Risk Assessment And Pre-warning Technique For Soft Yoke Mooring System Based On Monitoring

Soft Yoke Mooring System (SYMS) is widely used and exhibits the good performance in Bohai Bay. As one of the major facilities, SYMS is designed to moor ship to the jacket platform and transmit oil and power. However, due to the complexity of environmental load and structure motions, soft yoke mooring system accidents happened several times in recent years, which cause huge economical losses. So it is in urgent need to find a good way to assess and control the risk of SYMS to guarantee the safety of offshore oil and gas development.Owing to the highly complex environmental factors and shortage of history accident statistics, traditional methods have some shortages in risk assessment for SYSM. In this paper, a in-side monitoring system is built up to make the safety assessment on FPSO in Bohai Bay. By monitoring, long-term real-time information of environmental factors and structural response are obtained, which can be used to assess and control the risks. The main contents are shown as following:First of all, based on the consideration of SYMS and FPSO, a monitoring system is built to gain the environmental factors and structural response of the structure. On the comparison with the theoretical structural response, the accuracy of measuring data can be verified and be used as the basis of risk assessment.And then, according to the data of monitoring and operation experience, failure modes of SYMS have been studied and the hierarchy structure of risk factors has been constructed. As an example, quantitative risk assessment for overload has been selected and carried out. Based on the monitoring data, the probability distribution model of the mooring force is built to calculate the probability distribution. An empirical formula is present to assess the consequence. Then the risk can be quantitatively evaluated with the method of risk matrix.Finally, according to the results of risk assessment, the risk of overload must be controlled to insure safety. Focus on the nonlinearity and parameters quantization of the numerical calculations, a neural network is present to make the prediction of the mooring force in shallow water based on existing monitoring data. By adjusting parameters of the function, the quality of RBF network is improved. On the comparison with the mooring force in prototype monitoring, the present algorithm and simulated results show the good accuracy and generalization ability and provide the acceptable efficiency in engineering.