CN114279061B - Method and device for controlling air conditioner and electronic equipment - Google Patents
Method and device for controlling air conditioner and electronic equipment Download PDFInfo
- Publication number
- CN114279061B CN114279061B CN202111426134.6A CN202111426134A CN114279061B CN 114279061 B CN114279061 B CN 114279061B CN 202111426134 A CN202111426134 A CN 202111426134A CN 114279061 B CN114279061 B CN 114279061B
- Authority
- CN
- China
- Prior art keywords
- power sequence
- sequence
- predicted
- air conditioner
- network
- 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.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
Landscapes
- Air Conditioning Control Device (AREA)
Abstract
The application discloses a method and device for controlling an air conditioner and electronic equipment. Wherein the method comprises the following steps: acquiring a power sequence of an air conditioner; generating a random sequence according to noise, taking an environmental factor of an air conditioner as a tag vector, and inputting the random sequence and the tag vector into a generation network to obtain a predicted power sequence; inputting the predicted power sequence and the power sequence into a discrimination network to obtain an objective function, and inputting the objective function into a generation network; according to the predicted power sequence and the objective function, the first parameter and the second parameter are adjusted until the first parameter and the second parameter are unchanged; and comparing the first power sequence with the second power sequence and the total power sequence respectively, selecting a target power sequence from the first power sequence and the second power sequence according to the comparison result, and controlling the air conditioner according to the target power sequence. The method and the device solve the technical problems of single load prediction type and low precision in the related technology.
Description
Technical Field
The application relates to the field of artificial intelligence, in particular to a method, a device and electronic equipment for controlling an air conditioner.
Background
With the rapid development of national economy and the continuous improvement of the living standard of people, the rapid increase of air conditioning load has become an important cause of the deterioration of power grid load characteristics and the shortage of power in summer.
The large-scale air-conditioning load of the public building has certain peak regulation capacity, so that the peak regulation potential of the large-scale air-conditioning load of the public building is excavated, the peak load is reduced by adopting a reasonable regulation and control means, and the method has important significance for reducing the pressure of a power grid. In the prior art, with the advent of artificial intelligence, machine learning and deep learning methods have been widely applied to short-term load prediction instead of physical model methods, but with the increase of data volume, the prediction accuracy of the application effect thereof is inferior to that of a neural network model, so that condition generation countermeasure neural network is gradually applied to load prediction. However, in the existing neural network scheme applied to building air conditioner load prediction, the input and output data types of a single network are single, the accuracy of output data is not enough, and errors are larger after the later generation of prediction data is applied.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the application provides a method, a device and electronic equipment for controlling an air conditioner, which are used for at least solving the technical problems of single load prediction type and low precision in the related technology.
According to an aspect of an embodiment of the present application, there is provided a method of controlling an air conditioner, including: acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner; generating a random sequence according to noise, taking an environmental factor of an air conditioner as a tag vector, inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence; inputting the predicted power sequence and the power sequence into a discrimination network in a generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not; according to the predicted power sequence and the objective function, adjusting a first parameter and a second parameter until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter for generating a network, and the second parameter is a parameter for judging the network; and respectively comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence, selecting a target power sequence from the first power sequence and the second power sequence according to the comparison result, and controlling the air conditioner according to the target power sequence.
Optionally, acquiring a power sequence of the air conditioner includes: and acquiring power data of the air conditioner, performing data cleaning on the power data, and converting the power data into a one-dimensional sequence to obtain a power sequence.
Optionally, the sub-power sequence of each subsystem of the air conditioner includes: the system comprises an air conditioning water chilling unit power sequence, an air conditioning chilled water pump system power sequence, an air conditioning cooling tower system power sequence, an air conditioning fan coil system power sequence and an air conditioning cooling water pump system power sequence.
Alternatively, the objective function is 0 or 1, and when the objective function is 0, it indicates that the predicted power sequence is not a real power sequence, and when the objective function is 1, it indicates that the predicted power sequence is a real power sequence.
Optionally, the comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence includes: calculating a first correlation coefficient between the first power sequence and the total power sequence; calculating a second correlation coefficient between the second power sequence and the total power sequence; and determining a target power sequence corresponding to the maximum value according to the maximum value in the first correlation coefficient and the second correlation coefficient, and controlling the air conditioner according to the target power sequence.
Optionally, the first correlation coefficient and the second correlation coefficient range from 0,1, and a correlation closer to 1 indicates a higher correlation and a correlation closer to 0 indicates a lower correlation.
Optionally, the size of the predicted power sequence that generates the network output is calculated from the number of input sequences, the convolution kernel side length, the convolution step size, and the number of fills.
According to another aspect of the embodiments of the present application, there is also provided an apparatus for controlling an air conditioner, including: the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner; the first input module is used for generating a random sequence according to noise, taking an environmental factor of an air conditioner as a tag vector, inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence; the second input module is used for inputting the predicted power sequence and the power sequence into a discrimination network in the generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not; the adjusting module is used for adjusting the first parameter and the second parameter according to the predicted power sequence and the objective function until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter for generating a network, and the second parameter is a parameter for judging the network; and the comparison module is used for comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence respectively, selecting a target power sequence from the first power sequence and the second power sequence according to the comparison result, and controlling the air conditioner according to the target power sequence.
