Equipment reliability parameter prediction method and system considering weather factors
By acquiring and analyzing the historical weather and external force data of the power system, establishing a model to predict equipment failure rate, solving the problem of insufficient accuracy of the failure rate model in the existing technology, and achieving higher prediction accuracy and reliability.
Patent Information
- Application Number
- CN202410926607.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-11
- Publication Date
- 2024-11-26
AI Technical Summary
The existing equipment failure rate model is insufficient in considering weather factors and external influences, and the calculation depends on sample number and statistical accuracy, which cannot effectively reflect the complex mechanism of action of internal and external factors and the individual differences in equipment.
By obtaining the historical weather data and external force impact data of the target power system, a weather condition model and external force failure model are established, combined with the preset correlation analysis logic, the equipment failure rate is predicted, and the LSTM neural network is used to predict rainfall, wind and temperature levels. The final weather score and external force damage health are calculated through the deduction table and weights, and the reliability parameters are corrected to improve the model accuracy.
Improve the accuracy and reliability of equipment failure rate prediction, can solve the model under some data conditions, use cloud model to analyze correlation, further enhance the accuracy of the model, and correct the reliability parameters through inverse problem theory, improving the model's Accuracy.
Smart Images

Figure CN119026451A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of equipment reliability parameter prediction taking weather factors into account, and in particular to a method and system for predicting equipment reliability parameters taking weather factors into account. Background Art
[0002] The basic mission of the power system is to provide users with continuous power supply in a safe, reliable and economical way. The output of renewable energy is random and volatile, and its proportion in the power system is gradually increasing, which may pose a threat to the safe and reliable operation of the power system. Therefore, it is crucial to evaluate the reliability of the power system and take timely intervention measures.
[0003] Although the reliability assessment of distribution systems has a relatively complete theoretical foundation, it still faces some problems and challenges. Statistical analysis of reliability parameters is the basis for improving the level of design and operation management, and is also the most basic and important work in the research of distribution system reliability. In order to establish an equipment failure rate model, it is necessary to use historical failure rate data and relevant knowledge of mathematical statistics to model from a mathematical perspective.
[0004] Existing equipment failure rate modeling methods can be divided into two types according to the degree of application of historical data: statistical models based on partial historical failure rates and statistical models based on complete historical failure rates.
[0005] The statistical model based on complete historical failure rate data only requires historical data such as equipment failure rate and service time, and uses common reliability analysis distributions (such as Weibull distribution, Poisson distribution, and lognormal distribution) to fit the failure rate time curve. This method is simple to understand and easy to implement, but it ignores the complex mechanism of internal and external influencing factors and individual differences in equipment, and its calculation accuracy is completely dependent on the number of samples and statistical accuracy.
[0006] The statistical model based on some historical failure rate data can make up for the shortcomings of the above model. It considers the impact of other factors on the failure rate while considering aging failure. Under the premise that massive data is constantly accumulating and power equipment is gradually being managed leanly, this method is more practical. However, the correlation between elements needs to be considered, which has a greater impact on the accuracy of the statistical model. Summary of the invention
[0007] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.
[0008] In view of the above existing problems, the present invention is proposed.
[0009] Therefore, the present invention provides a method and system for predicting equipment reliability parameters taking weather factors into account, which can solve the problems mentioned in the background technology.
[0010] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0011] In a first aspect, the present invention provides a method for predicting equipment reliability parameters taking weather factors into account, comprising:
[0012] Acquire historical weather data and external force impact data when equipment fails in the target power system, wherein the historical weather data includes rainfall, wind, and temperature level data, and the external force impact data includes at least impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors;
[0013] Establishing a weather condition model based on the historical weather data, and establishing an external force damage model based on the external force impact data;
[0014] According to the weather condition model and the external force damage model, combined with the preset correlation analysis logic, the equipment failure rate is predicted.
[0015] As a preferred solution of the equipment reliability parameter prediction method taking weather factors into account described in the present invention, it also includes: introducing the inverse problem theory of power system reliability assessment to correct the predicted reliability parameters.
[0016] As a preferred solution of the equipment reliability parameter prediction method taking weather factors into account described in the present invention, wherein: the acquisition of historical weather data and external force influence data when equipment fails in the target power system, the historical weather data includes rainfall, wind, and temperature level data, and the external force influence data includes at least the influence data of failure caused by animal factors, foreign body short circuit factors, and external construction influence factors, including:
[0017] The rainfall, wind and temperature grade data include a weather condition deduction table, where the rainfall grade is divided into moderate rain, heavy rain, rainstorm, yellow, orange and red, the wind grade is divided into level 3, level 4, level 5, level 6, blue, yellow, orange and red, and the temperature grade is divided into high temperature, orange and red;
[0018] Different levels correspond to different deduction mechanisms.
