CN116452041A - Power distribution network reliability assessment method based on cross entropy important sampling - Google Patents
Power distribution network reliability assessment method based on cross entropy important sampling Download PDFInfo
- Publication number
- CN116452041A CN116452041A CN202310340364.3A CN202310340364A CN116452041A CN 116452041 A CN116452041 A CN 116452041A CN 202310340364 A CN202310340364 A CN 202310340364A CN 116452041 A CN116452041 A CN 116452041A
- Authority
- CN
- China
- Prior art keywords
- reliability
- distribution network
- power distribution
- unavailability
- load
- 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
- 238000005070 sampling Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 title claims description 60
- 230000005540 biological transmission Effects 0.000 claims abstract description 38
- 238000011156 evaluation Methods 0.000 claims abstract description 30
- 230000003679 aging effect Effects 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 10
- 230000008439 repair process Effects 0.000 claims description 7
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 6
- 230000009471 action Effects 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010187 selection method Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000013433 optimization analysis Methods 0.000 description 2
- 208000025274 Lightning injury Diseases 0.000 description 1
- 238000012614 Monte-Carlo sampling Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Optimization (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Pure & Applied Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Analysis (AREA)
- Development Economics (AREA)
- Computational Mathematics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Evolutionary Biology (AREA)
- Public Health (AREA)
- Quality & Reliability (AREA)
- Probability & Statistics with Applications (AREA)
Abstract
The invention discloses a power distribution network reliability evaluation method based on cross entropy important sampling, which is characterized by comprising the following steps of: s1: considering the influence of the operation condition on the unavailability of the transmission line; s2: considering the influence of the aging effect on the unavailability of the transmission line; s3: combining the operation condition and the aging effect, and establishing a power transmission line unavailability model; s4: establishing a reliability index analysis model of the power distribution network; s5: solving coefficients of the power distribution network reliability index analysis model established in the step S4; s6: and taking the influence of new energy output and load uncertainty into consideration, training by using a BP neural network to obtain a reliability index analysis model coefficient of the power distribution network, and calculating the reliability index of the power distribution network. The reliability evaluation method of the power distribution network based on the cross entropy important sampling can improve the reliability evaluation speed and provide more reliable basis for the reliable and safe operation or planning of the power system.
Description
Technical Field
The invention belongs to the technical field of rapid reliability assessment of a power distribution network system, and particularly relates to a power distribution network reliability assessment method based on cross entropy important sampling.
Background
With the continuous development of social economy, the contradiction between the increase of energy demand and the shortage of energy is increasingly prominent. However, during the operation of the power system, the reliability of the element is often rapidly reduced due to the influence of sudden events such as extreme weather, etc., and thus a large-scale fault is caused.
Component faults caused by extreme weather are main factors for inducing energy supply interruption of a power system, the medium-long term reliability assessment is mainly based on statistical average value to describe the reliability of the component, the influence of the sudden factors on the reliability of the component is difficult to be described, and the guiding value for system operation and risk prevention and control is limited.
The reliability evaluation of the operation focuses on the reliability level of the system in the operation process, can sense and quantify the influence of the internal and external environment change on the reliability of the element, evaluates the reliability level of the system in real time and guides the development of a risk pre-control strategy. However, the power system has various elements, complex coupling relation, and the reliability evaluation process involves the problem of multi-energy flow optimization, so that the computational complexity is greatly increased. In addition, the time-varying characteristic of the element reliability parameter and the uncertainty caused by multiple types of loads further increase the calculation burden, so that how to effectively accelerate the operation reliability assessment of the power system is needed to be solved, and the design of the operation reliability assessment method of the power system meeting the aging requirement is a problem to be solved.
The reliability evaluation improvement algorithm proposed by the current research is difficult to process multidimensional uncertainty factors in the operation reliability evaluation of the power system, and the data driving method is high in acquisition cost of a training set due to high dimension of a required sample, so that the generalization capability of the model is difficult to ensure. Meanwhile, the element time-varying reliability model is simply exponential distribution, and the time-varying characteristics of the failure rate of the multiple types of elements in the power system are not fully considered.
Therefore, how to consider the component time-varying reliability model to perform rapid reliability evaluation on the power distribution system to obtain an accurate reliability evaluation result is a technical problem to be solved urgently by those skilled in the art at present.
