Research On Corrosion Trend Prediction And Reliability Modeling Of Storage Oil Tank

At present a lots of storage tanks are being served on all oilfields in our country, and most are in the mid-late service. Due to the backwardness and extensiveness of tank management, many storage tanks are running in spite of illness. Currently the management is not good to grasp of health of storage oil tanks, which leads to many defects not being discovered in time, so the latent defect not being predicted, assessed and treated commendably in the normal course of operation, which is seriously threatening the safety of petrochemical plant. In order to set up the necessary maintenances to repair storage tanks before disaster, the corrosion rule needs to be understood, and the reliability of the storage tank needs to be correctly assessed and forecasted. Therefore the corrosion trend prediction and reliability modeling of storage oil tanks were studied in this paper.Firstly, the main failure mode of oil storage tank was analyzed. Then the models of statistical analysis were set up for the corrosion of oil tank, and the large amounts of datum about corrosion of oil tank were analyzed. Normal distribution was adopted for uniform corrosion depth and extremum statistic analysis for localized corrosion depth, in which generalized extreme value distribution was used as statistical model to analyze the statistical regularity of corrosion depth, and was applied to the case of storage oil tank. And the model got through the goodness-of-fit test. The model was compared with the traditional model, and the results show that the model could better fit the statistical regularity of local corrosion. At the same time, the shortcomings of the parameter calculation in the GEV distribution are improved.Aimed at the complexity of the storage tank corrosion and the randomness of tested data, probability distribution of GM(1,1) prediction model was established by using the multi data sample of one inspection, which had been set up to predict the corrosion depth and rate of storage oil tanks. And the instance analysis and verification had been done. In the process of model building, the logical operation of sample probability has been used to quantify the errors in the sample, and the result is more objective and accurate. The result can provide reference for the maintenance plan of storage oil tank.The finite element model of storage tank containing corrosion pits was established firstly. The strength of storage tank was analyzed containing single pit and two pits with different sizes and depths. And the stress concentration coefficient of pit can be expressed as nonlinear equation of pit depth and diameter by statistical analysis of the results. And it has been verified that fitting value of stress concentration coefficient of different sizes of the pits is close to the finite element value. At the same time the pit was treated as a equivalent crack, and the carrying capacity of pit was calculated. And the results of two calculation methods were analyzed. Next, stress interference of pits were analyzed at circumferential direction and at axial direction respectively.Various factors were analyzed which impacted the strength, stability and perforation of tank, and the degradation model of structural resistance of each component was established. A model on corrosion depth with the service time was established according to the past test data. The reliability models were established on the strength, stability and local corrosion depth of the tank on the basis of reliability theory, and the reliability index were calculated by using the improved particle swarm algorithm, then the reliability were calculated. And the reliability of all tank components was analyzed with the growth of service time. The system reliability model was established on the basis of comprehensive consideration of the main failure modes, and system reliability of storage tank was analysed along with the growth of the service time considering the interaction between the main failure modes.The results can provide technical support to the safe operation of tank.The repair cycle of storage tanks was studied from different angles. To guarantee the safe operation of storage oil tank, the repair cycle was gotten according to given reliability. The repair cycle was optmized based on the principle of minimum cost per unit time considering the operation ecomomy of the storage oil tank. Considering the security and economy of storage oil tank, the maintenance cycle was calculated based on the principle of minimum cost per unit time under given reliability. Research results provide theoretical basis for the condition-based maintenance of storage oil tanks.Finally the practical application of related technologies was discussed based on the enterprise scientific and technological project from Shengli Oil Field Shengli Engineering Design Consulting Co. Ltd. Based on Web a integrity management system for the oilfield union station equipments has been developed for the actual management needs of oilfield union station. It was introduced in detail that the composition of the system architecture, the design of the related information models and the main function modules. And the running instance of the system was given. Then the application effect was analyzed and summarized.