Power distribution reliability evaluation method based on time-varying failure rate and non-homogeneous Poisson

Abstract

The invention relates to the field of power system management, and discloses a power distribution reliability evaluation method based on time-varying fault rate and non-homogeneous Poisson, comprising the following steps: S1, collecting operation data of a power distribution system, including historical fault record and operation duration of each component; s2, establishing a time-varying fault rate model of each element in the power distribution system; s3, constructing a non-homogeneous Poisson process model according to the time-varying fault rate model; s4, identifying a key transmission path in the power distribution system by using a minimum path method; s5, calculating a reliability index of the power distribution system; and S6, outputting a reliability evaluation result of the power distribution system. According to the method, the non-homogeneous Poisson process model based on the time-varying failure rate is established, so that the time-varying failure behavior of the element in the power distribution system can be simulated and predicted more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management, and in particular to a distribution reliability evaluation method based on time-varying failure rate and non-homogeneous Poisson. Background Art

[0002] With the rapid development of science and technology in my country, especially the widespread application of power electronics and computer technology, users have increasingly higher requirements for the reliability of power distribution systems. Power departments and government organizations are paying more and more attention to improving the reliability of distribution systems, because this is directly related to the stability and economic benefits of power supply. Failures in power equipment often lead to power grid safety accidents. Therefore, timely detection and repair of these faulty equipment is the key to reducing economic losses.

[0003] In traditional distribution system reliability assessment, the failure rate of components is usually regarded as a constant value. Although this assumption simplifies the modeling process, it will introduce large errors when facing large-scale components. In fact, due to external environmental influences and internal aging, the failure rate of components changes over time. The assumption of a constant failure rate cannot truly reflect the changes in the failure rate of components throughout their life cycle, thus affecting the overall reliability assessment of the distribution system.

[0004] Although current scientific research provides a variety of methods for obtaining time-varying failure rates, these methods still have obvious limitations in practical applications. The main problem is that although these simulation-based evaluation methods can improve the accuracy of calculations, they also greatly increase the complexity and time cost of calculations. In the increasingly complex and dynamic operating environment of power systems, rapid response of evaluations becomes particularly important. The long calculation process not only increases the complexity of operations, but may also affect the decision-making of system operation and maintenance due to the inability to provide evaluation results in real time. This situation is particularly insufficient today when the distribution network is changing rapidly and emergencies are frequent. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson, which solves the problems of long calculation time and insufficient accuracy in the existing distribution system reliability assessment method. By introducing an assessment method based on time-varying failure rate and non-homogeneous Poisson process, the efficiency and accuracy of the assessment are improved, so as to more effectively guide the planning and operation of the distribution network.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A distribution reliability evaluation method based on time-varying failure rate and non-homogeneous Poisson, comprising the following steps:

[0007] Collect the operating data of the power distribution system, including the historical fault records and operating hours of each component;

[0008] Establish a time-varying failure rate model for each component in the power distribution system;

[0009] According to the time-varying failure rate model, a non-homogeneous Poisson process model is constructed;

[0010] Use the least path method to identify critical transmission paths in the distribution system;

[0011] Calculate reliability indicators of power distribution systems;

[0012] Output the reliability assessment results of the power distribution system.

[0013] Preferably, the time-varying failure rate model includes modeling of component aging failure and random failure, the aging failure is described by Weibull distribution, and the random failure is described by exponential distribution.

[0014] Preferably, the aging failure is described by Weibull distribution, and the shape parameter m and scale parameter η of the Weibull distribution are extracted from historical failure data by regression analysis. The shape parameter is used to describe the variation trend of the failure rate over time. The time-varying failure rate of the Weibull distribution is defined as:

[0015]

[0016] Where λ(t) is the instantaneous failure rate at time t, m is the shape parameter, and η is the scale parameter.

[0017] Preferably, the random failure is described by exponential distribution, the failure rate of random failure is defined as a constant λ0, and the failure rate of random failure is calculated by the following formula:

[0018]

[0019] in, is the constant failure rate of random failures, a is the number of random failures, t i is the operating time of the faulty device, t j is the running time of normal equipment, Z1 and Z2 are the sets of faulty and normal equipment respectively.

[0020] Preferably, the non-homogeneous Poisson process model is used to describe the failure behavior of components over time, and the model calculates the expected number of events by the following formula:

[0021]

[0022] Where m(t) represents the expected number of failure events at time t, and λ(s) represents the instantaneous failure rate at time s.

[0023] Preferably, the probability distribution of the non-homogeneous Poisson process is given by the following formula:

[0024]

[0025] Wherein, P{N(t)=n} represents the probability of n failures occurring at time t, and m(t) is the expected number of failure events at time t.

[0026] Preferably, the method further comprises the following steps for identifying and evaluating critical transmission paths in the power distribution system:

[0027] Use the minimum path method to analyze all possible transmission paths in the distribution system to determine the critical paths that may affect the overall power supply reliability of the system;

[0028] For each identified critical path, the impact of the time-varying failure rate of each component on the path on the reliability of the path is calculated based on the constructed non-homogeneous Poisson process model;

[0029] The reliability assessment of all critical paths is integrated to determine the overall reliability level of the distribution system.

[0030] The present invention also provides a power distribution reliability evaluation device based on time-varying failure rate and non-homogeneous Poisson, comprising:

[0031] Data acquisition module, used to collect distribution system operation data;

[0032] Model building module for building time-varying failure rate models and non-homogeneous Poisson process models;

[0033] Reliability analysis module, used to perform system reliability analysis using the minimum path method;

[0034] A calculation module is used to calculate various reliability indicators of the power distribution system;

[0035] Output module, used to generate reliability assessment report.

[0036] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the above method is implemented.

[0037] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0038] The present invention provides a distribution reliability evaluation method based on time-varying failure rate and non-homogeneous Poisson.

[0039] It has the following beneficial effects:

[0040] 1. The present invention can more accurately simulate and predict the time-varying failure behavior of components in the power distribution system by establishing a non-homogeneous Poisson process model based on time-varying failure rate. Compared with the traditional static failure rate model, this method takes into account the aging of components and changes in usage conditions, thereby providing a more realistic reliability assessment. Through accurate prediction, maintenance cycles and resource allocation can be planned more effectively, unexpected power outages can be reduced, and the operating efficiency and reliability of the entire system can be improved.

[0041] 2. The present invention can identify the key transmission paths and weak links in the system through detailed reliability analysis of the power distribution system, so that maintenance work can be carried out more targetedly. This not only helps power companies make more reasonable maintenance decisions under limited resources, but also reduces the occurrence of major failures through preventive maintenance, ultimately reducing maintenance costs and improving service quality.

[0042] 3. The evaluation method of the present invention is highly adaptable and flexible and can be adjusted according to different types of distribution systems and changing operating conditions. This flexibility makes the method not only suitable for power systems of all sizes, from small enterprises to urban distribution networks, but also able to adapt to future technological developments and changes in operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0044] Figure 2 A flow chart of time-varying fault rate extraction according to an embodiment of the present invention;

[0045] Figure 3 This is an IEEE RBTS-BUS2 test diagram of an embodiment of the present invention;

[0046] Figure 4 is a reliability assessment flow chart of an embodiment of the present invention;

[0047] Figure 5 Schematic diagram of the average annual failure rate of LP 1 according to an embodiment of the present invention;

[0048] Figure 6 This is a schematic diagram of the annual average failure time of LP 1 in an embodiment of the present invention;

[0049] Figure 7 Schematic diagram of average fault duration of LP 1 according to an embodiment of the present invention;

[0050] Figure 8 It is a schematic diagram of the structure of the device of the present invention;

[0051] Fig. 9 It is a schematic diagram of the computer device structure of the present invention.

[0052] Among them, 100, data acquisition module; 200, model building module; 300, reliability analysis module; 400, calculation module; 500, output module; 40, computer equipment; 41, processor; 42, memory; 43, storage medium. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Please refer to the attached Figure 1 -Attached Figure 7 The embodiment of the present invention provides a method for evaluating the reliability of power distribution based on time-varying failure rate and non-homogeneous Poisson, comprising the following steps:

[0055] S1. Collecting the operating data of the power distribution system

[0056] In the implementation of the distribution reliability assessment method, step S1 is the basis and starting point of the entire assessment process. The purpose of this step is to collect all necessary distribution system data to ensure the accuracy and reliability of the time-varying failure rate model and the non-homogeneous Poisson process model in the subsequent steps. The comprehensiveness and accuracy of the data directly affect the effect of the fault prediction model and the quality of the assessment results. Therefore, the detailed collection and accurate processing of data are the key to achieving efficient distribution system management and optimizing operation and maintenance strategies.

[0057] In this embodiment, step S1 involves the detailed collection of distribution system operation data. Specifically, this includes but is not limited to the collection of the following types of data:

[0058] Historical fault records: These records provide the fault conditions of each component in past operation, including the time of the fault, the type of fault, the time of fault handling, and the recovery after the fault. These data are very important for analyzing the failure mode and frequency of the component.

[0059] Operating time: The operating time of a component from installation to the present. This is an important indicator for evaluating the health of a component and predicting its future probability of failure. Especially for aging equipment, operating time is a key factor in determining its maintenance and replacement cycle.

[0060] Environmental factors: Environmental factors such as temperature, humidity, pollution, etc. may affect the performance and life of components. Collecting this data helps to consider the impact of the environment on the failure rate in the model, making the failure prediction more accurate.

[0061] Load changes: The load changes to which the components are subjected are also recorded, as frequent or sudden load changes may accelerate wear and aging of the equipment.

[0062] In this embodiment, the collected data will be systematically organized and analyzed for the subsequent establishment of a time-varying failure rate model and a non-homogeneous Poisson process model. The accurate collection and processing of these data ensures the scientificity and practicality of the reliability assessment method of the entire distribution system, and provides solid data support for the operation and maintenance decision-making of the power system.

[0063] S2. Establish a time-varying failure rate model for each component in the power distribution system

[0064] After completing the data collection in step S1, the task of step S2 is to establish a time-varying failure rate model for each distribution system component based on the collected data. This model is the basis for the construction of the non-homogeneous Poisson process in the subsequent steps and directly affects the accuracy of the reliability assessment of the entire system. By constructing a time-varying failure rate model, the failure probability of components changing over time can be accurately described, providing strong support for the overall failure prediction of the system.

[0065] In this embodiment, the specific implementation of step S2 is to establish a time-varying failure rate model of each component in the power distribution system. The model is intended to reflect how the failure rate of the component changes over time, thereby providing a basis for the construction of a non-homogeneous Poisson process model.

[0066] 1. Aging failure model:

[0067] For the modeling of aging failure, Weibull distribution is used to describe the failure rate of components. Weibull distribution can well describe the change of failure rate of components at different aging stages. Its time-varying failure rate model is expressed as:

[0068]

[0069] Among them, λ(t) is the instantaneous failure rate at time t, m is the shape parameter, which determines the growth trend of the failure rate over time, and η is the scale parameter, which determines the time scale of the failure rate.

[0070] Shape parameter m: extracted from historical failure data through regression analysis. When the shape parameter is greater than 1, the failure rate increases with time, which is applicable to aging failures.

[0071] Scale parameter η: It represents the characteristic life of a component and is usually obtained based on the design life of the equipment or statistically from failure data.

[0072] 2. Random failure model:

[0073] For the modeling of random failure, exponential distribution is used to describe it. Unlike aging failure, the failure rate of random failure is constant over time, and its model formula is as follows:

[0074]

[0075] in, is the constant failure rate of random failures, a is the number of random failures, t i is the operating time of the faulty device, t j is the running time of normal equipment, Z1 and Z2 are the sets of faulty and normal equipment respectively.

[0076] Constant failure rate Obtained through statistical analysis of historical data, it is applicable to random failure events that do not change significantly over time.

[0077] In this embodiment, the establishment of the time-varying failure rate model provides accurate input parameters for the subsequent non-homogeneous Poisson process. These models can describe the failure behavior of components in different time periods in detail, thereby improving the accuracy of the overall reliability assessment of the system. Through accurate model construction, the maintenance and operation strategies of the distribution system can be more effectively optimized and improved.

[0078] S3. Construct a non-homogeneous Poisson process model based on the time-varying failure rate model

[0079] After establishing the time-varying failure rate models of each component, the task of step S3 is to use these models to construct a non-homogeneous Poisson process model. The non-homogeneous Poisson process is a statistical model that can describe the time-varying frequency of events (such as failures) in the system. Through this process, the time dependence of component failures in the distribution system can be simulated, thereby providing dynamic support for reliability assessment. This step is crucial for accurately predicting the reliability indicators of the system because it combines the previously established time-varying failure rate model to provide a comprehensive fault prediction framework.

[0080] In this embodiment, the specific implementation of step S3 is to construct a non-homogeneous Poisson process model based on the time-varying failure rate model established in step S2. This model is used to simulate the time distribution characteristics of component failures in the power distribution system, especially the frequency of events that vary with time at time t.

[0081] 1. Construction of non-homogeneous Poisson process:

[0082] The intensity function of the non-homogeneous Poisson process is different from the homogeneous Poisson process. The intensity function of the latter is a constant of time, while the intensity function of the former is a function of time t. The intensity function of the non-homogeneous Poisson process is determined by the time-varying failure rate model λ of each component. i (t) is provided, and its total intensity function can be expressed as:

[0083]

[0084] Among them, λ i (t) is the failure rate of the ith component at time t, and n is the total number of components in the system.

[0085] 2. Calculation of expected number of events m(t):

[0086] An important feature of the non-homogeneous Poisson process is its cumulative intensity function m(t), which represents the total number of failure events expected to occur by time t, and is given by:

[0087]

[0088] Where m(t) represents the expected number of failure events at time t, and λ(s) represents the instantaneous failure rate at time s.

[0089] By accumulating the failure rates in each time period, the failure trend of the entire system within a given time range can be obtained.

[0090] 3. Calculation of the probability of an event occurring:

[0091] Based on the cumulative strength function m(t), the probability of a system failing n times at time t can be calculated as follows:

[0092]

[0093] Where P{N(t)=n} represents the probability of n failures occurring at time t, and m(t) is the expected number of events at time t.

[0094] In this embodiment, by constructing a non-homogeneous Poisson process model, the time distribution characteristics of fault events in the power distribution system can be fully described. The advantage of this model is that it can accurately capture the dynamic characteristics of the failure rate in the system over time, thereby providing strong theoretical support for reliability assessment. This process ensures the accuracy of system reliability analysis and helps to better plan and manage the maintenance and operation strategies of the system.

[0095] S4. Use the minimum path method to identify the key transmission paths in the distribution system

[0096] After constructing the non-homogeneous Poisson process model, the key task of step S4 is to use the minimum path method to evaluate the reliability of the distribution system. The goal of this evaluation is to analyze the reliability of the distribution system by identifying the critical paths in the system. The failure of these critical paths may seriously affect the power supply capacity of the entire system, so identifying these paths is crucial to preventing system-level failures.

[0097] In this embodiment, step S4 uses the minimum path method to evaluate the reliability of the power distribution system based on the established time-varying failure rate model and the non-homogeneous Poisson process. The following is the specific content of this step:

[0098] 1. The basis for the application of the minimum path method:

[0099] In power distribution systems, the minimum path method is used to identify critical paths from source nodes to load nodes. These paths consist of a series of power components, each of which has its own specific time-varying failure rate λ(t), which changes with the aging and service life of the components.

[0100] 2. Critical path identification:

[0101] By calculating the cumulative failure rate of all components on each path, the path most likely to cause system failure can be identified. i The failure probability F(P i ) can be expressed as:

[0102]

[0103] Among them, λ j (t) is the path P i The time-varying failure rate of the jth component at time t.

[0104] 3. Calculation of expected value of path failure rate:

[0105] Based on the time-varying failure rate, the expected value of the component failure rate is calculated year by year, thereby obtaining the reliability index of each load point. This index reflects the impact of component aging on system reliability over time.

[0106] In this embodiment, step S4 effectively identifies the critical path in the power distribution system and evaluates the reliability of the system by applying the minimum path method in combination with the time-varying failure rate model and the non-homogeneous Poisson process.

[0107] S5. Calculate the reliability index of the power distribution system

[0108] After identifying the critical transmission paths in the distribution system, step S5 involves calculating the reliability indicators of the distribution system. This step is a comprehensive application of the previous steps, converting the analysis results of the time-varying failure rate and the critical path into specific reliability quantitative indicators, thereby providing decision support for the operation and maintenance of the distribution system. These indicators not only measure the overall performance of the system, but also reflect the reliability of the system's services to end users.

[0109] In this embodiment, step S5 is specifically implemented as follows:

[0110] 1. Definition of reliability index:

[0111] Reliability index is a key parameter to measure the performance of power distribution system, mainly including:

[0112] a) SAI FI, system average outage frequency index: the average number of consecutive power outages for each user within a predetermined period of time (usually one year).

[0113] b) SAIDI, System Average Outage Duration Index: The average total failure time experienced by each user in the system during a specified period of time.

[0114] c) CAIDI, user average outage duration index: the average fault duration of each user in the system within a specified time period.

[0115] d) ASAI, average power supply availability index: the ratio of the total actual power supply hours to the total power supply hours required by users within a specified time.

[0116] e) EENS, Expected Power Shortage: The total power shortage of the system during a specified period of time.

[0117] f) AENS, average expected power shortage: the average power shortage for each user within a specified time.

[0118] 2. Calculation method of reliability index:

[0119] SAI FI calculation formula:

[0120]

[0121] The calculation formula of SAID I is:

[0122]

[0123] CA ID I calculation formula:

[0124]

[0125] The calculation formula of ASA I is:

[0126]

[0127] EENS calculation formula:

[0128]

[0129] The calculation formula of AENS is:

[0130]

[0131] Among them, λ j, U j 、E j are the annual average number of power outages, annual average failure time, and annual average power shortage at load point j, respectively. j is the number of users at load point j.

[0132] 3. Calculation implementation:

[0133] The analysis of the critical path from step S4 is combined with the time-varying failure rate model and the non-homogeneous Poisson process established from steps S2 and S3.

[0134] First, the annual average number of failures and outage duration of each node are determined, which are calculated based on the component failure data and their time-varying failure rates on the critical path. Then, SAIFI, SAIDI, CAIDI, ASAI, EENS, and AENS are calculated.

[0135] 4. Result analysis:

[0136] Based on the calculated values, the reliability changes of the system in different time periods (such as 0 to 30 years and 30 to 40 years) are evaluated. These results help to understand the changing trends of system performance during the aging process of components.

[0137] The analysis results of reliability indicators are used to optimize maintenance strategies and resource allocation to improve the overall reliability and performance of the system.

[0138] In this embodiment, step S5 not only quantifies the reliability of the power distribution system, but also provides an intuitive understanding of the changes in system performance over time, making the management and maintenance of the power distribution system more scientific and efficient. Through detailed analysis of these indicators, reliable decision support can be provided for the continuous improvement of the power system.

[0139] S6. Reliability assessment results of output power distribution system

[0140] Step S6 is the final stage of the entire distribution reliability assessment process, in which the main task is to output the reliability assessment results of the distribution system. This step integrates the various reliability indicators and analysis data calculated in the previous steps and presents them to system operators and decision makers in an easy-to-understand format.

[0141] In this embodiment, the specific implementation of step S6 is as follows:

[0142] 1. Integrate the calculation results:

[0143] Summarize the data generated in the previous steps, such as the failure probability of the critical path, the annual average failure rate of each part of the system, and the power outage time.

[0144] 2. Result output format:

[0145] The output format is designed as reports and charts, which makes it easy for operations and management personnel to quickly understand the status and performance of the system.

[0146] 3. Report preparation:

[0147] Produce a detailed evaluation report, including a comprehensive analysis of system reliability, explanation of the meaning of each indicator, and recommendations for improvement based on data analysis.

[0148] In this embodiment, step S6 not only ensures that the results of the distribution system reliability assessment can be clearly and accurately communicated to relevant personnel, but also helps to make scientific system maintenance and optimization decisions by providing data support and explanations. This method enhances the practicality and influence of the assessment and makes the management of the distribution system more efficient and purposeful.

[0149] Example:

[0150] 1. Reliability evaluation indicators

[0151] The reliability evaluation index of the distribution system is the basis of reliability analysis. By calculating the reliability index of the system, the reliability level of the system is quantitatively reflected. The reliability index of the distribution system can be divided into user reliability index and system reliability index according to the evaluation object. The user reliability index is used to describe the power supply reliability of each load point in the distribution system, and the system reliability index reflects the power supply capacity of the system.

[0152] In this embodiment, the reliability index includes a user reliability index and a system reliability index.

[0153] User reliability indicators:

[0154] According to the network structure and component parameters of the distribution network, the reliability level of each load point in the system is evaluated.

[0155] Average annual failure number: The expected number of failures per year at a load point, expressed in failures / year, denoted by λ.

[0156] Average annual failure time: the expected value of the total failure time of the load point per year, in hours, denoted as r.

[0157] Mean fault duration: the expected value of the duration of each fault at the load point, expressed in hours / time, denoted as U.

[0158] System reliability indicators:

[0159] In order to reflect the overall reliability level of the distribution system, some performance indicators are defined based on the load point reliability index to characterize the power outage statistics from the perspective of the system. j , U j 、E jare the annual average number of power outages, annual average failure time, and annual average power shortage at load point j, respectively. j is the number of users at load point j, so the following system reliability index can be calculated:

[0160] a) SAIFI, System Average Outage Frequency Index: The average number of consecutive power outages for each user within a predetermined period of time (usually one year).

[0161]

[0162] b) SA ID I, system average outage duration indicator: the average total failure time experienced by each user in the system within a specified time period.

[0163]

[0164] c) CA ID I, user average power outage duration indicator: the average fault duration of each user in the system within a specified time period.

[0165]

[0166] d) ASAI, average power supply availability index: the ratio of the total actual power supply hours to the total power supply hours required by users within a specified time.

[0167]

[0168] e) EENS, Expected Power Shortage: The total power shortage of the system during a specified period of time.

[0169] EENS=∑E j

[0170] f) AENS, average expected power shortage: the average power shortage for each user within a specified time.

[0171]

[0172] 2. Extraction of time-varying failure rate

[0173] Accurate failure rate is an important indicator for measuring component reliability and system reliability, and is also an important indicator for measuring power system reliability. Accurately predicting the time-varying failure rate can estimate the changing trend of the real power system reliability index. In addition, when users need a more reliable power supply, accurate reliability indicators are needed. If accurate reliability indicators are extracted, the system reliability can be effectively evaluated, and on this basis, the planning and operation of the distribution network can be better guided.

[0174] According to the cause of the failure, the failure can be divided into two categories. One is the aging failure caused by time, and the other is the random failure caused by unexpected events. According to the theory of the competing risk model, the failure rate of each failure type is calculated separately, and the final total failure rate can be expressed as the sum of the aging failure rate and the random failure rate.

[0175] The exponential distribution and Weibull distribution are used to describe the normal operation stage and the aging stage respectively. In the normal operation stage, the failure is an event caused by unexpected factors, there is no memory, and as a result, there is no connection between each failure. Therefore, the exponential distribution function can be used. In the aging stage, since the failure varies with the operating time of the equipment, the Weibull distribution function is used, which is easy to change the shape of the probability distribution function with parameter changes. Equations (2.1)-(2.3) represent the probability distribution function, average life and time-varying failure rate of aging failure respectively.

[0176]

[0177] Where η is the scale parameter, m is the shape parameter, and t is the device operating time.

[0178] However, these exponential and Weibull distribution functions of failure rates are independent of each other. Therefore, it is necessary to assume that the equipment has k different failure modes. In order to consider the competing risk model, the following three assumptions are required:

[0179] a) each failure mode leading to a particular type of failure operates independently of every other mode, at least until the failure occurs;

[0180] b) The device fails when the first of all competing failure processes reaches the failure state;

[0181] c) Each of the k failure models has a known distribution function.

[0182] Based on the above assumptions, the competing risk model can be used, and the reliability, probability distribution function and failure rate of the i-th failure mode can be expressed as R(t), F(t) and λ i (t), the formula is as follows:

[0183]

[0184] Therefore, the failure rate function of the component is:

[0185]

[0186] Where λ0 is the constant failure rate when the random failure model adopts exponential distribution.

[0187] The procedure for extracting time-varying failure rates is as follows: Figure 2When the failure is a random failure, the exponential distribution function is used to extract the constant failure rate from the operating time and the failure occurrence time. On the other hand, if the failure is an aging failure, the Weibull distribution function is used to extract the time-varying failure rate from the cumulative failure probability and regression analysis.

[0188] The random failure rate can be calculated by formula (2.8) as follows:

[0189]

[0190] Where a is the number of random failures, t i is the operating time of the faulty device, t j is the running time of normal equipment, Z1 and Z2 are the sets of faulty and normal equipment respectively.

[0191] The procedure for extracting aging failure rate is as follows: Figure 2 First, the Kaplan-Meier estimation method is used to solve the problem of different installation times of equipment in the overall range.

[0192]

[0193] Among them, r i It is the reverse value of the i-th fault data, R0=0.

[0194] Then, the failure data is sorted from low to high values ​​according to the failure time, and the Weibull probability plot is drawn. Next, the probability distribution function and trend line are estimated through the Weibull probability plot. Whether the result is good is determined as the goodness of fit. The horizontal axis and vertical axis of the Weibull probability plot are the natural logarithm of the life time t and the corrected cumulative probability of the failed device until time t, respectively, as shown in the following formula:

[0195] lnln[1 / (1-F(t))]=mln(t)-mln(η)|

[0196] The above equation is in the form of a first order function y = mx + K. Therefore, the parameters can be estimated based on the trend line extracted from the regression analysis. The shape parameter m is the slope of the trend line; the scale parameter η is expressed as exp(-k / m).

[0197] After estimating the Weibull parameters, the time-varying failure rate can be calculated by formula (2.7).

[0198] 3. Non-homogeneous Poisson distribution

[0199] Let N(t) denote the number of failures in time (0, t], then {N(t), t ≥ 0} is called a Poisson process of strength λ. The homogeneous Poisson process model describes a repairable system with the following characteristics:

[0200] The life parameter of the system follows an exponential distribution;

[0201] After the system fails, it is repaired as good as new, that is, the intervals between consecutive failures are independent and identically distributed.

[0202] When the intensity function of the homogeneous Poisson process is no longer a constant but is related to time t, that is, λ(t), the homogeneous Poisson process is generalized to a non-homogeneous Poisson process.

[0203] A non-homogeneous Poisson process satisfies the following five conditions:

[0204] a) N(0) = 0

[0205] b) N(t) has independent increments

[0206] c)P{N(s)-N(t)=1}=λ(t)h+o(h)

[0207] d) P{N(s)-N(t)≥0}=o(h)

[0208] e) For any t ≥ 0, s ≥ 0, N(t+s)-N(t) is is a Poisson distribution with parameters

[0209]

[0210] According to the time-varying failure rate model and the non-homogeneous Poisson process, the minimum path method is used to evaluate the reliability of the distribution system. This embodiment takes the typical urban distribution system of IEEE RBTS-BUS2 as an example. Figure 3 As shown, according to Figure 4 The reliability of the power distribution system is evaluated according to the evaluation process. The service life of the components is set to 40 years. Based on the time-varying failure rate, the expected value of the component failure rate is calculated year by year to obtain the reliability index of the load.

[0211] Calculate the average annual failure rate of LP 1 from 0 to 40 years on an annual basis, such as Figure 5 As shown. Figure 5 It can be seen that due to the consideration of the time-varying failure rate pattern, the trend of the annual average number of failures of the customer reliability index over time is shown as a component. It can be seen that within 0 to 30 years, the user reliability index shows a constant trend because the impact of component aging has not yet become prominent. In the period of 30 to 40 years, due to the aging of components, the failure rate of components continues to increase, and the corresponding user reliability index also deteriorates. From the perspective of the full life cycle of components, the impact of component aging on user reliability indicators cannot be ignored.

[0212] Figure 6 and 7The mean annual failure time and mean failure duration of LP 1 are shown. Their trends over time are similar to Figure 5 same.

[0213] Since the failure rate is almost unchanged in the first 30 years, the system reliability index is almost the same. Therefore, the system reliability index is divided into two periods: 0 to 30 years and 30 to 40 years, as shown in Table 1. For comparative analysis, the reliability index when the component failure rate is constant is given.

[0214] As can be seen from the table, the system reliability index from 0 to 30 years is almost the same as the index under constant failure rate. However, within 30 to 40 years, due to the deterioration of component failure rate, the system reliability index also decreases, especially ASAI, which can characterize the overall reliability of the system.

[0215] From the analysis results of load and system reliability indicators, it is necessary to consider the time-varying failure rate model of components in the reliability evaluation process of distribution system. The impact of equipment aging on failure rate cannot be ignored.

[0216] Table 1 System reliability index

[0217]

[0218] The power distribution reliability assessment device based on time-varying failure rate and non-homogeneous Poisson described below and the power distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson described above can be referenced to each other.

[0219] Please refer to the attached Figure 8 The present invention also provides a distribution reliability evaluation device based on time-varying failure rate and non-homogeneous Poisson, comprising:

[0220] The data collection module 100 is used to collect the operation data of the power distribution system;

[0221] Model building module 200, used to build a time-varying failure rate model and a non-homogeneous Poisson process model;

[0222] Reliability analysis module 300, used to perform system reliability analysis using the minimum path method;

[0223] A calculation module 400 is used to calculate various reliability indicators of the power distribution system;

[0224] The output module 500 is used to generate a reliability evaluation report.

[0225] The device of this embodiment can be used to execute the above method embodiment, and its principles and technical effects are similar, which will not be repeated here.

[0226] Please refer to the attached Fig. 9The present invention further provides a computer device 40, comprising: a processor 41 and a memory 42, wherein the memory 42 stores a computer program executable by the processor, and when the computer program is executed by the processor, the above method is executed.

[0227] The present invention further provides a storage medium 43 on which a computer program is stored. When the computer program is run by the processor 41, the above method is executed.

[0228] Among them, the storage medium 43 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0229] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson, characterized in that: The following steps are involved: Collect the operating data of the power distribution system, including the historical fault records and operating hours of each component; Establish a time-varying failure rate model for each component in the power distribution system; According to the time-varying failure rate model, a non-homogeneous Poisson process model is constructed; Use the minimum path method to identify critical transmission paths in the distribution system; Calculate reliability indicators of power distribution systems; Output the reliability assessment results of the power distribution system.

2. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson according to claim 1, characterized in that: The time-varying failure rate model includes modeling of component aging failure and random failure. The aging failure is described by Weibull distribution, and the random failure is described by exponential distribution.

3. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson according to claim 2, characterized in that: The aging failure is described by Weibull distribution. The shape parameter m and scale parameter η of the Weibull distribution are extracted from historical failure data through regression analysis. The shape parameter is used to describe the variation trend of the failure rate over time. The time-varying failure rate of the Weibull distribution is defined as: Where λ(t) is the instantaneous failure rate at time t, m is the shape parameter, and η is the scale parameter.

4. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson according to claim 2, characterized in that: The random failure is described by exponential distribution, the failure rate of random failure is defined as a constant λ0, and the failure rate of random failure is calculated by the following formula: in, is the constant failure rate of random failures, a is the number of random failures, t i is the operating time of the faulty device, t j is the running time of normal equipment, Z1 and Z2 are the sets of faulty and normal equipment respectively.

5. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson according to claim 1, characterized in that: The non-homogeneous Poisson process model is used to describe the failure behavior of components over time. The model calculates the expected number of events by the following formula: Where m(t) represents the expected number of failure events at time t, and λ(s) represents the instantaneous failure rate at time s.

6. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson according to claim 5, characterized in that: The probability distribution of the non-homogeneous Poisson process is given by the following formula: Wherein, P{N(t)=n} represents the probability of n failures occurring at time t, and m(t) is the expected number of failure events at time t.

7. A distribution reliability assessment method based on time-varying failure rate and non-homogeneous Poisson according to claim 5, characterized in that: The method further comprises the following steps for identifying and evaluating critical transmission paths in the power distribution system: Use the minimum path method to analyze all possible transmission paths in the distribution system to determine the critical paths that may affect the overall power supply reliability of the system; For each identified critical path, the impact of the time-varying failure rate of each component on the path on the reliability of the path is calculated based on the constructed non-homogeneous Poisson process model; The reliability assessment of all critical paths is integrated to determine the overall reliability level of the distribution system.

8. A power distribution reliability assessment device based on time-varying failure rate and non-homogeneous Poisson, used to implement the method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, used to collect distribution system operation data; Model building module for building time-varying failure rate models and non-homogeneous Poisson process models; Reliability analysis module, used to perform system reliability analysis using the minimum path method; A calculation module is used to calculate various reliability indicators of the power distribution system; Output module, used to generate reliability assessment report.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images