Nuclear power plant equipment common cause failure data acquisition and analysis method and device

Abstract

The invention relates to a nuclear power plant equipment common cause failure data acquisition and analysis method and device. The method comprises the following steps: S1, acquiring an original data record of nuclear power plant equipment from a plurality of data source systems, wherein the original data record comprises an equipment failure event; and S2, common cause failure data of common cause equipment is determined. And S3, processing the common cause failure data by adopting an event influence vector method to obtain an influence vector of the corresponding equipment failure event. And S4, performing parameter estimation on a predetermined common cause failure model according to the influence vector to obtain a corresponding equipment common cause failure parameter. According to the invention, the results of risk assessment and system reliability analysis of the nuclear power plant better conform to the actual situation.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power common cause failure, and in particular to a method and device for collecting and analyzing common cause failure data of nuclear power plant equipment. Background Art

[0002] In the design of nuclear power plant safety-related systems, redundant design is usually adopted to ensure that the system meets the safety and reliability requirements, that is, the system will fail only when multiple identical components that perform the same function in the system fail. In the early reliability engineering, when calculating the reliability of systems with redundant designs, it was usually assumed that the failure of redundant components was independent of each other. Later, based on the long-term experience accumulated in the industry, it was found that the contribution of common cause failure of redundant components to the overall failure of the system is much greater than the contribution of independent failure of components alone to the overall failure of the system. Common cause failure is an important factor that leads to the unavailability of safety systems.

[0003] The main purpose of common cause failure data analysis is to provide common cause basic event data required for quantitative analysis of the fault tree of the probabilistic safety assessment / analysis (PSA) system. Due to the inherent correlation between equipment due to common reasons such as manufacturing, installation, commissioning and working environment, several series of equipment used as redundant backups may fail at the same time. Usually, the probability of this type of failure mode is much greater than the probability of simultaneous failure of several completely independent devices without correlation. It can also be seen from the PSA analysis of nuclear power plants that common cause failures are the main contributor to the frequency of core damage in units.

[0004] Determining the appropriate equipment common cause failure parameters has a great impact on the accuracy of nuclear power plant risk assessment. However, equipment common cause failure analysis is still in its initial stage in China, and there is no data on common cause failure of nuclear power plant equipment. Equipment common cause failure analysis often directly refers to external general data, which has great uncertainty. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method and device for collecting and analyzing common cause failure data of nuclear power plant equipment in view of the above-mentioned defects.

[0006] The technical solution adopted by the present invention to solve the technical problem is: a method for collecting and analyzing common cause failure data of nuclear power plant equipment, comprising the following steps:

[0007] S1. Acquire original data records of nuclear power plant equipment from multiple data source systems, wherein the original data records include equipment failure events;

[0008] S2. Determine the common cause failure data of common cause equipment;

[0009] S3. Processing the common cause failure data using an event impact vector method to obtain an impact vector of a corresponding equipment failure event;

[0010] S4. Estimating parameters of a predetermined common cause failure model according to the influence vector to obtain corresponding equipment common cause failure parameters.

[0011] In one embodiment, step S2 comprises:

[0012] The original data records are screened, distributed, and counted according to a preset common cause failure identification principle to obtain the common cause failure data.

[0013] In one embodiment, step S2 comprises:

[0014] determining a grouping factor based on the device failure event;

[0015] Grouping all the equipment failure events of the nuclear power plant according to the grouping factors to form different failure groups;

[0016] It is determined whether the failure group meets a preset standard condition; if so, the failure group is determined to be a common cause failure event.

[0017] In one embodiment, the grouping factors include power plant, system, equipment, failure mode, and failure date.

[0018] In one embodiment, the preset standard conditions include the following four conditions at the same time:

[0019] Condition 1: Two or more devices in the failure group have failed or degraded;

[0020] Condition 2: The equipment failure interval is within or close to the PSA task time;

[0021] Condition 3: Equipment failures have a common cause and are related through coupling mechanisms;

[0022] Condition 4: Equipment failure is caused by component failure within the boundary range.

