IAFSA-GRNN在油田集输管道CO2腐蚀速率预测中的应用

为提高油田集输管道CO₂腐蚀速率预测的准确性,针对原始广义回归神经网络(GRNN)预测精度低的问题,提出改进的群智能算法优化原始GRNN的预测模型;分别使用GRNN模型、人工鱼群算法(AFSA)优化的GRNN(AFSA-GRNN)模型和自适应改进的AFSA-GRNN(IAFSA-GRNN)模型预测X65管线钢的CO₂腐蚀速率。结果表明:采用AFSA和IAFSA优化光滑因子S后,能大大提高GRNN模型的预测精度,预测结果的平均相对误差由36.09%分别减小至7.20%和6.90%;与AFSA相比,IAFSA优化的GRNN不仅具有更高的预测精度,还具有更快的收敛速度。AFSA-GRNN在第164次迭代计算时收敛,而IAFSA-GRNN在第109次迭代计算时收敛,说明AFSA经自适应优化能提高优化过程的收敛速度和GRNN的预测精度。     

Longitudinal monitoring of neutral and adaptive genomic diversity in a reintroduction

Restoration programs in the form of ex-situ breeding combined with reintroductions are becoming critical to counteract demographic declines and species losses. Such programs are increasingly using genetic management to improve conservation outcomes. However, the lack of long-term monitoring of genetic indicators following reintroduction prevents assessments of the trajectory and persistence of reintroduced populations. We carried out an extensive monitoring program in the wild for a threatened small-bodied fish (southern pygmy perch, Nannoperca australis) to assess the long-term genomic effects of its captive breeding and reintroduction. The species was rescued prior to its extirpation from the terminal lakes of Australia's Murray-Darling Basin, and then used for genetically informed captive breeding and reintroductions. Subsequent annual or biannual monitoring of abundance, fitness, and occupancy over a period of 11 years, combined with postreintroduction genetic sampling, revealed survival and recruitment of reintroduced fish. Genomic analyses based on data from the original wild rescued, captive born, and reintroduced cohorts revealed low inbreeding and strong maintenance of neutral and candidate adaptive genomic diversity across multiple generations. An increasing trend in the effective population size of the reintroduced population was consistent with field monitoring data in demonstrating successful re-establishment of the species. This provides a rare empirical example that the adaptive potential of a locally extinct population can be maintained during genetically informed ex-situ conservation breeding and reintroduction into the wild. Strategies to improve biodiversity restoration via ex-situ conservation should include genetic-based captive breeding and longitudinal monitoring of standing genomic variation in reintroduced populations.