Supercritical solution impregnation (SSI) is a novel technology in which low molecular weight substances are loaded into the polymer matrix and has been recently used in the preparation of control-released drug/polymer complex microparticles, films and fibers.This process utilizes mainly the high diffusion coefficient and low viscosity of supercritical fluid and the swelling of polymer in the supercritical fluid.The advantages of SSI are the easy changes of the polymer matrix structure, which makes possible the application of SC-CO₂ in control-released drug delivery system.In this review, the principle, apparatus and application of SSI in the preparation of control-released drugs are introduced, and the development of this technology is also presented.

A thermodynamic model of ejector was established by introducing refrigerant property.The change in pressure and temperature in the ejector and the entrainment ratio of ejector for several refrigerants with similar normal boiling point were compared, and the difference in injector performance was analyzed according to refrigerant properties.The refrigeration performances of several pairs of refrigerants with similar normal boiling points were also compared.The results indicate that higher specific enthalpy of driving vapor results in higher entrainment ratio of ejector.Among the refrigerants studied, R717 gives the best performance and R152a takes the second place, and their COP difference increases with generation temperature.

Based on the boundary layer theory，a new simple numerical model is proposed by taking into account the effect of variable physical properties.Numerical simulations are carried out to investigate the turbulent convective heat transfer of water in tubes at the supercritical pressure（SCP）.The comparison of the simulation with experiment data shows that the proposed model gives correct results about the effects of pressure，mass flux and wall heat flux on the heat transfer for variable-property flow at the SCP and saves considerable computation time.Two low Reynolds number k-ε turbulence models by Jones and Launder（JL）to predict heat transfer of water at the SCP are evaluated.The JL1 model(c1=1.55)agrees well with the constant-property flow but greatly underestimates the heat transfer coefficient under the SCP condition.The JL2 model(c1=1.45)〖JP〗significantly overestimates the heat transfer coefficient for the constant-property flow，but it exhibits a good prediction in a wide range of parameters under the SCP condition.

A detailed description of the heat transfer and mass transfer during ammonia-water bubble absorption in air-cooled vertical finned tube is presented.The tube is externally finned.The ammonia-water absorbent and ammonia vapor enter the absorber at the bottom and leave at the top.A differential mathematical model for the heat and mass transfer is proposed based on mass and energy balances.The changes of the two-phase flow, such as churn flow, slug flow, and bubble flow, are analyzed.The heat and mass transfer between the vapor phase and the liquid phase in different two-phase flow, and the heat transfer from the two-phase flow to the air outside the tube are considered.Solving the differential equations, local values of some important parameters temperature,molar ratio and their variations along the absorber length are obtained.

Based on the Chebyshev collocation spectral method and Schur decomposition, two direct solvers for radiative discrete ordinates equations for a three-dimensional rectangular furnace with absorbing-emitting medium are developed.For the three-dimensional matrix equation from discretization of the radiative transfer equation, one solver is based on two-dimensional Schur decomposition after the transform of three-dimensional matrix equation to two-dimensional one using tensor product, and the other is based on three-dimensional Schur decomposition, which is developed in the present work.With the same parameters, the numerical experiments indicate that the new solvers can provide good accuracy.Compared with the standard discrete ordinates method, the new solvers are much faster, leading to less CPU time.Specially, the CPU time of the new solver based on the three-dimensional Schur decomposition is just 10%—1% of the standard discrete ordinates method.

The mathematical models of gas-liquid two-phase flow were presented, and the Cross-WLF model was selected as the viscosity model of the melt.The gas penetration process was simulated by using the Level Set /SIMPLEC methods, which could capture the moving interfaces at different times, including the gas-melt interface and the front of the melt.The physical features, such as velocity and temperature at different times were described.The influences of melt temperature, gas delay time and injection pressure on gas penetration time and penetration length were analyzed.The numerical results showed that the Level Set/SIMPLEC methods could precisely trace the two moving interfaces in the gas penetration process and the penetration length was affected significantly by melt temperature, gas delay time and injection pressure.

