Bayesian sequential tests for comparing the number of species in two populations

We study a problem of scquentially comparing the number of species in two populations. We consider a sequential Bayesian approach which incorporates a sampling cost and a misclassification loss, and examine optimal and sub-optimal stopping rules. The optimal stopping rule is shown to be truncated when the prior distribution of the number of species is bounded. or when the prior is unbounded and satisfies certain conditions.     

A predictive method for the solubility of drug in supercritical carbon dioxide

We present a predictive approach for the solubility of drug in supercritical CO₂. The fugacity of drug in the solid phase is estimated from its melting temperature and heat of fusion, and the fugacity of the drug in its hypothetical liquid state. The fugacity of the drug is calculated from the Peng–Robinson (PR) EOS. Temperature and composition dependence of the interaction parameters a(T,x) and b(x) of the PR EOS are obtained from the quantum mechanics-based, COSMO-SAC solvation model. As a consequence, the method does not require input of experimental data of the mixture. The average logarithmic deviation (ALD-x) in predicted solubility of 46 drugs in subcritical and supercritical carbon dioxide (T = 293.15–473 K, P = 8.5–50 MPa, and 1160 solubility data ranging from 10⁻⁷ to 10⁻²) was found to be 0.81 (a factor of 5.3). The same method was also examined for solid solubility in a variety of solvents (60 solids including 34 different solvents (with different polarities) and 190 drug-solvent pairs) at ambient pressure. The ALD-x was found to be slightly better (0.58 or a factor of 2.89). The proposed method, the PR + COSMOSAC EOS, is thus a useful tool for a priori prediction of solid solubility in scCO₂, as well as for other solvents.     

A hybrid neural model (HNM) for the on‐line monitoring of lipase production by Candida rugosa

A mechanistic model was proposed by Gordillo for the representation of lipase production by Candida rugosa, with the bioreactor in batch and fed‐batch operation. However, the model was not able to represent the lipolytic activity. The objective of the present study is to propose an efficient hybrid neural‐phenomenological model (HNM) for this process. The experimental data used corresponded to fed‐batch operation with constant substrate feed rate at 2.8 × 10^?7; 5.6 × 10^?7 and 9.7 × 10^?7 kg s^?1. Artificial neural networks (ANNs) were trained to represent the aqueous and intracellular lipase activity and were further associated with a reduced version of the mechanistic model of the proposed HNM. When compared to the experimental data, the HNM exhibited higher accuracy. The HNM can be employed in process monitoring using only on‐line measurements of CO₂ and substrate feed rate to infer enzyme activities and also substrate and biomass concentrations. Copyright © 2007 Society of Chemical Industry     

A novel dynamic bayesian network‐based networked process monitoring approach for fault detection,propagation identification,and root cause diagnosis

A novel networked process monitoring, fault propagation identification, and root cause diagnosis approach is developed in this study. First, process network structure is determined from prior process knowledge and analysis. The network model parameters including the conditional probability density functions of different nodes are then estimated from process operating data to characterize the causal relationships among the monitored variables. Subsequently, the Bayesian inference‐based abnormality likelihood index is proposed to detect abnormal events in chemical processes. After the process fault is detected, the novel dynamic Bayesian probability and contribution indices are further developed from the transitional probabilities of monitored variables to identify the major faulty effect variables with significant upsets. With the dynamic Bayesian contribution index, the statistical inference rules are, thus, designed to search for the fault propagation pathways from the downstream backwards to the upstream process. In this way, the ending nodes in the identified propagation pathways can be captured as the root cause variables of process faults. Meanwhile, the identified fault propagation sequence provides an in‐depth understanding as to the interactive effects of faults throughout the processes. The proposed approach is demonstrated using the illustrative continuous stirred tank reactor system and the Tennessee Eastman chemical process with the fault propagation identification results compared against those of the transfer entropy‐based monitoring method. The results show that the novel networked process monitoring and diagnosis approach can accurately detect abnormal events, identify the fault propagation pathways, and diagnose the root cause variables. © 2013 American Institute of Chemical Engineers AIChE J, 59: 2348–2365, 2013     

CO2混合工质的气液相平衡的混合规则对比与预测研究

CO₂混合工质兼顾高效和环境友好的特点,在新一代热功转化循环中受到广泛关注。混合工质气液相平衡性质是循环分析与计算的基础。为了提高CO₂混合工质气液相平衡数据的计算精度,采用PR状态方程结合三种混合规则（vdW、MHV1、WS）,对7种CO₂+HFCs/HFOs及4种CO₂+HCs混合工质的气液相平衡性质进行了计算。结果表明,对CO₂+HCs混合工质,vdW混合规则可达较好的结果;对CO₂+HFCs/HFOs混合工质,在亚临界区三者计算精度相近,但在超临界区,WS混合规则对计算精度提升明显。最后,提出了一种差值模型预测CO₂混合工质气液相平衡性质,预测的AARD(p)值为2.03%,AAD(y)值为0.0120,预测精度较高。