According to still another aspect of the embodiments of the present application, there is also provided an electronic device, including: a memory for storing program instructions; and the processor is connected with the memory and is used for realizing the following functions when executing program instructions: acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner; generating a random sequence according to noise, taking an environmental factor of an air conditioner as a tag vector, inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence; inputting the predicted power sequence and the power sequence into a discrimination network in a generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not; according to the predicted power sequence and the objective function, adjusting a first parameter and a second parameter until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter for generating a network, and the second parameter is a parameter for judging the network; and respectively comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence, selecting a target power sequence from the first power sequence and the second power sequence according to the comparison result, and controlling the air conditioner according to the target power sequence.
According to still another aspect of the embodiments of the present application, there is also provided a nonvolatile storage medium including a stored program, wherein the apparatus in which the nonvolatile storage medium is controlled to execute the above method of controlling an air conditioner when the program runs.
In the embodiment of the application, total power data of an air conditioner and sub-power data of all sub-systems of the air conditioner are obtained, a tag vector is determined according to environmental factors of the air conditioner, a plurality of groups of generating networks are used, a random sequence formed according to noise and the tag vector are input into different generating networks to obtain a predicted power sequence, wherein the predicted power sequence comprises the predicted total power sequence and the predicted sub-power sequence, the predicted power sequence output by the generating networks is input into a judging network to judge, a target function is output, then the target function is input into the generating network, parameters of the generating network and the judging network are continuously adjusted until the generating network is optimized, a first power sequence obtained by summing the finally generated predicted sub-power sequences and a second power sequence corresponding to the finally generated predicted total power sequence are respectively compared with the total power sequence, and according to a comparison result, the target power sequence is selected from the first power sequence and the second power sequence to control the air conditioner, and the aim of optimizing the structure of the neural network is achieved, so that the technical effects of carrying out power prediction on different power types of the air conditioner are achieved, and the problems of single load prediction type and low precision in related technologies are solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
fig. 1 is a hardware block diagram of a computer terminal for implementing a method of controlling an air conditioner according to the present application;
FIG. 2 is a flow chart of a method of controlling an air conditioner according to an embodiment of the present application;
FIG. 3a is a system architecture diagram of a first generation method of reactive network training according to embodiments of the present application;
FIG. 3b is a system architecture diagram of a second method of generating countermeasure network training according to embodiments of the present application;
FIG. 3c is a system architecture diagram of a third method of generating countermeasure network training according to embodiments of the present application;
FIG. 3d is a system architecture diagram of a fourth generation countermeasure network training methodology in accordance with an embodiment of the present application;
FIG. 3e is a system architecture diagram of a fifth generation countermeasure network training methodology in accordance with an embodiment of the present application;
FIG. 3f is a system architecture diagram of a sixth generation countermeasure network training methodology in accordance with an embodiment of the present application;
FIG. 3g is a block diagram of a generation network according to an embodiment of the present application;
FIG. 3h is a block diagram of a discrimination network according to an embodiment of the present application;
FIG. 4 is a flow chart of determining a target power sequence according to an embodiment of the present application;
fig. 5 is a block diagram of an apparatus for controlling an air conditioner according to an embodiment of the present application.
Detailed Description
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
According to the embodiments of the present application, there is provided a method embodiment of controlling an air conditioner, it should be noted that the steps illustrated in the flowcharts of the drawings may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different from that herein.
The method embodiments provided by the embodiments of the present application may be performed in a mobile terminal, a computer terminal, or similar computing device. Fig. 1 shows a hardware block diagram of a computer terminal (or electronic device) for implementing a method of controlling an air conditioner. As shown in fig. 1, the computer terminal 10 (or electronic device 10) may include one or more processors 102 (shown as 102a, 102b, … …,102 n) which may include, but are not limited to, a microprocessor MCU or a processing device such as a programmable logic device FPGA, a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, the method may further include: a display, an input/output interface (I/O interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of the I/O interface), a network interface, a power supply, and/or a camera. It will be appreciated by those of ordinary skill in the art that the configuration shown in fig. 1 is merely illustrative and is not intended to limit the configuration of the electronic device described above. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
It should be noted that the one or more processors 102 and/or other data processing circuits described above may be referred to generally herein as "data processing circuits. The data processing circuit may be embodied in whole or in part in software, hardware, firmware, or any other combination. Furthermore, the data processing circuitry may be a single stand-alone processing module, or incorporated, in whole or in part, into any of the other elements in the computer terminal 10 (or electronic device). As referred to in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of the path of the variable resistor termination to interface).
The memory 104 may be used to store software programs and modules of application software, such as program instructions/data storage devices corresponding to the method for controlling an air conditioner in the embodiments of the present application, and the processor 102 executes the software programs and modules stored in the memory 104, thereby performing various functional applications and data processing, that is, implementing the method for controlling an air conditioner described above. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission module 106 is used to receive or transmit data via a network. The specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module for communicating with the internet wirelessly.
The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computer terminal 10 (or electronic device).