The rheological behavior of Chinese Shenfu coal ash slag was investigated experimentally by using a high-temperature rheometer at temperatures between 1200℃ and 1360℃.The effects of shear rate and temperature on the rheological behavior of the coal ash slag were studied.A metallurgical microscope was used to study the crystalline phase of the quenched sample at different temperatures.The computer software package FactSage was used to predict the proportions of solids and the compositions of the remaining liquid phase.The liquidus Urbain model was used to describe the viscosities of the homogeneous liquid in coal ash slag in a wide range of temperature.The modified Einstein-Roscoe equation was used to describe the effect of presence of solid suspension on the viscosity of partly crystallized coal ash slag.The results showed that，at its liquidus temperature calculated by FactSage，the coal ash slag behaved as a Newtonian fluid, below its liquidus temperature，however，the rheological behavior of the coal ash slag became non-Newtonian owing to its increased solid-phase content.Meanwhile，at a high crystal content，coal ash slag exhibited shear thinning behavior.The models provided a good description of the experimental data of fully liquid and liquid+solid mixtures.The modified Einstein-Roscoe equation satisfactorily estimated viscosities of the coal ash slag at different crystal contents.

The instability characteristics of swirling flow in the cyclone were studied based on measured real tangential velocity by hot wire anemometry（HWA）.The results showed that the real tangential velocity fluctuated obviously in the center region of the cyclone，however，near the wall of the cyclone，the real tangential velocity had a little fluctuation.The real tangential velocity was analyzed with Fast Fourier Transform(FFT).There was a basic frequency near the center region of the cyclone，which was about 44Hz.And the amplitude in the center region was bigger than that near the wall.The real tangential velocity fluctuation indicated that there existed an instability of swirling flow in the cyclone.This instability resulted in the rotation center of the swirling flow deviating from the geometric center of the cyclone，which produced the quasi-periodic oscillation of swirling flow and gave the real tangential velocity fluctuation.

Rice husk ash (RHA) was used as support to synthesize MnOx/RHA catalyst by incipient wetness impregnation.The influence of preparation conditions on the activity of MnOx/RHA for catalytic oxidation of NO was investigated in a fixed bed reactor.The nature, surface species and morphology of the catalyst were analyzed by XRD, SEM, EDX and XRF.It was found that most of MnOx supported on RHA was micron-size particle in the form of amorphous phase.The preparation conditions except for calcination time had a significant effect on the activity of catalyst.With increasing calcination temperature, both the particles size of MnOx and the fraction of crystalline MnOx increased.When the calcination temperature was raised from 200℃ to 650℃, the particle size of MnO_x was increased from 0.1 μm to 1.0 μm.The surface structure of RHA was determined by ashing temperature.With increasing ashing temperature, the surface structure changed from plate to honeycomb and filamentous.

Decomposition of a fraction of hydrogen peroxide to oxygen during epoxidation of 3-chloropropene to epichlorohydrin results in explosion hazards in the gas phase of the reactor.Explosion datus of 3-chloropropene/oxygen/nitrogen mixtures was measured, and the limiting oxygen concentration (LOC) of the mixture was determined as 10.9%.Furthermore, the effect of the reaction conditions on oxygen concentration in the gas phase was investigated.The results showed that in order to reduce oxygen concentration in the gas phase, catalyst quantity should be reduced as low as possible, and stirring velocity should be increased properly, and H2O2 quantity should be controlled strictly.Moreover, H2O2 should be fed in batches after the liquid phase boils, and the feeding quantity of H2O2 in the initial stage of the reaction should be kept at a low level, which would avoid feeding H2O2 instantaneously or suddenly.