[0023] In one embodiment, the common cause failure data includes common cause failure events, and step S3 includes:

[0024] Using an event impact vector method to process the common cause failure event to obtain an impact vector corresponding to the failure event;

[0025] Identify events that include equipment degraded states, events that include failure distribution over time, and events that include common cause uncertainty;

[0026] The impact vectors corresponding to the determined events are vector-combined to obtain a final impact vector.

[0027] In one embodiment, the common cause failure model is an α factor model, and step S4 includes:

[0028] Add the elements corresponding to each impact vector in the common cause failure event to obtain vector n j , where j = 0, 1, ..., m;

[0029] The equipment common cause failure parameters of the α factor model are obtained according to the following estimation method:

[0030]

[0031] Among them, n k represents the total number of events of common cause failure of k devices, satisfying:

[0032]

[0033] in, Represents the kth element of the average impact vector of event i.

[0034] In one embodiment, the method further comprises:

[0035] The equipment common cause failure parameters and the common cause failure events are stored to form a common cause failure database.

[0036] In addition, the present invention also provides a nuclear power plant equipment common cause failure data collection and analysis device, comprising:

[0037] An acquisition unit, used to acquire original data records of nuclear power plant equipment from multiple data source systems, wherein the original data records include equipment failure events;

[0038] A determination unit, used for determining common cause failure data of common cause equipment;

[0039] A processing unit, configured to process the common cause failure data using an event impact vector method to obtain an impact vector corresponding to the equipment failure event;

[0040] A parameter calculation unit is used to perform parameter estimation on a predetermined common cause failure model according to the influence vector to obtain corresponding equipment common cause failure parameters.

[0041] In one of the embodiments, the determination unit is used to screen, distribute, and count the original data records according to a preset common cause failure identification principle to obtain the common cause failure data.

[0042] The method and device for collecting and analyzing common cause failure data of nuclear power plant equipment implementing the present invention have the following beneficial effects: the present invention collects common cause failure data of a specific power plant to form common cause failure data parameters of equipment of the specific power plant, thereby facilitating the results of risk assessment and system reliability analysis of the nuclear power plant to be more in line with actual conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0044] Figure 1 It is a flow chart of a method for collecting and analyzing common cause failure data of nuclear power plant equipment provided by an embodiment of the present invention;

[0045] Figure 2 It is a schematic diagram of a nuclear power plant equipment common cause failure data collection and analysis device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0047] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0048] In order to facilitate the understanding of the present invention, the present invention will be described more fully below, and the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0050] In one of the embodiments provided by the present invention, reference Figure 1 The method for collecting and analyzing common cause failure data of nuclear power plant equipment in this embodiment includes the following steps:

[0051] S1. Obtain the original data records of nuclear power plant equipment from multiple data source systems. It can be understood that the original data records include equipment failure events and equipment degradation events. It should be noted that the preparatory work that needs to be done before determining the common cause failure data mainly includes the following parts: 1) Division of equipment categories and equipment boundaries; 2) Determination of the list of common cause equipment and failure modes; 3) Selection of common cause failure models. Failure modes and equipment categories are used in combination, such as pump startup failure and valve refusal to open; different types of equipment will not define the same failure mode.

[0052] S2. Determine the common cause failure data of common cause equipment. It is understandable that the source of common cause failure data of nuclear power plants involves information such as equipment operation, maintenance, and testing. It is necessary to collect the original data records in the power plant operation, maintenance, testing and other activities, and screen, allocate, and count these data records according to the identification principle of equipment common cause failure, and finally obtain the analysis and statistical results, and conduct corresponding data review and verification. It should be noted that common cause equipment refers to the analysis of equipment of the same type, or equipment of the same general category.

[0053] S3. Use the event impact vector method to process the common cause failure data to obtain the impact vector of the corresponding equipment failure event. It can be understood that this step uses the event impact vector method to analyze and process the equipment common cause failure data obtained through collection and statistics. Moreover, in the process of obtaining the impact vector, it also includes analyzing and determining the equipment degradation parameters, time factors and common cause factors of the event.

