CN116307138A - Electrical fire early warning method, device, terminal and medium based on federal learning - Google Patents
Electrical fire early warning method, device, terminal and medium based on federal learning Download PDFInfo
- Publication number
- CN116307138A CN116307138A CN202310188998.1A CN202310188998A CN116307138A CN 116307138 A CN116307138 A CN 116307138A CN 202310188998 A CN202310188998 A CN 202310188998A CN 116307138 A CN116307138 A CN 116307138A
- Authority
- CN
- China
- Prior art keywords
- early warning
- model
- warning model
- local
- load state
- 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
- 238000000034 method Methods 0.000 title claims abstract description 53
- 230000015654 memory Effects 0.000 claims description 20
- 238000012544 monitoring process Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 15
- 230000005611 electricity Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 10
- 230000007246 mechanism Effects 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000003672 processing method Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000003491 array Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002035 prolonged effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/06—Electric actuation of the alarm, e.g. using a thermally-operated switch
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention provides an electric fire early warning method, a system, a terminal and a computer storage medium based on federal learning, wherein the method comprises the following steps: acquiring a global early warning model transmitted by a server side as a local early warning model of a current period; acquiring power utilization state information of each acquisition moment, and acquiring first load state characteristics corresponding to each power utilization state information by utilizing the local early warning model; detecting whether each first load state feature meets the matching condition of the load state feature library or not, and performing fire early warning; meanwhile, whether the current state reaches a model update triggering condition is detected, if so, the local early warning model is updated based on the power utilization state information, and the global early warning model in the server is updated based on the updated local early warning model; the invention realizes the rapid and accurate early warning of the electrical load state of the user side.
Description
Technical Field
The invention relates to the field of fire safety, in particular to an electric fire early warning method, device, terminal and computer storage medium based on federal learning.
Background
With the increasing of the intelligentization and electrification degree of buildings, electric fires caused by overlarge electric loads, short circuits of lines and the like are frequent, and the number of the electric fires is in a trend of gradually rising. In order to discover hidden danger of an electric fire in time, a Non-invasive load monitoring method (Non-intrusive load monitoring, NILM) is mainly adopted at present, and the method is to install a measuring device at an inlet of a power system, and to extract the change of the electric load by monitoring and analyzing the state of the electric load data, so as to realize the identification and early warning of the electric fire load.
In order to obtain an accurate model calculation result and realize equipment management, the collected electricity consumption data is usually required to be transmitted to a government central server or a central server of a specific unit for analysis and processing; however, in the existing monitoring method, each data is transmitted to a central server, and the received data is analyzed based on an electric fire early-warning model stored on the central server to return a corresponding early-warning result, so that the early-warning process is easily influenced by the data processing capability of the server; when the data volume received by the server at the same time is large, the data processing speed is reduced, so that the early warning feedback period is prolonged, and the early warning effect is reduced; and because the electricity collection equipment used by different user ends is different, the uploaded electricity collection data has larger isomerism and other reasons, the collected data of different structures need to be processed by the electric fire early-warning model stored on the central server, so that the early-warning model is not easy to update, and when the electricity collection data of a new data structure appears, the early-warning effect of the electric fire early-warning model is further reduced when the previous electric fire early-warning model cannot detect the new electricity state characteristics.
In addition, the risk of data leakage is also high in the data acquisition, transmission and storage processes, so that the hidden privacy information of the user behaviors, habits and the like possibly generating economic benefits in the data is acquired by other people, and further the data security problem is generated.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the present invention aims to provide a fire early-warning method and device based on federal learning, which can solve the problems of the existing electrical fire early-warning method that when the data volume received by a server at the same time is large, the data processing speed is reduced, the early-warning feedback period is prolonged, and the early-warning effect is reduced.