Mixed-mode adsorption chromatography is a new kind of bioseparation technology，which normally combines the electrostatic and hydrophobic interactions.It has a unique property of salt-independent adsorption that can adsorb the target protein efficiently under either high-salt or low-salt condition.In the present work mixed-mode adsorption was used to separate immunoglobulin from porcine plasma.Firstly different methods were compared to precipitate the fibrinogen from porcine plasma，and the precipitation with ammonium sulfate was optimized.Then the adsorption isotherm and kinetics of mixed-mode adsorbent for immunoglobulin were investigated，and the conditions of adsorption and desorption were determined.Finally the mixed-mode adsorption chromatography was used to separate immunoglobulin directly from the supernate of ammonium sulfate precipitation.The yield was about 11.5 mg.8226ml^-1，and the purity of immunoglobulin reached about 77.4% with the image analysis of SDS-PAGE.The process developed in the present work provided a new way for the comprehensive utilization of porcine plasma resources.

Fault diagnosis is an important method to insure the safety and stability of chemical processes.Principle component analysis (PCA), one of the typical diagnostic methods, has been widely used in various chemical fault detections.However, PCA is not good at fault diagnosis of complex chemical processes.Artificial immune system (AIS) is an adaptive system inspired by theoretical immunology and observes immune functions, principles and models.Based on the principles of self/non-self discrimination in the immune system, fault diagnosis by using AIS is feasible.The ability of self-learning and self-adaptation makes AIS able to evolve during the online applications.A hybrid diagnostic system combining PCA and AIS was proposed in this paper for early fault diagnosis of chemical processes.A dynamic chemical simulation model was built with Honeywell’s UniSim platform, and the efficiency of the diagnostic system was validated.

In this paper, a soft sensor technique on the basis of support vector machines (SVM) was proposed to estimate the propylene concentration on the bottom of the distillation column, which took into consideration of the process knowledge and operation characteristics of the propylene distillation column.The advantage of SVM in small samples modeling was taken in the soft sensor technique.Furthermore, an intelligent control system was designed to real-time control the bottom propylene concentration in the distillation.A fuzzy PID was used as the master controller for the propylene concentration.One PID controller was used as the slave controller for the sensitive plate temperature.The two controllers constituted a cascade control system.In order to eliminate disturbance of the feed stock, a feedforward controller was designed in the system.The designed soft sensor and control strategy were used in one selected olefin plant and the results were demonstrated and discussed in this paper.The results were promising and showed that the constructed soft sensor was precise enough to satisfy the industrial on-site requirements.Moreover, the control strategy was efficient to guarantee the bottom propylene concentration under the specified limit.

The concept of process system engineering associates operation and management of business and supply chain with the operation of manufacturing execution system (MES), which requires the adaptive capacity of the constant change of manufacture processes and business processes, on condition that there is a transformation of organizational structure, technological process and product scheme.This paper proposes a configurable MES architecture with three layers (i.e.model layer, cooperation-platform layer and function layer).Firstly, business function is separated from description of factory by the model layer; then, business logic keeps apart from technical support via the platform layer and the overall process collaboration and global optimization can be put into practice by means of the same layer; ultimately, standardized function interface is designed for different products through the function layer.Taking the case of a configurable MES product(SMES)，the paper analyzes the implementation of the configurable architecture so as to adapt to constant change of business process.

Based on summarizing and analyzing the development，characteristics and problems of the existing architecture of manufacturing execution system（MES），a functional architecture of MES was proposed，which was oriented to manufacturing enterprises in China.By taking the production management model as the core，the proposed functional architecture defined the main functions of the production management model and the information flows transferring between these functions.Furthermore，maintenance management，quality management and inventory management，which would affect the production，were provided as the main functional models，and the information exchanges between them and production management were also presented.Additionally，in order to ensure scalability of this functional architecture，other functional models affecting the production were also defined.Finally，an application of the functional architecture to guide the design and establishment of the MES software oriented to the process industry of China was introduced to demonstrate the practicability and completeness.

