Process Technology: An Introduction - Haan A.B. 2015

18 Process safety
18.2 Development, design, and construction of safe plants

18.2.1 Introduction

It is obvious that most of the dangers in the operation of chemical plants have to do with the reactivities of the substances present and the way they are treated in the plant. A conflict arises because chemical processes are impossible without substances with hazardous properties and effects. Although chemistry requires reactive substances, it is this same reactivity that represents the danger they pose. The substances must therefore be reliably contained in the process equipment, and their reactivity must be governed so that uncontrolled chemical reactions cannot take place. The principal hazards of substances are caused by releases and spills outside the plant, and by uncontrolled chemical reactions between them. The larger the quantities of substances and energy released, the greater the danger. Therefore plant and process safety efforts should have the objective of minimizing the quantities of substances contained and finally controlling the potential risks that remain.

18.2.2 Safety assessment

Chemical processes and their associated technical facilities are developed in steps. After process development in the laboratory, testing on a pilot scale is commonly practiced before the project goes through the various planning stages. Each phase involves questions about the safety of the process and the plant which must all be answered before going to the next phase. One of the most important steps in safety assessment is the systematic identification of possible disturbing influences, their initiating events, and their effects during each stage of process development. In this process the main safety tasks for the process designer, engineer, and constructor are the correct identification and assessment of all hazards, followed by the appropriate steps to reduce and control them. These safety tasks concern the process itself, and the safe design and operation of the technical facility required for the process:

· (1) safe process designs by identifying all types of hazards, assessing their hazard potentials, minimizing the hazard potentials and deactivating the hazard potentials;

· (2) safe plant design and operation by systematically analyzing danger sources, evaluating their probabilities of occurrence, minimizing sources of trouble and error, and employing a fault-tolerant design

If an industrial plant is to be constructed for a process that has been safety optimized in this way, two analytical tasks followed by two design tasks remain to be performed:

· (1) the plant system must be systematically searched for danger sources that can activate the deactivated hazard potential;

· (2) when possible faults are identified, their frequencies or probabilities of occurrence must be evaluated so that appropriate safety measures can be taken;

· (3) all possibilities for minimizing sources of trouble and error must be exhausted;

· (4) as far as possible, the facility must be designed and equipped so that faults are “forgiven” without resulting in harm.

The most expedient way of creating a safe plant is thus to plan for safety studies. At each step in process and plant development, safety analyses must be done in order to pose the right questions and immediately seek solutions to the problems identified. Four phases of safety assessment can be identified:

· (1) create safety principles by compiling and determining safety, toxicological, and ecological data, identifying sources of danger in the process, examining possible safety solutions, establishing the safety concept for the process;

· (2) define the safety concept for the plant by performing systematic analysis and identifying technical protective measures;

· (3) perform a detailed safety analysis by analyzing all plausible forms of trouble as to cause, effect, and corrective measures, adopting the final detailed safety concept;

· (4) conduct the safety acceptance of the plant by doing a nominal/actual comparison and carrying out functional tests.

Appropriate safety engineering involves assessing the hazards as to both possible scope and probability of occurrence. The methodological aids available for use in these tasks are all characterized by clear and easily understood structures and systematic procedures. These methods are similar in that they are characterized as “deviation analysis” in that they look for possible hazards which can arise in the process, in the plant, or in plant operation if an error occurs, or if the state or sequence of actions deviate from the prescribed state of sequence. The selection of the method should depend on the aim of the study. If the aim is a purely qualitative evaluation focusing on a very small balance volume, the checklist method will in most cases be sufficient. A checklist enumerates points that, according to experience, are associated with hazards in the handling of substances and mixtures of substances, or in the performance of a technical process. It can be as detailed as desired, and must be suitable for the kind of analysis being carried out. Checklists can be used to ensure the completeness of the safety concept in later phases of a plant project so as to ensure that all possible events are included in the safety concept. They have the obvious advantage that they can be adapted to any problem. Their drawback is that things not included may not always be recognized and dealt with. Therefore more sophisticated methods have to be applied if the character of a system is more complex.

18.2.3 Structure of safety studies

Safety studies can be classified on the basis of the fishbone model shown in Fig. 18.3. They are commonly subdivided into hazard identification studies, effect quantification studies, hazard probability quantification studies, risk quantification studies, and risk assessment and risk control studies.

Hazard identification studies include the process safety analysis, substances dossier, case histories of incidents during conceptual engineering, design-phase hazard study during basic engineering, and the HAZOP (hazard and operability study) and process hazard review before commissioning of the newly built plant. The process safety analysis is a systematic investigation into the acute inherent hazards (explosion, fire, chemical reactions, toxicity, etc.) of a process performed during the conceptual engineering phase. The analysis takes place at the level of flow diagrams and identifies locations where enough ingredients are present for a hazard to occur. The substances dossier is a collection of all physical and SHE (safety, health, and environment) data on the raw materials, auxiliary materials, by-products, and final products to be used in the plant. This is the basis for starting a process safety analysis. The design-phase hazard study is performed to ensure that the plant design is such as to allow safe operation in terms of SHE under all circumstances. It concerns a systematic investigation during the engineering phase into all foreseeable deviations from normal process operation. The unwanted situations are found and classified into effect categories.

The HAZOP study is a final systematic inspection with regard to the SHE aspects of the design of a plant or part of a plant before commissioning. It is most probably the best known and most widely acknowledged of the qualitative methods. The method is very systematic and reaches a high degree of completeness with respect to the identification of possible process deviations. The prerequisite to the conduction of a HAZOP analysis is the existence of a thorough process description, including detailed information on the design and process data as well as complete piping and installation diagrams. The necessary effort may become extremely large for complex plants and is already quite significant for smaller units. The process hazard review is a HAZOP study for existing plants.

Fig. 18.3: The fishbone model.

Effect quantification studies include the maximum credible accident analysis and classification of unwanted situations into effect categories. The maximum credible accident (MCA) analysis is performed during conceptual engineering to determine the unwanted event in a plant which, while still being credible, has the most serious effects on the surrounding area. In principle an MCA study covers only the effects on humans. For a number of scenarios the maximum effects on the surroundings are calculated. These effects are pressure waves of an explosion, fatalities from exposure to an acutely toxic substance, and heat radiation from a fire. The scenario involving the largest effect distance is the MCA. Classification of unwanted situations into effect categories uses a flow chart to classify the effects into effect categories to enable risks to individuals, the environment, and economic interests to be controlled in a structured manner during basic engineering.

Hazard probability quantification studies include the classification of hazardous areas and risk (fault tree) analysis. During basic engineering, the classification of hazardous areas is used to classify plant areas where an ignitable or explosive atmosphere may be present, so that adequate measures can be taken with regard to potential ignition sources. Fault-tree analysis is one of the most important methods to clarify the logical connection of a disturbance and the events which may have caused it. The construction of a fault tree always begins with the top event. Having identified this top event, the immediate causes that will lead to it are searched for. In order to use fault trees for the quantitative determination of the probability of an occurrence of the top-event, probabilities and frequencies are attributed to the individual states and malfunctioning components.

Risk quantification is done through the quantitative risk analysis that provides information during conceptual engineering on the probability that a person located at a given place outside the fence dies as a consequence of an incident in the plant. This is called the individual risk. In addition, the group risk is determined for the plant, and this is done on the basis of the actual population density. Examples of risk assessment and risk control studies are the design basis for plants and buildings, minimum distances between process units, pressure-relieving and depressurizing venting systems, explosion hazards in spaces and equipment, and SHE assessment. For instance, the design basis for plants and buildings is made to ensure that in the event of a disaster in a process plant, buildings continue to perform their vital protective functions, and adjacent plants are adequately protected to prevent domino effects. It is investigated, for instance, whether buildings must be pressure resistant and/or gas tight.