Research And Application Of On-line Monitoring Data Optimization And Life Prediction Technoligy For Corrosion Of Heavy Oil Catalytic Cracking Unit

As one of the most important secondary processing units in the refinery,the heavy oil catalytic cracking unit has huge production capacity,but also faces severe production safety challenges.With the continuous improvement of petroleum refining capacity,the equipment corrosion problem faced by petrochemical enterprises in China is becoming more and more serious.In order to eliminate the potential safety hazards caused by equipment corrosion,petrochemical enterprises have actively set up a corrosion monitoring system,and plan to improve it with advanced monitoring technology.Affected by many factors such as time,technology and capital,the existing corrosion monitoring system is based on manual fixed-point thickness measurement technology,and the corrosion of equipment is analyzed by accumulating the wall thickness data of each measuring point for a long time.Because there are many defects in manual fixed-point thickness measurement technology,it can not provide timely and reliable corrosion early warning information.Therefore,based on the existing manual fixed-point thickness measurement technology of heavy oil catalytic cracking unit and combined with the corrosion loop of the unit,this paper improves the existing corrosion monitoring system of the unit by using online monitoring technology.At the same time,based on the corrosion monitoring data,the pipeline life prediction model is established,which accurately predicts the corrosion trend of the pipeline and improves the early warning ability of the corrosion monitoring system.The main research contents and conclusions are as follows.(1)According to the process flow and corrosion mechanism of a heavy oil catalytic cracking unit in a petrochemical enterprise,the corrosion status of the unit was analyzed.Then,the corrosion circuit diagram is drawn with the corrosion mechanism as the main reference and combined with the four aspects of medium,temperature,pressure and material.The key corrosion parts in the device are summarized,and the typical high-corrosion risk parts are selected for on-line monitoring according to the principle of point selection,and the corrosion protection measures are formulated in combination with the corrosion mechanism,equipment pipeline materials,corrosion monitoring indicators and other data of the measuring points.(2)Ultrasonic on-line corrosion monitoring system is used to monitor the residual wall thickness and corrosion rate of monitoring points in real time.In order to solve the problem of deviation of monitoring value caused by pipeline temperature change due to process adjustment,environmental temperature change and device shutdown,an ultrasonic probe temperature compensation experiment was designed.Comparing the corrosion trend of monitoring points before and after temperature compensation shows that the measurement error of pipeline wall thickness can be eliminated by modifying the set wave velocity of corrosion monitoring system.(3)Based on the wall thickness data accumulated by monitoring points,a grey prediction model(GM(1,1))optimized by sparrow search algorithm is established to predict the life of corroded pipelines.This model reduces the prediction error by optimizing the background value of GM(1,1)with the Sparrow Search Algorithm(SSA),and improves the optimization strategy of SSA to obtain the ISSA with faster iteration speed,and establishes the corrosion prediction model of ISSA-GM(1,1).Compared with GM(1,1),DGM(1,1)and SSA-GM(1,1)models,it is found that the fitting error and prediction error of ISSA-GM(1,1)model are the smallest.The results show that the ISSA-GM(1,1)corrosion prediction model proposed in this paper can accurately predict the corrosion trend of pipelines.