CN115619078B - Method and device for predicting hazard risk level of small animals in transformer substation - Google Patents

Method and device for predicting hazard risk level of small animals in transformer substation Download PDF

Info

Publication number
CN115619078B
CN115619078B CN202211311275.8A CN202211311275A CN115619078B CN 115619078 B CN115619078 B CN 115619078B CN 202211311275 A CN202211311275 A CN 202211311275A CN 115619078 B CN115619078 B CN 115619078B
Authority
CN
China
Prior art keywords
information
small animal
activity
equipment
prediction model
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
Application number
CN202211311275.8A
Other languages
Chinese (zh)
Other versions
CN115619078A (en
Inventor
鲁仁全
谢家豪
汪香念
饶红霞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong University of Technology
Original Assignee
Guangdong University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong University of Technology filed Critical Guangdong University of Technology
Priority to CN202211311275.8A priority Critical patent/CN115619078B/en
Publication of CN115619078A publication Critical patent/CN115619078A/en
Application granted granted Critical
Publication of CN115619078B publication Critical patent/CN115619078B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0635Risk analysis of enterprise or organisation activities
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06Energy or water supply
    • YGENERAL 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
    • Y04INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04SSYSTEMS 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
    • Y04S10/00Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to the technical field of risk management and control of substations, in particular to a method and a device for predicting hazard risk levels of small animals in substations, wherein the method comprises the following steps: acquiring equipment characteristic information in a to-be-detected area of a transformer substation; inputting the equipment characteristic information into a small animal activity prediction model to obtain activity information of the small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information; inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in a region to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information; the risk grade division is carried out on the equipment damage information through the fuzzy membership function, so that the risk grade of the small animal damage in the area to be detected is obtained, the risk grade of the small animal damage in the transformer substation is effectively predicted, and effective data support is provided for preventing and solving the problem of the small animal damage for power operation and maintenance staff.