It should be noted here that, in some alternative embodiments, the computer device (or the electronic device) shown in fig. 1 described above may include hardware elements (including circuits), software elements (including computer code stored on a computer readable medium), or a combination of both hardware elements and software elements. It should be noted that fig. 1 is only one example of a specific example, and is intended to illustrate the types of components that may be present in the computer device (or electronic device) described above.
In the above-described operating environment, the present application provides a method of controlling an air conditioner as shown in fig. 2. Fig. 2 is a method for controlling an air conditioner according to an embodiment of the present application, as shown in fig. 2, the method including the steps of:
step S202, acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner;
step S204, generating a random sequence according to noise, taking environmental factors of an air conditioner as tag vectors, inputting the random sequence and the tag vectors into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence;
step S206, inputting the predicted power sequence and the power sequence into a discrimination network in a generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence;
step S208, according to the predicted power sequence and the objective function, the first parameter and the second parameter are adjusted until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter for generating a network, and the second parameter is a parameter for judging the network;
Step S210, comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence, selecting a target power sequence from the first power sequence and the second power sequence according to the comparison result, and controlling the air conditioner according to the target power sequence.
According to the method, total power data of an air conditioner and sub-power data of all subsystems of the air conditioner are obtained, a tag vector is determined according to environmental factors of the air conditioner, a plurality of groups of generating networks are used, random sequences formed according to noise and tag vectors are input into different generating networks to obtain a predicted power sequence, wherein the predicted power sequence comprises the predicted total power sequence and the predicted sub-power sequence, the predicted power sequence output by the generating networks is input into a judging network to judge, an objective function is output, the objective function is input into the generating network, parameters of the generating network and the judging network are continuously adjusted until the generating network is optimized, a first power sequence obtained by summing the finally generated predicted sub-power sequences and a second power sequence corresponding to the finally generated predicted total power sequence are respectively compared with the total power sequence, and according to comparison results, the target power sequence is selected from the first power sequence and the second power sequence, and the air conditioner is controlled according to the target power sequence, so that the aim of optimizing a training model is achieved, and the technical effect of carrying out power prediction on different power types of the air conditioner is achieved, and the problems of single load prediction type and low precision in the related technology are solved.
Optionally, in step S202, a power sequence of the air conditioner is acquired, which specifically includes the following steps: the method comprises the steps of obtaining power data of an air conditioner, wherein the power data are the power used by the air conditioner in a building every minute in a day, the power data comprise total power data x1 of the air conditioner and sub-power data of all sub-systems of the air conditioner, the sub-power data comprise power data x2 of an air conditioner water chilling unit, power data x3 of an air conditioner chilled water pump system, power data x4 of an air conditioner cooling tower system, power data x5 of an air conditioner fan coil system and power data x6 of the air conditioner cooling water pump system, all the power data in the power data x1 to x6 are converted into one-dimensional sequences after being subjected to data cleaning, different power sequences are obtained, and the size of the power sequences can be M multiplied by 1, and M is the size of data quantity.
Optionally, in step S202, the sub-power sequence of each subsystem of the air conditioner specifically includes the following: the system comprises an air conditioning water chilling unit power sequence, an air conditioning chilled water pump system power sequence, an air conditioning cooling tower system power sequence, an air conditioning fan coil system power sequence and an air conditioning cooling water pump system power sequence.
Optionally, in step S204, the environmental factors in which the air conditioner is located include the time T for which the air conditioner is operated, the temperature T of the environment, and the humidity H of the environment, and the tag vector is denoted as y, then y may be expressed as 
Wherein k is more than or equal to 1 and less than or equal to M.
3 a-3 f are system architecture diagrams for generating a reactive network training method, as shown in FIG. 3 a-3 f, a random sequence generated by noise z1 and a label vector y are input into a generating network G1 to obtain a predicted total power sequence G (z 1, y), and the sequence size of G (z 1, y) is Mx1; inputting the random sequence generated by the noise z2 and the tag vector y into a generating network G2 to obtain a predicted air conditioner chiller power sequence G (z 2, y), wherein the sequence size of the G (z 2, y) is M multiplied by 1; and similarly, respectively inputting a random sequence generated by the noise z3 to z6 and the tag vector y into corresponding generating networks G3 to G6 to obtain a corresponding predicted air conditioner chilled water pump system power sequence G (z 3, y), a predicted air conditioner cooling tower system power sequence G (z 4, y), a predicted air conditioner fan coil system power sequence G (z 5, y) and a predicted air conditioner cooling water pump system power sequence G (z 6, y), wherein the size of the input sequence and the size of the output sequence are M multiplied by 1.
Optionally, in step S204, the size of the predicted power sequence of the generated network output is calculated from the number of input sequences, the convolution kernel side length, the convolution step size and the number of fills. In the present embodiment, the generation networks G1 to G6 are constituted by convolution layers and deconvolution layers.
