CN109946967A - A kind of building energy internet information interchanger of combination of edge intelligence - Google Patents
A kind of building energy internet information interchanger of combination of edge intelligence Download PDFInfo
- Publication number
- CN109946967A CN109946967A CN201910226987.1A CN201910226987A CN109946967A CN 109946967 A CN109946967 A CN 109946967A CN 201910226987 A CN201910226987 A CN 201910226987A CN 109946967 A CN109946967 A CN 109946967A
- Authority
- CN
- China
- Prior art keywords
- data
- building energy
- energy
- control
- cloud platform
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of building energy interconnection network switch of combination of edge intelligence, comprising: downlink communication mould group acquires multi-energy data;Uplink communication mould group receives control strategy;Microprocessing unit, bidirectional data transfers are carried out with cloud platform and building energy system respectively, the first control strategy is handled when receiving the first control strategy to obtain control instruction, obtains control instruction according to multi-energy data and pre-set processing locality strategy when receiving the second control strategy and based on transfer learning algorithm process;Mould group is controlled, control instruction is received and building energy system is sent to by isolation mode group;Power supply, beneficial effects of the present invention: the computing capability of edge termination is extended using cloud platform, network delay is in turn avoided to the obstruction of task timeliness, the requirement of real-time for meeting the complicated calculations ability need and control of equipment optimization simultaneously, so that becoming a reality to the complex optimization of various energy resources equipment in building energy internet.
Description
Technical field
The present invention relates to the interconnections of the building energy of building energy administrative skill field more particularly to a kind of combination of edge intelligence
Net message switching equipment.
Background technique
With the development of energy technology, energy technology has breached the limitation of the single energy, using energy internet as generation
The comprehensive energy operation and management of table becomes the trend of future development.Due to energy-storage battery, distributed generation resource, Chu Leng, heat accumulation, cold
The rise of the technologies such as thermoelectricity trilogy supply (Combined Cooling, Heating and Power, CCHP), heat pump, building interior
Energy network from it is unidirectional with can network gradually become two-way electric, the hot and cold energy internet mutually flowed.
In energy internet, energy source link is logical from the original single radial transmission from source to load that can be planted
Road, become multipotency kind mixing multiple sources, lotus, storage main body between polynary network.
Therefore it is required to the operation demand according to energy internet, provides energy information between different sources, storage, lotus
Connection and control, the distribution of the Optimized Operation energy.
The good operation of energy internet needs accurately to dispatch and manage, it is traditional it is artificial, it is mechanical and it is simple from
The energy O&M and dispatching technique of dynamicization can no longer meet demand.Simultaneously because gas supply, heat supply, the energy feature of cooling supply and electricity
Power has biggish difference, and traditional electric power, heating power, air conditioning managing equipment cannot well adapt to comprehensive energy management and running
Requirement, be badly in need of that uniformly monitoring, scheduling, control can be carried out to electricity, hot and cold source, storage, lotus simultaneously based on digital technology
New energy Internet of things control device.
With the fast development of information technology and energy technology, with 5G, big data, cloud platform, artificial intelligence, Internet of Things,
Distributed energy, energy internet, be " internet+wisdom energy " of representative from energy body is building energy row at energy storage technology
Industry has blown a burst of spring breeze, carries out reconstruction of energizing to traditional building energy system with completely new theory and framework, bring safety,
Stablize, is efficient, green, energy conservation, balance, autonomous, convenient, intelligent completely new energy-consuming experience.
Application No. is 201810883188.7 patents to disclose a kind of energy interchanger auxiliary device intelligent monitor system
And monitoring method, comprising: smart machine monitoring modular is communicated to connect for establishing with smart machine, connect from the smart machine
Receiving unit electric information, and transmit this information to data processing module;Data processing module, for the electricity based on the received
Gas information inquires each access port of energy interchanger, whether is physically connected to energy exchange with the determination smart machine
Machine, and determine connected access port, data memory module is sent by query result;Data memory module connects for basis
The query result received is updated or stores to data.Using the energy interchanger acquisition access device electric data and
The smart machine facility information that mode is sent by wireless communication and heartbeat message carry out online, the off-line monitoring of equipment.Meanwhile
For the conventional equipment of not wireless communication module, can be carried out by the data acquisition information of energy interchanger online, offline
Judgement.
The prior art is mainly directed towards the data acquisition and monitoring of energy device, is related to not to energy conversion Optimized Operation aspect
It is more.The realization for biasing toward communication function, for being related to seldom for intelligent control algorithm, framework is fairly simple, local computing capability
It is weaker, it is difficult to the big data analysis and intelligent algorithm of extended deployment complexity, for complicated calculations ability need and real-time control
Equilibrium is difficult between system.
Summary of the invention
Aiming at the problems existing in the prior art, the present invention relates to a kind of building energy internet of combination of edge intelligence letters
Cease interchanger.The present invention adopts the following technical scheme:
A kind of building energy internet information interchanger of combination of edge intelligence is suitable for building energy system, described to build
It builds energy internet information interchanger and a cloud platform communicates to connect, the building energy internet information interchanger is for acquiring
The multi-energy data of the building energy system simultaneously carries out being uploaded to the cloud platform after data processing, and receives the cloud platform
The control strategy issued, the control strategy include the first control strategy and the second control strategy, the building energy internet
Message switching equipment handles to obtain control instruction according to the control strategy, and controls the building energy according to the control instruction
System;The building energy internet information interchanger includes:
Downlink communication mould group, for acquiring the multi-energy data of the building energy system;
Uplink communication mould group, the control strategy issued for receiving the cloud platform;
Microprocessing unit connects the downlink communication mould group and the uplink communication mould group, for flat with the cloud respectively
Platform and the building energy system carry out bidirectional data transfers, and to described first when for receiving first control strategy
Control strategy is handled to obtain the control instruction and be exported, and when for receiving second control strategy according to
Multi-energy data and pre-set processing locality strategy are simultaneously obtained the control instruction and are exported based on transfer learning algorithm process;
Mould group is controlled, the microprocessing unit is connected, refers to for receiving the control that the miniature processing unit issues
It enables and exports;
Isolation mode group connects the control mould group, for receiving the control instruction and being sent to the building energy system
It unites to control the energy source dispensing of the building energy system;
Power supply, connect the uplink communication mould group, the memory, the microprocessing unit, the downlink communication mould group,
The control mould group and the isolation mode group, for being the building energy internet information power supply for exchange.