Description

Method and device for predicting hazard risk level of small animals in transformer substation
Technical Field
The invention relates to the technical field of transformer substation risk management and control, in particular to a method and a device for predicting risk levels of damage to small animals in a transformer substation.
Background
The transformer substation refers to a place for converting voltage and current in a power system, receiving electric energy and distributing electric energy, and is an indispensable key link in the power system, which affects the operation level of the power system. Therefore, in order to improve the stability of the operation of the power system, the safety and stability of the substation need to be improved.
One of the factors affecting the safety and stability of the substation is the small animal hazard. In a dark but incompletely closed space such as a transformer substation, mice, snakes, frogs and other small animals often appear, and the occurrence of the mice, snakes, frogs and other small animals often causes damage to transformer substation equipment, short circuit of a line, large load and the like. However, the prior art lacks evaluation and prediction of risk of damage to small animals in a transformer substation, and still needs to spend a great deal of manpower cost and time cost for manual investigation.
Disclosure of Invention
The invention provides a method and a device for predicting the risk level of the harm of a small animal in a transformer substation, which are used for predicting the risk level of the harm of the small animal in the transformer substation.
The invention provides a method for predicting hazard risk level of small animals in a transformer substation, which comprises the following steps:
acquiring equipment characteristic information in a to-be-detected area of a transformer substation;
Inputting the equipment characteristic information into a small animal activity prediction model to obtain activity information of the small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information;
inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information;
and carrying out risk grading on the equipment damage information through a fuzzy membership function to obtain the risk grade of the small animal damage in the area to be detected.
Optionally, the method further comprises:
grid division is carried out on the indoor ground area of the transformer substation, and grid areas are obtained;
acquiring historical equipment characteristic information and historical activity information of the small animals in each grid area from historical data of the transformer substation;
respectively carrying out data preprocessing on the historical equipment characteristic information and the historical activity information to obtain a training data set of the equipment characteristic information and a training data set of the activity information;
wherein the training dataset of device characteristic information is used as input to the animal activity prediction model; the training dataset of activity information is used as an output of the animal activity prediction model.
Optionally, the training step of the small animal activity prediction model comprises:
constructing an initial small animal activity prediction model according to an XGBoost algorithm;
based on the training data set of the equipment characteristic information and the training data set of the activity information, performing optimization training on the initial small animal activity prediction model by using a Bayesian optimization algorithm to obtain a small animal activity prediction model with a mapping relation between the equipment characteristic information and the activity information;
the initial small animal activity prediction model is:
Figure BDA0003908014610000021
wherein X is i 4*1 column vector represents the device characteristic information of the ith grid area; f (f) j (X i ) A predictive score of a j-th regression tree representing the small animal activity predictive model corresponding to the i-th grid region;
Figure BDA0003908014610000022
a 2*1 column vector representing the small animal activity information of the ith grid area; />
Figure BDA0003908014610000023
A leaf node value of the small animal activity prediction model corresponding to the ith grid area is represented; q (Xi) represents the mapping relation between the leaf nodes of the small animal activity prediction model and the ith grid area;
the objective function expression of the initial small animal activity prediction model is as follows:
Figure BDA0003908014610000024
Figure BDA0003908014610000025
wherein t is the number of small animal activity prediction regression trees to be optimized,
Figure BDA0003908014610000026
as a loss function, Ω (f (t) ) And gamma is a first super parameter of the regular term, lambda is a second super parameter of the regular term, and T is the number of leaf nodes of the current regression tree.
Optionally, the training step of the small animal hazard risk assessment model includes:
acquiring historical equipment damage information in each grid area;
performing data preprocessing on the historical equipment damage information to obtain a historical equipment damage information training data set;
and taking the training data set of the activity information as input quantity and the training data set of the equipment damage information as output quantity, and performing iterative training on a pre-constructed small animal hazard risk assessment model to obtain the small animal hazard risk assessment model with the mapping relation between the activity information and the equipment damage information.
Optionally, the small animal hazard risk assessment model is:
Figure BDA0003908014610000031
wherein Z is i Device damage information for ith grid area, Y i Information on the activities of the animals in the ith grid area including the number of occurrences of the animals y i1 Residence time y i2 P (Z is less than or equal to j|Y) is the cumulative probability of the risk of damage to the small animal; j is the class of damaged results of the device; alpha and beta are the first and second effector parameters of the small animal hazard risk assessment model.
Optionally, the number of the areas to be measured is a plurality, and the method further includes:
and drawing a risk level distribution diagram according to the risk level of the small animal hazard in each region to be tested.
The invention provides a device for predicting hazard risk level of small animals in a transformer substation, which comprises the following components:
the first acquisition module is used for acquiring equipment characteristic information in a to-be-detected area of the transformer substation;
the first input module is used for inputting the equipment characteristic information into a small animal activity prediction model to obtain the activity information of the small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information;
the second input module is used for inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information;
and the risk grade classification module is used for carrying out risk grade classification on the equipment damage information through a fuzzy membership function to obtain the risk grade of the small animal damage of the area to be detected.
Optionally, the apparatus further comprises:
the division module is used for carrying out gridding division on the indoor ground area of the transformer substation to obtain grid areas;
The second acquisition module is used for acquiring historical equipment characteristic information and historical activity information of the small animals in each grid area from the historical data of the transformer substation;
the first data preprocessing module is used for respectively preprocessing the historical equipment characteristic information and the historical activity information to obtain a training data set of the equipment characteristic information and a training data set of the activity information;
wherein the training dataset of device characteristic information is used as input to the animal activity prediction model; the training dataset of activity information is used as an output of the animal activity prediction model.
Optionally, the apparatus further comprises:
a third obtaining module, configured to obtain historical equipment damage information in each grid area;
the second data preprocessing module is used for preprocessing the data of the historical equipment damage information to obtain a historical equipment damage information training data set;
the second training module is used for performing iterative training on the pre-built small animal hazard risk assessment model by taking the training data set of the activity information as input quantity and the training data set of the equipment damage information as output quantity to obtain the small animal hazard risk assessment model with the mapping relation between the activity information and the equipment damage information.
