CN115473595B - Prediction method for packet loss rate of wireless sensing equipment in transformer substation - Google Patents
Prediction method for packet loss rate of wireless sensing equipment in transformer substation Download PDFInfo
- Publication number
- CN115473595B CN115473595B CN202211001228.3A CN202211001228A CN115473595B CN 115473595 B CN115473595 B CN 115473595B CN 202211001228 A CN202211001228 A CN 202211001228A CN 115473595 B CN115473595 B CN 115473595B
- Authority
- CN
- China
- Prior art keywords
- wireless sensing
- sensing equipment
- packet loss
- loss rate
- base station
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013528 artificial neural network Methods 0.000 claims abstract description 49
- 238000012549 training Methods 0.000 claims abstract description 27
- 230000005540 biological transmission Effects 0.000 claims abstract description 24
- 230000002068 genetic effect Effects 0.000 claims abstract description 15
- 230000005684 electric field Effects 0.000 claims abstract description 12
- 210000000349 chromosome Anatomy 0.000 claims description 43
- 230000001537 neural effect Effects 0.000 claims description 12
- 108090000623 proteins and genes Proteins 0.000 claims description 12
- 230000035772 mutation Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 4
- 210000000225 synapse Anatomy 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 6
- 238000005259 measurement Methods 0.000 abstract description 3
- 238000012216 screening Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
- H04B17/318—Received signal strength
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/373—Predicting channel quality or other radio frequency [RF] parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Electromagnetism (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
A prediction method of packet loss rate of wireless sensing equipment in a transformer substation is characterized in that the transmission frequency of a 5G base station antenna of the transformer substation, the transmission power of the 5G base station antenna, the distance between the wireless sensing equipment and the 5G base station antenna, the electric field intensity at the wireless sensing equipment and the data transmission rate of the wireless sensing equipment are used as input parameters of a BP neural network, and the packet loss rate of the wireless sensing equipment is used as output parameters. And continuously screening the threshold value and the weight value during BP neural network training by utilizing a genetic algorithm, so as to obtain the optimal prediction of the packet loss rate of the wireless sensing equipment. The method realizes the real-time monitoring of the interference degree of the wireless sensing equipment; and the complex flow in the packet loss rate measurement process is avoided, and the method is more suitable for actual application of workers in the station.
Description
Technical Field
The invention relates to the technical field of monitoring of wireless sensing equipment in a transformer substation, in particular to a prediction method of packet loss rate of the wireless sensing equipment in the transformer substation.
Background
In recent years, with the continuous innovation of power grid technology, the demand for monitoring data transmission by wireless sensing devices in power infrastructures such as substations is increasing, and in order to perform faster, safer and more stable real-time online monitoring on various parameters of operation of the internal devices of the substations, 5G communication technology is gradually popularized in the power infrastructures such as the substations. But simultaneously, the radiation beam of the 5G base station antenna is more concentrated and the radiation power is stronger than that of the traditional base station antenna, so that the original complex and severe electromagnetic environment of the transformer substation is frosted, and the electromagnetic compatibility problem faced by the wireless sensing equipment sensitive in the transformer substation is clearly more outstanding.
Therefore, the packet loss phenomenon of the wireless sensing device may become more serious due to the introduction of the 5G base station antenna while the data transmission speed of the wireless sensing device is increased. Considering that the existing packet loss rate in engineering is obtained through passive measurement, namely, the equipment transmits a series of test data to the terminal, the packet loss rate is measured by utilizing a software program according to the total data received by the terminal equipment, the process is very troublesome, and the actual operation of transformer substation staff is inconvenient. Therefore, a new method is needed to be found, and the packet loss rate of the wireless sensing equipment interference factor in the substation can be directly predicted according to the monitoring index of the wireless sensing equipment interference factor.
Disclosure of Invention
The invention provides a method for predicting the packet loss rate of wireless sensing equipment in a transformer substation, which can achieve the purpose of predicting the packet loss rate of the wireless sensing equipment in the transformer substation by measuring the transmitting frequency of a 5G base station antenna in the transformer substation, the electric field intensity at the wireless sensing equipment, the distance between a 5G base station signal at the wireless sensing equipment and a 5G base station and the data transmission rate of the wireless sensing equipment. Thereby realizing the real-time monitoring of the interference degree of the wireless sensing equipment; and the complex flow in the packet loss rate measurement process is avoided, and the method is more suitable for actual application of workers in the station.
