CN113224852A - Power communication management method and system based on AI auxiliary decision - Google Patents
Power communication management method and system based on AI auxiliary decision Download PDFInfo
- Publication number
- CN113224852A CN113224852A CN202110594898.XA CN202110594898A CN113224852A CN 113224852 A CN113224852 A CN 113224852A CN 202110594898 A CN202110594898 A CN 202110594898A CN 113224852 A CN113224852 A CN 113224852A
- Authority
- CN
- China
- Prior art keywords
- information
- environment information
- power
- neuron
- communication management
- 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
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- 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
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Human Computer Interaction (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention relates to an AI (AI-aid decision) -based power communication management method, which comprises the following steps of: step A, collecting image information and extracting features, analyzing the safety and peripheral safety conditions of equipment through the features, sending an alarm if a safety problem exists, and simultaneously sending the place and time with the safety problem; b, collecting environment information, comparing the change conditions of the current environment information and the previous environment information, if the change of the environment information is positively correlated with the change of the power consumption, correspondingly increasing or decreasing the power dispatching according to the increase or decrease of the environment information, and if the change of the environment information is negatively correlated with the change of the power consumption, correspondingly decreasing or increasing the power dispatching according to the increase or decrease of the environment information; and C, acquiring line load data and scheduling according to the circuit load power. The invention can realize the optimal configuration of power resources and improve the automation and the unmanned performance of a power system.
Description
Technical Field
The invention relates to the technical field of power transmission, in particular to a power communication management method and system based on AI auxiliary decision.
Background
With the improvement of the living standard of people, the popularization range of power transmission is wider and wider. And the load power of the electric power is getting larger. The management and maintenance of electric power facilities are extremely difficult. The traditional SCADA system has the defects of slow speed and more small transmission devices, and causes difficulty in monitoring and decision-making of the devices. Compared with the mode of investing a large amount of labor, how to reduce the safety hazard of high voltage to maintenance personnel, and the visual and more convenient management of the client side is the direction of power system intellectualization.
In the existing system, the service for the equipment of the basic level is not synchronous with the service of the business hall, and the place and the time are often required to be reported after the fault occurs; nearby personnel are rescheduled for a corresponding treatment and the power facility is often also in an environment that can pose a safety hazard to the personnel.
On the other hand, the process of power utilization can generate centralized power utilization along with the problems of time, weather and the like. And relatively idle during other periods. The problem of the concentration of the electrical load makes timely scheduling very important.
For the aspect of power grid maintenance, data changes can be caused by abnormal lines because damage to power equipment is not sudden. The on-off state of a knife switch in the traditional power grid needs manual operation, so that the switch is missed or missed; because the place of the electric power facility is often spacious, the danger of the electric power facility or the electric shock danger of the surrounding organisms can be caused. How to increase monitoring and management and improve the safety of electric power facilities is also an urgent problem to be solved.
Utility management of electricity is not only directed to technicians. But also for people on non-technical posts. The traditional power grid has higher requirements on non-technical post personnel, and the requirements on data monitoring and visualization for improving the processing efficiency are more and more urgent.
Disclosure of Invention
The invention aims to provide an AI-aid decision-based power communication management method and system which combine multiple factors to provide an intelligent solution aiming at the defects in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a power communication management method based on AI auxiliary decision comprises the following steps:
step A, collecting image information and extracting features, analyzing the safety and peripheral safety conditions of equipment through the features, sending an alarm if a safety problem exists, and simultaneously sending the place and time with the safety problem;
b, collecting environment information, comparing the change conditions of the current environment information and the previous environment information, if the change of the environment information is positively correlated with the change of the power consumption, correspondingly increasing or decreasing the power dispatching according to the increase or decrease of the environment information, and if the change of the environment information is negatively correlated with the change of the power consumption, correspondingly decreasing or increasing the power dispatching according to the increase or decrease of the environment information;
and C, acquiring line load data, increasing the power dispatching of the line node if the line load is increased, and reducing the power dispatching of the line node if the line load is reduced.
Further, the step a further includes detecting biological information, and comprehensively determining whether a biological exists near the power equipment through the infrared detected temperature information and the image information acquired by the image, and if so, sending an alarm to remind.