Optionally, the number of the areas to be measured is plural, and the apparatus further includes:
and the drawing module is used for drawing a risk level distribution diagram according to the risk level of the small animal hazard in each region to be tested.
From the above technical scheme, the invention has the following advantages:
according to the method, equipment characteristic information in a region to be detected of a transformer substation is acquired, and is input into a small animal activity prediction model to obtain activity information of a small animal; the small animal activity prediction model is a mapping relation model of equipment characteristic information and activity information, and the activity information is input into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information, risk grade division is carried out on the equipment damage information through a fuzzy membership function, the small animal hazard risk grade of the area to be detected is obtained, the small animal hazard risk grade in the transformer substation is effectively predicted, effective data support is provided for electric operation and maintenance staff to prevent and solve the small animal hazard problem, and time cost and labor cost are greatly reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained from these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for predicting risk levels of damage to small animals in a transformer substation according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a method for predicting risk levels of damage to small animals in a transformer substation according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a device for predicting risk levels of damage to small animals in a transformer substation according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method and a device for predicting the risk level of the damage of a small animal in a transformer substation, which are used for predicting the risk level of the damage of the small animal in the transformer substation.
In order to make the objects, features and advantages of the present invention more comprehensible, the technical solutions in the embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it is apparent that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment one:
referring to fig. 1, fig. 1 is a flow chart of a method for predicting risk levels of damage to small animals in a transformer substation according to an embodiment of the present invention.
The method for predicting the risk level of the small animal hazard in the transformer substation provided by the embodiment comprises the following steps:
101. acquiring equipment characteristic information in a to-be-detected area of a transformer substation;
it should be noted that the area to be measured refers to a substation area that needs to be inspected, and the area to be measured can be selected according to operation and maintenance requirements. The equipment characteristic information comprises equipment information, equipment wiring information, equipment occupied area and equipment density. The device information refers to device type information and the like. The equipment density refers to the ratio of the total equipment footprint to the area occupied by the equipment in the area. The device routing information refers to information of a connection cable between devices in an area.
102. Inputting the equipment characteristic information into a small animal activity prediction model to obtain activity information of the small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information.
The small animal activity prediction model is a model which is built in advance and is trained, has a mapping relation between equipment characteristic information and activity information, and correspondingly outputs the activity information of the small animal when the equipment characteristic information is input. The device characteristic information is input in the form of a column vector, and the vector is a column vector of 4*1.
The activity information output by the animal activity prediction model is predicted according to the equipment characteristic information and is output in a vector form, and the vector form is 2*1 columns of vectors. The small animal activity information includes: residence time of the small animals and number of times the small animals appear.
103. Inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information.
It should be noted that the equipment damage information refers to the damage degree of the equipment, and can be used to represent the risk degree of damage to the small animals. The small animal hazard risk assessment model is a model which is built in advance and is trained, and has a mapping relation between activity information and equipment damage information. After the activity information of the small animals obtained in the step 102 is input into the small animal hazard risk assessment model, the small animal hazard risk assessment model correspondingly outputs equipment damage information.
104. And carrying out risk grading on the equipment damage information through a fuzzy membership function to obtain the risk grade of the small animal damage in the area to be detected.
It should be noted that the number of the regions to be measured may be plural, and the fuzzy membership function is:
Figure BDA0003908014610000061
wherein A is i (z) is the risk level of the small animal hazard in the ith test area; z is equipment damage information, a, b, c, d is the first risk level respectivelyA second risk level, a third risk level, a fourth risk level, wherein a<b<c<d。
Specific values of the first risk level, the second risk level, the third risk level and the fourth risk level can be determined by adopting an intuition method, namely, a class interval of the membership function is determined through the understanding and understanding of the fuzzy concept of the hazard risk level of the small animals by the transformer substation staff.
In this embodiment, risk classification is performed on the equipment damage information through the fuzzy membership function to obtain the risk level of the small animal damage in the area to be detected, so that prediction of the risk condition of the small animal damage in each place of the transformer substation is realized, operation and maintenance personnel can pay attention to the transformer substation area possibly having serious risk of damage in time, corresponding measures are performed on the serious risk area in time, and the safety of the transformer substation is improved.
As an application example, the prior art is to expel small animals by arranging a small animal expelling device so as to reduce the harm of the animals to a transformer substation. However, when setting up the toy and driving the device, do not set up pertinently, cause the device that drives to put in too much, the problem of extravagant resource easily, perhaps drive the device and throw too little, be difficult to obtain the effectual problem of expelling effect. By the prediction method provided by the embodiment, after the risk level of the small animals in each area of the transformer substation is obtained, operation and maintenance personnel can set a corresponding number of driving devices according to the risk level of each area, for example, a larger number of driving devices are set in the area with high risk level, fewer driving devices are set in the area with low risk level, and the like, so that the risk brought by the small animals to the transformer substation is reduced, and meanwhile, the resource waste is avoided. Therefore, the embodiment provides targeted data support for the position setting and the placement number of the small animal driving device by the power operation staff.
According to the method, equipment characteristic information in a region to be detected of a transformer substation is acquired, and is input into a small animal activity prediction model to obtain activity information of a small animal; the small animal activity prediction model is a mapping relation model of equipment characteristic information and activity information, and the activity information is input into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information, risk grade division is carried out on the equipment damage information through a fuzzy membership function, the small animal hazard risk grade of the area to be detected is obtained, the small animal hazard risk grade in the transformer substation is effectively predicted, effective data support is provided for electric operation and maintenance staff to prevent and solve the small animal hazard problem, and time cost and labor cost are greatly reduced.
In another preferred embodiment, the number of the areas to be measured is plural, and step 104 further includes:
