WO2021185177A1 - 基于多数据采集的石化常压储油罐健康状态评估方法 - Google Patents

基于多数据采集的石化常压储油罐健康状态评估方法 Download PDF

Info

Publication number
WO2021185177A1
WO2021185177A1 PCT/CN2021/080526 CN2021080526W WO2021185177A1 WO 2021185177 A1 WO2021185177 A1 WO 2021185177A1 CN 2021080526 W CN2021080526 W CN 2021080526W WO 2021185177 A1 WO2021185177 A1 WO 2021185177A1
Authority
WO
WIPO (PCT)
Prior art keywords
oil storage
storage tank
health status
parameter
parameters
Prior art date
Application number
PCT/CN2021/080526
Other languages
English (en)
French (fr)
Inventor
张洪华
曾钦达
曾远跃
郑耿峰
黄春榕
林尧
林宁
肖婷
Original Assignee
福建省特种设备检验研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 福建省特种设备检验研究院 filed Critical 福建省特种设备检验研究院
Priority to US18/027,122 priority Critical patent/US20240028937A1/en
Publication of WO2021185177A1 publication Critical patent/WO2021185177A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the invention relates to the application field of equipment health evaluation, and in particular discloses a method for evaluating the health of petrochemical atmospheric oil storage tanks based on multi-data collection.
  • atmospheric storage tanks in the petrochemical industry mainly use a safety management mode that combines regular manual inspections and fixed-point monitoring and alarm systems to ensure safety in the reservoir area.
  • the tank specifications are large-scale and precise, and the existing mode detection appears to be relatively low in accuracy and efficiency.
  • the purpose of the present invention is to provide a method for evaluating the health of petrochemical atmospheric storage tanks based on multiple data collection, which scientifically comprehensively evaluates the health of the storage tanks, and improves the safety of the use of the storage tanks. .
  • the present invention adopts the following scheme to realize: a method for evaluating the health status of petrochemical atmospheric oil storage tanks based on multi-data collection.
  • the evaluation method includes the following steps: Step 1. Determine the influencing factors of the health status of the oil storage tank and the parameters of the influencing factors Collect and obtain the abnormal probability of each parameter;
  • Step 2 Establish the probability membership distribution function of abnormal parameters in the health state, and obtain the health state grade membership matrix under the influence of probability;
  • Step 3 Establish the distribution function of the degree of membership of the health status, and obtain the matrix of the degree of membership of the health status under the influence of the abnormal severity of the parameter;
  • Step 4 Obtain the membership degree vector of the abnormal severity of the parameter under the comprehensive influence to the health state
  • Step 5 Determine the health status of the dynamic monitoring parameters of the oil storage tank
  • Step 6 Establish the state set and state evaluation set of the oil storage tank, and obtain the weight coefficient of the importance of each basic parameter of the oil storage tank;
  • Step 7 Determine the degradation degree of each basic parameter of the oil storage tank
  • Step 8 Establish the judgment matrix of the deterioration degree of the basic parameters, and perform the fuzzy comprehensive evaluation of the basic parameters of the oil storage tank;
  • Step 9 Determine the basic health status of the oil storage tank according to the principle of maximum membership
  • Step 10 Take the health status of the dynamic monitoring parameters of the oil storage tank and the severity level in the basic health status of the oil storage tank to determine the final health status of the oil storage tank.
  • step 1 further specifically includes the following steps: step 11, through the analysis of the health status of the oil storage tank, the parameters selected for online monitoring include but are not limited to the five parameters: the temperature in the tank is denoted as parameter A, the pressure in the tank Mark it as parameter B, the liquid level in the tank as parameter C, the vibration data of the pipeline as parameter D, and the lightning protection grounding resistance as parameter E; the monitored parameters are collected and transmitted to the data processing server through the network;
  • Step 12 Compare each parameter with the corresponding set normal range value. If it exceeds the normal range, it will be recorded as abnormal, and the number of abnormalities will be counted for test data analysis;
  • Step 13 Analyze the test data to obtain the probability that the parameter is abnormal. The smaller the probability, the better the health of the oil storage tank.
  • step 2 further specifically includes the following steps: step 21, according to the characteristics of the p distribution of the abnormality probability of each parameter, within a set confidence interval, the lower the probability value of the monitored parameter abnormality, the health status The more it tends to be better, the triangular distribution is chosen as the distribution function of the probability of abnormality of the parameters in the healthy state.
  • step 21 according to the characteristics of the p distribution of the abnormality probability of each parameter, within a set confidence interval, the lower the probability value of the monitored parameter abnormality, the health status The more it tends to be better, the triangular distribution is chosen as the distribution function of the probability of abnormality of the parameters in the healthy state.
  • Step 21 Substitute the abnormal probability values corresponding to the monitored parameters A, B, C, D, and E into the probability membership distribution function, and the health status membership vectors under the influence of a single factor can be obtained as v A1 , V B1 , v C1 , v D1 , v E1 .
  • step 3 further specifically includes the following steps: step 31, setting the severity level q of the parameter abnormality, where the parameter abnormal severity and the probability of occurrence of the parameter abnormality have the same influence characteristics on the health status, and the triangular distribution is also selected As the distribution function of the degree of membership of the health status of the abnormal severity of the parameter, there is
  • Step 32 Select the maximum score value of each severity level and substitute it into the distribution function of the membership degree of the health status level, and the membership degree vectors of the health status under the influence of the abnormal severity of the single factor parameter are respectively v A2 , v B2 , v C2 , v D2 , v E2 .
  • step 4 is further specifically:
  • k is the parameters A, B, C, D, E;
  • the factor i is 1, 2;
  • j 1,...,5;
  • k is the parameters A, B, C, D, E;
  • the factor i is 1, 2;
  • j 1,...,5;
  • the vectors v A1 and v A2 , v B1 and v B2 , v C1 and v C2 , v D1 and v D2 , v E1 and v E2 are respectively composed of matrices V A , V B , V C , V D and V E , And substitute
  • H k R k ⁇ V k
  • k is the parameters A, B, C, D, E;
