本文選題：NUTS + 生態(tài)模型��； 參考：《地球信息科學(xué)學(xué)報》2017年10期

【摘要】：參數(shù)優(yōu)化方法是準確估計生態(tài)模型參數(shù)、降低其不確定性的有效手段。本文提出一種基于貝葉斯機器學(xué)習(xí)的No-U-Turn Sampler(NUTS)生態(tài)模型參數(shù)優(yōu)化方法。NUTS是一種高效的參數(shù)優(yōu)化方法,每次取樣中利用遞歸算法生成候選參數(shù)集(二叉樹)推斷參數(shù)的后驗信息,如果滿足約束條件"非U型回轉(zhuǎn)",不斷構(gòu)建子樹更新參數(shù);否則,記錄本次抽樣的"最優(yōu)"參數(shù)集,并開始下一次取樣,直到獲取足夠樣本。該算法在每次取樣中充分優(yōu)化參數(shù),避免因隨機游走行為產(chǎn)生冗余抽樣,提高了參數(shù)優(yōu)化效率。本文以千煙洲亞熱帶人工針葉林碳通量模擬為例,基于Pymc3框架利用NUTS參數(shù)優(yōu)化方法實現(xiàn)了碳通量(Net Ecosystem Exchange,NEE)模型參數(shù)反演,并與Metropolis-Hastings(MH)方法進行對比。結(jié)果表明,本文算法的參數(shù)值達到穩(wěn)定波動時的抽樣次數(shù)減少了85%左右,參數(shù)優(yōu)化效率提升3倍左右。參數(shù)優(yōu)化后,2種NEE模型中7個參數(shù)不確定性降低10%~53%。此外,NEE模擬效果明顯提升,模擬值與實測值的R2分別提高23%和17%,RMSE分別降低3%和4%。綜上所述,本文提出的參數(shù)優(yōu)化方法對生態(tài)領(lǐng)域的參數(shù)估計或數(shù)據(jù)同化工作具有一定的借鑒意義。
[Abstract]:Parameter optimization method is an effective method to estimate the parameters of ecological model accurately and reduce its uncertainty. In this paper, a No-U-Turn Samplern NUTS-based ecological model parameter optimization method based on Bayesian machine learning. NUTS is an efficient parameter optimization method. The recursive algorithm is used to generate the posterior information of the candidate parameter set (binary tree) in every sampling. If the constraint "non-U rotation" is satisfied, subtree update parameters are constantly constructed; otherwise, the "optimal" parameter set of this sampling is recorded and the next sampling begins until sufficient samples are obtained. The algorithm optimizes parameters in every sampling, avoids redundant sampling due to random walk behavior, and improves the efficiency of parameter optimization. In this paper, the numerical simulation of carbon flux of artificial coniferous forest in Qianyanzhou subtropics is taken as an example. Based on Pymc3 framework, the parameter inversion of net Ecosystem Exchange nee) model is realized by using Nuts parameter optimization method, and compared with Metropolis-HastingsMH method. The results show that the sampling times of the algorithm are reduced by about 85% and the efficiency of parameter optimization is increased by about 3 times when the parameters of the algorithm reach stable fluctuation. After parameter optimization, the uncertainty of 7 parameters in the two kinds of NEE models is reduced by 10% and 53%. In addition, the simulation effect of nee was significantly improved, the R2 of simulated value and measured value were increased by 23% and 17%, respectively, and RMSE decreased by 3% and 4%, respectively. To sum up, the parameter optimization method proposed in this paper can be used for reference in the field of ecological parameter estimation or data assimilation.
【作者單位】： 沈陽農(nóng)業(yè)大學(xué);中國科學(xué)院地理科學(xué)與資源研究所生態(tài)系統(tǒng)網(wǎng)絡(luò)觀測與模擬重點實驗室;中國科學(xué)院大學(xué);
【基金】：國家重點研發(fā)計劃(2016YFC0500204) 國家自然科學(xué)基金項目(31501217、41571424) 遼寧省科學(xué)技術(shù)計劃項目(2014201001)
【分類號】：O212;TP18

【相似文獻】

相關(guān)碩士學(xué)位論文 前1條

1 李紅梅;機器學(xué)習(xí)方法和統(tǒng)計建模方法的預(yù)測比較研究[D];云南師范大學(xué);2016年

，

本文編號：2046106