[0054] S4. Parameter estimation is performed on a predetermined common cause failure model according to the influence vector to obtain corresponding equipment common cause failure parameters. It can be understood that the original data (such as the number of equipment common cause failures and the total number of equipment failures) is obtained from the nuclear power plant, and the predetermined common cause failure model is used to perform classical estimation on the original data to obtain the corresponding common cause parameters. It can also include using a method that combines power plant specific data with general data for Bayesian estimation. Optionally, the equipment common cause failure parameters are stored to form a common cause failure database.

[0055] This embodiment collects common cause failure data of a specific power plant to form equipment common cause failure data parameters of the specific power plant, thereby facilitating the results of risk assessment and system reliability analysis of the nuclear power plant to be more in line with actual conditions.

[0056] In a specific implementation, the specific implementation steps are described as follows.

[0057] 1. Preparation for Common Cause Failure Data Collection

[0058] The preparation work before the common cause failure data collection mainly includes two parts:

[0059] 1) Division of device classes and device boundaries

[0060] In nuclear power plants, many devices have the same or similar process performance, functional requirements and operating conditions. In order to increase the size of the common cause failure data collection sample space and improve the accuracy of the data statistical processing process, these devices with the same or similar process performance, functional requirements and operating conditions are classified into the same category, called equipment class (for example: electric isolation valves may be manufactured by many different manufacturers and installed on various systems, but they may have similar process performance, functions and operating conditions).

[0061] In the process of dividing equipment classes, it should be followed that the sample devices contained in a equipment class should conform to the definition of the equipment class; the sample devices in the equipment class should come from multiple systems as much as possible and be representative; and the degree of subdivision of the equipment class and the size of the sample space should be weighed.

[0062] For each device class, the boundary of the device class needs to be defined. The purpose of device boundary definition is to define the boundary range of the device class, clarify the allocation and ownership of common cause failure data records of specific devices in the data source, and support statistics on the number of data records or the time when data records occurred.

[0063] The boundaries of the equipment class should include all equipment-specific components; and the boundaries of the equipment class should be clearly defined without boundary crossing. For example, in the PSA model, each switch, relay, or control circuit contactor of the pump is usually not modeled, because such details are difficult to obtain directly from the original data records of the power plant; the failure of these equipment components is usually counted in the failure record of the pump.

[0064] 2) Determination of common cause equipment and failure mode list

[0065] In order to determine the scope of the equipment common cause failure data analysis, it is necessary to first determine the common cause equipment that needs to be analyzed and the list of failure modes that cause equipment failure.

[0066] Generally speaking, common cause devices need to consider multiple devices that are basically in symmetrical positions on different redundant columns and have the same function or role, such as redundant pumps in different injection columns, electric valves in corresponding redundant columns, etc.

[0067] Failure mode refers to the manifestation of a device's inability to achieve a specific function. Each device can have multiple failure modes. For each failure mode, a clear definition is given. Table 1 gives the typical failure mode definitions commonly used in PSA, which can provide a reference for nuclear power plants to assign failure events to specific failure modes.

[0068] Table 1 Definition of typical failure modes in PSA

[0069]

[0070] 3) Common cause failure model selection

[0071] In the PSA model, the common cause failure of equipment is represented in the form of equipment common cause groups and common cause basic events. For different equipment common cause groups, appropriate common cause failure models will be selected for them. The MGL model and the α factor model are currently the most commonly used common cause failure models, and the two are essentially consistent. In the embodiment of the present invention, the α factor model is preferably used as the common cause failure model. Of course, in some embodiments, the MGL model can also be used.

[0072] When collecting CCF data, the α factor model and parameters are selected. The α factor model is a ratio model that uses a series of failure ratios and the total failure probability of components to determine the frequency of common cause failures.

[0073] α k It refers to the proportion of the total failure frequency of all components in the system that results from the failure of k components due to a common cause.

[0074]

[0075] Among them, α1 = the share of the frequency of independent events in the total failure frequency; α2 = the share of the frequency of two components failing due to a common cause in the total failure frequency, and so on. The α factor model has the following relationship with the basic parameter model:

[0076]

[0077] Where Q t Represents the total failure frequency of each component (including all independent events and common cause events).

[0078] For example, when n=2, the probability of the two-component system including independent failure of component A and common cause failure of component AB can be expressed as:

[0079] P(A)=Q1,P(AB)=Q2,then

[0080] Q1=α1Q t , Q2=α2Q t

[0081] When n = 3, Q1 = α1Q t , Q3=α3Q t . And so on.

[0082] 2. Data collection of common cause failures in power plants

[0083] The source of common cause failure data of nuclear power plants involves information such as equipment operation, maintenance, and testing. It is necessary to collect the original data records in the power plant operation, maintenance, testing and other activities, and screen, distribute, and count these data records to finally obtain statistical results. In other words, the original data records are screened, distributed, and counted according to the preset common cause failure identification principles to obtain common cause failure data.

[0084] The process of collecting common cause failure data of PSA equipment is as follows.

[0085] 1) Sources of power plant common cause data

[0086] The power plant specific data source refers to the information on the common cause failure data of the equipment accumulated during the actual operation of the power plant. The accumulation and preservation of common cause failure data of power plants involves multiple departments such as operation, maintenance, and testing. It is difficult to have an original data source that can cover these data, so these data sources must be fully utilized in the process of common cause failure data collection. Different power plants have different data systems and methods for recording equipment operation, testing, failure and maintenance events. In general, the data sources for nuclear power plants to provide typical information on common cause failures of equipment include: experience feedback system, power plant operation event report, event ticket, production process management system, maintenance work order / report, periodic test report, operator log, PSA equipment reliability database.

[0087] 2) Common cause failure data collection process

[0088] The event records in the above-mentioned experience feedback system, production process management system, power plant operation event report and other data sources are analyzed and identified according to the following steps.

[0089] A) First, determine the system, device, failure mode, and degradation level based on the event record description. Event records without device failure or degradation or event records containing devices not in the scope of the CCF study are not considered. Preferably, compare the failure records of each device to eliminate duplicate events. Specific examples are shown in Table 2.

[0090] B) Determine grouping factors based on equipment failure events. Grouping factors include power plant, system, equipment, failure mode, and failure date. Group all equipment failure events in nuclear power plants according to grouping factors to form different failure groups. In other words, group all equipment failure records according to power plant, system, equipment, failure mode, and failure date. This grouping can help analysts identify failure records that occur within a specified time interval and may be related to CCF events (i.e., common cause failure events). Compare the date of each failure record with the dates of other failure records of the power plant, and group all failure records within the time range where CCF events may occur for screening.

[0091] C) Determine whether the failure group meets the preset standard conditions. If so, determine that the failure group is a common cause failure event. Specifically, if the following preset standard conditions are met at the same time, the failure group will be identified as a CCF event:

[0092] Condition 1: Two or more devices have failed or degraded.

[0093] Condition 2: The equipment failure interval is within or close to the PSA task time. For spare equipment, the time interval can be assumed to be the specified test interval plus a 25% margin.

[0094] Condition 3: These device failures have a common cause and are related through a coupling mechanism.

[0095] Condition 4: Equipment failure is caused by component failure within the boundary range.

[0096] It can be understood that the preset standard conditions of this embodiment also correspond to the rules of step 1 to step 3 in the parameter quantification below. That is to say, in the parameter quantification process, the degradation parameters, time factors and common cause factors are assigned values ​​according to the preset standard conditions.

[0097] Table 2 Examples of equipment failure event codes for nuclear power plants

[0098]

[0099] It is understandable that the degree of degradation, failure date, failure cause, and coupling factors need to be clearly stated in the data collection results to facilitate the subsequent common cause parameter processing. When identifying common cause failures of equipment, the equipment failure mechanism does not have to be exactly the same, and the failed subcomponents or parts are not necessarily the same.

[0100] 3) Common cause failure data review

[0101] In the process of collecting equipment common cause failure data, a lot of judgment and analysis work is required, and the data collectors and analysts need to have considerable knowledge of operation, maintenance, periodic testing, and reliability. In this embodiment, this step can also be selected by data collection professionals to collect data, and experts from operation and maintenance departments can professionally verify the collected data to ensure the accuracy of the data.

[0102] 3. Quantification of common cause failure model parameters

[0103] The event impact vector method is used to process the common cause failure events to obtain the impact vector of the corresponding failure event. In other words, the event impact vector method is used to process the equipment common cause failure data collected and statistically obtained. The impact vector is a numerical representation of the CCF event. For a device group with a sample size of m, the impact vector has m+1 elements. F krepresents the (k+1)th element. If exactly K devices fail, then F k is equal to 1, and the others are equal to 0.

[0104] For example, if the size of the common cause device group is equal to 2, then the possible impact vectors are as follows:

[0105] [1, 0, 0] indicates no device failure.

[0106] [0, 1, 0] means that one and only one device fails.

[0107] [0, 0, 1] indicates that two devices have common cause failure.

[0108] The above impact vectors are given when the event classification is clear. To classify an event as a common cause failure, multiple devices must fail simultaneously due to a common cause. Simultaneity and failure are defined based on certain performance criteria. For such events, the definition of the impact vector is unique and clear.

[0109] It should be noted that the event impact vector method of this embodiment can refer to the prior art. When processing common cause failure data, the impact vector method (also known as the event impact vector method) is usually used. The impact vector method is a vector expression method for failure events, which can express the impact of failure events on devices in a common cause group. For example, if a redundant system consists of m devices, and one of the common cause events is the simultaneous failure of k devices, then its impact vector can be expressed as:

[0110]

[0111] However, in actual data collection, many event descriptions lack sufficient details to make accurate judgments. For example, due to known common causes and coupling factors, the event includes two devices in a degraded state, which does not meet the criteria for classifying the "faulty equipment state" as a complete common cause failure. For another example, key information about the individual faults involved in the CCF event may be lacking, including the difficulty in identifying the root cause of the equipment failure in the event.

[0112] Therefore, the classification of events includes the evaluation of their impact vectors, which requires the establishment of several dimensional hypotheses, each of which represents a different explanation of the event. The probability of the hypothesis is related to the degree of influence of the failure cause and the coupling factors, and can of course be evaluated by the analyst.

[0113] The evaluation is performed as follows:

[0114] Step 1: Identify events that include a device degraded state. That is, identify events that include a device degraded state.

[0115] Step 2: Identify events that contain failures distributed over time (including multiple equipment failure events that occur in close time but not simultaneously).

[0116] Step 3: Identify events that contain common cause uncertainty (events that contain multiple failures but it is not entirely certain that these failures are due to a common cause.)

[0117] Step 4: Combine the results of the above three steps into vectors, that is, combine the impact vectors corresponding to the determined events into vectors to obtain an average impact vector.

[0118] The above steps are described in detail as follows.

[0119] Step 1: Identify events that include a degraded device state

[0120] For events in this category, the analyst needs to assess the severity of degradation of each device in the event using the device performance standard as a reference. In other words, given a degradation state, the analyst assesses the probability that this level of degradation will lead to failure during the PSA mission. This is called the device degradation value, denoted as p i And 0≤p i ≤1 (see Table 3-1 for recommended values)

[0121] Table 3-1 Correspondence between device degradation degree and device degradation parameters

[0122]

[0123] Step 2: Identify events that contain the distribution of failures over time

[0124] In this case, it is determined that there is a common cause for the equipment conditions, but the equipment conditions (failures, degradations, etc.) did not occur or were not detected at the same time. In other words, they were recorded at different but closely related times (or test cycles). In this case, a probability q can be assigned that reflects the extent to which the event (equipment degradation) represents a CCF event during the mission time of interest for the analysis (e.g., a typical PSA mission time).

[0125] The q value is affected by the operating characteristics of the equipment. For operating equipment, the probability of assigning an event delay q is explicit and is based only on the reporting time of the fault; there are no assumptions about the time of the fault or whether multiple faults or degraded states occur simultaneously. For standby equipment, the situation is more complicated. See Table 3-2 and Table 3-3 for specific q value assignments.

[0126] Table 3-2 Weight values ​​of time factors for obvious failure

[0127]

[0128] Table 3-3 Potential failure time factor weight values

[0129] Serial number The time interval between the failure of devices A and B Relevance Time factor q 1 There is no need for device B to succeed high 1.0 2 Device B has only one success requirement middle 0.5 3 Device B has only two successful requests Low 0.1 4 Device B has more than two successful requests Not CCF 0.0

[0130] It should be noted that MT in Table 3-2 above represents PSA task time. Time interval description based on PSA task time. In Table 3-3, for "medium" events, there is one success between failures, and for "low" events, there are two successes between failures.

[0131] Regardless of how q is determined, the average impact vectors for these cases are obtained from two sets of impact vectors, one representing the common cause assumption with probability q, and the other representing the independent events assumption. Probability q is the probability that these mechanisms will lead to a common cause failure under actual demand.

[0132] For example, if two of the three devices fail at different times due to a common cause, the set of impact vectors is as follows:

[0133] For common cause failures: I CCF =q[0,0,1,0]=[0,0,q,0]

[0134] For independent failure of device 1: I C1 =(1-q)[0,1,0,0]=[0,1-q,0,0]

[0135] For independent failure of device 2: I C2 =(1-q)[0,1,0,0]=[0,1-q,0,0]

[0136] The average influence vector for this particular case is:

[0137] Typically, for an event involving delayed failure of k devices in a system of m redundant devices, there are k+1 impact vectors as follows:

[0138] For common cause failures: I CCF =[0,0,K,q,K,0]

[0139] For independent failure of device 1: I Cl =[0, 1-q, 0, K, 0]

[0140] For independent failure of device k: I Ck =[0, 1-q, 0, K, 0]

[0141] The average influence vector for this particular case is:

[0142] where q is the k+1 elements of the vector.

[0143] Step 3: Identify events that contain common cause uncertainty

[0144] The uncertainty caused by insufficient information about the state of the relevant equipment and the time of failure can be implicit in the equipment degradation parameter Pi and the time factor q, respectively. The uncertainty caused by the inability to determine whether multiple failures are caused by a common cause requires an independent analysis parameter because it involves an important and different element of the common cause failure event (i.e., the coupling factor). For this purpose, a parameter called the "common cause factor" c is introduced as the analyst's confidence in the existence of a common cause in the event. The value of c can be 0≤c≤1, as follows:

[0145] 1.0 is used when the cause of multiple failures is the same, often resulting in the same failure / degradation mechanism and affecting the same subcomponents in each device. The corrective actions taken for each device involved in the incident are also usually the same. For example, an investigation shows that all redundant valves have similar damage.

[0146] 0.5 This value is used when the event description does not directly indicate that multiple failures are due to the same cause, the same failure mechanism, or affect the same subcomponent, but there is evidence that the underlying root cause of multiple failures is the same. For example, sticking is observed in two check valves, one due to hinge pin wear and the other due to hinge pin misalignment, but there is evidence that the procedures related to the repair and maintenance of these valves are deficient.

[0147] 0.1 Use this value when the event description states that multiple failures are from different causes, different failure mechanisms, or affect different subcomponents, but there is still some evidence that the underlying root cause of multiple failures is the same. For example, one motor has cooling water leaking due to a loose fitting, and another motor has sealing water leaking due to a missing gasket, but it can be inferred that the underlying root cause of the problem is lack of training, inadequate procedures, etc. Since this assumption is based on information in the event record, the common cause factor is small.

[0148] The effect of this factor on the event impact vector can be similar to the time factor q. More specifically, after replacing q with c, the following equation can be obtained:

[0149] For common cause failures:

[0150] For independent failure of device 1:

[0151] For independent failure of device k: I Cl =[0, 1-c, 0, K, 0]

[0152] The average impact vector for this particular case is: I Ck =[0, 1-c, 0, K, 0]

[0153] where c is the k+1 elements of the vector.

[0154] Step 4: Identify events that include equipment degradation status, time delays, and common cause uncertainties

[0155] If an event includes the above three situations of equipment degradation, time delay and common cause uncertainty, the impact vector of the event can be determined in two steps: first, the uncertainty of the time factor and the common cause factor is not considered, and only the degradation parameters of the equipment are considered to establish the impact vector, and then the impact vector is corrected to reflect the uncertainty of the time factor and the common cause factor. The formula of the impact vector is as follows:

[0156] I CCF =[cqF0, cqF1, ..., cqF m ]

[0157] Device 1: I C1 =[(1-cq)(1-p1),(1-cq)p1,0,...,0]

[0158] …

[0159] Device m: I Cm =[(1-cq)(1-p m ), (1-cq)p m , 0, ..., 0]

[0160] Among them, c is the common cause factor; q is the time factor; p i is the degradation parameter of the ith device; F i Available through p i The average impact vector of the event in this case is calculated as:

[0161]

[0162] 4. Parameter estimation to form a common cause failure database

[0163] In the system to be analyzed, once the average impact vector of all events is determined, the elements corresponding to each impact vector can be added to obtain the vector (n0, n1, ..., n m ),in:

[0164]

[0165] n k represents the total number of common cause failure events of k devices, Represents the kth element of the average impact vector of event i.

[0166] For the α factor model used in equipment common cause failure analysis, the common cause parameters (also known as equipment common cause failure parameters) satisfy:

[0167]

[0168] α k It indicates the ratio of the common cause failures of k devices to the total failures in a system with m identical devices.

[0169] The parameters of the α factor model can be calculated using the following maximum likelihood estimation method:

[0170]

[0171] For example, in the specific data collection of nuclear power plants, the impact vector of the common cause group of three equipments of emergency diesel generator set startup failure is [0,892.579,25.294,17.527].

[0172] The parameters of the α factor model can be calculated using the maximum likelihood estimation method above. For the three common cause groups of emergency diesel generator set startup failure:

[0173]

[0174] In summary, a specific quantitative report of the power plant with failure to start emergency diesel generator sets can be obtained, see Table 4.

[0175] Table 4 Specific quantitative report on emergency diesel generator set start failure

[0176]

[0177] For most equipment, there are very few records of common cause failures; directly using the collected power plant data and the classical estimation method to calculate may result in a problem of too small sample space. In this case, the prior data and the power plant-specific data are comprehensively considered and processed using the Bayesian estimation method: the publicly released equipment common cause failure data is used as the prior data, and the characteristic data collected by the power plant is used as the sample data, and the posterior data is obtained by Bayesian processing.

[0178] The technical method used in the present invention is to collect common cause failure records and data of nuclear power plants through a set of standardized nuclear power plant equipment common cause failure data collection and analysis processing methods and models to form equipment common cause failure data parameters and database.

[0179] The main advantages are as follows:

[0180] (1) Collect common cause failure data of specific power plants to form common cause failure data parameters and database of power plant equipment; the results of risk assessment and system reliability analysis of power plants are more in line with actual conditions;

[0181] (2) This method is widely applicable. It can be used to carry out analysis and evaluation of nuclear power units with different reactor designs as long as the equipment has sufficient event data.

[0182] In another embodiment provided by the present invention, reference Figure 2 The nuclear power plant equipment common cause failure data collection and analysis device of this embodiment includes:

[0183] The acquisition unit is used to acquire original data records of nuclear power plant equipment from multiple data source systems, where the original data records include equipment failure events.

[0184] A determination unit is used to determine common cause failure data of common cause equipment.

[0185] The processing unit is used to process the common cause failure data by adopting the event impact vector method to obtain the impact vector of the corresponding equipment failure event.

[0186] The parameter calculation unit is used to estimate the parameters of the predetermined common cause failure model according to the influence vector to obtain the corresponding equipment common cause failure parameters.

[0187] This embodiment collects common cause failure data of a specific power plant to form equipment common cause failure data parameters of the specific power plant, thereby facilitating the results of risk assessment and system reliability analysis of the nuclear power plant to be more in line with actual conditions.

[0188] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0189] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0190] It can be understood that the above embodiments only express the preferred implementation modes of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the patent scope of the present invention. It should be pointed out that, for ordinary technicians in this field, the above technical features can be freely combined without departing from the concept of the present invention, and several deformations and improvements can be made, which all belong to the protection scope of the present invention. Therefore, all equivalent changes and modifications made to the scope of the claims of the present invention should belong to the coverage of the claims of the present invention.

Claims

1. A method for collecting and analyzing common cause failure data of nuclear power plant equipment, characterized in that: The following steps are involved: S1. Acquire original data records of nuclear power plant equipment from multiple data source systems, wherein the original data records include equipment failure events; S2. Determine the common cause failure data of common cause equipment; S3. Processing the common cause failure data using an event impact vector method to obtain an impact vector of a corresponding equipment failure event; S4. Estimating parameters of a predetermined common cause failure model according to the influence vector to obtain corresponding equipment common cause failure parameters.

2. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 1, characterized in that: Step S2 includes: The original data records are screened, distributed, and counted according to a preset common cause failure identification principle to obtain the common cause failure data.

3. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 2, characterized in that: Step S2 includes: determining a grouping factor based on the device failure event; Grouping all the equipment failure events of the nuclear power plant according to the grouping factors to form different failure groups; It is determined whether the failure group meets a preset standard condition; if so, the failure group is determined to be a common cause failure event.

4. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 3, characterized in that: The grouping factors include power plant, system, equipment, failure mode, and failure date.

5. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 3, characterized in that: The preset standard conditions also include the following four conditions: Condition 1: Two or more devices in the failure group have failed or degraded; Condition 2: The equipment failure interval is within or close to the PSA task time; Condition 3: Equipment failures have a common cause and are related through coupling mechanisms; Condition 4: Equipment failure is caused by component failure within the boundary range.

6. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 1, characterized in that: The common cause failure data includes common cause failure events, and step S3 includes: Using an event impact vector method to process the common cause failure event to obtain an impact vector corresponding to the failure event; Identify events that include equipment degraded states, events that include failure distribution over time, and events that include common cause uncertainty; The impact vectors corresponding to the determined events are vector-combined to obtain a final impact vector.

7. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 6, characterized in that: The common cause failure model is an α factor model, and step S4 includes: Add the elements corresponding to each impact vector in the common cause failure event to obtain vector n j , where j = 0, 1, ..., m; The equipment common cause failure parameters of the α factor model are obtained according to the following estimation method: Among them, n k represents the total number of events of common cause failure of k devices, satisfying: in, Represents the kth element of the average impact vector of event i.

8. The method for collecting and analyzing common cause failure data of nuclear power plant equipment according to claim 6, characterized in that: The method further includes: The equipment common cause failure parameters and the common cause failure events are stored to form a common cause failure database.

9. A nuclear power plant equipment common cause failure data collection and analysis device, characterized in that: include: An acquisition unit, used to acquire original data records of nuclear power plant equipment from multiple data source systems, wherein the original data records include equipment failure events; A determination unit, used for determining common cause failure data of common cause equipment; A processing unit, configured to process the common cause failure data using an event impact vector method to obtain an impact vector corresponding to the equipment failure event; A parameter calculation unit is used to perform parameter estimation on a predetermined common cause failure model according to the influence vector to obtain corresponding equipment common cause failure parameters.

10. The nuclear power plant equipment common cause failure data acquisition and analysis device according to claim 9, characterized in that: The determination unit is used to screen, distribute and count the original data records according to a preset common cause failure identification principle to obtain the common cause failure data.

Images