تجزیه و تحلیل حساسیت در مدیریت ریسک حفاظت از رعد و برق

The total risk R can be expressed by the following equation: R = NPL, where N is the number of dangerous events, P is the probability of damage and L measures the amount of loss. These parameters can be regarded as macroparameters in the sense that they can be given as a function of additional parameters (microparameters). Methods of calculation, check and refinement of parameter values have always been general themes of the standardization. The comparison of standards IEC 62305-2:2006 Ed. 1 and IEC 62305-2:2010 Ed. 2 is a good example of these efforts, as changes underwent in the determination of the collection area of service and the value of some parameters, moreover new parameters were introduced in order to break existent parameters into more components. Though the international standard contains the value of parameters, it does not give any information about the procedure how these values were determined, nor about the theoretical considerations behind. Therefore it is proper to ask whether the changes of a given parameter value or introduction of a new parameter make radical changes in resulting risk value or not. In this study sensitivity analysis was used to investigate how the individual macro- and micro parameters influence the risk. Moreover it was determined whether or not new parameters and calculations have an effect on the importance of the respective components of the resulting risk. The sensitivity analysis was done for two case studies in the standard, a country house and an office building respectively, with both editions of the standard.

The aim of the second part of the international lightning protection standard [2] and [3] is to decide the necessity of lightning protection of an object. For this purpose a relatively complicated calculation method is used. At the beginning the losses (the loss of human life, of public services, of cultural heritage and of economical value) shall be determined regarding the striking point of the flash (direct or indirect strike to the structure and to the connecting networks), then the risk values of these losses shall be calculated. The sum of these risk components gives the total risk. If the total risk is higher than the tolerable risk, then protection is needed, decisions shall be made which techniques to adopt in order to reduce the total risk regarding the given structure. The parameter values needed in the risk calculations are chosen based on the specifications of the building in question (e.g. existing lighting protection measures, coordinated SPD system, etc.). These parameter values are published in the standards and obviously have a decisive effect on the calculated total risk, based on which decisions can be made regarding further protection measures. Therefore it is important to investigate how sensitive the total risk is to the individual parameters given by the standard. This paper presents such a study. Sensitivity analysis was done for two case studies in the standard: a country house and an office building respectively, using both editions of the standard. Based on the results the order of the resulting risk sensitivities to the different parameters were determined, than a comparison was made between the two editions. This gives an overview about which parameters are the most relevant, and which ones are unimportant in these two cases.

In this paper sensitivity analysis was done for two problems, a country house and an office building (case studies of the standards), using both editions. This enables us to focus on the exact determination of those parameter values for which the total risk is most sensitive. The results show that in the studied cases the total risk is most sensitive to the lightning ground flash density and to the parameters regarding physical damage. The location of the building and the lines (as well as the cl, cE parameters characterizing them) are also relevant. However the sensitivity to the collection area is not affected significantly, despite the change in the formula for its calculation. Breaking down the losses to further parameters in the new edition gives the advantage of unambiguously determining the probabilities. Moreover the majority of the newly introduced parameters has a significant effect on the total risk. Finally, the total risk sensitivity to parameters regarding the injuries of living creatures is relatively small, even large changes in the parameter values result in only minor variations in the total risk. The probability of physical damage is much higher than the probability of injury to living beings; this is shown by the positions of the relating parameters in the list. According to the results in the case of the country house the first four, and in the case of the office building the first six parameters should be investigated in more details. The risk sensitivity is relevant for these parameters (S≈1)(S≈1). Of course, each objects need a separate model, therefore sensitivity analysis is suggested for more types of buildings. Using these results standardization can be made more efficient in the future, by focusing on the exact determination of the parameters for which the total risk is most sensitive.