The technical scheme adopted by the invention is as follows:
a prediction method for packet loss rate of wireless sensing equipment in a transformer substation comprises the following steps:
step one: collecting signal frequency emitted by 5G base station antennas in N groups of substations, electric field size at wireless sensing equipment, 5G base station signal strength at the wireless sensing equipment, distance between the wireless sensing equipment and the 5G base station antennas, data transmission rate of the wireless sensing equipment and packet loss rate of the wireless sensing equipment at the moment as samples;
step two: all N groups of samples acquired in the first step are normalized, the data transmission rate of the wireless sensing equipment, the signal frequency emitted by the 5G base station antenna, the 5G base station signal at the wireless sensing equipment, the electric field intensity at the wireless sensing equipment and the distance between the wireless sensing equipment and the 5G antenna are taken as input parameters, and the packet loss rate of the wireless sensing equipment at the moment is taken as an output parameter to determine the topology structure of the BP neural network;
step three: setting an initial weight omega and a threshold v of each synapse of the BP neural network, inputting the input parameters normalized in the second step into the BP neural network for learning training, and outputting a prediction result after the maximum learning times X are reached; then comparing with the actually measured packet loss rate, and stopping training if the error range is met; if not, entering a step four;
step four: if the training requirement is not met, taking all predicted packet loss rates of the BP neural network as individuals, sending weights and thresholds after the training of each individual to a genetic algorithm, and carrying out real number coding on the weights and thresholds of the BP neural network corresponding to the individuals by using the genetic algorithm to obtain a weight chromosome library inherited by the individuals: omega (1) =[ω (1) 1 ,ω (1) 2 ,…,ω (1) m ]、ω (2) =[ω (2) 1 ,ω (2) 2 ,…,ω (2) m ]、…、ω (N) =[ω (N) 1 ,ω (N) 2 ,…,ω (N) m ];ω (1) 、ω (2) 、…、ω (N) N weight matrixes obtained by the neural network after training by the genetic algorithm are respectively represented; omega (N) 1 ,ω (N) 2 ,…,ω (N) m Representing m weights in the retrieved Nth weight matrix;
chromosome library for threshold values: v (V) (1) =[v (1) 1 ,v (1) 2 ,…,v (1) m ]、V (2) =[v (2) 1 ,v (2) 2 ,…,v (2) m ]、…、V (N) =[v (N) 1 ,v (N) 2 ,…,v (N) m ]。V (1) 、V (2) 、…、V (N) Respectively representing N threshold matrixes obtained by the neural network after training by a genetic algorithm; v (N) 1 ,v (N) 2 ,…,v (N) m Representing m thresholds within the retrieved nth threshold matrix;
step five: using the formulaN represents the total number of N individuals, namely the number of wireless sensing devices needing to predict the packet loss rate;
calculating the fitness of the ith individual, f i For the fitness of the ith individual, the smaller the fitness is, the better the prediction result is; d (D) i Predicted packet loss rate for the ith individual, D i ' is the actual packet loss rate of the ith individual.
Step six: randomly selecting two chromosomes from the weight value and threshold chromosome library in the fourth step, wherein: the probability of selection of each chromosome isThe selected chromosomes generate two new weight and threshold chromosomes through crossing and mutation;
step seven: generating N new weight and threshold chromosomes, transmitting the weight and the genes on the threshold chromosomes to the BP neural network, and re-executing the operation of the third step until the optimal BP neural network meeting the training error requirement and the weight and the threshold thereof are obtained.
In the first step, data measured by the wireless sensing device are transmitted to the terminal computer at the same time interval Δt, and the packet loss rate is read for N times by the packet loss rate measuring software installed by the terminal computer, and the data transmission rate of each wireless sensing device, the signal frequency transmitted by the 5G base station antenna, the 5G base station signal at the wireless sensing device, the electric field intensity at the wireless sensing device, and the distance from the wireless sensing device to the 5G antenna are recorded.
In the second step, the collected N groups of sample data pass through the formulaNormalizing;
wherein: x' is the normalized value of the parameter of a certain index, x is the value of the parameter of the index, and the index x max Maximum value of parameter value, the index x min Minimum value of parameter values.