Further, in the step B, a BP neural network for predicting the grid load is constructed by using the environmental information, and the following neurons are constructed:
first neuronRepresents the lowest load of the previous day; second neuronThe highest temperature of the previous day is the highest temperature,is the lowest temperature of the previous day; third neuronIs the highest humidity of the previous day,the lowest humidity for the previous day; fourth neuron Typei-1Representing the weather type of the previous day; fifth neuronRepresenting the lowest load of the first two days; the sixth neuronThe highest temperature of the first two days,the lowest temperature of the first two days;the first neuronThe highest humidity of the first two days,the lowest humidity for the first two days; type of the eighth neuroni-2Representing the weather type of the first two days;
the mathematical expression of the BP neural network is:
where B is the noise data, G is the noise data, F is the coefficient,
defining a target error function
Computing
Ok=Ok-1+D
This is assigned to the desired output y.
Further, the GA optimizes BP as follows:
step M1, calculating an adaptive function, and calculating the weight and the threshold of the population group by using the BP neural network, so that the individual fitness:
wherein m is the number of neurons in the output layer
Desired output y, predicted output o;
step M2, the selection operation,
selecting excellent individuals to be inherited to the next generation according to the individual fitness, and selecting a wheel disc calculation method;
fi=S/Fi
wherein F is the fitness of the individual, F is the fitness function of the individual, and S is a selection coefficient; p is the individual selection probability;
step M3, the cross-over operation,
the GA algorithm adopts a real number intersection method:
axi=axi(1-b)+ayib
ayi=ayi(1-b)+axib
axiis the ith position of the x chromosome, ayiPosition i of the y chromosome;
in step M4, the mutation operation,
after selecting an individual, the probability of some genes is converted into other alleles:
amaxis a of a geneijUpper bound, aminIs a of a geneijUpper bound, r2Is a random number, G is the number of iterations, GmaxMaximum number of iterations, r random number.
More specifically, the method includes classifying the image information by a cascade convolution calculation:
step D1, inputting a picture with a resolution of 100 × 100, performing 1 × 1 convolution on the first convolution layer to obtain a 100 × 100 picture, and performing processing by using a relu nonlinear activation function; convolution with 5 × 5 is used in the second convolution layer, so that a 96 × 96 picture is obtained and processed by using a relu nonlinear activation function;
step D2, performing a first layer of downsampling, using 2 × 2 downsampling to obtain 48 × 48 pictures;
step D3, using convolution of 1 × 1 on the third convolution layer to obtain 48 × 48 pictures, and processing them by using relu nonlinear activation function; using 5 × 5 convolution on the fourth convolution layer to obtain 44 × 44 pictures, and processing the pictures by using a relu nonlinear activation function;
step D4, performing a second layer of downsampling, using 2 × 2 downsampling to obtain 22 × 22 pictures;
step D5, fully connecting 1024 neurons in the first fully-connected layer, and processing by using a relu nonlinear activation function;
and D6, fully connecting 2 neurons in the second fully-connected layer to obtain an output layer.
An AI-aided decision-based power communication management system, comprising:
the acquisition module is used for acquiring image information, line information and environment information;
the transmission module is used for transmitting the information acquired by the acquisition module to the decision module;
and the decision-making module is used for providing operation safety and surrounding safety warning according to the image information acquired by the acquisition module and providing power utilization scheduling analysis according to the line information and the environment information acquired by the acquisition module.
To illustrate, the acquisition module includes:
the video detection device is used for acquiring video images;
and the infrared detection device is used for collecting biological infrared information.
To illustrate, the acquisition module includes:
date detection means for recording a date;
the temperature detection device is used for acquiring temperature information;
and the humidity detection device is used for acquiring humidity information.
The technical scheme can bring the following beneficial effects:
1. the patent proposes that circuit scheduling is realized through calculation self-adaption of a neural network according to collection of weather, temperature, time and humidity. The power utilization efficiency is improved, and the optimal configuration of power resources is realized.