Disclosure of Invention
In view of the above, the present invention aims to provide a method for evaluating the reliability of a power distribution network based on cross entropy important sampling, which is used for improving the speed of reliability evaluation and providing a more reliable basis for the reliable and safe operation or planning of a power system.
In order to achieve the above object, the present invention provides the following technical solutions:
the evaluation method of the reliability evaluation method of the power distribution network based on cross entropy important sampling comprises the following steps:
s1: considering the influence of the operation condition on the unavailability of the transmission line;
s2: considering the influence of the aging effect on the unavailability of the transmission line;
s3: combining the operation condition and the aging effect, and establishing a power transmission line unavailability model;
s4: establishing a reliability index analysis model of the power distribution network;
s5: solving coefficients of the power distribution network reliability index analysis model established in the step S4;
s6: and taking the influence of new energy output and load uncertainty into consideration, training by using a BP neural network to obtain a reliability index analysis model coefficient of the power distribution network, and calculating the reliability index of the power distribution network.
Further, in step S1, the influence of the operation condition on the power transmission line unavailability is considered to be specifically:
in FOR line Indicating the unavailability of the transmission line, I 0 And I trip Respectively setting a rated current value and a protection action setting value; FOR (FOR) en Referring to the probability of line failure caused by the external weather environment, parameters a and b are determined by historical data, and can be calculated by the following formula:
in the method, in the process of the invention,setting the maximum fault probability of the heavy load of the power transmission line by combining operation data and experience; />The method is characterized by comprising the steps of obtaining historical outage statistical data, wherein lambda and mu respectively represent the failure rate and the repair rate of the power transmission line, and the failure rate and the repair rate are the basic unavailability rate of the power transmission line.
Further, in step S2, the influence of the aging effect on the power transmission line unavailability is specifically:
let the service time of the line be T 0 Its unavailability in the i-th period can be expressed as:
wherein f (t) is a Weibull distribution function.
Further, in step S3, the operation conditions and the aging effect are combined, and the power transmission line unavailability model is established specifically as follows:
further, in step S4, the establishing a reliability index analysis model of the power distribution network includes the following steps:
s4.1: the load point reliability index can be expressed as:
in U i Annual average outage time, lambda, for load point i i Annual average outage rate for load point i; event(s)Indicating that the load point i is out of load, T is 8760h, T total Is the total analog time length, t i (s) is the duration of the system event that causes load point i to be out of load; i i (s) is a 0-1 indicating variable, if the load point I is not lost in the event preceding the system event s i (s) =1, otherwise I i (s)=0;
S4.2: based on the full probability formula and the conditional probability formula, transforming the reliability index formula of each load point in the step S4.1:
the key elements are assumed to have two states of working and failure, and faults among the elements are independent; for M key elements, the number of combined states is m=2 m The method comprises the steps of carrying out a first treatment on the surface of the Setting event F j (j=1, 2,., M) represents a set of system events with M key elements in the j-th combined state, obviously event F 1 -F M Forming a complete event group, namely mutually exclusive two by two and combining the complete event group into a complete set; combining the definition of the full probability formula and the conditional probability, the above formula is transformed as follows:
the above indicates that the key element is in the combined state F j At the time of an eventConditional probability of occurrence; x under the condition of unchanged system topology structure, electrical parameters, operation parameters and the like i j Depending on the reliability level of the non-critical elements of the system, it is assumed that the non-critical elements do not change during system operation, and therefore +.>Is a constant, called analytical model coefficient;
s4.3: deducing an analytical expression of the annual average outage rate index of the load points:
analyzing and expressing the reliability index of each load point as an explicit function of the reliability parameter of the key element, and when the reliability parameter of the key element changes, combining the corrected key element with the state occurrence probability P (F j ) Substituting the load point reliability index into the model to obtain the load point reliability index;
s4.4: calculating the reliability index of the power distribution network:
the SAIFI is the average power failure frequency of the system; SAIDI is the average outage time for the system and CAIDI is the average outage duration.
