CN113915743A - Air conditioning unit optimization control method and device based on load prediction - Google Patents

Air conditioning unit optimization control method and device based on load prediction Download PDF

Info

Publication number
CN113915743A
CN113915743A CN202111202473.6A CN202111202473A CN113915743A CN 113915743 A CN113915743 A CN 113915743A CN 202111202473 A CN202111202473 A CN 202111202473A CN 113915743 A CN113915743 A CN 113915743A
Authority
CN
China
Prior art keywords
component
feature
neuron
features
air conditioning
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.)
Withdrawn
Application number
CN202111202473.6A
Other languages
Chinese (zh)
Inventor
彭磊
靳璇如
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Foshan Pinzhi Information Technology Co ltd
Original Assignee
Foshan Pinzhi Information Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Foshan Pinzhi Information Technology Co ltd filed Critical Foshan Pinzhi Information Technology Co ltd
Priority to CN202111202473.6A priority Critical patent/CN113915743A/en
Publication of CN113915743A publication Critical patent/CN113915743A/en
Withdrawn legal-status Critical Current

Links

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
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/54—Control or safety arrangements characterised by user interfaces or communication using one central controller connected to several sub-controllers
    • 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
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/56—Remote control
    • F24F11/58—Remote control using Internet communication
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2134—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on separation criteria, e.g. independent component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mechanical Engineering (AREA)
  • Combustion & Propulsion (AREA)
  • Chemical & Material Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Mathematical Physics (AREA)
  • Signal Processing (AREA)
  • Human Computer Interaction (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The application provides an air conditioning unit optimization control method and device based on load prediction, which are used for improving the stability and robustness of prediction of load change of an air conditioning unit. The method comprises the following steps: acquiring operation data of the air conditioning unit, wherein the operation data comprises various types of data; processing the operation data through a first neural network module to obtain multiple components corresponding to the operation data, wherein the first neural network module comprises a feature extraction module and a feature reconstruction module, the feature extraction module is used for converting the multiple types of data into multiple types of features, and the feature reconstruction module is used for separating the multiple types of features to obtain the multiple component features corresponding to the operation data; determining a key component feature from the plurality of component features; and processing the key component characteristics through the second neural network module, and predicting the operation load of the air conditioning unit in the next time period.

Description

Air conditioning unit optimization control method and device based on load prediction
Technical Field
The application relates to the technical field of artificial intelligence, in particular to an air conditioning unit optimization control method and device based on load prediction.
Background
With the continuous development and progress of technology, artificial intelligence has been widely applied in various fields. For example, in the field of temperature control, the operation data of the air conditioning unit can be processed through a Neural network model, such as a Convolutional Neural Network (CNN), a Fast Convolutional Neural network (F-CNN), and the like, so as to predict the load variation trend of the air conditioning unit, and pre-regulate and control the operation of the air conditioning unit according to the prediction result.
However, because the types of the operation data of the air conditioning unit are various, the current neural network model is not stable enough when load change prediction is carried out, and the robustness is not good enough.
Disclosure of Invention
The embodiment of the application provides an air conditioning unit optimization control method and device based on load prediction, which are used for improving the stability and robustness of prediction of load change of an air conditioning unit.
In order to achieve the purpose, the technical scheme is as follows:
in a first aspect, an embodiment of the present application provides an air conditioning unit optimization control method based on load prediction, where the method includes: obtaining operation data of an air conditioning unit, wherein the operation data comprises various types of data; processing the operating data through a first neural network module to obtain multiple components corresponding to the operating data, wherein the first neural network module comprises a feature extraction module and a feature reconstruction module, the feature extraction module is used for converting the multi-class data into multi-class features, and the feature reconstruction module is used for separating the multi-class features to obtain the multiple component features corresponding to the operating data; determining a key component feature from the plurality of component features; and processing the key component characteristics through a second neural network module, and predicting the operation load of the air conditioning unit in the next time period.
Based on the method in the first aspect, the characteristic extraction module and the characteristic reconstruction module of the first neural network module can convert the multiple types of operation data of the air conditioning unit into multiple component characteristics, and extract key component characteristics from the multiple component characteristics, that is, the component characteristics which play the most important role in load change of the air conditioning unit or have the greatest influence. Therefore, the second neural network module can more accurately predict the operation load of the air conditioning unit in the next time period by using the key component characteristics so as to improve the stability and robustness of predicting the load change of the air conditioning unit.
In a possible design, the feature reconstruction module includes a plurality of self-encoders, the number of the self-encoders is inversely related to the size of the air conditioning unit, the network structures of the self-encoders are different, each self-encoder is configured to perform feature separation on the plurality of types of features to obtain one component feature of the operating data, and the self-encoders output the plurality of component features in common. It can be understood that, under the condition that the size of the air conditioning unit is large, the load fluctuation is usually small, and the randomness of key component characteristics influencing the load change is weak, so that a small number of self-encoders can be adopted to reduce the network size and improve the operation efficiency. On the contrary, under the condition that the scale of the air conditioning unit is small, the load fluctuation is large generally, and the randomness of key component characteristics influencing the load change is also strong, so that a large number of self-encoders can be adopted to increase the network scale and improve the accuracy of the calculation of the key component characteristics.
In one possible design, each of the self-encoders is configured to perform feature separation on the multiple classes of features, and obtaining a component feature of the operation data is: and each self-encoder is used for carrying out feature separation on the multiple types of features to obtain a low-dimensional feature, and carrying out dimension increasing processing on the low-dimensional feature to obtain a component feature of the operating data. Wherein, each of the self-encoders is used for performing feature separation on the multiple classes of features to obtain a low-dimensional feature: each self-encoder can decompose the multiple types of features by using a plurality of basis vectors to obtain multiple types of basis vectors, so that each self-encoder can calculate the content of each type of basis vector, wherein the content is high and is a main component, namely the low-dimensional features, and each self-encoder can accurately calculate a component feature corresponding to the operation data.
In one possible design, determining a key component feature from the plurality of component features includes: determining M component feature sets from the plurality of component features, wherein each component feature set in the M component feature sets comprises the same component feature, and M is a positive integer; and determining a key component feature set from the M component feature sets, wherein the key component feature set is the component feature set with the most component features in the M component feature sets, and the component features contained in the key component feature set are the key component features. That is, the key component feature is a feature that occurs the most frequently among various component features. This also reflects exactly how the key component characteristics can have the greatest or most critical effect on load changes. Therefore, the load prediction can be more accurately carried out by subsequently using the key component characteristics determined in the manner.
In one possible design, for an ith component feature set and a jth component feature set in the M component feature sets, i and j are different, i and j are integers from 1 to M, a euclidean distance between any two component features included in the ith component feature set is less than or equal to a distance threshold, a euclidean distance between any two component features included in the jth component feature set is less than or equal to the distance threshold, and a euclidean distance between any one component feature included in the ith component feature set and any one component feature included in the jth component feature set is greater than the distance threshold. It can be understood that the various component feature sets can be accurately determined by calculating the Euclidean distances of the features and considering similar features with small distances as the same feature.
In a possible design, the air conditioning unit includes N air conditioners, N is an integer greater than 1, each of the N air conditioners corresponds to one of the first sub-neurons and one of the second sub-neurons, and there are N of the first sub-neurons and N of the second sub-neurons in total, the first neuron is determined by weighting the N first sub-neurons, the second neuron is determined by weighting the N second sub-neurons, and the second neural network module is determined by combining the first neuron and the second neuron. And each air conditioner corresponds to the weight of one first sub-neuron and one second sub-neuron, and the weight depends on the importance of the air conditioner in the air conditioning unit. The importance may refer to functional importance, i.e., the more critical the air conditioner plays in cooling or heating the air conditioning unit, the higher the importance of the air conditioner. In this way, the first neuron and the second neuron are functionally coupled with the air conditioning unit through weighted calculation, so that the load prediction can be more accurately performed by using the second neural network module, and the stability and robustness of the load prediction are further improved.
In one possible design, the determining by the second neural network module from the first neuron and the second neuron combination means: combining the first neuron and the second neuron into a horizontally opposed network structure, the first neuron being located on a diagonal of the horizontally opposed network structure, the second neuron being located in a position other than the diagonal in the horizontally opposed network structure; alternatively, the second neuron is located on a diagonal of the horizontally opposed network structure and the first neuron is located in a position other than the diagonal in the horizontally opposed network structure. In this way, no matter what order the characteristics in the key characteristic components are input into the model, the load prediction result is not changed, and the practicability of the model is greatly improved.
In one possible design, the second neural network module includes W hidden layers, the horizontally opposed network structures s in the s hidden layer include the first neuron s and the second neuron s, the horizontally opposed network structures t in the t feature layer include the first neuron t and the second neuron t, the first neuron s is different from the first neuron t, the second neuron s is different from the second neuron t, W is an integer greater than 1, s and t are integers from 1 to M, and s and t are different. Therefore, the differentiation degree of the model can be improved to a certain degree, and the robustness of the model is effectively improved.
In a second aspect, an embodiment of the present application provides an air conditioning unit optimization control device based on load prediction, the device including: the receiving and sending module is used for obtaining the operation data of the air conditioning unit, and the operation data comprises various types of data; the processing module is used for processing the operating data through a first neural network module to obtain multiple components corresponding to the operating data, wherein the first neural network module comprises a feature extraction module and a feature reconstruction module, the feature extraction module is used for converting the multiple types of data into multiple types of features, and the feature reconstruction module is used for separating the multiple types of features to obtain the multiple component features corresponding to the operating data; determining a key component feature from the plurality of component features; and processing the key component characteristics through a second neural network module, and predicting the operation load of the air conditioning unit in the next time period.
In a possible design, the feature reconstruction module includes a plurality of self-encoders, the number of the self-encoders is inversely related to the size of the air conditioning unit, the network structures of the self-encoders are different, each self-encoder is configured to perform feature separation on the plurality of types of features to obtain one component feature of the operating data, and the self-encoders output the plurality of component features in common.
In one possible design, each of the self-encoders is configured to perform feature separation on the multiple classes of features, and obtaining a component feature of the operation data is: and each self-encoder is used for carrying out feature separation on the multiple types of features to obtain a low-dimensional feature, and carrying out dimension increasing processing on the low-dimensional feature to obtain a component feature of the operating data.
In one possible design, the processing module is further configured to determine M component feature sets from the plurality of component features, where each of the M component feature sets includes a same component feature, and M is a positive integer; and determining a key component feature set from the M component feature sets, wherein the key component feature set is the component feature set with the most component features in the M component feature sets, and the component features contained in the key component feature set are the key component features.
In one possible design, for an ith component feature set and a jth component feature set in the M component feature sets, i and j are different, i and j are integers from 1 to M, a euclidean distance between any two component features included in the ith component feature set is less than or equal to a distance threshold, a euclidean distance between any two component features included in the jth component feature set is less than or equal to the distance threshold, and a euclidean distance between any one component feature included in the ith component feature set and any one component feature included in the jth component feature set is greater than the distance threshold.
In a possible design, the air conditioning unit includes N air conditioners, N is an integer greater than 1, each of the N air conditioners corresponds to one of the first sub-neurons and one of the second sub-neurons, and there are N of the first sub-neurons and N of the second sub-neurons in total, the first neuron is determined by weighting the N first sub-neurons, the second neuron is determined by weighting the N second sub-neurons, and the second neural network module is determined by combining the first neuron and the second neuron.
In one possible design, the determining by the second neural network module from the first neuron and the second neuron combination means: combining the first neuron and the second neuron into a horizontally opposed network structure, the first neuron being located on a diagonal of the horizontally opposed network structure, the second neuron being located in a position other than the diagonal in the horizontally opposed network structure; alternatively, the second neuron is located on a diagonal of the horizontally opposed network structure and the first neuron is located in a position other than the diagonal in the horizontally opposed network structure.
In one possible design, the second neural network module includes W hidden layers, the horizontally opposed network structures s in the s hidden layer include the first neuron s and the second neuron s, the horizontally opposed network structures t in the t feature layer include the first neuron t and the second neuron t, the first neuron s is different from the first neuron t, the second neuron s is different from the second neuron t, W is an integer greater than 1, s and t are integers from 1 to M, and s and t are different.
Optionally, the transceiver module may include a receiving module and a transmitting module. Wherein the receiving module is configured to implement a receiving function of the apparatus according to the second aspect. The sending module is configured to implement a sending function of the apparatus according to the second aspect.
Optionally, the apparatus of the second aspect may further comprise a storage module storing the program or the instructions. The program or instructions, when executed by the processing module, cause the apparatus to perform the method of the first aspect.
It should be noted that the apparatus according to the second aspect may be a network device, a chip (system) or other component or assembly that can be disposed in the network device, or an apparatus including the network device, and the present application is not limited thereto.
In addition, for technical effects of the apparatus according to the second aspect, reference may be made to technical effects of the method according to the first aspect, and details are not repeated here.
In a third aspect, an air conditioning unit optimization control apparatus based on load prediction is provided. The air conditioning unit optimization control device based on load prediction comprises: a processor and a memory; the memory is adapted to store a computer program which, when executed by the processor, causes the apparatus to perform the method of the first aspect.
In one possible design, the apparatus of the third aspect may further include a transceiver. The transceiver may be a transmit-receive circuit or an interface circuit. The transceiver may be for the apparatus of the third aspect to communicate with other apparatuses.
In this application, the apparatus according to the third aspect may be a network device, or a chip (system) or other component or assembly that can be disposed in the network device, or an apparatus that includes the network device.
In addition, for technical effects of the apparatus according to the third aspect, reference may be made to technical effects of the method according to the first aspect, and details are not repeated here.
In a fourth aspect, the present application provides a computer-readable storage medium, on which program code is stored, and when the program code is executed by the computer, the method according to the first aspect is executed.
Drawings
Fig. 1 is a schematic structural diagram of an air conditioning unit optimization control system based on load prediction according to an embodiment of the present application;
fig. 2 is a flowchart of an air conditioning unit optimization control method based on load prediction according to an embodiment of the present application;
fig. 3 is a first schematic structural diagram of an air conditioning unit optimization control device based on load prediction according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a second air conditioning unit optimization control device based on load prediction according to an embodiment of the present application.
Detailed Description
The technical solution in the present application will be described below with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present application provides a load prediction-based air conditioning unit optimization control system, which may include one or more devices, such as a first device.
The first device may be, for example, a terminal device, such as a user terminal, which may also be referred to as a user equipment, an access terminal, a subscriber unit, a subscriber station, a mobile station, a remote terminal, a mobile device, a user terminal, a wireless communication device, a user agent, or a user equipment. The terminal device in the embodiment of the present application may be a mobile phone (mobile phone), a tablet computer (Pad), a computer with a wireless transceiving function, a Virtual Reality (VR) terminal device, an Augmented Reality (AR) terminal device, a wireless terminal in industrial control (industrial control), a wireless terminal in self driving (self driving), a wireless terminal in remote medical (remote medical), a wireless terminal in smart grid (smart grid), a wireless terminal in transportation safety (transportation safety), a wireless terminal in smart city (smart city), a wireless terminal in smart home (smart home), a vehicle-mounted terminal, an RSU with a terminal function, and the like.
Alternatively, the first device may also be a network device, such as a device with wireless transceiving functionality or a chip (system) or other component or assembly of a device, including but not limited to: an Access Point (AP) in a wireless fidelity (WiFi) system, such as a home gateway, a router, a server, a switch, a bridge, etc., an evolved Node B (eNB), a Radio Network Controller (RNC), a Node B (NB), a Base Station Controller (BSC), a Base Transceiver Station (BTS), a home base station (e.g., home evolved Node B, or home Node B, HNB), a Base Band Unit (BBU), a wireless relay Node, a wireless backhaul Node, a transmission point (transmission and reception point, TRP or transmission point, etc.), and may be 5G, such as a new radio interface (NR) system, a TP, a Transmission Point (TP), a group of antennas including one or more antenna panels (antenna panels) in the system, alternatively, the network node may also be a network node forming a gNB or a transmission point, such as a baseband unit (BBU), or a Distributed Unit (DU), a roadside unit (RSU) having a base station function, or a wired access gateway.
The details will be described below in conjunction with the method.
Referring to fig. 2, an embodiment of the present application provides an air conditioning unit optimization control method based on load prediction. The method may be applied to a first device in the system shown in fig. 1. The method comprises the following steps:
s201, obtaining operation data of the air conditioning unit.
Wherein the operational data includes multiple types of data.
S202, processing the operation data through the first neural network module to obtain multiple components corresponding to the operation data.
The first neural network module comprises a feature extraction module and a feature reconstruction module, the feature extraction module is used for converting the multi-class data into the multi-class features, and the feature reconstruction module is used for separating the multi-class features to obtain the multi-component features corresponding to the operating data.
Specifically, the characteristic reconstruction module comprises a plurality of self-encoders, the number of the self-encoders is inversely related to the scale of the air conditioning unit, the network structures of the self-encoders are different, each self-encoder is used for carrying out characteristic separation on multiple types of characteristics to obtain one component characteristic of the operation data, and the self-encoders output multiple component characteristics. It can be understood that, under the condition that the size of the air conditioning unit is large, the load fluctuation is usually small, and the randomness of key component characteristics influencing the load change is weak, so that a small number of self-encoders can be adopted to reduce the network size and improve the operation efficiency. On the contrary, under the condition that the scale of the air conditioning unit is small, the load fluctuation is large generally, and the randomness of key component characteristics influencing the load change is also strong, so that a large number of self-encoders can be adopted to increase the network scale and improve the accuracy of the calculation of the key component characteristics.
Further, each self-encoder is configured to perform feature separation on the multiple types of features, and obtaining one component feature of the operation data is: each self-encoder is used for carrying out feature separation on the multiple types of features to obtain a low-dimensional feature, and carrying out dimension increasing processing on the low-dimensional feature to obtain a component feature of the operation data. Wherein, each self-encoder is used for carrying out feature separation on multiple types of features to obtain a low-dimensional feature, and the low-dimensional feature refers to: each self-encoder can decompose the multiple types of features by using some base vectors to obtain multiple base vectors, so that each self-encoder can calculate the content of each base vector, wherein the content is high in principal component, namely the low-dimensional features, and each self-encoder can accurately calculate a component feature corresponding to the operation data.
S203, a key component feature is determined from the multiple component features.
Wherein determining a key component feature from the plurality of component features comprises: determining M component feature sets from the multiple component features, wherein each component feature set in the M component feature sets comprises the same component feature, and M is a positive integer; and determining a key component feature set from the M component feature sets, wherein the key component feature set is the component feature set with the most component features in the M component feature sets, and the component features contained in the key component feature set are key component features. That is, the key component feature is a feature that occurs the most frequently among various component features. This also reflects exactly how the key component characteristics can have the greatest or most critical effect on load changes. Therefore, the load prediction can be more accurately carried out by subsequently using the key component characteristics determined in the manner.
Illustratively, for an ith component feature set and a jth component feature set in the M component feature sets, i and j are different, i and j are integers from 1 to M, a euclidean distance between any two component features included in the ith component feature set is less than or equal to a distance threshold, a euclidean distance between any two component features included in the jth component feature set is less than or equal to a distance threshold, and a euclidean distance between any one component feature included in the ith component feature set and any one component feature included in the jth component feature set is greater than the distance threshold. It can be understood that the various component feature sets can be accurately determined by calculating the Euclidean distances of the features and considering similar features with small distances as the same feature.
And S204, processing the key component characteristics through the second neural network module, and predicting the operation load of the air conditioning unit in the next time period.
The air conditioning unit comprises N air conditioners, N is an integer larger than 1, each air conditioner in the N air conditioners corresponds to a first sub-neuron and a second sub-neuron, the N first sub-neurons and the N second sub-neurons are total, the first neuron is determined according to the N first sub-neurons in a weighted mode, the second neuron is determined according to the N second sub-neurons in a weighted mode, and the second neural network module is determined according to the combination of the first neuron and the second neuron. The weight of each air conditioner corresponding to one first sub-neuron and one second sub-neuron depends on the importance of the air conditioner in the air conditioning unit. The importance may refer to functional importance, i.e., the more critical the air conditioner plays in cooling or heating the air conditioning unit, the higher the importance of the air conditioner. In this way, the first neuron and the second neuron are functionally coupled with the air conditioning unit through weighted calculation, so that the load prediction can be more accurately performed by using the second neural network module, and the stability and robustness of the load prediction are further improved.
In one possible design, the determining, by the second neural network module, from the first neuron and the second neuron combination means: combining a first neuron and a second neuron into a horizontally opposed network structure, the first neuron being located on a diagonal of the horizontally opposed network structure, the second neuron being located at a position other than the diagonal in the horizontally opposed network structure; alternatively, the second neuron is located on a diagonal of the horizontally opposed network structure and the first neuron is located in a position other than the diagonal in the horizontally opposed network structure. In this way, no matter what order the characteristics in the key characteristic components are input into the model, the load prediction result is not changed, and the practicability of the model is greatly improved.
In one possible design, the second neural network module includes W hidden layers, the horizontally-opposed network structures s in the s-th hidden layer include a first neuron s and a second neuron s, the horizontally-opposed network structures t in the t-th feature layer include a first neuron t and a second neuron t, the first neuron s is different from the first neuron t, the second neuron s is different from the second neuron t, W is an integer greater than 1, s and t are integers from 1 to M, and s and t are different. Therefore, the differentiation degree of the model can be improved to a certain degree, and the robustness of the model is effectively improved.
In summary, due to the feature extraction module and the feature reconstruction module of the first neural network module, the multi-class operation data of the air conditioning unit can be converted into the multi-component features, and the key component features are extracted from the multi-component features, namely, the most key effect is played on the load change of the air conditioning unit, or the most influenced component features are obtained. Therefore, the second neural network module can more accurately predict the operation load of the air conditioning unit in the next time period by using the key component characteristics so as to improve the stability and robustness of predicting the load change of the air conditioning unit.
Referring to fig. 3, the present embodiment further provides an air conditioning unit optimization control apparatus 300 based on load prediction, for performing the above method.
Specifically, the apparatus 300 includes: the transceiving module 301 is configured to obtain operation data of the air conditioning unit, where the operation data includes multiple types of data; a processing module 302, configured to process the operating data through a first neural network module to obtain multiple components corresponding to the operating data, where the first neural network module includes a feature extraction module and a feature reconstruction module, the feature extraction module is configured to convert the multiple types of data into multiple types of features, and the feature reconstruction module is configured to separate the multiple types of features to obtain multiple component features corresponding to the operating data; determining a key component feature from the plurality of component features; and processing the key component characteristics through a second neural network module, and predicting the operation load of the air conditioning unit in the next time period.
In a possible design, the feature reconstruction module includes a plurality of self-encoders, the number of the self-encoders is inversely related to the size of the air conditioning unit, the network structures of the self-encoders are different, each self-encoder is configured to perform feature separation on the plurality of types of features to obtain one component feature of the operating data, and the self-encoders output the plurality of component features in common.
In one possible design, each of the self-encoders is configured to perform feature separation on the multiple classes of features, and obtaining a component feature of the operation data is: and each self-encoder is used for carrying out feature separation on the multiple types of features to obtain a low-dimensional feature, and carrying out dimension increasing processing on the low-dimensional feature to obtain a component feature of the operating data.
In one possible design, the processing module 302 is further configured to determine M component feature sets from the plurality of component features, where each component feature set in the M component feature sets includes a same component feature, and M is a positive integer; and determining a key component feature set from the M component feature sets, wherein the key component feature set is the component feature set with the most component features in the M component feature sets, and the component features contained in the key component feature set are the key component features.
In one possible design, for an ith component feature set and a jth component feature set in the M component feature sets, i and j are different, i and j are integers from 1 to M, a euclidean distance between any two component features included in the ith component feature set is less than or equal to a distance threshold, a euclidean distance between any two component features included in the jth component feature set is less than or equal to the distance threshold, and a euclidean distance between any one component feature included in the ith component feature set and any one component feature included in the jth component feature set is greater than the distance threshold.
In a possible design, the air conditioning unit includes N air conditioners, N is an integer greater than 1, each of the N air conditioners corresponds to one of the first sub-neurons and one of the second sub-neurons, and there are N of the first sub-neurons and N of the second sub-neurons in total, the first neuron is determined by weighting the N first sub-neurons, the second neuron is determined by weighting the N second sub-neurons, and the second neural network module is determined by combining the first neuron and the second neuron.
In one possible design, the determining by the second neural network module from the first neuron and the second neuron combination means: combining the first neuron and the second neuron into a horizontally opposed network structure, the first neuron being located on a diagonal of the horizontally opposed network structure, the second neuron being located in a position other than the diagonal in the horizontally opposed network structure; alternatively, the second neuron is located on a diagonal of the horizontally opposed network structure and the first neuron is located in a position other than the diagonal in the horizontally opposed network structure.
In one possible design, the second neural network module includes W hidden layers, the horizontally opposed network structures s in the s hidden layer include the first neuron s and the second neuron s, the horizontally opposed network structures t in the t feature layer include the first neuron t and the second neuron t, the first neuron s is different from the first neuron t, the second neuron s is different from the second neuron t, W is an integer greater than 1, s and t are integers from 1 to M, and s and t are different.
Alternatively, the transceiving module 301 may include a receiving module and a transmitting module. Wherein the receiving module is configured to implement a receiving function of the apparatus according to the second aspect. The sending module is configured to implement a sending function of the apparatus according to the second aspect.
Optionally, the load prediction-based air conditioning unit optimization control device may further include a storage module, where the storage module stores a program or instructions. The program or instructions, when executed by the processing module 302, cause the load prediction based air conditioning unit optimization control to perform the method described above with respect to fig. 2.
The air conditioning unit optimization control device based on load prediction may be a network device, a chip (system) or other components or assemblies that can be installed in the network device, or a device including the network device, which is not limited in the present application.
In addition, the technical effects of the air conditioning unit optimization control device based on load prediction can refer to the technical effects of the above methods, and are not described herein again.
The following describes each component of the air conditioning unit optimization control device 400 based on load prediction in detail with reference to fig. 4:
the processor 401 is a control center of the air conditioning unit optimization control apparatus 400 based on load prediction, and may be a single processor or a collective term for a plurality of processing elements. For example, the processor 401 is one or more Central Processing Units (CPUs), or may be an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application, such as: one or more microprocessors (digital signal processors, DSPs), or one or more Field Programmable Gate Arrays (FPGAs).
Alternatively, the processor 401 may perform various functions of the air conditioning unit optimization control apparatus 400 based on load prediction by running or executing a software program stored in the memory 402 and calling up data stored in the memory 402.
In particular implementations, processor 401 may include one or more CPUs such as CPU0 and CPU1 shown in fig. 4 as an example.
In one embodiment, the load prediction based air conditioning unit optimization control apparatus 400 may also include a plurality of processors, such as the processor 401 and the processor 404 shown in fig. 4. Each of these processors may be a single-Core Processor (CPU) or a multi-Core Processor (CPU). A processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (e.g., computer program instructions).
The memory 402 is configured to store a software program for executing the scheme of the present application, and is controlled by the processor 401 to execute the software program.
Alternatively, memory 402 may be a read-only memory (ROM) or other type of static storage device that may store static information and instructions, a Random Access Memory (RAM) or other type of dynamic storage device that may store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to such. The memory 402 may be integrated with the processor 401, or may be independent, and is coupled to the processor 401 through an interface circuit (not shown in fig. 4) of the air conditioning unit optimization control device 400 based on load prediction, which is not specifically limited in this embodiment of the present application.
A transceiver 403 for communication with other devices. For example, the air conditioning unit optimization control apparatus 400 based on load prediction is a network device, and the transceiver 403 may be used to communicate with a terminal device or communicate with another network device.
Optionally, the transceiver 403 may include a receiver and a transmitter (not separately shown in fig. 4). Wherein the receiver is configured to implement a receive function and the transmitter is configured to implement a transmit function.
Alternatively, the transceiver 403 may be integrated with the processor 401, or may be independent and coupled to the processor 401 through an interface circuit (not shown in fig. 4) of the air conditioning unit optimization control device 400 based on load prediction, which is not specifically limited in this embodiment of the present application.
It should be noted that the configuration of the apparatus 400 shown in fig. 4 does not constitute a limitation of the load prediction-based air conditioning unit optimization control apparatus, and the actual load prediction-based air conditioning unit optimization control apparatus may include more or less components than those shown, or some components may be combined, or a different arrangement of components may be used.
In addition, the technical effects of the method of the above method embodiment can be referred to for the technical effects of the apparatus 400, and are not described herein again.
It should be understood that the processor in the embodiments of the present application may be a Central Processing Unit (CPU), and the processor may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It will also be appreciated that the memory in the embodiments of the subject application can be either volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. The non-volatile memory may be a read-only memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an electrically Erasable EPROM (EEPROM), or a flash memory. Volatile memory can be Random Access Memory (RAM), which acts as external cache memory. By way of example, but not limitation, many forms of Random Access Memory (RAM) are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct bus RAM (DR RAM).
The above embodiments may be implemented in whole or in part by software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. The procedures or functions according to the embodiments of the present application are generated in whole or in part when a computer instruction or a computer program is loaded or executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored on a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more collections of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
It should be understood that the term "and/or" herein is merely one type of association relationship that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. In addition, the "/" in this document generally indicates that the former and latter associated objects are in an "or" relationship, but may also indicate an "and/or" relationship, which may be understood with particular reference to the former and latter text.
In the present application, "at least one" means one or more, "a plurality" means two or more. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.
It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some feature fields may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
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 network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. An air conditioning unit optimization control method based on load prediction is characterized by comprising the following steps:
obtaining operation data of an air conditioning unit, wherein the operation data comprises various types of data;
processing the operating data through a first neural network module to obtain multiple components corresponding to the operating data, wherein the first neural network module comprises a feature extraction module and a feature reconstruction module, the feature extraction module is used for converting the multi-class data into multi-class features, and the feature reconstruction module is used for separating the multi-class features to obtain the multiple component features corresponding to the operating data;
determining a key component feature from the plurality of component features;
and processing the key component characteristics through a second neural network module, and predicting the operation load of the air conditioning unit in the next time period.
2. The method according to claim 1, wherein the characteristic reconstruction module comprises a plurality of self-encoders, the number of the self-encoders is inversely related to the size of the air conditioning unit, the network structures of the self-encoders are different, each self-encoder is used for performing characteristic separation on the plurality of types of characteristics to obtain one component characteristic of the operation data, and the self-encoders output the plurality of types of component characteristics together.
3. The method of claim 2, wherein each of the plurality of self-encoders is configured to perform feature separation on the plurality of classes of features, and obtaining a component feature of the operation data is: and each self-encoder is used for carrying out feature separation on the multiple types of features to obtain a low-dimensional feature, and carrying out dimension increasing processing on the low-dimensional feature to obtain a component feature of the operating data.
4. A method according to any one of claims 1 to 3, wherein determining a key component feature from the plurality of component features comprises:
determining M component feature sets from the plurality of component features, wherein each component feature set in the M component feature sets comprises the same component feature, and M is a positive integer;
and determining a key component feature set from the M component feature sets, wherein the key component feature set is the component feature set with the most component features in the M component feature sets, and the component features contained in the key component feature set are the key component features.
5. The method according to claim 4, wherein i and j are different for an ith component feature set and a jth component feature set in the M component feature sets, i and j are integers from 1 to M, a Euclidean distance between any two component features included in the ith component feature set is smaller than or equal to a distance threshold, a Euclidean distance between any two component features included in the jth component feature set is smaller than or equal to the distance threshold, and a Euclidean distance between any one component feature included in the ith component feature set and any one component feature included in the jth component feature set is greater than the distance threshold.
6. The method of claim 1, wherein the set of air conditioners includes N air conditioners, N is an integer greater than 1, each of the N air conditioners corresponds to one of the first sub-neurons and one of the second sub-neurons, N are the first sub-neurons and N are the second sub-neurons, the first neuron is determined by weighting the N first sub-neurons, the second neuron is determined by weighting the N second sub-neurons, and the second neural network module is determined by combining the first neuron and the second neuron.
7. The method of claim 6, wherein the second neural network module determines from the first neuron and the second neuron combination that: combining the first neuron and the second neuron into a horizontally opposed network structure, the first neuron being located on a diagonal of the horizontally opposed network structure, the second neuron being located in a position other than the diagonal in the horizontally opposed network structure; alternatively, the second neuron is located on a diagonal of the horizontally opposed network structure and the first neuron is located in a position other than the diagonal in the horizontally opposed network structure.
8. The method of claim 7, wherein the second neural network module comprises W hidden layers, wherein the horizontally opposed network structures s in an s-th hidden layer comprise the first neurons s and the second neurons s, wherein the horizontally opposed network structures t in a t-th feature layer comprise the first neurons t and the second neurons t, wherein the first neurons s are different from the first neurons t, wherein the second neurons s are different from the second neurons t, wherein W is an integer greater than 1, wherein s and t are integers from 1 to M, and wherein s and t are different.
9. An air conditioning unit optimization control device based on load prediction, characterized in that the device comprises:
the receiving and sending module is used for obtaining the operation data of the air conditioning unit, and the operation data comprises various types of data;
the processing module is used for processing the operating data through a first neural network module to obtain multiple components corresponding to the operating data, wherein the first neural network module comprises a feature extraction module and a feature reconstruction module, the feature extraction module is used for converting the multiple types of data into multiple types of features, and the feature reconstruction module is used for separating the multiple types of features to obtain the multiple component features corresponding to the operating data; determining a key component feature from the plurality of component features; and processing the key component characteristics through a second neural network module, and predicting the operation load of the air conditioning unit in the next time period.
10. A computer-readable storage medium, characterized in that the storage medium has stored thereon a program code which, when executed by the computer, performs the method according to any one of claims 1-8.
CN202111202473.6A 2021-10-15 2021-10-15 Air conditioning unit optimization control method and device based on load prediction Withdrawn CN113915743A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111202473.6A CN113915743A (en) 2021-10-15 2021-10-15 Air conditioning unit optimization control method and device based on load prediction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111202473.6A CN113915743A (en) 2021-10-15 2021-10-15 Air conditioning unit optimization control method and device based on load prediction

Publications (1)

Publication Number Publication Date
CN113915743A true CN113915743A (en) 2022-01-11

Family

ID=79240673

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111202473.6A Withdrawn CN113915743A (en) 2021-10-15 2021-10-15 Air conditioning unit optimization control method and device based on load prediction

Country Status (1)

Country Link
CN (1) CN113915743A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117910518A (en) * 2024-03-19 2024-04-19 青岛创新奇智科技集团股份有限公司 Method and system for analyzing generated data

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117910518A (en) * 2024-03-19 2024-04-19 青岛创新奇智科技集团股份有限公司 Method and system for analyzing generated data
CN117910518B (en) * 2024-03-19 2024-06-11 青岛创新奇智科技集团股份有限公司 Method and system for analyzing generated data

Similar Documents

Publication Publication Date Title
CN106464526B (en) System and method for detecting anomalies
US20220416921A1 (en) Frequency spectrum management device, system, method and computer readable storage medium
CN112084959B (en) Crowd image processing method and device
CN111050387B (en) Base station dormancy method and device based on energy efficiency estimation, electronic equipment and medium
CN109743738A (en) Frequency spectrum sharing method, device and the electronic equipment of spectra sharing system
CN113915743A (en) Air conditioning unit optimization control method and device based on load prediction
CN110784837B (en) Indoor positioning method, device, medium and electronic equipment
Le et al. Fingerprinting indoor positioning method based on kernel ridge regression with feature reduction
US20230336940A1 (en) Electronic device for configuring geofence and operation method thereof
CN112214677A (en) Interest point recommendation method and device, electronic equipment and storage medium
CN113630895A (en) Data packet transmission method and device
CN113672669B (en) Data query method and device based on distributed storage
CN113744033A (en) Order estimation method and device
CN113783833A (en) Method and device for constructing computer security knowledge graph
CN113891323A (en) WiFi-based user tag acquisition system
EP3905746A1 (en) Frequency spectrum management apparatus, frequency spectrum management method and computer-readable storage medium
CN113301541B (en) SL communication video knowledge graph construction method and device
US11086976B2 (en) Identification preprocessing
CN114867104B (en) Positioning method and device based on multiple beams
KR102360264B1 (en) Method for providing communication service and apparatus for performing the method
CN115941563A (en) Task monitoring method and device integrating information of multiple command platforms
KR102163214B1 (en) Machine Learning based proactive caching in CoMP-enabled small cell networks
Lyu et al. Deep Learning‐Based Bluetooth Low‐Energy 5.1 Multianchor Indoor Positioning with Attentional Data Filtering
CN112507358B (en) Scene-demand-based lightweight encryption algorithm selection method
US11665596B2 (en) Planning of fixed wireless internet

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20220111

WW01 Invention patent application withdrawn after publication