and drawing a risk level distribution diagram according to the risk level of the small animal hazard in each region to be tested.
It should be noted that, according to the risk levels of the small animals in each area to be detected obtained in steps 101-104, performing contour fitting on the areas with the same risk level to obtain each risk level distribution block surrounded by a plurality of irregular and mutually disjoint contour lines, and finally filling the blocks with different risk levels by adopting colors with different depths to obtain a small animal risk level distribution map.
Therefore, in this embodiment, according to the risk level of the small animal hazard in each region to be tested, a risk level distribution map is drawn, so that the risk level condition of each region in the transformer substation is intuitively displayed, more intuitive data is provided for operation and maintenance personnel, and different preventive measures are more conveniently performed for regions with different risk levels.
Embodiment two:
referring to fig. 2, fig. 2 is a flow chart of a method for predicting risk levels of damage to small animals in a transformer substation according to an embodiment of the present invention.
The method provided by the embodiment comprises the following steps:
201. grid division is carried out on the indoor ground area of the transformer substation, and grid areas are obtained;
in this embodiment, first, grid division is performed on the indoor ground area of the substation to obtain each grid area.
The size of the grid area may be a multi-step size of 30cm by 30cm,50cm by 50cm,100cm by 100cm,200cm by 200cm,300cm by 300cm, etc. When the grid area is divided into large areas, the corresponding grid quantity is small, and when the grid area is divided into small areas, the grid quantity is large. The number of grids is increased, so that the calculation accuracy is improved, but the calculation scale is also improved, and the risk prediction effect is affected when the grid area is smaller than the equipment floor area, so that the comprehensive consideration needs to be carried out by combining the factors of the calculation accuracy, the calculation scale, whether the grid area is smaller than the equipment floor area and the like when determining the size of the grid area. In this embodiment, the mesh area size is preferably: 100cm x 100cm.
202. Acquiring historical equipment characteristic information and historical activity information of the small animals in each grid area from historical data of the transformer substation;
the historical device feature information includes: historical equipment information, historical equipment wiring information and historical equipment occupied area. Historical activity information for small animals includes: historical residence time of small animals and historical number of times the small animals appeared.
The historical equipment wiring information refers to historical data corresponding to the equipment wiring information of the grid area.
The calculation formula of the equipment wiring information is as follows:
Figure BDA0003908014610000091
wherein n is the total number of single cables with average height less than 1 meter, l i Indicating the length of the cable, h i Representing the average height, x, of an individual cable i3 Device routing information representing an ith grid area.
The closer the cable is to the ground, the more easily the cable is damaged when the cable is clung to the ground, the cable is basically not damaged when the cable is one meter or more away from the ground, (1-h) i ) Representing a single unitThe extent to which the cable is vulnerable,
Figure BDA0003908014610000092
indicating the extent to which the cables are vulnerable within the grid area. Thus, the device routing information may be used to represent the vulnerability of the cables within the area.
In this embodiment, the historical data corresponding to the equipment feature information and the historical data corresponding to the activity information of the small animal are obtained to be used as training data of the small animal activity prediction model, so as to train the small animal activity prediction model.
203. Respectively carrying out data preprocessing on the historical equipment characteristic information and the historical activity information to obtain a training data set of the equipment characteristic information and a training data set of the activity information;
it should be noted that the data preprocessing includes data cleansing and data transformation. In this embodiment, after data cleaning is performed on the historical equipment feature information, data conversion is performed on the historical equipment feature information subjected to data cleaning, so as to obtain a training data set of the equipment feature information. Data preprocessing of historical activity information is the same.
The data cleaning process comprises the following steps:
1) Data consistency processing: renaming data with the same column name and different meanings or the same meaning with different column names. For example: and the device information is the device information, but the specific characteristics of different devices are represented, and the specific characteristics of the different devices are subjected to distinguishing naming again. Animal residence time, animal activity time and animal loitering time with different column names are all animal residence time, so that the data of the three are re-named as animal residence time.
2) Data deduplication: for multiple sets of repeated data, only one set of effective data is reserved, and other repeated input data are deleted.
3) Missing value processing: and (5) re-complementing the missing data. If the equipment characteristic information is missing, positioning is carried out according to the position information of the equipment, and the equipment characteristic information is complemented according to the actual information. And if the small animal activity information is missing, deleting the missing corresponding small animal activity information, and resampling to complement.
4) Outlier processing: and data which does not conform to the data specification or does not conform to the underlying logic of the data are collected again or deleted. If the small animal residence time unit is not in conformity with the specification, the small animal residence time is collected again, and if the corresponding equipment coordinates of the collected equipment characteristic information are out of the indoor coordinate range of the transformer substation, the small animal residence time unit is collected again or deleted.
The data transformation refers to data normalization processing of the equipment characteristic information and the animal activity information after data cleaning.
In the equipment characteristic information and the small animal activity information, the residence time data of the small animals are unevenly distributed and the interval span is larger, compared with the average distribution, the fitting degree of the residence time data and the normal distribution is higher, so that the z-score standardization method is adopted for transformation, the comparability between the data can be ensured on the premise of simple transformation calculation, and the data set is more standard.
The z-score normalization formula is as follows:
Figure BDA0003908014610000101
wherein y is i2 Is the residence time of small animals, mu is the mean value thereof, sigma is the standard deviation thereof, y i2 ' is the standard value after z-score transformation.
In the equipment characteristic information and the small animal activity information, all data except the residence time of the small animal are relatively average, the intervals among the data are more integer words such as number and times, and the influence degree can be directly measured, so that the numerical values are only uniformly distributed in the [0,1] interval, and therefore, when the data except the residence time of the small animal are subjected to data transformation, the data are directly transformed by adopting a min-max standardization method, and the data are ensured to be compared in the same dimension while the variation dimension and the variation range are eliminated.
Taking the data standardization of the equipment floor space as an example, the standardized formula is as follows:
Figure BDA0003908014610000102
wherein x is i2 For raw data of the equipment footprints, max is the maximum value of the equipment footprints, min is the minimum value of the equipment footprints, x i2 * The standard value after min-max conversion.
After the data cleaning and data transformation, a training data set of the equipment characteristic information and a training data set of the small animal activity information are obtained.