Further, in step S5, coefficients of the reliability index analysis model of the power distribution networkThe solving comprises the following steps:
s5.1: order theT total = 0 Assuming that the initial states of all elements are normal working states, generating a system state by adopting a sequential Monte Carlo method;
s5.2: let the current system state be s, the time length t corresponding to the system state s s Is determined by the state of the shortest duration in all the elements, let T be the element l total =T total +t s The method comprises the steps of carrying out a first treatment on the surface of the Judging whether each load point is out of load or not under the system event s, if the load point i is out of load, enabling t to be the same as the load point i i (s)=t s And find I i (s) matching s to set F based on the combined state of key elements in system event s j (j=1, 2,., M); if the system state s belongs to the set F j Order in principle
S5.3: judging the simulation time T total Whether or not it is smaller than the set simulation total time T set The method comprises the steps of carrying out a first treatment on the surface of the If yes, generating a next system state, and turning back to the step S5.2; otherwise, the analytical model coefficient is obtained by calculation according to the formula shown in the step S4.2, and the algorithm is terminated.
Further, step S6 includes the steps of:
s6.1: selecting a system key line according to the power transmission line unavailability model and the key element selection method established in the step S3;
s6.2: generating input data samples based on a grid search method by taking new energy output and load level as input characteristic quantities, establishing an analytical model for each input data sample, and obtaining analytical model coefficients corresponding to reliability indexes of each load point to serve as actual values of output data;
s6.3: building a BP neural network aiming at each load point, inputting a training sample into the neural network for learning, and obtaining a mapping relation between system operation parameters and analysis model coefficients;
s6.4: substituting the predicted value of the system load level and the new energy output in a future short period into a trained neural network model to obtain a corresponding analysis model coefficient; solving the unavailability rate of the key element, and substituting the unavailability rate into an analysis model to obtain the reliability index of each load point;
s6.5: and calculating and taking a system reliability index.
Compared with the prior art, the evaluation method for the reliability of the power distribution network based on the cross entropy important sampling has the following advantages:
the invention provides a power distribution network reliability evaluation method based on cross entropy important sampling, which comprises the steps of firstly, deducing a reliability index analysis calculation model considering component availability change based on full probability and conditional probability, and rapidly solving analysis model coefficients through a cross entropy important sampling method. Secondly, embedding the reliability index analysis model into the traditional reliability parameter optimization problem to realize analysis expression of the reliability parameter optimization model. The analysis model provided only needs to execute the reliability evaluation calculation based on the cross entropy important samples once in the solving process, thereby avoiding the defect that the reliability evaluation needs to be carried out for many times due to the change of the reliability parameters and greatly improving the solving efficiency of the reliability parameter optimization problem.
Drawings
Fig. 1 is a flowchart of a power distribution network reliability evaluation method based on cross entropy important sampling.
Fig. 2 is a topology diagram of an IEEE-RBTS-BUS4 system to which the present invention is applied.
Detailed Description
The invention is further described below with reference to the drawings and the detailed description.
In order to make the purposes, technical effects and technical solutions of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention are clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention; it will be apparent that the described embodiments are some of the embodiments of the present invention. Based on the present disclosure as examples, other embodiments that may be obtained by one of ordinary skill in the art without undue burden are within the scope of the present disclosure.
The invention provides a power distribution network reliability evaluation method based on cross entropy important sampling, which comprises the following steps:
s1: considering the influence of the operation condition on the unavailability of the transmission line;
s2: considering the influence of the aging effect on the unavailability of the transmission line;
s3: combining the operation condition and the aging effect, and establishing a power transmission line unavailability model;
s4: establishing a reliability index analysis model of the power distribution network;
s5: solving coefficients of the power distribution network reliability index analysis model established in the step S4;
s6: and taking the influence of new energy output and load uncertainty into consideration, training by using a BP neural network to obtain a reliability index analysis model coefficient of the power distribution network, and calculating the reliability index of the power distribution network.
In step S1, the influence of the operation condition on the power transmission line unavailability is specifically considered as follows: as the load current of the line increases, the heating value of the power transmission line increases, and the mechanical strength of the power transmission line gradually decreases, so that the line unavailability rate increases.
In FOR line Indicating the unavailability of the transmission line, I 0 And I trip Respectively setting a rated current value and a protection action setting value; FOR (FOR) en The method refers to the probability of line fault caused by external weather environment, and in operation evaluation, corresponding values are required to be selected according to the real-time external environment, and the method classifies weather into normal weather, severe weather (rain and snow, lightning stroke, strong wind and the like) and extremely severe weather (earthquake, flood, typhoon and the like), wherein FOR in each class en The temporary hold is constant.
