Disclosure of Invention
The embodiment of the invention provides a method and a device for remotely controlling a central air-conditioning system, which are used for solving the technical problems of untimely response, low efficiency and weak computing capacity of the existing central air-conditioning system control method.
The embodiment of the invention provides a method for remotely controlling a central air-conditioning system, which comprises the following steps:
the method comprises the steps that a cloud platform determines a control strategy based on operation data of all energy equipment of a central air conditioning system collected by an edge computing gateway, and sends the control strategy to the edge computing gateway;
the edge computing gateway receives the control strategy and inputs the control strategy into a trained network model to obtain a control instruction;
and the edge computing gateway controls the operation of controlled energy equipment in the central air-conditioning system according to the control instruction.
Optionally, the determining, by the cloud platform, a control policy based on operation data of all energy devices of the central air conditioning system collected by the edge computing gateway, and sending the control policy to the edge computing gateway includes:
the cloud platform is used for preprocessing all energy equipment operation data of the central air conditioning system collected by the edge computing gateway;
the cloud platform inputs all the preprocessed energy equipment operation data of the central air conditioning system collected by the edge computing gateway into a trained neural network to obtain a control strategy;
and the cloud platform sends the control strategy to the edge computing gateway.
Optionally, the pre-processing comprises: filtering, cleaning and/or rejecting dead spots.
Optionally, before the control strategy is input into the trained network model and a control instruction is obtained, the method for remotely controlling the central air conditioning system further includes:
the cloud platform trains a pre-stored network by using training sample data and sends the trained pre-stored network to the edge computing gateway;
and the edge computing gateway receives the trained pre-stored network sent by the cloud platform, and trains the trained pre-stored network again by using the collected running data of all energy equipment of the central air-conditioning system of the building to obtain a trained network model.
The embodiment of the invention provides a device for remotely controlling a central air-conditioning system, which comprises:
the device comprises a determining and sending module, a receiving and inputting module and a control module;
the determining and sending module is used for determining a control strategy based on all energy equipment operation data of the central air conditioning system acquired by the edge computing gateway by the cloud platform and sending the control strategy to the edge computing gateway;
the receiving and inputting module is used for receiving the control strategy by the edge computing gateway and inputting the control strategy into a trained network model to obtain a control instruction;
and the control module is used for controlling the operation of controlled energy equipment in the central air-conditioning system by the edge computing gateway according to the control instruction.
Optionally, the determining and sending module is specifically configured to:
the cloud platform is used for preprocessing all energy equipment operation data of the central air conditioning system collected by the edge computing gateway;
the cloud platform inputs all the preprocessed energy equipment operation data of the central air conditioning system collected by the edge computing gateway into a trained neural network to obtain a control strategy;
and the cloud platform sends the control strategy to the edge computing gateway.
Optionally, the pre-processing comprises: filtering, cleaning and/or rejecting dead spots.
Optionally, before the receiving and inputting module inputs the control strategy into the trained network model to obtain the control command, the apparatus for remotely controlling the central air conditioning system further includes: the training module is used for training the pre-stored network by the cloud platform by using training sample data and sending the trained pre-stored network to the edge computing gateway;
and the edge computing gateway receives the trained pre-stored network sent by the cloud platform, and trains the trained pre-stored network again by using the collected running data of all energy equipment of the central air-conditioning system of the building to obtain a trained network model.
According to the embodiment of the invention, through the architecture of 'cloud + edge', the computing capability of the edge computing gateway is expanded by using a cloud platform, the obstruction of network delay to the timeliness of tasks is avoided, and the complex computing capability requirement of equipment optimization and the real-time requirement of control are met; the transfer learning algorithm further utilizes the architectural characteristics of 'cloud + edge', so that the complex optimization of various devices in the central air-conditioning system becomes practical.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart illustrating a method for remotely controlling a central air conditioning system according to an embodiment of the present invention, including:
and S11, the cloud platform determines a control strategy based on the operation data of all energy equipment of the central air conditioning system collected by the edge computing gateway, and sends the control strategy to the edge computing gateway.