The training data set for obtaining the equipment characteristic information and the training data set for obtaining the animal activity information can form an animal hazard data set of an animal activity prediction model, and the animal hazard data set can be directly used as input and output data of the model. Taking the training data set of the equipment characteristic information as the input of the small animal activity prediction model, taking the training data set of the activity information as the output of the small animal activity prediction model, and taking the input of the small animal activity prediction model as X i =[x i1 ;x i2 ;x i3 ;x i4 ] T I represents the ith grid area, x i1 Indicating the device type, x i2 Representing the footprint of the device, x i3 Representing equipment wiring information, x i4 Indicating the device density. The output is: y is Y i =[y i1 ;y i2 ] T I represents the position of the grid area where the data is located, y i1 Indicating the number of times of appearance of small animals, y i2 Indicating the residence time of the small animals.
204. Constructing an initial small animal activity prediction model according to an XGBoost algorithm;
the method comprises the steps of constructing an initial small animal activity prediction model based on an XGBoost algorithm; wherein, the initial small animal activity prediction model is:
Figure BDA0003908014610000111
wherein X is i For 4*1 column vector, the device characteristic information of the ith grid area comprises device category x i1 Area x of equipment i2 Device trace information x i3 Degree of equipment Density x i4 ;f j (X i ) A predictive score of a j-th regression tree representing the small animal activity predictive model corresponding to the i-th grid region;
Figure BDA0003908014610000112
for 2*1 column vector, the small animal activity information of the ith grid area comprises the occurrence number y of small animals i1 And residence time y i2 ;/>
Figure BDA0003908014610000113
A leaf node value of the small animal activity prediction model corresponding to the ith grid area is represented; q (Xi) represents the mapping relation between the leaf nodes of the small animal activity prediction model and the ith grid area;
the objective function expression of the initial small animal activity prediction model is as follows:
Figure BDA0003908014610000114
Figure BDA0003908014610000115
wherein t is the number of small animal activity prediction regression trees to be optimized,
Figure BDA0003908014610000116
as a loss function, Ω (f (t) ) And gamma is a first super parameter of the regular term, lambda is a second super parameter of the regular term, and T is the number of leaf nodes of the current regression tree.
Wherein, gamma and lambda are used for controlling punishment force.
In order to further improve the prediction precision of the small animal activity prediction model, the embodiment performs second-order taylor expansion on the objective function of the initial small animal activity prediction model to find the optimal solution of the objective function, thereby obtaining the optimal objective function.
The second order taylor expansion of the objective function is as follows:
Figure BDA0003908014610000121
wherein c is a complex constant term of the previous t-1 tree, g i 、h i The method comprises the following steps:
Figure BDA0003908014610000122
Figure BDA0003908014610000123
substituting the regular term Ω (f) (t) ) And then, further simplifying to obtain:
Figure BDA0003908014610000124
wherein:
Figure BDA0003908014610000125
Figure BDA0003908014610000126
because of H j +λ>0, when
Figure BDA0003908014610000127
And obtaining an optimal function for the optimal solution:
Figure BDA0003908014610000128
205. and optimizing and training the initial small animal activity prediction model by using a Bayesian optimization algorithm based on the training data set of the equipment characteristic information and the training data set of the activity information to obtain the small animal activity prediction model with the mapping relation between the equipment characteristic information and the activity information.
According to the method, the super-parameters in the initial small animal activity prediction model are optimized and trained by using a Bayesian optimization algorithm, so that the optimal super-parameters are obtained, and the prediction precision of the small animal activity prediction model is further improved.
The Bayesian optimization process comprises the following steps: initializing G in a small animal activity prediction model j 、H j Defining domain intervals of internal parameters and super parameters gamma and lambda; device characteristic information data training set (device category x i1 Area x of equipment i2 Device routing case x i3 Degree of equipment Density x i4 ) As the input of the initial animal activity prediction model, observing and solving the number y of animal appearance in the corresponding grid area i1 And residence time y i2 The method comprises the steps of carrying out a first treatment on the surface of the Estimating an objective function of an initial small animal activity prediction model by taking TPE based on a Gaussian mixture model as a probability agent model to obtain a global optimal parameter; inputting global optimal parameters into an initial small animal activity prediction model for training; and repeatedly executing the optimization operation until the prediction precision of the initial small animal activity prediction model reaches a preset minimum prediction precision threshold value, and obtaining the small animal activity prediction model with the mapping relation between the equipment characteristic information and the activity information. The optimal super-parameters are the super-parameters corresponding to the initial small-activity prediction model meeting the preset prediction precision minimum threshold.
The objective function of the bayesian optimization algorithm is:
Figure BDA0003908014610000131
wherein f (a) is Bayesian optimization objective function, and a is small animal livingHyper-parameter set in dynamic prediction model, comprising first hyper-parameter gamma of regular term and second hyper-parameter lambda, a of regular term * Optimum parameter values for minimizing the objective function.
The Gaussian mixture model-based TPE is a model-based sequential optimization method, a probability agent model is established in a classification mode, new sample points are not generated too greedily, the situation of local optimization can be prevented, and the method has good global exploration capacity.
According to the embodiment, the indoor ground area of the transformer substation is subjected to gridding division to obtain grid areas, historical equipment characteristic information and historical activity information of animals in the grid areas are obtained from historical data of the transformer substation, and the historical equipment characteristic information and the historical activity information are respectively subjected to data preprocessing, so that a data set (a training data set of equipment characteristic information and a training data set of activity information) which can be directly used as input and output of an animal activity prediction model is obtained; and constructing and training an initial small animal activity prediction model according to the XGBoost algorithm, optimizing and training the initial small animal activity prediction model by using a Bayesian optimization algorithm based on the training data set of the equipment characteristic information and the training data set of the activity information to obtain a small animal activity prediction model with the mapping relation between the equipment characteristic information and the activity information, so that the small animal activity prediction model is constructed and trained, and the prediction precision of the small animal activity prediction model is improved by using the Bayesian optimization algorithm.
In another preferred embodiment, after the small animal activity prediction model is obtained, the accuracy and stability of the optimized small animal activity prediction model is also checked by K-cross validation.
206. And acquiring historical equipment damage information in each grid area.
In this embodiment, historical equipment damage information in each grid area is obtained from historical data of the transformer substation. Historical device impairment information refers to the degree of historical impairment of a device.
207. And carrying out data preprocessing on the historical equipment damage information to obtain a historical equipment damage information training data set.
It should be noted that, the data preprocessing may refer to step 203. In this embodiment, after the data cleaning is performed on the damaged information of the historical equipment, the data conversion is performed on the damaged information of the historical equipment after the data cleaning. When data transformation is carried out, a min-max standardization method is adopted for transformation.
208. And taking the training data set of the activity information as input quantity and the training data set of the equipment damage information as output quantity, and performing iterative training on a pre-constructed small animal hazard risk assessment model to obtain the small animal hazard risk assessment model with the mapping relation between the activity information and the equipment damage information.
In the embodiment, a small animal hazard risk assessment model is built in advance according to an ordered logistic regression model.
The small animal hazard risk assessment model formula is as follows:
Figure BDA0003908014610000141
wherein Z is i Device damage information for ith grid area, Y i Information on the activities of the animals in the ith grid area including the number of occurrences of the animals y i1 Residence time y i2 P (Z is less than or equal to j|Y) is the accumulated probability of the risk of damage to the small animal, and j is the category of the damaged result of the equipment; alpha and beta are the first and second effector parameters of the small animal hazard risk assessment model.
And then, taking the training data set of the activity information as the input of the small animal hazard risk assessment model, taking the training data set of the equipment damage information as the output of the small animal hazard risk assessment model, and training the small animal hazard risk assessment model to obtain optimal effect parameters alpha and beta, thereby obtaining the small animal hazard risk assessment model with the mapping relation of the activity information and the equipment damage information.
It will be appreciated that the order of steps 206-207 may follow step 202 or step 205, and the present embodiment is not specifically limited herein.
209. Acquiring equipment characteristic information in a to-be-detected area of a transformer substation;
It should be noted that the area to be measured refers to a substation area that needs to be inspected, and the area to be measured can be selected according to operation and maintenance requirements. The number of areas to be measured may be plural.
210. And inputting the equipment characteristic information into a small animal activity prediction model to obtain the activity information of the small animal.
The device characteristic information is input in the form of a column vector, and the column vector is 4*1, which corresponds to the form of the training data set. And inputting the equipment characteristic information into the small animal activity prediction model obtained in the step 205 to obtain predicted small animal activity information.
211. And inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in the region to be detected.
It should be noted that, the small animal activity information predicted in step 210 is input into the small animal hazard risk assessment model trained in step 208, so as to obtain the equipment damage information of the area to be tested. The equipment damage information output by the small animal hazard risk assessment model is sequencing type data, and can be used for comparing sizes and sequencing.
212. And carrying out risk grading on the equipment damage information through a fuzzy membership function to obtain the risk grade of the small animal damage in the area to be detected.
In particular, reference may be made to step 104, which is not described herein.
In another preferred embodiment, after acquiring the device feature information in the to-be-detected area of the substation, the method further includes:
carrying out data preprocessing on the equipment characteristic information;
the method for inputting the equipment characteristic information into the animal activity prediction model to obtain the animal activity information specifically comprises the following steps:
and inputting the device characteristic information subjected to data preprocessing into a small animal activity prediction model to obtain small animal activity information.
It can be understood that, in order to improve the prediction accuracy of the small animal activity prediction model, when the small animal activity prediction model is used to predict the region to be detected, the device feature information of the region to be detected is input into the small animal activity prediction model after the data processing operation in step 203 is performed, so as to obtain the corresponding small animal activity information.
Embodiment III:
referring to fig. 3, fig. 3 is a schematic structural diagram of a device for predicting risk levels of damage to small animals in a transformer substation according to an embodiment of the present invention.
The device for predicting risk level of small animal hazard in transformer substation provided by the embodiment comprises:
the first acquiring module 301 is configured to acquire device feature information in a to-be-detected area of the substation;
The first input module 302 is configured to input the device feature information into a small animal activity prediction model to obtain activity information of a small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information;
the second input module 303 is configured to input the activity information into a small animal hazard risk assessment model, so as to obtain equipment damage information in the area to be tested; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information;
and the risk classification module 304 is configured to perform risk classification on the equipment damage information through a fuzzy membership function, so as to obtain a risk class of the small animal hazard in the area to be detected.
Further, the apparatus further comprises:
the division module is used for carrying out gridding division on the indoor ground area of the transformer substation to obtain grid areas;
the second acquisition module is used for acquiring historical equipment characteristic information and historical activity information of the small animals in each grid area from the historical data of the transformer substation;
the first data preprocessing module is used for respectively preprocessing the historical equipment characteristic information and the historical activity information to obtain a training data set of the equipment characteristic information and a training data set of the activity information;
Wherein the training dataset of device characteristic information is used as input to the animal activity prediction model; the training dataset of activity information is used as an output of the animal activity prediction model.
Further, the apparatus further comprises:
the construction module is used for constructing an initial small animal activity prediction model according to the XGBoost algorithm;
the first training module is used for carrying out optimization training on the initial small animal activity prediction model by using a Bayesian optimization algorithm based on the training data set of the equipment characteristic information and the training data set of the activity information to obtain a small animal activity prediction model with the mapping relation between the equipment characteristic information and the activity information;
further, the apparatus further comprises:
a third obtaining module, configured to obtain historical equipment damage information in each grid area;
the second data preprocessing module is used for preprocessing the data of the historical equipment damage information to obtain a historical equipment damage information training data set;
the second training module is used for performing iterative training on the pre-built small animal hazard risk assessment model by taking the training data set of the activity information as input quantity and the training data set of the equipment damage information as output quantity to obtain the small animal hazard risk assessment model with the mapping relation between the activity information and the equipment damage information.
Further, the number of the areas to be measured is plural, and the apparatus further includes:
and the drawing module is used for drawing a risk level distribution diagram according to the risk level of the small animal hazard in each region to be tested.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the systems, apparatuses and modules described above may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing module, or each functional module may exist separately and physically, or two or more functional modules may be integrated in one processing module. The integrated modules may be implemented in hardware or in software functional modules.
The integrated modules, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RAM, randomAccessMemory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (6)

1. A method for predicting risk levels of small animal hazards in a substation, the method comprising:
grid division is carried out on the indoor ground area of the transformer substation, and grid areas are obtained;
acquiring historical equipment characteristic information and historical activity information of the small animals in each grid area from historical data of the transformer substation;
respectively carrying out data preprocessing on the historical equipment characteristic information and the historical activity information to obtain a training data set of the equipment characteristic information and a training data set of the activity information;
wherein the training data set of the device characteristic information is used as an input of a small animal activity prediction model; the training data set of the activity information is used as the output of the animal activity prediction model;
Acquiring equipment characteristic information in a to-be-detected area of a transformer substation;
inputting the equipment characteristic information into a small animal activity prediction model to obtain activity information of the small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information;
inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information;
carrying out risk grade division on the equipment damage information through a fuzzy membership function to obtain the risk grade of the small animal damage of the area to be detected;
the training step of the small animal activity prediction model comprises the following steps:
constructing an initial small animal activity prediction model according to an XGBoost algorithm;
based on the training data set of the equipment characteristic information and the training data set of the activity information, performing optimization training on the initial small animal activity prediction model by using a Bayesian optimization algorithm to obtain a small animal activity prediction model with a mapping relation between the equipment characteristic information and the activity information;
wherein the initial small animal activity prediction model is:
Figure QLYQS_1
Wherein X is i 4*1 column vector represents the device characteristic information of the ith grid area; f (f) j (X i ) A predictive score of a j-th regression tree representing the small animal activity predictive model corresponding to the i-th grid region;
Figure QLYQS_2
a 2*1 column vector representing the small animal activity information of the ith grid area; />
Figure QLYQS_3
A leaf node value of the small animal activity prediction model corresponding to the ith grid area is represented; q (Xi) represents the mapping relation between the leaf nodes of the small animal activity prediction model and the ith grid area;
the objective function expression of the initial small animal activity prediction model is as follows:
Figure QLYQS_4
Figure QLYQS_5
wherein t is the number of small animal activity prediction regression trees to be optimized,
Figure QLYQS_6
as a loss function, Ω (f (t) ) The method is characterized in that the method is a regular term, gamma is a first super parameter of the regular term, lambda is a second super parameter of the regular term, and T is the number of leaf nodes of a current regression tree;
the small animal hazard risk assessment model is as follows:
Figure QLYQS_7
wherein Z is i Device damage information for ith grid area, Y i Information on the activities of the animals in the ith grid area including the number of occurrences of the animals y i1 Residence time y i2 P (Z is less than or equal to j|Y) is the cumulative probability of the risk of damage to the small animal; j is the class of damaged results of the device; alpha and beta are the first and second effector parameters of the small animal hazard risk assessment model.
2. The method of claim 1, wherein the training step of the small animal hazard risk assessment model comprises:
acquiring historical equipment damage information in each grid area;
performing data preprocessing on the historical equipment damage information to obtain a historical equipment damage information training data set;
and taking the training data set of the activity information as input quantity and the training data set of the equipment damage information as output quantity, and performing iterative training on a pre-constructed small animal hazard risk assessment model to obtain the small animal hazard risk assessment model with the mapping relation between the activity information and the equipment damage information.
3. The method of claim 1, wherein the number of regions to be measured is a plurality, the method further comprising:
and drawing a risk level distribution diagram according to the risk level of the small animal hazard in each region to be tested.
4. A device for predicting risk levels of damage to small animals in a substation, the device comprising:
the division module is used for carrying out gridding division on the indoor ground area of the transformer substation to obtain grid areas;
the second acquisition module is used for acquiring historical equipment characteristic information and historical activity information of the small animals in each grid area from the historical data of the transformer substation;
The first data preprocessing module is used for respectively preprocessing the historical equipment characteristic information and the historical activity information to obtain a training data set of the equipment characteristic information and a training data set of the activity information;
wherein the training data set of the device characteristic information is used as an input of a small animal activity prediction model; the training data set of the activity information is used as the output of the animal activity prediction model;
the construction module is used for constructing an initial small animal activity prediction model according to the XGBoost algorithm;
the first training module is used for carrying out optimization training on the initial small animal activity prediction model by using a Bayesian optimization algorithm based on the training data set of the equipment characteristic information and the training data set of the activity information to obtain a small animal activity prediction model with the mapping relation between the equipment characteristic information and the activity information;
the first acquisition module is used for acquiring equipment characteristic information in a to-be-detected area of the transformer substation;
the first input module is used for inputting the equipment characteristic information into a small animal activity prediction model to obtain the activity information of the small animal; the animal activity prediction model is a mapping relation model of equipment characteristic information and activity information;
The second input module is used for inputting the activity information into a small animal hazard risk assessment model to obtain equipment damage information in the area to be detected; the small animal hazard risk assessment model is a mapping relation model of activity information and equipment damage information;
the risk grade classification module is used for carrying out risk grade classification on the equipment damage information through a fuzzy membership function to obtain the risk grade of the small animal damage of the area to be detected;
wherein the initial small animal activity prediction model is:
Figure QLYQS_8
wherein X is i 4*1 column vector represents the device characteristic information of the ith grid area; f (f) j (X i ) A predictive score of a j-th regression tree representing the small animal activity predictive model corresponding to the i-th grid region;
Figure QLYQS_9
a 2*1 column vector representing the small animal activity information of the ith grid area; />
Figure QLYQS_10
A leaf node value of the small animal activity prediction model corresponding to the ith grid area is represented; q (Xi) represents the mapping relation between the leaf nodes of the small animal activity prediction model and the ith grid area;
the objective function expression of the initial small animal activity prediction model is as follows:
Figure QLYQS_11
Figure QLYQS_12
wherein t is the number of small animal activity prediction regression trees to be optimized,
Figure QLYQS_13
As a loss function, Ω (f (t) ) The method is characterized in that the method is a regular term, gamma is a first super parameter of the regular term, lambda is a second super parameter of the regular term, and T is the number of leaf nodes of a current regression tree;
the small animal hazard risk assessment model is as follows:
Figure QLYQS_14
wherein Z is i Device damage information for ith grid area, Y i Information on the activities of the animals in the ith grid area including the number of occurrences of the animals y i1 Residence time y i2 P (Z is less than or equal to j|Y) is the cumulative probability of the risk of damage to the small animal; j is the class of damaged results of the device; alpha and beta are the first and second effector parameters of the small animal hazard risk assessment model.
5. The apparatus of claim 4, wherein the apparatus further comprises:
a third obtaining module, configured to obtain historical equipment damage information in each grid area;
the second data preprocessing module is used for preprocessing the data of the historical equipment damage information to obtain a historical equipment damage information training data set;
the second training module is used for performing iterative training on the pre-built small animal hazard risk assessment model by taking the training data set of the activity information as input quantity and the training data set of the equipment damage information as output quantity to obtain the small animal hazard risk assessment model with the mapping relation between the activity information and the equipment damage information.
6. The apparatus of claim 4, wherein the number of regions to be measured is a plurality, the apparatus further comprising:
and the drawing module is used for drawing a risk level distribution diagram according to the risk level of the small animal hazard in each region to be tested.
CN202211311275.8A 2022-10-25 2022-10-25 Method and device for predicting hazard risk level of small animals in transformer substation Active CN115619078B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211311275.8A CN115619078B (en) 2022-10-25 2022-10-25 Method and device for predicting hazard risk level of small animals in transformer substation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211311275.8A CN115619078B (en) 2022-10-25 2022-10-25 Method and device for predicting hazard risk level of small animals in transformer substation