In the second step, the method for determining the topology structure of the BP neural network comprises the following steps:
taking the normalized signal frequency transmitted by the 5G base station antenna, the 5G base station signal at the wireless sensing device, the 5G base station signal intensity at the wireless sensing device, the distance between the wireless sensing device and the 5G base station antenna and the data transmission rate of the wireless sensing device as input parameters, and taking the normalized packet loss rate of the wireless sensing device as output parameters;
according to the input signal frequency of a 5G base station antenna in the parametric transformer substation, the signal intensity of the 5G base station at the wireless sensing equipment, the distance between the wireless sensing equipment and the 5G base station antenna and the data transmission rate at the moment, the used BP neural network input layer neural node can be determined to be 5; according to the packet loss rate of the output parameter wireless sensing equipment, the neural node of the output layer of the used BP neural network can be determined to be 1.
The method comprises the following steps:
determining the number N of neurons of the hidden layer h 8;
wherein: n (N) i Representing the number of neural nodes of an input layer of the BP neural network; n (N) o The number of the neural nodes of the output layer of the BP neural network; n (N) s Is the number of samples of the training set; alpha is in the range of [2,10]Arbitrary value between, here taken as 25/12.
The sixth step further comprises:
the crossover operations of the chromosomes are as follows:
wherein: A. b is a weight value and a threshold chromosome selected from the step six, A 'and B' are genes after crossing operation, and lambda is a random number between (0 and 1).
The chromosome mutation operation after crossing is as follows:
in the formula (3), r 1 ,r 2 Are all arbitrary values generated between (0, 1) at each iteration, a i The weight value selected for step six, the i gene on the threshold chromosome, i.e., the i weight value, the threshold value, a min A is the minimum weight value, threshold value, a on the chromosome max And g is the current iteration number, and K is the maximum iteration number, wherein g is the maximum weight and threshold value on the chromosome.
The invention discloses a prediction method for packet loss rate of wireless sensing equipment in a transformer substation, which has the following technical effects:
1): according to the invention, a prediction model of the packet loss rate of the wireless sensing equipment of the transformer substation is established, and the weight and the threshold value in the BP neural network are optimized by a genetic algorithm, so that the optimal prediction of the packet loss rate of the wireless sensing equipment is achieved.
2): according to the invention, the packet loss rate of the wireless sensing equipment is calculated in an algorithm prediction mode, so that a complex flow of measuring the packet loss rate by a transformer station worker is avoided, and the actual engineering application of the worker in the station and the interference degree of the wireless sensing equipment are more convenient to evaluate.
Drawings
Fig. 1 is a topology diagram of a BP neural network of the prediction method of the present invention.
FIG. 2 is a schematic diagram of a genetic algorithm optimized BP neural network flow.
Detailed Description
A prediction method of packet loss rate of wireless sensing equipment in a transformer substation is characterized in that the transmission frequency of a 5G base station antenna of the transformer substation, the transmission power of the 5G base station antenna, the distance between the wireless sensing equipment and the 5G base station antenna, the electric field intensity at the wireless sensing equipment and the data transmission rate of the wireless sensing equipment are used as input parameters of a BP neural network, and the packet loss rate of the wireless sensing equipment is used as output parameters. And continuously screening the threshold value and the weight value during BP neural network training by utilizing a genetic algorithm, so as to obtain the optimal prediction of the packet loss rate of the wireless sensing equipment. The method comprises the following steps:
step one: and transmitting data measured by the wireless sensing equipment to the terminal computer at the same time interval delta t, reading the packet loss rate for N times through packet loss rate measurement software installed by the terminal computer, and recording the data transmission rate of each wireless sensing equipment, the signal frequency transmitted by the 5G base station antenna, the 5G base station signal at the wireless sensing equipment, the electric field intensity at the wireless sensing equipment and the distance between the wireless sensing equipment and the 5G antenna.
Wherein: the signal frequency emitted by the 5G base station antenna, the 5G base station signal at the target wireless sensing device is measured by adopting an ETS3117 receiving antenna and a 3075X-R spectrometer, the electric field intensity at the target wireless sensing device is measured by adopting a Narda NBM550 field intensity measuring instrument, and the distance between the target wireless sensing device and the 5G antenna is measured by adopting a laser range finder.
