CN113971259B

CN113971259B - Error reliability parameter identification and correction method for power generation and transmission system element

Info

Publication number: CN113971259B
Application number: CN202111234908.5A
Authority: CN
Inventors: 胡博; 谢开贵; 李春燕; 邵常政; 曹侃; 周鲲鹏; 彭吕斌; 林铖嵘; 卢慧; 李维展
Original assignee: Chongqing University; State Grid Corp of China SGCC; Electric Power Research Institute of State Grid Hubei Electric Power Co Ltd
Current assignee: Chongqing University; State Grid Corp of China SGCC; Electric Power Research Institute of State Grid Hubei Electric Power Co Ltd
Filing date: 2021-10-22
Publication date: 2024-07-12
Anticipated expiration: 2041-10-22

Abstract

The application discloses a method for identifying and correcting error reliability parameters of elements of a power generation and transmission system, which comprises the following steps: the error reliability parameter identification and correction is constructed as a nonlinear optimization problem with the deviation between the reliability index estimated value and the statistic value minimized as an objective function, and whether the error parameter exists in the reliability parameters of the elements of the power transmission and reception system is judged; simplifying the reliability parameters of the power generation and transmission system; dividing the reliability parameter into a sensitive parameter and a non-sensitive parameter according to the sensitivity; and estimating the unit parameters connected with the load nodes, carrying out sensitive parameter estimation and identification, estimating the whole parameters, identifying the error parameters and correcting the error parameters by adopting an interval algorithm. The reliability evaluation inverse problem of the power generation and transmission system is considered, and the wrong element reliability parameters can be identified and corrected from the reliability index of the power generation and transmission system.

Description

Error reliability parameter identification and correction method for power generation and transmission system element

Technical Field

The invention belongs to the technical field of reliability evaluation of power systems, and relates to a method for identifying and correcting error reliability parameters of elements of a power generation and transmission system.

Background

The inverse problem of power system reliability refers to the use of reliability indicators of the entire system (or node) to identify and correct erroneous component reliability parameters.

In power system reliability assessment, the accuracy of component reliability parameters has a significant impact on system reliability assessment, and inaccurate reliability parameters may lead to diametrically opposed assessment results. The reliability parameters of the components are emphasized when the components of the power system are marked, and the parameters are often derived from historical reliability statistics of the similar components. The erroneous historical data can mislead the bidding result, and influence the reliability and safety of the future power system.

Component reliability parameters are often obtained from historical power outage records, but these data are of poor quality. On the one hand, this is because the utility company needs to count huge data, and there is inevitably error and unavailable data; on the other hand, the automatic data collection equipment on the element side is not fully put into use yet, and it is difficult to fully record erroneous data. The wrong component reliability parameters are mixed in a large number of component reliability parameters, and it is necessary to perform identification and correction.

Currently, research on the inverse problem of system reliability is mainly focused on power distribution systems, and no related research on power transmission systems exists. Therefore, how to identify and correct the wrong reliability parameters in the power generation and transmission system and ensure the reliability and the safety of the future power system is a technical problem to be solved by the present technicians in the field.

With the increasing popularity of intelligent detection devices (such as PMUs and smart meters), reliability indexes of the whole power system or nodes are more easily obtained, and it is possible to invert element reliability parameters from the reliability indexes of the system (or nodes). Based on this, reliability inverse problem techniques have developed.

Disclosure of Invention

In order to solve the defects in the prior art, the application provides a method for identifying and correcting error reliability parameters of a power generation and transmission system element.

In order to achieve the above object, the present invention adopts the following technical scheme:

A method for identifying and correcting error reliability parameters of a power generation and transmission system element comprises the following steps:

step 1: the error reliability parameter identification and correction is constructed as a nonlinear optimization problem with the deviation between the reliability index estimated value and the statistic value minimized as an objective function, whether the error parameter exists in the reliability parameters of the elements of the power transmission and detection system is judged, if so, the step 2 is entered, and otherwise, the error parameter and the correction value are output;

step 2: simplifying the reliability parameters of the power generation and transmission system;

Step 3: calculating the sensitivity of the reliability parameter according to a multi-measure quantization method, and dividing the reliability parameter into a sensitive parameter and a non-sensitive parameter according to the sensitivity;

Step 4: estimating unit parameters connected with the load nodes to obtain a suspected error unit parameter set S _GECP and an estimated value of the parameters;

Step 5: based on the estimation of S _GECP and parameters in the S _GECP, performing sensitive parameter estimation and identification by adopting a global optimization algorithm to obtain a suspected error parameter set S _ECP and the estimation of parameters in the S _ECP;

Step 6: based on the estimation of S _ECP and parameters therein, estimating the whole parameters by adopting a CIVR-LSE algorithm, and optimizing a suspected error parameter set S _ECP;

Step 7: and identifying the error parameters in S _ECP according to the deviation of the parameter statistic value and the estimated value, and correcting the error parameters by adopting an interval algorithm.

The invention further comprises the following preferable schemes:

Preferably, in step 1, a system of function equations of the reliability index with respect to the element reliability parameter is first constructed, where the system of function equations of the reliability index with respect to the element reliability parameter is:

Wherein N _C is the number of elements, and N _id is the number of known reliability indexes; Is a known system or node reliability index statistic, e ₁,e₂,……,e_Nid element reliability parameter to each reliability index mapping relation, Is a component failure rate parameter that is a function of the failure rate,Is a component repair rate parameter;

Then based on equation (1), the false reliability parameter identification and correction is constructed as a nonlinear optimization problem with the reliability index estimation and statistics deviation minimized as an objective function:

s.t.x^LB≤x≤x^UB

wherein x is the parameter vector to be solved in the formula (1); x ^LB and x ^UB are the reliability parameter value ranges obtained from engineering experience respectively.

Preferably, in step 1, reliability evaluation is performed based on the parameter statistics, and a deviation percentage gamma _i,i∈S_B of the reliability index evaluation value and the statistics thereof is calculated;

if any node i exists, enabling gamma _i to be larger than a threshold gamma ⁰, marking the node of the power transmission and generation system, and turning to the step 2; otherwise, outputting error parameters and sum correction values;

s _B is a total node set of the power generation and transmission system;

preferably, step 2 specifically comprises:

if no injection power exists on the node bus between the two power transmission lines connected in series, the two lines are regarded as an equivalent line;

And the reliability parameters of the same-model units installed on the same node or the same-model parallel transmission lines in the same transmission corridor are regarded as the same, and the same variables to be solved are used for representing the same.

Preferably, step 3 specifically includes:

Step 3.1: the reliability parameter sensitivity was quantified using the following three measures:

Measure one: quantifying sensitivity to reliability parameters based on a system reliability index measure;

and (2) measuring: quantifying the sensitivity of the reliability parameter based on an objective function f (x) measure that minimizes the deviation of the reliability index estimate from the statistic;

And (3) measuring: quantifying reliability parameter sensitivity based on an apportionment ratio measure of the element to the system reliability index;

Step 3.2: based on the reliability parameter sensitivity of the three measure quantification, the parameters are clustered, and the reliability parameters are divided into sensitive parameters and non-sensitive parameters.