FIG. 3g is a block diagram of a generating network, as shown in FIG. 3g, the random sequence data formed by input noise z1, z2, z3, z4, z5, z6 and a tag vector y composed of the running time of the air conditioner, the temperature of the environment and the humidity of the environment, the random sequence size is Mx1, the size of the receptive field is K1x1 through a first layer convolution layer, the convolution step size is S1, and the output sequence size of the first layer convolution layer is
In the convolutional neural network, the area size of an input layer corresponding to one element in an output result of a certain layer is determined and is called a receptive field, the larger the convolution step length is, the larger the receptive field is, the output sequence is input into a second layer of convolution, the receptive field size is set to be K2×1, the convolution step length is set to be S2, and the output sequence size of the output sequence of the second layer of convolution layer is set to be>
Denoted as M2×1, the sequence M2×1 is input into the third upsampling layer, and the sequence size is outputAnd finally, inputting the output sequence of the third layer into the up-sampling layer of the fourth layer, wherein the output sequence has the size of Mx1. The constructed generation network and parameter settings are only exemplary references, the specific internal structure of which is modified according to the actual situation, and the pooling layer can be added after the convolution layer according to the actual situation. Where M represents the size of the data amount of the input sequence, K represents the convolution kernel side length, S represents the convolution step length, and P represents the padding number, typically 0.
Optionally, in step S206, the predicted total power sequence G (z 1, y) output by the generating network G1 is taken as an input of the markov discriminating network D1, meanwhile, the total power data x1 of the air conditioner is input to D1, the D1 performs probability evaluation on G (z 1, y) and x1, and an objective function R (x 1, y) is output, where the objective function is used to indicate whether the predicted power sequence is a real power sequence, the objective function is divided into 0 or 1, and when the objective function is 0, it indicates that the predicted power sequence is not a real power sequence, and when the objective function is 1, it indicates that the predicted power sequence is a real power sequence.
Fig. 3h is a block diagram of a discrimination network, and as shown in fig. 3h, the predicted air-conditioning chiller power sequence G (z 2, y) output by the generation network G2 is used as an input of the markov discrimination network D2, meanwhile, a power sequence corresponding to the air-conditioning chiller power data x2 is input to D2, and the D2 performs probability evaluation on the power sequence corresponding to the G (z 2, y) and x2, and outputs an objective function R (x 2, y). Similarly, the power sequences corresponding to G (z 3, y) and x3, the power sequences corresponding to G (z 4, y) and x4, the power sequences corresponding to G (z 5, y) and x5, and the power sequences corresponding to G (z 6, y) and x6 are respectively input into D3 to D6, and the power sequences corresponding to G (z 3, y) and x3, the power sequences corresponding to G (z 4, y) and x4, the power sequences corresponding to G (z 5, y) and x5, and the power sequences corresponding to G (z 6, y) and x6 are respectively subjected to probability evaluation by D3 to D6, so that corresponding objective functions R (x 3, y), R (x 4, y), R (x 5, y), and R (x 6, y) are output.
Inputting the obtained R (x 1, y) into a loss function g×1, where the loss function g×1 has the formula:
in the above formula, G is a generation network, the discrimination network R is a generic term for the discrimination networks D1 to D6, λ is a parameter of the loss function, L CGAN Taking logarithms of the probability that the predicted power sequence judged by the judging network is a real power sequence and the probability that the predicted power sequence is an unreal power sequence, and then respectively calculating expected sums; l (L) L1 Improving the quality of the production of a production network for adding L1 distance loss to the production network, wherein L CGAN The formula of (2) is as follows:
L CGAN (G,R)＝E x,y [logR(x,y)]+E x,z [log(1-R(x,G(x,z)))]
at L CGAN In the formula of (a), x is power data of an air conditioner, z is noise of an input generation network G, y is a label vector formed by factors of an environment where the air conditioner is located, G (x, z) represents a predicted power sequence generated by the generation network G, R (x, y) represents probability of judging whether the predicted power sequence is a real power sequence or not by a judgment network R, and E represents an expected value.
L L1 The formula of (2) is as follows:
L L1 (G)＝E x,y,z [||y-G(x,z)|| 1 ]
obtaining a loss function G (1) according to the formula, inputting G (1) into a generating network G1, generating a new predicted total power sequence G (z 1, y) by G1, taking G (z 1, y) as an input of a Markov judging network D1 according to the step, inputting total power data x1 of an air conditioner into D1, carrying out probability evaluation on G (z 1, y) and x1 by D1, outputting a new objective function R (x 1, y), adjusting a first parameter corresponding to the generating network G1 and a second parameter corresponding to the judging network D1 according to the output of the generating network and the judging network, so that the judging network maximizes a function (log R (x), log (1-R (G (z))), and minimizes a function min (log (1-R (G (z)))) by the generator until the generating network G1 reaches optimization, namely, when the parameters of the generating network G1 do not change, and the generating network model corresponding to the generating network G1 is finished.
Similarly, R (x 2, y) is input to the loss function g×2, R (x 3, y) is input to the loss function g×3, R (x 4, y) is input to the loss function g×4, R (x 5, y) is input to the loss function g×5, R (x 6, y) is input to the loss function g×6, parameters corresponding to the generation networks G2 to G6 are adjusted, parameters corresponding to the networks D2 to D6 are determined, and until the parameters corresponding to the generation networks G2 to G6 do not change, the generation countermeasure network model corresponding to the generation networks G2 to G6 ends training.