2. This patent is to power equipment's control, the data monitoring of line transmission. Remote 5G repair is started for simple abnormal problems, and large equipment is damaged. The system can self-locate the damaged address and time, and is convenient for maintenance personnel to maintain the equipment.
3 this patent proposes visualization data and processing. In the visualization center, the client receives all the collected data, and the data are calculated by the server. The prior ticket distribution, pre-judgment, in-process treatment and evidence collection are realized. And (5) overload processing. The automation and the unmanned degree of the power system are improved.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is a schematic diagram of one embodiment of the present invention;
FIG. 2 is an image classification flow diagram of one embodiment of the present invention;
FIG. 3 is a table of neural network parameters for one embodiment of the present invention;
FIG. 4 is a schematic diagram of neural network parameter relationships according to one embodiment of the present invention;
FIG. 5 is a diagram of a neural network BP of one embodiment of the present invention;
FIG. 6 is a code map of an evolved neural network structure and chromosomes according to one embodiment of the invention.
Detailed Description
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
As shown in fig. 1 to 6, a power communication management method based on AI-assisted decision includes the following steps:
step A, collecting image information and extracting features, analyzing the safety and peripheral safety conditions of equipment through the features, sending an alarm if a safety problem exists, and simultaneously sending the place and time with the safety problem;
b, collecting environment information, comparing the change conditions of the current environment information and the previous environment information, if the change of the environment information is positively correlated with the change of the power consumption, correspondingly increasing or decreasing the power dispatching according to the increase or decrease of the environment information, and if the change of the environment information is negatively correlated with the change of the power consumption, correspondingly decreasing or increasing the power dispatching according to the increase or decrease of the environment information;
and C, acquiring line load data, increasing the power dispatching of the line node if the line load is increased, and reducing the power dispatching of the line node if the line load is reduced.
The power facilities are usually located in a relatively open area, and the aging failure of the power facilities and the safety of the surrounding environment are difficult to guarantee. Carry out characteristic analysis through gathering image information, can carry out remote monitoring to electrical equipment's outward appearance, if damaged, then can remind the staff to go the maintenance at once. Meanwhile, the life of the equipment can be comprehensively judged by combining image information according to the used age of the electrical equipment, for example, the remaining life of the electrical equipment is consistent with the normal service life of the electrical equipment under the condition that the appearance of the electrical equipment is intact, and the remaining life of the electrical equipment is shorter than the normal service life of the electrical equipment under the condition that the appearance of the electrical equipment is damaged. The opening and closing conditions of the disconnecting link can be judged through the image information. Similarly, the fault condition of the electrical equipment can be judged through image acquisition, whether the working environment is safe can be judged through image acquisition of the external environment, and basic field information is provided according to the working environment, so that maintenance personnel can be helped to prepare before going to a fault place.
Different environmental information has different effects on the electrical load. Taking date as an example, in spring festival, New year's day and other traditional festivals, the electricity consumption of residents will increase sharply in the spring festival, but on the contrary, the electricity consumption of industry will decrease relative to working days. In the case of weather, the electricity utilization conditions in different regions are different in cold and hot weather, and regional centralized electricity utilization can occur in the weather. Taking humidity as an example, different humidities can affect the operation of enterprises, and under the condition that the change of the humidity of the external environment is obvious, centralized power utilization can also occur when the indoor temperature and humidity are kept constant. The environmental factors are comprehensively analyzed, and the relevance between the power utilization conditions of different areas in the previously collected data and the environmental factors is judged, so that the subsequent power utilization conditions are reasonably presumed, and a power utilization scheduling decision is provided.
The detection of the line load can not only provide reference information of power utilization scheduling in real time, but also reasonably predict the relationship between the power utilization condition and the environmental information through the algorithm by collecting the power utilization condition and the environmental information, and provide decision reference for the power utilization scheduling.
Further, the step a further includes detecting biological information, and comprehensively determining whether a biological exists near the power equipment through the infrared detected temperature information and the image information acquired by the image, and if so, sending an alarm to remind.
If there is a living thing around the electrical equipment, not only the safety of the electrical equipment itself will be affected, but also the life health of the animal or human will be damaged, and whether the image is a living thing or not may not be accurately judged by only depending on the image, and the biological information can be accurately judged by means of the infrared temperature information.
Further, in the step B, a BP neural network for predicting the grid load is constructed by using the environmental information, and the following neurons are constructed:
first neuronRepresents the lowest load of the previous day; second neuronThe highest temperature of the previous day is the highest temperature,is the lowest temperature of the previous day; third neuronIs the highest humidity of the previous day,the lowest humidity for the previous day; fourth neuron Typei-1Representing the weather type of the previous day; fifth neuronRepresenting the lowest load of the first two days; the sixth neuronThe highest temperature of the first two days,the lowest temperature of the first two days; the first neuronThe highest humidity of the first two days,the lowest humidity for the first two days; type of the eighth neuroni-2Representing the weather type of the first two days;
the mathematical expression of the BP neural network is:
where B is the noise data, G is the noise data, F is the coefficient,
defining a target error function
Computing
Ok=Ok-1+D
This is assigned to the desired output y.
As shown in fig. 3, which is a neuron table of date, weather, temperature, and humidity, fig. 4 shows the influence of each parameter, and the mathematical model constructed using the model of fig. 4 shows the constraints between variables, the lower vertices represent the states, and the upper vertices represent the observed values. The BP neural network can realize the nonlinear calculation of input and output, and has high self-adaption and fault-tolerant rate. And (4) calculating weights of neurons in different layers of the BP neural network by the evolved neural network according to a Genetic Algorithm (GA) principle. And (5) carrying out chromosome dimension coding. The first gene of the chromosomal sequence is w15, connecting neuron 1 with neuron 5 weight; the second gene, W16, is linked to neurons 1 and 6, and so on in a fully linked fashion. y represents what the load of the desired output is, O represents the value of the output to be estimated, k and k-1 represent different times, and if k is set to tomorrow, then k-1 is today. By calculating O to be equal to y, the expectation is calculated first because there will be a difference between the load actually output and the expected load. Thereby predicting the load value on the day. The influence of the date is attributed to noise, i.e. G in the above formula, being a given value. And performing effective dispatching on the upstream of the power grid to enable the power grid to be intelligent.
Further, the GA optimizes BP as follows:
step M1, calculating an adaptive function, and calculating the weight and the threshold of the population group by using the BP neural network, so that the individual fitness:
wherein m is the number of neurons in the output layer
Desired output y, predicted output o;
step M2, the selection operation,
selecting excellent individuals to be inherited to the next generation according to the individual fitness, and selecting a wheel disc calculation method;
fi=S/Fi
wherein F is the fitness of the individual, F is the fitness function of the individual, and S is a selection coefficient; p is the individual selection probability;
step M3, the cross-over operation,
the GA algorithm adopts a real number intersection method:
axi=axi(1-b)+ayib
ayi=ayi(1-b)+axib
axiis the ith position of the x chromosome, ayiPosition i of the y chromosome;
in step M4, the mutation operation,
after selecting an individual, the probability of some genes is converted into other alleles:
amaxis a of a geneijUpper bound, aminIs a of a geneijUpper bound, r2Is a random number, G is the number of iterations, GmaxMaximum number of iterations, r random number.
The estimated value is not always perfect, sometimes abnormal values occur, the abnormal values are not eliminated, and the system output is unstable. The load output value with larger error is eliminated, and the residual load output value is more accurate.
More specifically, the method includes classifying the image information by a cascade convolution calculation:
step D1, inputting a picture with a resolution of 100 × 100, performing 1 × 1 convolution on the first convolution layer to obtain a 100 × 100 picture, and performing processing by using a relu nonlinear activation function; convolution with 5 × 5 is used in the second convolution layer, so that a 96 × 96 picture is obtained and processed by using a relu nonlinear activation function;
step D2, performing a first layer of downsampling, using 2 × 2 downsampling to obtain 48 × 48 pictures;
step D3, using convolution of 1 × 1 on the third convolution layer to obtain 48 × 48 pictures, and processing them by using relu nonlinear activation function; using 5 × 5 convolution on the fourth convolution layer to obtain 44 × 44 pictures, and processing the pictures by using a relu nonlinear activation function;
step D4, performing a second layer of downsampling, using 2 × 2 downsampling to obtain 22 × 22 pictures;
step D5, fully connecting 1024 neurons in the first fully-connected layer, and processing by using a relu nonlinear activation function;
and D6, fully connecting 2 neurons in the second fully-connected layer to obtain an output layer.
The process is a visual processing process, can perform visual processing on the acquired video information, and specifically comprises the following steps: the detection equipment can provide templates for subsequent image processing, the camera detects surrounding invaded animals and people, the closing condition of the disconnecting link and the safety of workers, the online working time is counted to ensure safety, the weather condition of the surrounding place of the camera is detected, and scenes such as sunny days, rainy days and cloudy days are judged.
An AI-aided decision-based power communication management system, comprising:
the acquisition module is used for acquiring image information, line information and environment information;
the transmission module is used for transmitting the information acquired by the acquisition module to the decision module;
and the decision-making module is used for providing operation safety and surrounding safety warning according to the image information acquired by the acquisition module and providing power utilization scheduling analysis according to the line information and the environment information acquired by the acquisition module.
To illustrate, the acquisition module includes:
the video detection device is used for acquiring video images;
and the infrared detection device is used for collecting biological infrared information.
To illustrate, the acquisition module includes:
date detection means for recording a date;
the temperature detection device is used for acquiring temperature information;
and the humidity detection device is used for acquiring humidity information.
The above description is only a preferred embodiment of the present invention, and for those skilled in the art, the present invention should not be limited by the description of the present invention, which should be interpreted as a limitation.
Claims (8)
1. An AI-aided decision-making-based power communication management method is characterized by comprising the following steps of:
step A, collecting image information and extracting features, analyzing the safety and peripheral safety conditions of equipment through the features, sending an alarm if a safety problem exists, and simultaneously sending the place and time with the safety problem;
b, collecting environment information, comparing the change conditions of the current environment information and the previous environment information, if the change of the environment information is positively correlated with the change of the power consumption, correspondingly increasing or decreasing the power dispatching according to the increase or decrease of the environment information, and if the change of the environment information is negatively correlated with the change of the power consumption, correspondingly decreasing or increasing the power dispatching according to the increase or decrease of the environment information;
and C, acquiring line load data, increasing the power dispatching of the line node if the line load is increased, and reducing the power dispatching of the line node if the line load is reduced.
2. The AI-assisted decision-based power communication management method of claim 1, wherein: and step A, detecting biological information, comprehensively judging whether organisms exist nearby the power equipment or not through the infrared detected temperature information and the image information acquired by the image, and sending an alarm to remind if the organisms exist nearby the power equipment.
3. The AI-assisted decision-based power communication management method of claim 1, wherein: in the step B, a BP neural network for predicting the load of the power grid is constructed by utilizing the environmental information, and the following neurons are constructed:
first neuronRepresents the lowest load of the previous day; second neuron The highest temperature of the previous day is the highest temperature,is the lowest temperature of the previous day; third neuron Is the highest humidity of the previous day,the lowest humidity for the previous day; fourth neuron Typei-1Representing the weather type of the previous day; fifth neuronRepresenting the lowest load of the first two days; the sixth neuron The highest temperature of the first two days,the lowest temperature of the first two days; the first neuron The highest humidity of the first two days,the lowest humidity for the first two days; type of the eighth neuroni-2Representing the weather type of the first two days;
the mathematical expression of the BP neural network is:
where B is the noise data, G is the noise data, F is the coefficient,
defining a target error function
Computing
Ok=Ok-1+D
This is assigned to the desired output y.