The scheduling problem of on-line crude oil blending is important to decrease the variation of the blended oil’s properties, support the implementation of advanced control and optimize the refining plan.Because that sequential and continuous variables are interlaced, the scheduling problem of crude oil blending does not have an explicit structure, and its complexity is so high that the conventional methods cannot work real time.For the double-deck structure, an optimization scheme based on order was presented.By the simulation on the practical data of crude oil distillation range, the algorithm based on order could improve the computational efficiency significantly.

The effect of non-isothermal mixing on system energy target in heat integrated water networks was addressed，and the concept of “split temperature” was proposed.For homogeneous mixing，system utility requirement will not decrease.For heterogeneous mixing，it can identify whether system utility requirement would increase，decrease or keep the same，by locating the split temperature and comparing the cumulative heat of the mixing streams above the split temperature and the heat demand in the lower temperature intervals.Based on the ideas above，some rules of non-isothermal mixing are presented，which could be used to estimate whether energy penalty exists or not，and to eliminate the penalty by designing new mixing temperatures.Combining them with improved “separate system”，a systematic method is proposed to design energy efficient water networks.

A mathematical programming model was proposed to optimize a batch water-using system with wastewater reuse and centralized treatment in coordination.In order to achieve an optimal operation in the repeated cycle under multiple cycles, a back-calculation strategy was proposed to dampen fluctuations of residual water and contaminant concentrations in buffer tanks.Optimal water-using schemes could be obtained for different operating cycles.A case study of a batch water-using system with a single contaminant was used to illustrate the effectiveness of the proposed method.It provided insight into the design and operation of the batch water-using system with centralized treatment running from start-up to steady state during multiple operating cycles.

Most of the dynamic optimal control problems in chemical engineering can be written in the form of differential algebraic equations (DAEs).Simultaneous approach is an important method for solving these problems.The discretization strategies often used in this approach are local methods, for instance, orthogonal collocation on finite elements (OCFE), which has many advantages.Pseudo-spectral (PS), as a global method, has its own unique properties.It may offer a rapid convergence rate for the approximation of analytic functions and has high precision and low computational effort with a simple structure.As the representations of local method and global method respectively, OCFE and PS were compared.The collocation points and their different distributions were presented, and the degree of freedom (DOF) of non-linear programming (NLP) after discretization was discussed, as well as the criteria were offered to ensure the DOF of the NLPs.At last, a continuous case and a discontinuous case were studied, and it was concluded that if the problem was smooth enough then the convergence of PS was better than that of OCFE.

Expert system based on expert knowledge is one of the most common technologies in chemical fault diagnosis.Knowledge acquisition is the bottleneck of expert system, so knowledge extraction is the key technology of expert system.In this paper, an extracting knowledge technology based on gene expression programming (GEP) for fault diagnosis of chemical processes was presented.Fuzzy processing of the data was performed with the ambiguity function, and then from the database the anomalies and the reasons of these anomalies were identified by using GEP evolution properties.Therefore the knowledge rules were obtained which could be used in fault diagnosis.Practical case study showed that the technology combined with experts in the field could effectively extract fault diagnosis knowledge, and could be used as a knowledge acquisition tool in expert system for fault diagnosis of chemical processes.

Hydrogen is an important utility in refineries.More and more refineries start retrofitting their hydrogen networks to meet the more strict environmental regulation and product standards.The production and consumption of hydrogen fluctuate with crude oil supply and product demand.Consequently, the traditional deterministic design methods are not enough to deal with such retrofit problems.In this paper, an integrated model that considered design and scheduling together was established.Two-stage stochastic programming based on scenario was proposed to handle these uncertainties.Decision variables were divided into first-stage variables and second-stage variables, to minimize the summary cost of first-stage and second-stage.Example study demonstrated the effectiveness of the method as compared with the traditional deterministic method.

Based on the receding horizon optimized control strategy, an optimal control method of grade transition was presented for the large-scale bi-tubular polypropylene production process.Firstly, with mechanism analysis, the Hammerstein model was used to represent the dynamics of resin quality in the grade transition process.Secondly, by receding optimization of the performance index of grade transition, the controlled variables in all reactors were computed in each sampling time, such as polymerization temperature, ratio of hydrogen to propylene and ratio of ethylene to propylene.Moreover, introducing the two-tie control structure and design of state estimator yielded receding optimized closed-loop control of polypropylene grade transition and predicted the dynamic trajectories of the polymer quality in all bi-tubular reactors.Finally, an example of simulation of multi-grade transition control was used to illustrate the effectiveness of the result in this paper.

Through modification of unconstrained model algorithmic predictive control（PC）algorithm based on the impulse response model of the system，a novel predictive control algorithm of linearly constrained system（LCS-PC）was developed by using linear programming optimization technique.Theoretical analysis showed that LCS-PC possessed the same stability and robustness of PC.Simulation results of an atmospheric distillation column control demonstrated that LCS-PC could not only satisfy the linearly constrained requirements of the column，but also give better control response than PC and linear quadratic optimal state feedback（LQSF）control.Comparison with quadratic programming（QP），LCS-PC had higher computation efficiency，so it could satisfy the real time control requirements of industrial processes much better than QP.

Traditional support vector machines (SVM) is originally designed for binary classification.How to effectively extend it to multi-class classification is worthy to research. This paper compared some common support vector machines for multi-class classification problems, and proposed a multi-class support vector machine based on fuzzy kernel clustering algorithm.The classification accuracy and classification speed of support vector machine depended on the tree structure.This multi-class support vector machine used fuzzy kernel clustering algorithm to generated fuzzy class, and combined with the multi-class SVM based on binary tree classification algorithm for multi-class classification.Experimental results showed that the proposed method was more effective and accurate.

To handle the modeling problem for high-dimension data, the independent component subspace method (ICSM) was proposed.As a new ensemble learning method, ICSM could overcome the main drawback of the random subspace method.It constructed subspaces according to independent components (ICs) contributions on different process variables.As a result, the modeling requirement of the ensemble learning method was satisfied, and its physical meaning was also well presented.Moreover, a new fault detection method named ICSM-PCA was also developed.Firstly, PCA monitoring models were build on different subspaces, then the weighted value of each model was computed based on T² and SPE statistics.Finally，two ensemble statistics could be built for monitoring industrial processes.A case study of the Tennessee-Eastman (TE) process illustrated that the proposed method showed good modeling performance and exhibited satisfactory fault detection ability.

To optimize the global production indices of catalytic reforming process, based on a 17-lumped kinetics model and a catalytic reforming process model and considering various constraints, a multi-objective optimization model was proposed to maximize the aromatics yield and minimize the yield of heavy aromatics.Then an improved multi-objective hybrid genetic algorithm（HNAGA）, was proposed by integrating a genetic algorithm with traditional local optimization algorithms and was then used to solve the model.Finally, the industrial simulation result proved that the hybrid algorithm HNAGA was better than the genetic algorithm in obtaining Pareto optimal solutions.The model and algorithm could effectively improve the accuracy of decision-making in operation optimization of the catalytic reforming unit.

Because static soft sensor modeling can not reflect the dynamic information of industrial processes，which lead to worse estimation precision and robustness.A dynamic soft sensor modeling based on least square vector machine（LS-SVM）and ARMA time series prediction modeling was presented.A static soft sensor model based on LS-SVM was established firstly，and then dynamic correction in the static model was made by using the dynamic estimation of prediction error in ARMA to improve the dynamic response characteristics.Finally，the proposed LSSVM-ARMA was used to predict the concentration of ethane in ethylene distillation.Simulation indicated that this method featured good approximation and good generalization ability as compared with LSSVM，and could be used in soft sensor.

Laser surface texturing (LST) has been used in mechanical seal for nearly 10 years, and the modeling and experiments have been made for determining the dimensions that characterize LST, but many works still remains to be done in choosing suitable texture and optimizing the dimension designs to help retain lubricant and enhances the hydrodynamic effect at the interface in the running fields.Based on the study of collective effect of a laser-textured mechanical seal with a radial distributed micropore face, a theoretical analysis model for a laser-textured mechanical seal with an oblique distributed micropore face was developed, by which the fluid film pressure distribution between the faces was obtained.The variation of fluid film stiffness and opening force with oblique angle, pore column number, and face radial ratio under different operation conditions was studied, and the load carrying capacity and its mechanisms were presented.The results showed that when the oblique angle was between 60°and 90°, both the radial pore number and the circumferential pore number was equal or the number differences were controlled within the range of ±1, excellent sealing performance could be obtained for a laser-textured mechanical seal with an oblique distributed pore face.

The anodic oxidation of 1,4-dichlorobenzene in acetonitrile-aqueous solution was investigated on a platinum electrode.The cyclic voltammetric (CV) profiles showed that the oxidation potential region of 1,4-dichlorobenzene was about 2.0—2.3 V(vs SCE).Moreover, the effect of different scan rates on CV curves implied that oxidation reaction of 1,4-dichlorobenzene was controlled by diffusion process.Additionally, the major intermediates detected by liquid chromatography (LC) and liquid chromatography/mass spectrometry (LC/MS) were parachlorophenol, 1,4-benzoquinone and 2,5-dichloro-p-benzoquinone.The variation of the concentrations of related anions (oxalate, acetate, formate, maleate and chloride ion) during the reaction process was ascertained by ion chromatography (IC).On the basis of these findings, two tentative degradation pathways, hydroxyl radical attacking of the atom of carbon in benzene ring and dechlorination of 1,4-dichlorobenzene were described.

Outer space mutation of violacein-producing bacterium was carried out by the space piggyback of manned spacecraft Shenzhou-7.The original strain for the mutation was engineered by introducing violacein-synthesizing gene cluster into Citrobacter freundii.The plate screening and liquid fermentation experiments showed that the mutation ratio of violacein-producing strain reached to 70%, and the mutation range for violacein productivity was from -61.2% to 7.3％.The violacein concentration produced by the selected efficient strain M_S4 was 2.16 g·L^-1 using 0.45% (vol) of glycerol as the carbon source in shake flask culture for 32 h at 20℃, which was about 143.8% higher than that produced by the original engineered strain.By serial subculture experiment, HPLC, and DNA sequencing, the screened mutant strain showed high genetics stability and obvious difference from the original strain on the component of crude violaceins, but there was no change in the DNA sequence of the violacein biosynthesis gene cluster,indicating that the space piggyback is likely having impacts on the gene regulatory networks of space-mutated strain.

The physical and ecological characteristics of cultivated sludge for simultaneous anaerobic sulfide and nitrate removal were studied.The cultivated sludge showed good settling ability, whose diameter, settling velocity and density were 0.54—3.99 mm, 56.13—171.43 m·h^-1 and 1.08 kg·m^-3, respectively.It was irregular-shaped and was composed of subunits (such as zoogloea and floc) under optical microscope.It was observed that the dominant microorganisms on the surface were bacilli, but those in the interior were diversiform under scanning electron microscope.The results from PCR-DGGE analysis exhibited rich diversity of microbial populations and dominant kinds of microorganisms Proteobacteria.

There are many factors which affect the removal rate of nitrogen oxides and sulfur oxides in the reactor energized by dielectric barrier discharge （DBD），such as oxygen content，relative humidity and SO2 initial concentration in the flue gas.In this paper，SO2 initial concentration，oxygen content and relative humidity in the flue gas were changed to study the reaction mechanism of NO and SO₂ in the dielectric barrier discharge plasma process.The results showed that the initial concentration of SO₂ varying even in a larger range had little effect on the removal rate of NO and SO₂ in the N₂/SO₂/NO system.Increasing the relative humidity in the N₂/NO/SO₂/H₂O system could reduce the concentration of SO2 significantly in the flue gas，and its influence on SO2 is more obvious than that on NO.In the system of N₂/NO/SO₂/O₂，the influence of oxygen content on the removal rate of SO2 was not obvious，but more obvious on the reaction of NO in the DBD process.Adding O₂ could promote the conversion of NO into NO₂ and other nitrogen oxides.On the other hand，the generation of NO was also accelerated under some circumstances.Analysis of the reaction mechanism based on the experiment demonstrated that the SO₂ reaction was mainly related to the active free radicals produced in the DBD process，OH and e^-_aq，while the NO reaction was related to O and N.

Microbes play an important role in bioremediation technologies,the aim of this article was to investigate functional strains for soil bio-remediation.In this study benzo［a］pyrene(BaP),benzo［a］anthracene(BaA),benzo［b］fluoranthene(BbF),benzo［k］fluoranthene(BkF),indeno［1,2,3-cd］pyrene(IPY) were used as carbon and energy sources and 50 strains were isolated by steps including enrichment culture and plate cultivation from the contaminated soil at a coking plant in Beijing.The result of 16S rRNA gene sequencing analysis revealed that these strains were clustered into 15 groups and belonged to genera Sphingomonas，Methylobacterium，Burkholderia，Rhodococcus，Bradyrhizobium，Phyllobacterium，Chryseobacterium，Microbacteriumm，of which Sphingomonas was the dominant species.The highest degradation rates of BaA，BaP，BbF，BkF，IPY were 39.64%，33.52%，38.57%，25.37% and 31.17% respectively when tested with strain 3-6-12.The experimental results will be useful for bioremediation of PAHs-contaminated soil.

A direct-expansion type solar-air source heat pump water heating system was developed.An experimental prototype was investigated under different weather conditions in summer and winter in Nanjing，China.Experimental results showed that the prototype could heat water up to 55℃ efficiently by using different heat sources.When there was a large fluctuation of solar radiation on a sunny day，the collector/evaporator absorbed heat from both solar radiation and air，the prototype operated steadily in the solar-air dual source mode，with a high averaged energy efficiency ratio（EER）of 4.83.Meanwhile，it operated in the air-source mode with an averaged EER of 3.97 in a rainy day.Also，on a cold sunny day in winter，it operated in the solar-source mode with a relatively higher evaporation temperature compared with air-source heat pump，thus avoiding defrosting for the evaporator.

In this paper the effect of ultrasonic energy density on sludge dewatering performance was investigated and the related mechanism was discussed.Experimental results showed that time of ultrasonic action had little influence on sludge dewatering performance at the lower energy density.Sludge dewatering performance was improved when energy density increased from 0.25 kW·L^-1 to 0.5 kW·L^-1.At the ultrasonic energy density of 0.5 kW·L^-1, sludge dewatering performance reached the highest at the time of 10 s with mud cake water content 83.8%, specific resistance filtration (SRF) 0.8×109 s2·g^-1, viscosity 16.8 mPa·s and capillary suction time (CST) 10.7 s.At the higher energy density, sludge floc was destroyed and dispersed which resulted in the increase of filtration resistance and the deterioration of sludge dewatering performance.At the energy density of 4 kW·L^-1, and time of ultrasonic action 120 s, mud cake water content, SRF, viscosity and CST were up to 89.4%, 3.1×10⁹ s²·g^-1, 43.2 mPa·s and 56.2 s, respectively.

The effect of aeration rate on the efficiency of CANON (completely autotrophic nitrogen-removal over nitrite) process was studied to improve total nitrogen removal of ammonium-rich wastewater.With simulated wastewater as influent, two kinds of CANON reactors, which used sponge and modified polyethylene as carriers respectively, were studied in tests.The results showed that total nitrogen removal efficiency and removal load were closely related to aeration rate, the increase of aeration rate would improve total nitrogen removal of CANON reactor when aeration rate was less than the limiting aeration rate.For the CANON reactor which used sponge as carrier, when aeration rate was 35.8 m³·m^-3·h^-1, the total nitrogen removal efficiency reached the maximum value of 79.48%, and total nitrogen removal load was 2.32 kg·m^-3·d^-1.When aeration rate was more than the limiting aeration rate, the effect on total nitrogen removal remained stable, because the structure of sponge protected the Anammox bacteria from inhibition by dissolved oxygen.With sponge as carrier, more biomass could be retained in the CANON reactor, especially for Anammox bacteria, which guaranteed the stable operation of the CANON reactor.For the CANON reactor which used modified polyethylene as carrier, when aeration rate was 6.3 m³·m^-3·h^-1, the total nitrogen removal efficiency reached the maximum value of 77.6%, and totalnitrogen removal load was 1.01 kg·m^-3·d^-1.When aeration rate was more than the limiting aeration rate, the effect on total nitrogen removal deteriorated because Anammox bacteria were inhibited by dissolved oxygen.Less biomass could be retained, so the reactor was less stable.

Considering the specific behavior of indirect measurement of flow rate in the polymer extrusion process，a quasi-static digital filtering method，intermediate between dynamic and static measurements，was proposed.By means of this method a direct measurement of the dynamic variables was transferred into an indirect measurement of static variables which demonstrated relative stability in comparison with dynamic variables.Based on this concept，the influences of stochastic and periodic disturbances in extrusion process on measurement accuracy of flow rate were eliminated effectively by using limit filtering and recursive-average filtering.The optimal filtering parameters，including sampling time T and recursive number N，were determined by experiment.Experimental results showed that the output value of quasi-static digital filtering fitted with the actual value very well.

Velocity and pressure result in the filling pattern of plastic melt in injection molding.The temperature field plays an important role and should be simulated correctly and stably because it changes the viscosity significantly, which interacts with velocity and pressure.The GLS(Galerkin/least-squares) method based on the PG( Petrov-Galerkin) method was employed to solve velocity and pressure field stably when equal order interpolation functions were used for velocity and pressure field.The GGLS(Galerkin gradient least-squares) and SUPG(streamline upwind/Petrov-Galerkin) methods were combined to solve temperature equation in a stabilized way in which the SUPG method was used to prevent the oscillation caused by dominated convection, and the GGLS method was used to avoid pseudo-increase of temperature because of small thermal conduction.Numerical examples showed that the temperature results from the GGLS method were more reasonable than the classic Galerkin method.When the filling stage was simulated, the combined GLS/SUPG method gave a valid means for the stabilized and exact solution of temperature field, and at the same time,melt velocity, pressure and flow front could be exactly solved by using the GLS method.

Phosphorylated sodium alginate (P-SA) membrane was synthesized from phosphorus pentaoxide, triethyl phosphate and phosphoric acid, and then modified by ferrocene as a linking reagent. Chitosan(CS) was modified by acetylferrocene. P-ferrocene-SA/acetylferrocene-CS bipolar membrane was prepared with a paste method, and was named as P-mSA/mCS BPM. The bipolar membrane was characterized by charge density, H⁺ permeability, impedance, I-V curves and SEM. Zinc/nickel layer was plated down to the surface of cation exchange layer to realize zero polar distance electrolysis in the cathode room. The experimental results showed that 2,3,5,6-tetrachlopyridine (2,3,5,6-TCP[BFQ]) was electrocatalytically synthesized at the bipolar membrane by degradation of pentachloropyridine.The current efficiency was 70.1%, yield of 2,3,5,6-TCP was 96% ,when the current density was 30 mA·cm^-2. Compared with the traditional Zn reduction method, the electro-reduction technology saved the zinc powder and eliminated the pollution to the environment.