The cloud platform is a platform for issuing a control strategy to the edge computing gateway. The edge computing gateway is a terminal which collects all energy equipment operation data of the central air-conditioning system, analyzes and processes the operation data and sends the analyzed and processed operation data to the cloud platform. The collected operation data of all energy equipment of the central air-conditioning system comprises but is not limited to measured data of electric energy, gas, cold energy and/or heat energy, sensor data, equipment operation state data and/or alarm data. The control strategy is a strategy for controlling the central air conditioning system controlled energy device, including but not limited to the pressure of cooling water and/or the pressure of chilled water.
And S12, the edge computing gateway receives the control strategy and inputs the control strategy into the trained network model to obtain a control instruction.
And the edge computing gateway receives the control strategy sent by the cloud platform and takes the control strategy as the input of the trained network model. The output of the trained network model is the control instruction. The control command is a command for controlling controlled energy equipment in the central air-conditioning system.
And S13, the edge computing gateway controls the operation of the controlled energy equipment in the central air-conditioning system according to the control instruction.
Wherein the control command is a command for controlling a controlled energy device in the central air-conditioning system. The edge computing gateway can control the operation of the controlled energy equipment in the central air-conditioning system according to the control instruction.
The embodiment of the invention expands the computing capability of the edge computing gateway by using the cloud platform through the architecture of 'cloud + edge', avoids the obstruction of network delay to the timeliness of tasks, and simultaneously meets the complex computing capability requirement of equipment optimization and the real-time requirement of control; the transfer learning algorithm further utilizes the architectural characteristics of 'cloud + edge', so that the complex optimization of various devices in the central air-conditioning system becomes practical.
Further, on the basis of the above method embodiment, the determining, by the cloud platform, a control policy based on operation data of all energy devices of the central air conditioning system collected by the edge computing gateway, and sending the control policy to the edge computing gateway includes:
the cloud platform is used for preprocessing all energy equipment operation data of the central air conditioning system collected by the edge computing gateway;
the cloud platform inputs all the preprocessed energy equipment operation data of the central air conditioning system collected by the edge computing gateway into a trained neural network to obtain a control strategy;
and the cloud platform sends the control strategy to the edge computing gateway.
The operation data of all energy devices of the central air-conditioning system collected by the edge computing gateway contains useless data such as redundant data and/or dead data, so that the cloud platform is required to preprocess the operation data of all energy devices of the central air-conditioning system collected by the edge computing gateway so as to improve the computing capacity. And after the operation data of all energy equipment of the central air conditioning system collected by the edge computing gateway after the preprocessing is used as the input of a trained neural network, obtaining a control strategy after the operation data passes through the trained neural network. And the cloud platform issues the control strategy to the edge computing gateway.
According to the embodiment of the invention, the computing capability of the cloud platform is improved by preprocessing the useless data such as redundant data and/or dead point data in the running data of all the energy equipment of the central air-conditioning system, which is acquired by the edge computing gateway.
Further, on the basis of the above method embodiment, the preprocessing includes: filtering, cleaning and/or rejecting dead spots.
Wherein the filtering operation may remove redundant data. The scrubbing operation may correct recognizable errors in the data. The culling bad point operation may remove the bad point data.
According to the embodiment of the invention, the computing capacity of the cloud platform is improved by filtering, cleaning and/or extracting the dead pixel of the data.
Further, on the basis of the above method embodiment, before the control strategy is input into the trained network model to obtain the control command, the method for remotely controlling the central air conditioning system further includes:
the cloud platform trains a pre-stored network by using training sample data and sends the trained pre-stored network to the edge computing gateway;
and the edge computing gateway receives the trained pre-stored network sent by the cloud platform, and trains the trained pre-stored network again by using the collected running data of all energy equipment of the central air-conditioning system of the building to obtain a trained network model.