Preferably, the building energy system includes multiple energy devices and the multiple energy device of connection to constitute pair
Answer the energy flow path of the energy network of the building energy system;
The multi-energy data includes the operation data of the corresponding energy device and the energy of the corresponding energy flow path
Source flow data.
Preferably, energy flowing data packet include in the energy network MAC Address of each energy device node and
The port of transmission path;
The control instruction includes the optimal path to energy conversion and distribution, will be each described by the control instruction
The MAC Address of energy device and port are matched so that the energy flow path meets the optimal path.
Preferably, first control strategy is preset non-to judge that the building energy system meets in the cloud platform
In real time, long period data and when operational decision making scene, the building energy internet information interchanger is according to the cloud platform
The control strategy issued handles to obtain the control instruction, and the energy of the building energy system is controlled according to the control instruction
Source allotment;
Second control strategy is to judge that the building energy system meets preset real-time, short in the cloud platform
When cycle data and local decision-making scene, the building energy internet information interchanger is according to the multi-energy data and in advance
The processing locality strategy that is arranged simultaneously obtains the control instruction based on the transfer learning algorithm process, and according to the control
System instruction controls the energy source dispensing of the building energy system.
Preferably, the building energy internet information interchanger further include:
Encryption and decryption mould group, connects the microprocessing unit and the power supply, for send the miniature processing unit
Communication data is uploaded to uplink communication mould group after carrying out data encryption operation.
Preferably, the building energy internet information interchanger further include:
Memory connects the microprocessing unit and the power supply, for all operations to the miniature processing unit
Operation data is stored;
Clock connects the microprocessing unit and the power supply, for providing time reference for the miniature processing unit.
Preferably, the microprocessing unit includes:
Upstream data communication interface, for establishing communication connection for the microprocessing unit and the cloud platform;
Edge calculations platform connects the upstream data communication interface, is previously provided with the processing locality strategy, described
Processing locality strategy is based on the transfer learning algorithm;
Downlink data communication interface connects the edge calculations platform, for being the microprocessing unit and the building
Energy resource system establishes communication connection.
Preferably, the edge calculations platform includes:
Edge calculations gateway, for the processing locality strategy to be stored in advance;
Data collaborative and task cooperation unit, connection are described between the edge calculations platform and the cloud platform
Data are decomposed with task, are distributed, are shared and is synchronous;
Intelligence learning frame unit, for carrying out data prediction, Heterogeneous Computing, light weight according to the processing locality strategy
Data analysis and the processing of transfer learning local network;
ETL process unit, for carrying out data extraction, data conversion and data loading processing;
Internet of Things protocol stack, for formatting operation to the multi-energy data to obtain Uniform data format and upload
To the cloud platform.
Preferably, the transfer learning algorithm uses BP deep neural network algorithm;Or
The transfer learning algorithm uses vector machine algorithm;Or
The transfer learning algorithm uses decision Tree algorithms;Or
The transfer learning algorithm uses factorial analysis machine algorithm.
Preferably, the cloud platform is located in one or more data centers, passes through high-speed communication between the cloud platform
Network connection group is combined into the control centre for carrying out information processing and control, and the cloud platform is realized using edge calculations technology
Data acquisition, analysis and control operation;
The building energy internet information interchanger is located in building, and is joined by wide-area communication network and cloud platform
Logical, the building energy internet information interchanger realizes data acquisition, analysis and control operation using edge calculations technology.
Beneficial effects of the present invention: the entirety management to energy resource system, including monitoring, control and optimization are realized.Pass through
The framework of " cloud+edge ", the computing capability of edge termination is extended using cloud computing platform, in turn avoids network delay to task
The obstruction of timeliness, while meeting the requirement of real-time of the complicated calculations ability need and control of equipment optimization;
Transfer learning algorithm further utilizes the feature of the framework of " cloud+edge ", so as to more in building energy internet
The complex optimization of kind energy device becomes a reality.
Detailed description of the invention
Fig. 1 is a kind of building energy internet information friendship of combination of edge intelligence in a kind of preferred embodiment of the present invention
The functional block diagram changed planes;
Fig. 2 is the functional block diagram of microprocessing unit in a kind of preferred embodiment of the present invention;
Fig. 3 is the functional block diagram of edge calculations platform in a kind of preferred embodiment of the present invention.
Specific embodiment
It should be noted that in the absence of conflict, following technical proposals be can be combined with each other between technical characteristic.