Step two: the acquired N groups of sample data are utilized by a formulaNormalizing;
wherein: x' is the normalized value of the parameter of a certain index, x is the value of the parameter of the index, and the index x max Maximum value of parameter value, the index x min Minimum value of parameter values.
Taking the normalized transmitting frequency of the 5G base station antenna, the 5G base station signal at the wireless sensing device, the electric field intensity at the target wireless sensing device, the distance from the 5G base station and the data transmission rate of the wireless sensing device as input parameters, and taking the normalized packet loss rate of the wireless sensing device as output parameters. According to the input transmission frequency and transmission power of the 5G base station antenna in the parametric transformer substation, the radiation field intensity of the 5G antenna at the target wireless sensing equipment, the distance from the 5G base station and the data transmission rate at the moment, the used BP neural network input layer neural node can be determined to be 5; according to the packet loss rate of the output parameter wireless sensing equipment, the neural node of the output layer of the used BP neural network can be determined to be 1. Can be defined by:
determining the number N of neurons of the hidden layer h Is 8, where N i Representing the number, N, of neural nodes of the BP neural network input layer o Is the number of the neural nodes of the BP neural network output layer and N s Is the number of samples of the training set. Alpha is in the range of [2,10]Arbitrary value between, here taken as 25/12. Therefore, the topology structure diagram of the last determined BP neural network is shown in FIG. 1.
Step three: giving the initial weight omega and threshold v to all synapses from the BP neural network input layer to the hidden layer and from the hidden layer to the output layer, inputting the normalized input parameters into the BP neural network for learning training, outputting a prediction result after the maximum learning times X are reached, comparing with the actually measured packet loss rate, stopping training if the error range is met, and entering the fourth step if the error range is not met.
Step four: regarding BP neural networks which do not meet the training requirement, regarding all predicted packet loss rates as individuals, transmitting weights and thresholds after the training of each individual to a genetic algorithm, and carrying out real number coding on the weights and thresholds of the BP neural networks corresponding to the individuals by using the genetic algorithm to obtain a weight chromosome library omega inherited by the individuals (1) =[ω (1) 1 ,ω (1) 2 ,…,ω (1) m ]、ω (2) =[ω (2) 1 ,ω (2) 2 ,…,ω (2) m ]、…、ω (N) =[ω (N) 1 ,ω (N) 2 ,…,ω (N) m ]And a threshold chromosome pool V (1) =[v (1) 1 ,v (1) 2 ,…,v (1) m ]、V (2) =[v (2) 1 ,v (2) 2 ,…,v (2) m ]、…、V (N) =[v (N) 1 ,v (N) 2 ,…,v (N) m ]。
Step five: using the formulaCalculating the fitness of the ith individual, f i For the fitness of the ith individual, a smaller fitness indicates a better prediction result, D i Predicted packet loss rate for the ith individual, D i ' is the actual packet loss rate of the ith individual.
Step six: randomly selecting two genes from the weight value and the threshold gene sequence in the third step, wherein the selection probability of each gene is that
Further, since the weights and threshold chromosomes in the neural network are both encoded by real numbers, chromosome crossover operation is performed on the selected chromosomes by a real number crossover method:
wherein: A. b is a weight value and a threshold chromosome selected from the step six, A 'and B' are genes after crossing operation, and lambda is a random number between (0 and 1).
Then, the crossed chromosomes were subjected to mutation as follows:
in the middle ofWherein r is 1 ,r 2 Are all of any value between (0, 1), a i The weight value selected for step six, the i gene on the threshold chromosome, i.e., the i weight value, the threshold value, a min A is the minimum weight value, threshold value, a on the chromosome max For the maximum weight and threshold on the chromosome, g is the current iteration number, and K is the maximum iteration number.
Two new weight and threshold chromosomes are generated after crossing and mutation.
Step seven: generating N new weight and threshold chromosomes by adopting the method of the step six, sending genes on the chromosomes to the BP neural network, and re-executing the operation of the step three until the optimal BP neural network meeting the training error requirement and the weight and threshold thereof are obtained.