In the embodiment of the invention, the whole set of migration learning framework is deployed by depending on a cloud + edge architecture. The whole network is a multi-layer Back Propagation (BP) deep learning network framework. The process of obtaining the trained network model is as follows: establishing an original version learning framework org-NN-V0.0 on the cloud platform, obtaining labeled training sample data (P0) depending on a test point scene, performing preliminary training on org-NN-V0.0 on the cloud platform depending on pre-labeled training sample data (P0), determining an input layer, an output layer and a deviation feedback function of a multi-layer BP deep learning network, performing preliminary training on an upper layer part of a middle hidden layer, and forming the cloud platform pre-stored network org-NN-V1.0. When the edge computing gateway of the invention belonging to the building NB1 is accessed, the cloud platform issues the org-NN-V1.0 to the migration learning local network of the edge computing gateway of the invention through the uplink communication interface, and further trains the org-NN-V1.0 network by using the operation data of all energy equipment of the central air conditioning system of the building NB1 collected by the edge computing gateway to form the local edge network NN-NB1 suitable for the actual scene of the building NB 1. The edge computing gateway synchronizes all the energy equipment operation data of the central air conditioning system of the building NB1 subjected to data preprocessing to the cloud platform, and the cloud platform further trains org-NN-V1.0 by using the data to optimize and form the cloud platform pre-storage network org-NN-V2.0. When the edge computing gateway of the invention belonging to the building NB2 is accessed, the cloud platform issues the org-NN-V2.0 to the migration learning local network of the edge computing gateway of the invention through the uplink communication interface, and further trains the org-NN-V2.0 network by using the operation data of all energy equipment of the central air conditioning system of the building NB2 collected by the edge computing gateway to form the local edge network NN-NB2 suitable for the actual scene of the building NB 2. The edge computing gateway synchronizes all the energy equipment operation data of the central air conditioning system of the building NB12 subjected to data preprocessing to the cloud platform, and the cloud platform further trains org-NN-V2.0 by using the data to optimize and form the cloud platform pre-stored network org-NN-V3.0. By analogy, the pre-stored network org-NN of the cloud platform continuously iterates, and robustness is continuously improved. Each local edge network NN-NBn also conforms to the actual scene of the building to which it belongs. The local edge network NN-NBn is a well-trained network model.
The migration learning algorithm provided by the embodiment of the invention utilizes the architecture characteristics of 'cloud + edge' to realize the complex optimization of various devices in the central air-conditioning system.
Fig. 2 is a schematic structural diagram illustrating an apparatus for remotely controlling a central air conditioning system according to an embodiment of the present invention, including: a determination and transmission module 21, a reception and input module 22 and a control module 23;
the determining and sending module 21 is configured to determine a control strategy based on operation data of all energy devices of the central air conditioning system collected by the edge computing gateway by the cloud platform, and send the control strategy to the edge computing gateway;
the receiving and inputting module 22 is configured to receive the control policy by the edge computing gateway, and input the control policy into the trained network model to obtain a control instruction;
and the control module 23 is configured to control, by the edge computing gateway, operation of the controlled energy device in the central air conditioning system according to the control instruction.
Further, on the basis of the above device embodiment, the determining and sending module 21 is specifically configured to: the cloud platform is used for preprocessing all energy equipment operation data of the central air conditioning system collected by the edge computing gateway;
the cloud platform inputs all the preprocessed energy equipment operation data of the central air conditioning system collected by the edge computing gateway into a trained neural network to obtain a control strategy;
and the cloud platform sends the control strategy to the edge computing gateway.
Further, on the basis of the above apparatus embodiment, the preprocessing includes: filtering, cleaning and/or rejecting dead spots.
Further, on the basis of the above device embodiment, before the receiving and inputting module inputs the control strategy into the trained network model to obtain the control command, the device for remotely controlling the central air conditioning system further includes: the training module is used for training the pre-stored network by the cloud platform by using training sample data and sending the trained pre-stored network to the edge computing gateway;
and the edge computing gateway receives the trained pre-stored network sent by the cloud platform, and trains the trained pre-stored network again by using the collected running data of all energy equipment of the central air-conditioning system of the building to obtain a trained network model.
FIG. 3 is a logic block diagram of an electronic device according to an embodiment of the invention; the electronic device includes: a processor (processor)31, a memory (memory)32, and a bus 33;
wherein, the processor 31 and the memory 32 complete the communication with each other through the bus 33; the processor 31 is configured to call the program instructions in the memory 32 to execute the method for remotely controlling the central air conditioning system provided in the above method embodiment.
An embodiment of the present invention further provides a non-transitory computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements a method for performing remote control on a central air conditioning system provided in the foregoing embodiments.
The above-described embodiments of the apparatus are merely illustrative, and 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 network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.