A specific embodiment of the invention is further described with reference to the accompanying drawing:
As shown in Figure 1-3, a kind of building energy internet information interchanger of combination of edge intelligence, is suitable for building energy
System, above-mentioned building energy internet information interchanger and a cloud platform communicate to connect, and above-mentioned building energy internet information is handed over
It changes planes and the multi-energy data for acquiring above-mentioned building energy system and is uploaded to above-mentioned cloud platform, Yi Jijie after carrying out data processing
The control strategy that above-mentioned cloud platform issues is received, above-mentioned control strategy includes the first control strategy and the second control strategy, above-mentioned to build
It builds energy internet information interchanger to be handled to obtain control instruction according to above-mentioned control strategy, and is controlled according to above-mentioned control instruction
Above-mentioned building energy system;Above-mentioned building energy internet information interchanger includes:
Downlink communication mould group 5, for acquiring the above-mentioned multi-energy data of above-mentioned building energy system;
Uplink communication mould group 1, the above-mentioned control strategy issued for receiving above-mentioned cloud platform;
Microprocessing unit 2 connects above-mentioned downlink communication mould group 5 and above-mentioned uplink communication mould group 1, for respectively with above-mentioned cloud
Platform and above-mentioned building energy system carry out bidirectional data transfers, and to above-mentioned the when for receiving above-mentioned first control strategy
One control strategy is handled to obtain above-mentioned control instruction and be exported, and when for receiving above-mentioned second control strategy according to
It states multi-energy data and pre-set processing locality strategy and above-mentioned control instruction and defeated is obtained based on transfer learning algorithm process
Out;
Mould group 3 is controlled, connects above-mentioned microprocessing unit 2, the above-mentioned control issued for receiving above-mentioned miniature processing unit
It instructs and exports;
Isolation mode group 4 connects above-mentioned control mould group 3, supreme states building energy for receiving above-mentioned control instruction and sending
System is to control the energy source dispensing of above-mentioned building energy system;
Power supply connects above-mentioned uplink communication mould group 1, above-mentioned memory 7, above-mentioned microprocessing unit 2, above-mentioned downlink communication
Mould group 5, above-mentioned control mould group 3 and above-mentioned isolation mode group 4, for being above-mentioned building energy internet information power supply for exchange.
In the present embodiment, the present invention is based on cloud+edges " framework, downwards acquisition build in all energy devices fortune
Row data (including electricity/gas/cold heat measured data, other sensors data, equipment running status data, alarm data etc.), warp
It crosses after ownship information processing and acquisition data is uploaded into energy management cloud platform, receive the scheduling and control strategy that cloud platform passes down;
Scheduling and control instruction are issued to energy device to lower after the intelligent algorithm of the machine, it is adjusted and is in not
Same operating condition, to reach the different target such as safety, energy conservation, economic, efficient.
Realize the entirety management to energy resource system, including monitoring, control and optimization.By the framework of " cloud+edge ",
The computing capability that edge termination is extended using cloud platform in turn avoids network delay to the obstruction of task timeliness, simultaneously completely
Foot the complicated calculations ability need of equipment optimization and the requirement of real-time of control.Transfer learning algorithm further utilizes " cloud+side
The feature of the framework of edge ", so that becoming a reality to the complex optimization of various energy resources equipment in building energy internet.
In preferred embodiment, above-mentioned building energy system includes multiple energy devices and the above-mentioned multiple energy devices of connection
To constitute the energy flow path of the energy network of corresponding above-mentioned building energy system;
Above-mentioned multi-energy data includes the operation data of corresponding above-mentioned energy device and the energy of corresponding above-mentioned energy flow path
Source flow data.
In preferred embodiment, above-mentioned energy flowing data packet includes each energy device node in above-mentioned energy network
The port of MAC Address and transmission path;
Above-mentioned control instruction includes the optimal path to energy conversion and distribution, will be each above-mentioned by above-mentioned control instruction
The MAC Address of energy device and port are matched so that above-mentioned energy flow path meets above-mentioned optimal path.
In preferred embodiment, above-mentioned first control strategy is to judge that above-mentioned building energy system meets in above-mentioned cloud platform
When preset non real-time, long period data and operational decision making scene, above-mentioned building energy internet information interchanger is according to upper
It states the control strategy that cloud platform issues to handle to obtain above-mentioned control instruction, and above-mentioned building energy is controlled according to above-mentioned control instruction
The energy source dispensing of system;
Above-mentioned second control strategy is to judge that above-mentioned building energy system meets preset real-time, short in above-mentioned cloud platform
When cycle data and local decision-making scene, above-mentioned building energy internet information interchanger is according to above-mentioned multi-energy data and in advance
The above-mentioned processing locality strategy that is arranged simultaneously obtains above-mentioned control instruction based on above-mentioned transfer learning algorithm process, and according to above-mentioned control
System instruction controls the energy source dispensing of above-mentioned building energy system.
In the present embodiment, the division of labor of cloud platform and building energy internet information interchanger are as follows: be responsible for by cloud computing non-
In real time, long period data, operational decision making scene, and edge calculations gateway 221 stresses to determine in real-time, short cycle data, local
The scenes such as plan.Data, analysis processing are converged by edge calculations gateway 221 under normal circumstances, and data are extracted into result and upload cloud
Platform is completed further analysis and is shown;Cloud platform issues control strategy, and it is complete to be decomposed into specific instruction by edge calculations gateway 221
It is executed at control;It is directly controlled in special circumstances by cloud platform.
Seven layers of mode of communication are analogous to, building energy internet information interchanger works in the data link of the energy
Layer (MAC).The present invention is connect by data-interface with source, storage, lotus equipment and transmission line, the pipe network for building interior each energy,
The energy flow-data in the operation data and energy pipe network, route of each energy resource system is acquired, is made in information data level each
Interconnection between energy device obtains the MAC Address of each energy device node and the port of transmission path, and by artificial
The optimal path of energy conversion and distribution is calculated in intelligent algorithm, will be each
The MAC Address of a energy device and port are matched, and are that energy conversion scheduling bridge joint forms number in data plane
According to channel, Optimized Operation is completed, plays the role of energy exchange scheduling.
In preferred embodiment, above-mentioned building energy internet information interchanger further include:
Encryption and decryption mould group connects above-mentioned microprocessing unit 2 and above-mentioned power supply, for send above-mentioned miniature processing unit
Communication data is uploaded to uplink communication mould group 1 after carrying out data encryption operation.
In preferred embodiment, above-mentioned building energy internet information interchanger further include:
Memory 7 connects above-mentioned microprocessing unit 2 and above-mentioned power supply, for all fortune to above-mentioned miniature processing unit
Row operation data is stored;
Clock 6 connects above-mentioned microprocessing unit 2 and above-mentioned power supply, for providing time base for above-mentioned miniature processing unit
It is quasi-.
In the present embodiment, it includes power supply mould group, miniature processing that the hardware of building energy internet information interchanger, which is constituted,
Unit, clock 6, memory 7, control mould group 3 and relevant isolation mode group 4, uplink communication mould group 1 and relevant encryption and decryption
Mould group, downlink communication mould group 5.Wherein communications module is the power supply of other above-mentioned mould groups;Clock 6 provides the time for miniature processing unit
Benchmark;Memory 7 provides data storage function for miniature processing unit;Control module receives the control that miniature processing unit issues
Instruction, and sent outside by isolation mode group 4 to controlled energy device;Communication data with cloud platform is sent to by miniature processing unit
Encryption module completes data encryption, and passes through uplink communication mould group 1 and cloud platform two-way communication;Miniature processing unit passes through downlink
Communications module 5, which is realized, acquires the data of energy device and other sensors.
In preferred embodiment, above-mentioned microprocessing unit 2 includes:
Upstream data communication interface 21, for establishing communication connection for above-mentioned microprocessing unit 2 and above-mentioned cloud platform;
Edge calculations platform 22 connects above-mentioned upstream data communication interface 21, is previously provided with above-mentioned processing locality strategy,
Above-mentioned processing locality strategy is based on above-mentioned transfer learning algorithm;
Downlink data communication interface 23 connects above-mentioned edge calculations platform 22, for for above-mentioned microprocessing unit 2 and above-mentioned
Building energy system establishes communication connection.
In the present embodiment, the software of building energy internet information interchanger mainly operates on miniature processing unit,
Using embedded architecture.Operating system, upstream data communication interface 21, downlink data communication interface 23 and side including bottom
Edge computing platform 22;Wherein edge calculations platform 22 is based on edge platform both service mode (EPaaS), including edge calculations gateway
221, data collaborative, task cooperation, ETL process, Internet of Things agreement station and intelligence learning frame;Wherein intelligence learning frame is based on
Edge software both service mode (ESaaS), including the analysis of data prediction, Heterogeneous Computing, lightweight data, transfer learning local network
Network.
Software according to the present invention in practice can be there are many form, and such as simple embedded system is based on edge
Form, form of edge IaaS+SaaS framework of IaaS+PaaS+SaaS framework etc..
In preferred embodiment, above-mentioned edge calculations platform 22 includes:
Edge calculations gateway 221, for above-mentioned processing locality strategy to be stored in advance;
Data collaborative and task cooperation unit 222 connect above-mentioned for flat in above-mentioned edge calculations platform 22 and above-mentioned cloud
Data are decomposed with task between platform, are distributed, are shared and is synchronous;
Intelligence learning frame unit 223, for according to above-mentioned processing locality strategy carry out data prediction, Heterogeneous Computing,
Lightweight data analysis and the processing of transfer learning local network;
ETL process unit 224, for carrying out data extraction, data conversion and data loading processing;
Internet of Things protocol stack 225, for formatting operation to above-mentioned multi-energy data to obtain Uniform data format simultaneously
It is uploaded to above-mentioned cloud platform.
In the present embodiment, wherein ETL process is data extraction, data conversion, data load.Wherein data prediction packet
It includes the filtering to initial data, cleaning, polymerization, reject bad point, decomposition, interception.Wherein data collaborative and task cooperation are edges
To the decomposition of data and task, distribution, shared and synchronous between terminal and cloud platform.Wherein lightweight data analysis includes that equipment is drawn
Picture, failure tree analysis (FTA), with can the simple local data analysis algorithm such as abnormal variation association analysis, exception information polymerization, specifically
It is configured, can be issued by uplink communication interface by cloud platform according to application demand.Wherein Heterogeneous Computing is miniature according to terminal
The hardware configuration of processing unit is packaged the hardware interface of different processors, allows task at the place of different structure
Synthetic operation in Li Qi group, including control, optimization and analytical calculation.
In addition to data collaborative and task cooperation, it may also include resource coordinating, management collaboration, business cooperation, peace in practice
Full collaboration etc..