Preferably, in step 3.1, the sensitivity is represented by measuring a partial derivative of a system reliability index with respect to a reliability parameter of a certain element, and the larger the partial derivative is, the more sensitive the system reliability index is to the change of the reliability parameter of the element, and the more easily the reliability of the whole system is improved;

The measure II changes the repair rate mu _j or the failure rate lambda _j of the element j by a small amount E, calculates the increment of the related reliability index based on f (x), and the larger the increment of the reliability index is, the more sensitive the reliability index is to the change of the reliability parameter;

the measure three-order element distributes the allocation of the system reliability index proportionally, wherein the unreliability of the system is borne by the fault element, and the normal operation element does not participate in the allocation of the system reliability index.

Preferably, in step 3.2, K-Means clustering is adopted, the reliability parameters are respectively clustered into two groups according to the quantized results of the three measures of sensitivity, and the final sensitive parameter set is the union set of sensitive parameter sets corresponding to the measures.

Preferably, step 4 specifically includes:

step 4.1: recording the load node set of all the connected units as S ^NG, and suspected error unit parameter set Setting the threshold value of the index value deviation as gamma ^NG and the parameter judgment threshold value as beta ⁰;

According to the order of the absolute value of the deviation between the node reliability index evaluation value and the statistical value thereof from big to small, ordering the nodes in S ^NG and marking the nodes as a vector D ^NG;

The kth element of D ^NG is denoted D ^NG (k), let k=1;

Step 4.2: if D ^NG(k)≥α^NG, the fault rate and the repair rate of all the units connected by the node DNG (k) are used as parameters to be solved, and the reliability index value of the load loss probability LOLP, the power loss frequency LOLF and the expected power loss EENS of the node is utilized to solve the nonlinear optimization problem of minimizing the deviation between the reliability index estimation value and the statistic value as an objective function; otherwise, turning to step 4.4;

Step 4.3: the set of reliability parameters of all units on the node is recorded as The maximum deviation d _m of the estimated value of the parameter m from its statistical value is:

Wherein, Is an estimate of m and,Is the statistical value of m;

If d _m≥β⁰, record And adding m to S _GECP;

let k=k+1, go to step 4.2;

step 4.4: the set S _GECP and the estimates of the parameters therein are output.

Preferably, the global optimization algorithm in step 5 is a mixed optimization algorithm based on a PSO algorithm and TRR, the outer layer of the mixed optimization algorithm adopts the PSO algorithm, one particle refers to a vector of the whole reliability parameter value, in the inner layer, for each particle, the reliability parameter value is taken as an initial value, the TRR algorithm is adopted to search a local optimal solution near the particle, the local optimal solution is used for replacing the original particle, the objective function value of the local solution is taken as the fitness of the particle, and the algorithm marks a sensitive parameter with larger deviation between an estimated value and a statistic value as a suspected error parameter.

Preferably, step 5 specifically includes:

Step 5.1: giving all reliability index statistical values and initial value ranges of all parameters;

the set of the sensitive reliability parameters is S ^sensi, and the set of the suspected error parameters is The parameter judgment threshold value is beta ⁰;

The iteration number of the PSO algorithm is N _{PSO_iter}, and the population scale is N _{PSO_size};

randomly generating an initial population, wherein individual particles are denoted as x;

The number of iterations k=0;

Step 5.2: for the particle x, taking x as an initial value, and solving by adopting a trust domain reflection method to obtain a solution x 'and an adaptability value f (x');

Let x=x', if k=n _{PSO_iter}, go to step 5.5;

step 5.3: updating the individual optimum and the global optimum according to the particle fitness;

Step 5.4: updating the speed and the position of each particle, wherein k=k+1, and turning to step 5.2;

step 5.5: outputting a nonlinear optimization problem global optimal solution and a target function value thereof;

Step 5.6: for parameter m, m epsilon S _sensi, calculating the deviation d _m between the estimated value x _m and the statistical value in the optimal solution;

If d _m≥β⁰, adding m to set S _ECP, storing an estimate of m, x _m;

if m ε S _ECP∩S_GECP, then the estimate of m needs to be corrected to x _m and And stored.

Preferably, step 6 specifically includes:

Step 6.1: setting an initial judgment threshold value beta ^init, a minimum judgment threshold value beta ^min, and calculating the number K ^roll of scrolling, wherein the current judgment threshold value is recorded as beta;

The initial value of the parameter m in each LSE is recorded as The estimate obtained for this LSE is x _m;

If m is S _ECP, then Initializing the sensitive parameter estimation value obtained in the step 5;

Otherwise, let

Let the iteration number k=1, β=β ^init;

Step 6.2: taking the whole parameters as variables, executing an LSE algorithm to obtain a parameter estimated value x _m;

step 6.3: calculating the parameter deviation d _m, if d _m is more than or equal to beta, then

Step 6.4: if k=k ^roll, go to step 6.5;

Otherwise, let β=β - (β ^init-β^min)/(K^roll -1), k=k+1, go to step 6.2;

step 6.5: output of And x _m, for each parameter m, if d _m≥β^min, add m to S _ECP.

Preferably, step 7 specifically includes:

Step 7.1: solving a function equation set of the reliability index about the element reliability parameter by adopting an interval algorithm;

step 7.2: in step 7.1, if the equation set has a solution, all node reliability indexes corresponding to each solution are calculated respectively, and whether a certain solution exists or not is judged to meet the reliability condition, namely:

Judging that a certain solution exists, if any node i can meet gamma _i≤γ⁰, then the error parameter identification can be considered to be completed, the solution is an error parameter correction value, and the step is switched to 7.4;

if the equation set has no solution or each solution cannot meet the reliability condition, turning to the step 7.3;

step 7.3: expanding the set S _ECP, removing the suspected correct parameters from the S _ECP, and returning to the step 7.1;

Step 7.4: and finally, outputting the set S _ECP and the corresponding correction value.

Preferably, step 7.1 uses the parameters in S _ECP as the variables to be solved, gives the initial value range, selects the reliability indexes equal to the number of the variables, constructs the function equation set of the reliability indexes with respect to the element reliability parameters, and adopts interval algorithm to solve.

Preferably, the method of expanding the set S _ECP in step 7.3 is:

And (3) sorting the parameters except the S _ECP from large to small, adding the parameters and the combination thereof into the S _ECP according to the sorting, and performing interval verification until the verification passes, and completing expansion of the set S _ECP.

Preferably, the step 7.3 excludes the suspected correct parameter from S _ECP, specifically:

Before each interval check, a trusted region reflection algorithm is adopted to carry out least square estimation on the parameters in S _ECP, and then the minimum deviation c _m between the estimated value of the parameter m and the statistical value of the estimated value is calculated:

Setting a parameter exit threshold as beta ^exit, if c _m<β^exit, considering m as a correct parameter, and excluding the parameter from S _ECP; if c _m≥β^exit, then remain in S _ECP.

The application has the beneficial effects that:

Because the number of the system/node indexes is less than the number of the reliability parameters of all the elements, the reliability parameter correction problem of the power transmission and transmission system can theoretically have a plurality of optimal solutions, and the element reliability parameter true value is one solution.

Drawings

FIG. 1 is a flow chart of the method of the present invention;

FIG. 2 is a schematic diagram of a load node connected to a unit in an embodiment of the present invention;

FIG. 3 is a flowchart of a rolling iterative algorithm for initial value change in an embodiment of the invention;

FIG. 4 is a flowchart of error parameter correction based on interval algorithm in an embodiment of the present invention;

FIG. 5 is a flow chart of the error reliability parameter identification and correction according to an embodiment of the present invention;

FIG. 6 shows the variation of the estimation of some parameters in CIVR-LSE iteration process according to the embodiment of the present invention.