4. The AI-assisted decision-based power communication management method of claim 3, wherein the GA optimization BP comprises the following steps:
step M1, calculating an adaptation function, calculating the weight and the threshold of the population by using the BP neural network,
then the individual fitness is:
wherein m is the number of neurons in the output layer
Desired output y, predicted output o;
step M2, the selection operation,
selecting excellent individuals to be inherited to the next generation according to the individual fitness, and selecting a wheel disc calculation method;
fi=S/Fi
wherein F is the fitness of the individual, F is the fitness function of the individual, and S is a selection coefficient; p is the individual selection probability;
step M3, the cross-over operation,
the GA algorithm adopts a real number intersection method:
axi=axi(1-b)+ayib
ayi=ayi(1-b)+axib
axiis the ith position of the x chromosome, ayiPosition i of the y chromosome;
in step M4, the mutation operation,
after selecting an individual, the probability of some genes is converted into other alleles:
amaxis a of a geneijUpper bound, aminIs a of a geneijUpper bound, r2Is a random number, G is the number of iterations, GmaxMaximum number of iterations, r random number.
5. The AI-assisted decision-based power communication management method of claim 2, further comprising classifying image information by cascaded convolution calculations:
step D1, inputting a picture with a resolution of 100 × 100, performing 1 × 1 convolution on the first convolution layer to obtain a 100 × 100 picture, and performing processing by using a relu nonlinear activation function; convolution with 5 × 5 is used in the second convolution layer, so that a 96 × 96 picture is obtained and processed by using a relu nonlinear activation function;
step D2, performing a first layer of downsampling, using 2 × 2 downsampling to obtain 48 × 48 pictures;
step D3, using convolution of 1 × 1 on the third convolution layer to obtain 48 × 48 pictures, and processing them by using relu nonlinear activation function; using 5 × 5 convolution on the fourth convolution layer to obtain 44 × 44 pictures, and processing the pictures by using a relu nonlinear activation function;
step D4, performing a second layer of downsampling, using 2 × 2 downsampling to obtain 22 × 22 pictures;
step D5, fully connecting 1024 neurons in the first fully-connected layer, and processing by using a relu nonlinear activation function;
and D6, fully connecting 2 neurons in the second fully-connected layer to obtain an output layer.
6. An AI-aided decision-making based power communication management system, comprising:
the acquisition module is used for acquiring image information, line information and environment information;
the transmission module is used for transmitting the information acquired by the acquisition module to the decision module;
and the decision-making module is used for providing operation safety and surrounding safety warning according to the image information acquired by the acquisition module and providing power utilization scheduling analysis according to the line information and the environment information acquired by the acquisition module.
7. The AI-assisted decision-based power communication management system of claim 6, wherein the collection module comprises:
the video detection device is used for acquiring video images;
and the infrared detection device is used for collecting biological infrared information.
8. The AI-assisted decision-based power communication management system of claim 6, wherein the collection module comprises:
date detection means for recording a date;
the temperature detection device is used for acquiring temperature information;
and the humidity detection device is used for acquiring humidity information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110594898.XA CN113224852A (en) | 2021-05-28 | 2021-05-28 | Power communication management method and system based on AI auxiliary decision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110594898.XA CN113224852A (en) | 2021-05-28 | 2021-05-28 | Power communication management method and system based on AI auxiliary decision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113224852A true CN113224852A (en) | 2021-08-06 |
Family
ID=77099519
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110594898.XA Pending CN113224852A (en) | 2021-05-28 | 2021-05-28 | Power communication management method and system based on AI auxiliary decision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113224852A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103699873A (en) * | 2013-09-22 | 2014-04-02 | 杭州电子科技大学 | Lower-limb flat ground walking gait recognition method based on GA-BP (Genetic Algorithm-Back Propagation) neural network |
CN105303247A (en) * | 2015-09-16 | 2016-02-03 | 北京国电通网络技术有限公司 | Garden type hot and cold energy mixed application energy network regulation method and system |