Wherein transfer learning local network is the edge deployment part of transfer learning.A whole set of transfer learning frame relies on " cloud
The deployment of+edge " framework.Whole network is Multi-Layer BP Neural deep learning network frame.Most first edition learning framework is established in cloud platform
Org-NN-V0.0 relies on pilot scene to obtain the training sample data (P0) for completing mark, marks in advance in cloud platform support
Training sample data (P0) carry out initial training to org-NNV0.0, determine input layer, the output of Multi-Layer BP Neural deep learning network
Layer and deviation feedback function form cloud platform and prestore network org-NN- to the top section initial training of intermediate hidden layers
V1.0.When there is the above-mentioned terminal access of the invention for belonging to building NB1, cloud platform is by uplink communication interface, by org-NN-
V1.0 is issued to the transfer learning local network of the above-mentioned terminal of the present invention, utilizes terminal affiliated building NB1 energy system collected
System operation data further trains org-NN-V1.0 network, forms this rand of building NB1 actual scene where being suitble to
Hoddy network NN-NB1.The above-mentioned terminal of the present invention is simultaneously by the affiliated building NB1 energy system operation after data prediction
Data are also synchronized to cloud platform, and platform further trains org-NN-V1.0 using the data, and optimization forms cloud platform
Prestore network org-NN-V2.0.When there is the above-mentioned terminal access of the invention for belonging to building NB2, cloud platform passes through uplink communication
Org-NN-V2.0 is issued to the transfer learning local network of the above-mentioned terminal of the present invention, utilizes terminal institute collected by interface
Belong to building NB2 energy system operation data further to train org-NN-V2.0 network, is formed and build NB2 where being suitble to
The local edge network NN-NB2 of actual scene.The above-mentioned terminal of the present invention is simultaneously by the affiliated building after data prediction
NB12 energy system operation data are also synchronized to cloud platform, and platform further instructs org-NN-V2.0 using the data
Practice, optimization forms cloud platform and prestores network org-NN-V3.0.And so on, the network org-NN that prestores of cloud platform constantly changes
Generation, robustness are constantly promoted.Each local edge network NN-NBn also complies with the respectively affiliated actual scene built.
Wherein transfer learning algorithm is mainly completed to electricity, hot and cold energy energy equipment, energy storage device, with can equipment
Operating status optimizes scheduling.By energy-storage battery control, Lighting control, photovoltaic control, micro-capacitance sensor scheduling is realized;Pass through pot
Furnace control, heat accumulation control, pipe network control, airconditioning control, heat pump control, existing heat supply network scheduling;By pipe network control, airconditioning control,
Heat pump control, vacant, cold-storage control of divulging information, realize cold net scheduling, pass through the association of micro-capacitance sensor scheduling, heat supply network scheduling, cold net scheduling
Together, the energy give-and-take conditions that electric, hot, cold, four, gas can plant are controlled, realize the scheduling of energy internet global optimization.
Wherein specific Optimized Operation formula are as follows:
foptimal(t)=min (fcost(t)), wherein t is time variable, foptimalIt (t) is optimal result, fcost(t) it is
Exogenous factor power.
Wherein: ec(t) outsourcing electrical power, hcIt (t) is outsourcing thermal power, gc(t)
For outsourcing gas velocity.
ec(t)=eload(t)+ees(t)+eloss(t)-edg(t), in which: eloadIt (t) is load electric power, ees(t)
For energy storage charge-discharge electric power, elossIt (t) is line transmission loss power, edgIt (t) is distributed generation resource generated output.
eload(t)=ecold(t)+elight(t)+eheat(t)+eair(t)+eelse(t), wherein ecoldIt (t) is refrigeration electricity consumption
Power, elightIt (t) is electric consumption on lighting power, eheatIt (t) is heating electric power, eairIt (t) is air draft electric power, eelse
It (t) is other electric powers.
ees(t)=echarge(t)+eoperate(t)-edischarge(t), wherein echargeIt (t) is charge power, edischarge
It (t) is discharge power, eoperateIt (t) is operation electric power.
edg(t)=epv(t)+ewg(t)+ecchp(t), wherein epvIt (t) is photovoltaic generation power, ewgIt (t) is wind turbine power generation function
Rate, ecchpIt (t) is cold, heat and electricity triple supply (CCHP) generated output.
ecchp(t)=TCCHp_e(gCCHP(t)), wherein TCCHP_eFor the fuel gas generation energy conversion efficiency function of CCHP, gCCHP
It (t) is the gas flow rate function of CCHP operation consumption.
Wherein hloadIt (EV) is load hot merit
Rate,
For hloss(t) Heat transmission dissipated power, hboilerIt (t) is boiler heats power, hCCHPIt (t) is CCHP heats power.
hcchp(t)=TCCHP_h(gCCHP(t)), wherein TCCHP_hFor combustion gas heating capacity transfer efficiency function, gCCHP(t) it is
The gas flow rate function of CCHP operation consumption.
Wherein EV is by the temperature T of all measuring points of Discussion on architecturen(t), humidity Tn(t) matrix constituted.
hboiler(t)=Tboiler(gboiler(t)), wherein TboilerFor the combustion gas heating capacity transfer efficiency function of boiler,
gboilerIt (t) is the gas flow rate function of boiler operatiopn consumption.
hac(t)=Tac(ecold(t)), wherein TacFor the cooling power transfer efficiency function of air-conditioning.
Wherein gcookIt (t) is kitchen gas flow velocity.
The input layer parameter of neural network are as follows: gboiler(t)、gcook(t)、gCCHP(t)、EV、 hboiler(t)、hCCHP(t)、
hac(t)、epv(t)、ewg(t)、ecchp(t)、ees(t)、 ecold(t)、elight(t)、eheat(t)、eair(t)、eelse(t)。
Output parameter is ec(t)、hc(t)、gc(t)、eloss(t)、hloss(t).With
As deviation feedback condition.
In preferred embodiment, above-mentioned transfer learning algorithm uses BP deep neural network algorithm;Or
Above-mentioned transfer learning algorithm uses vector machine algorithm;Or
Above-mentioned transfer learning algorithm uses decision Tree algorithms;Or
Above-mentioned transfer learning algorithm uses factorial analysis machine algorithm.
In this embodiment, the algorithm in addition to BP deep neural network, such as supporting vector also can be used in transfer learning algorithm
Machine, decision tree, factorial analysis machine scheduling algorithm.