  • the five parameters A, B, C, D, and E that are feasible to the oil storage tank under the comprehensive influence of the probability of occurrence of parameter abnormality and the severity of parameter abnormality, the health status membership vectors are respectively H A , H B , H C , HD , HE .
  • step 5 is further specifically: setting the dynamic monitoring parameter health status level of the oil storage tank under the comprehensive influence of the dynamic monitoring parameter abnormality probability and the dynamic monitoring parameter abnormal severity: healthy, good, attention, deterioration, disease; then the principle of maximum degree of membership, the membership of a dynamic state of health tank vector H a, H B, H C , H D, H E can have storage tanks a, B, C, D, E corresponding to five kinds of parameters Monitoring parameter health status level.
  • step 7 is further specifically: for the basic parameter U1 of the date of commissioning and transformation, the degradation degree is calculated according to the actual use time of the oil storage tank; that is, the degradation degree calculation formula is:
  • i 1
  • T is the average failure life of the oil storage tank
  • k is the failure index, and k is 1 or 2;
  • X, Y, Z are coefficients whose values are between 0 and 1, 0 means health, 1 means complete deterioration; P 1 , P 2 , and P 3 are the weights of designers, quality inspectors, and experts in the industry, respectively ;
  • t is the use time of the oil storage tank
  • T is the average failure life of the oil storage tank
  • k is the failure index, and k is 1 or 2;
  • step 8 is further specifically:
  • the membership degree of the health status level is obtained, and the membership degree function of the ridge distribution is adopted:
  • R i (r I (l i ), r II (l i ), r III (l i ), r IV (l i ), r V (l i ))
  • step 9 is further specifically: the value of the oil storage tank belonging to health, good, attention, deterioration, and disease can be obtained from the result of the fuzzy comprehensive evaluation, and then the basic parameters of the storage tank can be judged according to the principle of maximum membership. What is the state of health, goodness, attention, deterioration, and disease.
  • the present invention discloses a method for evaluating the health status of petrochemical atmospheric oil storage tanks based on multi-data collection, using sensor monitoring to collect data related to equipment failure and safety, and combining basic data of the storage tank , A scientific and comprehensive assessment of the health of the oil storage tank not only improves the safety of the oil storage tank, but also ensures the service life of the petrochemical atmospheric oil storage tank.
  • Figure 1 is a schematic flow diagram of the method of the present invention.
  • a method for evaluating the health status of petrochemical atmospheric storage tanks based on multiple data collection includes the following steps: Step 1. Determine the factors affecting the health of the storage tanks. The parameters of the factors are collected and the abnormal probability of each parameter is obtained;
  • Step 2 Establish the probability membership distribution function of abnormal parameters in the health state, and obtain the health state grade membership matrix under the influence of probability;
  • Step 3 Establish the distribution function of the degree of membership of the health status, and obtain the matrix of the degree of membership of the health status under the influence of the abnormal severity of the parameter;
  • Step 4 Obtain the membership degree vector of the abnormal severity of the parameter under the comprehensive influence to the health state
  • Step 5 Determine the health status of the dynamic monitoring parameters of the oil storage tank
  • Step 6 Establish the state set and state evaluation set of the oil storage tank, and obtain the weight coefficient of the importance of each basic parameter of the oil storage tank;
  • Step 7 Determine the degradation degree of each basic parameter of the oil storage tank
  • Step 8 Establish the judgment matrix of the deterioration degree of the basic parameters, and perform the fuzzy comprehensive evaluation of the basic parameters of the oil storage tank;
  • Step 9 Determine the basic health status of the oil storage tank according to the principle of maximum membership
  • Step 10 Take the health status of the dynamic monitoring parameters of the oil storage tank and the severity level in the basic health status of the oil storage tank to determine the final health status of the oil storage tank.
  • Step S1 Considering the factors affecting the safety of the oil storage tank, the following online monitoring parameters are selected including but not limited to the five parameters: tank temperature (denoted as A), tank pressure (denoted as B) ), the liquid level in the tank (denoted as C), the vibration data of important pipelines (denoted as D), and the lightning protection grounding resistance (denoted as E). Corresponding parameter collection sensors are respectively installed in the appropriate part of the oil storage tank.
  • Step S2 The data collection device collects and preprocesses the data of each sensor, and the preprocessed data is transmitted to the data processing server through the network, and the data is processed and managed.
  • the health status evaluation system compares each parameter with the corresponding set normal range value. If it exceeds the range, it will be recorded as abnormal, and the number of abnormalities will be counted; used for test data analysis;
  • Step S3 Analyze the probability of parameter abnormality through test data analysis. The smaller the probability, the better the health of the storage tank, that is, the probability of each parameter abnormality (the number of abnormalities in the official operation days) /Official operating days);
  • Step S4 The severity level of abnormal parameters corresponding to each parameter: I (strong), II (strong), III (medium), IV (mild); see Table 1 below
  • Step S5 Establish a membership distribution function. According to the characteristics of the abnormal probability distribution, within a certain confidence interval, the smaller the probability value of the abnormality of the monitoring parameter, the better the health status.
  • the triangular distribution can be selected as the distribution function of the health status membership degree of the parameter abnormal probability factor, as follows:
  • Step S6 Calculate the membership degree vector of the health state
  • the membership degree vectors of the health state under the influence of a single factor can be obtained as
  • Step S7 Determine the health level of the atmospheric storage tank under the influence of probability
  • the respective values of the health status levels under the influence of abnormal state parameters of A, B, C, D, and E can be obtained (the health status levels are divided into "healthy”, “good”, “attention”, “Deterioration” and “Disease”).
  • Step S8 Establish a severity level scoring standard
  • the severity level is scored using a 10-point system. Levels I to IV correspond to 1 to 10 points, and each level corresponds to 2 to 3 points. For easy analysis, the corresponding scores can be compressed to between 0.1 and 1.0, as shown in Table 2. List.