In the discrimination networks D1 to D6, the internal structure is as follows: the markov discrimination network used in the embodiment of the present application is composed of a convolution layer and a full connection layer, and generates a predicted power sequence G (z, y) output in the network as an input of the discrimination network, and inputs the predicted power sequence G (z, y) into a markov discrimination network R, wherein the discrimination network R is a generic term of discrimination networks D1 to D6, the input sequence size is mx 1, and simultaneously, the acquired power data x of the air conditioner is input into the corresponding markov discrimination network R, x is a generic term of x1 to x6, the input sequence size is mx 1, the input sequence corresponding to x is input into a first layer convolution layer, the set receptive field size is k3x1, the convolution step size is S3, and the lakyrelu activation function is used, and the example is normalized, and the output sequence size is 
The method is characterized by comprising the following steps: m3×1, inputting the output sequence of the first layer convolution layer into the second layer convolution layer, setting the receptive field size as K4×1, the convolution step length as S4, using the LeakyRelu activation function, and normalizing the example, wherein the output sequence size is +.>
The method is characterized by comprising the following steps: m4×1, inputting the output sequence of the second layer convolution layer into the third layer convolution layer, setting the receptive field size as K5×1, the convolution step length as S5, using the LeakyRelu activation function, and normalizing the example, wherein the output sequence size is->
The method is characterized by comprising the following steps: m5×1, inputting the output sequence of the third layer convolution layer into the fourth layer convolution layer, setting the receptive field size as K6×1, the convolution step length as S6,using the LeakyRelu activation function and instance normalization, the output sequence size was +.>
The method is characterized by comprising the following steps: and M6 multiplied by 1, inputting the output sequence of the fourth convolution layer into the fifth full-connection layer, wherein the number of neurons is 1, and the activation function is a Sigmod function. Wherein, the liquid crystal display device comprises a liquid crystal display device,
the LeakyReLu function is:
the Sigmod function is:
in the LeakyReLu function and the Sigmod function, i is the value of the output sequence.
In the above process, the output tensor of the convolution layer has the following size:
o represents the size of the output sequence, M represents the size of the data amount of the input sequence, K represents the convolution kernel side length, S represents the convolution step length, P represents the filling number, which is generally 0, and the discrimination network R is the generic term for the discrimination networks D1 to D6.
It should be noted that the markov discrimination network and the parameter setting constructed in the embodiments of the present application are only exemplary references, and the specific internal structure of the markov discrimination network and the parameter setting are modified according to actual situations.
Optionally, in step S210, the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence are compared with the total power sequence, respectively, as shown in fig. 4, and specifically includes the following steps:
step S402, calculating a first correlation coefficient between a first power sequence and a total power sequence;
step S404, calculating a second correlation coefficient between the second power sequence and the total power sequence;
step S406, determining a target power sequence corresponding to the maximum value according to the maximum value in the first correlation coefficient and the second correlation coefficient, and controlling the air conditioner according to the target power sequence.
In the step S402, the finally generated predicted sub-power sequences G (z 2, y), G (z 3, y), G (z 4, y), G (z 5, y), G (z 6, y) are input into the following summation formula:
G total ＝G(z2,y)+G(z3,y)+G(z4,y)+G(z5,y)+G(z6,y)
g according to the formula total For the first power sequence, the first power sequence is an air conditioner predicted total power sequence G obtained by using subsystem power prediction total Calculating a first power sequence G by pearson correlation coefficient analysis total And a first correlation coefficient between the total power sequences, wherein the total power sequences are one-dimensional power sequences after data cleaning according to the total power data x1 of the air conditioner, and the pearson correlation coefficient is calculated by the following formula:
in the above formula, X is the first power sequence, Y is the total power sequence, E ((X-E (X)) (Y-E (Y))) is referred to as the covariance of the random variables X and Y, denoted cov (X, Y), i.e., cov (X, Y) =e ((X-E (X)) (Y-E (Y))), and the quotient of the covariance and standard deviation between the two variables is referred to as the correlation coefficient of the random variables X and Y, denoted ρ.
In the step S404, a second correlation coefficient between the finally generated predicted total power sequence G (z 1, y) and the total power sequence is calculated according to the pearson correlation coefficient formula, wherein the total power sequence is based on the total power data x of the air conditioner 1 And carrying out one-dimensional power sequence after data cleaning.
Determining the maximum of the first and second correlation coefficients by comparing the first and second correlation coefficients, e.g.The first power sequence G is determined if the value of the first correlation coefficient is larger than the value of the second correlation coefficient total As a target power sequence; if the value of the second correlation coefficient is larger than that of the first correlation coefficient, the second power sequence, namely the finally generated predicted total power sequence G (z 1, y) is used as a target power sequence, and the control and scheduling decision of the air conditioner is carried out according to the target power sequence.
In the above steps S402 to S406, the range of the first correlation coefficient and the second correlation coefficient is [0,1], and the closer the correlation coefficient is to 1, the higher the correlation between the first power sequence or the second power sequence and the total power sequence is, the more accurate the calculation result is, and the closer the correlation is to 0, the lower the correlation is.
It should be noted that, the size of the input/output sequence in the embodiment of the present application is only referred to, and the specific size of the input sequence is modified according to the size of the actual acquired data volume.