Examples:
taking a certain transformer substation of 500kV as an example, comparing the actual packet loss rate under each parameter with the predicted packet loss rate calculated by the method, wherein the training iteration number is set to be 1000, the number of hidden layer nodes of the BP neural network is set to be 8, 85% of sample data are selected as a training set, and 15% of data are selected as a test set. After training, the error between the wireless sensing device packet loss rate and the actual packet loss rate predicted by the BP neural network based on the genetic algorithm improvement is shown in table 1:
table 1 actual packet loss rate and predicted packet loss rate
It can be seen that the errors between the predicted values and the actual values are controlled within the error range of 5%, and the predicted requirements can be considered to be satisfied.
Claims (5)
1. A prediction method for packet loss rate of wireless sensing equipment in a transformer substation is characterized by comprising the following steps:
step one: collecting signal frequency emitted by 5G base station antennas in N groups of substations, electric field size at wireless sensing equipment, 5G base station signal strength at the wireless sensing equipment, distance between the wireless sensing equipment and the 5G base station antennas, data transmission rate of the wireless sensing equipment and packet loss rate of the wireless sensing equipment at the moment as samples;
step two: all N groups of samples acquired in the first step are normalized, the data transmission rate of the wireless sensing equipment, the signal frequency emitted by the 5G base station antenna, the 5G base station signal at the wireless sensing equipment, the electric field intensity at the wireless sensing equipment and the distance between the wireless sensing equipment and the 5G antenna are taken as input parameters, and the packet loss rate of the wireless sensing equipment at the moment is taken as an output parameter to determine the topology structure of the BP neural network;
step three: setting an initial weight omega and a threshold v of each synapse of the BP neural network, inputting the input parameters normalized in the second step into the BP neural network for learning training, and outputting a prediction result after the maximum learning times X are reached; then comparing with the actually measured packet loss rate, and stopping training if the error range is met; if not, entering a step four;
step four: if the training requirement is not met, taking all predicted packet loss rates of the BP neural network as individuals, sending weights and thresholds after the training of each individual to a genetic algorithm, and carrying out real number coding on the weights and thresholds of the BP neural network corresponding to the individuals by using the genetic algorithm to obtain a weight chromosome library inherited by the individuals:
ω (1) =[ω (1) 1 ,ω (1) 2 ,…,ω (1) m ]、ω (2) =[ω (2) 1 ,ω (2) 2 ,…,ω (2) m ]、…、ω (N) =[ω (N) 1 ,ω (N) 2 ,…,ω (N) m ];
ω (1) 、ω (2) 、…、ω (N) n weight matrixes obtained by the neural network after training by the genetic algorithm are respectively represented;
ω (N) 1 ,ω (N) 2 ,…,ω (N) m representing m weights in the retrieved Nth weight matrix;
chromosome library for threshold values:
V (1) =[v (1) 1 ,v (1) 2 ,…,v (1) m ]、V (2) =[v (2) 1 ,v (2) 2 ,…,v (2) m ]、…、V (N) =[v (N) 1 ,v (N) 2 ,…,v (N) m ];
V (1) 、V (2) 、…、V (N) respectively representing N threshold matrixes obtained by the neural network after training by a genetic algorithm;
v (N) 1 ,v (N) 2 ,…,v (N) m representing m thresholds within the retrieved nth threshold matrix;
step five: using the formulaN represents the number of wireless sensing devices needing to predict the packet loss rate;
calculating the fitness of the ith individual, f i For the fitness of the ith individual, the smaller the fitness is, the better the prediction result is; d (D) i Predicted packet loss rate for the ith individual, D i ' is the actual packet loss rate of the ith individual;
step six: randomly selecting two chromosomes from the weight value and threshold chromosome library in the fourth step, wherein: the probability of selection of each chromosome isThe selected chromosomes generate two new weight and threshold chromosomes through crossing and mutation;
step seven: generating N new weight and threshold chromosomes, transmitting the weight and the genes on the threshold chromosomes to the BP neural network, and re-executing the operation of the third step until the optimal BP neural network meeting the training error requirement and the weight and the threshold thereof are obtained.