Publications (2)

Publication Number Publication Date
CN115619078A CN115619078A (en) 2023-01-17
CN115619078B true CN115619078B (en) 2023-06-02

Family

ID=84863612

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211311275.8A Active CN115619078B (en) 2022-10-25 2022-10-25 Method and device for predicting hazard risk level of small animals in transformer substation

Country Status (1)

Country Link
CN (1) CN115619078B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110598726A (en) * 2019-07-16 2019-12-20 广东工业大学 Transmission tower bird damage risk prediction method based on random forest

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160071059A1 (en) * 2014-09-05 2016-03-10 Shafer, Kline & Warren, Inc. Infrastructure management, model, and deliverable creation system and method of use
CN113807570B (en) * 2021-08-12 2024-02-02 水利部南京水利水文自动化研究所 XGBoost-based reservoir dam risk level assessment method and system
CN114492973A (en) * 2022-01-18 2022-05-13 苏州热工研究院有限公司 Method for predicting marine organisms in peripheral sea area of nuclear power plant
CN115050047A (en) * 2022-05-30 2022-09-13 广东电网有限责任公司 Method and device for protecting animals in transformer substation, electronic equipment and storage medium

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110598726A (en) * 2019-07-16 2019-12-20 广东工业大学 Transmission tower bird damage risk prediction method based on random forest

Also Published As

Publication number Publication date
CN115619078A (en) 2023-01-17

Similar Documents

Publication Publication Date Title
Flores et al. Evolutive design of ARMA and ANN models for time series forecasting
CN108700851A (en) System, method and platform based on cloud for predicting energy expenditure
Decanini et al. Robust fault diagnosis in power distribution systems based on fuzzy ARTMAP neural network-aided evidence theory
CN114519514B (en) Low-voltage transformer area reasonable line loss value measuring and calculating method, system and computer equipment
CN115293326A (en) Training method and device of power load prediction model and power load prediction method
CN113705688B (en) Abnormal electricity consumption behavior detection method and system for power users
Nugroho et al. A review of simulation urban growth model
Eliades et al. Iterative deepening of Pareto solutions in water sensor networks
Xie et al. Autoencoder-based deep belief regression network for air particulate matter concentration forecasting
CN112241836A (en) Virtual load dominant parameter identification method based on incremental learning
CN115619078B (en) Method and device for predicting hazard risk level of small animals in transformer substation
CN110956281A (en) Power equipment abnormity detection alarm system based on Log analysis
Cao et al. Fast and explainable warm-start point learning for AC Optimal Power Flow using decision tree
CN112363012A (en) Power grid fault early warning device and method
Qin et al. Modeling data envelopment analysis by chance method in hybrid uncertain environments
Pan et al. Assessment of MV XLPE cable aging state based on PSO-XGBoost algorithm
CN112668905B (en) Multi-parameter power distribution cable health state evaluation method and device, computer equipment and storage medium
CN114118759A (en) Distribution transformer area load overload state assessment method and device
Salski et al. A fuzzy and neuro-fuzzy approach to modelling cattle grazing on pastures with low stocking rates in Central Europe
Peters et al. Reinforcement Learning with Pattern-based Rewards.
CN114201825A (en) Method and system for evaluating equipment performance degradation state based on combination characteristics
Shao et al. Wetland Ecotourism Development Using Deep Learning and Grey Clustering Algorithm from the Perspective of Sustainable Development
Aminudin et al. Voltage collapse risk index prediction for real time system's security monitoring
Qian et al. High load function prediction model based on decision tree
Dai et al. Life prediction method of hydrogen energy battery based on MLP and LOESS

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