CN108898243A - A kind of Transmission Network of Power System security test system - Google Patents
A kind of Transmission Network of Power System security test system Download PDFInfo
- Publication number
- CN108898243A CN108898243A CN201810591879.XA CN201810591879A CN108898243A CN 108898243 A CN108898243 A CN 108898243A CN 201810591879 A CN201810591879 A CN 201810591879A CN 108898243 A CN108898243 A CN 108898243A
- Authority
- CN
- China
- Prior art keywords
- module
- prediction
- sequence
- weight
- obtains
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 54
- 238000012360 testing method Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 claims abstract description 30
- 230000003252 repetitive effect Effects 0.000 claims abstract description 14
- 230000008859 change Effects 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 238000004891 communication Methods 0.000 claims abstract description 9
- 238000003062 neural network model Methods 0.000 claims description 49
- 230000006870 function Effects 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 19
- 238000012986 modification Methods 0.000 claims description 17
- 230000004048 modification Effects 0.000 claims description 17
- 230000003044 adaptive effect Effects 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 12
- 238000005065 mining Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 9
- 230000002068 genetic effect Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000007405 data analysis Methods 0.000 claims description 6
- 239000000427 antigen Substances 0.000 claims description 3
- 102000036639 antigens Human genes 0.000 claims description 3
- 108091007433 antigens Proteins 0.000 claims description 3
- 238000013459 approach Methods 0.000 claims description 3
- 238000009412 basement excavation Methods 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000007418 data mining Methods 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 description 23
- 230000005611 electricity Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to electric network security test technical fields, disclose a kind of Transmission Network of Power System security test system, including:Test instruction input module, initial parameter configuration module, central control module, wireless communication module, expert advice module, alarm module, display module, load forecast module.The present invention provides expert advice module can provide more professional guiding opinion to staff, ensure the professional of power grid security test, accident occurrence probability caused by reduction is artificial;The present invention passes through the characteristics of this cyclically-varying trend of load forecast module analysis electric load, basic model using SIN function model as prediction data, have the advantages that be bonded Power system load data actual change, in order to overcome the disadvantage for using SIN function relatively large as basic model error, the method that the present invention also uses repetitive learning improves the accuracy of prediction.
Description
Technical field
The invention belongs to electric network security test technical field more particularly to a kind of Transmission Network of Power System safeties
Test macro.
Background technique
Currently, the prior art commonly used in the trade is such:
Electric system by power plant, send power transformation route, power supply and distribution and the electrical energy production that forms of the links such as electricity consumption with disappear
Charge system.Its function be the non-renewable energy of nature is converted to electric energy by generation power device, then through transmission of electricity, power transformation and
Distribution supplies power to each user.To realize this function, electric system also has accordingly in links and different levels
Information and control system, the production process of electric energy is measured, adjusted, controlled, protect, communicates and dispatched, to guarantee to use
Family obtains safety, good electric energy.However, since test job personnel specialty level is different, being surveyed when existing power grid security is tested
Examination is easy to appear unprofessional, leads to operating accident;It is existing big to load forecast error simultaneously, it is unfavorable for surveying power grid security
The accuracy of examination.
So-called " Transmission Network of Power System safety ", refers in scheduling problem, the object that is scheduled is generally by N number of work
The set of part (Job) composition, referred to as example (Instance).Example, J are indicated with IjIndicate j-th of workpiece therein.Each
Workpiece JjThere is respective arrival time (Release time) RjWith process time Pj(Processing time).It is dispatching
In scheme S, on-stream time Sj(Start time), completion date Cj(Completion time), and electric system is transmitted electricity
Network security, that is, ∑ Cj(Total completion time).So-called " Transmission Network of Power System safe prediction ", refers to and uses section
Mathematical model prediction is gone out on missions or the Transmission Network of Power System of project safety.
In engineer application, it is widely present the demand of Transmission Network of Power System safe prediction.Such as in manufacturing enterprise
In, there is various types of product, miscellaneous processing method and the different equipment of production capacity, production is adjusted
Degree can more reasonably coordinate various activities, reduce production cost simultaneously thereby increasing productivity.Formulate the scheduling scheme (packet of science
Include and predict reasonable Transmission Network of Power System safety), the Work in Process in workshop can be efficiently controlled, product is improved and hands over
Delivery date Service Efficiency and shortening product manufacturing period, while job guide is provided to shop layer personnel, facilitating tension management, person comments
The working condition of valence shop layer.Therefore, the research of Transmission Network of Power System safe prediction is to raising enterprises production efficiency, enhancing
Enterprise competitiveness has very strong practical significance.
With deeply becoming increasingly complex with the development of science and technology, the external environment that manufacturing enterprise faces for economic globalization
It is changeable, such as turn of the market is rapid, intensified competition, client's diversification.A core of the production scheduling problems as manufacture system
Heart problem, excellent scheduling result can help enterprise to shorten the production cycle, improve production efficiency, enhance the competitiveness, and minimum
The production scheduling problems for changing completion date obtain more concerns.Therefore, it is necessary to the electric system in production scheduling problems
Power transmission network is predicted safely, to optimize the production procedure of enterprise, improves the production efficiency of enterprise.
For manufacturing enterprise, the factor for influencing Transmission Network of Power System safe prediction is not limited solely to some vehicle
Between, but it is related to the links such as product design, manufacture, O&M, data involved in these links belong to industrial big number
According to scope, the networked data etc. comprising sensing data, controller data and device systems.Therefore, electric system is transmitted electricity
Network security prediction needs to organically blend with the mining analysis of industrial big data, that is, needs to obtain by the analysis of industrial big data
The every historical data and influence factor of product power system power transmission network safety are influenced, and then utilizes these influence factors and phase
The other historical datas closed complete the prediction of product power system power transmission network safety.
Transmission Network of Power System safety predicting method in engineer application field and manufacturing enterprise there has been very much
Research, but up to the present, it there is no one kind to carry out analysis mining to influence factor under the driving of industrial big data, and will dynamic spy
Property incorporate prediction model Transmission Network of Power System safety predicting method.
In conclusion problem of the existing technology is:
When existing power grid security is tested, since test job personnel specialty level is different, test is easy to appear unprofessional, leads
Cause operating accident;It is existing big to load forecast error simultaneously, it is unfavorable for the accuracy tested power grid security.
The prior art not can be carried out Accurate Prediction safely to Transmission Network of Power System, cannot optimize the safety of power transmission network
Property prediction, improve production efficiency, enterprise can not be adapted to because of various change caused by seasonal variations.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of Transmission Network of Power System security test systems
System.
The invention is realized in this way a kind of Transmission Network of Power System security test system includes:
Test instruction input module, initial parameter configuration module, central control module, wireless communication module, expert advice
Module, alarm module, display module, load forecast module;
Instruction input module is tested, is connect with central control module, for input test instruction operation;
Initial parameter configuration module, connect with central control module, for configuring the initial parameter of test environment;
Central control module, with test instruction input module, initial parameter configuration module, wireless communication module, Zhuan Jiajian
Module, alarm module, display module, the connection of load forecast module are discussed, is worked normally for dispatching modules;
Wireless communication module is connect with central control module, for carrying out remote testing control behaviour by wireless transmitter
Make;
Expert advice module, connect with central control module, for carrying out suggestion to test operation by expert's net;
Alarm module is connect with central control module, for being alarmed by alarm test failure;
Display module is connect with central control module, for showing test information data;
Load forecast module, connect with central control module, for predicting electric load.
Further, the load forecast module includes:Data acquisition module, study module, prediction module;
Data acquisition module acquires the historical data of electric load for block in chronological order;
Study module obtains basic sequence according to SIN function for preprocessing sequence fitting to be obtained a SIN function
Column;
Prediction module for obtaining forecasting sequence according to prediction SIN function model, and carries out anti-normalizing to forecasting sequence
Change processing, obtains prediction data sequence.
Further, the load forecast module prediction technique is as follows:
Step 1, data acquisition module acquire the historical data of electric load in chronological order, obtain Load Time Series
Load Time Series Y is normalized in Y, obtains preprocessing sequence O;
Step 2, preprocessing sequence O fitting is obtained a SIN function S (t) by study module, according to SIN function S (t)
Obtain basic sequence L1,
Step 3, study module, which subtracts each other preprocessing sequence O and basic sequence L1, obtains base residual sequence e;
Step 4, study module find out the root-mean-square error of base residual sequence e, by the root-mean-square error and preset valve
Value comparison;
If being less than preset threshold, SIN function S (t) is prediction SIN function model W (t), and prediction module is according to prediction
SIN function model W (t) obtains forecasting sequence f, and carries out anti-normalization processing to forecasting sequence f, obtains prediction data sequence
F;
If more than preset threshold, then following steps (e) to (g) is continued to execute;
Step 5, study module are fitted base residual sequence e, residual error SIN function Q (t) are obtained, according to residual error
SIN function Q
Step 6 obtains residual sequence M;
Step 7, study module find out the root-mean-square error of residual sequence M, by the root-mean-square error and preset threshold values pair
Than;
If being less than preset threshold, SIN function S (t) and residual error SIN function Q (t) is superimposed, obtains the sinusoidal letter of prediction
Exponential model W (t), prediction module obtain forecasting sequence f according to prediction SIN function model W (t), and carry out to forecasting sequence f anti-
Normalized obtains prediction data sequence F;
If more than preset threshold, then enter repetitive learning step:
According to residual sequence Mi (i=1,2,3 ..., n), fitting obtain prediction residual SIN function Pi (t) (i=1,2,
3 ..., n), prediction residual sequence Ni is obtained according to Pi (t), using formula M+1=Mi-Ni, finds out next residual sequence Mi+
1, Pi+1 (t) is obtained according to Mi+1 fitting, prediction residual sequence Ni+1 is obtained according to Pi+1 (t);
After the prediction residual sequence Ni that each operation obtains, the prediction residual sequence Ni root-mean-square error is calculated, by this
Square error and preset threshold values compare, if being less than preset threshold, stop repetitive learning step, if more than preset threshold, then
Continue repetitive learning step;
Step 8, after stopping repetitive learning step, by SIN function S (t) and with residual error SIN function Q (t) and all pre-
Residual error SIN function Pi (t) superposition is surveyed, obtains prediction SIN function model W (t), prediction module is according to prediction SIN function model
W (t) obtains forecasting sequence f, and carries out anti-normalization processing to forecasting sequence f, obtains forecasting sequence F.
The anti-normalization processing of prediction module includes:
S1, the big data based on Hadoop building inclusion relation type database data, sensing data and controller data
Analytical unit goes to step S2;
S2, under MapReduce frame use Apriori association rules mining algorithm, in big data analysis unit into
Row analysis and excavation, obtain Transmission Network of Power System Safety Influence Factors, go to step S3;
S3, in conjunction with Transmission Network of Power System Safety Influence Factors and Transmission Network of Power System security history data, structure
Neural network model BP is built, the initial weight of neural network model BP is generated, goes to step S4;
S4, the weight and threshold value of neural network model BP are dynamically refined, obtain dynamic neural network model DBP,
The weight and threshold value for generating dynamic neural network model DBP, go to step S5;
S5, with adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP, obtain prediction model
AIGA-DBP calculates Transmission Network of Power System safe prediction value according to prediction model AIGA-DBP, goes to step S6;
S6, judge Transmission Network of Power System safe prediction value and the error of Transmission Network of Power System safety expectation value is
The no condition for meeting setting, if so, going to step S7;Otherwise step S5 is re-executed;
S7, output power system power transmission network safe prediction value and forecasting sequence F terminate.
Further, step S1 specifically includes following steps:
Relevant database data, sensing data and controller data are uploaded to distributed field system by Sqoop
Unite HDFS, and stores into NoSQL database;Using MapReduce Computational frame to relevant database data, sensor number
Mining analysis is carried out according to controller data, NoSQL database is written into the data analyzed, and show by Web.
Further, in step S2 under MapReduce frame with Apriori association rules mining algorithm specifically include with
Lower step:
S201, the set L of frequent 1 item collection is obtained using MapReduce computation module1, generate the set C of candidate's k item collectionk;
S202, in Map function processing stage, Transaction Information of each Map task computation handled by it concentrates each affairs
C is included in recordkIn Item Sets frequency of occurrence, for each Map task, if some of candidate's k item collection
Collection (including k project) appears in a transaction journal, then Map function is generated and exported<Some item collection, 1>Key-value pair is given
Combiner function gives Reduce function after being handled by Combiner function;
S203, in Reduce function processing stage, Reduce function adds up CkIn Item Sets frequency of occurrence, obtain institute
There is the support frequency of Item Sets, it is all that the minimum Item Sets for supporting frequency of frequency >=setting is supported to form frequent item set LkCollection
Close, if the maximum the number of iterations of k < and for sky, execute k++, be transferred to step S202;Otherwise, terminate operation.
Further, the method that neural network model BP initial weight is generated described in step S3 is to appoint in following 4 kinds of methods
It anticipates one kind:
Method one:Randomly initial weight is selected between section [- 1,1];
Method two:Randomly initial weight is selected between the section [- 0.01,0.01] near zero;
Method three:There are two-level networks in neural network model BP:Input layer and hidden layer and between network, hidden layer
The initial weight of network between output layer, two-level network uses different selection modes:Input layer is to hidden layer connection weight
Value is initialized as random number, and the connection weight of hidden layer to output layer is initialized as -1 or 1;
Method four:By random number of the weight initialization between [a, b], wherein a, b are the integer for meeting following equation:
Wherein H is network node in hidden layer;
Step S4 specifically includes following steps:
Weight w between S401, adjustment neural network model BP hidden layer and output layerkj;
Adjust wkjPurpose be desirable to the new output o of output node j* pjO is exported than currentlypjCloser to target value tpj, fixed
Justice:
Wherein α represents the degree of approach, remains unchanged in each cycle of training, and becomes smaller with the adjustment of node in hidden layer H,
Do not consider threshold value, then has:
Wherein wkjAnd w* kjRespectively update the weight of front and back, ypkFor hidden layer output, Δ wkjFor wkjKnots modification;
Δ w is obtained according to formula (3)kjSolution equation:
Wherein,
Equation (4) are solved according to least square and error principle, obtain Δ wkjApproximate solution:
It is connected to the hidden layer node k of output node j to each, calculates the weight changes delta w between k and jkj, update
Weight simultaneously calculates error of sum square E, then in one optimal k of k ∈ [1, H] interval selection, so that E is minimum;
Weight v between S402, adjustment neural network model BP input layer and hidden layerik
Adjust vikPurpose be once neural network algorithm falls into local minimum point, it is minimum that modification weight can jump out this
Point, judge neural network algorithm fall into local minimum point condition be error E change rate Δ E=0, and E > 0;
Do not consider threshold value, the change of the weight of hidden layer node k is solved by following equation:
Wherein δpj=f-1(ypk+Δypk)-f-1(ypk), M is natural number, then hidden layer exports ypkSolution formula is:
Wherein Δ ypkFor ypkKnots modification, then have:
According to the matrix equation that least square and error principle solution formula (6) construct, can calculate:
Aggregative formula (6) and (10) calculate the consecutive mean knots modification of weight between hidden layer and output layer
Calculate the consecutive mean knots modification of weight between input layer and hidden layer
M takes the natural number between 10~20 in formula (12), obtains neural network model BP according to formula (11) and (12)
Consecutive mean weight, according to the consecutive mean weight of neural network model BP obtain dynamic neural network model DBP;
With the power plant of adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP and threshold in step S5
Value obtains prediction model AIGA-DBP and specifically includes following steps:
Antigen recognizing is carried out first, i.e. the identification of mean square error generates initial antibody (dynamic neural network model DBP
Weight and threshold value) after, carry out antibody coding, then calculating antibody fitness and antibody concentration are carried out according to the result of calculating
Adaptive antibody selection operation based on concentration obtains suitable weight and threshold value if this operation meets condition, and by result
Output;If this operation is unsatisfactory for condition, need to carry out adaptive crossover operation and TSP question operation, and carry out weight and threshold
The update of value, then antibody fitness and antibody concentration are recalculated, weight and threshold value until obtaining the condition that meets, last basis
Meet the weight and threshold value of condition, obtains prediction model AIGA-DBP.
Advantages of the present invention and good effect are:
The present invention provides expert advice module can provide more professional guiding opinion to staff, ensure power grid peace
That tests entirely is professional, reduces artificial caused accident occurrence probability;Meanwhile the present invention passes through load forecast module analysis
The characteristics of this cyclically-varying trend of electric load, the basic model using SIN function model as prediction data, with
The prior art is compared, and has the advantages that be bonded Power system load data actual change, in order to overcome using based on SIN function
The relatively large disadvantage of model error, the method that the present invention also uses repetitive learning improve the accuracy of prediction.
The present invention constructs big data analysis unit first, and then excavating total complete time with association rule algorithm influences
Factor, and neural network model BP is constructed, the weight and threshold value of neural network model BP are dynamically refined, to be moved
State neural network model DBP, then predicted with adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP
Model AIGA-DBP finally calculates Transmission Network of Power System safe prediction value with prediction model AIGA-DBP, according to electricity
Force system power transmission network safe prediction value can optimize production procedure, improve production efficiency.
Dynamic neural network model DBP in the present invention can adapt to various change caused by season.
(3) big data analysis technology has been used in the present invention, so that the digging of Transmission Network of Power System Safety Influence Factors
Dig it is highly efficient and accurate, Transmission Network of Power System Safety Influence Factors consider it is more comprehensive, effectively improve the accurate of prediction
Property.
Detailed description of the invention
Fig. 1 is Transmission Network of Power System security test system structure diagram provided in an embodiment of the present invention.
In figure:1, instruction input module is tested;2, initial parameter configuration module;3, central control module;4, it wirelessly communicates
Module;5, expert advice module;6, alarm module;7, display module;8, load forecast module.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing
Detailed description are as follows.
Structure of the invention is explained in detail with reference to the accompanying drawing.
As shown in Figure 1, Transmission Network of Power System security test system provided by the invention includes:Test instruction input
Module 1, initial parameter configuration module 2, central control module 3, wireless communication module 4, expert advice module 5, alarm module 6,
Display module 7, load forecast module 8.
Instruction input module 1 is tested, is connect with central control module 3, for input test instruction operation;
Initial parameter configuration module 2 is connect with central control module 3, for configuring the initial parameter of test environment;
Central control module 3, with test instruction input module 1, initial parameter configuration module 2, wireless communication module 4, specially
Family's suggestion module 5, alarm module 6, display module 7, load forecast module 8 connect, for dispatching the normal work of modules
Make;
Wireless communication module 4 is connect with central control module 3, for carrying out remote testing control by wireless transmitter
Operation;
Expert advice module 5 is connect with central control module 3, for carrying out suggestion to test operation by expert's net;
Alarm module 6 is connect with central control module 3, for being alarmed by alarm test failure;
Display module 7 is connect with central control module 3, for showing test information data;
Load forecast module 8 is connect with central control module 3, for predicting electric load.
Load forecast module 5 provided by the invention includes:Data acquisition module, study module, prediction module;
Data acquisition module acquires the historical data of electric load for block in chronological order;
Study module obtains basic sequence according to SIN function for preprocessing sequence fitting to be obtained a SIN function
Column;
Prediction module for obtaining forecasting sequence according to prediction SIN function model, and carries out anti-normalizing to forecasting sequence
Change processing, obtains prediction data sequence.
8 prediction technique of load forecast module provided by the invention is as follows:
Step 1, data acquisition module acquire the historical data of electric load in chronological order, obtain Load Time Series
Load Time Series Y is normalized in Y, obtains preprocessing sequence O;
Step 2, preprocessing sequence O fitting is obtained a SIN function S (t) by study module, according to SIN function S (t)
Obtain basic sequence L1,
Step 3, study module, which subtracts each other preprocessing sequence O and basic sequence L1, obtains base residual sequence e;
Step 4, study module find out the root-mean-square error of base residual sequence e, by the root-mean-square error and preset valve
Value comparison;
If being less than preset threshold, SIN function S (t) is prediction SIN function model W (t), and prediction module is according to prediction
SIN function model W (t) obtains forecasting sequence f, and carries out anti-normalization processing to forecasting sequence f, obtains prediction data sequence
F;
If more than preset threshold, then following steps (e) to (g) is continued to execute;
Step 5, study module are fitted base residual sequence e, residual error SIN function Q (t) are obtained, according to residual error
SIN function Q
Step 6 obtains residual sequence M;
Step 7, study module find out the root-mean-square error of residual sequence M, by the root-mean-square error and preset threshold values pair
Than;
If being less than preset threshold, SIN function S (t) and residual error SIN function Q (t) is superimposed, obtains the sinusoidal letter of prediction
Exponential model W (t), prediction module obtain forecasting sequence f according to prediction SIN function model W (t), and carry out to forecasting sequence f anti-
Normalized obtains prediction data sequence F;
If more than preset threshold, then enter repetitive learning step:
According to residual sequence Mi (i=1,2,3 ..., n), fitting obtain prediction residual SIN function Pi (t) (i=1,2,
3 ..., n), prediction residual sequence Ni is obtained according to Pi (t), using formula M+1=Mi-Ni, finds out next residual sequence Mi+
1, Pi+1 (t) is obtained according to Mi+1 fitting, prediction residual sequence Ni+1 is obtained according to Pi+1 (t);
After the prediction residual sequence Ni that each operation obtains, the prediction residual sequence Ni root-mean-square error is calculated, by this
Square error and preset threshold values compare, if being less than preset threshold, stop repetitive learning step, if more than preset threshold, then
Continue repetitive learning step;
Step 8, after stopping repetitive learning step, by SIN function S (t) and with residual error SIN function Q (t) and all pre-
Residual error SIN function Pi (t) superposition is surveyed, obtains prediction SIN function model W (t), prediction module is according to prediction SIN function model
W (t) obtains forecasting sequence f, and carries out anti-normalization processing to forecasting sequence f, obtains forecasting sequence F.
When the present invention tests, staff passes through test 1 input test of instruction input module instruction operation;Then, pass through
The initial parameter of the configuration test environment of initial parameter configuration module 2;Central control module 3 passes through channel radio for feedback information is tested
Letter module 4 is sent to staff, and staff obtains Test Suggestion by expert advice module 5;Pass through load forecast
Module 8 predicts electric load;If test is abnormal, alarmed by alarm module 6 test failure;Finally, logical
Cross the display test information data of display module 7.
Below with reference to concrete analysis, the invention will be further described.
The anti-normalization processing of prediction module includes:
S1, the big data based on Hadoop building inclusion relation type database data, sensing data and controller data
Analytical unit goes to step S2;
S2, under MapReduce frame use Apriori association rules mining algorithm, in big data analysis unit into
Row analysis and excavation, obtain Transmission Network of Power System Safety Influence Factors, go to step S3;
S3, in conjunction with Transmission Network of Power System Safety Influence Factors and Transmission Network of Power System security history data, structure
Neural network model BP is built, the initial weight of neural network model BP is generated, goes to step S4;
S4, the weight and threshold value of neural network model BP are dynamically refined, obtain dynamic neural network model DBP,
The weight and threshold value for generating dynamic neural network model DBP, go to step S5;
S5, with adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP, obtain prediction model
AIGA-DBP calculates Transmission Network of Power System safe prediction value according to prediction model AIGA-DBP, goes to step S6;
S6, judge Transmission Network of Power System safe prediction value and the error of Transmission Network of Power System safety expectation value is
The no condition for meeting setting, if so, going to step S7;Otherwise step S5 is re-executed;
S7, output power system power transmission network safe prediction value and forecasting sequence F terminate.
Step S1 specifically includes following steps:
Relevant database data, sensing data and controller data are uploaded to distributed field system by Sqoop
Unite HDFS, and stores into NoSQL database;Using MapReduce Computational frame to relevant database data, sensor number
Mining analysis is carried out according to controller data, NoSQL database is written into the data analyzed, and show by Web.
Following steps are specifically included with Apriori association rules mining algorithm under MapReduce frame in step S2:
S201, the set L of frequent 1 item collection is obtained using MapReduce computation module1, generate the set C of candidate's k item collectionk;
S202, in Map function processing stage, Transaction Information of each Map task computation handled by it concentrates each affairs
C is included in recordkIn Item Sets frequency of occurrence, for each Map task, if some of candidate's k item collection
Collection (including k project) appears in a transaction journal, then Map function is generated and exported<Some item collection, 1>Key-value pair is given
Combiner function gives Reduce function after being handled by Combiner function;
S203, in Reduce function processing stage, Reduce function adds up CkIn Item Sets frequency of occurrence, obtain institute
There is the support frequency of Item Sets, it is all that the minimum Item Sets for supporting frequency of frequency >=setting is supported to form frequent item set LkCollection
Close, if the maximum the number of iterations of k < and for sky, execute k++, be transferred to step S202;Otherwise, terminate operation.
The method that neural network model BP initial weight is generated described in step S3 is any one in following 4 kinds of methods:
Method one:Randomly initial weight is selected between section [- 1,1];
Method two:Randomly initial weight is selected between the section [- 0.01,0.01] near zero;
Method three:There are two-level networks in neural network model BP:Input layer and hidden layer and between network, hidden layer
The initial weight of network between output layer, two-level network uses different selection modes:Input layer is to hidden layer connection weight
Value is initialized as random number, and the connection weight of hidden layer to output layer is initialized as -1 or 1;
Method four:By random number of the weight initialization between [a, b], wherein a, b are the integer for meeting following equation:
Wherein H is network node in hidden layer;
Step S4 specifically includes following steps:
Weight w between S401, adjustment neural network model BP hidden layer and output layerkj;
Adjust wkjPurpose be desirable to the new output o of output node j* pjO is exported than currentlypjCloser to target value tpj, fixed
Justice:
Wherein α represents the degree of approach, remains unchanged in each cycle of training, and becomes smaller with the adjustment of node in hidden layer H,
Do not consider threshold value, then has:
Wherein wkjAnd w* kjRespectively update the weight of front and back, ypkFor hidden layer output, Δ wkjFor wkjKnots modification;
Δ w is obtained according to formula (3)kjSolution equation:
Wherein,
Equation (4) are solved according to least square and error principle, obtain Δ wkjApproximate solution:
It is connected to the hidden layer node k of output node j to each, calculates the weight changes delta w between k and jkj, update
Weight simultaneously calculates error of sum square E, then in one optimal k of k ∈ [1, H] interval selection, so that E is minimum;
Weight v between S402, adjustment neural network model BP input layer and hidden layerik;
Adjust vikPurpose be once neural network algorithm falls into local minimum point, it is minimum that modification weight can jump out this
Point, judge neural network algorithm fall into local minimum point condition be error E change rate Δ E=0, and E > 0;
Do not consider threshold value, the change of the weight of hidden layer node k is solved by following equation:
Wherein δpj=f-1(ypk+Δypk)-f-1(ypk), M is natural number, then hidden layer exports ypkSolution formula is:
Wherein Δ ypkFor ypkKnots modification, then have:
According to the matrix equation that least square and error principle solution formula (6) construct, can calculate:
Aggregative formula (6) and (10) calculate the consecutive mean knots modification of weight between hidden layer and output layer
Calculate the consecutive mean knots modification of weight between input layer and hidden layer
M takes the natural number between 10~20 in formula (12), obtains neural network model BP according to formula (11) and (12)
Consecutive mean weight, according to the consecutive mean weight of neural network model BP obtain dynamic neural network model DBP;
With the power plant of adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP and threshold in step S5
Value obtains prediction model AIGA-DBP and specifically includes following steps:
Antigen recognizing is carried out first, i.e. the identification of mean square error generates initial antibody (dynamic neural network model DBP
Weight and threshold value) after, carry out antibody coding, then calculating antibody fitness and antibody concentration are carried out according to the result of calculating
Adaptive antibody selection operation based on concentration obtains suitable weight and threshold value if this operation meets condition, and by result
Output;If this operation is unsatisfactory for condition, need to carry out adaptive crossover operation and TSP question operation, and carry out weight and threshold
The update of value, then antibody fitness and antibody concentration are recalculated, weight and threshold value until obtaining the condition that meets, last basis
Meet the weight and threshold value of condition, obtains prediction model AIGA-DBP.
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form,
Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to
In the range of technical solution of the present invention.
Claims (5)
1. a kind of Transmission Network of Power System security test system, which is characterized in that the Transmission Network of Power System safety
Property test macro includes:
Instruction input module is tested, is connect with central control module, for input test instruction operation;
Load forecast module, connect with central control module, for predicting electric load;
The load forecast module includes:Data acquisition module, study module, prediction module;
Data acquisition module acquires the historical data of electric load for block in chronological order;
Study module obtains basic sequence according to SIN function for preprocessing sequence fitting to be obtained a SIN function;
Prediction module for obtaining forecasting sequence according to prediction SIN function model, and carries out at renormalization forecasting sequence
Reason, obtains prediction data sequence;
The prediction technique of load forecast module is as follows:
Step 1, data acquisition module acquire the historical data of electric load in chronological order, obtain Load Time Series Y, right
Load Time Series Y is normalized, and obtains preprocessing sequence O;
Preprocessing sequence O fitting is obtained a SIN function S (t), is obtained according to SIN function S (t) by step 2, study module
Basic sequence L1,
Step 3, study module, which subtracts each other preprocessing sequence O and basic sequence L1, obtains base residual sequence e;
Step 4, study module find out the root-mean-square error of base residual sequence e, by the root-mean-square error and preset threshold values pair
Than;
If being less than preset threshold, SIN function S (t) is prediction SIN function model W (t), and prediction module is sinusoidal according to prediction
Function model W (t) obtains forecasting sequence f, and carries out anti-normalization processing to forecasting sequence f, obtains prediction data sequence F;
If more than preset threshold, then following steps five are continued to execute to step 7;
Step 5, study module are fitted base residual sequence e, obtain residual error SIN function Q (t), according to residual error sine
Function Q;
Step 6 obtains residual sequence M;
Step 7, study module find out the root-mean-square error of residual sequence M, and the root-mean-square error and preset threshold values are compared;
If being less than preset threshold, SIN function S (t) and residual error SIN function Q (t) is superimposed, obtains prediction SIN function mould
Type W (t), prediction module obtains forecasting sequence f according to prediction SIN function model W (t), and carries out anti-normalizing to forecasting sequence f
Change processing, obtains prediction data sequence F;
If more than preset threshold, then enter repetitive learning step:
According to residual sequence Mi (i=1,2,3 ..., n), fitting obtain prediction residual SIN function Pi (t) (i=1,2,3 ...,
N), prediction residual sequence Ni is obtained according to Pi (t), using formula M+1=Mi-Ni, finds out next residual sequence Mi+1, according to
Mi+1 fitting obtains Pi+1 (t), obtains prediction residual sequence Ni+1 according to Pi+1 (t);
After the prediction residual sequence Ni that each operation obtains, the prediction residual sequence Ni root-mean-square error is calculated, by the root mean square
Error and preset threshold values compare, if being less than preset threshold, stop repetitive learning step, if more than preset threshold, then continue
Repetitive learning step;
Step 8, stop repetitive learning step after, by SIN function S (t) and with residual error SIN function Q (t) and it is all predict it is residual
Poor SIN function Pi (t) superposition, obtains prediction SIN function model W (t), prediction module is according to prediction SIN function model W (t)
It obtains forecasting sequence f, and anti-normalization processing is carried out to forecasting sequence f, obtain forecasting sequence F;
The anti-normalization processing of prediction module includes:
S1, the big data analysis based on Hadoop building inclusion relation type database data, sensing data and controller data
Unit goes to step S2;
S2, Apriori association rules mining algorithm is used under MapReduce frame, divided in big data analysis unit
Analysis and excavation, obtain Transmission Network of Power System Safety Influence Factors, go to step S3;
S3, in conjunction with Transmission Network of Power System Safety Influence Factors and Transmission Network of Power System security history data, building mind
Through network model BP, the initial weight of neural network model BP is generated, step S4 is gone to;
S4, the weight and threshold value of neural network model BP are dynamically refined, obtain dynamic neural network model DBP, generated
The weight and threshold value of dynamic neural network model DBP, goes to step S5;
S5, with adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP, obtain prediction model AIGA-
DBP calculates Transmission Network of Power System safe prediction value according to prediction model AIGA-DBP, goes to step S6;
S6, judge whether the error of Transmission Network of Power System safe prediction value and Transmission Network of Power System safety expectation value is full
The condition set enough, if so, going to step S7;Otherwise step S5 is re-executed;
S7, output power system power transmission network safe prediction value and forecasting sequence F terminate.
2. Transmission Network of Power System security test system as described in claim 1, which is characterized in that step S1 is specifically included
Following steps:
Relevant database data, sensing data and controller data are uploaded to distributed file system by Sqoop
HDFS, and store into NoSQL database;Using MapReduce Computational frame to relevant database data, sensing data
Mining analysis is carried out with controller data, NoSQL database is written into the data analyzed, and show by Web.
3. Transmission Network of Power System security test system as described in claim 1, which is characterized in that in step S2
Following steps are specifically included with Apriori association rules mining algorithm under MapReduce frame:
S201, the set L of frequent 1 item collection is obtained using MapReduce computation module1, generate the set C of candidate's k item collectionk;
S202, in Map function processing stage, Transaction Information of each Map task computation handled by it concentrates each transaction journal
In be included in CkIn Item Sets frequency of occurrence, for each Map task, if some item collections (packet of candidate's k item collection
Containing k project) it appears in a transaction journal, then Map function generates and exports some item collection of <, and 1 > key-value pair is given
Combiner function gives Reduce function after being handled by Combiner function;
S203, in Reduce function processing stage, Reduce function adds up CkIn Item Sets frequency of occurrence, obtain all items
The support frequency of mesh collection, it is all that the minimum Item Sets for supporting frequency of frequency >=setting is supported to form frequent item set LkSet,
If the maximum the number of iterations of k < and for sky, execute k++, be transferred to step S202;Otherwise, terminate operation.
4. Transmission Network of Power System security test system as described in claim 1, which is characterized in that produced described in step S3
The method of raw neural network model BP initial weight is any one in following 4 kinds of methods:
Method one:Randomly initial weight is selected between section [- 1,1];
Method two:Randomly initial weight is selected between the section [- 0.01,0.01] near zero;
Method three:There are two-level networks in neural network model BP:Input layer and hidden layer and between network, hidden layer with it is defeated
The initial weight of network between layer out, two-level network uses different selection modes:At the beginning of input layer to hidden layer connection weight
Beginning turns to random number, and the connection weight of hidden layer to output layer is initialized as -1 or 1;
Method four:By random number of the weight initialization between [a, b], wherein a, b are the integer for meeting following equation:
Wherein H is network node in hidden layer;
Step S4 specifically includes following steps:
Weight w between S401, adjustment neural network model BP hidden layer and output layerkj;
Adjust wkjPurpose be desirable to the new output o of output node j* pjO is exported than currentlypjCloser to target value tpj, definition:
Wherein α represents the degree of approach, remains unchanged in each cycle of training, and becomes smaller with the adjustment of node in hidden layer H, does not examine
Consider threshold value, then has:
Wherein wkjAnd w* kjRespectively update the weight of front and back, ypkFor hidden layer output, Δ wkjFor wkiKnots modification;
Δ w is obtained according to formula (3)kjSolution equation:
Wherein,
Equation (4) are solved according to least square and error principle, obtain Δ wkjApproximate solution:
It is connected to the hidden layer node k of output node j to each, calculates the weight changes delta w between k and jkj, update weight
And error of sum square E is calculated, then in one optimal k of k ∈ [1, H] interval selection, so that E is minimum;
Weight v between S402, adjustment neural network model BP input layer and hidden layerik;
Adjust vikPurpose be once neural network algorithm falls into local minimum point, modification weight can jump out the minimal point, sentence
Disconnected neural network algorithm fall into local minimum point condition be error E change rate Δ E=0, and E > 0;
Do not consider threshold value, the change of the weight of hidden layer node k is solved by following equation:
Wherein δpj=f-1(ypk+Δypk)-f-1(ypk), M is natural number, then hidden layer exports ypkSolution formula is:
Wherein Δ ypkFor ypkKnots modification, then have:
According to the matrix equation that least square and error principle solution formula (6) construct, can calculate:
Aggregative formula (6) and (10) calculate the consecutive mean knots modification of weight between hidden layer and output layer
Calculate the consecutive mean knots modification of weight between input layer and hidden layer
M takes the natural number between 10~20 in formula (12), obtains the dynamic of neural network model BP according to formula (11) and (12)
State average weight obtains dynamic neural network model DBP according to the consecutive mean weight of neural network model BP;
With the power plant of adaptive immune genetic AIGA algorithm optimization dynamic neural network model DBP and threshold value in step S5, obtain
It obtains prediction model AIGA-DBP and specifically includes following steps:
Antigen recognizing is carried out first, i.e. the identification of mean square error generates the initial antibody (power of dynamic neural network model DBP
Value and threshold value) after, antibody coding is carried out, then calculating antibody fitness and antibody concentration, are based on according to the result of calculating
The adaptive antibody selection operation of concentration obtains suitable weight and threshold value, and result is defeated if this operation meets condition
Out;If this operation is unsatisfactory for condition, need to carry out adaptive crossover operation and TSP question operation, and carry out weight and threshold value
Update, then recalculate antibody fitness and antibody concentration, weight and threshold value until obtaining the condition that meets are last according to full
The weight and threshold value of sufficient condition obtain prediction model AIGA-DBP.
5. Transmission Network of Power System security test system as described in claim 1, which is characterized in that the electric system is defeated
Electric network security test system further includes:
Initial parameter configuration module, connect with central control module, for configuring the initial parameter of test environment;
Central control module, with test instruction input module, initial parameter configuration module, wireless communication module, expert advice mould
Block, alarm module, display module, the connection of load forecast module, work normally for dispatching modules;
Wireless communication module is connect with central control module, for carrying out remote testing control operation by wireless transmitter;
Expert advice module, connect with central control module, for carrying out suggestion to test operation by expert's net;
Alarm module is connect with central control module, for being alarmed by alarm test failure;
Display module is connect with central control module, for showing test information data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810591879.XA CN108898243A (en) | 2018-06-04 | 2018-06-04 | A kind of Transmission Network of Power System security test system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810591879.XA CN108898243A (en) | 2018-06-04 | 2018-06-04 | A kind of Transmission Network of Power System security test system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108898243A true CN108898243A (en) | 2018-11-27 |
Family
ID=64344524
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810591879.XA Pending CN108898243A (en) | 2018-06-04 | 2018-06-04 | A kind of Transmission Network of Power System security test system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108898243A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109636062A (en) * | 2018-12-25 | 2019-04-16 | 湖北工业大学 | A kind of students ' behavior analysis method and system based on big data analysis |
CN109840891A (en) * | 2019-01-07 | 2019-06-04 | 重庆工程学院 | A kind of intelligence strand tapered anchorage and prestressed monitoring method and detection system, terminal |
CN109881538A (en) * | 2019-02-18 | 2019-06-14 | 杭州西红柿环保科技有限公司 | A kind of enhanced method for paper pulp molding production |
CN110014656A (en) * | 2018-12-13 | 2019-07-16 | 闽南理工学院 | A kind of 3D printing personalization shoes, print control system and print control program |
CN111314801A (en) * | 2020-02-13 | 2020-06-19 | 中国铁道科学研究院集团有限公司铁道建筑研究所 | Data acquisition system and method supporting dynamic scheduling |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101766001A (en) * | 2007-06-05 | 2010-06-30 | 阿斯特里姆有限公司 | Remote testing system and method |
CN104636479A (en) * | 2015-02-15 | 2015-05-20 | 西安电子科技大学 | Industrial big data driven total completion time prediction method |
CN106056233A (en) * | 2016-05-11 | 2016-10-26 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Power load prediction method |
CN108040070A (en) * | 2017-12-29 | 2018-05-15 | 北京奇虎科技有限公司 | A kind of network security test platform and method |
-
2018
- 2018-06-04 CN CN201810591879.XA patent/CN108898243A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101766001A (en) * | 2007-06-05 | 2010-06-30 | 阿斯特里姆有限公司 | Remote testing system and method |
CN104636479A (en) * | 2015-02-15 | 2015-05-20 | 西安电子科技大学 | Industrial big data driven total completion time prediction method |
CN106056233A (en) * | 2016-05-11 | 2016-10-26 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Power load prediction method |
CN108040070A (en) * | 2017-12-29 | 2018-05-15 | 北京奇虎科技有限公司 | A kind of network security test platform and method |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110014656A (en) * | 2018-12-13 | 2019-07-16 | 闽南理工学院 | A kind of 3D printing personalization shoes, print control system and print control program |
CN109636062A (en) * | 2018-12-25 | 2019-04-16 | 湖北工业大学 | A kind of students ' behavior analysis method and system based on big data analysis |
CN109840891A (en) * | 2019-01-07 | 2019-06-04 | 重庆工程学院 | A kind of intelligence strand tapered anchorage and prestressed monitoring method and detection system, terminal |
CN109881538A (en) * | 2019-02-18 | 2019-06-14 | 杭州西红柿环保科技有限公司 | A kind of enhanced method for paper pulp molding production |
CN111314801A (en) * | 2020-02-13 | 2020-06-19 | 中国铁道科学研究院集团有限公司铁道建筑研究所 | Data acquisition system and method supporting dynamic scheduling |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108898243A (en) | A kind of Transmission Network of Power System security test system | |
CN107784397B (en) | Power grid material demand prediction system and prediction method thereof | |
CN104102212B (en) | Dispatching method, apparatus and system for gas and steam system in iron and steel enterprises | |
WO2021189739A1 (en) | Production plan-based iron and steel plant oxygen load prediction method | |
CN109885907A (en) | A kind of Satellite Attitude Control System health state evaluation and prediction technique based on cloud model | |
CN108491969A (en) | Spatial Load Forecasting model building method based on big data | |
CN107315884A (en) | A kind of building energy consumption modeling method based on linear regression | |
CN104574002B (en) | Asset management information processing method and processing device based on big data analysis | |
CN107239849A (en) | One kind is based on improved exponential smoothing gray model Methods of electric load forecasting | |
CN104636479A (en) | Industrial big data driven total completion time prediction method | |
Li et al. | An improved risk-benefit collaborative grey target decision model and its application in the decision making of load adjustment schemes | |
CN106199174A (en) | Extruder energy consumption predicting abnormality method based on transfer learning | |
CN107918368B (en) | The dynamic prediction method and equipment of iron and steel enterprise's coal gas yield and consumption | |
CN108921322A (en) | A kind of ultra-short term node load prediction technique and system based on Spot Price | |
CN109086930A (en) | A kind of user power utilization behavior analysis method based on electric power K line chart and depth network | |
CN105631528A (en) | NSGA-II and approximate dynamic programming-based multi-objective dynamic optimal power flow solving method | |
Tian et al. | Real-time shop floor scheduling method based on virtual queue adaptive control: Algorithm and experimental results | |
CN106505624A (en) | Determine regulator control system and the method for power distribution network distributed power source optimum ability to arrange jobs | |
CN101916999B (en) | 500kV regional control center control domain optimization system | |
Lao et al. | The optimized multivariate grey prediction model based on dynamic background value and its application | |
CN101540009B (en) | Method for predicting facility and equipment failure | |
CN114331041A (en) | Smart power plant maturity assessment method and system based on information physical system architecture | |
Singh et al. | Multi-objective stochastic heuristic methodology for tradespace exploration of a network centric system of systems | |
CN116011722A (en) | Large power grid-oriented man-machine cooperation regulation and control method, module and device | |
Shafiq et al. | Virtual engineering objects (VEO): designing, developing and testing models |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181127 |