[0019] As a preferred solution of the equipment reliability parameter prediction method taking weather factors into account according to the present invention, wherein: the weather condition model established according to the historical weather data comprises:
[0020] Collect the rainfall, wind, and temperature level data for the past two years, and classify them according to the weather type in the weather condition penalty table. Put these data into three arrays, each of which includes training data and test data;
[0021] Build three LSTM neural networks, namely the rainfall level neural network, wind level neural network and temperature level neural network, use the training data to train each neural network, and then use the test data to perform performance tests on the trained neural networks to verify their prediction accuracy and performance, and obtain three independent neural networks to predict the rainfall level, wind level and temperature level at the next moment respectively;
[0022] Rainfall level neural network, the input of which is the rainfall level per hour in the previous three months, and the output is the statistical number of rainfall levels in the next week represents the total number of occurrences of rainfall level b1 in the next week, where the rainfall levels are divided into moderate rain, heavy rain, rainstorm, yellow, orange and red, b1∈{moderate rain, heavy rain, rainstorm, yellow, orange, red};
[0023] Wind level neural network, the input of which is the wind level per hour in the previous three months, and the output is the wind level statistics for the next week represents the total number of occurrences of wind level b2 in the next week, where wind levels are divided into level 3, level 4, level 5, level 6, blue, yellow, orange and red, b2∈{level 3, level 4, level 5, level 6, blue, yellow, orange, red};
[0024] Temperature level neural network, the input of the neural network is the temperature level of each hour in the previous three months, and the output is the total number of temperature levels in the next week It represents the total number of occurrences of temperature level b3 in the next week, where the temperature levels are divided into high temperature, orange and red, b3∈{high temperature, orange, red}.
[0025] As a preferred solution of the equipment reliability parameter prediction method taking weather factors into account according to the present invention, wherein: the establishment of a weather condition model according to the historical weather data further comprises:
[0026] The formula for calculating the deduction points for weather conditions in different regions is as follows:
[0027]
[0028] Among them, k j is the deduction value of the jth meteorological factor; The jth meteorological factor is at level b j The number of failures that occurred; is the occurrence level b of the jth meteorological factor jThe total number of times;
[0029] When calculating the final deduction points, each weather risk factor is deducted individually and weighted and accumulated. The final weather score S1 is as follows:
[0030]
[0031] Among them, ω j is the importance of the jth meteorological factor. The larger its value is, the more important the factor is in affecting the fault in the area.
[0032] As a preferred solution of the equipment reliability parameter prediction method taking weather factors into account according to the present invention, the establishment of the external force damage model according to the external force impact data includes:
[0033] The calculation of external force destruction health S2 is as follows:
[0034]
[0035] Among them, m i is the score of the i-th factor. The higher the score, the smaller the probability of equipment failure caused by this factor. i is the importance of the i-th factor. The larger the value, the more important the factor is. A represents the number of factors that affect the data.
[0036] As a preferred solution of the equipment reliability parameter prediction method taking weather factors into account according to the present invention, the prediction of equipment failure rate according to the weather condition model and the external force damage model combined with the preset correlation analysis logic includes:
[0037] The dynamic failure rate model of the device after the preset correlation analysis logic is as follows:
[0038]
[0039] Among them, λ e is the equipment failure rate at time t, α0, α1, α2 and β2 are the parameters of Weibull distribution, θ is the failure rate influencing factor, and T1 and T2 represent the preset equipment operation time.
[0040] In a second aspect, the present invention provides a device reliability parameter prediction system taking weather factors into account, comprising:
[0041] A data acquisition module, used to acquire historical weather data and external force impact data when equipment fails in the target power system, wherein the historical weather data includes rainfall, wind, and temperature level data, and the external force impact data includes at least impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors;
[0042] A model building module, used to build a weather condition model based on the historical weather data, and to build an external force damage model based on the external force impact data;
[0043] The prediction module is used to predict the equipment failure rate based on the weather condition model and the external force damage model in combination with a preset correlation analysis logic.
[0044] In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method when executing the computer program.
[0045] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method described above when executed by a processor.
[0046] Compared with the prior art, the present invention has the following beneficial effects: the present invention proposes a method and system for predicting equipment reliability parameters taking weather factors into account, obtaining historical weather data and external force influence data when equipment fails in the target power system, the historical weather data including rainfall, wind, and temperature level data, the external force influence data at least including the influence data of failures caused by animal factors, foreign body short circuit factors, and external construction influence factors; establishing a weather condition model based on the historical weather data, and establishing an external force damage model based on the external force influence data; predicting the equipment failure rate based on the weather condition model and the external force damage model, combined with the preset correlation analysis logic. Compared with the statistical model based on complete historical failure rate data, the present invention only needs part of the data to solve the model, and compared with the statistical model based on part of the historical failure rate data, the present invention uses the cloud model to analyze the correlation to make the model more accurate. In addition, the present invention corrects the reliability parameters based on the inverse problem theory, further improving the accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:
[0048] Figure 1 A method flow chart of a method and system for predicting equipment reliability parameters taking weather factors into account provided by one embodiment of the present invention;
[0049] Figure 2A schematic diagram of a two-state model of a device in a device reliability parameter prediction method and system taking weather factors into account provided by an embodiment of the present invention;
[0050] Figure 3 A schematic diagram of a conditional cloud generator of a device reliability parameter prediction method and system taking weather factors into account provided by one embodiment of the present invention;
[0051] Figure 4 A schematic diagram of cloud model inference rules of a device reliability parameter prediction method and system taking weather factors into account provided by one embodiment of the present invention;
[0052] Figure 5 A schematic diagram of a power distribution system structure of a method and system for predicting equipment reliability parameters taking weather factors into account provided by one embodiment of the present invention;
[0053] Figure 6 A device reliability parameter prediction method taking weather factors into account and a two-dimensional state cloud distribution map of weather conditions and external force damage health of the system provided by an embodiment of the present invention;
[0054] Figure 7 A cloud droplet diagram simulated by a Γ half cloud fall model of a device reliability parameter prediction method and system taking weather factors into account provided by an embodiment of the present invention;
[0055] Figure 8 A schematic diagram of a line failure rate fitting surface of a device reliability parameter prediction method and system taking weather factors into account provided by an embodiment of the present invention;
[0056] Fig. 9 An internal structural diagram of a computer device of a method and system for predicting equipment reliability parameters taking weather factors into account provided by one embodiment of the present invention. DETAILED DESCRIPTION
[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
[0058] Example 1
[0059] Reference Figure 1-Figure 9 , which is the first embodiment of the present invention, provides a method and system for predicting equipment reliability parameters taking weather factors into account, including:
[0060] In the related art, existing equipment failure rate modeling methods can be divided into two types according to the degree of application of historical data: a data statistical model based on partial historical failure rates and a data statistical model based on complete historical failure rates.