  • Severity level Grading Compressed scoring criteria IV (mild) 1, 2, 3 0.1, 0.2, 0.3 III (medium) 4, 5, 6 0.4, 0.5, 0.6
  • Step S9 Establish the distribution function of the abnormal severity of the parameter
  • the abnormal severity of parameters and the probability of abnormal parameters have the same effect on the health status, so the triangular distribution is also selected as the health status membership distribution function of the abnormal severity of the parameters.
  • Step S10 Calculate the membership degree vector of the abnormal severity of the dynamic monitoring parameter to the health state
  • Step S11 Determine the health level of the atmospheric oil storage tank under the influence of severity
  • the respective values of the health status levels under the influence of abnormal state parameters of A, B, C, D, and E can be obtained (the health status levels are divided into “healthy”, “good”, “attention”, Five kinds of "deterioration” and “disease”).
  • Step S12 Calculate the membership degree vector of severity to the health state under the comprehensive influence
  • k is the parameters A, B, C, D, E;
  • the factor i is 1, 2;
  • j 1,...,5;
  • k is the parameters A, B, C, D, E;
  • the factor i is 1, 2;
  • j 1,...,5;
  • the vectors v A1 and v A2 , v B1 and v B2 , v C1 and v C2 , v D1 and v D2 , v E1 and v E2 are respectively composed of matrices V A , V B , V C , V D and V E , And substitute
  • H k R k ⁇ V k
  • k is the parameters A, B, C, D, E;
  • the five parameters A, B, C, D, and E that are feasible to the oil storage tank under the comprehensive influence of the probability of occurrence of parameter abnormality and the severity of parameter abnormality, the health status membership vectors are respectively H A , H B , H C , HD , HE .
  • the health status membership vector v A1 of the A parameter under the influence of factor 1 and the health status membership vector v A2 of the A parameter under the influence of factor 2 are used as the comparison series, and v 0j is the reference sequence to solve the correlation coefficient, correlation and weight , Get the weight vector R A , specifically:
  • Step S1 Solve the correlation coefficient:
  • k is the parameter A (k is the corresponding parameter when calculating the weight vector of other parameters)
  • the factor i is 1, 2;
  • j is 1 (ask the other j types of health status vector v 0j, the corresponding parameters are 2, 3, 4, 5);
  • k is the parameter A (k is the corresponding parameter when calculating the weight vector of other parameters);
  • the factor i is 1, 2;
  • j is 1 (ask the other j types of health status vector v 0j, the corresponding parameters are 2, 3, 4, 5);
  • Step S2 then take this case 2,3,4,5 j, respectively, then v 0j as a reference sequence, according to the calculation at step S1, the parameter A to obtain the degree of association r A1j, r A2j;
  • the weight vector R A (r' A1 , r'A2 ) can be calculated
  • Step S3 According to the method of step S1 and step S2, k is substituted by B, C, D, and E respectively. Similarly, R B , R C , R D , and R E can be obtained;
  • the vectors v A1 and v A2 , v B1 and v B2 , v C1 and v C2 , v D1 and v D2 , v E1 and v E2 are respectively composed of matrices V A , V B , V C , V D and V E , And substitute
  • H k R k ⁇ V k
  • k is the parameters A, B, C, D, E;
  • the five parameters A, B, C, D, and E that are feasible to the oil storage tank under the comprehensive influence of the probability of occurrence of parameter abnormality and the severity of parameter abnormality, the health status membership vectors are respectively H A , H B , H C , HD , HE .
  • Step S13 Determine the dynamic monitoring health status level of the oil storage tank
  • the three dynamic monitoring parameters A, B, C, D, and E of the oil storage tank can be obtained.
  • the dynamic monitoring health status levels under the combined influence of the dynamic monitoring parameter abnormality probability and the dynamic monitoring parameter abnormal severity are: (health status The level is divided into five categories: "health”, “good”, “attention”, “worse” and “disease”).
  • the basic data on the health status of oil storage tanks mainly include: the date of commissioning and renovation, the installation quality of coating, insulation and lining, the data of the previous inspection and testing of atmospheric storage tanks, the construction materials of each layer of wall and bottom plate, Nominal thickness, these four basic data are sequentially compiled as U1, U2, U3, U4.
  • Step S14 Determine the state set and state evaluation set of the oil storage tank
  • V (I, II, III, IV, V)
  • Step S15 Determine the importance of the basic data
  • Step S16 Determine the degradation degree of each basic data
  • the average failure life is determined based on the design life and other design data and a large number of statistical data, and the degradation degree calculation formula is:
  • X, Y, Z are coefficients, and their values are between 0 and 1, 0 means health, 1 means complete deterioration; P 1 , P 2 , and P 3 are designers, quality inspectors, and experts in the industry, respectively The weight of, its value reflects the level and authority of the scoring personnel; 2) Calculate the final degradation degree based on the average failure life determined based on the design life and other data and a large number of statistical data.
  • t is the use time of the oil storage tank
  • T is the average failure life of the oil storage tank
  • k is the failure index, and k is 1 or 2;
  • Step S17 establishes a basic parameter deterioration degree judgment matrix
  • the membership degree of the health status level is calculated. Because the ridge-shaped distribution has the characteristics of wide main value interval and smooth transition zone, it can better reflect the fuzzy relationship of the state space of the deterioration degree of the oil storage tank. Therefore, the ridge-shaped distribution is adopted. Distribution membership function:
  • R i (r I (l i ), r II (l i ), r III (l i ), r IV (l i ), r V (l i ))
  • Step S18 perform fuzzy comprehensive evaluation of the basic parameters of the oil storage tank
  • W (W 1 , W 2 , W 3 , W 4 )
  • Step S19 Comprehensively monitor the health status and basic health status of the storage tank to confirm the final status
  • the basic health status of the oil storage tank (result of step S18) take the more serious level of the health status of the oil storage tank dynamic monitoring parameters and the basic health status of the oil storage tank as the final health State evaluation value.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Computational Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Algebra (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Fuzzy Systems (AREA)
  • Biomedical Technology (AREA)
  • Automation & Control Theory (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Mining & Mineral Resources (AREA)
  • Operations Research (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Agronomy & Crop Science (AREA)
  • Animal Husbandry (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)

