Fatty acid methyl esters(FAMEs) are the main components of biodiesels. In order to obtain their thermophysical properties data, the sound speeds of three FAMEs were measured using Brillouin light scattering method(BLS) at the temperature from 293.15 to 463.15 K and pressure of 0.1 MPa. The correlations for the sound speed of FAMEs in literature and this paper were also fitted as the function of temperature to satisfy the usage in engineering areas. The AADs were 0.29% for methyl caproate, 0.24% for methyl heptanoate and 0.27% for methyl caprylate, respectively. And the experimental data were also used to predict the density and the surface tension by Wada's model and Auerbach's model, respectively. It showed that the predictive ability of Wada's model for the density was good, while Auerbach's model did not perform well in the prediction of the surface tension.

The flow patterns of liquid inside nozzles have great effect on spray characteristics in flashing spray. A transparent nozzle was manufactured to investigate the flow characteristic inside nozzles during flashing spray and its effects on spray characteristics. The flow patterns of R134a inside the nozzle and spray outside the nozzle were recorded by a high-speed CCD camera under different injection pressure of 0.77, 0.8, 0.9 and 1.5 MPa. The relationship between the internal flow inside the nozzle and external flashing spray were analyzed. It was found that spray patterns will expand immediately around the exit of the nozzle when vaporization occurred inside the nozzle, while the further development of spray radius could be restricted in the same condition. The spray cone angle caused by internal flashing spray was smaller than that caused by external flashing spray in the same condition for the flashing spray of R134a which had high superheated degree at room temperature and atmospheric pressure. The internal flow pattern was extremely unstable under a certain scope of injection pressure, and it was helpful to form stabilized spray patterns when homogeneous bubble flow was formed inside the nozzle.

SiO₂, Al₂O₃ and TiO₂ nanofluids based on deionized water with a particle volume fraction of 0.1% were prepared by the two-step dispersion method. Surfactants were added into the nanofluids to reduce particle aggregation and enhance stability. An ultraviolet spectrophotometer was used to test the absorbance of nanofluids as the absorbance decreased with decreasing concentration of nanoparticles suspended in liquid. Based on the principle of transient plane source(TPS) method, the 2500 S thermal constant analyzer was employed to conduct the thermal conductivity of nanofluids. In order to investigate the heat transfer performance in a triangular microchannel heat sink using nanofluids, an Infrared Thermal Camera(ImageIR 3350, Germany) was inverted and hanged immediately over the microchannel heat sink to observe the temperature distribution on the substrate. The condition of heat dissipation was imitated by a DC power supply(34420A, Agilent, China), which would energize to the thin film heater at a heat flux of q=200 W·cm^-2. As different nanofluids were studied, DI-water was used to clean the experimental system after the previous experiment was done to avoid the residues of nanoparticles. The results reflected that the surfactants had effect on the absorbance of nanofluids, and the particles would aggregate with the increase of standing time. The thermal conductivity and convective heat transfer were improved by adding nanoparticles. The average temperature on substrate was cooled down, and the uniformity of temperature was also repaired. As the result, TiO₂ nanofluids had a better behavior than SiO₂ and Al₂O₃ nanofluids.

Experiments of pulsating heat pipe(PHP) were conducted with graphene nanoplate(GNP) nanofliuds at different mass fractions(0.01%, 0.05% %, 0.08% and 0.10%) under various filling ratios(45%-90%) and heat inputs(10-100 W). The results indicated that when the filling ratio was 45%, GNP nanofluids can significantly improve the dry state in PHP. The filling ratio ranged from 55%-70% and PHP with GNP nanofluids at the mass fraction of 0.01% showed better heat transfer performance and the maximum reduction on thermal resistance of PHP with nanofluids was 83.33% compared to that with deionized water. The main reasons for improving the heat transfer performance of PHP were higher thermal conductivity and better surface wettability of GNP nanofluids.

The purpose of this paper is to investigate the heat transfer performance of the tubes with twisted elements at different twist ratio variation Tv. On the constant wall temperature boundary condition, the characteristics of heat transfer and pressure drop are investigated experimentally and numerically. Tv are within the scope of -5-5 and water is used as working fluid．The comprehensive heat transfer and heat enhancement mechanism are analyzed at different Tv. The results show that the condition of Tv>0 along the flow direction is superior to the condition of Tv=0. The average heat transfer enhancement ratio of Tv=2.5 is 5.0% higher than that of Tv=0. However，the heat transfer effect of Tv<0 is inferior to Tv=0, and thus the layout should be avoided. The flow field structures are altered by the change of the twist ratio of the elements. For the case of Tv>0, the vorticity and the action range of the detour-flow vortex within more than half of the element area are improved significantly with the increase of Tv. At the same time the coordination of pressure, velocity and temperature fields is improved and heat transfer is enhanced accordingly.

Computational particle fluid dynamics(CPFD) approach was employed to investigate the gas-solid flow behavior in the recirculation system of the dense transport bed. The influences of the drag model and the particle close packing limit on the simulation results were analyzed, and the appropriate model parameters were determined. Through the comparison of the simulation results under three operation conditions, the variation trends of the pressure distribution and solids circulating rate with the aeration flowrate were obtained, which was in agreement with experimental observation. The pressure gradient distribution, the gas flow direction and the solids concentration distribution were analyzed. The absolute value of the pressure gradient at the aeration inlet was the local maximum. When the aeration flowrate was zero, the pressure gradient was close to zero in a big area above the inlet. When the aeration flowrate was big enough, there was bubble around the inlet. A high pressure gradient would be formed in the lower area of the standpipe due to the lack of aeration at higher elevation, which would slow down the solids moving. The role of aeration gas on loosening particles just existed in a limited region near the inlet and its role on uniform pressure gradient distribution was more important to enable uniform gas-solid flow in the standpipe. In the design of aeration conditions, the standpipe pressure drop was determined by the solids circulating flowrate, and it needed appropriate aeration to keep constant gas velocity, leading to more uniform pressure gradient in the standpipe.

The plume dynamic characteristics from underwater pipe downward leakage are experimentally and theoretically investigated. The Lagrangian integral method is used to establish the dynamic model of downward plume. The model is numerically solved and the main plume parameters are obtained. The simulated results show that the simulated values in plume radius and length are generally consistent with the experimental values, but the simulated radius is higher than experiment value in area near plume terminal. The experimental and simulated results also show that for downward leakage the initial momentum of plume quickly decays and the plume length is very short compared with upward leakage. With increasing Froude number, the maximum length of upward plume rapidly grows while the maximum length of downward plume slowly grows.

In order to study the effects of porous media on supercooling degree of ethanol solution, the experimental system of porous media with different porosities and thermal conductivities was built. The effect of ethanol concentration on the supercooling degree was investigated by changing the ethanol concentration. The method of heat transfer enhancement was explored by changing the porosity and thermal conductivity of porous media. The data were processed by statistical method and the experiment at each condition was repeated 32 times. The experimental results showed that the porosity and thermal conductivity of porous media had effects on the supercooling degree of ethanol solution. With decreasing porosity and increasing thermal conductivity, the average supercooling degree decreased, and the stability was improved. High concentration had obvious effect on decreasing the supercooling degree. The higher the concentration, the lower the average value and the maximum value of supercooling degree, while the change of concentration had little effect on the stability of supercooling degree.

When impact velocity of liquid droplets reaches to a critical breakthrough condition, subdroplets will be generated and passed through mesh screen membrane in droplet-capturing devices of gas-liquid separators, which could cause severe performance and safety concern of downstream processes. Critical breakthrough velocity was measured on various mesh screens by visualization experiment and was studied for its dependence on factors including diameter and wettability of liquid droplets as well as structure parameters and inclination angle of mesh screen. Critical breakthrough was governed by normal dynamic pressure together with water hammer pressure in competing with capillary pressure. The normal dynamic pressure was a dynamic pressure component with regards to inclination angle of mesh screen. Water hammer pressure was created by significant water compression within a very short time of droplet impact on mesh screen membrane, and was related to the number of mesh pores in droplet projection area. The capillary pressure was related to location of liquid-gas interface in mesh pore and wettability of mesh screen. Both water hammer pressure and capillary pressure were affected by structure parameters of mesh screen. A non-dimensional critical criterion for liquid droplet breakthrough was established from force analysis of liquid-gas interface in mesh pore in combination with consideration of all influential factors. The critical criterion was matched well to experimental results and would be useful for designing gas-liquid separator with mesh screen membrane. The increase of inclination angle, decrease of droplet diameter, and increase of mesh screen hydrophobicity could increase critical impacting velocity and avoid occurrence of breakthrough condition.

It was of great significance correctly to predict the vapor diffusion and emission inside the tank for the researches of gasoline evaporation loss and vapor pollution control during the operation of loading gasoline into a tank. Then, two key parameters of the volumetric ratio λ of the displacement mixture gas of the vapor-air to the loaded gasoline and the qualitative ratio η of the evaporation loss to the loaded gasoline were mainly considered. Based on the models of volume of fluid(VOF), mass transfer and RNG k-ε turbulence, the evaporation loss was numerically simulated and experimentally investigated during the splash loading operation, and the oil vapor-air diffusion was analyzed and compared for the different loading exit heights, different loading velocity and the different initial vapor mass fraction. Meanwhile, an experimental system of evaporation loss in loading into a tank was built up to verify the numerical simulation, and the results of the numerical simulation were agreed well with the experimental data. The simulation results furthermore showed that the higher of the loading exit, the greater the qualitative ratio η of the evaporation loss to the loaded gasoline. The qualitative ratio of high exit was at around 0.34% and the qualitative ratio of low exit at around 0.025% by the increase of the loading velocity. The qualitative ratio η of high exit was at around of 0.44%, the qualitative ratio of mid exit at around 0.21% and the qualitative ratio of low exit at around 0.043% by increasing the initial vapor mass fraction. It was recommended that the effect of loading velocity and the initial vapor mass fraction should be considered in API loss formula by using a clean tank and low exit when loading and reducing loading speed appropriately before the loading pipe exit was submerged.

Surfaces with patterned domains of different wettabilities can be used for controlling the impact process. The dynamic behavior of droplet impact on hybrid surfaces with different wettabilities is simulated using the VOF method by dimensionless. Effects of Weber number and hydrophilic region diameter are studied with constant Reynold number. A drop impact regime map is generated, in which the impact dynamics is characterized as a function of Weber number and relative diameter of hydrophilic region. With the increase of Weber number, the droplet presents three different status: deposition, pinch-off as a short liquid column and break up as a long liquid column. The threshold Weber number where pinch-off occurs will increase with the increase of hydrophilic region diameter. For high Weber number, the maximum spreading factors are the same for different hydrophilic region diameter. While for low Weber number, the maximum spreading factors will increase with it.

To improve efficiency of Ce-based desulfurization agent in deep desulfurization of flue gas in fluid catalytic cracking(FCC) petroleum processing, reaction mechanism of SO₂ and O₂ over CeO₂ surface were investigated in both experiments and calculations by density functional theory(DFT). SO₂ removal from simulated flue gas with O₂ presence was carried out over CeO₂ under 60℃ and 200℃. CeO₂ characterization before and after desulfurization by XRD, IR and XPS was used to study changes in valence and relative amount of S and Ce as well as to explore possible oxidative reaction path of SO₂. Models of monomolecular O₂ adsorption and bimolecular O₂ and SO₂ adsorption on O-deficient CeO₂(111) and MgAl₂O₄(111) were established from DFT simulation. SO₂ oxidation and energy barriers of each reaction step were calculated using LST/QST method. Results indicated that CeO₂(111) had better catalytic effect for SO₂ desulfurization than MgAl₂O₄(111) and CeO₂ could transport lattice oxygen in oxidation with SO₃ desorption as rate-control step.

A Pt/SAPO-11 dual functional catalyst was prepared from hydrothermally synthesized SAPO-11 molecular sieve and used to study hydroisomerization of n-C₈, n-C₁₂ and n-C₁₆ alkanes. Results indicated that this catalyst had good isomerization activity and high selectivity as a result of appropriate acid center and hierarchical pore distribution. Yields for monomethyl-branched isomers and total branched isomers, ideal components for low freezing point diesel and high-grade lubricant, were above 60% and 75% respectively. The distribution of isomerization products was almost the same for n-alkanes of different chain lengths, which monomethyl-branched isomers were about 75% of total isomerization products and 40% were monomethyl-branched isomers with methyl group at middle chain. Isomerization and hydrocracking reactions were more likely to occur on longer chain n-alkanes. Under the same conversion, long-chain n-alkanes showed increase in hydrocracking but decrease in isomerization selectivity.

15% MnO_x/5% WO₃/ TiO₂ catalyst for low-temperature selective catalytic reduction of NO_x was synthesized by an impregnation method. NH₃-SCR mechanism was explored systematically and possible pathway was proposed through in situ FT-IR study on microscopic transient and steady states under various conditions of single or mixed gas feed with multiple pre-treatment. The results revealed that NH₃-SCR reaction over the 15%MnO_x/5%WO₃/TiO₂ catalyst proceeded mainly by the Eley-Rideal mechanism with the Langmuir-Hinshclwood pathway only at certain temperature. NH₃ adsorbed on the Lewis acid sites of the catalyst surface was the main source of reductant; however, more NH₄⁺ adsorbed on the Brønsted acid sites of the catalyst surface involved in the NO_x reduction with increase of reaction temperature. The adsorption and activation of NH₃ was key step for the whole SCR reaction. NO₂ easily reacted with adsorbed NH₃ with stronger affinity to catalyst surface than NO. The bidentate nitrate, which formed considerably at oxidation active centers of the catalyst by NO adsorption, didn't participate in NH₃-SCR reaction but obstructed adsorption and activation of NH₃. O₂ could promote NH₃-SCR by blocking formation of bidentate nitrate. The adsorption of NO and NH₃ on the catalyst surface was competitive at room temperature, which NO adsorption was higher than NH₃ at O₂ absence. The absorbed NH₃ could be massively activated and reacted with NO_x upon temperature reached to 100℃.

Three dimensional(3D) graphene aerogels(EGA) with interconnected networks were fabricated from graphite oxide(GO) and ethylenediamine(EDA) via chemical crosslinking method. The properties of EGA were characterized by scanning electron microscope(SEM), transmission electron microscope(TEM) and selected area electron diffraction(SAED). Adsorption performance of the aerogel in removing diesel oil in water was investigated. The results showed that the EGA adsorption capacities rose quickly in the first 5 min, and reached adsorption equilibrium in 30 min. The kinetic analysis indicated that the EGA adsorption behavior can be accurately fitted with a pseudo second-order model, and the adsorption rate increased with the increase of temperature. The activation energy(E_a), determined by Arrhenius law, was 23.94 kJ·mol^-1, suggesting physical adsorption dominated the adsorption of diesel oil onto aerogels. The intraparticle diffusion model(IPD) fitting results revealed that the adsorption sites for aerogels consisted of external surface, large inner pore spaces and small-size pore spaces between the graphene sheets. The adsorption isotherms were well fitted with Freundlich model.

Amino-functionalized task-specific ionic liquids are thought to have good application prospects in terms of CO₂ fixation and separation, due to their strong CO₂ absorption with selectivity under ambient conditions. In this study, four task-specific ionic liquids were synthesized, with their structures and properties characterized by IR, ¹H NMR. The performances of these four task-specific ionic liquids in terms of CO₂ absorption and regeneration were investigated. It was found that these four task-specific ionic liquids are supervisor to traditional ionic liquids in terms of CO₂ absorption. In addition, these four task-specific ionic liquids showed excellent regeneration performance, which indicated that they could be reused for CO₂ absorption. The solubility of CO₂ in ionic liquids was affected significantly by the viscosity, decreased by increasing the temperature, increased by increasing the pressure and absorbent concentration. Enhanced mass transfer can improve the regeneration efficiency. However, no significant effect on the absorption properties of ionic liquids was observed after multiple regenerations.

This paper focuses on the simulation, optimization and control of the reactive dividing wall column for the production of n-butyl acetate through ester exchange reaction. The optimal operating conditions are obtained with the minimum total annual cost(TAC) as the target function. Different matching relationships between manipulate variables and control variables are determined through the sensitive analysis of steady state relative gain and the Relative Gain Array(RGA) criterion, and then three control structures are built in Aspen Dynamics. It shows that the use of the reactant feed ratio to control the stage temperature is effective to handle the feed disturbances, which presents superior ability in maintaining stability of this system. At last, an improved control structure(CS3) is proposed, which deletes the ratio of reboiler duty and mixture feed rate and shows a great advantage in reducing overshoot of product purities. The specific control structure of CS3 is as follows: the reboiler load of reactive distillation column controls the temperature of the 16th plate of the reactive distillation column directly; ratio of feed rate for n-butanol and methyl acetate/methanol mixture controls the temperature of 28th plate of the reactive distillation column; and the reboiler load of stripping column controls that of the 3rd plate of stripping column directly.

When there is a fault in the chemical plant, the operators may not be able to find the root cause of the fault. Data-driven method can be a great help to reduce the uncertainty of root cause, even to locate the root cause. By analysis of chemical process data, the disturbance propagation way can be detected, which can help to locate the root cause. In this article, convergent cross mapping(CCM) with the characteristic of time is proposed to detect the causality in dynamic chemical process. Furthermore, the usage of Akaike information criterion is proposed to determine the most proper embedding dimension. To prove the effectiveness of the method, the new method is applied to the ecosystem examples, causality testing benchmark model and CSTR model. Comparing the result calculated by the original CCM, the effectiveness of the new method is found.

Organic Rankine cycle(ORC) and Kalina cycle are both promising ways for low temperature waste heat utilization, and these two technologies have their own advantages and disadvantages on using waste heat. In refineries, there is considerable waste heat. It is significant to choose a proper cycle system considering different waste heat sources for efficient utilization of energy. Thermal efficiency and exergy efficiency are two key parameters to evaluate energy performance of power cycle systems. In this paper, the waste heat sources are classified into three types(i.e., sensible heat source, combined heat source and latent heat source). An organic Rankine cycle and a Kalina cycle for low waste heat recovery are simulated by Aspen Hysys considering the characteristics of waste heat sources. The results show that when the waste heat is sensible heat source, the energy performance of Kalina cycle is better than that of ORC, while when waste heat is combined heat source and the ratio of latent heat source and sensible heat source(R) is equal to 1 or when waste heat is latent heat source, the energy performance of ORC is better than that of Kalina cycle.

Nonsharp split can improve the thermodynamic efficiency of the distillation system due to its avoidance of the remixing effects which are inherent to sharp split. Based on the Data Structure Theory, a method is introduced to synthesize and optimize the nonsharp complex distillation sequences. For an N-component mixture separation, the distillation sequences is comprised of N-1 columns, the nonsharp splits of which contain an arbitrary number of middle components. In order to synthesize the separation sequences above, the weighted directed digraphs can be applied to establish a new model which is suitable for the distillation columns. Additionally, based on the dynamic array “vector”, a storage structure of the weighted directed graphs, which combines the advantages of arrays and chain tables, is proposed. A series of operations are defined in the synthesis procedure and the breadth-first strategy is chosen to improve the synthesis efficiency. An example problem is solved to illustrate the high efficiency of this method. The results show that the optimal sequence of the example problem can effectively reduce investment cost and energy consumption.

The process system engineering methods for refinery production process have drawn increasing concerns in both academic and industrial communities due to the fierce global competition. Due to the complexity of refinery production process, the effective process model is still an open problem, which hampers process system engineering application. In some senses, the process model is the fundamental basis for successful application. To break this bottleneck, an integration between unit-wide optimal process control system and plant-wide scheduling system based modelling framework is proposed. The whole refinery production processes are divided into several different classes, and each class unit has a unique and well-designed model structure. Based on the big operational data collected from the unit-wide optimal control system, the multi-mode models are obtained to take varying crudes and operating conditions into account. This modelling mechanism can provide the concrete model for smart or intelligent refinery.

In hybrid wind/diesel/battery power systems, uncertainty of wind speed and volatility of user load have great influences on optimal design and operation of the systems and their battery storage systems. An optimization model of a hybrid power system with the battery storage system was proposed, in which the k-means algorithm was used for partitioning periods and analyzing uncertainty of wind speed and volatility of user load. In this model, the life cycle cost of battery was taken as a penalty function, and the cycle number of batteries was taken as a constraint. The objective function was the levelized cost of energy(LCOE) of the power system. The influence of uncertainty in wind speed on the design of the power system was investigated. A hybrid wind/diesel/battery system with a user load of 793 kW·h·d^-1 was adopted as an example case to demonstrate the proposed optimization procedure. The results show that the cycle number of battery declines sharply by considering the period partitioning based on fluctuations of wind speed and user load, which is helpful to effectively extend the lifespan of battery. When the running period is divided based on the change of wind speed, the cycle number of battery in one year is the least, and the LCOE of the power system is the lowest. If the maximum cycle number of battery could be prolonged from 2000 cycles to 10000 cycles, the LCOE of the power system reduces by 8.3%-16.6%.

In chemical processes, fault propagation pathways and root cause identification could be discovered though analysis of interactions and time delay relationships among different process variables. Because of its importance in improving process safety and operation profit, fault discovery has been a popular and challenging research topic. Common methods such as correlation and entropy transfer functions, which usually cannot get accurate time delay and interaction strength between variables by limited applicability for linear and weak nonlinear systems or high computation demand, have experienced many disadvantages in actual application. Recently, a new convergent cross mapping(CCM) algorithm in ecology has been considered suitable for causality analysis and time delay identification for nonlinear coupling process variables. However, CCM fails to find application in externally disturbed chemical processes because it cannot establish stable embedded flow from process data. An improved CCM, disturbance filtered cross mapping method(DFCM), overcame many challenges of creating stable embedded flow by analyzing external disturbance, filtering disturbed process data, and applying filtrated data to CCM calculation. Case studies showed good results of time delays and causality analysis, thus DFCM could be applied to chemical processes under external disturbance.

For fault detection in dynamic processes, a novel dimensionality reduction method was proposed on the basis of improved neighbor selection in neighborhood preserving embedding algorithm, i.e., time constrained neighborhood preserving embedding(TCNPE). Compared to neighborhood preserving embedding(NPE), which selected neighborhood only by Euclidean distance, TCNPE selected neighborhoods within certain timeframe by k-nearest neighboring method with a consideration of time series correlation between data points and constructed a localized constraining relationship between near and distant neighborhoods within this time window. First, TCNPE algorithm extracted main features of the process data and performed linear dimensionality reduction. Next, Hotelling's T² and SPE statistics were established for online process monitoring and kernel density estimation(KDE) was used to determine control limits. Case study and simulation of Tennessee-Eastman Process demonstrated efficacy of the proposed method.

A novel random walk algorithm with compulsive evolution(RWCE) was proposed on the basis of different heuristic methods for global optimization of heat exchanger networks. In RWCE algorithm, both integer(e.g., number of heat exchanger units) and continuous(e.g., area of heat exchanger) variables were optimized simultaneously by expanding or contracting randomly area of heat exchangers in the direction of targeting cost reduction. Moreover, when individuals walked around local optima, the RWCE algorithm could compulsively accept imperfect networks at certain probability such that it had strong capability of jumping out of the local optima and continuing global optimization. Several case studies indicated that the proposed RWCE algorithm, compared to other heuristic methods, possessed characteristics of simple evolution strategy, strong algorithm suitability and global searchability, which significantly improved optimization performance.

A novel design method was proposed on the basis of distributed deviation for consensus pass temperature control on industrial feed heaters, in order to overcome limitations of centralized balance control schemes on pass temperatures. In the method, only temperature deviations of adjacent passes were used as input in control algorithm to achieve uniform temperature across all passes while maintaining constant total feed flow. Compared to the traditional centralized balance control, this distributed deviation control scheme by using only temperature information of adjacent passes was more cost effective especially for cases with large number of passes. Sufficient conditions for the distributed deviation controller on industrial feed heaters with time delay and random disturbance were derived by Lyapunov function. The simulation results illustrated that the new controller design method was effective and the distributed deviation control scheme was feasible even for a class of industrial feed heaters with mismatched models.

A fault detection method for chemical process based on LPP-GNMF algorithm is proposed. NMF(non-negative matrix factorization) is a novel dimensionality reduction algorithm, with characteristics of positive pure additivity of latent variables in the mechanism, thus, when compressing the data, the information can be described based on the local characteristics inner the data. Compared to the traditional multivariate statistical process monitoring methods such as principal component analysis(PCA), NMF offers a better ability for data explanation. However, firstly, NMF requires the original data to meet the requirements of non-negative, which can not be guaranteed in the actual chemical process, in order to relax the non-negative requirements of the original data, a generalized non-negative matrix factorization(GNMF) algorithm is quoted. Secondly, GNMF does not take the local structure and geometric properties into account during the process of decomposition, which may not be accurate to deal with the problem of data. Aiming at this problem, the algorithm of combining GNMF with LPP(locality preserving projection) is proposed. The proposed LPP-GNMF algorithm is applied to the Tennessee Eastman process to evaluate the monitoring performance. The simulation results show the feasibility of the proposed algorithm compared with the PCA algorithm, the NMF algorithm and the SNMF algorithm.

To solve the issue of low accuracy of the traditional soft sensor methods of polypropylene melt index, an approach based on deep belief network and extreme learning machine（DBN-ELM）was used to the melt index prediction of polypropylene. Traditional deep belief network(DBN) applied the deep learning to the learning process of the deep neural networks. Different from traditional deep belief network, this approach applied the extreme learning machine algorithm(ELM) to the learning process of DBN to improve the DBN model. Firstly, deep belief network was employed to extract effective features from vibration data by numerical analysis. Then, the effective features were put into the extreme learning machine to proceed model training to obtain the soft sensor model. The experimental validation showed that the method was more accuracy than the traditional method.

Heat exchanger network synthesis can be described by a mixed integer non-linear programming(MINLP) model, which features non-convex, non-linear and non-differentiable optimization. The parallel computing technology based on GPU provides an efficient support for solving large scale models. In this work, a hybrid algorithm combined branch and bound method(BB) with sequential quadratic programming(SQP) is proposed to overcome difficulties in the existing parallel SQP algorithm, such as too many combinations of integer variables, dependency of initial values and etc.. The BB method is adopted in the hybrid algorithm instead of the exhaustive method. It can not only reduce the combinations of integer variables, but also select feasible initial values for the SQP algorithm. The solution quality is improved. The results of the examples show that the proposed parallel hybrid algorithm can solve the heat exchanger network synthesis problems efficiently. Compared to the serial algorithm, the proposed parallel algorithm has much higher executive speed with the speedup ratio of 39.

In petrochemical industry, operating conditions of heat exchanger networks(HEN) are always changed, which particularly fouling-induced heat resistance keeps increasing over time. To overcome variation of operating conditions, several optimizing methods on design and retrofit of multi-period HENs are available with no consideration of fouling effect on HEN. The inevitable HEN fouling and increase of heat resistance will drive operation deviate from optimum value and make it difficult to sustain energy saving. A synthesis design method of multi-period HENs based on continuous energy saving was proposed which optimal synthesis was achieved by targeting total accumulative cost and considering fouling influence. First, case study to analyze fouling influence on multi-period HENs showed that optimal value of multi-period HEN in each period had been changed due to fouling. Then, a model on optimizing multi-period HEN was established by considering continuous energy saving and objective function of total accumulative cost. Finally, case study indicated effectiveness of the design optimization method.

Crude oil fouling is a crucial problem in refinery heat exchanger network(HEN) that reduces heat transfer and affects routine production. Conventional strategy to mitigate fouling is to clean heat exchanger regularly and to optimize cleaning schedule accordingly. Backup heat exchangers are used in many industries with severe fouling concern in order to avoid reduction of heat recycle caused by taking the heat exchanger off-line, however, such practice has not been deployed in crude oil pre-heating systems. Hence, a method for fouling mitigation in HEN was proposed to optimize cleaning schedule with backup for key heat exchangers, which key heat exchangers were first identified and then simulated annealing algorithm(SA) was used to optimize schedule for fouling removal. A case study of crude oil preheat train showed that backup on key heat exchangers offered higher energy saving and better economic efficiency than current methods.

A self-adaptive differential evolution algorithm with multiple strategies(SMDE) was proposed to overcome premature or localized optimization of differential evolution(DE) as a result of fixed parameter settings. Based on basic framework of classical DE, the first step in SMDE was to create a candidate set of mutation strategy, scale factor(F) and crossover rate(CR). In the followed searching process, mutation strategy, F and CR for each individual variable in next evolutionary generation were determined self-adaptively from the corresponding candidate set according to knowledge learnt from previous searches, so that proper mutation strategies and control parameters could be set at various evolution stages. Compared to other famous DE variants on optimizing 10 routine standard testing problems, SMDE had better search precision and faster convergence rate. Moreover, study on estimation of uncertain parameters in dynamic process systems of chemical engineering showed that SMDE could effectively solve engineering optimization challenges.

Sealing performance is significantly affected by the change of seal face topography produced by thermal and force deformation or machining. A mathematical model of spiral groove liquid film seal considering radial taper and circumferential waviness was established. The dynamic Reynolds equation was solved by partial derivative method and the opening force, stiffness, dynamic stiffness and dynamic damping coefficients of liquid film seal were obtained by means of finite element method, and then the influence of radial taper and circumferential waviness on steady and dynamic characteristics of liquid film seal were analyzed. The results indicated that the opening force of liquid film seal decreased with the increase of taper and increased with the increase of waviness amplitude. Under the same waviness amplitude, the opening force was reduced with the increase of waviness number. Stiffness of liquid film gradually reduced with the increase of taper and the variation trend with waviness amplitude was different with the change of waviness number. The absolute values of dynamic stiffness coefficients of liquid film decreased with the increase of taper and the effects of waviness on axial stiffness coefficient and angular stiffness coefficients were more obvious. Dynamic damping coefficients of liquid film gradually decreased with increasing taper and increased with increasing of waviness amplitude.

Macromolecular polyurethane surfactants(PUS) with equal degree of polymerization but different compositions and structures were designed and synthesized by reacting PCDL, PBA and PPG with isophorone diisocyanate(IPDI). Methyl methacrylate(MMA) monomer was polymerized in PUS micelles to prepare WPUA emulsions and dried latex films with different latex structure and morphology. Transmission electron microscopy(TEM), Zeta nanosizer, and rotating viscometer were employed to characterize latex morphology, particle size, and shear viscosity of WPUA emulsions. ATR-FTIR, water absorption measurement, electromechanical universal testing machine, and UV-Vis spectroscopy were employed to characterize structures, water resistance, mechanical properties, and physical aging behaviors of WPUA films. PUS composition and structure had a significant influence on morphology and structure of WPUA latexes and properties of dried films. Distinct core-shell structure was achieved in WPUA latex using PCDL-based PUS as macromolecular surfactant. WPUA-PCDL film exhibited good mechanical and aging-resistant properties.

Ionic liquids, as a kind of novel green solvents, have been widely applied to structural stability study of protein due to their unique physicochemical properties. In this study, insulin was chosen as a heat-sensitive protein medicine to research the structural stability of protein in various imidazolium ionic liquids from a molecular insight by using molecular dynamics simulation. The results indicate that the ionic liquids are able to stabilize the molecular structure of insulin effectively at room temperature compared with aqueous solution. The RMSD values and the secondary structure of insulin show that the weaker the hydrogen-bond basicity of anion is, the shorter the alkyl chain of cation is, the more stable the molecular structure of insulin is. In order to explain thoroughly the effect of alkyl chain length of ionic liquid on the stability of the protein, the interaction and the number of contact between the insulin and dicyanamide imidazolium ionic liquids with different alkyl chain length are further analyzed. It is found that the cation stabilize effectively the molecular structure of insulin which relies mainly on the electrostatic interaction between the insulin and the 1-methylimidazole of cation. Comparing to ionic liquids with long alkyl chain, the interaction between the insulin and ionic liquids is more powerful, and the insulin also adsorb more anion and 1-methylimidazole of cation in ionic liquids with short alkyl chain. All these results indicate that the ionic liquids with short chain alkyl indeed improve the molecular structural stability of insulin. Generally, the molecular dynamic simulation, as adopted to this research, provides a molecular insight to investigate the stability of insulin in ionic liquids, which is important and essential for molecular design of ionic liquids and stability study of heat-sensitive protein drugs.

The chemical looping based on Fe₃O₄/FeO oxygen carrier was employed in the CO₂/CH₄ reforming to produce syngas. The process was simulated and analyzed using the Aspen Plus software to evaluate the performance of the system. CH₄ conversion rate, CO₂ conversion rate, energy efficiency and H₂/CO ratio in the syngas produced were calculated and the optimum operation conditions were obtained. Sensitivity analysis of the system including reactor temperature and pressure, and Fe₃O₄/CH₄ and CO₂/CH₄ ratios were also performed. It was found that CH₄ conversion rate of 97.91%, CO₂ conversion rate of 32.76%, energy efficiency of 93.77% and H₂/CO ratio of 0.93 in the syngas produced were obtained at the optimum operating conditions in the system. The results indicated that the system could potentially bring about considerable conversion rates of CH₄ and CO₂. Furthermore, the produced syngas with a lower H₂/CO ratio was suitable for the synthesis of dimethyl ether.

The new-type mechanical rotary disk(MRD), which was used for the experiment of excess sludge disintegration combination with ultrasound, was manufactured by independent design to improve the disintegration effect. To investigate the MRD disintegration effect, the experiments under different operation conditions were carried out. The results showed that d(the distance between the plates) and T_m(the disintegration time) were the important factors to influence SCOD₊(the added value of solubility chemical oxygen demand) and EDR(energy disintegration ratio). The SCOD₊ was 3130 mg·L^-1 after 25 min of MRD disintegration under the condition that d was 1.5 mm and the revolving speed was 2300 r·min^-1, which presented that the disintegration was effective. According to the center combination experimental design of Central-Composite based on the response surface method, SCOD₊ and EDR were considered as response to paid attention to the energy efficiency. SCOD₊ and EDR were 7871.78 mg·L^-1 and 8.475 mg·L^-1·kJ^-1 under the optimal conditions that d, T_m and T_u(ultrasound disintegration time) were 2.76 mm, 10.89 min and 17.2 min, respectively. The results were respectively 3.02 times and 2.7 times of What used MRD（2610 mg·L^-1，3.11 mg·L^-1·kJ^-1）and 2.48 times and 2.6 times of that used ultrasonic（3170 mg·L^-1，3.26 mg·L^-1·kJ^-1）under the same total disintegration time.

The coupling characteristics between ash deposit and dew point corrosion of deep-cooling flue gas on ND steel and 316 L stainless steel(control) was investigated in a 65 t·h^-1 circulating fluidized bed biofuel boiler. The ash deposit and corrosion layer were studied by X-ray fluorescence spectroscopy(XRF), X-ray diffraction(XRD), scanning electron microscope(SEM) and energy dispersive spectrometer(EDS). Results indicated that the deposits on tube surface could be classified as layers of corrosion, coupling and ash deposit(NH₄Cl、SiO₂ and(NH₄)₂SO₄). The coupling layer contained ash deposit and metal corrosion by-product, which was prone to spall off because of strong adhesion to the ash deposit layer. The corrosion layer, mainly composed of metal compounds of chlorides and oxides, reduced dramatically in thickness with the decrease of inlet water temperature. The degree of corrosion of ND steel increased sharply when wall temperature was below water dew point of flue gas. ND steel showed corrosion resistance inferior to 316 L stainless steel at experimental conditions.

The effect of sulfolane on carbon dioxide(CO₂) absorption and desorption by monoethanolamine(MEA) aqueous solution was measured by gas-liquid stirring rig and true heat flow calorimeter, including CO₂ cyclic loading, absorption rate, heat of absorption and desorption. The experimental results show that sulfolane effect is significant for absorption rate, CO₂ cyclic loading and CO₂ desorption. In the condition of sulfolane addition, CO₂ absorption rate into MEA solutions decreases and the decreasing amplitude increases with CO₂ loading increase. Sulfolane can also accelerate desorption process and improve desorption level. Meanwhile, the heat duty, condensation duty and energy consumption are to some extent decreased. Under the condition of flue gas of coal-based power station and 20% sulfolane addition, the apparent CO₂ absorption rate averagely decreases 10%, CO₂ cyclic loading increases 24% and unit energy consumption for CO₂ desorption decreases 18% in comparison with 20% MEA solution.

Conventional dissolved air flotation(DAF) has the problems of low efficiency for microbubble meshing and particle adhesion, and unstability of the adhesion between microbubbles and particles. In this work, a novel countercurrent-cocurrent dissolved air flotation(CCDAF) was developed, of which the contact room was consisted of collision and adhesion contact room, and each room was introduced dissolved air water. The results showed that the CCDAF significantly enhanced the adhesion efficiency of microbubbles-floc. The average removal efficiency of turbidity and algae were 96.4% and 96.50%, respectively. The diameter of particles for effluent was mainly ranged in 2-7 μm. Most of the removed substance was macromolecules and hydrophobic organic compounds. COD_Mn, UV₂₅₄, DOC had achieved 37.6%, 46.3% and 32.11%, respectively, indicating the significantly higher removal efficiency of CCDAF than the traditional DAF. The analysis of the removal mechanism between microbubbles and particles showed that the collision, adhesion and copolymerization in countercurrent room and collision, adhesion, wrapped, meshing and adsorption-bridging in cocurrent-contact room were probably the reasons to enhance the removal efficiency of this CCDAF.

A novel A₂N₂ system was developed to deal with low C/N domestic wastewater. The operating performance under different nitrification duration distribution ratio in N-SBR unit was investigated. The experiment was conducted as follows: under the condition of A²/O-SBR unit anaerobic was set as 1.5 h, anoxic 2 h, aerobic 0.5 h and the aeration flow rate of A²/O-SBR and N-SBR units was 100 and 120 L·h^-1, respectively, and then the nitrification time distribution ratios were set as 5:1, 4:1, 3.5:1, 3:1, 7:1 and 8:1 to determine the optimal operation parameter. The results indicated that the efficient utilization of carbon sources was achieved in the A²/O-SBR unit, while the nitrification duration distribution ratio had little impact on the removal of organic substances. In order to ensure a better nitrification and denitrification and phosphorus removal performance, the first nitrification duration time must be≥3.5 h. Under the condition of total nitrification time was unchanged, an appropriate increase in the first nitrification time would greatly improve the system TN removal. An appropriate increase in the secondary nitrification time can reduce the effluent NH₄⁺-N concentration to meet the first A discharge standard in China(GB 18918-2002). When nitrification time allocation ratio was 4:1, the system achieved the optimal nitrogen and phosphorus removal performance with average effluent concentrations of TN and PO₄^3--P of 11.5 and 0 mg·L^-1, respectively, and the corresponding average removal efficiency was 75% and 100%, respectively.

A new method to improve the CO₂ capture performance of carbide slag during the cyclic calcination/carbonation process by adding HCl intermittently(indirect chlorination) was proposed in this work. The effects of intermittent addition of HCl on the cyclic carbonation characteristics of carbide slag under different reaction conditions such as cycle number when HCl was added, carbonation temperature and volume ratio of CO₂/HCl were investigated in a dual fixed-bed reactor by adding HCl into carbonation atmosphere. The results showed that the CO₂ capture performance of carbide slag during the cyclic calcination/carbonation process can be improved by adding HCl intermittently during the carbonation. The optimum CO₂ capture performance of carbide slag was achieved when HCl was only added during the first 4 cycles and the CO₂ capture capacity after 10 cycles was 51% higher than that without addition of HCl. Chlorination happened between HCl and CaCO₃, which decreased CO₂ diffusion resistance through the compact product layer of CaCO₃, and thus higher carbonation conversion was obtained. The optimum carbonation temperature was still 700℃ in the presence of HCl. As the volume ratio of CO₂/HCl increased, the positive effect of HCl on the CO₂ capture performance of carbide slag decreased.

Supercapacitors were constructed with graphene as electrodes and ionic liquid N-methyl-N-propylpyrrolidinium cyanate([C₃mpyr] [OCN])/propylene carbonate(PC) as the mixed electrolytes. The electrochemical properties of constructed supercapacitors were tested at 50, 60 and 70℃ by cyclic voltammetry(CV), electrochemical impedance spectroscopy(EIS) and galvanostatic charge/discharge(GCD). The results indicated that the constructed supercapacitors exhibited good electrochemical properties at high operating temperatures. The supercapacitors gave the specific capacity of 295 F·g^-1 and delivered an energy density as high as 118 W·h·kg^-1 at 70℃ with the current density of 0.5 A·g^-1. Additionally, the constructed supercapacitors showed a good cycling stability.

Previous study have examined many materials to remove the vanadium ions from aqueous solution, but it is difficult to meet the national discharge standard. In this paper, a mesoporous chromium hydroxide(Cr(OH)₃), which has a high specific surface area(312.70 m²·g^-1) and a large pore volume(0.48 cm³·g^-1), was synthesized by adding NaOH solution into CrCl₃·6H₂O solution drop by drop. The mesoporous Cr(OH)₃ powders were used as adsorbents to remove vanadium ions from aqueous solution, which showed high efficiency in vanadium removing at pH from 2.0~9.0 and concentration of V⁵⁺ ranging from 100 to 500 mg·L^-1. Under the optimal conditions, the adsorption rate of vanadium approached to 100% and the concentration of V⁵⁺ decreased from 500 to 0.81 mg·L^-1. The adsorption process belonged to the Langmuir isothermal adsorption and the coefficient was higher than 0.99. Dynamic analysis showed that the pseudo-second-order model could describe the adsorption process very well.

The release of NO, NH₃ and HCN during chrome shavings pyrolysis was investigated using Fourier transform infrared spectrometer. The effect of the temperature was discussed as time increased. The results showed that the release of NO, NH₃ and HCN during pyrolysis was different from each other. The release of nitrogen-product was relative low in low temperature zone with the domination of NH₃. With increasing temperature, the yields of both NO and HCN increased, especially at 550℃ and the release of HCN over 500 ml ·m^-3 was higher than that of NH₃ exceeded, while the release of NH₃ was kept at 10%-15%. In the high temperature zone, the release of HCN increased rapidly, while the release of NO was relative low. The release of NH₃ was related to the destruction of amino in collagen, such as acid amides and amide band. The HCN was mainly formed by the second splitting decomposition of nitrogen-containing compound in the pyrolyzed char, as well as of the nitrogen-containing heterocyclic compound in pyrolyzed liquid oil. The release of NO was associated to the oxygen-containing function groups such as C=O，C-O/C-O-C, and the ratio of N/O contained in precursor.

Compared with conventional aqueous electrolyte, deep eutectic solvent(DES) has been studied as the electrolyte of flow battery for its unique advantages. However, its physical and electrochemical properties are more sensitive to the temperature and related researches can hardly be found in open literature. The aim of the work is to study the physical and electrochemical properties of vanadium ions in DES through varying the operating temperatures. The curves of cyclic voltammetry indicate that vanadium ions show a quasi-reversible reaction. As the temperature increased from ambient temperature to 55℃, the difference between oxidation and reduction peaks decreases from 0.271 V to 0.249 V at the scan rate of 100 mV·s^-1. At the same time, both the anodic and the cathodic peak current densities of the V(Ⅱ)/V(Ⅲ) have the similarly incremental trend. The conductivity also has an obvious increment from 2.2 mS·cm^-1 at ambient temperature to 11.16 mS·cm^-1 at 55℃, but the viscosity of DES vanadium ions have a huge drop. The results prove that the operating temperature plays a vital effect on the characteristics of DES and it deserves a further study.

Introducing the supercritical water oxidation technology into the clean combustion of coal to solve serious problems of low combustion efficiency and pollution of traditional coal combustion, this paper analyzed the reaction characteristics of coal in two different oxidizing atmosphere of air and supercritical water. Furthermore, based on different reaction paths and by means of the energy utilization diagram methodology, the energy conversion characteristics of coal oxidizing in different atmosphere were analyzed, and the new mechanism of energy release of coal oxidizing in aqueous phase was revealed. Finally, the exergy analysis of practical application process of two kinds of oxidation was carried out. The results showed that the aqueous phase oxidation of coal reduced chemistry exergy level from 1 to 0.83, this is to say, the grade difference between chemistry energy and heat source was decreased, indicating that it reduced the irreversible loss of the conversion process from chemical energy into physical energy, exergy losses decreased about 6.04% and thermal exergy increased by 5.25%. The exergy efficiency of coal oxidation in supercritical water was as high as 80.1% about 24.2% higher than oxidation in air, and thus it can greatly improve the exergy efficiency.

The effects of hydrogen addition on explosion characteristics of gas under the condition of obstacles were investigated. The hydrogen fraction in the methane-hydrogen mixture was varied from 0 to 1 at equivalence ratio of 1. The result indicated that the maximum pressure and propagation speed of flame both increased with increasing fraction of hydrogen and obstacle quantity. In addition, the hydrogen addition can also change flame color and flame front. However, the influence of the obstacle quantity on average propagation velocity of flame was slight, and the effects of the obstacle quantity on flame propagation speed of methane as well as hydrogen were different. Increasing trend of the maximum pressure and flame color with the fraction of hydrogen was more and more obvious as the obstacle quantity increased. The influence of the obstacle quantity on explosion characteristics of hydrogen was more obvious than that of methane.