[0061] The statistical model based on complete historical failure rate data only requires historical data such as equipment failure rate and service time, and uses common reliability analysis distributions (such as Weibull distribution, Poisson distribution, and lognormal distribution) to fit the failure rate time curve. This method is simple to understand and easy to implement, but it ignores the complex mechanism of internal and external influencing factors and individual differences in equipment, and its calculation accuracy is completely dependent on the number of samples and statistical accuracy.
[0062] The statistical model based on some historical failure rate data can make up for the shortcomings of the above model. It considers the impact of other factors on the failure rate while considering aging failure. Under the premise that massive data is constantly accumulating and power equipment is gradually being managed leanly, this method is more practical. However, the correlation between elements needs to be considered, which has a greater impact on the accuracy of the statistical model.
[0063] The present application provides a method that can effectively solve the above-mentioned problems. Next, multiple embodiments will be combined to explain in detail how to implement the device reliability parameter prediction method taking weather factors into account.
[0064] It should be noted that the distribution network is composed of many primary and secondary power equipment, such as lines, transformers, circuit breakers, disconnectors and relay protection devices. Most of the equipment can be repaired, so in equipment reliability analysis, a two-state transition model is often used to describe the operating state of the equipment, namely the normal operation state and the fault repair state. Figure 2 shown.
[0065] In order to accurately establish the state model of the equipment, it is necessary to calculate the state transition parameters, including the failure rate λ and the repair rate μ. There are many common definitions of equipment failure rate, of which the two most widely used definitions are statistics and density functions. In the statistical method, the failure rate λ represents the number of equipment failures per unit time; in the density function, the equipment failure probability density function f(t) is used to calculate the probability of equipment failure in different time periods. As shown in the following formula.
[0066]
[0067] Where P is the probability value of equipment failure, R(t) is the reliability function, which can be expressed as R(t)=P(T>t), Δt is the time interval, and f(t) is the equipment failure probability density function.
[0068] In reliability analysis, different life distributions are usually used for fitting analysis according to actual conditions and data. Common life distributions of continuous variables include Poisson distribution, normal distribution, exponential distribution, Weibull distribution, and Gamma distribution. The mathematical characteristics of these life distributions are listed in Table 1, which can be used to analyze and predict the failure rate and reliability of equipment.
[0069] Table 1 Life distribution commonly used in reliability analysis
[0070]
[0071] It should be noted that, according to the failure rate data of the transformer operating years, the distribution function fitting method was adopted, and the Weibull distribution, Gamma distribution and Poisson distribution were compared. The goodness of fit test was performed by determining the coefficient and the residual sum of squares. The closer the determination coefficient is to 1 and the smaller the residual sum of squares is, the better the fitting effect of the model is. The results are shown in Table 2. Considering that the Weibull distribution has the largest determination coefficient, the smallest residual sum of squares, and the smallest test statistic. The Kolmogorov-Smirnov test was further used to determine the results as shown in Table 3.
[0072] According to Table 2 and Table 3, the estimation of unknowns of Gamma distribution requires the calculation of a series of special functions, while the fitting effect of Weibull distribution is the best and the estimation is relatively simple. Therefore, the present invention chooses to use a staged model and double Weibull distribution to describe the failure rate of the equipment, as shown in the following formula.
[0073]
[0074] Among them, α1≥0, α2≥0 are the shape parameters of the double Weibull distribution, and β2≥0 is the size parameter of the double Weibull distribution.
[0075] Table 2 Comparison of distribution function fitting results
[0076]
[0077] Table 3 K-S test results
[0078]
[0079] It should be noted that the cloud model is mainly used to convert uncertainty between qualitative concepts and quantitative descriptions. The principles include fuzziness (ambiguity of boundaries) and randomness (probability of event occurrence) in mathematics.
[0080] In the cloud model, the domain U = {x} is represented by a quantitative and precise numerical set, and T is a qualitative concept corresponding to the domain U. For each element x in T, its certainty u(x) for T can be regarded as a random number with a stable tendency.
[0081] For example, to describe whether a person is a "tall" qualitative concept, a cloud model can be used to describe the relationship between people of different heights and this concept. For example, a person with a height of 2 meters can be said to be 100% tall, while a person with a height of 1.7 meters may only have a membership of 55%, and a person with a height of 1.5 meters may only have a membership of 10%. Here, "tall" is a qualitative concept, and each person's height and their membership of this concept constitute a distribution of a cloud model, describing the uncertainty of this concept.
[0082] The overall characteristics of the cloud model can be described by three numerical characteristics, namely, the expected value E x , Entropy value E n and the super entropy value H e Expected E x It reflects the center of gravity of the cloud and is a typical sample point for concept quantification. n It is used to describe the degree of discreteness of cloud droplets, that is, the degree of uncertainty of qualitative concepts, including the measure of randomness and the range of cloud droplet values that can be accepted by the concept. The degree of condensation of cloud droplets is reflected by the super entropy H e , which describes the uncertainty of entropy, is an important indicator for measuring whether sample data can form a concept.