Parameters a and b are determined from historical data and can be calculated using the following equation:
in the method, in the process of the invention,the maximum fault probability for the heavy load of the power transmission line can be set by an expert in combination with operation data and experience; />The method is characterized by comprising the steps of obtaining historical outage statistical data, wherein lambda and mu respectively represent the failure rate and the repair rate of the power transmission line, and the failure rate and the repair rate are the basic unavailability rate of the power transmission line.
In step S2, the influence of the aging effect on the power transmission line unavailability is specifically:
let the service time of the line be T 0 Its unavailability in the i-th period can be expressed as:
wherein f (t) is a Weibull distribution function.
In step S3, the operation conditions and the aging effect are combined, and the power transmission line unavailability model is established specifically as follows:
step S4 is to establish a distribution network reliability index analysis model, and firstly, based on a full probability formula and a conditional probability formula, an analysis expression of the reliability index of the load point relative to the reliability parameters of the key elements is deduced, so that the distribution network reliability index analysis model is obtained. The method specifically comprises the following steps:
s4.1: the load point reliability index can be expressed as:
in U i Annual average outage time, lambda, for load point i i Annual average outage rate for load point i; event(s)Indicating that the load point i is out of load, T is 8760h, T total Is the total analog time length, t i (s) is the duration of the system event that causes load point i to be out of load; i i (s) is an indicator variable of 0 or 1, if the load point I is not lost in the previous event of the system event s i (s) =1, otherwise I i (s)=0。
S4.2: based on the full probability formula and the conditional probability formula, transforming the reliability index formula of each load point in the step S4.1:
the key elements are assumed to have two states of working and failure, and faults among the elements are independent; for M key elements, the number of combined states is m=2 m The method comprises the steps of carrying out a first treatment on the surface of the Setting event F j (j=1, 2,., M) represents a set of system events with M key elements in the j-th combined state, obviously event F 1 -F M Forming a complete event group, namely mutually exclusive two by two and combining the complete event group into a complete set; combining the definition of the full probability formula and the conditional probability, the above formula is transformed as follows:
the above indicates that the key element is in the combined state F j At the time of an eventConditional probability of occurrence; x under the condition of unchanged system topology structure, electrical parameters, operation parameters and the like i j Depending on the level of reliability of non-critical elements of the system. Since the invention assumes that non-critical elements do not change during system operation, there is +.>Is a constant and is called an analytical model coefficient.
S4.3: deducing an analytical expression of the annual average outage rate index of the load points:
analyzing and expressing the reliability index of each load point as an explicit function of the reliability parameter of the key element, and when the reliability parameter of the key element changes, combining the corrected key element with the state occurrence probability P (F j ) Substituting the load point reliability index into the model to obtain the load point reliability index.
S4.4: calculating the reliability index of the power distribution network:
the SAIFI is the average power failure frequency of the system; SAIDI is the average outage time for the system and CAIDI is the average outage duration.
In step S5, coefficients of the reliability index analysis model of the power distribution networkThe solving comprises the following steps:
s5.1: order theT total =0, assuming that all the initial states of the elements are normal working states, generating a system state by adopting a sequential monte carlo method;
s5.2: let the current system state be s, the time length t corresponding to the system state s s Is determined by the state of the shortest duration in all the elements, let T be the element l total =T total +t s The method comprises the steps of carrying out a first treatment on the surface of the Judging whether each load point is out of load or not under the system event s, if the load point i is out of load, enabling t to be the same as the load point i i (s)=t s And find I i (s) matching s to set F based on the combined state of key elements in system event s j (j=1, 2,., M); if the system state s belongs to the set F j Order in principle
S5.3: judging the simulation time T total Whether or not it is smaller than the set simulation total time T set The method comprises the steps of carrying out a first treatment on the surface of the If yes, generating a next system state, and turning back to the step S5.2; otherwise, the analytical model coefficient is obtained by calculation according to the formula shown in the step S4.2, and the algorithm is terminated.
As can be seen from the above flow, the analytical model coefficient obtaining process can be embedded into a sequential Monte Carlo simulation algorithm, and after one complete reliability evaluation is performed, the analytical model coefficient can be obtained.