Abstract

一种基于多数据采集的石化常压储油罐健康状态评估方法，其中，常压储油罐的健康状态受多种因素的影响，通过采集相应传感数据，再综合储油罐基础数据情况对储油罐的健康状态进行评估，取储油罐动态监测参数健康状态和储油罐基础健康状态中的严重级别，确定最终储油罐的健康状态，该方法对储油罐健康状态进行科学的综合评估，提高了储油罐使用的安全性。

Description

基于多数据采集的石化常压储油罐健康状态评估方法 技术领域
本发明涉及装备健康状态评估应用领域，特别是公开了一种基于多数据采集的石化常压储油罐健康状态评估方法。
背景技术
目前石化行业常压储油罐主要通过定期人工巡检和定点监测警报系统相结合的安全管理模式来保障库区安全问题，但是随着石化产业的蓬勃发展、储油罐规模日益扩大、储油罐规格大型化、精密化，现有模式检测显得精度和效率较为低下。
影响储油罐的安全因素众多，目前还没有一种涉及石化储油罐的多因素综合，高效、科学的健康状态评估系统。
发明内容
为克服上述问题，本发明的目的是提供一种基于多数据采集的石化常压储油罐健康状态评估方法，对储油罐健康状态进行科学的综合评估，提高了储油罐使用的安全性。
本发明采用以下方案实现：一种基于多数据采集的石化常压储油罐健康状态评估方法，所述评估方法包括如下步骤：步骤1、确定储油罐健康状态影响因素，对影响因素的参数进行采集并得到每种参数发生异常的概率；
步骤2、建立健康状态下参数发生异常的概率隶属度分布函数，获取概率影响下健康状态等级隶属度矩阵；
步骤3、建立健康状态等级隶属度分布函数，获取参数异常严酷度影响下健康状态等级隶属度矩阵；
步骤4、获取综合影响下参数异常严酷度对健康状态隶属度向量；
步骤5、确定储油罐动态监测参数健康状态；
步骤6、建立储油罐状态集和状态评价集，获取储油罐各基础参数重要度权重系数；
步骤7、确定储油罐各基础参数劣化度；
步骤8、建立基础参数劣化度判断矩阵，进行储油罐基础参数模糊综合评估；
步骤9、按最大隶属度原则确定储油罐基础健康状态；
步骤10、取所述储油罐动态监测参数健康状态和所述储油罐基础健康状态中的严重级别，确定最终储油罐的健康状态。
进一步的，所述步骤1进一步具体包括如下步骤：步骤11、通过储油罐健康状态影响分析，选取在线监测的参数包括但不限于该五项参数：罐内温度记为参数A、罐内压力记为参数B、罐内液位记为参数C、管道的振动数据记为参数D、防雷接地电阻记为参数E；对监测的参数进行采集经过网络传输至数据处理服务器；
步骤12、对每种参数与对应设置好的正常范围值进行比对，若超出正常范围则记为异常，统计异常次数，用于测试数据分析；
步骤13、通过测试数据分析得到参数发生异常的概率，概率越小，则储油罐健康状态越好。
进一步的，所述步骤2进一步具体包括如下步骤：步骤21、根据每种参数发生异常概率p分布的特性，在一设定的置信区间内，监测的参数异常发生的概率值越小，健康状态越趋于优，则选择三角分布作为健康状态下参数发生异常的概率隶属度分布函数，有：
Figure PCTCN2021080526-appb-000001
Figure PCTCN2021080526-appb-000002
Figure PCTCN2021080526-appb-000003
Figure PCTCN2021080526-appb-000004
Figure PCTCN2021080526-appb-000005
步骤21、将监测的参数A、参数B、参数C、参数D、参数E对应的发生异常的概率值代入概率隶属度分布函数，可得单因素影响下的健康状态隶属度向量分别为v A1、v B1、v C1、v D1、v E1。
进一步的，所述步骤3进一步具体包括如下步骤：步骤31、设置参数异常的严酷度级别q，其中，参数异常严酷度和参数异常发生的概率对健康状态的影响特性相同，则同样选取三角分布作为参数异常严酷度的健康状态等级隶属度分布函数，有
Figure PCTCN2021080526-appb-000006
Figure PCTCN2021080526-appb-000007
Figure PCTCN2021080526-appb-000008
Figure PCTCN2021080526-appb-000009
Figure PCTCN2021080526-appb-000010
步骤32、选取各严酷度级别的最大评分值代入健康状态等级隶属度分布函数，可得单因素参数异常严酷度影响下健康状态隶属度向量分别为v A2、v B2、v C2、v D2、v E2。
进一步的，所述步骤4进一步具体为：
将动态监测参数异常概率影响下各参数的健康状态隶属度向量v A1、v B1、v C1、v D1、v E1和参数异常严酷度影响下各参数的健康状态隶属度向量v A2、v B2、v C2、v D2、v E2与第j种健康状态等级向量v 0j分别进行灰色关联；其中，j为健康状态等级分健康、良好、注意、恶化和疾病，记作1，…，5；即向量v 0j表示为：v 01＝(1,0,0,0,0)、v 02＝(0,1,0,0,0)、v 03＝(0,0,1,0,0)、v 04＝(0,0,0,1,0)、v 05＝(0,0,0,0,1)；
依据式
Figure PCTCN2021080526-appb-000011
式中m为1，…，5；
k为参数A、B、C、D、E；
因素i为1，2；
j为1，…，5；
Figure PCTCN2021080526-appb-000012
为二级最小差，
Figure PCTCN2021080526-appb-000013
为二级最大差，|v 0j(m)-v ki(m)|为绝对差值；
求得ξ kij(m)
再利用式
Figure PCTCN2021080526-appb-000014
式中m为1，…，5；
k为参数A、B、C、D、E；
因素i为1，2；
j为1，…，5；
求得r kij
再利用式
Figure PCTCN2021080526-appb-000015
计算得到r’ ki
能计算得到权重向量R k＝(r’ k1,r’ k2),即：R A＝(r’ A1,r’ A2),R B＝(r’ B1,r’ B2),R C＝(r’ C1,r’ C2),R D＝(r’ D1,r’ D2),R E＝(r’ E1,r’ E2),
由v A1与v A2、v B1与v B2、v C1与v C2、v D1与v D2、v E1与v E2向量分别组成矩阵V A、V B、V C、V D和V E，
Figure PCTCN2021080526-appb-000016
并代入
H k＝R k·V k
式中k为参数A、B、C、D、E；
可行到储油罐的A、B、C、D、E五种参数在参数异常发生概率和参数异常严酷度综合影响下的健康状态隶属度向量分别为H A、H B、H C、H D、H E。
进一步的，所述步骤5进一步具体为：设置动态监测参数异常概率与动态监测参数异常严酷度综合影响下的储油罐动态监测参数健康状态等级为：健康、良好、注意、恶化、疾病；则根据最大隶属度原则，通过健康状态隶属度向量H A、H B、H C、H D、H E能得储油罐的A、B、C、D、E五种参数对应的储油罐动态监测参数健康状态等级。
进一步的，所述步骤6进一步具体为：所述储油罐各基础参数包括投用、改造日期，涂层、保温和衬里的安装质量，常压储油罐历次检验和检测情况数据，各层壁板和底板的建造材料、名义厚度，将此四项基础数据依次编为U1，U2，U3，U4；根据储油罐各基础数据，则储油罐状态集为：U＝(U1，U2，U3，U4)；根据储油罐动态监测参数健康状态等级：健康、良好、注意、恶化、疾病；则设定储油罐的健康状态等级分别对应为I，II，III，IV，V，则储油罐状态评价集为G＝(I，II，III，IV，V)；根据储油罐状态集和状态评价集，确定四项基础参数的权重系数分别为：权重W 1、权重W 2、权重W 3、权重W 4。
进一步的，所述步骤7进一步具体为：针对投用、改造日期的基础参数U1，根据储油罐实际使用时间计算劣化度；即劣化度计算公式为：
l i＝(t/T) k
式中：i＝1,t为储油罐的使用时间；T为该储油罐的平均故障寿命；k为故障指数，k取1或2；
针对涂层、保温和衬里的安装质量U2，常压储油罐历次检验和检测情况数据U3，各层壁板和底板的建造材料、名义厚度U4，这些基础参数先经过劣化度估算公式：
l i′＝(X·P 1+Y·P 2+Z·P 3)/(P 1+P 2+P 3)，i＝2,3,4
式中：X，Y，Z为系数其值介于0～1之间，0代表健康，1代表完全劣化；P 1、P 2、P 3分别为设计人员、质检人员、行内专家的权重；
求解，再结合储油罐的平均故障寿命计算，利用公式：
Figure PCTCN2021080526-appb-000017
式中：t为储油罐的使用时间；T为该储油罐的平均故障寿命；k为故障指数，k取1或2；
计算基础参数U2，U3，U4的劣化度。
进一步的，所述步骤8进一步具体为：
根据各基础参数劣化度求其健康状态等级的隶属度，采用岭形分布隶属度函数：
Figure PCTCN2021080526-appb-000018
Figure PCTCN2021080526-appb-000019
Figure PCTCN2021080526-appb-000020
Figure PCTCN2021080526-appb-000021
Figure PCTCN2021080526-appb-000022
由此可得到以劣化度为评价标准的模糊评判矩阵为：
R i＝(r I(l i)，r II(l i)，r III(l i)，r IV(l i)，r V(l i))
Figure PCTCN2021080526-appb-000023
则储油罐基础参数的模糊综合评估：
E＝W·R
其中W为四项基础参数的权重系数W＝(W 1，W 2，W 3，W 4)。
进一步的，所述步骤9进一步具体为：从模糊综合评估结果能得到该储油罐属于健康、良好、注意、恶化、疾病的数值，再按最大隶属度原则能判断储油罐基础参数所处的是健康、良好、注意、恶化、疾病中的哪一个状态。
本发明的有益效果在于：本发明公布了一种基于多数据采集的石化常压储油罐健康状态评估方法，运用传感监测收集与设备故障、安全相关的数据，结合合储油罐基础数据，对储油罐健康状态进行科学的综合评估，不仅提高了储油罐使用的安全性，而且也确保了石化常压储油罐的使用寿命。
附图说明
图1是本发明的方法流程示意图。
具体实施方式
下面结合附图对本发明做进一步说明。
请参阅图1所示，本发明的一种基于多数据采集的石化常压储油罐健康状态评估方法，所述评估方法包括如下步骤：步骤1、确定储油罐健康状态影响因素，对影响因素的参数进行采集并得到每种参数发生异常的概率；
步骤2、建立健康状态下参数发生异常的概率隶属度分布函数，获取概率影响下健康状态等级隶属度矩阵；
步骤3、建立健康状态等级隶属度分布函数，获取参数异常严酷度影响下健康状态等级隶属度矩阵；
步骤4、获取综合影响下参数异常严酷度对健康状态隶属度向量；
步骤5、确定储油罐动态监测参数健康状态；
步骤6、建立储油罐状态集和状态评价集，获取储油罐各基础参数重要度权重系数；
步骤7、确定储油罐各基础参数劣化度；
步骤8、建立基础参数劣化度判断矩阵，进行储油罐基础参数模糊综合评估；
步骤9、按最大隶属度原则确定储油罐基础健康状态；
步骤10、取所述储油罐动态监测参数健康状态和所述储油罐基础健康状态中的严重级别，确定最终储油罐的健康状态。
下面对本发明进一步说明：步骤S1：综合考虑影响储油罐安全的因素，选取以下在线监测的参数包括但不限于该五项参数：罐内温度(记为A)、罐内压力(记为B)、罐内液位(记为C)，重要管道的振动数据(记为D)，防雷接地电阻(记为E)。在储油罐合适部分别装设相应参数采集传感器。
步骤S2：数据采集设备采集各传感器数据并进行预处理，预处理后的数据经过网络传输至数据处理服务器，对数据进行处理、管理。健康状态评估系统对每种参数与对应设置好的正常范围值进行比对，若超出范围则记为异常，统计异常次数；用于测试数据分析；
步骤S3：通过测试数据分析得到参数发生异常的概率，概率越小，则储油罐健康状态越好，即根据历史正式运行天数统计每种参数发生异常的概率(正式运行天数内发生异常的 次数/正式运行天数)；
步骤S4：每种参数对应的参数异常严酷度级别：I(强)，II(较强)，III(中等)，IV(轻度)；见下表1
监测参数 参数异常概率(p) 参数异常严酷度级别(q)
A 历史正式运行天数统计p A 专家评估q A
B 历史正式运行天数统计p B 专家评估q B
C 历史正式运行天数统计p C 专家评估q C
D 历史正式运行天数统计p D 专家评估q D
E 历史正式运行天数统计p E 专家评估q E
表1
步骤S5：建立隶属度分布函数。根据异常概率分布的特性，在一定的置信区间内，监测参数异常发生的概率值越小，健康状态越趋于优。可选择三角分布作为参数异常概率因素的健康状态隶属度分布函数，有：
Figure PCTCN2021080526-appb-000024
Figure PCTCN2021080526-appb-000025
Figure PCTCN2021080526-appb-000026
Figure PCTCN2021080526-appb-000027
Figure PCTCN2021080526-appb-000028
步骤S6：计算健康状态隶属度向量；
将A、B、C、D、E状态参数异常概率值代入隶属度分布函数，可得单因素影响下的健康状态隶属度向量分别为
Figure PCTCN2021080526-appb-000029
步骤S7：确定概率影响下常压储油罐健康状态等级；
按最大隶属度原则，根据S6计算结果，可得到A、B、C、D、E状态参数异常影响下的健康状态等级分别值(健康状态等级分“健康”、“良好”、“注意”、“恶化”和“疾病”)。
参数异常严酷度因素分析
步骤S8：建立严酷度等级评分标准；
严酷度等级的评分采用10分制，I～IV级对应1～10分，每级对应2～3个分数点，为便于分析，可将对应分数压缩到0.1～1.0之间，如表2所列。
表2严酷度等级的评分标准
严酷度等级 评分标准 压缩评分标准
IV(轻度) 1，2，3 0.1，0.2，0.3
III(中等) 4，5，6 0.4，0.5，0.6
II(较强) 7，8 0.7，0.8
I(强) 9，10 0.9，1.0
步骤S9：建立参数异常严酷度隶属度分布函数；
参数异常严酷度和参数异常概率对健康状态的影响特性相同，所以同样选取三角分布作为参数异常严酷度的健康状态隶属度分布函数，类似有
Figure PCTCN2021080526-appb-000030
Figure PCTCN2021080526-appb-000031
Figure PCTCN2021080526-appb-000032
Figure PCTCN2021080526-appb-000033
Figure PCTCN2021080526-appb-000034
步骤S10：计算动态监测参数异常严酷度对健康状态隶属度向量；
针对A、B、C、D、E状态的参数状态严酷度级别(见表1)，按照表2的严酷度等级评分标准，选取各严酷度等级的最大评分值代入隶属度分布函数，可得单因素影响下的健康状态隶属度向量分别为v A2、v B2、v C2、v D2、v E2；
步骤S11：确定严酷度影响下常压储油罐健康状态等级；
按最大隶属度原则，根据S10计算结果，可得到A、B、C、D、E状态参数异常影响下的健康状态等级分别值(健康状态等级分“健康”、“良好”、“注意”、“恶化”和“疾病”五种)。
常压储油罐动态健康状态综合评定
步骤S12：计算综合影响下严酷度对健康状态隶属度向量；
将动态监测参数异常概率影响下各参数的健康状态隶属度向量v A1、v B1、v C1、v D1、v E1和参数异常严酷度影响下各参数的健康状态隶属度向量v A2、v B2、v C2、v D2、v E2与第j种健康状态等级向量v 0j分别进行灰色关联；其中，j为健康状态等级分健康、良好、注意、恶化和疾病，记作1，…，5；即向量v 0j表示为：v 01＝(1,0,0,0,0)、v 02＝(0,1,0,0,0)、v 03＝(0,0,1,0,0)、v 04＝(0,0,0,1,0)、v 05＝(0,0,0,0,1)；
依据式
Figure PCTCN2021080526-appb-000035
式中m为1，…，5；
k为参数A、B、C、D、E；
因素i为1，2；
j为1，…，5；
Figure PCTCN2021080526-appb-000036
为二级最小差，
Figure PCTCN2021080526-appb-000037
为二级最大差，|v 0j(m)-v ki(m)|为绝对差值；
求得ξ kij(m)
再利用式
Figure PCTCN2021080526-appb-000038
式中m为1，…，5；
k为参数A、B、C、D、E；
因素i为1，2；
j为1，…，5；
求得r kij
再利用式
Figure PCTCN2021080526-appb-000039
计算得到r’ ki
能计算得到权重向量R k＝(r’ k1,r’ k2),即：R A＝(r’ A1,r’ A2),R B＝(r’ B1,r’ B2),R C＝(r’ C1,r’ C2),R D＝(r’ D1,r’ D2),R E＝(r’ E1,r’ E2),
由v A1与v A2、v B1与v B2、v C1与v C2、v D1与v D2、v E1与v E2向量分别组成矩阵V A、V B、V C、V D和 V E，
Figure PCTCN2021080526-appb-000040
并代入
H k＝R k·V k
式中k为参数A、B、C、D、E；
可行到储油罐的A、B、C、D、E五种参数在参数异常发生概率和参数异常严酷度综合影响下的健康状态隶属度向量分别为H A、H B、H C、H D、H E。
为了让本领域技术人员更加清楚地理解各参数A、B、C、D、E对应的权重向量的求解方式，下面对权重向量R A作进一步的说明：
如因素1影响下A参数的健康状态隶属度向量v A1与因素2影响下A参数的健康状态隶属度向量v A2作为比较系列，v 0j为参考数列，进行关联系数、关联度和权重的求解，得到权重向量R A，具体为：
步骤S1、求解关联系数：
v 0j中取j＝1,有v 01＝(v 01(1)、v 01(2)、v 01(3)、v 01(4)、v 01(5))
依据式
Figure PCTCN2021080526-appb-000041
式中m为1，…，5；
k为参数A(求其他参数权重向量时k则为对应的参数)
因素i为1，2；
j为1(求其他j种健康状态等级向量v 0j则为对应的参数为2、3、4、5)；
Figure PCTCN2021080526-appb-000042
为二级最小差，
Figure PCTCN2021080526-appb-000043
为二级最大差，|v 0j(m)-v ki(m)|为绝对差值；
求得ξ A11(m)和ξ A21(m)；
再利用式
Figure PCTCN2021080526-appb-000044
式中m为1，…，5；
k为参数A(求其他参数权重向量时k则为对应的参数)；
因素i为1，2；
j为1(求其他j种健康状态等级向量v 0j则为对应的参数为2、3、4、5)；
求得r A11，r A21；
步骤S2、此时j再取2，3，4，5，再分别以v 0j作为参考数列，根据步骤S1的计算方式，得到参数A的关联度r A1j，r A2j；
再利用式
Figure PCTCN2021080526-appb-000045
计算得到r’ A1和r’ A2
能计算得到权重向量R A＝(r’ A1,r’ A2)；
步骤S3、再按步骤S1和步骤S2的方法，k以B、C、D、E分别代入，同理可求得R B，R C,R D,R E；
由v A1与v A2、v B1与v B2、v C1与v C2、v D1与v D2、v E1与v E2向量分别组成矩阵V A、V B、V C、V D和V E，
Figure PCTCN2021080526-appb-000046
并代入
H k＝R k·V k
式中k为参数A、B、C、D、E；
可行到储油罐的A、B、C、D、E五种参数在参数异常发生概率和参数异常严酷度综合影响下的健康状态隶属度向量分别为H A、H B、H C、H D、H E。
步骤S13：确定储油罐动态监测健康状态等级；
根据最大隶属度原则，可得储油罐的A、B、C、D、E三种动态监测参数异常概率与动态监测参数异常严酷度综合影响下的动态监测健康状态等级分别值：(健康状态等级分为“健康”、“良好”、“注意”、“恶化”和“疾病”五种)。
确定储油罐基础健康状态
储油罐影响健康状态的基础数据主要有：投用、改造日期，涂层、保温和衬里的安装质量，常压储油罐历次检验和检测情况数据，各层壁板和底板的建造材料、名义厚度，此四项基础数据依次编为U1，U2，U3，U4。
步骤S14：确定储油罐状态集和状态评价集；
根据储油罐基础数据，其状态集为：
U＝(U1，U2，U3，U4)
设定储油罐的健康状态分为“健康”、“良好”、“注意”、“恶化”和“疾病”5个等级，则状态评价集为
V＝(I，II，III，IV，V)
步骤S15：确定基础数据的重要度；
通过对石化储油罐专业资料的分析，关于4项基础参数的重要程度分析结果，最终确定四项基础参数的权重如表3：
表3石化储油罐基础参数及权重
Figure PCTCN2021080526-appb-000047
Figure PCTCN2021080526-appb-000048
步骤S16：确定各基础数据的劣化度；
针对表3中不同基础参数采用不同的劣化度计算方法，具体为：
1.针对“投用、改造日期(U1)”，根据装备实际使用时间计算劣化度。
因投用、改造日期难以监测及检测，其变化与时间之间具有近似的线性关系，根据设计寿命等设计资料及大量统计数据确定其平均故障寿命，则其劣化度计算公式为：
l i＝(t/T) k
式中：t为储油罐的使用时间；T为该储油罐的平均故障寿命；k为故障指数，通常可取1或2。
2.针对“涂层、保温和衬里的安装质量(U2)”，“常压储油罐历次检验和检测情况数据(U3)”“各层壁板和底板的建造材料、名义厚度(U4)”，因这些参数的劣化度与本身质量和使用时间均有关系，故采用“打分估计”和“实际使用时间”综合计算的方法。
1)根据设计人员、质检人员、行内专家计算劣化度。
其劣化度估算公式为：
l i′＝(X·P 1+Y·P 2+Z·P 3)/(P 1+P 2+P 3)i＝2,3,4
式中：X，Y，Z分别为系数，其值介于0～1之间，0代表健康，1代表完全劣化；P 1、P 2、P 3分别为设计人员、质检人员、行内专家的权重，其值反映打分人员的水平和权威性；2)综合根据设计寿命等资料及大量统计数据确定的平均故障寿命，计算最终劣化度。
则其最终劣化度计算公式为：
Figure PCTCN2021080526-appb-000049
式中：t为储油罐的使用时间；T为该储油罐的平均故障寿命；k为故障指数，k取 1或2；
计算基础参数U2，U3，U4的劣化度。
步骤S17建立基础参数劣化度判断矩阵；
根据劣化度求其健康状态等级的隶属度，由于岭形分布具有主值区间宽、过渡带平缓的特点，能较好的反映储油罐劣化度的状态空间的模糊关系，因此，采用岭形分布隶属度函数：
Figure PCTCN2021080526-appb-000050
Figure PCTCN2021080526-appb-000051
Figure PCTCN2021080526-appb-000052
Figure PCTCN2021080526-appb-000053
Figure PCTCN2021080526-appb-000054
由此可得到以劣化度为评价标准的模糊评判矩阵为：
R i＝(r I(l i)，r II(l i)，r III(l i)，r IV(l i)，r V(l i))
Figure PCTCN2021080526-appb-000055
步骤S18进行储油罐基础参数的模糊综合评判；
E＝W·R
其中W为据表3中的
W＝(W 1，W 2，W 3，W 4)
从模糊综合评判结果可以得到该装备系统属于“健康”、“良好”、“注意”、“恶化”和“疾病”的程序值，依据隶属最大原则可以判断储油罐基础参数所处状态。
综合储油罐动态监测健康状态和基础健康状态进行最终状态的确认
步骤S19：综合储油罐动态监测健康状态和基础健康状态进行最终状态的确认
根据储油罐动态监测参数健康状态(步骤S13结果)储油罐基础健康状态(步骤S18结果)，取储油罐动态监测参数健康状态和储油罐基础健康状态中的较严重级别为最终健康状态评估值。
以上所述仅为本发明的较佳实施例，凡依本发明申请专利范围所做的均等变化与修饰，皆应属本发明的涵盖范围。

Claims (10)

  1. 一种基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述评估方法包括如下步骤：步骤1、确定储油罐健康状态影响因素，对影响因素的参数进行采集并得到每种参数发生异常的概率；
    步骤2、建立健康状态下参数发生异常的概率隶属度分布函数，获取概率影响下健康状态等级隶属度矩阵；
    步骤3、建立健康状态等级隶属度分布函数，获取参数异常严酷度影响下健康状态等级隶属度矩阵；
    步骤4、获取综合影响下参数异常严酷度对健康状态隶属度向量；
    步骤5、确定储油罐动态监测参数健康状态；
    步骤6、建立储油罐状态集和状态评价集，获取储油罐各基础参数重要度权重系数；
    步骤7、确定储油罐各基础参数劣化度；
    步骤8、建立基础参数劣化度判断矩阵，进行储油罐基础参数模糊综合评估；
    步骤9、按最大隶属度原则确定储油罐基础健康状态；
    步骤10、取所述储油罐动态监测参数健康状态和所述储油罐基础健康状态中的严重级别，确定最终储油罐的健康状态。
  2. 根据权利要求1所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤1进一步具体包括如下步骤：步骤11、通过储油罐健康状态影响分析，选取在线监测的参数包括但不限于该五项参数：罐内温度记为参数A、罐内压力记为参数B、罐内液位记为参数C、管道的振动数据记为参数D、防雷接地电阻记为参数E；对监测的参数进行采集经过网络传输至数据处理服务器；
    步骤12、对每种参数与对应设置好的正常范围值进行比对，若超出正常范围则记为异常，统计异常次数，用于测试数据分析；
    步骤13、通过测试数据分析得到参数发生异常的概率，概率越小，则储油罐健康状态越好。
  3. 根据权利要求2所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征 在于：所述步骤2进一步具体包括如下步骤：步骤21、根据每种参数发生异常概率p分布的特性，在一设定的置信区间内，监测的参数异常发生的概率值越小，健康状态越趋于优，则选择三角分布作为健康状态下参数发生异常的概率隶属度分布函数，有：
    Figure PCTCN2021080526-appb-100001
    Figure PCTCN2021080526-appb-100002
    Figure PCTCN2021080526-appb-100003
    Figure PCTCN2021080526-appb-100004
    Figure PCTCN2021080526-appb-100005
    步骤21、将监测的参数A、参数B、参数C、参数D、参数E对应的发生异常的概率值代入概率隶属度分布函数，可得单因素影响下的健康状态隶属度向量分别为v A1、 v B1、v C1、v D1、v E1。
  4. 根据权利要求3所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤3进一步具体包括如下步骤：步骤31、设置参数异常的严酷度级别q，其中，参数异常严酷度和参数异常发生的概率对健康状态的影响特性相同，则同样选取三角分布作为参数异常严酷度的健康状态等级隶属度分布函数，有
    Figure PCTCN2021080526-appb-100006
    Figure PCTCN2021080526-appb-100007
    Figure PCTCN2021080526-appb-100008
    Figure PCTCN2021080526-appb-100009
    Figure PCTCN2021080526-appb-100010
    步骤32、选取各严酷度级别的最大评分值代入健康状态等级隶属度分布函数，可得单因素参数异常严酷度影响下健康状态隶属度向量分别为v A2、v B2、v C2、v D2、v E2。
  5. 根据权利要求4所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤4进一步具体为：
    将动态监测参数异常概率影响下各参数的健康状态隶属度向量v A1、v B1、v C1、v D1、v E1和参数异常严酷度影响下各参数的健康状态隶属度向量v A2、v B2、v C2、v D2、v E2与第j种健康状态等级向量v 0j分别进行灰色关联；其中，j为健康状态等级分健康、良好、注意、恶化和疾病，记作1，…，5；即向量v 0j表示为：v 01＝(1,0,0,0,0)、v 02＝(0,1,0,0,0)、v 03＝(0,0,1,0,0)、v 04＝(0,0,0,1,0)、v 05＝(0,0,0,0,1)；
    依据式
    Figure PCTCN2021080526-appb-100011
    式中m为1，…，5；
    k为参数A、B、C、D、E；
    因素i为1，2；
    j为1，…，5；
    Figure PCTCN2021080526-appb-100012
    为二级最小差，
    Figure PCTCN2021080526-appb-100013
    为二级最大差，|v 0j(m)-v ki(m)|为绝对差值；
    求得ξ kij(m)
    再利用式
    Figure PCTCN2021080526-appb-100014
    式中m为1，…，5；
    k为参数A、B、C、D、E；
    因素i为1，2；
    j为1，…，5；
    求得r kij
    再利用式
    Figure PCTCN2021080526-appb-100015
    计算得到r’ ki
    能计算得到权重向量R k＝(r’ k1,r’ k2),即：R A＝(r’ A1,r’ A2),R B＝(r’ B1,r’ B2),R C＝(r’ C1,r’ C2),R D＝(r’ D1,r’ D2),R E＝(r’ E1,r’ E2),
    由v A1与v A2、v B1与v B2、v C1与v C2、v D1与v D2、v E1与v E2向量分别组成矩阵V A、V B、V C、V D和V E，
    Figure PCTCN2021080526-appb-100016
    并代入
    H k＝R k·V k
    式中k为参数A、B、C、D、E；
    可行到储油罐的A、B、C、D、E五种参数在参数异常发生概率和参数异常严酷度综合影响下的健康状态隶属度向量分别为H A、H B、H C、H D、H E。
  6. 根据权利要求5所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤5进一步具体为：设置动态监测参数异常概率与动态监测参数异常严酷度综合影响下的储油罐动态监测参数健康状态等级为：健康、良好、注意、恶化、疾病；则 根据最大隶属度原则，通过健康状态隶属度向量H A、H B、H C、H D、H E能得储油罐的A、B、C、D、E五种参数对应的储油罐动态监测参数健康状态等级。
  7. 根据权利要求6所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤6进一步具体为：所述储油罐各基础参数包括投用、改造日期，涂层、保温和衬里的安装质量，常压储油罐历次检验和检测情况数据，各层壁板和底板的建造材料、名义厚度，将此四项基础数据依次编为U1，U2，U3，U4；根据储油罐各基础数据，则储油罐状态集为：U＝(U1，U2，U3，U4)；根据储油罐动态监测参数健康状态等级：健康、良好、注意、恶化、疾病；则设定储油罐的健康状态等级分别对应为I，II，III，IV，V，则储油罐状态评价集为G＝(I，II，III，IV，V)；根据储油罐状态集和状态评价集，确定四项基础参数的权重系数分别为：权重W 1、权重W 2、权重W 3、权重W 4。
  8. 根据权利要求7所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤7进一步具体为：针对投用、改造日期的基础参数U1，根据储油罐实际使用时间计算劣化度；即劣化度计算公式为：
    l i＝(t/T) k
    式中：i＝1,t为储油罐的使用时间；T为该储油罐的平均故障寿命；k为故障指数，k取1或2；
    针对涂层、保温和衬里的安装质量U2，常压储油罐历次检验和检测情况数据U3，各层壁板和底板的建造材料、名义厚度U4，这些基础参数先经过劣化度估算公式：
    l iˊ＝(X·P 1+Y·P 2+Z·P 3)/(P 1+P 2+P 3)，i＝2,3,4
    式中：X，Y，Z为系数其值介于0～1之间，0代表健康，1代表完全劣化；P 1、P 2、P 3分别为设计人员、质检人员、行内专家的权重；
    求解，再结合储油罐的平均故障寿命计算，利用公式：
    Figure PCTCN2021080526-appb-100017
    式中：t为储油罐的使用时间；T为该储油罐的平均故障寿命；k为故障指数，k取1或2；
    计算基础参数U2，U3，U4的劣化度。
  9. 根据权利要求8所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤8进一步具体为：
    根据各基础参数劣化度求其健康状态等级的隶属度，采用岭形分布隶属度函数：
    Figure PCTCN2021080526-appb-100018
    Figure PCTCN2021080526-appb-100019
    Figure PCTCN2021080526-appb-100020
    Figure PCTCN2021080526-appb-100021
    Figure PCTCN2021080526-appb-100022
    由此可得到以劣化度为评价标准的模糊评判矩阵为：
    R i＝(r I(l i)，r II(l i)，r III(l i)，r IV(l i)，r V(l i))
    Figure PCTCN2021080526-appb-100023
    则储油罐基础参数的模糊综合评估：
    E＝W·R
    其中W为四项基础参数的权重系数W＝(W 1，W 2，W 3，W 4)。
  10. 根据权利要求9所述的基于多数据采集的石化常压储油罐健康状态评估方法，其特征在于：所述步骤9进一步具体为：从模糊综合评估结果能得到该储油罐属于健康、良好、注意、恶化、疾病的数值，再按最大隶属度原则能判断储油罐基础参数所处的是健康、良好、注意、恶化、疾病中的哪一个状态。
PCT/CN2021/080526 2020-03-16 2021-03-12 基于多数据采集的石化常压储油罐健康状态评估方法 WO2021185177A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/027,122 US20240028937A1 (en) 2020-03-16 2021-03-12 Method for evaluating health status of petrochemical atmospheric oil storage tank using data from multiple sources

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010182513.4 2020-03-16
CN202010182513.4A CN111368451B (zh) 2020-03-16 2020-03-16 基于多数据采集的石化常压储油罐健康状态评估方法

Publications (1)

Publication Number Publication Date
WO2021185177A1 true WO2021185177A1 (zh) 2021-09-23

Family

ID=71208865

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/080526 WO2021185177A1 (zh) 2020-03-16 2021-03-12 基于多数据采集的石化常压储油罐健康状态评估方法

Country Status (3)

Country Link
US (1) US20240028937A1 (zh)
CN (1) CN111368451B (zh)
WO (1) WO2021185177A1 (zh)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114091268A (zh) * 2021-11-24 2022-02-25 南京航空航天大学 一种基于节点重要度与层次分析法的无人机健康状况评估方法
CN114444569A (zh) * 2021-12-22 2022-05-06 北京航天测控技术有限公司 一种电源控制系统健康状态评估算法
CN116089787A (zh) * 2023-03-08 2023-05-09 中国人民解放军海军工程大学 基于层次分析法的船舶子系统运行状态分析方法及系统
CN116118010A (zh) * 2023-04-17 2023-05-16 武昌理工学院 一种用于非对称配钢型钢混凝土柱的能源管理系统
CN116186888A (zh) * 2022-12-28 2023-05-30 北京控制工程研究所 航天器健康状态量化评估方法、装置、电子设备及介质
CN116311594A (zh) * 2023-05-11 2023-06-23 中国人民解放军海军工程大学 一种船舶子系统状态分析方法、装置及存储介质
CN117129815A (zh) * 2023-10-27 2023-11-28 南京中鑫智电科技有限公司 一种基于物联网的多劣化绝缘子综合检测方法和系统
CN117150934A (zh) * 2023-10-30 2023-12-01 南京中鑫智电科技有限公司 变压器套管状态综合监测系统及其在线数据处理方法
WO2023246185A1 (zh) * 2022-06-20 2023-12-28 东方电气集团东方电机有限公司 一种评估方法、装置、电子设备和存储介质
CN117688514A (zh) * 2024-02-04 2024-03-12 广东格绿朗节能科技有限公司 基于多源数据的遮阳篷健康状况检测方法及系统
CN117909200A (zh) * 2024-03-19 2024-04-19 中国电子科技集团公司第十研究所 一种信息保障体系能力增量对比评估方法、设备及系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111368451B (zh) * 2020-03-16 2023-03-31 福建省特种设备检验研究院 基于多数据采集的石化常压储油罐健康状态评估方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107203947A (zh) * 2017-05-22 2017-09-26 武汉大学 一种数字化变电站继电保护系统状态评价方法
US20190005400A1 (en) * 2016-09-20 2019-01-03 Southwest Petroleum University A fuzzy evaluation and prediction method for running status of mechanical equipment with occurrence probability of failure modes
CN111353236A (zh) * 2020-03-16 2020-06-30 福建省特种设备检验研究院 基于多因素的石化常压储油罐健康状态评估系统
CN111368451A (zh) * 2020-03-16 2020-07-03 福建省特种设备检验研究院 基于多数据采集的石化常压储油罐健康状态评估方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102928507B (zh) * 2012-10-17 2014-12-10 浙江省电力公司电力科学研究院 气体绝缘全封闭组合电器罐体健康监测装置及其监测方法
US10877465B2 (en) * 2016-10-24 2020-12-29 Fisher-Rosemount Systems, Inc. Process device condition and performance monitoring
CN107941537B (zh) * 2017-10-25 2019-08-27 南京航空航天大学 一种机械设备健康状态评估方法
CN108035838B (zh) * 2017-12-07 2020-11-03 武汉四创自动控制技术有限责任公司 全厂水轮机调速系统健康状态评估及优化方法
CN108874733A (zh) * 2018-04-25 2018-11-23 明阳智慧能源集团股份公司 一种大型半直驱机组健康状态评估方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190005400A1 (en) * 2016-09-20 2019-01-03 Southwest Petroleum University A fuzzy evaluation and prediction method for running status of mechanical equipment with occurrence probability of failure modes
CN107203947A (zh) * 2017-05-22 2017-09-26 武汉大学 一种数字化变电站继电保护系统状态评价方法
CN111353236A (zh) * 2020-03-16 2020-06-30 福建省特种设备检验研究院 基于多因素的石化常压储油罐健康状态评估系统
CN111368451A (zh) * 2020-03-16 2020-07-03 福建省特种设备检验研究院 基于多数据采集的石化常压储油罐健康状态评估方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YAN DONGDONG: "Research on Safety Evaluation of Large Crude Oil Tanks Based on Fuzzy Comprehensive Evaluation Method", MANAGEMENT & TECHNOLOGY OF SME, 25 September 2015 (2015-09-25), pages 285 - 288, XP055851411 *

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114091268A (zh) * 2021-11-24 2022-02-25 南京航空航天大学 一种基于节点重要度与层次分析法的无人机健康状况评估方法
CN114444569A (zh) * 2021-12-22 2022-05-06 北京航天测控技术有限公司 一种电源控制系统健康状态评估算法
CN114444569B (zh) * 2021-12-22 2024-05-10 北京航天测控技术有限公司 一种电源控制系统健康状态评估算法
WO2023246185A1 (zh) * 2022-06-20 2023-12-28 东方电气集团东方电机有限公司 一种评估方法、装置、电子设备和存储介质
CN116186888B (zh) * 2022-12-28 2024-01-23 北京控制工程研究所 航天器健康状态量化评估方法、装置、电子设备及介质
CN116186888A (zh) * 2022-12-28 2023-05-30 北京控制工程研究所 航天器健康状态量化评估方法、装置、电子设备及介质
CN116089787A (zh) * 2023-03-08 2023-05-09 中国人民解放军海军工程大学 基于层次分析法的船舶子系统运行状态分析方法及系统
CN116118010A (zh) * 2023-04-17 2023-05-16 武昌理工学院 一种用于非对称配钢型钢混凝土柱的能源管理系统
CN116118010B (zh) * 2023-04-17 2023-06-30 武昌理工学院 一种用于非对称配钢型钢混凝土柱的能源管理系统
CN116311594A (zh) * 2023-05-11 2023-06-23 中国人民解放军海军工程大学 一种船舶子系统状态分析方法、装置及存储介质
CN116311594B (zh) * 2023-05-11 2023-09-19 中国人民解放军海军工程大学 一种船舶子系统状态分析方法、装置及存储介质
CN117129815A (zh) * 2023-10-27 2023-11-28 南京中鑫智电科技有限公司 一种基于物联网的多劣化绝缘子综合检测方法和系统
CN117129815B (zh) * 2023-10-27 2024-02-02 南京中鑫智电科技有限公司 一种基于物联网的多劣化绝缘子综合检测方法和系统
CN117150934B (zh) * 2023-10-30 2024-01-26 南京中鑫智电科技有限公司 变压器套管状态综合监测系统及其在线数据处理方法
CN117150934A (zh) * 2023-10-30 2023-12-01 南京中鑫智电科技有限公司 变压器套管状态综合监测系统及其在线数据处理方法
CN117688514A (zh) * 2024-02-04 2024-03-12 广东格绿朗节能科技有限公司 基于多源数据的遮阳篷健康状况检测方法及系统
CN117688514B (zh) * 2024-02-04 2024-04-30 广东格绿朗节能科技有限公司 基于多源数据的遮阳篷健康状况检测方法及系统
CN117909200A (zh) * 2024-03-19 2024-04-19 中国电子科技集团公司第十研究所 一种信息保障体系能力增量对比评估方法、设备及系统
CN117909200B (zh) * 2024-03-19 2024-06-11 中国电子科技集团公司第十研究所 一种信息保障体系能力增量对比评估方法、设备及系统

Also Published As

Publication number Publication date
CN111368451B (zh) 2023-03-31
US20240028937A1 (en) 2024-01-25
CN111368451A (zh) 2020-07-03

Similar Documents

Publication Publication Date Title
WO2021185177A1 (zh) 基于多数据采集的石化常压储油罐健康状态评估方法
CN111353236B (zh) 基于多因素的石化常压储油罐健康状态评估系统
WO2022252505A1 (zh) 一种基于多指标集群分析的设备状态监测方法
CN110703214B (zh) 一种气象雷达状态评估和故障监测方法
CN105912857B (zh) 一种配电设备状态监测传感器的选配方法
CN109858140B (zh) 一种基于信息熵离散型贝叶斯网络冷水机组故障诊断方法
KR101776350B1 (ko) 압축기를 진단하는 방법 및 시스템
CN113506001B (zh) 一种作业现场安全风险精益化智慧管控辅助决策方法
CN116777223A (zh) 一种城市地下管网安全综合风险评估方法及系统
CN113343177A (zh) 基于模糊综合评价理论的电梯设备健康状态诊断方法
CN115018384A (zh) 一种建筑工地安全风险评估方法及系统
CN114862267A (zh) 一种输油气管道报警管理体系的评价方法及系统
CN114021915A (zh) 基于改进均衡权重和可变模糊集的电气火灾风险评估方法
CN111626646B (zh) 一种装备完好性检查信息融合方法
CN116167659B (zh) 一种碳市场碳排放数据质量评价方法
CN116384732A (zh) 场站管道风险智能评估方法、系统、存储介质及计算设备
CN109784777B (zh) 基于时序信息片段云相似度度量的电网设备状态评估方法
CN116628976A (zh) 一种水轮机组设备状态变权综合评价方法
CN115796832A (zh) 基于多维参量的变电设备健康状态综合评估方法
CN116596302A (zh) 基于动态分析的埋地钢制燃气管道检验周期确定方法、电子设备和存储介质
CN110533213A (zh) 基于支持向量机的输电线路缺陷风险建模及其预测方法
CN116739399A (zh) 一种高压电缆运行状态评价方法
CN115345414A (zh) 一种输油气管道工控网络信息安全性评价方法及系统
CN113239436B (zh) 一种钢桥状态等级评估与预测方法
CN111861271B (zh) 一种管道保温性能的评价方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21771173

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21771173

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 10.07.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21771173

Country of ref document: EP

Kind code of ref document: A1