[0083] It should also be noted that the algorithm for generating clouds is called a cloud generator. When a specific value x in the domain U1 is given, the cloud generator generates the membership distribution of the specific value x in the qualitative concept T1. The cloud generator at this time is called an X-condition cloud generator. When a specific membership u(x) is given, the cloud generator generates the cloud droplet distribution of the qualitative concept T2 in the domain U2 that satisfies the membership. The cloud generator at this time is called a Y-condition cloud generator. It can be seen that the input and output of the X-condition cloud and the Y-condition cloud are inverse. Figure 3 The implementation schematic diagram of the X-condition cloud and Y-condition cloud generators is shown. By connecting the X-condition cloud generator and the Y-condition cloud generator, cloud language reasoning rules can be formed to realize uncertain reasoning between different concepts.
[0084] Figure 1 A method flow chart of a device reliability parameter prediction method and system taking weather factors into account is shown, including:
[0085] S101, obtaining historical weather data and external force impact data when equipment in the target power system fails, the historical weather data including rainfall, wind, and temperature level data, and the external force impact data including at least impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors;
[0086] Among them, historical weather data and external force impact data when equipment fails in the target power system are obtained. The historical weather data include rainfall, wind, and temperature level data. The external force impact data at least include the impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors. The rainfall, wind, and temperature level data include the development of a weather condition deduction table. The rainfall level is divided into moderate rain, heavy rain, rainstorm, yellow, orange, and red. The wind level is divided into level 3, level 4, level 5, level 6, blue, yellow, orange, and red. The temperature level is divided into high temperature, orange, and red.
[0087] It should be noted that different levels correspond to different deduction mechanisms.
[0088] In the embodiment of the present application, the weather condition deduction table is shown in Table 4. For equipment whose main functional parts are exposed to the outside world, the main cause of equipment failure is external factors, so it is necessary to model weather conditions and external force damage. According to the data, the meteorological factors of equipment failure usually include four risk factors: rainfall, lightning, wind and temperature. If a certain meteorological factor exceeds the threshold and becomes the main risk factor, the equipment failure rate model should adopt the component fragility curve related to the meteorological factor to establish an independent failure rate model. To this end, it is necessary to formulate a weather condition deduction table.
[0089] Table 4 Deduction values of different meteorological factors at different meteorological levels
[0090]
[0091]
[0092] S102, establishing a weather condition model based on historical weather data, and establishing an external force damage model based on external force impact data;
[0093] It should be noted that in order to make weather forecasts for the next moment using past weather data, an LSTM (Long Short-Term Memory) neural network can be used. The LSTM neural network consists of a series of connected storage units, each of which contains three "gates", namely the input gate, output gate, and forget gate, which are used to control the flow of information. The forget gate determines whether to remember or forget the information of the previous time, the input gate processes the information of the previous time, and the output gate passes the updated information of the current time to the next time. These three gates allow information to flow through special self-connected units, avoiding the gradient vanishing problem, thereby achieving long-term dependency learning.
[0094] Among them, establishing a weather condition model based on historical weather data includes: collecting rainfall, wind, and temperature level data for the past two years, and classifying them according to the weather types in the weather condition deduction table, and putting these data into three arrays, each of which includes training data and test data;
[0095] Build three LSTM neural networks, namely the rainfall level neural network, wind level neural network and temperature level neural network, use the training data to train each neural network, and then use the test data to perform performance tests on the trained neural networks to verify their prediction accuracy and performance, and obtain three independent neural networks to predict the rainfall level, wind level and temperature level at the next moment respectively;
[0096] Rainfall level neural network, the input of which is the rainfall level per hour in the previous three months, and the output is the statistical number of rainfall levels in the next week represents the total number of occurrences of rainfall level b1 in the next week, where the rainfall levels are divided into moderate rain, heavy rain, rainstorm, yellow, orange and red, b1∈{moderate rain, heavy rain, rainstorm, yellow, orange, red};
[0097] Wind level neural network, the input of which is the wind level per hour in the previous three months, and the output is the wind level statistics for the next week represents the total number of occurrences of wind level b2 in the next week, where wind levels are divided into level 3, level 4, level 5, level 6, blue, yellow, orange and red, b2∈{level 3, level 4, level 5, level 6, blue, yellow, orange, red};
[0098] Temperature level neural network, the input of the neural network is the temperature level of each hour in the previous three months, and the output is the total number of temperature levels in the next week It represents the total number of occurrences of temperature level b3 in the next week, where the temperature levels are divided into high temperature, orange and red, b3∈{high temperature, orange, red}.
[0099] Furthermore, building a weather condition model based on historical weather data also includes:
[0100] The formula for calculating the deduction points for weather conditions in different regions is as follows:
[0101]
[0102] Among them, k j is the deduction value of the jth meteorological factor; The jth meteorological factor is at level b j The number of failures that occurred; is the occurrence level b of the jth meteorological factor j The total number of times;
[0103] Furthermore, when calculating the final deduction points, each weather risk factor is deducted individually and weighted and accumulated. The final weather score S1 is as follows:
[0104]
[0105] Among them, ω j is the importance of the jth meteorological factor. The larger its value is, the more important the factor is in affecting the fault in the area.