As can be seen from steps S4 and S5, the analysis expression of the reliability index of the power distribution network constructed by the present invention is only applicable when the system operation parameters (such as new energy output, load level, etc.) are not changed, and when the system operation parameters are changed, the analysis model needs to be reconstructed. The influence of the load level and the new energy output level on the analysis model coefficient is difficult to analyze and express, the influence of the system load level and the new energy output on the analysis model coefficient is described through a data driving method, and a power distribution network operation reliability assessment model-data hybrid driving method is established.
BP neural network is one of the most widely used neural network models, and comprises an input layer, a hidden layer and an output layer. The input layer to hidden layer feature extraction process can be expressed as:
h 1 =σ 1 (W 1 x+b 1 )
wherein: h is a 1 Hiding the feature vector of the layer for the first layer; sigma (sigma) 1 Is an activation function; w (W) 1 A weight coefficient matrix for the first hidden layer; x refers to input characteristic quantity, and the output force of each fan and the load level of each load point are adopted in the invention; b 1 Is a hidden layer bias vector.
Activation function sigma 1 A sigmoid function is generally used, as shown in the following formula.
Assuming that there are X hidden layers in common, the analytical model coefficients can be expressed as:
the error between the predicted value and the actual value of the analytical model coefficient can be expressed as
The error of the n-th hidden layer obtained by reverse derivation is
The weight coefficient matrix of each layer after error update is as follows
The method is one iteration in the BP neural network training process, and the accurate training model can be obtained by adjusting the learning parameters through multiple iterations, and the analysis model coefficient can be quickly and accurately obtained when the system operation parameters are changed.
Further, step S6 includes the steps of:
s6.1: selecting a system key line according to the power transmission line unavailability model and the key element selection method established in the step S3;
s6.2: generating input data samples based on a grid search method by taking new energy output and load level as input characteristic quantities, establishing an analytical model for each input data sample, and obtaining analytical model coefficients corresponding to reliability indexes of each load point to serve as actual values of output data;
s6.3: building a BP neural network aiming at each load point, inputting a training sample into the neural network for learning, and obtaining a mapping relation between system operation parameters and analysis model coefficients;
s6.4: substituting the predicted value of the system load level and the new energy output in a future short period into a trained neural network model to obtain a corresponding analysis model coefficient; solving the unavailability rate of the key element, and substituting the unavailability rate into an analysis model to obtain the reliability index of each load point;
s6.5: and (5) solving a system reliability index.
Specific examples:
the component reliability parameter optimization analysis model based on the cross entropy important sampling method provided by the invention is applied to an RBTS system, and the specific implementation process is as follows:
first, an RBTS system is obtained, including component failure rate and component repair time. The system is improved to a certain extent, and the key element set is selected as S. Two fans are additionally arranged at the 11 nodes and the 75 nodes, the capacity of a single fan is 1MW, and the forced outage rate is 0.05.
And secondly, obtaining a reliability evaluation analysis model by utilizing the steps, and testing the accuracy of the reliability evaluation analysis model. From the foregoing, it can be seen that critical component unavailability varies as system operating conditions vary, and that non-critical component unavailability may be approximately considered constant during system operation. The present invention therefore assumes that the unavailability of the remaining elements other than the set of critical elements S does not change during system operation.
And when the system load level is 1p.u and the fan output is 0.5p.u, establishing an analytical model. The reliability parameters of key elements are changed and substituted into an analytical model to obtain the reliability index of the power distribution network, and the reliability index is compared with index results obtained by the traditional sequential Monte Carlo method, so that the accuracy of the analytical modeling method is demonstrated. Taking the result obtained by the sequential MCS as a reliability index true value, and taking the reliability index true value as a reference for error calculation, defining a calculation formula of the relative error delta as follows:
wherein X is AN And X is SMCS And respectively calculating reliability indexes of the analysis model and the sequential MCS. The case parameter settings are shown in table 1 and the calculation results are shown in table 2.
Table 1 case parameter settings
TABLE 2 off-stream Rate index for certain load points
As can be seen from tables 1-2, under different parameter variation amplitudes, compared with a standard sequential Monte Carlo sampling method, the reliability evaluation analysis model of the power distribution network built by the invention has the advantage that the calculation error of the reliability index of the load point is not more than 2%. Meanwhile, the relative error results of each case in tables 1-2 also show that the closer the critical element unavailability is to the numerical value used for modeling (the original unavailability is the invention), the higher the analytical model accuracy is. In table 3, the relative error of the CAIDI index is significantly higher than the SAIDI and SAIDI indexes, because the calculation error of the annual average failure rate and annual average outage time of the load point may be accumulated in the calculation process of the CAIDI index, resulting in a phenomenon that the error of the CAIDI index is larger. As can be seen from Table 3, the maximum error of the CAIDI index is not more than 3%.
TABLE 3 off-stream time index for certain load points
/>
Table 4 reliability index of distribution network
/>
And thirdly, taking the evaluation result of the operation reliability of the power distribution network with the uncertainty of the source load into consideration. Setting the variation range of the system load level as [0.1,1], and the variation range of the fan wind power output as [0.3,1], taking 0.02 as step length, generating 1575 source load data combinations as input data samples by adopting a grid search method, and calculating an analytic model coefficient as an output data sample by the method. And offline training the BP neural network model by adopting 1575 training sample sets to obtain a data driving model of analysis model coefficients related to the system load level and the fan output. Setting three groups of scenes under different source load levels, substituting the system load and the fan output level in each scene into a trained BP neural network model to obtain corresponding analysis model coefficients, calculating the unavailability rate of key elements, and substituting the analysis model to obtain the operation reliability evaluation result of the power distribution network, wherein the evaluation result is shown in table 5.
Comparison of the hybrid driving method and the calculation result of the conventional sequential Monte Carlo method provided in Table 5
As can be seen from the table, compared with the traditional method, the algorithm provided by the invention has great advantages in calculation speed, because the traditional method needs to repeatedly evaluate the reliability of the system when the element state probability and the operation parameters change, and the calculation time is long, so that the requirement of real-time evaluation cannot be met. The method provided by the invention can quickly obtain the reliability index of the power distribution network by establishing a hybrid drive model offline and inputting real-time data such as load, fan output, component unavailability and the like online, and the calculation error is less than 3%. The modeling time of the hybrid drive model is mainly consumed in training sample generation, and this process can be performed offline.
The invention provides an element reliability parameter optimization analysis model based on a cross entropy important sampling method. Firstly, a reliability index analysis calculation model considering the change of the availability of the element is deduced based on conditional probability, and analysis model coefficients are rapidly obtained through a cross entropy important sampling method. Secondly, embedding the reliability index analysis model into the traditional reliability parameter optimization problem to realize analysis expression of the reliability parameter optimization model. The analysis model provided only needs to execute the reliability evaluation calculation based on the cross entropy important samples once in the solving process, thereby avoiding the defect that the reliability evaluation needs to be carried out for many times due to the change of the reliability parameters and greatly improving the solving efficiency of the reliability parameter optimization problem. The calculation result shows that the model provided by the invention can quickly solve the reliability of the system on the basis of uncertainty of the element model and uncertainty of new energy load, does not lose precision, and provides more reliable basis for quick and reliable evaluation and safe operation or planning of the power system.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. The power distribution network reliability evaluation method based on cross entropy important sampling is characterized by comprising the following steps of:
s1: considering the influence of the operation condition on the unavailability of the transmission line;
s2: considering the influence of the aging effect on the unavailability of the transmission line;
s3: combining the operation condition and the aging effect, and establishing a power transmission line unavailability model;
s4: establishing a reliability index analysis model of the power distribution network;
s5: solving coefficients of the power distribution network reliability index analysis model established in the step S4;
s6: and taking the influence of new energy output and load uncertainty into consideration, training by using a BP neural network to obtain a reliability index analysis model coefficient of the power distribution network, and calculating the reliability index of the power distribution network.