Detailed Description

The application is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present application, and are not intended to limit the scope of the present application.

The invention relates to a method for identifying and correcting error reliability parameters of elements of a power generation and transmission system, which starts from known reliability indexes and checks or obtains the reliability parameters of the elements, belongs to the reliability inverse problem, and the related concept of the reliability inverse problem is shown in a table 1:

TABLE 1 reliability inverse problem correlation concept

The invention can find and correct the wrong reliability parameter by using the known node/system reliability index statistic value and the function of the reliability index about the element reliability parameter (CRP), and the mapping relation between the e ₁,e₂,……,e_Nid element reliability parameter and each reliability index is generally expressed as follows:

Wherein N _C is the number of elements, and N _id is the number of known reliability indexes; Is a known system or node reliability index statistic, e ₁,e₂,……,e_Nid element reliability parameter to each reliability index mapping relation, Is a component failure rate parameter that is a function of the failure rate,Is a component repair rate parameter.

In the formula (1), the reliability parameter of each element includes two parameters of a failure rate and a repair rate. Thus, in a system containing N _C elements, there are 2N _C reliability parameters. Assume that there are N _E error parameters and N _E<N_id among the 2N _C reliability parameters, i.e., the number of unknowns is less than the number of equations. Note that in this regard, equation set (1) may be converted into a nonlinear optimization problem that minimizes the deviation of the reliability index estimate from the statistics as an objective function, and the optimization problem is solved under the constraint of the actual reliability parameter range of the engineering, so that the estimate of the reliability parameter may be obtained. Then, the suspected error parameter can be identified and corrected by comparing the parameter estimation value with the statistic value. The compact form of the specific optimization problem is:

s.t.x^LB≤x≤x^UB

Wherein x is the parameter vector to be solved in the formula (1); x ^LB and x ^UB are the reliability parameter value ranges obtained from engineering experience, and e _i (x) is a vector (e ₁,e₂,……,e_Nid)'. Theoretically, the overall reliability parameter truth values constitute a globally optimal solution of the formula.

As can be seen from equation (2), the reliability index is a multidimensional, nonlinear function of the reliability parameter, and its dimension and degree of nonlinearity increase as the number of elements increases. The number of variables to be solved in the optimization problem (2) is recorded as N _x, and N _x＝2N_C is obtained. Generally, N _x is much greater than N _id. Taking IEEE-RTS as an example, N _id＝54,N_x = 140. Thus, in theory, there may be multiple globally optimal solutions for equation (2).

Because the dimension and nonlinearity degree of the variables are too high, the solution is directly solved by adopting a deterministic mathematical programming algorithm (such as an interior point method, a trusted region reflection method and the like) or an intelligent optimization algorithm, and the obtained solution is often a local optimal solution. In a locally optimal solution, both the wrong and the correct parameters deviate from their true values, which makes accurate correction of the parameters difficult. The interval optimization algorithm can theoretically obtain a globally optimal solution, but the interval searching times of the interval optimization algorithm can exponentially increase along with the dimension of the variable, so that the efficiency is extremely low in solving.

In order to obtain a parameter true value, when solving the formula (2), the invention divides the parameter estimation into two steps of sensitive parameter estimation and overall parameter estimation, firstly obtains a good estimation value of the sensitive parameter, and takes part in the subsequent overall parameter estimation as an initial value. This is because when the initial value error of the sensitive parameter is large, the accuracy of the overall parameter estimation is greatly affected. In the overall parameter estimation, the estimation result is corrected by using the parameter statistics. When the deviation between the parameter estimation value and the statistic value is smaller, the statistic value is still used as the initial value of the subsequent parameter estimation; only when an estimate of a parameter deviates significantly from its statistical value, it is identified as an erroneous parameter and the statistical value is replaced by the estimate as an initial value for the subsequent parameter estimation. This is because, assuming that there are only a small number of parameter errors, the parameter truth vector is in the vicinity of the parameter statistics vector over the search space. By searching around the parameter statistics vector, the above-described correction strategy can promote the probability of the parameter estimate converging to a true value, preventing the estimate from moving away from the true value, as will be explained below.

Since the error reliability parameter correction of the power generation and transmission system is a high-dimensional and strong nonlinear problem, a method capable of identifying all error parameters and accurately correcting the error parameters is difficult to find, and the aim of the invention is to identify the error parameters as far as possible and calculate true values of the error parameters. Good parameter correction results should not only enable the corresponding reliability index evaluation values to fit the reliability index statistic values accurately, but also be sufficiently close to the parameter true values. Further, from the aspects of error of correction value and true value, the invention provides a plurality of indexes for evaluating the correction result, and the indexes are specifically shown in table 2.

TABLE 2 index of parameter correction Effect

As shown in fig. 1, in the implementation, the present invention includes the following steps:

Step 1: based on the parameter statistics value, reliability evaluation is carried out, whether error parameters exist in the reliability parameters of the elements of the power generation and transmission system or not is judged, if so, the step 2 is carried out, and otherwise, the error parameters and the correction value are output;

in specific implementation, all node sets are marked as S _B, suspected error parameter sets A threshold γ ⁰ is set.

Performing reliability evaluation based on the parameter statistic value, and calculating the deviation percentage gamma _i,i∈S_B of the reliability index evaluation value and the statistic value; for example, the reliability index evaluation value is f _e, the statistical value is f _s, and the deviation percentage is

Step 2: the reliability parameters of the power generation and transmission system are simplified, and the method specifically comprises the following steps:

Step 3: calculating the sensitivity of the reliability parameter according to a multi-measure quantization method, and dividing the reliability parameter into a sensitive parameter and a non-sensitive parameter according to the sensitivity, wherein the method specifically comprises the following steps:

Further, for the power generation and transmission system, commonly used reliability evaluation indexes comprise system LOLP, LOLF and EENS indexes, and other indexes can be deduced through the three indexes; common component reliability parameters include failure rate λ, repair rate μ, component availability a, and the like. The system reliability index is a function of the element reliability parameter.

Let the system have n elements, wherein the engineering interested element is m, its availability is A ₁,A₂,…,A_m respectively, then the relationship between the system LOLP, LOLF and EENS index and the element availability is expressed as:

LOLP＝f₁(A₁,A₂,…,A_m)

LOLF＝f₂(A₁,A₂,…,A_m)

EENS＝f₃(A₁,A₂,…,A_m)

Where f ₁,f₂,…,f₃ is a nonlinear function with respect to A ₁,A₂,…,A_m.

The sensitivity is characterized by the partial derivative of the system reliability index with respect to the reliability parameter of a certain element, e.g. the sensitivity of the system reliability indexes LOLP, LOLF and EENS with respect to the reliability parameter of element l is:

each element has two states of normal and fault, and the engineering interested element has 2 ^m combined states.

Any combination state j therein can be further divided into working element state setsAnd a set of failed element statesI (l) is an indicating variable, satisfying:

the larger the partial derivative, the more sensitive the system reliability index is to the change of the element reliability parameter, and the more easily the element reliability is improved, the whole system reliability is easily improved.

And (2) measuring: quantifying the sensitivity of the reliability parameter based on the objective function f (x) measure of equation (2);

the calculation formula of the second measure is:

Where j=1, 2, … …, N _C.

The sensitivity based on the above equation is a partial derivative of the function to the argument at a certain point, and cannot reflect the function variation over a wide range of the argument. In addition, the values of the parameters under study may themselves be incorrect.

Therefore, the classification of the parameter may be misled only according to the partial derivative of the current value of the parameter. In order to overcome the defect, the sensitivity is characterized by the measure of the invention by using a perturbation method. In order to analyze the influence of the component reliability parameter variation value on the system reliability index, the repair rate mu _j or the failure rate lambda _j of the component j is changed by a small amount epsilon, and the increment size of the related reliability index is calculated and can be defined as:

where ε represents the relative change of the independent variable and is generally 0.1. The larger the reliability index increment, the more sensitive the reliability index is to changes in the reliability parameter.

The first measure of sensitivity may also be re-quantified in a similar way.

The first measure of sensitivity reflects the extent to which the parameter affects the system/node reliability index;

And the second measure reflects the influence of the parameter on the deviation of the reliability index evaluation value from the statistical value. It was found in the study whether the parameter was erroneously valued, typically without significant impact on its first sensitivity, and possibly causing a large change in its second sensitivity. This is because the second measure is more susceptible to parameter values than the first measure. For example, when the values of the parameters approach the locally optimal solution (point of zero gradient), the second sensitivity of each parameter gradually approaches 0, while the first sensitivity does not change significantly. The two sensitivities can complement each other, so that the effect of error parameter identification is improved.

For the third measure of sensitivity, the apportionment ratio of the element to the reliability index may reflect the degree of influence of the element reliability level on the reliability index, positively correlated with the first measure.

The measure three-order element distributes the allocation of the system reliability index proportionally, wherein the unreliability of the system is borne by the fault element, and the normal operation element does not participate in the allocation of the system reliability index;

Further, a two-element system is illustrated in which a failure event U is caused by a failure of elements X ₁ and X ₂. The correlation reliability index f (X ₁,X₂,…,X_n) can be decomposed into 3 parts:

The first part is part f ₁(X₁ related to element X ₁ only);

Event f ₂(X₂ associated with second component X ₂);

The third part being related to elements other than X ₁ and X ₂

Thus, the first and second substrates are bonded together,

Let f (k→1) be the sensitivity index corresponding to the faulty element X ₁, and f (k→1) be the sensitivity index corresponding to the faulty element X ₂. Then, define the sensitivity index as

Step 3.2: after the sensitivity calculation is completed, based on the reliability parameter sensitivity quantified by the three measures, the parameters are clustered, and the reliability parameters are divided into sensitive parameters and non-sensitive parameters.

The clustering algorithm can be K-Means clustering, fuzzy C-Means clustering and other algorithms. Because the clustering task is simpler, the K-Means clustering method is used for clustering the K-Means. The reliability parameters are grouped into two groups according to three measures of sensitivity. The final sensitive parameter set is the union of the sensitive parameter sets corresponding to the measures.

and 4-5, calculating an estimated value of the sensitive parameter by adopting a global optimization algorithm so as to provide a good initial value for the subsequent overall parameter estimation.

Note that when a load-shedding strategy is adopted after a system accident, if a unit connected with a load node fails, the load of the node is firstly reduced. In other words, the reliability index of the node is greatly affected by the reliability of the connected units. Therefore, before the sensitive parameters are estimated as a whole, the reliability index of the load node can be used for estimating the reliability parameters of the connected unit.

The specific steps are sequentially described as follows:

(1) Parameter estimation of unit connected with load nodes under consideration of load-cut-off strategy

For a certain load node connected with the unit, if the reliability index evaluation value and the statistic value of the load node have significant differences, the reliability parameter of the unit connected with the node is more likely to be wrong. Therefore, the reliability index of the load node can be utilized to estimate the reliability parameter of the unit connected with the node.

The load node connected unit parameter estimation (algorithm 1) comprises the following steps:

step 4.1: the load node set of all the connected units is recorded as S ^NG, and the suspected error unit parameter set The threshold value of the index value deviation is denoted as γ ^NG, and the parameter determination threshold value is denoted as β ⁰. And ordering the nodes in S ^NG according to the order of the absolute value of the deviation of the node reliability index evaluation value and the statistical value thereof from big to small, and marking the k element of the vector D ^NG.D^NG as D ^NG (k). Let k=1. FIG. 2 is a schematic diagram of load nodes connected to a unit, and G1-G4 are units.

Step 4.2: if D ^NG(k)≥α^NG, using the failure rate and repair rate of all units connected with the node D ^NG (k) as parameters to be solved, and using LOLP, LOLF and EENS index values of the node to establish an optimization problem as in (2) and solve the optimization problem; otherwise, go to step 4.4.

Step 4.3: the set of reliability parameters of all units on the node is recorded asThe maximum deviation d _m of the estimated value of the parameter m from its statistical value is:

Wherein, Is an estimate of m in the present algorithm,Is a statistical value of m. If d _m≥β⁰, recordAnd m is added to S _GECP. Let k=k+1, go to step 4.2.

Step 4.4: the algorithm ends and the set S _GECP and the estimates of the parameters therein are output.

It should be noted that: because the reliability index information of the local nodes is utilized, compared with the sensitive parameter estimation and identification method based on the global optimization algorithm, which is proposed in the following step (2), the error of the parameter estimation can be larger. Therefore, the parameter estimation obtained later should be comprehensively considered to determine the suspected error parameter.

Further, for a certain parameter in S _GECP obtained here, if it is not determined as an error parameter in the sensitive parameter estimation and identification based on the global optimization algorithm set forth in the following (2), it should be excluded from S _GECP.

(2) Sensitive parameter estimation and identification based on global optimization algorithm

The invention establishes the optimization problem as shown in the formula (2) by using all reliability index statistical values, and estimates the sensitive parameters.

Although only the estimation of sensitive parameters is of interest here, the optimization variables need to be overall reliability parameters, not just sensitive parameters. This is because if the optimization variables contain only sensitive parameters, erroneous parameters in non-sensitive parameters may interfere with the sensitive parameter estimation, resulting in large errors in the sensitive parameter estimation.

The local search accuracy of the conventional mathematical programming algorithm is high, but the quality of the solution depends on the initial value, and can only converge when the initial value is sufficiently close to the true value. The group intelligent optimization algorithm can search in the whole space and has certain global convergence characteristic, however, the local searching capability is weak, and the calculation accuracy can not meet the identification requirement.

The invention adopts a mixed optimization algorithm based on particle swarm optimization (PARTICLE SWARM optimization, PSO) algorithm and trusted region reflection algorithm (TRR) to calculate the estimation value of the sensitive parameter. The algorithm fully combines the global searching characteristic of the group intelligent algorithm and the local searching capability of the conventional algorithm, and can give the approximate value of the sensitive parameter.

The outer layer of the hybrid optimization algorithm adopts a PSO algorithm, and one particle refers to a vector of the total reliability parameter values. In the inner layer, for each particle, the reliability parameter value is taken as an initial value, a TRR algorithm is adopted to search a local optimal solution near the particle, the local solution is used for replacing the original particle, and the objective function value of the local solution is taken as the fitness of the particle. After the algorithm is completed, the sensitive parameters with larger deviation between the estimated value and the statistic value are marked as suspected error parameters.

The sensitive parameter initial value estimation algorithm (algorithm 2) comprises the following steps:

Step 5.1: giving all reliability index statistical values and initial value ranges of all parameters.

The set of sensitive reliability parameters is marked as S ^sensi, and the set of suspected error parametersThe parameter decision threshold is noted as beta ⁰.

The iteration number of the PSO algorithm is N _{PSO_iter}, and the population size is N _{PSO_size}.

The initial population is randomly generated, where individual particles are denoted as x.

The number of iterations k=0.

Step 5.2: for the particle x, taking x as an initial value, solving by adopting a trusted region reflection method to obtain a solution x 'and an adaptability value f (x').

Let x=x'. If k=n _{PSO_iter}, go to step 5.5.

Step 5.3: and updating the individual optimum and the global optimum according to the particle fitness.

Step 5.4: the velocity and position of each particle are updated, k=k+1. Turning to step 5.2.

Step 5.5: and outputting a global optimal solution and an objective function value of the formula (2).

Step 5.6: for parameter m (m e S _sensi), the deviation d _m of its estimate x _m from the statistic in the optimal solution is calculated using equation (5).

If d _m≥β⁰, adding m to set S _ECP, storing an estimate of m, x _m;

The estimate of the S _ECP parameter obtained by the algorithm will be used as its initial value in the subsequent estimation algorithm.

Step 6, starting from the error sensitive parameter estimation value, estimating the whole reliability parameter, providing a rolling estimation algorithm for correcting an initial value solution by adopting a parameter statistical value, and estimating the whole parameter, wherein the algorithm can improve the accuracy of the parameter estimation value, and the method is specific:

And utilizing the parameter statistical value to provide a variable initial value rolling iteration least squares estimation (CHANGING INITIAL values rolling LSE, CIVR-LSE) algorithm.

The algorithm executes multiple rounds of LSE by scrolling in hopes of obtaining a solution that is sufficiently close to the true value. Then, parameters suspected of being wrong are identified by comparing the deviation of the statistic value and the estimated value of each parameter.

Since only a small number of erroneous parameters are assumed to be present in the system, the parameter statistics vectors are closer to their true vectors in the search space of the optimization problem. Thus, in parameter estimation, the parameter statistics should be "trusted". And only when the deviation between the parameter estimated value and the statistic value is large, judging the parameter estimated value as a suspected error parameter, taking the estimated value as an initial value of LSE of the next round, and otherwise, taking the statistic value as the initial value. In this way, the algorithm both corrects the suspected error parameter and ensures that the true value is searched for around the parameter statistics. As the number of scrolls increases, the CIVR-LSE solution gradually approaches true values.

In CIVR-LSE, the maximum deviation d _m of the estimate of parameter m from its statistic is:

wherein S ^para is the overall reliability parameter set.

And (3) recording the judgment threshold value as beta in the formula (6), if d _m is more than or equal to beta, judging that the parameter m is wrong, and taking the estimated value x _m as the initial value of the LSE of the next round.

The flow chart of CIVR-LSE algorithm (Algorithm 3) is shown in FIG. 3, which includes the steps of:

Step 6.1: an initial judgment threshold value beta ^init, a minimum judgment threshold value beta ^min, the number of times of rolling calculation K ^roll and a current judgment threshold value beta are set. The initial value of the parameter m in each LSE is recorded as The estimate obtained for this LSE is x _m. If m is S _ECP, thenInitializing the sensitive parameter estimation value obtained in the step 5; otherwise, let Let the number of iterations k=1, β=β ^init.

Step 6.2: with the overall parameters as variables, LSE is performed to obtain a parameter estimate x _m.

Step 6.3: the parameter deviation d _m is calculated according to equation (5). If d _m is greater than or equal to beta, then make

Step 6.4: if k=k ^roll, go to step 6.5; otherwise, let β=β - (β ^init-β^min)/(K^roll -1), k=k+1. Turning to step 6.2.

Step 6.5: output ofAnd x _m. For each parameter m, if d _m≥β^min, then m is added to S _ECP.

The research finds that after the parameter estimation in the steps 4 and 5, most error parameters can be identified, which meets the actual requirement of engineering for improving the accuracy of the parameters, but error parameters with less influence on the reliability index by parts can be omitted. In order to form a closed loop method system for identifying and correcting error parameters, the invention provides a parameter identification completion degree test method based on an interval algorithm, which tests whether all error parameters are identified. If the test is passed, the solution of the interval algorithm is the correction value of the error parameter; otherwise, enumerating parameters or combinations thereof according to the order of the parameter sensitivity from large to small, and respectively adding the enumerated parameters or combinations thereof into the set S _ECP for verification until the verification passes.

① Parameter identification completion degree inspection and correction based on interval algorithm

Based on the analysis of steps 1 to 6, a set of suspected error parameters S _ECP is obtained.

However, it cannot be confirmed whether S _ECP contains all error parameters, and a check is required. And (3) taking the error parameter in S _ECP as a variable, and establishing and solving the LSE problem as shown in the formula (2) by utilizing all reliability indexes.

If S _ECP contains all error parameters, the objective function value is close to 0; otherwise, the function value is not 0. The premise of ensuring the validity of the verification thought is that a solution algorithm can obtain a global optimal solution. However, the conventional nonlinear optimization algorithm cannot determine whether the result is an optimal solution. Therefore, it cannot be determined whether S _ECP contains all error parameters by whether the objective function value is close to 0.

The interval algorithm can judge the existence of the solution on the studied interval, and can ensure that global optimum is obtained. Because the interval optimization algorithm has lower calculation efficiency, an interval iteration algorithm for the equation set is adopted here.

Taking the parameters in S _ECP as the variables to be solved, giving an initial value range, selecting reliability indexes equal to the number of the variables, constructing an equation set shown as the formula (1), and solving.

If a solution exists in the interval algorithm, reliability evaluation is performed based on the solution. If all the system/node reliability indexes are matched with the statistical values, all the error parameters can be judged to be identified, and the solution is the correction value of the error parameters. Otherwise, it may be determined that there are still erroneous parameter omissions.

At this time, the missing error parameters need to be searched for using the two methods of the following ②、③ th.

② Missing error parameter search based on sensitivity ordering

And enumerating parameters or combinations thereof except the step S _ECP, respectively adding the parameters or combinations thereof to the step S _ECP, and performing interval verification until verification passes. When the parameters are too many, the key to improving the searching efficiency is how to determine the parameter priority order of the participating interval verification.

Based on the sensitivity idea, the parameters except S _ECP are ordered from large to small according to the system reliability index change caused by the deviation of the parameter estimation value and the statistical value. The parameters and their combinations are then added to S _ECP in this order.

③ Method for eliminating correct reliability parameters

Since some correct parameters may be misjudged as incorrect parameters, the set S _ECP may contain several correct parameters. If the parameters in S _ECP are too many, the calculation efficiency of the interval algorithm will be affected, so the number of parameters in S _ECP should be reduced as much as possible.

Setting a parameter exit threshold as beta ^exit, if c _m<β^exit, considering m as the correct parameter, and excluding it from S _ECP; if c _m≥β^exit, then remain in S _ECP. The threshold β ^exit should be small to avoid excluding false parameters.

Fig. 4 shows a simple flow of error parameter correction based on the interval algorithm.

Namely, the step 7 specifically comprises the following steps:

Step 7.1: according to the ① method, solving an equation set shown as a formula (1) by adopting an interval algorithm;

Step 7.3: expanding the set S _ECP according to the ② method described above; removing the suspected correct parameters from the S _ECP according to the ③ method, and returning to the step 7.1;

Examples

The following describes the technical scheme in the present invention further with reference to fig. 5, and adopts the modified RBTS, IEEE-RTS and 91 node power system (hereinafter referred to as CS system) to verify the reliability parameter error value identification and correction method of the present invention.

The overall flow of the error reliability parameter identification and correction method is shown in fig. 5, and the steps are as follows:

In the original data of the test system, the reliability parameters of the same type of elements are identical and independent of each other. For example, there are four types of power transmission branches in an IEEE-RTS transmission system, and the original reliability parameters are shown in table 3.

TABLE 3 IEEE original reliability parameters of RTS System Branch

Note that: repair time of two 138kV cables in IEEE-RTS is different and is respectively 16h and 35h. The branch numbers corresponding to the branch types are shown in Table 4.

TABLE 4 IEEE correspondence between branches and branch types for RTS systems

Branch type	Corresponding branch number
		138KV cable	1,10
138KV overhead line	2-6,8-13
		138/230KV transformer	7,14-17
230KV overhead line	18-38

In order to embody the difference of the reliability levels of the same type of elements of the power transmission system, the invention sets the reliability parameters with different values for the same type of power transmission elements. For example, for a 138kV overhead line of an IEEE-RTS system, a set of random numbers obeying normal distribution is generated as true values of unit fault rates of the lines of the type by taking the original value of the unit fault rate of the overhead line as a mean value and taking 1/10 of the unit fault rate as a variance.

To simulate the error parameters in engineering practice, error parameter embodiments are set by modifying the reliability parameter values of part of the elements of the test system in advance. And taking the value of each parameter before modification as a true value, and taking the value after modification as a parameter statistic value. Before the parameter is modified, the system and the node reliability index are calculated based on the true value, and the true value is used as the statistical value of the reliability index in the calculation example. And after the parameters are modified, in the process of parameter identification and correction, carrying out reliability assessment based on parameter statistical values or estimated values, wherein the obtained reliability index is used as a reliability index estimated value.

In combination with the actual situation of the reliability of the elements in the power system, step 2 of the algorithm of fig. 5 makes the following simplifying assumptions:

a) The reliability parameters of the same model units installed on the same node are the same;

b) The reliability parameters of the same type parallel transmission lines connected between the same two nodes are the same.

The component reliability parameters are identical, meaning that their failure rate (or repair rate) shares a to-be-solved variable in the parameter correction problem. These two hypotheses reduce the likelihood of multiple solutions being present.

Assume that a parameter whose parameter statistic and its true value have an error exceeding 10% is taken as an error parameter. The determination threshold value γ ⁰ =1% of the reliability index, the determination threshold value β ⁰ =20% in the sensitive parameter initial value estimation, the initial determination threshold value β ^init =20% in CIVR-LSE, and the minimum determination threshold value β ^min =10%. Correct parameter decision threshold β ^exit =1%. The value interval of each parameter can be obtained by engineering experience, and the value interval is assumed to be 0.5-2 times of the true value.

Reliability parameter identification and correction results

(1) Testing of RBTS and IEEE-RTS systems

Error parameter embodiments Case 1-Case 2 and Case 3-Case 7 are set here for RBTS and IEEE-RTS, respectively, to verify the validity of the proposed method. The error parameters for each example are shown in Table 5

TABLE 5 RBTS component reliability parameter error values for IEEE-RTS

Note that: lambda _G1-20 denotes the failure rate of all 20MW units mounted on node 1, L7 representing leg 7.

The error reliability parameter identification results for Case 1-Case 7 are shown in Table 6.

TABLE 6 RBTS error reliability parameter identification results for various embodiments of IEEE-RTS systems

Comparing the identification results of each embodiment with the set error parameters, it can be seen that the identification results of each embodiment include all the error reliability parameters. Table 7 lists the reliability parameter correction values for some embodiments.

TABLE 7 RBTS and IEEE-RTS System part embodiment reliability parameter correction results

As can be seen from the results in Table 7, the interval algorithm can accurately correct the error values of the reliability parameters, and the correction interval of all the error parameters contains true values of the parameters, wherein the maximum error between the interval boundary and the true values is not more than 0.1%.

(2) Testing of 91 node systems (CS systems)

The 91-node CS system is used here to verify the validity of the proposed method, the error parameters of Case 8 are as follows:

TABLE 8 element reliability parameter error values for CS System

Case 8 includes 10 error parameters, involving 23 elements, which account for 9.7% of the total system elements. The error reliability parameter identification result of Case 8 is as follows:

Case 8: lambda _G52-45、λ_L1、λ_L40、λ_L139、μ_G1-550、μ_G23-100、μ_G33-175、μ_G66-200 and mu _G42-330.

Comparing the Case 8 identification result with the error parameters set in table 8, it can be seen that: the method provided by the invention recognizes 9 parameters, and omits the parameter lambda _L3, and the method does not cover all error parameters, but contains all sensitive error parameters. The reliability parameter correction results are shown in the following table.

Table 9 CS reliability parameter correction results for systems

As can be seen from the results of table 9, although the Case 8 identification result does not contain all error parameters, the identified error parameters can be checked by the interval algorithm, and the correction value is very close to the true value. The results indicate that the missing error parameters do not affect the accuracy of correction to other parameters, probably because the missing error parameters are non-sensitive parameters, which have very little effect on the reliability index.

Reliability parameter identification and correction process analysis

Taking Case 7 of the IEEE-RTS system as an example, the error reliability parameter identification and correction process is shown in steps, wherein the error reliability parameter identification and correction process comprises unit parameter estimation connected with a load node, initial value estimation of sensitive parameters, overall parameter estimation and identification and parameter correction based on interval algorithm.

Step 1-step 3:

Case 7 data is shown in table 5, this example includes 6 error parameters, involving 11 elements, 15.7% of the total system elements, and 14.7% of the total machine capacity.

Before identification and correction, the reliability parameters are clustered first.

The reliability parameters are divided into two types of sensitive parameters and non-sensitive parameters by comprehensively considering three measures of sensitivity. The sensitive parameter sets are as follows:

S^sensi＝{λ_G18-400,λ_G21-400,λ_G23-350,λ_G13-197,λ_G23-155,λ_G1-76,λ_G2-76,λ_G16-155,λ_G22-50,λ_G1-20,μ_G18-400,μ_G21-400,μ_G23-350,μ_G13-197,μ_G23-155,μ_G15-12}.

From the clustering result, the sensitive parameters of the IEEE-RTS system are the unit fault rate and the repair rate. Notably, failure of any one unit can directly reduce power supply adequacy, which in turn can lead to reduced system reliability. Thus, if the total number of units of the system is small, the reliability parameters of the whole unit can be added to the sensitive parameter set.

Step 4: parameter estimation of a group of load nodes connected thereto

The corresponding algorithm of parameter estimation of the unit connected with the load node is algorithm 1. And sequencing the load nodes according to the order of the deviation of the reliability index evaluation value and the statistical value thereof from big to small. Since only one rough estimate of the unit parameters is made, this step mainly focuses on load nodes with large deviations in reliability index, and therefore a deviation threshold γ ^NG =20% is set. The following table shows the load nodes with index value deviations greater than the threshold value gamma ^NG.

TABLE 10 deviation of reliability index evaluation value from statistical value for partial load node

Node numbering	18	1	3	19	6
						Deviation of	32.41％	27.27％	25.96％	21.70％	21.24％

The index value deviation of the node 18 is the largest, and the node is connected with a 400MW unit. The set failure rate and repair rate estimates are obtained using the LOLP, LOLF, and EENS metrics of node 18, and the results are shown in table 11.

Table 11 reliability parameter estimation for units connected to node 18

Assuming a decision threshold of β ⁰ =20%, a parameter with a deviation greater than β ⁰ will be marked as a suspected error parameter and its estimate recorded. As can be seen from table 11, the deviation of lambda _G18-400 is 46.00%, should be marked as a suspected error parameter, and its estimate 7.751 is recorded. The deviation of parameter mu _G18-400 does not exceed threshold beta ⁰, and is still at a statistical value of 58.4 as its initial value in the subsequent parameter estimation. At this time, a set of suspected error parameters S _GECP＝{λ_G18-400. Next, the second node 1 is ranked for the index value deviation degree, and the parameters of the group to which it is connected are estimated.

Table 12 reliability parameter estimation of the set connected to node 1

Parameters (parameters)	Statistics value	Estimation	Deviation of the estimate from the statistic
				λ_G1-20	12.978	20.119	55.02％
λ_G1-76	4.469	6.0335	35.01％
				μ_G1-20	175.2	187.97	7.29％
μ_G1-76	219	188.03	16.47％

As can be seen from table 12, λ _G1-20 and λ _G1-76 should be marked as suspected error parameters and their estimates recorded. At this time, S _GECP＝{λ_G18-400,λ_G1-20,λ_G1-76 }. The subsequent nodes in table 11, nodes 3, 19 and 6, are not connected to the aggregate, and thus the parameter estimation here ends. In addition, as described in (1) of step 4, if a certain parameter in the current S _GECP is not determined as an error parameter in the subsequent step, the parameter should be excluded.

Step 5: initial estimation and identification of sensitive reliability parameters

Initial value estimation and identification of sensitive reliability parameters correspond to algorithm 2. Let the parameter decision threshold β ⁰ =20%, the population size of the PSO algorithm is 30, and the maximum iteration number is 200. After the algorithm is executed, the identified suspected error parameters and their estimated values are shown in table 13.

TABLE 13 list of suspected error parameters among the sensitive parameters

Parameters (parameters)	Statistics value	Estimation	Deviation of the estimate from the statistic
				λ_G1-20	12.978	21.397	60.28％
λ_G18-400	5.309	8.950	64.84％
				μ_G15-12	112.308	159.949	55.60％

At this time, S _ECP＝{λ_G18-400,λ_G1-20,μ_G15-12 }. According to the algorithm flow step 6, three reliability indexes of the systems LOLP, EENS and LOLF are selected, and the parameters in S _ECP are solved by adopting an interval algorithm. However, the interval algorithm has no solution, which means that all error parameters are not currently recognized.

Step 6: overall reliability parameter estimation and identification

The overall reliability parameter estimation and identification corresponds to algorithm 3. And using the estimated value of the suspected error parameter obtained in the previous step as an initial value to participate in rolling least square estimation (CIVR-LSE) of the initial value. Let the initial decision threshold β ^init =20%, the minimum decision threshold β ^min =10%, and the number of scroll iterations K ^roll =15. When the first LSE of CIVR-LSE is completed, there are 11 parameter estimates that deviate from the statistics by more than β ^min. The estimated variation of these 11 parameters during CIVR-LSE rolling iterations is shown in fig. 6. The vertical axis represents the percentage of parameter truth, statistics and estimates relative to the respective truth, and the horizontal axis is the number of 11 parameters. As can be seen from fig. 6, the estimates of these 11 parameters gradually return to true values during the CIVR-LSE iteration. This variation illustrates that CIVR-LSE solutions gradually approach true values.

After CIVR-LSE is completed, the following set of suspected error parameters is obtained S _ECP:

S_ECP＝{λ_G1-20,λ_G18-400,λ_L7,μ_G15-12,μ_L11}

As can be seen by comparison with the error parameter list of table 5, current S _ECP already contains exactly all error parameters except for parameter lambda _L23. According to the algorithm flow step 6, the parameters in the set S _ECP are solved by adopting an interval algorithm, and the algorithm has no solution. This means that all error parameters have not been recognized currently, and thus, it is necessary to continue searching for missing error parameters.

Step 7: error reliability parameter correction based on interval algorithm

According to the algorithm flow in fig. 5, steps 7.1 to 7.3 are repeated at this time. Through enumeration and verification, when the parameter lambda _L23 is added into S _ECP, the interval algorithm has solutions and can pass the verification of all reliability indexes, and all error parameters can be recognized, and the identification and correction process is finished. The final output set S _ECP is: s _ECP＝{λ_G1-20,λ_G18-400,λ_L7,λ_L23,μ_G15-12,μ_L11 }.

It can be seen that S _ECP contains exactly all error parameters. At this time, the solution of the interval algorithm is the correction value of the error parameter. The parameter correction values of this embodiment (Case 7) are shown in table 14.

Comparison of the inventive method with conventional algorithms

For comparison purposes, the parameter identification results of the four steps are shown in Table 14, wherein the first row of the table lists 6 actual error parameters.

Table 14 identification results of actual error parameters at each step

Note that: the term "v" indicates that the parameter is included in the suspected error parameter set when the corresponding step is completed, and the term "x" indicates that the parameter is not included.

As can be seen from table 14, only two error parameters are identified when step 4 and step 5 are completed; when the step 6 is completed, only one parameter is left to be unrecognized; after the algorithm execution is completed, all 6 parameters are identified. The identification process of the embodiment can be known that the 4 main steps set by the proposed method are organically combined, and the steps are advanced layer by layer, so that all error parameters are effectively identified.

In order to prove the effectiveness of the proposed method, 4 sections of evaluation indexes are adopted, and Case 7 is taken as an example, and the correction effect of the proposed method and a conventional algorithm is compared. The following four algorithms were tested:

method A: starting from the parameter statistical value, directly executing LSE;

method B: starting from the parameter statistical value, skipping the initial value estimation of the sensitive parameter, and executing CIVR-LSE;

method C: after estimating the initial value of the sensitive parameter, executing single LSE; without performing CIVR-LSE;

Method D (method of the invention): the initial values of the sensitive parameters are estimated and then CIVR-LSE is performed.

It should be noted that the method D does not include a correction process based on an interval algorithm. The comparison of method B with method D may exhibit the effect of an initial estimate of the sensitive parameter, and the comparison of method C with method D may exhibit the effect of the CIVR-LSE algorithm.

The results of the identification of each method are shown in Table 14, the first row of the table being 6 actual error parameters.

Table 15 identification results of each method on actual error parameters

Note that: the "v" indicates that the parameter is included in the suspected error parameter set obtained by the corresponding method, and the "x" indicates that the parameter is not included.

As can be seen from table 15, for 6 error parameters, method a identified 4 of them, while the other three identified 5, indicating that performing LSE directly was less effective. The partial evaluation indexes of the recognition and correction effects of the respective methods are shown in Table 16.

Table 16 partial evaluation index of identification and correction effect of each method

As can be seen from table 16: the indices of method D are all optimal, while method a is worst, with methods B and C in between. As can be seen from the selectivity index, although the number of error parameters successfully identified by the methods B and C is the same as that of the method D, the methods B and C misjudge a part of correct parameters as error parameters, so that the selectivity of the two methods is low. The suspected error parameters identified by the method D are just actual error parameters. The method A misjudges more correct parameters, and has the lowest selectivity. From the average error index of the estimates, method D is significantly better than the other three methods, with errors of 7.4%, 25.4% and 15.4% for the other three methods, respectively. As can be seen from the comparison of the evaluation indexes of the method B, C and the method D, the lack of any one of the two steps of sensitive parameter initial value estimation and CIVR-LSE can obviously reduce the effect of error parameter identification and correction, which proves the effectiveness of the method D, namely the method of the invention.

Due to missing part of the component outage records, there is an error in the statistics of the reliability parameters of a small part of the components, and it is necessary to identify and correct the erroneous reliability parameters. Under the background, the invention establishes an inverse problem model oriented to error reliability parameter correction and provides a solving method. Since the error reliability parameter is mixed in the overall reliability parameter, the overall parameter needs to be estimated. Because the estimation of the sensitive parameters is usually accurate and has a large influence on the estimation accuracy of the whole parameters, the invention provides a two-step parameter estimation strategy. Firstly, a particle swarm optimization algorithm with a local search mechanism is adopted to estimate sensitive parameters. Then, a variable initial value algorithm is adopted to carry out rolling estimation on all parameters. In the two-step parameter estimation process, identifying error parameters according to the deviation of the parameter estimation value and the statistic value; for the wrong parameter, its estimate will be used instead of its statistic as the initial value for the subsequent estimation. Finally, judging whether the identification process of the error parameter is finished or not by using an interval algorithm, and correcting the parameter. Aiming at the error parameter embodiment Case 7 of the IEEE-RTS, compared with three algorithms such as conventional LSE, the provided algorithm realizes the accurate identification of the error parameter and has optimal selectivity; the parameter estimation errors of the proposed algorithm are only 7.4%, 25.4% and 15.4% of the other three methods respectively; compared with the true value of the error parameter, the correction value error obtained by the method is not more than 1%. The invention has important significance for improving the reliability and safety of future power systems.

While the applicant has described and illustrated the embodiments of the present invention in detail with reference to the drawings, it should be understood by those skilled in the art that the above embodiments are only preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not to limit the scope of the present invention, but any improvements or modifications based on the spirit of the present invention should fall within the scope of the present invention.

Claims

1. A method for identifying and correcting error reliability parameters of elements of a power generation and transmission system is characterized in that:

The method comprises the following steps:

the global optimization algorithm is a mixed optimization algorithm based on a PSO algorithm and TRR, the outer layer of the mixed optimization algorithm adopts the PSO algorithm, one particle refers to a vector of the whole reliability parameter value, in the inner layer, for each particle, the reliability parameter value is taken as an initial value, the TRR algorithm is adopted to search a local optimal solution near the particle, the local solution is used for replacing the original particle, the objective function value of the local solution is taken as the adaptability of the particle, and the algorithm marks a sensitive parameter with larger deviation between an estimated value and a statistic value as a suspected error parameter;

the step 5 specifically comprises the following steps:

The number of iterations k=0;

Let x=x', if k=n _{PSO_iter}, go to step 5.5;

Step 5.6: for parameter m, m epsilon S ^sensi, calculating the deviation d _m between the estimated value x _m and the statistical value in the optimal solution;

If d _m≥β⁰, adding m to set S _ECP, storing an estimate of m, x _m;

if m ε S _ECP∩S_GECP, then the estimate of m needs to be corrected to x _m and Is stored;

Step 6: based on the estimation of the S _ECP and the parameters therein, the total parameters are estimated by adopting CIVR-LSE algorithm, and the suspected error parameter set S _ECP is optimized, which specifically comprises:

Otherwise, let Is the statistical value of m;

let the iteration number k=1, β=β ^init;

Step 6.4: if k=k ^roll, go to step 6.5;

Otherwise, let β=β - (β ^init-β^min)/(K^roll -1), k=k+1, go to step 6.2;

step 6.5: output of And x _m, for each parameter m, if d _m≥β^min, adding m to S _ECP;

Step 7: according to the deviation of the parameter statistic value and the estimated value, identifying the error parameter in S _ECP, and correcting the error parameter by adopting an interval algorithm, wherein the method specifically comprises the following steps:

Step 7.1, taking the parameters in S _ECP as variables to be solved, giving an initial value range, selecting reliability indexes equal to the number of the variables, constructing a function equation set of the reliability indexes with respect to the reliability parameters of the elements, and solving by adopting an interval algorithm;

Judging that a certain solution exists, if any node i can meet gamma _i≤γ⁰, then the error parameter identification can be considered to be completed, the solution is an error parameter correction value, turning to step 7.4, gamma _i is the deviation percentage of the reliability index evaluation value and the statistic value thereof, and gamma ⁰ is the threshold value of the deviation percentage of the reliability index evaluation value and the statistic value thereof;

The method for expanding the set S _ECP in the step 7.3 is as follows:

Sorting the parameters except S _ECP from large to small, adding the parameters and the combination thereof into S _ECP according to the sorting, and performing interval verification until the verification passes, and completing expansion of the set S _ECP;

The step 7.3 of excluding the suspected correct parameters from S _ECP specifically includes:

Setting a parameter exit threshold as beta ^exit, if c _m<β^exit, considering m as a correct parameter, and excluding the parameter from S _ECP; if c _m≥β^exit, continuing to remain in S _ECP;

2. The method for identifying and correcting error reliability parameters of a power generation and transmission system element according to claim 1, wherein:

in step 1, firstly, a function equation set of a reliability index about an element reliability parameter is constructed, wherein the function equation set of the reliability index about the element reliability parameter is as follows:

s.t.x^LB≤x≤x^UB

3. The method for identifying and correcting error reliability parameters of a power generation and transmission system element according to claim 1 or 2, wherein:

In the step 1, reliability evaluation is performed based on the parameter statistics, and the deviation percentage gamma _i,i∈S_B of the reliability index evaluation value and the statistics thereof is calculated;

S _B is a total node set of the power generation and transmission system.

4. A method for error reliability parameter identification and correction of a power generation and transmission system component according to claim 3, wherein:

The step 2 is specifically as follows:

5. The method for identifying and correcting error reliability parameters of a power generation and transmission system element according to claim 2, wherein:

The step 3 specifically comprises the following steps:

6. The method for identifying and correcting error reliability parameters of power generation and transmission system components of claim 5, wherein:

in step 3.1, the sensitivity is represented by measuring the partial derivative of a system reliability index on the reliability parameter of a certain element, and the larger the partial derivative is, the more sensitive the system reliability index is to the change of the reliability parameter of the element, and the more easily the reliability of the element is improved, the whole system is easy to be improved;

7. The method for identifying and correcting error reliability parameters of power generation and transmission system components of claim 5, wherein:

In step 3.2, K-Means clustering is adopted, reliability parameters are respectively clustered into two groups according to three measure quantization results of sensitivity, and a final sensitive parameter set is a union set of sensitive parameter sets corresponding to the measures.

8. The method for identifying and correcting error reliability parameters of a power generation and transmission system element according to claim 2, wherein:

The step 4 specifically comprises the following steps:

The kth element of D ^NG is denoted D ^NG (k), let k=1;

Wherein, Is an estimate of m and,Is the statistical value of m;

If d _m≥β⁰, record And adding m to S _GECP;

let k=k+1, go to step 4.2;