[0106] Furthermore, the external force damage model is established based on the external force impact data, including:
[0107] The calculation of external force destruction health S2 is as follows:
[0108]
[0109] Among them, m i is the score of the i-th factor. The higher the score, the smaller the probability of equipment failure caused by this factor. i is the importance of the i-th factor. The larger the value, the more important the factor is. A represents the number of factors that affect the data.
[0110] It should be noted that when determining the importance of weather conditions and external force damage indicators, a combined weighting method can be used, which combines subjective and objective weighting methods to ensure the objectivity and accuracy of the weights. First, an expert group or decision maker can be invited to give the discriminant matrix of each indicator, and these discriminant matrices can be analyzed using the hierarchical analysis method to determine the initial weights. Next, the entropy weight method can be used to correct the subjective weights obtained previously. The entropy weight method is an objective weight determination method that can calculate the weights based on the information entropy between indicators, thereby avoiding the influence of human subjectivity.
[0111] S103, predicting the equipment failure rate based on the weather condition model and the external force damage model in combination with the preset correlation analysis logic.
[0112] In the embodiment of the present application, there is a certain correlation between the weather condition score S1, the external damage health level S2 and the failure rate influencing factor θ. Weather conditions and external damage are uncertain and random. The relationship between them can be described in fuzzy language as follows: when the external damage health level and weather conditions are both good, the failure rate influencing factor is generally small. In order to implement this language rule, a dual-condition cloud model is used. It is necessary to establish a two-dimensional state cloud representing the weather condition score and the external damage health level, and a one-dimensional Y condition cloud representing the comprehensive influencing factor of the failure rate. For example Figure 4 shown.
[0113] If the sample input data exceeds 10 groups, the cloud digital features obtained by reverse cloud computing have high accuracy, and the error is usually less than 0.01. Therefore, the present invention predicts the current failure rate influencing factor by using the historical data of the equipment in the past 10 years. Considering that the selected historical state data samples are small and the membership cannot be determined in advance by the forward cloud, the present invention first uses the second-order central moment and fourth-order central moment based on the sample to calculate the digital feature quantity of the cloud. The specific calculation steps are as follows:
[0114] Suppose there are N cloud droplet samples x i ,i=1,2,…,N。
[0115] According to x i Calculate the sample mean Sample variance (second-order center distance) Sample fourth-order center distance
[0116] Calculate the centroid of the sample, that is
[0117] Calculate the entropy of the sample, that is
[0118] Calculate the super entropy of the sample, that is
[0119] In the embodiment of the present application, a common two-dimensional normal cloud is used to describe the conversion relationship between S1 and S2 and the degree of bad equipment status. The membership of the cloud droplet describes the degree to which S1 and S2 belong to a low failure rate. The Γ half-fall cloud model is used to describe the conversion relationship between the comprehensive influencing factor of the failure rate and the failure rate, that is, as the comprehensive influencing factor of the failure rate increases, the possibility of equipment failure gradually increases, and the membership of the cloud droplet describes the degree to which the comprehensive influencing factor belongs to the good state of the equipment at time t.
[0120] In the embodiment of the present application, the corresponding reasoning steps are as follows:
[0121] With (E nD ,EnE ) is the expectation, (H nD ,H nE ) is the variance, and a two-dimensional random number (E) of the directional concept D and E is generated by the two-dimensional forward cloud generator. nD ′,E nE ′).
[0122] The membership of the two-dimensional forward cloud generator is calculated when the input is (S1, S2), as shown in the following formula:
[0123]
[0124] Among them, E xD is the expectation of the sample of directional concept D, E xE is the expectation of the sample of directional concept E.
[0125] E nZ For expectation, H nZ is the variance, and a one-dimensional random number E of the qualitative concept Z is generated by the one-dimensional half-cloud generator Γ nZ ′.
[0126] Calculate the output value θ(t) when the input value is (S1, S2) = E z -lnu·E n ' z
[0127] Since the distribution of comprehensive influencing factors of failure rate under the same S1 and S2 tends to be stable, they can be used to establish the connection between weather conditions, external force damage and influencing factors, thereby realizing the uncertain reasoning prediction of influencing factors.
[0128] According to the weather condition model and external force damage model, combined with the preset correlation analysis logic, the equipment failure rate is predicted including:
[0129] The dynamic failure rate model of the device (exposed device) after the preset correlation analysis logic is as follows:
[0130]
[0131] Among them, λ e is the equipment failure rate at time t, α0, α1, α2 and β2 are the parameters of Weibull distribution, θ is the failure rate influencing factor, and T1 and T2 represent the preset equipment operation time.
[0132] In an optional embodiment, the inverse problem theory of power system reliability assessment can be introduced to correct the predicted reliability parameters. The steps S101-S103 of the present application can quantify the impact of weather conditions and external force damage on component failure rates. This process is subjective and direct application to reliability assessment may lead to inaccurate results. Therefore, it is necessary to introduce the inverse problem theory of power system reliability assessment to correct the predicted reliability parameters.
[0133] The inverse problem of reliability evaluation (IPRE) theory of power system reliability evaluation aims to use known system reliability indicators to infer unknown or inaccurate component reliability parameters. This is an important research direction in the field of power system reliability evaluation. It not only helps to reveal the true operating status of the existing system, but also can be used to guide system planning and improve the stability and security of power supply. Reliability indicators of distribution systems, such as the average system interruption frequency index (SAIFI), average energy not supplied (AENS) and average system interruption duration index (SAIDI), are usually calculated by power grid companies based on the statistical number of interruptions, interruption duration and load interruption at each load point. By performing inverse reliability evaluation on the available reliability indicators, unknown or erroneous component reliability parameters can be quantified and corrected.
[0134] In an optional embodiment, the present invention assumes that the given reliability indicators are accurate and that, except for the component reliability parameter (CRP) to be determined, the remaining CRPs are known. In the present invention, the parameter to be corrected is the CRP obtained by the dynamic failure rate model of the device, i.e., the exposed device, and the parameter to be corrected is w e , σ e As shown in the following formula.
[0135]
[0136] Using the known reliability index and the functional relationship between the reliability index and the CRP to be determined, a nonlinear equation for expressing IPRE can be constructed. Its general form is as follows:
[0137]
[0138] Among them, N pais the number of CRPs required, N id is the number of known reliability indicators, is a known system reliability index, For CRP, Represents the functional relationship between reliability index and CRP.
[0139] In an alternative embodiment, the nonlinear equation for calculating IPRE is pa With N id The difference between the two, IPRE can be divided into the following three situations:
[0140] Ⅰ. If N pa =N id , then the nonlinear equations of IPRE are a set of Nid-dimensional nonlinear equations.
[0141] Ⅱ. If N pa <N id , then the nonlinear equation of IPRE can be transformed into a nonlinear optimization problem for obtaining the unknown CRP.
[0142] III. If N pa >N id , then the nonlinear equation of IPRE may have infinite solutions. In this case, in order to confirm a unique set of solutions, it is necessary to combine the relevant information of the actual engineering scenario and construct the corresponding inverse problem model.
[0143] In an optional embodiment, the calculation formula of the average service availability index (ASAI) is as follows:
[0144]
[0145] Among them, λ i and r i is the failure rate and repair time of component i, Nc is the number of components in the distribution system, Nload is the number of load points in the distribution system, Ni is the number of users who shed load when the i-th component fails, and Nj is the number of users connected to the j-th load point.
[0146] Substituting the known CRP value, the above formula can be converted to the following formula:
[0147]
[0148] Among them, a 1,i and b1 are simplified constants.
[0149] Similarly, the calculation formula for the average energy not supplied (AENS) is as follows:
[0150]
[0151] Among them, P i is the average load at load point i.
[0152] The calculation formula of the system average interruption frequency index (SAIFI) is as follows:
[0153]
[0154] The calculation formula of the customer average interruption frequency index (CAIFI) is as follows:
[0155]
[0156] Among them, M j is the number of load shedding users connected to load point j.
[0157] In an optional embodiment, according to the relevant principles of least squares estimation (LSE), a nonlinear optimization model (denoted as P1) can be obtained to solve the problem. The specific formula is as follows:
[0158]
[0159] x LB ≤x≤x UB
[0160] In this model, vector x represents the variables used in decision model P1, which usually include failure rate, repair rate, etc. i Represents input data, usually including known SAIFI, CAIFI and AENS index values. The upper and lower limits of CRP obtained from engineering experience are expressed as x UB 、x LB The reliability index analytical function is used to set the minimum sum of squares of the deviations between the calculated value of the reliability index of the objective function and the known value. LB ≤x≤x UB are the upper and lower bounds of x.
[0161] In practical applications, when N pa >N idWhen the model P1 is used, there may be an infinite number of solutions. Such a solution lacks practical significance and cannot be directly applied to engineering practice. It is necessary to combine the specific engineering scenario to determine the only solution, set boundary conditions on the basis of model P1, and thus construct an inverse problem model (i.e., model P2) corresponding to the actual scenario. This section will illustrate the construction method of model P2 by calculating the reliability parameters under the planning scenario. The goal of model P2 is to minimize the total cost of component transformation and optimize the reliability parameters of the design components. The calculation formula is:
[0162]
[0163] U i =λ i / (λ i +μ i )
[0164] i∈S COM
[0165] e j (λ,μ)≤y j j=1,2,…,N id
[0166] λ LB ≤λ≤λ UB ,μ LB vμ≤μ UB
[0167] Among them, S COM represents the set of components including the CRP to be requested; represents the investment cost of component i; represents the annual operation and maintenance cost of component i, the discount rate and the economic life of the equipment are represented by the annuity present value coefficient α; with U 0,i They represent the benchmark investment cost and benchmark unavailability rate of component i respectively; U max,i , U min,i Respectively represent the upper and lower limits of the unavailability rate of component i; τ i is a constant about component i; λ and μ represent the failure rate and repair rate vectors to be determined; e j (λ,μ) is the calculated value of the jth reliability index corresponding to λ and μ; λ LB , UB Respectively represent the upper and lower limits of the failure rate parameter vector to be determined; μ LB , μ UB Respectively represent the upper and lower limits of the repair rate parameter vector to be determined; y j represents the constraint of the jth indicator. The first formula represents the total cost consisting of component investment cost and operation and maintenance cost; the second formula indicates For U iThe third formula is the reliability index constraint to be satisfied; the fourth formula is the range of CRP to be determined. In model P2, the decision variable is the CRP to be optimized.
[0168] Model P1 is a general basic inverse problem model that is applicable to a variety of situations, including solving existing CRPs and determining future CRPs in planning scenarios. However, when Model P1 has infinite solutions, Model P2 is needed. Model P2 incorporates additional information about specific engineering problems to make it more targeted. The Model P2 example in this section focuses on reliability parameter optimization in a planning scenario, and can be applied to other problems by simply applying the corresponding additional information to improve the construction of Model P2.
[0169] In summary, the present invention proposes a method for predicting equipment reliability parameters taking weather factors into account, obtaining historical weather data and external force influence data when equipment fails in the target power system, the historical weather data including rainfall, wind, and temperature level data, the external force influence data at least including influence data of failures caused by animal factors, foreign body short circuit factors, and external construction influence factors; establishing a weather condition model based on the historical weather data, and establishing an external force damage model based on the external force influence data; predicting the equipment failure rate based on the weather condition model and the external force damage model, combined with preset correlation analysis logic. Compared with the statistical model based on complete historical failure rate data, the present invention only needs part of the data to solve the model, and compared with the statistical model based on part of the historical failure rate data, the present invention uses the cloud model to analyze the correlation to make the model more accurate. In addition, the present invention corrects the reliability parameters based on the inverse problem theory, further improving the accuracy of the model.
[0170] This embodiment also provides a device reliability parameter prediction system taking weather factors into account, including:
[0171] A data acquisition module is used to acquire historical weather data and external force impact data when equipment fails in the target power system. The historical weather data includes rainfall, wind, and temperature level data. The external force impact data at least includes impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors.
[0172] A model building module is used to build a weather condition model based on historical weather data and an external force damage model based on external force impact data;
[0173] The prediction module is used to predict the equipment failure rate based on the weather condition model and the external force damage model combined with the preset correlation analysis logic.
[0174] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.
[0175] This embodiment also provides a computer device, which may be a terminal, and its internal structure diagram may be as shown in FIG. Fig. 9 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for predicting equipment reliability parameters taking into account weather factors is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.
[0176] This embodiment further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:
[0177] Obtain historical weather data and external force impact data when equipment fails in the target power system. The historical weather data includes rainfall, wind, and temperature level data. The external force impact data includes at least the impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors.
[0178] Establish a weather condition model based on historical weather data, and establish an external force damage model based on external force impact data;
[0179] The equipment failure rate is predicted based on the weather condition model and the external force damage model, combined with the preset correlation analysis logic.
[0180] Example 2
[0181] Reference Figure 2-Figure 8 , which is an embodiment of the present invention, provides a method and system for predicting equipment reliability parameters taking weather factors into account. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.
[0182] Table 1 Load data
[0183]
[0184]
[0185] The present invention is based on the IEEE RBTS BUS 6-node power distribution system for example analysis. Figure 5 and as shown in Table 1. The system is also connected to 10 1.5MW photovoltaic panels, whose output power follows the Beta distribution with shape parameters of 2.06 and 2.5. The power line failure rate is 0.05f / (a·km), and the repair time is 4h; the circuit breaker failure rate is 0.002f / a, and the repair time is 4h; the indoor transformer failure rate is 0.015f / a, and the repair time is 200h; the switching time of the section switch is 1 minute. The historical statistical data of three groups of transformers in a regional distribution system are selected as the input for the prediction of the failure rate of transformers connected to load points LP1-LP3, including service years, weather conditions, external force damage and corresponding failure rates. S1, S2, expectation, entropy and super entropy are calculated, and these values are used as the input of the two-dimensional normal cloud model. In the cloud model, each cloud droplet represents a realization of a qualitative concept, so increasing the number of cloud droplets can better reflect the state characteristics of the qualitative concept. In this study, the number of cloud droplets was set to 5000, and the cloud droplet distribution map of the two-dimensional normal cloud was obtained, as shown in Figure 6 The results show that the two-dimensional normal cloud can effectively describe the probability distribution characteristics and uncertainties of the original cloud droplets.
[0186] The entropy of the cloud model is calculated according to the expected equation expression of the Γ cloud, and the super entropy is set to 0.05. Then, the membership degree obtained from the two-dimensional state cloud is used as input to obtain the one-dimensional influence factor cloud. Figure 7 The scatter plot of cloud droplet composition obtained by the Γ half-cloud model is shown. As the weather becomes more severe and the degree of external force damage increases, the value of the comprehensive impact factor also increases.
[0187] Since there is uncertainty in each reasoning process, the average value of the results of 1000 program runs is used as the final influencing factor result. When actually calculating the super entropy, the difference may be negative, resulting in a complex number obtained from the square root operation, which is actually meaningless. To solve this problem, the present invention considers changing the super entropy calculation formula to use the absolute value under the square root. Based on the obtained data set, the present invention calculates the equivalent service age of the line respectively, and uses it as the input of the equivalent operating life. The present invention uses the LM method to fit the optimal parameters, and the specific parameters are shown in Table 2.
[0188] Table 2 Optimal parameters
[0189] parameter <![CDATA[T2]]> <![CDATA[α0]]> <![CDATA[α1]]> <![CDATA[α2]]> <![CDATA[β2]]> Mean square error Correlation coefficient Numeric 3 0.01568 0.01907 0.0011 2.7377 0.0072 0.9904
[0190] Substitute the calculated parameters into the dynamic failure rate model shown, and draw the fitting surface as shown in Figure 8 As shown, it can be seen that the data samples are basically located on the fitting surface.
[0191] From the analysis of the surface, we can see that from the perspective of comprehensive influencing factors, the failure rate satisfies the trend of increasing upward slope as the influencing factor increases. In the time dimension, the failure rate increases with the increase of equivalent operating years. When both the comprehensive influencing factor and the equivalent operating years are large, the failure rate increases exponentially, which is consistent with the actual situation.
[0192] The results of applying the method proposed in the present invention to the IEEE RBTS BUS 6-node power distribution system are shown in Table 3. It can be seen that the accuracy of reliability evaluation can be guaranteed by using the method proposed in the present invention. Among them, the true value of the reliability evaluation index is the reliability index calculated based on the true value of CRP.
[0193] Table 3 Solution results of the proposed algorithm
[0194]
[0195] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
[0196] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.
[0197] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0198] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0199] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0200] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.
[0201] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. A method for predicting equipment reliability parameters taking weather factors into account, It is characterized in that include: Acquire historical weather data and external force impact data when equipment fails in the target power system, wherein the historical weather data includes rainfall, wind, and temperature level data, and the external force impact data includes at least impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors; Establishing a weather condition model based on the historical weather data, and establishing an external force damage model based on the external force impact data; According to the weather condition model and the external force damage model, combined with the preset correlation analysis logic, the equipment failure rate is predicted.
2. The equipment reliability parameter prediction method taking weather factors into account according to claim 1, It is characterized in that Also includes: The inverse problem theory of power system reliability assessment is introduced to correct the predicted reliability parameters.
3. The equipment reliability parameter prediction method taking weather factors into account as claimed in claim 2, It is characterized in that The acquisition of historical weather data and external force impact data when equipment fails in the target power system, wherein the historical weather data includes rainfall, wind, and temperature level data, and the external force impact data includes at least impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors, including: The rainfall, wind and temperature grade data include a weather condition deduction table, where the rainfall grade is divided into moderate rain, heavy rain, rainstorm, yellow, orange and red, the wind grade is divided into level 3, level 4, level 5, level 6, blue, yellow, orange and red, and the temperature grade is divided into high temperature, orange and red; Different levels correspond to different deduction mechanisms.
4. The equipment reliability parameter prediction method taking weather factors into account as claimed in claim 3, It is characterized in that The establishing of a weather condition model according to the historical weather data comprises: Collect the rainfall, wind, and temperature level data for the past two years, and classify them according to the weather type in the weather condition penalty table. Put these data into three arrays, each of which includes training data and test data; Build three LSTM neural networks, namely the rainfall level neural network, wind level neural network and temperature level neural network, use the training data to train each neural network, and then use the test data to perform performance tests on the trained neural networks to verify their prediction accuracy and performance, and obtain three independent neural networks to predict the rainfall level, wind level and temperature level at the next moment respectively; Rainfall level neural network, the input of which is the rainfall level per hour in the previous three months, and the output is the statistical number of rainfall levels in the next week Indicates that the rainfall level in the next week is b 1 The total number of occurrences of the rainfall levels are divided into moderate rain, heavy rain, rainstorm, yellow, orange and red, b 1 ∈{moderate rain, heavy rain, rainstorm, yellow, orange, red}; Wind level neural network, the input of which is the wind level per hour in the previous three months, and the output is the wind level statistics for the next week Indicates that the stroke level in the next week will be B 2 The total number of occurrences of wind levels is divided into 3, 4, 5, 6, blue, yellow, orange and red, b 2 ∈{level 3, level 4, level 5, level 6, blue, yellow, orange, red}; Temperature level neural network, the input of the neural network is the temperature level of each hour in the previous three months, and the output is the total number of temperature levels in the next week Indicates that the temperature level in the next week will be b 3 The total number of occurrences of the temperature levels are divided into high temperature, orange and red, b 3 ∈{high temperature, orange, red}.
5. The equipment reliability parameter prediction method taking weather factors into account according to claim 4, It is characterized in that The establishing of a weather condition model according to the historical weather data further comprises: The formula for calculating the deduction points for weather conditions in different regions is as follows: Among them, k j is the deduction value of the jth meteorological factor; The jth meteorological factor is at level b j The number of failures that occurred; is the occurrence level b of the jth meteorological factor j The total number of times; When calculating the final deduction score, each weather risk factor is deducted individually and weighted and accumulated. The final weather score S 1 As shown below: Among them, ω j is the importance of the jth meteorological factor. The larger its value is, the more important the factor is in affecting the fault in the area.
6. The equipment reliability parameter prediction method taking weather factors into account according to claim 5, It is characterized in that The establishing of the external force damage model according to the external force impact data comprises: External force destroys health S 2 The calculation is as follows: Among them, m i is the score of the i-th factor. The higher the score, the smaller the probability of equipment failure caused by this factor. i is the importance of the i-th factor. The larger the value, the more important the factor is. A represents the number of factors that affect the data.
7. The equipment reliability parameter prediction method taking weather factors into account according to claim 6, It is characterized in that The predicting of the equipment failure rate according to the weather condition model and the external force damage model in combination with the preset correlation analysis logic includes: The dynamic failure rate model of the device after the preset correlation analysis logic is as follows: Among them, λ e is the equipment failure rate at time t, α 0 , α 1 , α 2 and β 2 is the parameter of Weibull distribution, θ is the failure rate influencing factor, T 1 With T 2 Indicates the time when the preset device is running.
8. A equipment reliability parameter prediction system taking weather factors into account, It is characterized in that include: A data acquisition module, used to acquire historical weather data and external force impact data when equipment fails in the target power system, wherein the historical weather data includes rainfall, wind, and temperature level data, and the external force impact data includes at least impact data of failures caused by animal factors, foreign body short circuit factors, and external construction factors; A model building module, used to build a weather condition model based on the historical weather data, and to build an external force damage model based on the external force impact data; The prediction module is used to predict the equipment failure rate based on the weather condition model and the external force damage model in combination with a preset correlation analysis logic.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.
Citation Information
Cited By
Running state risk assessment and early warning method, system and equipment of electric power system under extreme weather condition and medium
CN119648469A