CN108537277A (en) * | 2018-04-10 | 2018-09-14 | 湖北工业大学 | A kind of image classification knowledge method for distinguishing |
CN108777049A (en) * | 2018-04-27 | 2018-11-09 | 国网浙江桐庐县供电有限公司 | A kind of animal alarm system for substation |
CN109524968A (en) * | 2018-12-17 | 2019-03-26 | 广西电网有限责任公司电力科学研究院 | A kind of public building load participates in the control system of peak load regulation network |
CN110580588A (en) * | 2019-09-19 | 2019-12-17 | 国网上海市电力公司 | power grid dispatching system and power grid dispatching method based on ubiquitous Internet of things |
CN110601193A (en) * | 2019-10-09 | 2019-12-20 | 中电科新型智慧城市研究院有限公司 | Power load prediction system and method based on environment variable perception and hysteresis |
-
2021
- 2021-05-28 CN CN202110594898.XA patent/CN113224852A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103699873A (en) * | 2013-09-22 | 2014-04-02 | 杭州电子科技大学 | Lower-limb flat ground walking gait recognition method based on GA-BP (Genetic Algorithm-Back Propagation) neural network |
CN105303247A (en) * | 2015-09-16 | 2016-02-03 | 北京国电通网络技术有限公司 | Garden type hot and cold energy mixed application energy network regulation method and system |
CN108537277A (en) * | 2018-04-10 | 2018-09-14 | 湖北工业大学 | A kind of image classification knowledge method for distinguishing |
CN108777049A (en) * | 2018-04-27 | 2018-11-09 | 国网浙江桐庐县供电有限公司 | A kind of animal alarm system for substation |
CN109524968A (en) * | 2018-12-17 | 2019-03-26 | 广西电网有限责任公司电力科学研究院 | A kind of public building load participates in the control system of peak load regulation network |
CN110580588A (en) * | 2019-09-19 | 2019-12-17 | 国网上海市电力公司 | power grid dispatching system and power grid dispatching method based on ubiquitous Internet of things |
CN110601193A (en) * | 2019-10-09 | 2019-12-20 | 中电科新型智慧城市研究院有限公司 | Power load prediction system and method based on environment variable perception and hysteresis |
Non-Patent Citations (1)
Title |
---|
王天安等: "基于智能优化的电力负荷预测方法研究", 《电子设计工程》, vol. 29, no. 2, 31 January 2021 (2021-01-31), pages 129 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Serrano-Guerrero et al. | A new interval prediction methodology for short-term electric load forecasting based on pattern recognition | |
Kolhe et al. | GA-ANN for short-term wind energy prediction | |
CN110807550A (en) | Distribution transformer overload identification early warning method based on neural network and terminal equipment | |
CN111160626B (en) | Power load time sequence control method based on decomposition fusion | |
CN115796434A (en) | Management and control method and device for power distribution network, electronic equipment and storage medium | |
CN115730749B (en) | Power dispatching risk early warning method and device based on fusion power data | |
CN111461921A (en) | Load modeling typical user database updating method based on machine learning | |
Xie et al. | Short-term power load forecasting model based on fuzzy neural network using improved decision tree | |
Ningrum et al. | Fuzzy mamdani logic inference model in the loading of distribution substation transformer SCADA system | |
Cai et al. | Exploratory analysis of massive data for distribution fault diagnosis in smart grids | |
CN117239713A (en) | Intelligent security management and control method and system based on power distribution network dispatching | |
CN117614137A (en) | Power distribution network optimization system based on multi-source data fusion | |
CN117200352A (en) | Photovoltaic power generation regulation and control method and system based on cloud edge fusion | |
CN113224852A (en) | Power communication management method and system based on AI auxiliary decision | |
CN111861587A (en) | System and method for analyzing residential electricity consumption behavior based on hidden Markov model and forward algorithm | |
CN113723670B (en) | Photovoltaic power generation power short-term prediction method with variable time window | |
CN114936808A (en) | Cloud-edge cooperative task management system and method for substation fault detection | |
CN112821456B (en) | Distributed source-storage-load matching method and device based on transfer learning | |
CN112862249B (en) | Lean management method and system for intelligent power distribution equipment | |
CN111857015A (en) | Power transmission and transformation cloud intelligent controller | |
Dutta et al. | Machine Learning‐Based Intelligent Power Systems | |
CN117113267B (en) | Prediction model training method based on big data and photovoltaic power generation performance detection method | |
CN111897655B (en) | Power transmission and transformation high in clouds intelligent control ware | |
CN117728583B (en) | Distributed photovoltaic cluster energy control monitoring system based on transfer learning | |
CN117113157B (en) | Platform district power consumption fault detection system based on artificial intelligence |
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: 20210806 |
|
RJ01 | Rejection of invention patent application after publication |