To achieve the above and other related objects, a first aspect of the present invention provides an electrical fire early warning method based on federal learning, which is applicable to a user terminal and is used for performing each period of monitoring on an electrical load state of the user; the user terminals are connected with each other through a block chain network and are connected with the server through the block chain network; the service end stores a global early warning model for early warning of electric fire; the user terminals are stored with a preset load state feature library; the federal learning-based electric fire early warning method comprises the following steps when executing the current period: acquiring a global early warning model transmitted by a server through the blockchain network, and taking the global early warning model as a local early warning model of a current period; acquiring power utilization state information of each acquisition moment, and acquiring first load state characteristics corresponding to each power utilization state information by utilizing the local early warning model; performing fire early warning by detecting whether each first load state feature meets the matching condition of the load state feature library; and detecting whether the current state reaches a model update triggering condition while acquiring the power utilization state information, if so, updating the local early-warning model based on the acquired power utilization state information so as to update the global early-warning model in the server based on the updated local early-warning model.
In an embodiment of the present invention, the updating the local early warning model based on each of the electricity consumption status information includes: and training the local early warning model based on the power utilization state information in the current period to obtain a new local early warning model.
In an embodiment of the present invention, the updating the global early warning model in the server based on the updated local early warning model includes: the method comprises the steps of obtaining a model certification request of the new local early warning model, sending the model certification request and the new local early warning model to a blockchain network, and enabling the blockchain network to store the new local early warning model in an electrical fire monitoring model database on a chain by calling a pre-deployed related intelligent contract and executing a verification mechanism and a consensus mechanism after receiving the model certification request; and after receiving the model evidence obtaining request sent by the server, the blockchain network invokes a pre-deployed related intelligent contract, takes out the new local early warning model from an electrical fire monitoring model database on the chain, and sends the new local early warning model to the server for storage so as to update the global early warning model in the server.
In an embodiment of the present invention, the electrical fire early warning method based on federal learning further includes: and encrypting the updated local early warning model to update the global early warning model in the server based on the encrypted local early warning model.
In an embodiment of the present invention, the encryption processing manner of the updated local early warning model includes: and encrypting the updated local early warning model by using a bloom filtering and random response mechanism method to obtain an encrypted local early warning model.
In an embodiment of the present invention, the implementation manner of performing fire early warning by detecting whether each of the first load status features meets the matching condition of the load status feature library includes: detecting whether the characteristic value of the first load state characteristic is located in a threshold range of the corresponding load state characteristic in the load state characteristic library, if so, judging that the first load state characteristic meets the matching condition of the load state characteristic library, and if not, judging that the first load state characteristic does not meet the matching condition of the load state characteristic library.
The present invention provides in a second aspect an electrical fire early warning device based on federal learning for performing each cycle of monitoring of an electrical load state of a user; the electric fire early warning equipment is connected with the acquisition equipment to acquire electric power utilization state information of a user; and each electric fire early warning device is connected with each other and is connected with a service end through a block chain network: the service end stores a global early warning model for early warning of electric fire; the electrical fire early warning device includes: the system comprises a communication unit, a storage unit, an analysis early warning unit and a model updating unit; the communication unit is connected with the blockchain network and is used for acquiring the global early warning model transmitted by the server through the blockchain network and storing the global early warning model into the storage unit as a local early warning model of the current period; the storage unit is connected with the communication unit and used for storing the local early warning model; the storage unit is also used for storing a preset load state feature library and storing power utilization state information of each acquisition moment in the current period; the analysis early warning unit is connected with the storage unit and is used for acquiring power utilization state information at each acquisition moment and acquiring first load state characteristics corresponding to each power utilization state information by utilizing the local early warning model while acquiring each power utilization state information; and fire disaster early warning is carried out by detecting whether each first load state characteristic meets the matching condition of the load state characteristic library; the model updating unit is connected with the analysis early warning unit, the storage unit and the communication unit and is used for detecting whether the current state reaches a model updating triggering condition or not, if so, updating the local early warning model based on the power utilization state information, and controlling the communication unit to transmit the updated local early warning model to a blockchain network so as to update the global early warning model in the server.
In an embodiment of the invention, the electrical fire early warning device based on federal learning further includes: the encryption unit is connected with the model updating unit and the storage unit and is used for carrying out encryption processing on the updated local early warning model and storing the encrypted local early warning model into the storage unit.
The present invention also provides, in a third aspect, a terminal comprising: a processor and a memory; the memory is used for storing a computer program, and the processor is used for executing the computer program stored in the memory to execute the federal learning-based electric fire early warning method.
The present invention also provides in a fourth aspect a computer storage medium storing a computer program which when executed by a processor implements the federal learning-based electrical fire early warning method as any of the above.
As described above, the federal learning-based electric fire early warning method, system, terminal and computer storage medium provided by the invention form a local early warning model in a user terminal by transmitting the global early warning model in a server terminal to the user terminal for storage; load state feature extraction is carried out on the collected power utilization state information based on a local early warning model, and a monitoring result of whether the electric load state is normal or not in the current period is obtained by detecting the matching degree between the load state feature and a load state feature library; and updating the local early warning model by utilizing the power utilization state information acquired in the current period to update the service-side global early warning model based on the local model, so that the service-side global early warning model is updated, and the early warning information of the electric load state of the user side can be rapidly and accurately acquired; the method can also realize the rapid and high-frequency updating of the global early-warning model in the service end, further can update the local early-warning model in each user end at high frequency, and improves the accuracy and high efficiency of electric fire early warning of each user end in the whole network.
Drawings
FIG. 1 shows an application field of an embodiment of an electric fire early warning method based on federal learning according to the present invention
A scene diagram;
FIG. 2 is a flow chart of an electrical fire early warning method based on federal learning according to an embodiment of the present invention
A schematic diagram;
FIG. 3 shows a flow of an electrical fire warning method based on federal learning according to another embodiment of the present invention
A schematic program diagram;
FIG. 4 shows an electrical fire early warning device based on federal learning according to an embodiment of the present invention
A schematic diagram;
FIG. 5 shows a junction of the federal learning-based electric fire early warning device according to the present invention in another embodiment
Schematic construction;
FIG. 6 is a schematic diagram of a terminal according to an embodiment of the invention;
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic idea of the present invention by way of illustration.
In order to solve the technical problems in the prior art, the invention provides an electric appliance fire early warning method based on federal learning in a first aspect, which is suitable for a user side and is used for executing each period of monitoring on the electric load state of the user so as to obtain a monitoring result of whether the load state is normal in each period.
Alternatively, the duration of a single cycle is any number of 1 to 5 days.
Referring to fig. 1, a schematic diagram of an application scenario of the electrical fire early warning method provided by the invention in an embodiment is shown. As shown in fig. 1, the user side is connected with the acquisition device to acquire the electrical power utilization state information of the user in real time; the user terminals are connected with each other through a block chain network and are connected with the service terminal through the block chain network.
Wherein, the service end stores an electric fire early warning model; the electric fire early warning model is a model containing the relation between the electricity utilization state information and the load state characteristics so as to acquire the load state characteristics corresponding to the user based on the electricity utilization state information.
The user side stores a preset load state feature library; the load state feature library comprises matching conditions corresponding to the load state features.
Optionally, the feature matching condition is a change threshold range corresponding to each load state feature.
Optionally, the electricity consumption state information includes, but is not limited to, one or more of voltage value, current value, active power, reactive power and power factor.
Optionally, the service end is a fire early warning platform.
Referring to fig. 2, a schematic flow chart of the federal learning-based electric fire early warning method according to an embodiment of the present invention in a single cycle is shown; as shown in fig. 1, the method, when executing the current cycle, includes:
s100, acquiring a global early warning model transmitted by a server based on the blockchain network, and storing the global early warning model as a local early warning model of a current period;
the global early warning model is the latest electric fire early warning model stored in the server.
Specifically, when executing the current period, the server transmits the global early warning model stored in the server to each user terminal through a blockchain network.
And the user side stores the global early warning model as a local early warning model after receiving the global early warning model.
S200, acquiring power utilization state information of each acquisition time in the current period; after the electricity utilization state information of each acquisition moment is acquired, acquiring first load state characteristics corresponding to each piece of electricity utilization state information by utilizing the local early warning model; detecting whether each first load state feature meets the matching condition of the load state feature library; if so, judging that the corresponding acquisition time is in a normal load state; if not, judging the corresponding acquisition time as a risk load state, and carrying out fire early warning;
specifically, based on a preset acquisition frequency, acquiring the electric power utilization state information of a user by using acquisition equipment so as to obtain the electric power utilization state information acquired at each acquisition moment; and simultaneously, sequentially inputting the obtained power utilization state information into the local model for calculation so as to respectively obtain first load state characteristics corresponding to the acquisition time.
Comparing the first load state characteristics of each acquisition moment with each load state range in the load state characteristic library, and judging that the corresponding acquisition moment is a normal load state if the characteristic value corresponding to the first load state characteristics is in the load state range, namely judging that the first load state characteristics meet the matching condition of the load state characteristic library; if the characteristic value corresponding to the first load state characteristic is out of the load state range, namely, the first load state characteristic is judged to not meet the matching condition of the load state characteristic library, the corresponding acquisition time is judged to be a risk load state, and fire disaster early warning is carried out.
Specifically, the user terminal is connected with an early warning unit; and when the user side judges that the risk load state exists in the current period, generating a fire early-warning signal, and sending the fire early-warning signal to the early-warning unit for fire early-warning.
And S300, detecting whether the current state reaches a model update triggering condition while acquiring the power utilization state information, if so, updating the local early warning model based on the acquired power utilization state information, and updating the global early warning model in the server based on the updated local early warning model.
The model update triggering condition is a triggering condition for executing the local early warning model update;
optionally, the model update triggering condition includes reaching a preset time.
Optionally, the model update triggering condition includes that the total collection amount of the collected electricity utilization state information in the current period reaches a threshold value.
Training the local early warning model based on the total power utilization state information in the current period when the current state reaches the model updating triggering condition so as to obtain a new local early warning model; after the new local early-warning model is obtained, the new local early-warning model is uploaded to a server based on a blockchain network so as to update the global early-warning model in the server based on the new local early-warning model.
In one embodiment, the blockchain-based network uploads the new local alert model to a server, including:
the method comprises the steps of obtaining a model certification request of the new local early warning model, sending the model certification request and the new local early warning model to a blockchain network, and enabling the blockchain network to store the new local early warning model in an electrical fire monitoring model database on a chain by calling a pre-deployed related intelligent contract and executing a verification mechanism and a consensus mechanism after receiving the model certification request; and after receiving the model evidence obtaining request sent by the server, the blockchain network invokes a pre-deployed related intelligent contract, takes out the new model from the electrical fire monitoring model database on the chain, and sends the new model to the server for storage so as to update the global early warning model in the server.
Optionally, the updating manner of the global early warning model in the server includes:
the server obtains each new model from the blockchain network in the current period, decodes and aggregates each new model to obtain an aggregated total model; and storing the total model as a new global early warning model.
After the step S300 is performed, the process returns to the step S100 to perform the electrical fire early warning process of the next cycle, so as to implement electrical fire early warning for the user side.
In another embodiment, in the performing step S300, as shown in fig. 3, the federal learning-based electrical fire early warning method is as follows:
and S300', detecting whether the current state reaches a model update triggering condition while acquiring the power utilization state information, if so, updating the local early warning model based on the acquired power utilization state information, and encrypting the updated local early warning model to update the fire early warning model in the server based on the encrypted local early warning model.
Specifically, when judging that the risk load characteristics exist in the current period, updating the local early warning model; and carrying out differential privacy encryption on the updated local early warning model by using a bloom filtering and random response mechanism method so as to obtain the encrypted local early warning model.
In an embodiment, the differential privacy processing method based on the random response mechanism includes:
firstly, carrying out hash operation by using a bloom filter with specified hash number to obtain a binary string with specified bit number; then, carrying out first random response, and turning over or reserving each binary bit according to a preset random rule; when the unit function is executed on the local early warning model, the binary string is directly used for executing a second random response, namely each binary bit is turned over or reserved according to a preset random rule different from the first random rule, and a new binary string is obtained; and adding the new binary string into the updated local early-warning model to realize differential privacy treatment of the local early-warning model.
It should be noted that the bloom filter, the first random rule and the second random rule adopted in each of the clients are the same.
In order to solve the technical problems in the prior art, the invention provides an electric fire early warning device based on federal learning in a second aspect, which is suitable for a user side and is used for executing each period of monitoring on an electric load state of the user side so as to obtain a monitoring result of whether the load state is normal in each period.
The electric fire early warning device is connected with the acquisition equipment to acquire electric power utilization state information of a user; and each electric fire early warning device is connected with each other and is connected with the service end through a block chain network.
Referring to fig. 4, a schematic structural diagram of the electrical fire early warning device according to an embodiment of the present invention is shown; as shown in fig. 4, the electric fire early warning device 10 includes: the system comprises a communication unit 11, a storage unit 12, an analysis and early warning unit 13 and a model updating unit 14.
The communication unit 11 is connected to a blockchain network, and is configured to obtain a global early-warning model transmitted by a server through the blockchain network, and store the global early-warning model as a local early-warning model in a current period in the storage unit 12;
the storage unit 12 is connected with the communication unit 11 and is used for storing the local early warning model; the storage unit 12 is connected with the acquisition unit and is used for storing the power utilization state information of each acquisition time in the current period; and, the storage unit 12 is further configured to store a preset load state feature library;
the analysis and early warning unit 13 is connected with the storage unit 12, and is used for acquiring power utilization state information of each acquisition time in a current period, and acquiring first load state characteristics corresponding to each power utilization state information by utilizing the local early warning model while acquiring each power utilization state information; detecting whether each first load state feature meets the matching condition of the load state feature library; if yes, judging that the current period is in a normal load state; if not, judging that the risk load state exists in the current period, and carrying out fire early warning;
the model updating unit 14 is connected to the analysis and early warning unit 13, the storage unit 12, and the communication unit 11, and is configured to detect whether the current state reaches a model update triggering condition, if so, update the local early warning model based on the power consumption state information, and control the communication unit to transmit the updated local early warning model to a blockchain network, so as to update the global early warning model in the server.
Specifically, a specific implementation manner of updating the local early warning model based on the power consumption state information is the same as that in the foregoing embodiment, and will not be described herein again.
Likewise, a specific implementation manner of transmitting the updated local early warning model to the blockchain network to update the global early warning model in the server is the same as the implementation manner in the foregoing embodiment, and will not be described herein again.
In another embodiment, as shown in fig. 5, the electrical fire early warning device further includes: an encryption unit 15; the encryption unit 15 is connected to the model updating unit 14 and the storage unit 12, and is configured to encrypt the updated local early warning model, and store the encrypted local early warning model in the storage unit.
Specifically, the specific implementation manner of the encryption unit for performing encryption processing on the updated local early warning model is the same as that in the foregoing embodiment, and will not be described herein again.
In order to solve the technical problems in the prior art, the invention further provides a terminal; referring to fig. 6, a schematic diagram of the structure of the terminal in one embodiment is shown.
As shown in fig. 6, the terminal 40 includes: a memory 41 and a processor 42.
Wherein the memory 41 is configured to store a computer program: the processor 42 runs a computer program to perform the federal learning-based electrical fire early warning method as described above.
Alternatively, the number of the memories 41 may be one or more, and the number of the processors 42 may be one or more.
Optionally, the processor 42 loads one or more instructions corresponding to the process of the application program into the memory 41 according to the steps in the electric fire early warning method based on federal learning, and the processor 42 runs the application program stored in the first memory 41, so as to implement the steps in the electric fire early warning method based on federal learning; or the processor 42 may load one or more instructions corresponding to the process of the application program into the memory 41 according to the steps in the federal learning-based electric fire early warning method as described above, and the processor 42 executes the application program stored in the first memory 41, thereby implementing the steps in the federal learning-based electric fire early warning method as described above.
Optionally, the memory 41 includes, but is not limited to, high speed random access memory, nonvolatile memory. Such as one or more disk storage devices, flash memory devices, or other non-volatile solid-state storage devices; the processor 42 may include, but is not limited to, a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processing, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
Alternatively, the processor 42 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processing, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
To solve the technical problems in the prior art, the present invention further provides a computer readable storage medium, on which a computer program is stored, which when called by a processor, implements the federal learning-based electric fire early warning method as described above.
Wherein the computer-readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: portable computer disks, hard disks, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static Random Access Memory (SRAM), portable compact disk read-only memory (CD-ROM), digital Versatile Disks (DVD), memory sticks, floppy disks, mechanical coding devices.
The computer readable program described herein may be downloaded from a computer readable storage medium to a respective computing/processing device or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network interface card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium in the respective computing/processing device.
Computer program instructions for carrying out operations of the present application may be assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and a procedural programming language such as the "C" language or similar programming languages.
The computer readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In some embodiments, aspects of the present application are implemented by personalizing electronic circuitry, such as programmable logic circuitry, field Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), with state information for computer readable program instructions, which may execute the computer readable program instructions.
In summary, according to the federal learning-based electric fire early warning method, system, terminal and computer storage medium provided by the invention, the global early warning model in the server side is transmitted to the user side to form the local early warning model in the user side; load state feature extraction is carried out on the collected power utilization state information based on a local early warning model, and a monitoring result of whether the electric load state is normal or not in the current period is obtained by detecting the matching degree between the load state feature and a load state feature library; and updating the local early warning model by utilizing the power utilization state information acquired in the current period to update the service end global early warning model based on the local model, so that early warning information of the electric load state of the user end can be rapidly and accurately acquired; the global early warning model in the service end can be updated rapidly and at high frequency, so that the local early warning model in each user end can be updated at high frequency, and the accuracy and the high efficiency of electric fire early warning of each user end in the whole network are improved; in addition, the updated early warning model is encrypted by adopting a blockchain technology, so that the safety of data transmission is improved, and the leakage risk of user data privacy is greatly reduced.
The above embodiments are merely illustrative of the principles of the present invention and its effectiveness, and are not intended to limit the invention. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Accordingly, it is intended that all equivalent modifications and variations of the invention be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.
Claims (10)
1. The electric fire early warning method based on federal learning is characterized by being suitable for a user side and used for executing each period of monitoring on the electric load state of a user; the user terminals are connected with each other through a block chain network and are connected with the server through the block chain network; the service end stores a global early warning model for early warning of electric fire; the user terminals are stored with a preset load state feature library;
the federal learning-based electric fire early warning method comprises the following steps when executing the current period:
acquiring a global early warning model transmitted by a server through the blockchain network, and taking the global early warning model as a local early warning model of a current period;
acquiring power utilization state information of each acquisition moment, and acquiring first load state characteristics corresponding to each power utilization state information by utilizing the local early warning model; performing fire early warning by detecting whether each first load state feature meets the matching condition of the load state feature library;
and detecting whether the current state reaches a model update triggering condition while acquiring the power utilization state information, if so, updating the local early-warning model based on the acquired power utilization state information so as to update the global early-warning model in the server based on the updated local early-warning model.
2. The federal learning-based electrical fire early-warning method according to claim 1, wherein updating the local early-warning model based on each of the electricity consumption state information comprises:
and training the local early warning model based on the power utilization state information in the current period to obtain a new local early warning model.
3. The federal learning-based electrical fire early-warning method according to claim 2, wherein the updating the global early-warning model in the server based on the updated local early-warning model comprises:
the method comprises the steps of obtaining a model certification request of the new local early warning model, sending the model certification request and the new local early warning model to a blockchain network, and enabling the blockchain network to store the new local early warning model in an electrical fire monitoring model database on a chain by calling a pre-deployed related intelligent contract and executing a verification mechanism and a consensus mechanism after receiving the model certification request; and after receiving the model evidence obtaining request sent by the server, the blockchain network invokes a pre-deployed related intelligent contract, takes out the new local early warning model from an electrical fire monitoring model database on the chain, and sends the new local early warning model to the server for storage so as to update the global early warning model in the server.
4. The federal learning-based electrical fire early warning method according to claim 1, further comprising:
and encrypting the updated local early warning model to update the global early warning model in the server based on the encrypted local early warning model.
5. The federal learning-based electrical fire early warning method according to claim 4, wherein the encryption processing method of the updated local early warning model comprises:
and encrypting the updated local early warning model by using a bloom filtering and random response mechanism method to obtain an encrypted local early warning model.
6. The federal learning-based electric fire early warning method according to claim 1, wherein the implementation of fire early warning by detecting whether each of the first load status features satisfies the matching condition of the load status feature library comprises:
detecting whether the characteristic value of the first load state characteristic is located in a threshold range of the corresponding load state characteristic in the load state characteristic library, if so, judging that the first load state characteristic meets the matching condition of the load state characteristic library, and if not, judging that the first load state characteristic does not meet the matching condition of the load state characteristic library.
7. An electric fire early warning device based on federal learning is characterized by being used for performing each period of monitoring on the electric load state of a user; the electric fire early warning equipment is connected with the acquisition equipment to acquire electric power utilization state information of a user; and each electric fire early warning device is connected with each other and is connected with a service end through a block chain network: the service end stores a global early warning model for early warning of electric fire;
the electrical fire early warning device includes: the system comprises a communication unit, a storage unit, an analysis early warning unit and a model updating unit;
the communication unit is connected with the blockchain network and is used for acquiring the global early warning model transmitted by the server through the blockchain network and storing the global early warning model into the storage unit as a local early warning model of the current period;
the storage unit is connected with the communication unit and used for storing the local early warning model; the storage unit is also used for storing a preset load state feature library and storing power utilization state information of each acquisition moment in the current period;
the analysis early warning unit is connected with the storage unit and is used for acquiring power utilization state information at each acquisition moment and acquiring first load state characteristics corresponding to each power utilization state information by utilizing the local early warning model while acquiring each power utilization state information; and fire disaster early warning is carried out by detecting whether each first load state characteristic meets the matching condition of the load state characteristic library;
the model updating unit is connected with the analysis early warning unit, the storage unit and the communication unit and is used for detecting whether the current state reaches a model updating triggering condition or not, if so, updating the local early warning model based on the power utilization state information, and controlling the communication unit to transmit the updated local early warning model to a blockchain network so as to update the global early warning model in the server.
8. The federal learning-based electrical fire early warning apparatus of claim 7, further comprising: the encryption unit is connected with the model updating unit and the storage unit and is used for carrying out encryption processing on the updated local early warning model and storing the encrypted local early warning model into the storage unit.
9. A terminal, comprising: a processor and a memory; the memory is used for storing a computer program, and the processor is used for executing the computer program stored in the memory to execute the federal learning-based electric fire early warning method according to any one of claims 1 to 6.
10. A computer storage medium storing a computer program, wherein the computer program when executed by a processor implements the federal learning-based electrical fire early warning method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310188998.1A CN116307138A (en) | 2023-03-01 | 2023-03-01 | Electrical fire early warning method, device, terminal and medium based on federal learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310188998.1A CN116307138A (en) | 2023-03-01 | 2023-03-01 | Electrical fire early warning method, device, terminal and medium based on federal learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116307138A true CN116307138A (en) | 2023-06-23 |
Family
ID=86837278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310188998.1A Pending CN116307138A (en) | 2023-03-01 | 2023-03-01 | Electrical fire early warning method, device, terminal and medium based on federal learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116307138A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117237804A (en) * | 2023-09-15 | 2023-12-15 | 江苏三棱智慧物联发展股份有限公司 | Pyrotechnical recognition system and method based on federal learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363305A (en) * | 2019-07-17 | 2019-10-22 | 深圳前海微众银行股份有限公司 | Federal learning method, system, terminal device and storage medium |
| CN110533419A (en) * | 2019-07-24 | 2019-12-03 | 阿里巴巴集团控股有限公司 | Sharing method, device and the server of air control model based on block chain |
| CN113160021A (en) * | 2021-03-18 | 2021-07-23 | 天津中科物联科技有限公司 | Safe production early warning system based on multi-source heterogeneous data federal learning |
| CN113901412A (en) * | 2021-10-18 | 2022-01-07 | 国网山东省电力公司济宁供电公司 | Power quality disturbance detection method and device, electronic equipment and storage medium |
| CN113922361A (en) * | 2021-09-24 | 2022-01-11 | 西安理工大学 | Potential safety hazard and fault protection method based on non-invasive power distribution and utilization information decomposition |
| CN114358307A (en) * | 2021-11-30 | 2022-04-15 | 北京邮电大学 | Federal learning method and device based on differential privacy law |
| CN114781653A (en) * | 2022-06-21 | 2022-07-22 | 平安科技（深圳）有限公司 | Model training method, system, device and storage medium based on artificial intelligence |
-
2023
- 2023-03-01 CN CN202310188998.1A patent/CN116307138A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363305A (en) * | 2019-07-17 | 2019-10-22 | 深圳前海微众银行股份有限公司 | Federal learning method, system, terminal device and storage medium |
| CN110533419A (en) * | 2019-07-24 | 2019-12-03 | 阿里巴巴集团控股有限公司 | Sharing method, device and the server of air control model based on block chain |
| CN113160021A (en) * | 2021-03-18 | 2021-07-23 | 天津中科物联科技有限公司 | Safe production early warning system based on multi-source heterogeneous data federal learning |
| CN113922361A (en) * | 2021-09-24 | 2022-01-11 | 西安理工大学 | Potential safety hazard and fault protection method based on non-invasive power distribution and utilization information decomposition |
| CN113901412A (en) * | 2021-10-18 | 2022-01-07 | 国网山东省电力公司济宁供电公司 | Power quality disturbance detection method and device, electronic equipment and storage medium |
| CN114358307A (en) * | 2021-11-30 | 2022-04-15 | 北京邮电大学 | Federal learning method and device based on differential privacy law |
| CN114781653A (en) * | 2022-06-21 | 2022-07-22 | 平安科技（深圳）有限公司 | Model training method, system, device and storage medium based on artificial intelligence |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117237804A (en) * | 2023-09-15 | 2023-12-15 | 江苏三棱智慧物联发展股份有限公司 | Pyrotechnical recognition system and method based on federal learning |
| CN117237804B (en) * | 2023-09-15 | 2024-02-13 | 江苏三棱智慧物联发展股份有限公司 | Pyrotechnical recognition system and method based on federal learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112714193B (en) | Community data processing system, method, device and equipment based on Internet of things platform | |
| US10216959B2 (en) | Method and systems using privacy-preserving analytics for aggregate data | |
| WO2021063068A1 (en) | Operation and maintenance control and operation and maintenance analysis method and apparatus, system, and storage medium | |
| CN105723378B (en) | Protection system including safety regulation assessment | |
| US8751414B2 (en) | Identifying abnormalities in resource usage | |
| US20200293657A1 (en) | Anomalous behavior detection in processor based systems | |
| Fotovvat et al. | Comparative performance analysis of lightweight cryptography algorithms for IoT sensor nodes | |
| US20160258845A1 (en) | Self-learning simulation environments | |
| US9485222B2 (en) | Data stream traffic control | |
| CN109474603B (en) | Data packet grabbing processing method and terminal equipment | |
| CN108256322B (en) | Security testing method and device, computer equipment and storage medium | |
| CN116307138A (en) | Electrical fire early warning method, device, terminal and medium based on federal learning | |
| CN113225339B (en) | Network security monitoring method and device, computer equipment and storage medium | |
| La Marra et al. | D-BRIDEMAID: A Distributed Framework for Collaborative and Dynamic Analysis of Android Malware. | |
| Zhang et al. | Privacy-preserved mobile sensing through hybrid cloud trust framework | |
| Wu et al. | Bayesian model updating method based android malware detection for IoT services | |
| EP3647982B1 (en) | Cyber attack evaluation method and cyber attack evaluation device | |
| US11681938B2 (en) | Intelligent signal matching of disparate input data in complex computing networks | |
| US20220027490A1 (en) | Gan-based data obfuscation decider | |
| CN112165484B (en) | Network encryption traffic identification method and device based on deep learning and side channel analysis | |
| CN113132906B (en) | Virtual positioning identification method and device, storage medium and electronic equipment | |
| Saneja et al. | A scalable correlation‐based approach for outlier detection in wireless body sensor networks | |
| CN116519095A (en) | Instrument state diagnosis and response method, device, equipment and storage medium | |
| CN111667190B (en) | Electric power construction grounding monitoring method, device and server | |
| CN113507691A (en) | Information pushing system and method based on power distribution network cross-region service |
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 |