In the embodiment of the application, the structure of the neural network is optimized, the power input of the air conditioning subsystem is increased, and the process of the Pearson correlation coefficient analysis method is added, so that the output predicted power type is not single any more, the accuracy of the total power prediction of the air conditioner of the final building is improved, and the power prediction of the generated air conditioning subsystem is better so that the working modes of working equipment of all the subsystems of the air conditioner are optimized.
Fig. 5 is a structural view of an apparatus for controlling an air conditioner according to an embodiment of the present application, as shown in fig. 5, the apparatus including:
the acquiring module 50 is configured to acquire a power sequence of the air conditioner, where the power sequence includes a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner;
the first input module 52 is configured to generate a random sequence according to noise, take an environmental factor where an air conditioner is located as a tag vector, input the random sequence and the tag vector into a generation network in a generation countermeasure network, and obtain a predicted power sequence, where the predicted power sequence includes a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence;
a second input module 54, configured to input the predicted power sequence and the power sequence into a discrimination network in a generating countermeasure network, obtain an objective function, and input the objective function into the generating network, where the objective function is used to represent whether the predicted power sequence is a real power sequence;
the adjusting module 56 is configured to adjust a first parameter and a second parameter according to the predicted power sequence and the objective function until the first parameter and the second parameter do not change, where the first parameter is a parameter for generating a network, and the second parameter is a parameter for discriminating the network;
And the comparison module 58 is configured to compare the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence, select a target power sequence from the first power sequence and the second power sequence according to the comparison result, and control the air conditioner according to the target power sequence.
In the acquisition module 50, acquiring a power sequence of an air conditioner includes: and acquiring power data of the air conditioner, performing data cleaning on the power data, and converting the power data into a one-dimensional sequence to obtain a power sequence.
In the acquisition module 50, the sub-power sequence of each subsystem of the air conditioner includes: the system comprises an air conditioning water chilling unit power sequence, an air conditioning chilled water pump system power sequence, an air conditioning cooling tower system power sequence, an air conditioning fan coil system power sequence and an air conditioning cooling water pump system power sequence.
In the first input module 52, the magnitude of the predicted power sequence that generates the network output is calculated from the number of input sequences, the convolution kernel side length, the convolution step size, and the number of fills.
In the second input module 54, the objective function is 0 or 1, and when the objective function is 0, it indicates that the predicted power sequence is not a real power sequence, and when the objective function is 1, it indicates that the predicted power sequence is a real power sequence.
In the comparing module 58, the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence are compared with the total power sequence respectively, and the following process is specifically required to be executed: calculating a first correlation coefficient between the first power sequence and the total power sequence; calculating a second correlation coefficient between the second power sequence and the total power sequence; and determining a target power sequence corresponding to the maximum value according to the maximum value in the first correlation coefficient and the second correlation coefficient, and controlling the air conditioner according to the target power sequence.
In the comparison module 58, the first correlation coefficient and the second correlation coefficient range to be 0,1, and a correlation closer to 1 indicates a higher correlation and a correlation closer to 0 indicates a lower correlation.
It should be noted that, the apparatus for controlling an air conditioner shown in fig. 5 is used to execute the method for controlling an air conditioner shown in fig. 2-4, so the explanation of the method for controlling an air conditioner is also applicable to the apparatus for controlling an air conditioner, and is not repeated here.
The embodiment of the application also provides a nonvolatile storage medium, which comprises a stored program, wherein when the program runs, the equipment where the nonvolatile storage medium is controlled to execute the following method for controlling the air conditioner:
Acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner;
generating a random sequence according to noise, taking an environmental factor of an air conditioner as a tag vector, inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence;
inputting the predicted power sequence and the power sequence into a discrimination network in a generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not;
according to the predicted power sequence and the objective function, adjusting a first parameter and a second parameter until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter for generating a network, and the second parameter is a parameter for judging the network;
and respectively comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence, selecting a target power sequence from the first power sequence and the second power sequence according to the comparison result, and controlling the air conditioner according to the target power sequence.
In the above method, acquiring the power sequence of the air conditioner includes: and acquiring power data of the air conditioner, performing data cleaning on the power data, and converting the power data into a one-dimensional sequence to obtain a power sequence.
A sub-power sequence of each subsystem of an air conditioner, comprising: the system comprises an air conditioning water chilling unit power sequence, an air conditioning chilled water pump system power sequence, an air conditioning cooling tower system power sequence, an air conditioning fan coil system power sequence and an air conditioning cooling water pump system power sequence.
In the generating network, the size of the predicted power sequence output by the generating network is calculated by the number of input sequences, the convolution kernel side length, the convolution step length and the filling number.
In the objective function generated by the discrimination network, the objective function is 0 or 1, and when the objective function is 0, it indicates that the predicted power sequence is not a real power sequence, and when the objective function is 1, it indicates that the predicted power sequence is a real power sequence.
And respectively comparing the first power sequence obtained by summing the finally generated predicted sub-power sequences and the second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence, wherein the method comprises the following steps: calculating a first correlation coefficient between the first power sequence and the total power sequence; calculating a second correlation coefficient between the second power sequence and the total power sequence; and determining a target power sequence corresponding to the maximum value according to the maximum value in the first correlation coefficient and the second correlation coefficient, and controlling the air conditioner according to the target power sequence.
