مقایسه اثربخشی و کارایی در مشخصات فنی و بهینه سازی تعمیر و نگهداری

Optimization of technical specification requirements and maintenance (TS&M) has been found interesting from the very beginning at Nuclear Power Plants (NPPs). However, the resolution of such a kind of optimization problem has been limited often to focus only on individual TS&M-related parameters (STI, AOT, PM frequency, etc.) and/or adopting an individual optimization criterion (availability, costs, plant risks, etc.). Nevertheless, a number of reasons exist (e.g. interaction, similar scope, etc.) that justify the interest to focus on the coordinated optimization of all of the relevant TS&M-related parameters based on multiple criteria. The purpose of this paper is on signifying benefits and improvement areas in performing the coordinated optimization of TS&M through reviewing the effectiveness and efficiency of common strategies for optimizing TS&M at system level. A case of application is provided for a stand-by safety-related system to demonstrate the basic procedure and to extract a number of conclusions and recommendations from the results achieved. Thus, it is concluded that the optimized values depend on the particular TS&M-related parameters being involved and the solutions with the largest benefit (minimum risk or minimum cost) are achieved when considering the simultaneous optimization of all of them, although increased computational resources are also required. Consequently, it is necessary to analyze not only the value reached but also the performance of the optimization procedure through effectiveness and efficiency measures which lead to recommendations on potential improvement areas.

Technical specification and maintenance (TS&M) activities are associated with controlling risk or with satisfying requirements at Nuclear Power Plants (NPPs), which are candidate to be evaluated for their resource effectiveness in risk-informed applications [1]. The resource versus risk-control effectiveness principles formally enters in optimization problems where the cost or the burden is to be minimized while the risk or performance is constrained to be at a given level. This relationship applies when TS&M activities are optimized to minimize the cost while controlling unavailability of safety-related systems and plant risk, and vice versa. In the past, normally, the resolution of such a kind of optimization problems have been limited to focus only on a single parameter (STI, AOT, PM frequency, etc.) and/or adopting a single optimization criterion (availability, costs, plant risks, etc.) [2], [3], [4], [5], [6], [7] and [8]. However, the interaction that exists between TS&M requirements, and the similar purpose of the different TS&M tasks on increasing the system availability, which in some cases may be redundant, have brought a growing interest at present to focus on the coordinated optimization of relevant TS&M-related parameters based on multiple criteria [9], [10], [11], [12], [13] and [14]. The purpose of this paper is on highlighting benefits and improvement areas in performing the coordinated optimization of TS&M through reviewing the effectiveness and efficiency of common strategies used for optimizing TS&M at system level. A case of application is provided for a stand-by safety-related system to demonstrate the basic procedure and to extract a number of conclusions and recommendations from the results achieved. Particular attention is paid to show the interaction among these parameters by means of highlighting the impact that each one has on the various contributions to the unavailability and cost functions at system level. In addition, measures of effectiveness and efficiency used to compare the solutions found for the different optimization alternatives are introduced.

The interaction that exists between TS&M requirements, and the similar purpose of the different TS&M tasks on increasing the system availability, which in some cases is redundant, have brought a growing interest to focus on their simultaneous and multi-criteria optimization in the context of the use of PRA for risk-informed applications. In the simultaneous optimization of TS&M-related parameters based on risk and cost, one normally faces multi-modal and non-linear objective functions and a variety of both linear and non-linear constraints. This results in a complex and discrete search space with regions of feasible and unfeasible solutions for a discontinuous objective function that eventually presents local optima. The SSGA adapted to handle with the implicit constraints and with the penalization has proved their capability to solve this kind of problems. All of the optimized TS&M parameters in the example of application represent practical alternatives that can be implemented in a NPP under a real situation. In addition, the optimized values for the different parameters and the reduction in risk and cost achieved depend on the number of parameters considered and the best results (minimum risk or cost) are obtained for the simultaneous optimization of all the TS&M-related parameters (eight parameters herein) but depending on the objective function selected for the optimization problem, which demonstrate the importance of the simultaneous optimization of all the relevant TS&Mrelated parameters. However, convergence efficiency for the eight-parameter optimization achieves the worst result. Computation resources may be an important limiting factor, so that, improvement of the convergence efficiency should be further investigated. Thus, convergence efficiency could be improved by increasing the computation speed (for example, evaluation in parallel of the objective function) and/or improving the convergence effort, that is to say, reducing the iterations required to achieve a converged solution (e.g. improving the GA, or developing a new algorithm). In addition, none of the optimizations performed can be considered as risk and cost effective simultaneously, as the score reached for the multi-criteria effectiveness is very poor. In fact, the five-parameter cost optimization is the best one from the effectiveness point of view, which indicates that middle points in Fig. 3 are more effective as they are closer to the hypothetical best point. This result suggests that new procedures to perform the multi-criteria optimization should be further investigated considering a multiobjective optimization in order to guide the search towards that region in Fig. 3 (e.g. effectiveness-weighted risk and cost functions, Pareto optimal, etc.).