In transfer learning algorithm, it can be selected to run simultaneously data for training source from edge termination to cloud platform according to scene
Network is asynchronous.
In preferred embodiment, above-mentioned cloud platform is located in one or more data centers, passes through between above-mentioned cloud platform
High-speed communicating network connection group is combined into the control centre for carrying out information processing and control, and above-mentioned cloud platform utilizes edge calculations
Technology realizes data acquisition, analysis and control operation;
Above-mentioned building energy internet information interchanger is located in building, and is joined by wide-area communication network and cloud platform
Logical, above-mentioned building energy internet information interchanger realizes data acquisition, analysis and control operation using edge calculations technology.
In the present embodiment, above-mentioned " cloud+edge " framework (is built comprising the sub- equipment of cloud computing platform (cloud platform) and edge
Build energy internet information interchanger), wherein cloud computing platform is located in one or more data centers, passes through high-speed communication network
Network connection group is combined into an entirety, and as the center of information processing and control, the sub- equipment in edge is located in building, logical by wide area
Communication network and cloud computing platform connection;The sub- equipment utilization edge calculations technology of cloud computing platform and edge is realized data acquisition, is divided
Analysis and control task.
By description and accompanying drawings, the exemplary embodiments of the specific structure of specific embodiment are given, based on present invention essence
Mind can also make other conversions.Although foregoing invention proposes existing preferred embodiment, however, these contents are not intended as
Limitation.
For a person skilled in the art, after reading above description, various changes and modifications undoubtedly be will be evident.
Therefore, appended claims should regard the whole variations and modifications for covering true intention and range of the invention as.It is weighing
The range and content of any and all equivalences, are all considered as still belonging to the intent and scope of the invention within the scope of sharp claim.
Claims (10)
1. a kind of building energy internet information interchanger of combination of edge intelligence, is suitable for building energy system, feature exists
In the building energy internet information interchanger and a cloud platform communicate to connect, the building energy internet information exchange
Machine is used to acquire the multi-energy data of the building energy system and is uploaded to the cloud platform after carrying out data processing, and receives
The control strategy that the cloud platform issues, the control strategy include the first control strategy and the second control strategy, the building
Energy internet information interchanger handles to obtain control instruction according to the control strategy, and controls institute according to the control instruction
State building energy system;The building energy internet information interchanger includes:
Downlink communication mould group, for acquiring the multi-energy data of the building energy system;
Uplink communication mould group, the control strategy issued for receiving the cloud platform;
Microprocessing unit connects the downlink communication mould group and the uplink communication mould group, for respectively with the cloud platform and
The building energy system carries out bidirectional data transfers, and to first control when for receiving first control strategy
Strategy is handled to obtain the control instruction and be exported, and according to the energy when for receiving second control strategy
Data and pre-set processing locality strategy are simultaneously obtained the control instruction and are exported based on transfer learning algorithm process;
Mould group is controlled, the microprocessing unit is connected, for receiving the control instruction that the miniature processing unit issues simultaneously
Output;
Isolation mode group connects the control mould group, for receive the control instruction and be sent to the building energy system with
Control the energy source dispensing of the building energy system;
Power supply connects the uplink communication mould group, memory, the microprocessing unit, the downlink communication mould group, described
Mould group and the isolation mode group are controlled, for being the building energy internet information power supply for exchange.
2. building energy internet information interchanger according to claim 1, which is characterized in that the building energy system
Including multiple energy devices and the multiple energy device is connected to constitute the energy network of the corresponding building energy system
Energy flow path;
The multi-energy data includes the operation data of the corresponding energy device and the energy flow of the corresponding energy flow path
Dynamic data.
3. building energy internet information interchanger according to claim 2, which is characterized in that the energy flow-data
The port of MAC Address and transmission path including energy device node each in the energy network;
The control instruction includes the optimal path of energy conversion and distribution, is set each energy by the control instruction
Standby MAC Address and port is matched so that the energy flow path meets the optimal path.
4. building energy internet information interchanger according to claim 1, which is characterized in that first control strategy
To judge that the building energy system meets preset non real-time, long period data and operational decision making scene in the cloud platform
When, the building energy internet information interchanger handles to obtain the control according to the control strategy that the cloud platform issues and refers to
It enables, and controls the energy source dispensing of the building energy system according to the control instruction;
Second control strategy is to judge that the building energy system meets preset real-time, short cycle in the cloud platform
When data and local decision-making scene, the building energy internet information interchanger is according to the multi-energy data and presets
The processing locality strategy and the control instruction is obtained based on the transfer learning algorithm process, and referred to according to the control
Enable the energy source dispensing for controlling the building energy system.
5. building energy internet information interchanger according to claim 1, which is characterized in that the building energy interconnection
Net message switching equipment further include:
Encryption and decryption mould group, connects the microprocessing unit and the power supply, the communication for sending the miniature processing unit
Data are uploaded to uplink communication mould group after carrying out data encryption operation.
6. building energy internet information interchanger according to claim 1, which is characterized in that the building energy interconnection
Net message switching equipment further include:
Memory connects the microprocessing unit and the power supply, for all operations operation to the miniature processing unit
Data are stored;
Clock connects the microprocessing unit and the power supply, for providing time reference for the miniature processing unit.
7. building energy internet information interchanger according to claim 1, which is characterized in that the microprocessing unit packet
It includes:
Upstream data communication interface, for establishing communication connection for the microprocessing unit and the cloud platform;
Edge calculations platform connects the upstream data communication interface, is previously provided with the processing locality strategy, the local
Processing strategie is based on the transfer learning algorithm;
Downlink data communication interface connects the edge calculations platform, for being the microprocessing unit and the building energy
System establishes communication connection.
8. building energy internet information interchanger according to claim 1, which is characterized in that the edge calculations platform
Include:
Edge calculations gateway, for the processing locality strategy to be stored in advance;
Data collaborative and task cooperation unit, connection are described for the logarithm between the edge calculations platform and the cloud platform
According to being decomposed, distributed, shared and synchronous with task;
Intelligence learning frame unit, for carrying out data prediction, Heterogeneous Computing, lightweight data according to the processing locality strategy
Analysis and the processing of transfer learning local network;
ETL process unit, for carrying out data extraction, data conversion and data loading processing;
Internet of Things protocol stack, for formatting operation to the multi-energy data to obtain Uniform data format and be uploaded to institute
State cloud platform.
9. building energy internet information interchanger according to claim 1, which is characterized in that the transfer learning algorithm
Using BP deep neural network algorithm;Or
The transfer learning algorithm uses vector machine algorithm;Or
The transfer learning algorithm uses decision Tree algorithms;Or
The transfer learning algorithm uses factorial analysis machine algorithm.
10. building energy internet information interchanger according to claim 1, which is characterized in that the cloud platform is located at
In one or more data centers, it is combined by high-speed communicating network connection group for carrying out information processing between the cloud platform
With the control centre of control, the cloud platform realizes data acquisition, analysis and control operation using edge calculations technology;
The building energy internet information interchanger is located in building, and passes through wide-area communication network and cloud platform connection, institute
It states building energy internet information interchanger and realizes data acquisition, analysis and control operation using edge calculations technology.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910226987.1A CN109946967A (en) | 2019-03-25 | 2019-03-25 | A kind of building energy internet information interchanger of combination of edge intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910226987.1A CN109946967A (en) | 2019-03-25 | 2019-03-25 | A kind of building energy internet information interchanger of combination of edge intelligence |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109946967A true CN109946967A (en) | 2019-06-28 |
Family
ID=67011501
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910226987.1A Pending CN109946967A (en) | 2019-03-25 | 2019-03-25 | A kind of building energy internet information interchanger of combination of edge intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109946967A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110440396A (en) * | 2019-07-11 | 2019-11-12 | 雄安达实智慧科技有限公司 | The central air-conditioning global optimization energy-saving control method and system of cloud side end collaboration |
CN110705694A (en) * | 2019-09-30 | 2020-01-17 | 国网山东省电力公司电力科学研究院 | Electric larceny monitoring method facing edge data center based on feature extraction |
CN110852387A (en) * | 2019-11-13 | 2020-02-28 | 江苏能来能源互联网研究院有限公司 | Energy internet super real-time state studying and judging algorithm |
CN110887173A (en) * | 2019-11-27 | 2020-03-17 | 国网冀北电力有限公司 | Method and device for remotely controlling central air-conditioning system |
CN110989988A (en) * | 2019-12-10 | 2020-04-10 | 天津瑞能电气有限公司 | Micro-grid edge layer software platform based on edge calculation |
CN111556093A (en) * | 2020-03-27 | 2020-08-18 | 天津市普迅电力信息技术有限公司 | Multifunctional edge Internet of things agent device for power grid information acquisition |
CN111708635A (en) * | 2020-06-16 | 2020-09-25 | 深圳天海宸光科技有限公司 | Video intelligent grading processing system and method |
CN112162849A (en) * | 2020-08-28 | 2021-01-01 | 中冶华天工程技术有限公司 | Cloud and edge computing cooperative equipment intelligent control system and method |
CN113037837A (en) * | 2021-03-05 | 2021-06-25 | 东北大学 | Intelligent terminal facing comprehensive energy system and based on edge calculation and control method thereof |
CN113820979A (en) * | 2021-09-14 | 2021-12-21 | 蓝鲸智云智能科技南京有限公司 | Energy-consumption system energy-saving strategy cloud platform based on IOT technology |
CN114492769A (en) * | 2022-04-11 | 2022-05-13 | 飞腾信息技术有限公司 | Data processing method, related equipment and system |
CN117092918A (en) * | 2023-10-20 | 2023-11-21 | 北京明略政睿信息技术有限公司 | Cloud edge cooperation-based intelligent gas sensing control method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102010050726A1 (en) * | 2010-11-08 | 2012-05-10 | Alphaeos Gmbh & Co. Kg | Building automation system |
WO2014072054A1 (en) * | 2012-11-07 | 2014-05-15 | Guttenberg & Hördegen | System for the analysis and control of energy flows |
CN108494605A (en) * | 2018-04-16 | 2018-09-04 | 华东师范大学 | A kind of energy net connection monitoring method and storage medium based on block chain |
CN108808663A (en) * | 2018-06-12 | 2018-11-13 | 浙江大学 | It is a kind of based on the industrial user's heat demand response method provided multiple forms of energy to complement each other |
CN108958049A (en) * | 2018-07-10 | 2018-12-07 | 三智头智能科技(佛山)有限公司 | A kind of energy monitor management method and system |
CN109379420A (en) * | 2018-10-10 | 2019-02-22 | 上海方融科技有限责任公司 | A kind of comprehensive energy service platform system based on distributed structure/architecture |
-
2019
- 2019-03-25 CN CN201910226987.1A patent/CN109946967A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102010050726A1 (en) * | 2010-11-08 | 2012-05-10 | Alphaeos Gmbh & Co. Kg | Building automation system |
WO2014072054A1 (en) * | 2012-11-07 | 2014-05-15 | Guttenberg & Hördegen | System for the analysis and control of energy flows |
CN108494605A (en) * | 2018-04-16 | 2018-09-04 | 华东师范大学 | A kind of energy net connection monitoring method and storage medium based on block chain |
CN108808663A (en) * | 2018-06-12 | 2018-11-13 | 浙江大学 | It is a kind of based on the industrial user's heat demand response method provided multiple forms of energy to complement each other |
CN108958049A (en) * | 2018-07-10 | 2018-12-07 | 三智头智能科技(佛山)有限公司 | A kind of energy monitor management method and system |
CN109379420A (en) * | 2018-10-10 | 2019-02-22 | 上海方融科技有限责任公司 | A kind of comprehensive energy service platform system based on distributed structure/architecture |
Non-Patent Citations (2)
Title |
---|
吴悦明等: "边缘计算在智慧管廊建设中的应用探究", 《智能建筑》 * |
郑宇等: "一种基于能源互联网的能源USB系统方案设计", 《新型工业化》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110440396A (en) * | 2019-07-11 | 2019-11-12 | 雄安达实智慧科技有限公司 | The central air-conditioning global optimization energy-saving control method and system of cloud side end collaboration |
CN110705694A (en) * | 2019-09-30 | 2020-01-17 | 国网山东省电力公司电力科学研究院 | Electric larceny monitoring method facing edge data center based on feature extraction |
CN110852387B (en) * | 2019-11-13 | 2022-04-22 | 江苏能来能源互联网研究院有限公司 | Energy internet super real-time state studying and judging algorithm |
CN110852387A (en) * | 2019-11-13 | 2020-02-28 | 江苏能来能源互联网研究院有限公司 | Energy internet super real-time state studying and judging algorithm |
CN110887173A (en) * | 2019-11-27 | 2020-03-17 | 国网冀北电力有限公司 | Method and device for remotely controlling central air-conditioning system |
CN110989988A (en) * | 2019-12-10 | 2020-04-10 | 天津瑞能电气有限公司 | Micro-grid edge layer software platform based on edge calculation |
CN111556093A (en) * | 2020-03-27 | 2020-08-18 | 天津市普迅电力信息技术有限公司 | Multifunctional edge Internet of things agent device for power grid information acquisition |
CN111708635A (en) * | 2020-06-16 | 2020-09-25 | 深圳天海宸光科技有限公司 | Video intelligent grading processing system and method |
CN112162849A (en) * | 2020-08-28 | 2021-01-01 | 中冶华天工程技术有限公司 | Cloud and edge computing cooperative equipment intelligent control system and method |
CN113037837A (en) * | 2021-03-05 | 2021-06-25 | 东北大学 | Intelligent terminal facing comprehensive energy system and based on edge calculation and control method thereof |
CN113820979A (en) * | 2021-09-14 | 2021-12-21 | 蓝鲸智云智能科技南京有限公司 | Energy-consumption system energy-saving strategy cloud platform based on IOT technology |
CN114492769A (en) * | 2022-04-11 | 2022-05-13 | 飞腾信息技术有限公司 | Data processing method, related equipment and system |
CN114492769B (en) * | 2022-04-11 | 2022-08-12 | 飞腾信息技术有限公司 | Data processing method, related equipment and system |
CN117092918A (en) * | 2023-10-20 | 2023-11-21 | 北京明略政睿信息技术有限公司 | Cloud edge cooperation-based intelligent gas sensing control method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109946967A (en) | A kind of building energy internet information interchanger of combination of edge intelligence | |
WO2023201916A1 (en) | Distributed flexible resource aggregation control apparatus and control method | |
CN104238466A (en) | Smart power and energy using system platform based on Cloud of Things | |
CN203147907U (en) | Urban centralized heating linkage optimal control system | |
CN107329444B (en) | A kind of comprehensive energy Multiple Time Scales coordinated control system and method | |
CN110225075A (en) | A kind of building energy internet wisdom operation cloud operating system | |
CN106096772A (en) | Energy based on intelligent power and load corporate management system and implementation method | |
CN104571068A (en) | Optimized operation control method and system of distributed energy system | |
CN110221567A (en) | Intelligent managing and control system based on 5G technology of Internet of things | |
CN102236349A (en) | System energy efficiency controller, energy efficiency grain device and intelligent energy service system for energy utilization | |
CN102236833A (en) | Universal energy network for realizing energy optimal utilization and method for providing energy transaction and service | |
CN105258306B (en) | A kind of automatic demand response device and application method towards central air conditioner system | |
CN107590612A (en) | Demand response system, demand response method, apparatus and computer-processing equipment | |
CN106597862A (en) | Building energy consumption control device and building energy consumption control method based on association rule mining | |
CN102522783B (en) | Multi-user automatic-negotiation non-synchronization micro-grid scheduling system and method | |
CN108629444B (en) | Distributed multi-zone heat supply method and system based on multi-agent control method | |
CN111787123A (en) | Intelligent heat supply network operation and maintenance management system | |
CN105048515B (en) | For how wind farm grid-connected dispatching method | |
CN106839082B (en) | Wind power heating regulator control system and regulation method | |
CN106524277B (en) | Regional energy supply system of heat supply in winter of multipotency source form | |
CN106160009B (en) | A kind of interconnected electric power system dispersion coordinated dispatching method based on risk node electricity price | |
CN205593051U (en) | Combination control system operation optimizing apparatus among distributed energy system | |
CN115102953A (en) | Power distribution network cloud edge terminal cooperative control system and method | |
CN104836846A (en) | Energy interconnection communication network architecture system based on SDN technology | |
CN103138293A (en) | Optimal distribution method and system for heat-engine plant plant-level loads |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190628 |
|
RJ01 | Rejection of invention patent application after publication |