2. The method for predicting the packet loss rate of wireless sensing equipment in a transformer substation according to claim 1, wherein the method comprises the following steps: in the first step, data measured by the wireless sensing device are transmitted to the terminal computer at the same time interval Δt, and the packet loss rate is read for N times by the packet loss rate measuring software installed by the terminal computer, and the data transmission rate of each wireless sensing device, the signal frequency transmitted by the 5G base station antenna, the 5G base station signal at the wireless sensing device, the electric field intensity at the wireless sensing device, and the distance from the wireless sensing device to the 5G antenna are recorded.
3. The method for predicting the packet loss rate of wireless sensing equipment in a transformer substation according to claim 1, wherein the method comprises the following steps: in the second step, the collected N groups of sample data pass through the formulaNormalizing;
wherein: x' is the normalized value of the parameter of a certain index, x is the value of the parameter of the index, and the index x max Maximum value of parameter value, the index x min Minimum value of parameter values.
4. The method for predicting the packet loss rate of wireless sensing equipment in a transformer substation according to claim 1, wherein the method comprises the following steps: in the second step, the method for determining the topology structure of the BP neural network comprises the following steps:
taking the normalized signal frequency transmitted by the 5G base station antenna, the 5G base station signal at the wireless sensing device, the 5G base station signal intensity at the wireless sensing device, the distance between the wireless sensing device and the 5G base station antenna and the data transmission rate of the wireless sensing device as input parameters, and taking the normalized packet loss rate of the wireless sensing device as output parameters;
according to the input signal frequency of a 5G base station antenna in the parametric transformer substation, the signal intensity of the 5G base station at the wireless sensing equipment, the distance between the wireless sensing equipment and the 5G base station antenna and the data transmission rate at the moment, the used BP neural network input layer neural node can be determined to be 5; according to the packet loss rate of the output parameter wireless sensing equipment, the neural node of the output layer of the used BP neural network can be determined to be 1;
the method comprises the following steps:
wherein: n (N) i Representing the number of neural nodes of an input layer of the BP neural network; n (N) o The number of the neural nodes of the output layer of the BP neural network; n (N) s Is the number of samples of the training set; alpha is in the range of [2,10]Arbitrary values in between.
5. The method for predicting the packet loss rate of wireless sensing equipment in a transformer substation according to claim 1, wherein the method comprises the following steps:
the sixth step further comprises:
the crossover operations of the chromosomes are as follows:
wherein: A. b is a weight value and a threshold chromosome selected from the step six, A 'and B' are genes after crossing operation, and lambda is a random number between (0 and 1);
the chromosome mutation operation after crossing is as follows:
in the formula (3), r 1 ,r 2 Are all arbitrary values generated between (0, 1) at each iteration, a i The weight value selected for step six, the i gene on the threshold chromosome, i.e., the i weight value, the threshold value, a min A is the minimum weight value, threshold value, a on the chromosome max And g is the current iteration number, and K is the maximum iteration number, wherein g is the maximum weight and threshold value on the chromosome.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211001228.3A CN115473595B (en) | 2022-08-19 | 2022-08-19 | Prediction method for packet loss rate of wireless sensing equipment in transformer substation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211001228.3A CN115473595B (en) | 2022-08-19 | 2022-08-19 | Prediction method for packet loss rate of wireless sensing equipment in transformer substation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115473595A CN115473595A (en) | 2022-12-13 |
CN115473595B true CN115473595B (en) | 2024-04-02 |
Family
ID=84366206
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211001228.3A Active CN115473595B (en) | 2022-08-19 | 2022-08-19 | Prediction method for packet loss rate of wireless sensing equipment in transformer substation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115473595B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014056363A1 (en) * | 2012-10-09 | 2014-04-17 | 中兴通讯股份有限公司 | Uplink power control method and device based on genetic algorithm in communication network |
CN105913150A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | BP neural network photovoltaic power station generating capacity prediction method based on genetic algorithm |