2. The method for evaluating the reliability of a power distribution network based on cross entropy important sampling according to claim 1, wherein the consideration of the influence of the operation condition on the power transmission line unavailability in step S1 is specifically as follows:
in FOR line Indicating the unavailability of the transmission line, I 0 And I trip Respectively setting a rated current value and a protection action setting value; FOR (FOR) en Referring to the probability of line failure caused by the external weather environment, parameters a and b are determined by historical data, and can be calculated by the following formula:
in the method, in the process of the invention,setting the maximum fault probability of the heavy load of the power transmission line by combining operation data and experience; />The method is characterized by comprising the steps of obtaining historical outage statistical data, wherein lambda and mu respectively represent the failure rate and the repair rate of the power transmission line, and the failure rate and the repair rate are the basic unavailability rate of the power transmission line.
3. The method for evaluating reliability of a power distribution network based on cross entropy significant sampling according to claim 2, wherein considering the influence of the aging effect on the power transmission line unavailability in step S2 is specifically as follows:
let the service time of the line be T 0 Its unavailability in the i-th period can be expressed as:
wherein f (t) is a Weibull distribution function.
4. The method for evaluating the reliability of a power distribution network based on cross entropy important sampling according to claim 3, wherein in the step S3, the operation conditions and the aging effect are combined, and the power transmission line unavailability model is established specifically as follows:
5. the method for evaluating the reliability of a power distribution network based on cross entropy important sampling according to claim 4, wherein in step S4, establishing a power distribution network reliability index analysis model comprises the following steps:
s4.1: the load point reliability index can be expressed as:
in U i Annual average outage time, lambda, for load point i i Annual average outage rate for load point i; event(s)Indicating that the load point i is out of load, T is 8760h, T total Is the total analog time length, t i (s) is the duration of the system event that causes load point i to be out of load; i i (s) is an indicator variable of 0 or 1, if the load point I is not lost in the previous event of the system event s i (s) =1, otherwise I i (s)=0;
S4.2: based on the full probability formula and the conditional probability formula, transforming the reliability index formula of each load point in the step S4.1:
the key elements are assumed to have two states of working and failure, and faults among the elements are independent; for M key elements, the number of combined states is m=2 m The method comprises the steps of carrying out a first treatment on the surface of the Setting event F j (j=1, 2,., M) represents a set of system events with M key elements in the j-th combined state, obviously event F 1 -F M Forming a complete event group, namely mutually exclusive two by two and combining the complete event group into a complete set; combining the definition of the full probability formula and the conditional probability, the above formula is transformed as follows:
the above indicates that the key element is in the combined state F j At the time of an eventConditional probability of occurrence; in the case of unchanged system topology, electrical and operational parameters, etc., the +.>Depending on the reliability level of the non-critical elements of the system, it is assumed that the non-critical elements do not change during system operation, and therefore +.>Is a constant, called analytical model coefficient;
s4.3: deducing an analytical expression of the annual average outage rate index of the load points:
λ i =T×(Y i 1 P(F 1 )+Y i 2 P(F 2 )+…+Y i M P(F M ))
analyzing and expressing the reliability index of each load point as an explicit function of the reliability parameter of the key element, and when the reliability parameter of the key element changes, combining the corrected key element with the state occurrence probability P (F j ) Substituting the load point reliability index into the model to obtain the load point reliability index;
s4.4: calculating the reliability index of the power distribution network:
the SAIFI is the average power failure frequency of the system; SAIDI is the average outage time for the system and CAIDI is the average outage duration.
6. The method for evaluating the reliability of a power distribution network based on cross entropy significant sampling according to claim 5, wherein in step S5, the coefficients of the power distribution network reliability index analysis model are calculatedY i j The solving comprises the following steps:
s5.1: order theT total =0, assuming that all the initial states of the elements are normal working states, generating a system state by adopting a sequential monte carlo method;
s5.2: let the current system state be s, the time length t corresponding to the system state s s Is determined by the state of the shortest duration in all the elements, let T be the element l total =T total +t s The method comprises the steps of carrying out a first treatment on the surface of the Judging whether each load point is out of load or not under the system event s, if the load point i is out of load, enabling t to be the same as the load point i i (s)=t s And find I i (s) matching s to set F based on the combined state of key elements in system event s j (j=1, 2,., M); if the system state s belongs to the set F j Order in principle
S5.3: judging the simulation time T total Whether or not it is smaller than the set simulation total time T set The method comprises the steps of carrying out a first treatment on the surface of the If yes, generating a next system state, and turning back to the step S5.2; otherwise, the analytical model coefficient is obtained by calculation according to the formula shown in the step S4.2, and the algorithm is terminated.
7. The method for evaluating the reliability of a power distribution network based on cross entropy significant sampling according to claim 6, wherein step S6 comprises the steps of:
s6.1: selecting a system key line according to the power transmission line unavailability model and the key element selection method established in the step S3;
s6.2: generating input data samples based on a grid search method by taking new energy output and load level as input characteristic quantities, establishing an analytical model for each input data sample, and obtaining analytical model coefficients corresponding to reliability indexes of each load point to serve as actual values of output data;
s6.3: building a BP neural network aiming at each load point, inputting a training sample into the neural network for learning, and obtaining a mapping relation between system operation parameters and analysis model coefficients;
s6.4: substituting the predicted value of the system load level and the new energy output in a future short period into a trained neural network model to obtain a corresponding analysis model coefficient; solving the unavailability rate of the key element, and substituting the unavailability rate into an analysis model to obtain the reliability index of each load point;
s6.5: and (5) solving a system reliability index.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310340364.3A CN116452041A (en) | 2023-03-31 | 2023-03-31 | Power distribution network reliability assessment method based on cross entropy important sampling |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310340364.3A CN116452041A (en) | 2023-03-31 | 2023-03-31 | Power distribution network reliability assessment method based on cross entropy important sampling |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116452041A true CN116452041A (en) | 2023-07-18 |
Family
ID=87131406
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310340364.3A Pending CN116452041A (en) | 2023-03-31 | 2023-03-31 | Power distribution network reliability assessment method based on cross entropy important sampling |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116452041A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117094190A (en) * | 2023-08-03 | 2023-11-21 | 南京航空航天大学 | Combined beam structure importance analysis method based on symmetrical entropy |
-
2023
- 2023-03-31 CN CN202310340364.3A patent/CN116452041A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117094190A (en) * | 2023-08-03 | 2023-11-21 | 南京航空航天大学 | Combined beam structure importance analysis method based on symmetrical entropy |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108876054B (en) | Short-term power load prediction method based on improved genetic algorithm optimization extreme learning machine | |
CN112990556A (en) | User power consumption prediction method based on Prophet-LSTM model | |
CN110929927A (en) | Artificial intelligence prediction model construction method applying power distribution network universe big data | |
CN103971175B (en) | Short-term load prediction method of multistage substations | |
CN111967777B (en) | Comprehensive evaluation method for energy storage and clean energy utilization value promotion | |
CN104573857A (en) | Power grid load rate prediction method based on intelligent algorithm optimization and combination | |
CN110837915A (en) | Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning | |
CN116452041A (en) | Power distribution network reliability assessment method based on cross entropy important sampling | |
Zhang et al. | Research on AGC performance during wind power ramping based on deep reinforcement learning | |
CN114742283A (en) | Medium-voltage line loop closing current estimation and risk assessment method based on deep learning network | |
CN113988655A (en) | Power transmission line running state evaluation method considering multiple meteorological factors | |
CN111900713A (en) | Multi-scene power transmission network planning method considering load and wind power randomness under network source coordination | |
CN115130378A (en) | Monte Carlo algorithm-based power distribution network toughness evaluation method under typhoon disaster | |
CN104834975A (en) | Power network load factor prediction method based on intelligent algorithm optimization combination | |
CN111952962A (en) | Power distribution network low voltage prediction method based on T-S fuzzy neural network | |
CN111784020A (en) | Service life prediction method for intelligent substation relay protection device | |
Sun et al. | Hybrid reinforcement learning for power transmission network self-healing considering wind power | |
CN106548284A (en) | A kind of adaptive mode massing power grid security Alarm Assessment method towards operation regulation and control | |
Xu et al. | Short-term wind speed prediction based on GRU | |
CN108110756A (en) | Consider the industrial park distribution network planning method of uncertain factor | |
CN109412152B (en) | Power grid loss calculation method based on deep learning and elastic network regularization | |
Fan et al. | Ultra-short-term bus load forecasting method based on multi-source data and hybrid neural network | |
CN114759579A (en) | Power grid active power optimization control system, method and medium based on data driving | |
CN115330085A (en) | Wind speed prediction method based on deep neural network and without future information leakage | |
CN114707121A (en) | Carbon emission calculation method for distributed power supply random output power distribution network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |