables are selected as 1 parameter of indoor IT load heating value, the input layer variable structure form of the '5 +2+ 1' is formed, and other air conditioning system parameters are fixed values and do not participate in the calculation of a neural network model. Based on an improved neural network model, a neural network model structure with 8 nodes as an input layer, 20 nodes as a hidden layer and 1 node as an output layer is constructed. As shown in FIG. 5, the historical monitoring data is used as training data in a proportion of 70%, a gradient descent optimization algorithm is adopted for training, the average accuracy of the prediction of the neural network model with the structure is 99.33%, the average error is 0.67%, and the neural network model has high precision. For the energy consumption data which obviously deviates from a normal value and is generated in the related maintenance process of the air conditioning system, the neural network model can be automatically ignored; for the neural network model which is not trained again, as shown in fig. 6, the prediction accuracy of the neural network model is still kept above 97% under the multi-working-condition after 1 month, and the obvious change rule consistent with the true value is kept, so that the neural network model has strong generalization capability.
In the embodiment of the invention, BA system monitoring data is adopted to carry out variable decoupling on factors influencing air conditioner energy consumption, and the factors are divided into controllable variables, environment variables and demand variables which are used as input layer parameters of a neural network model, so that the structure is clear; the direct control parameters are adopted as controllable variables, so that the subsequent adoption of related automatic energy-saving management is facilitated. Determining a setting parameter directly influencing the energy consumption of the air conditioning system according to the air conditioning BA system, namely, a controllable variable X; determining an environment variable Y according to the environment monitoring quantity of the BA system; the demand variable Z is determined from the indoor load requirements. And (3) constructing an independent variable of 'X + Y + Z' influencing the energy consumption of the air conditioning system, and using the independent variable to an input layer of the neural network model. The neural network model can change controllable variables along with the upgrading of the BA system, does not concern internal automatic control coupling logic, and is convenient for energy-saving optimization of related parameter setting values; in addition, the energy-saving, safety and stability regulation and control of the refrigeration task of the air conditioning system can be carried out according to different indoor requirements (indoor heat dissipation requirements and temperature control requirements).
The method is characterized in that a standardized method is integrated into a traditional neural network model structure, an input layer, a hidden layer and an output layer are improved by the standardized method, a neural network model for improving the input layer, the hidden layer and the output layer is constructed, and an activation function adopts tanh. The improved input layer can greatly reduce the workload of the early-stage characteristic engineering and the conversion of input variables, and directly input and train the source data. The first normalization layer concentrates data distribution in an activation range of a tanh activation function, so that training efficiency can be improved; for error values, the error values are distributed outside the activation range of the tanh function, the influence degree on the training of the neural network model is low, overfitting can be prevented, and the generalization capability is improved. The improved output layer can further improve the training efficiency of the neural network model.
The embodiment of the invention is suitable for energy consumption prediction of a central air-conditioning system through the improved neural network model. The neural network model can reduce the workload of feature engineering of general machine learning, improve the training efficiency of the neural network model, prevent the over-fitting phenomenon, and has higher prediction accuracy and stronger generalization capability.
According to the embodiment of the invention, historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; the neural network model is adopted to predict the energy consumption of the central air-conditioning system, so that the training efficiency of the neural network model can be improved, and the prediction accuracy and generalization capability of the neural network model can be improved.
Fig. 7 is a schematic structural diagram illustrating an energy consumption prediction apparatus of a central air conditioning system according to an embodiment of the present invention. As shown in fig. 7, the central air-conditioning system energy consumption prediction apparatus includes: a training data acquisition unit 701, a model training unit 702, a model output unit 703, and an energy consumption prediction unit 704. Wherein:
the training data acquisition unit 701 is used for acquiring historical influence factors related to energy consumption of the central air conditioning system and corresponding historical energy consumption data; the model training unit 702 is configured to train a neural network model according to the historical influence factors, and output corresponding predicted energy consumption data, where the neural network model is composed of an input layer, a hidden layer including a full connection layer, a first normalization layer, and an activation layer, and an output layer; the model output unit 703 is configured to adjust a model parameter of the neural network model so that an error between the predicted energy consumption data and the historical energy consumption data satisfies a preset precision, and output the neural network model; the energy consumption prediction unit 704 is configured to predict the energy consumption of the central air conditioning system by using the neural network model.
In an alternative manner, the training data obtaining unit 701 is configured to: acquiring historical setting parameters, historical outdoor environment and historical indoor demand factors of the central air-conditioning system; selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors; and acquiring the historical energy consumption data corresponding to the historical influence factors.
In an alternative manner, the training data obtaining unit 701 is configured to: acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower; acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation; acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
In an alternative approach, the model training unit 702 is configured to: inputting the historical influencing factors into the input layer of a neural network model; carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer; normalizing the output data of the full connection layer through the first normalization layer of the hidden layer; performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer; and outputting the corresponding predicted energy consumption data through the output layer.
In an alternative approach, the model training unit 702 is configured to: adding a second normalization layer after the input layer, and performing normalization processing on the input historical influence factors through the second normalization layer; before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer.
In an alternative manner, the model training unit 702 is further configured to: normalizing the output data by applying the following relational expression, and concentrating most of the output data in a preset range:
y＝(x–xmean)/(δ+ε)，
wherein y is the result after standardization treatment, and xmeanIs the average of said historical influence of the input, δ is the input's placeThe standard deviation of the historical influencing factors, epsilon, is a very small value.
In an alternative manner, the model training unit 702 is further configured to: activating the output data in the preset range through the activation layer; and regarding the output data outside the preset range as error data.
According to the embodiment of the invention, historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; the neural network model is adopted to predict the energy consumption of the central air-conditioning system, so that the training efficiency of the neural network model can be improved, and the prediction accuracy and generalization capability of the neural network model can be improved.
The embodiment of the invention provides a nonvolatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the energy consumption prediction method of the central air-conditioning system in any method embodiment.
The executable instructions may be specifically configured to cause the processor to:
acquiring historical influence factors related to energy consumption of a central air-conditioning system and corresponding historical energy consumption data;
training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer;
adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model;
and predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
In an alternative, the executable instructions cause the processor to:
acquiring historical setting parameters, historical outdoor environment and historical indoor demand factors of the central air-conditioning system;
selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors;
and acquiring the historical energy consumption data corresponding to the historical influence factors.
In an alternative, the executable instructions cause the processor to:
acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower;
acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation;
acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
In an alternative, the executable instructions cause the processor to:
inputting the historical influencing factors into the input layer of a neural network model;
carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer;
normalizing the output data of the full connection layer through the first normalization layer of the hidden layer;
performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer;
and outputting the corresponding predicted energy consumption data through the output layer.
In an alternative, the executable instructions cause the processor to:
adding a second normalization layer after the input layer, and performing normalization processing on the input historical influence factors through the second normalization layer;
before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer.
In an alternative, the executable instructions cause the processor to:
normalizing the output data by applying the following relational expression, and concentrating most of the output data in a preset range:
y＝(x–xmean)/(δ+ε)，
wherein y is the result after standardization treatment, and xmeanIs the average value of the historical influencing factors input, delta is the standard deviation of the historical influencing factors input, and epsilon is a very small value.
In an alternative, the executable instructions cause the processor to:
activating the output data in the preset range through the activation layer;
and regarding the output data outside the preset range as error data.
According to the embodiment of the invention, historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; the neural network model is adopted to predict the energy consumption of the central air-conditioning system, so that the training efficiency of the neural network model can be improved, and the prediction accuracy and generalization capability of the neural network model can be improved.
Embodiments of the present invention provide a computer program product comprising a computer program stored on a computer storage medium, the computer program comprising program instructions that, when executed by a computer, cause the computer to perform the central air conditioning system energy consumption prediction method in any of the above-mentioned method embodiments.
The executable instructions may be specifically configured to cause the processor to:
acquiring historical influence factors related to energy consumption of a central air-conditioning system and corresponding historical energy consumption data;
training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer;
adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model;
and predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
In an alternative, the executable instructions cause the processor to:
acquiring historical setting parameters, historical outdoor environment and historical indoor demand factors of the central air-conditioning system;
selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors;
and acquiring the historical energy consumption data corresponding to the historical influence factors.
In an alternative, the executable instructions cause the processor to:
acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower;
acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation;
acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
In an alternative, the executable instructions cause the processor to:
inputting the historical influencing factors into the input layer of a neural network model;
carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer;
normalizing the output data of the full connection layer through the first normalization layer of the hidden layer;
performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer;
and outputting the corresponding predicted energy consumption data through the output layer.
In an alternative, the executable instructions cause the processor to:
adding a second normalization layer after the input layer, and performing normalization processing on the input historical influence factors through the second normalization layer;
before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer.
In an alternative, the executable instructions cause the processor to:
normalizing the output data by applying the following relational expression, and concentrating most of the output data in a preset range:
y＝(x–xmean)/(δ+ε)，
wherein y is the result after standardization treatment, and xmeanIs the average value of the historical influencing factors input, delta is the standard deviation of the historical influencing factors input, and epsilon is a very small value.
In an alternative, the executable instructions cause the processor to:
activating the output data in the preset range through the activation layer;
and regarding the output data outside the preset range as error data.
According to the embodiment of the invention, historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; the neural network model is adopted to predict the energy consumption of the central air-conditioning system, so that the training efficiency of the neural network model can be improved, and the prediction accuracy and generalization capability of the neural network model can be improved.
Fig. 8 is a schematic structural diagram of a computing device according to an embodiment of the present invention, and a specific embodiment of the present invention does not limit a specific implementation of the device.
As shown in fig. 8, the computing device may include: a processor (processor)802, a Communications Interface 804, a memory 806, and a communication bus 808.
Wherein: the processor 802, communication interface 804, and memory 806 communicate with one another via a communication bus 808. A communication interface 804 for communicating with network elements of other devices, such as clients or other servers. The processor 802 is configured to execute the program 810, and may specifically execute the relevant steps in the above-described central air conditioning system energy consumption prediction method embodiment.
In particular, the program 810 may include program code comprising computer operating instructions.
The processor 802 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present invention. The one or each processor included in the device may be the same type of processor, such as one or each CPU; or may be different types of processors such as one or each CPU and one or each ASIC.
The memory 806 stores a program 810. The memory 806 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 810 may be specifically configured to cause the processor 802 to perform the following operations:
acquiring historical influence factors related to energy consumption of a central air-conditioning system and corresponding historical energy consumption data;
training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer;
adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model;
and predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
In an alternative, the program 810 causes the processor to:
acquiring historical setting parameters, historical outdoor environment and historical indoor demand factors of the central air-conditioning system;
selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors;
and acquiring the historical energy consumption data corresponding to the historical influence factors.
In an alternative, the program 810 causes the processor to:
acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower;
acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation;
acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
In an alternative, the program 810 causes the processor to:
inputting the historical influencing factors into the input layer of a neural network model;
carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer;
normalizing the output data of the full connection layer through the first normalization layer of the hidden layer;
performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer;
and outputting the corresponding predicted energy consumption data through the output layer.
In an alternative, the program 810 causes the processor to:
adding a second normalization layer after the input layer, and performing normalization processing on the input historical influence factors through the second normalization layer;
before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer.
In an alternative, the program 810 causes the processor to:
normalizing the output data by applying the following relational expression, and concentrating most of the output data in a preset range:
y＝(x–xmean)/(δ+ε)，
wherein y is the result after standardization treatment, and xmeanIs the average value of the historical influencing factors input, delta is the standard deviation of the historical influencing factors input, and epsilon is a very small value.
In an alternative, the program 810 causes the processor to:
activating the output data in the preset range through the activation layer;
and regarding the output data outside the preset range as error data.
According to the embodiment of the invention, historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data are obtained; training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer; adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model; the neural network model is adopted to predict the energy consumption of the central air-conditioning system, so that the training efficiency of the neural network model can be improved, and the prediction accuracy and generalization capability of the neural network model can be improved.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the invention and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specified otherwise.
Claims (10)
1. A central air conditioning system energy consumption prediction method is characterized by comprising the following steps:
acquiring historical influence factors related to energy consumption of a central air-conditioning system and corresponding historical energy consumption data;
training a neural network model according to the historical influence factors, and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a first normalization layer and an activation layer, and an output layer;
adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model;
and predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
2. The method of claim 1, wherein the obtaining historical impact factors related to central air conditioning system energy consumption and corresponding historical energy consumption data comprises:
acquiring historical setting parameters, historical outdoor environment and historical indoor demand factors of the central air-conditioning system;
selecting part or all of the historical setting parameters, the historical outdoor environment factors and the historical indoor demand factors as the historical influence factors;
and acquiring the historical energy consumption data corresponding to the historical influence factors.
3. The method of claim 2, wherein the obtaining of historical setup parameters, historical outdoor environmental factors, and historical indoor demand factors for the central air conditioning system comprises:
acquiring setting parameters of an air conditioner host of the central air conditioning system, setting parameters of a chilled water pump, setting parameters of an air conditioner tail end, setting parameters of a cooling water pump and setting parameters of a cooling tower;
acquiring historical outdoor environmental factors of the central air-conditioning system including but not limited to outdoor temperature, outdoor relative humidity, wind speed and solar radiation;
acquiring the historical indoor demand factors of the central air conditioning system including but not limited to indoor heat dissipation capacity, indoor environment temperature control value and indoor environment humidity control value.
4. The method of claim 1, wherein training a neural network model based on the historical influencing factors to output corresponding predicted energy consumption data comprises:
inputting the historical influencing factors into the input layer of a neural network model;
carrying out full connection processing on the history influence factors through the full connection layer of the hidden layer;
normalizing the output data of the full connection layer through the first normalization layer of the hidden layer;
performing activation processing on the output data after the normalization processing through the activation layer of the hidden layer;
and outputting the corresponding predicted energy consumption data through the output layer.
5. The method of claim 4, wherein inputting the historical influence factors into the input layer of a neural network model, further comprises:
adding a second normalization layer after the input layer, and performing normalization processing on the input historical influence factors through the second normalization layer;
before outputting the corresponding predicted energy consumption data through the output layer, the method further includes: and adding a third normalization layer before the output layer, and normalizing the data output by the hidden layer through the third normalization layer.
6. The method of claim 4, wherein the normalizing the output data of the fully-connected layer by the first normalization layer of the hidden layer comprises:
normalizing the output data by applying the following relational expression, and concentrating most of the output data in a preset range:
y＝(x–xmean)/(δ+ε)，
wherein y is the result after standardization treatment, and xmeanIs the average value of the historical influencing factors input, delta is the standard deviation of the historical influencing factors input, and epsilon is a very small value.
7. The method according to claim 6, wherein the performing, by the active layer of the hidden layer, an activation process on the normalized output data comprises:
activating the output data in the preset range through the activation layer;
and regarding the output data outside the preset range as error data.
8. An energy consumption prediction apparatus for a central air conditioning system, the apparatus comprising:
the training data acquisition unit is used for acquiring historical influence factors related to the energy consumption of the central air-conditioning system and corresponding historical energy consumption data;
the training unit is used for training a neural network model according to the historical influence factors and outputting corresponding predicted energy consumption data, wherein the neural network model consists of an input layer, a hidden layer comprising a full connection layer, a normalization layer and an activation layer, and an output layer;
the model output unit is used for adjusting model parameters of the neural network model to enable the error between the predicted energy consumption data and the historical energy consumption data to meet preset precision, and outputting the neural network model;
and the energy consumption prediction unit is used for predicting the energy consumption of the central air-conditioning system by adopting the neural network model.
9. A computing device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the steps of the central air-conditioning system energy consumption prediction method according to any one of claims 1-7.
10. A computer storage medium having stored therein at least one executable instruction for causing a processor to perform the steps of the central air conditioning system energy consumption prediction method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010725321.3A CN113970170A (en) | 2020-07-24 | 2020-07-24 | Central air-conditioning system energy consumption prediction method and device and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010725321.3A CN113970170A (en) | 2020-07-24 | 2020-07-24 | Central air-conditioning system energy consumption prediction method and device and computing equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113970170A true CN113970170A (en) | 2022-01-25 |
Family
ID=79585864
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010725321.3A Pending CN113970170A (en) | 2020-07-24 | 2020-07-24 | Central air-conditioning system energy consumption prediction method and device and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113970170A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114791160A (en) * | 2022-01-27 | 2022-07-26 | 王艳茜 | Central air conditioner control method and device based on neural network model |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150009375A (en) * | 2013-07-16 | 2015-01-26 | 한국전자통신연구원 | Method and system for predicting power consumption |
| JP2017067427A (en) * | 2015-10-01 | 2017-04-06 | パナソニックＩｐマネジメント株式会社 | Air conditioning control method, air conditioning control device and air conditioning control program |
| 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 |
| CN109945402A (en) * | 2019-03-07 | 2019-06-28 | 东南大学 | A kind of central air conditioning water system power-economizing method |
| CN109959122A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network |
| CN110392515A (en) * | 2019-07-23 | 2019-10-29 | 上海朗绿建筑科技股份有限公司 | A kind of Cooling and Heat Source equipment room energy-conserving control method and system based on historical data |
| CN111037365A (en) * | 2019-12-26 | 2020-04-21 | 大连理工大学 | Cutter state monitoring data set enhancing method based on generative countermeasure network |
| CN111259798A (en) * | 2020-01-16 | 2020-06-09 | 西安电子科技大学 | Modulation signal identification method based on deep learning |
| CN111365828A (en) * | 2020-03-06 | 2020-07-03 | 上海外高桥万国数据科技发展有限公司 | Model prediction control method for realizing energy-saving temperature control of data center by combining machine learning |
-
2020
- 2020-07-24 CN CN202010725321.3A patent/CN113970170A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150009375A (en) * | 2013-07-16 | 2015-01-26 | 한국전자통신연구원 | Method and system for predicting power consumption |
| JP2017067427A (en) * | 2015-10-01 | 2017-04-06 | パナソニックＩｐマネジメント株式会社 | Air conditioning control method, air conditioning control device and air conditioning control program |
| 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 |
| CN109945402A (en) * | 2019-03-07 | 2019-06-28 | 东南大学 | A kind of central air conditioning water system power-economizing method |
| CN109959122A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network |
| CN110392515A (en) * | 2019-07-23 | 2019-10-29 | 上海朗绿建筑科技股份有限公司 | A kind of Cooling and Heat Source equipment room energy-conserving control method and system based on historical data |
| CN111037365A (en) * | 2019-12-26 | 2020-04-21 | 大连理工大学 | Cutter state monitoring data set enhancing method based on generative countermeasure network |
| CN111259798A (en) * | 2020-01-16 | 2020-06-09 | 西安电子科技大学 | Modulation signal identification method based on deep learning |
| CN111365828A (en) * | 2020-03-06 | 2020-07-03 | 上海外高桥万国数据科技发展有限公司 | Model prediction control method for realizing energy-saving temperature control of data center by combining machine learning |
Non-Patent Citations (3)
| Title |
|---|
| 刘玉良等: "《深度学习》", 31 January 2020, 西安电子科技大学出版社, pages: 108 - 110 * |
| 吉安卡洛•扎克尼: "TensorFlow深度学习", 机械工业出版社, pages: 58 - 60 * |
| 魏峥等: "基于机器学习的冷水机组能耗模型辨识方法研究", 《建筑科学》, vol. 34, no. 6, 30 June 2018 (2018-06-30), pages 115 - 122 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114791160A (en) * | 2022-01-27 | 2022-07-26 | 王艳茜 | Central air conditioner control method and device based on neural network model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11507033B2 (en) | HVAC control system with model driven deep learning | |
| CN110285532B (en) | Machine room air conditioner control method, device and system based on artificial intelligence | |
| CN106874581B (en) | Building air conditioner energy consumption prediction method based on BP neural network model | |
| CN113835344B (en) | Control optimization method of equipment, display platform, cloud server and storage medium | |
| US11774923B2 (en) | Augmented deep learning using combined regression and artificial neural network modeling | |
| US20190368759A1 (en) | Systems and methods to control energy consumption efficiency | |
| CN114383299B (en) | Central air-conditioning system operation strategy optimization method based on big data and dynamic simulation | |
| CN112084707A (en) | Refrigeration machine room energy-saving optimization method and system based on variable flow decoupling of chilled water and cooling water | |
| CN107392368B (en) | Meteorological forecast-based office building dynamic heat load combined prediction method | |
| Mahbod et al. | Energy saving evaluation of an energy efficient data center using a model-free reinforcement learning approach | |
| Su et al. | An agent-based distributed real-time optimal control strategy for building HVAC systems for applications in the context of future IoT-based smart sensor networks | |
| CN103912966A (en) | Optimal control method for ground source heat pump refrigerating system | |
| CN113739365A (en) | Central air-conditioning cold station group control energy-saving control method, device, equipment and storage medium | |
| US20210190349A1 (en) | Building control system with zone grouping based on predictive models | |
| CN114065994A (en) | Energy consumption optimization method, device and equipment for air conditioning system and computer storage medium | |
| CN117970986B (en) | Temperature and humidity control method, device and medium of cold and hot system | |
| CN115758912A (en) | Air conditioner energy consumption optimizing system | |
| Zhang et al. | Deep reinforcement learning towards real-world dynamic thermal management of data centers | |
| CN116501104A (en) | Temperature control device and method for liquid cooling heat dissipation system | |
| CN113970170A (en) | Central air-conditioning system energy consumption prediction method and device and computing equipment | |
| Zhang et al. | DRL-S: Toward safe real-world learning of dynamic thermal management in data center | |
| AU2021100960A4 (en) | Artificial Intelligence Based Cooling System for Managing the Energy Efficiency | |
| US20230085072A1 (en) | Hvac control system with model driven deep learning | |
| CN114528749A (en) | Model determination method, model determination device, electronic equipment and storage medium | |
| CN114777325A (en) | Boiler system regulation and control method, model building method, related equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220125 |