In the above process, the range of the first correlation coefficient and the second correlation coefficient is [0,1], and the closer the correlation coefficient is to 1, the higher the correlation is, and the closer the correlation is to 0, the lower the correlation is.
The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application and are intended to be comprehended within the scope of the present application.
Claims (9)
1. A method of controlling an air conditioner, comprising:
acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner;
generating a random sequence according to noise, taking an environmental factor of the air conditioner as a tag vector, and inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence;
inputting the predicted power sequence and the power sequence into a discrimination network in the generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not;
According to the predicted power sequence and the objective function, a first parameter and a second parameter are adjusted until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter of the generated network, and the second parameter is a parameter of the judging network;
comparing a first power sequence obtained by summing the finally generated predicted sub-power sequences and a second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence respectively, selecting a target power sequence from the first power sequence and the second power sequence according to a comparison result, and controlling the air conditioner according to the target power sequence;
the label vector comprises the running time of the air conditioner, the temperature of the environment and the humidity of the environment; the size of the predicted power sequence output by the generating network is calculated by the number of input sequences, the convolution kernel edge length, the convolution step length and the filling number; the final generated predicted sub-power sequence and the final generated predicted total power sequence are sequences obtained by continuously adjusting parameters of a generating network and a judging network until the generating network is optimized.
2. The method of claim 1, wherein the obtaining the power sequence of the air conditioner comprises: and acquiring power data of the air conditioner, performing data cleaning on the power data, and converting the power data into a one-dimensional sequence to obtain the power sequence.
3. The method of claim 1, wherein the sub-power sequence of each subsystem of the air conditioner comprises: the system comprises an air conditioning water chilling unit power sequence, an air conditioning chilled water pump system power sequence, an air conditioning cooling tower system power sequence, an air conditioning fan coil system power sequence and an air conditioning cooling water pump system power sequence.
4. The method of claim 1, wherein the objective function is 0 or 1, and wherein when the objective function is 0, it indicates that the predicted power sequence is not a real power sequence, and wherein when the objective function is 1, it indicates that the predicted power sequence is a real power sequence.
5. The method according to claim 1, wherein the summing the finally generated predicted sub-power sequences to obtain a first power sequence and the finally generated second power sequence corresponding to the predicted total power sequence are respectively compared with the total power sequence, and comprises:
Calculating a first correlation coefficient between the first power sequence and the total power sequence;
calculating a second correlation coefficient between the second power sequence and the total power sequence;
and determining the target power sequence corresponding to the maximum value according to the maximum value in the first correlation coefficient and the second correlation coefficient, and controlling the air conditioner according to the target power sequence.
6. The method of claim 5, wherein the first correlation coefficient and the second correlation coefficient range from 0,1, and wherein a closer correlation coefficient to 1 indicates a higher correlation and a closer correlation to 0 indicates a lower correlation.
7. An apparatus for controlling an air conditioner, comprising:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner;
the first input module is used for generating a random sequence according to noise, taking an environmental factor of the air conditioner as a tag vector, and inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence;
The second input module is used for inputting the predicted power sequence and the power sequence into a discrimination network in the generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not;
the adjusting module is used for adjusting a first parameter and a second parameter according to the predicted power sequence and the objective function until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter of the generating network, and the second parameter is a parameter of the judging network;
the comparison module is used for comparing a first power sequence obtained by summing the finally generated predicted sub-power sequences and a second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence respectively, selecting a target power sequence from the first power sequence and the second power sequence according to a comparison result, and controlling the air conditioner according to the target power sequence;
the label vector comprises the running time of the air conditioner, the temperature of the environment and the humidity of the environment; the size of the predicted power sequence output by the generating network is calculated by the number of input sequences, the convolution kernel edge length, the convolution step length and the filling number; the final generated predicted sub-power sequence and the final generated predicted total power sequence are sequences obtained by continuously adjusting parameters of a generating network and a judging network until the generating network is optimized.
8. An electronic device, comprising:
a memory for storing program instructions;
and the processor is connected with the memory and is used for realizing the following functions when executing the program instructions: acquiring a power sequence of an air conditioner, wherein the power sequence comprises a total power sequence of the air conditioner and a sub-power sequence of each subsystem of the air conditioner; generating a random sequence according to noise, taking an environmental factor of the air conditioner as a tag vector, and inputting the random sequence and the tag vector into a generation network in a generation countermeasure network to obtain a predicted power sequence, wherein the predicted power sequence comprises a predicted total power sequence and a predicted sub-power sequence, and the random sequence, the predicted power sequence and the power sequence are in one-to-one correspondence; inputting the predicted power sequence and the power sequence into a discrimination network in the generation countermeasure network to obtain an objective function, and inputting the objective function into the generation network, wherein the objective function is used for indicating whether the predicted power sequence is a real power sequence or not; according to the predicted power sequence and the objective function, a first parameter and a second parameter are adjusted until the first parameter and the second parameter are unchanged, wherein the first parameter is a parameter of the generated network, and the second parameter is a parameter of the judging network; comparing a first power sequence obtained by summing the finally generated predicted sub-power sequences and a second power sequence corresponding to the finally generated predicted total power sequence with the total power sequence respectively, selecting a target power sequence from the first power sequence and the second power sequence according to a comparison result, and controlling the air conditioner according to the target power sequence; the label vector comprises the running time of the air conditioner, the temperature of the environment and the humidity of the environment; the size of the predicted power sequence output by the generating network is calculated by the number of input sequences, the convolution kernel edge length, the convolution step length and the filling number; the final generated predicted sub-power sequence and the final generated predicted total power sequence are sequences obtained by continuously adjusting parameters of a generating network and a judging network until the generating network is optimized.