CN105913151A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | Photovoltaic power station power generation amount predication method based on adaptive mutation particle swarm and BP network |
CN107622319A (en) * | 2016-07-14 | 2018-01-23 | 上海电气分布式能源科技有限公司 | A kind of microgrid power Forecasting Methodology |
CN111147163A (en) * | 2019-12-17 | 2020-05-12 | 南京航空航天大学 | Wireless communication link loss prediction method based on DNN neural network |
WO2021026944A1 (en) * | 2019-08-09 | 2021-02-18 | 东北大学 | Adaptive transmission method for industrial wireless streaming media employing particle swarm and neural network |
-
2022
- 2022-08-19 CN CN202211001228.3A patent/CN115473595B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014056363A1 (en) * | 2012-10-09 | 2014-04-17 | 中兴通讯股份有限公司 | Uplink power control method and device based on genetic algorithm in communication network |
CN105913150A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | BP neural network photovoltaic power station generating capacity prediction method based on genetic algorithm |
CN105913151A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | Photovoltaic power station power generation amount predication method based on adaptive mutation particle swarm and BP network |
CN107622319A (en) * | 2016-07-14 | 2018-01-23 | 上海电气分布式能源科技有限公司 | A kind of microgrid power Forecasting Methodology |
WO2021026944A1 (en) * | 2019-08-09 | 2021-02-18 | 东北大学 | Adaptive transmission method for industrial wireless streaming media employing particle swarm and neural network |
CN111147163A (en) * | 2019-12-17 | 2020-05-12 | 南京航空航天大学 | Wireless communication link loss prediction method based on DNN neural network |
Non-Patent Citations (3)
Title |
---|
Solution and analysis of the electromagnetic scattering field of the substation under the radiation of the 5G base station antenna;Can Zhang等;《2022 IEEE 5th International Electrical and Energy Conference 》;20220811;全文 * |
基于BP神经网络的IGBT模块开关损耗求解;唐波等;《高压电器》;20190716;第55卷(第7期);28-31 * |
基于遗传算法优化BP神经网络的GIS设备放电故障诊断;周正思;刘林;程鹏;;电气开关;20160615(第03期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN115473595A (en) | 2022-12-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110262233B (en) | Optimization method for technological parameters of magnetic control film plating instrument | |
CN110717302B (en) | Edge computing terminal equipment layout method for real-time online monitoring service of power grid | |
CN110276475B (en) | Optimal configuration method for automatic assembly production of circuit breaker | |
CN112686429B (en) | Mobile robot and path planning method thereof based on adaptive genetic algorithm | |
CN110765700A (en) | Ultrahigh voltage transmission line loss prediction method based on quantum ant colony optimization RBF network | |
CN109829497B (en) | Supervised learning-based station area user identification and discrimination method | |
CA2436352A1 (en) | Process and system for developing a predictive model | |
CN103105246A (en) | Greenhouse environment forecasting feedback method of back propagation (BP) neural network based on improvement of genetic algorithm | |
CN103577679A (en) | Real-time computing method for theoretical line loss of low-voltage distribution room | |
CN104270281A (en) | Web Service QoS (Quality of Service) combined forecasting method | |
CN111598231A (en) | Cable connector wire temperature prediction based on GA (genetic algorithm) optimized serial BP (back propagation) network | |
CN113591299B (en) | Method and system for judging comprehensive performance of different types of micro-channel radiators | |
CN110111606A (en) | A kind of vessel traffic flow prediction technique based on EEMD-IAGA-BP neural network | |
CN112182961A (en) | Large-scale fading modeling prediction method for wireless network channel of converter station | |
Saber et al. | IoT based online load forecasting | |
CN111988786A (en) | Sensor network covering method and system based on high-dimensional multi-target decomposition algorithm | |
CN112084709A (en) | Large-scale generator insulation state evaluation method based on genetic algorithm and radial basis function neural network | |
CN114626769A (en) | Operation and maintenance method and system for capacitor voltage transformer | |
CN111612227A (en) | Load prediction method based on K-means clustering and bat optimization neural network | |
CN115473595B (en) | Prediction method for packet loss rate of wireless sensing equipment in transformer substation | |
CN114066250A (en) | Method, device, equipment and storage medium for measuring and calculating repair cost of power transmission project | |
CN112365074B (en) | Artificial intelligence decision-making method based on power grid regulation and control data | |
CN115526393B (en) | Manufacturing cost prediction method based on screening of key influencing factors of power transformation engineering | |
CN114004008A (en) | Airplane assembly line resource configuration optimization method based on neural network and genetic algorithm | |
CN115207921B (en) | Position optimization method and system for photovoltaic access power distribution network |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |