Risk level Assessment of the Desalter and Preflash Column of a Nigerian Crude Distillation Unit

A study of operability and assessment of the safety level of a crude distillation unit in a Nigerian refinery has been carried out. In a crude distillation unit of a refinery, the desalter and preflash column, as well as their associated pipe works and pumps, are considered to have high-risk factors that can present operability problems and possibly hazardous conditions. Applying the HAZOP methodology, a total of 25 guide words were suitably applied on 4 nodes to study the possible deviations that may occur. The 89 causes of these deviations identified in the study could be classified under human errors and equipment malfunctioning. From the nature of the 46 consequences identified, the plant is likely to experience more of operability problems and less of deadly hazardous scenes. Suitable recommendations that will improve the operability and safety levels of the plant were however presented in the study and presented in tables. If the 61 recommendations given in this study can be incorporated into the design with the few safeguards already present in the original design of the plant, the operability problems of the plant will be greatly minimized and safety level will improve drastically.


Introduction
Building and operating a chemical plant without accident-preventive measures can cause an unquantifiable magnitude of hazards. Over the years, a good number of hazards have been recorded in the chemical industry around the world [1] and most of the recorded accidents were largely due to human error. The 1974 Flixborough disaster that was initiated by a reactor leak killed 28 persons, injured other 36 and destroyed properties [2]. The 1976 Seveso disaster led to the pollution of a vast area of land and water bodies. Thousands of people were killed and properties worth billions of dollars destroyed in the 1984 Bhopal disaster which was a direct consequence of human error [3]. Facilities worth billions of dollars were also destroyed in the 2005 Buncefield disaster while the 2010 BPL refinery disaster caused much pollution to water bodies [4]. The accidents that have occurred in the process industry have prompted owners and operators of modern day's chemical plants to incorporate safety measures to prevent accidents. The major industrial accidents recorded between 1956 and 1998 showed a decline in accidents in the chemical and process industries. This decline is traceable to the recent attention given to the study of accident forecasting and loss prevention in the chemical and process industries [5] [6]. Preventive-mechanism entails recognizing possible hazardous scenes within the plant which may vary in size and which may be noticeable or not noticeable [7]. Accident in the chemical and process industries can arise from the process itself, properties of the chemicals and their handling such as fire, explosion and exposure to toxic substances. For example, over temperature can lead to over pressure which can cause fire, explosion or toxic release that can cause accident. The predictions and prevention of accidents can be done by recognizing the hazards and the corresponding actions to be taken [8]. The use of certain process schemes or materials in production in order to maximize profit may result in operability problems and increase accident risks in the plant. For example, steam stripping is very effective in the vapourisation of more products in refineries [9] but due to cost of steam, some refiners have employed the use of water steam which is less expensive [10] [11] but with its associated hazards and operational problems.
Over the years, experts in process plant safety have developed risk-assessment procedures to enhance safety levels of process plants. Some of the many risk-assessment procedures developed so far include the Preliminary Hazard Assessment (PHA), the Fault Tree Analysis (FTA), the Energy Tree and Barrier Analysis (ETBA), the Failure Mode and Effects Analysis (FMEA) as well as the Hazard and Operability (HAZOP) [12]. Each of these mechanisms has strengths and weaknesses and is specialized in handling a particular type of risk. Although HAZOP is time-consuming as it requires a considerable amount of time of preparation, it gives a proper, organized and critical examination of the process of new or existing facilities to evaluate the potential for equipment malfunctioning in terms of the resultant impacts [13]. HAZOP performs a structural investigation of each unit in a process to depict what kind of deviation from the ideal operation can occur and what harm may be caused by such deviation. It is adopted in HAZOP study that a system is safe when key operability parameters such as temperature, pressure, flow or levels are in their normal conditions. Operability problems if not identified in HAZOP can result in production losses H. O. Orugba et al. Journal of Materials Science and Chemical Engineering due to inferior product quality or process inefficiency. This means properly conducted HAZOP can help not just in plant and personnel safety but also prevents loss of continuity or loss of the product specification [14]. Initial HAZOP study helps identify suitable protection on measures that may be implemented to avoid impending accidents [5] [15]. HAZOP involves a study on how a plant might deviate from the intents while taking notes of the resulting appropriate solutions to these deviations. Since it is a group study, it creates a brainstorming environment that brings creativity and generates ideas. In HAZOP, a flow sheet on piping and instrumentation diagram (PID) for the plant is obtained. Each node on the PID is numbered where a series of disturbances are proposed. For each disturbance, potential causes and consequences are described and noted. Guide words are used to ensure that the design is explored in every conceivable way. It is paramount that a HAZOP team focuses only on consequences with serious effects since numerous consequences can be obtained. In this paper, the concept of HAZOP was applied to study the safety level of a crude distillation unit of a petroleum refinery. Crude from storage tank is pumped at 30˚C and 30 kg/sqcm through a valve into the first preheat train where its temperature is raised to about 125˚C and pressure reduced to 11 kg/sqcm. Water is injected into the crude both at upstream and downstream of the first preheating train to dissolve salts contained in the crude. Water injection upstream of preheating is manually controlled at  Liquid fraction mainly heavy naphtha is also withdrawn as side cut and sent into the atmospheric column and the bottom sent to the atmospheric heater to raise its temperature to 350˚C before it enters the atmospheric column flash zone. words were applied to each study node. In each node, a process parameter was identified and an intention was created for the node. For example, if the process parameter being considered was temperature, the first guide word like "low" was applied and a meaningful deviation like "low temperature" was developed. All the possible causes of low temperature as well as the likely consequences were determined. The study also identified existing operational safeguards but when a consequence is likely to pose a hazardous situation, recommendations were given for possible changes to be made to the system to eliminate or minimize hazard. The same process was carried out repeatedly for all the guide words on the same node. The next node was selected and the same activity was repeated on it.

Guide Words
The guide words used in the HAZOP study were as follows:

HAZOP Study of the Desalter
The P & ID of the desalter in Figure 1 was used to perform the HAZOP study and the details are presented in Table 1 and Table 2 as follows.

HAZOP Study of the Preflash Column
The P & ID of the preflash column of the crude distillation unit in Figure 2 was used to perform the HAZOP study and the details are presented in Table 3 and Table 4 as follows:

Results and Discussions
The main equipment of the crude distillation unit of the refinery considered in this work was the desalter, the preflash furnace, the preflash column and their associated pipe works and equipment like pumps. Using the method of HAZOP to evaluate the operability and safety level of the unit, 4 study nodes were identified. In the study of possible deviations that can occur in the nodes, 25 guide words were suitably applied on the nodes and 89 causes were identified. Most of the causes were due to equipment malfunctioning and a few may be classified as