9. A non-volatile storage medium, characterized in that the non-volatile storage medium comprises a stored program, wherein the device in which the non-volatile storage medium is controlled to execute the method of controlling an air conditioner according to any one of claims 1 to 6 when the program is run.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111426134.6A CN114279061B (en) | 2021-11-26 | 2021-11-26 | Method and device for controlling air conditioner and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111426134.6A CN114279061B (en) | 2021-11-26 | 2021-11-26 | Method and device for controlling air conditioner and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114279061A CN114279061A (en) | 2022-04-05 |
| CN114279061B true CN114279061B (en) | 2023-07-14 |
Family
ID=80870104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111426134.6A Active CN114279061B (en) | 2021-11-26 | 2021-11-26 | Method and device for controlling air conditioner and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114279061B (en) |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102212663B1 (en) * | 2018-05-22 | 2021-02-05 | 주식회사 석영시스템즈 | An apparatus for hvac system input power control based on target temperature and method thereof |
| US20200394553A1 (en) * | 2019-06-13 | 2020-12-17 | International Business Machines Corporation | Predicting functional tags of semantic role labeling |
| CN110445126A (en) * | 2019-06-25 | 2019-11-12 | 中国电力科学研究院有限公司 | A kind of non-intrusion type load decomposition method and system |
| CN110781433B (en) * | 2019-10-11 | 2023-06-02 | 腾讯科技（深圳）有限公司 | Data type determining method and device, storage medium and electronic device |
| CN112257850B (en) * | 2020-10-26 | 2022-10-28 | 河南大学 | Vehicle track prediction method based on generation countermeasure network |
| CN112413862B (en) * | 2020-11-18 | 2022-03-18 | 珠海格力电器股份有限公司 | Method and device for determining energy consumption coefficient, storage medium and electronic device |
| CN113361803A (en) * | 2021-06-28 | 2021-09-07 | 嘉兴国电通新能源科技有限公司 | Ultra-short-term photovoltaic power prediction method based on generation countermeasure network |
-
2021
- 2021-11-26 CN CN202111426134.6A patent/CN114279061B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114279061A (en) | 2022-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109659933B (en) | Electric energy quality prediction method for power distribution network with distributed power supply based on deep learning model | |
| US20210018209A1 (en) | Building management system with triggered feedback set-point signal for persistent excitation | |
| CN109140660A (en) | For the Intelligent temperature controller method and device of air-conditioning, air-conditioning, storage medium | |
| CN111178537B (en) | Feature extraction model training method and device | |
| CN116707331A (en) | Inverter output voltage high-precision adjusting method and system based on model prediction | |
| CN115682312A (en) | Air conditioner energy-saving control method, device and equipment and readable storage medium | |
| CN112272074A (en) | Information transmission rate control method and system based on neural network | |
| CN112288139A (en) | Air conditioner energy consumption prediction method and system based on chaotic time sequence and storage medium | |
| CN114970357A (en) | Energy-saving effect evaluation method, system, device and storage medium | |
| CN114279061B (en) | Method and device for controlling air conditioner and electronic equipment | |
| CN113705929A (en) | Spring festival holiday load prediction method based on load characteristic curve and typical characteristic value fusion | |
| WO2019227273A1 (en) | Hierarchical concept based neural network model for data center power usage effectiveness prediction | |
| CN114636212B (en) | GRNN-based running control method for multiple water chilling unit systems | |
| CN112598133A (en) | Vehicle data processing method, device, equipment and storage medium | |
| CN115629576B (en) | Non-invasive flexible load aggregation characteristic identification and optimization method, device and equipment | |
| CN111242266A (en) | Operation data management system | |
| CN113825955A (en) | Method, apparatus, system, storage medium, and processor for determining a temperature setting | |
| CN114909707A (en) | Heat supply secondary network regulation and control method based on intelligent balancing device and reinforcement learning | |
| CN115943399A (en) | Generation method, generation device, program, information processing method, and information processing device | |
| CN113033760A (en) | Method, testing device and system for testing operator precision in neural network | |
| CN117953811B (en) | LED huge curtain point-by-point correction system, method, equipment and medium | |
| CN117952283A (en) | Short-term load prediction method, system, terminal and medium based on membership degree | |
| CN116454890B (en) | Combined control method, device and equipment for unit based on SCUC model | |
| CN115640896B (en) | Household user power load prediction method under multi-user scene and related equipment | |
| CN116734411A (en) | Air conditioner control method and device based on user behaviors, air conditioner and storage 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |