Physical and neural network models of a silica-gel desiccant wheel

Two methods for modelling the performance of a desiccant wheel are presented: a physical model, based on mass and energy balances of the process, and a neural network model, based on the training of a black box model with real data. The physical model consists of a set of non-linear differential equations solved by finite differences techniques. The neural network model consists of a four-input–four-output network that calculates the outlet conditions from inlet ones. Real data are used to validate the physical model and to train the neural network. The physical model shows discrepancies between calculated and measured values mainly due to the fact that the system is assumed to be adiabatic. The heat losses in the ducts and the wheel are not considered in the model, but in the experimental facility these losses occur. In the case of the neural network model, the temperature and humidity ratio calculated for the outlet air are in accordance with the experimental data.     

Prediction of wind speed for the estimation of wind turbine power output from site climatological data using artificial neural network

In this paper, the wind speeds of Noupoort in the Western Cape region of South Africa are forecasted from the site climatological data using feed forward artificial neural network (ANN) with the back propagation training method. Different architectural designs are tested with different combinations of climatological data to obtain the most suitable ANN for predicting the wind speed of the site. The predicted wind speeds are compared with the actual measured wind speeds and the results are evaluated using mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE) and correlation coefficient (R). Some of the key results show that combination of temperature, wind direction and time of the day (TEM?+?WD?+?T) could effectively predict wind speed of Noupoort. The forecasted wind speed shows a strong agreement with the measured wind speed with R, RMSE, MAPE and MAE of 0.96, 0.56, 6.64% and 0.44, respectively.     

A neural network model for the prediction of compression ignition engine performance at different injection timings

Rapid depletion of fossil fuel and continuous increase in gasoline prices have stimulated the search of alternative fuels. This paper deals with the prediction of engine performance, emission and combustion characteristics of compression ignition engine fuelled with fish oil biodiesel using artificial neural network (ANN). Experimental investigations are carried out in a single cylinder constant speed direct injection diesel engine under variable load conditions at different injection timings?21⁰, 24⁰ and 27⁰ bTDC. The performance, combustion and emission characteristics are measured using an exhaust gas analyser, smoke meter, piezoelectric pressure transducer and crank angle encoder for different fuel blends and engine load conditions. For training the neural network, feed-forward back propagation algorithm is used. The developed ANN model predicts the performance, combustions and exhaust emissions with a correlation coefficients (R) of 0.97–0.99 and a mean relative error of 0.62–4.826%. The root mean square errors are found to be low. The developed model has found to predict accurately the engine performance, combustion and emission parameters at different injection timings.     

The effects of operational conditions of the desiccant wheel on the performance of desiccant cooling cycles

A desiccant cooling model is developed and applied to the ventilation, recirculation, makeup, and mix modes of the operating system. The mathematical model is based on the transient coupled heat and mass transfer and is used to predict the performance of the system under various design and operational conditions. The numerical results are validated using experimental measurements. The effects of the regeneration temperature and rotational speed of the desiccant wheel on the COP and output cycle temperature are investigated. The results show the availability of an optimum regeneration temperature and rotational speed in which the output cycle temperature has a minimum value. The optimum regeneration temperature and rotational speed are detected and shown on the Psychrometric charts. Calculating these values has a significant effect on the energy use of these cycles.     

Artificial neural network,ANN-PSO and ANN-ICA for modelling the Stirling engine

Nowadays, shortage of fossil fuels resources, especially oil, and also global attention to environmental hazards produced by the internal combustion process have caused extensive researches on the development of renewable energy engine technology. Among all kinds of renewable resources, solar energy Stirling engines have their own special situation for energy generation with lower pollutants and sustainable sources. The Stirling engine is an external combustion engine that uses any external heat source to generate mechanical power. Various parameters affect the performance of the Stirling engine. In this study, artificial neural network (ANN) was applied to estimate the power and torque values obtained from a Stirling heat engine (Philips M102C engine). It employs the Levenberg–Marquardt algorithm for training ANN with back propagation network for estimating the power and torque of the Stirling heat engine. The performances of the imperialist competitive algorithm (ICA)-ANN and ANN-particle swarm optimisation (PSO) are compared with the performance of the ANN based on mean square error (MSE) and correlation coefficient. PSO and ICAs are applied to determine the initial weights of the neural network. The obtained results indicate that ANN-PSO has a better performance than ICA-ANN and ANN alone; also the MSE for the ANN-PSO is lower as well. Considering the results obtained from this study, there is very good agreement between the output of the testing phase of the ANN-PSO model with experimental data and they are very close to each other.     

遗传BP神经网络在深基坑开挖监测中的应用

地铁车站等深基坑开挖施工中监测数据处理极其复杂 ,其经验多于理论。有鉴于此 ,本文提出了神经网络的处理模型 ,并用将遗传算法和BP最优化方法相结合所产生的一种高效率、高精度的算法来训练网络 ,并比较了各类训练网络方法的优缺点。由于测量数据存在误差 ,所以本文又着重分析了误差在网络中的前向传播与可控制性 ,首次提出了工程应用中网络的初值稳定性问题。最后对某地铁车站的开挖监测进行了实例分析。     

Simulation, performance analysis and optimization of desiccant wheels

Desiccant wheel is an important and crucial component that can be used in building HVAC systems in order to reach relevant energy savings and to use renewable sources. The optimization of air handling units based on desiccant wheels instead of conventional components is complex and it requires adequate simulation tools. In the present paper desiccant wheels performance and optimization are investigated. The analysis is carried out through a one-dimensional gas side resistance model which considers developing temperature and velocity profiles along the channels. Simulation results show a good agreement with experimental data available in literature in a wide range of operating conditions. The model is used to analyze the influence of working conditions on desiccant wheel performance and on the optimal revolution speed. Several performance criteria are introduced and each one of them is used to investigate and to find out the optimal desiccant wheel configuration. For each criterion the best process air angular sector and revolution speed are identified, and the obtained results are compared. Through a practical example it is finally shown how each criterion leads to different optimal configurations.     

Experimental validation of a simplified approach for a desiccant wheel model

This work presents an experimental validation of a simplified approach for a desiccant wheel model, based on the concept of the analogy method and the formulation proposed by Jurinak for the respective combined potentials. The present work experimentally investigates the validity of the assumption of the efficiency factors of the wheel, with regard to the combined potentials, remaining constant over a sufficiently wide range of operation conditions. The results prove the validity of the discussed assumption. The same analysis is implemented with regard to the technical data about the performance of the wheel, usually provided by the manufacturer in the form of software, the results being also positive. Thus, the respective efficiency factors can be calculated through a limited number of measurements, or even simpler, through the use of the manufacturer's performance software accompanying the product.     

Artificial neural network approach to predict the engine performance of fish oil biodiesel with diethyl ether using back propagation algorithm

An artificial neural network (ANN) model is developed to predict the engine performance of fish oil biodiesel blended with diethyl ether. Engine performance and emission characteristics such as brake thermal efficiency, hydrocarbon, exhaust gas temperature, oxides of nitrogen (NO_x), carbon monoxide (CO), smoke and carbon dioxide (CO₂) were considered. Experimental investigations on single-cylinder, constant speed, direct injection diesel engine are carried out under variable load conditions. The performance and emission characteristics are measured using an exhaust gas analyser, smoke metre, piezoelectric pressure transducer and crank angle encoder for different fuel blends and engine load conditions. In this model, a back propagation algorithm is used to predict the performance. Computational results clearly demonstrated that the developed ANN models produced less deviations and exhibited higher predictive accuracy with acceptable determination correlation coefficients of 0.97–1 and mean relative error of 0–3.061% with experimental values. The root mean square errors were found to be low. The developed model produces the idealised results and it has been found to be useful for predicting the engine performance and emission characteristics with limited number of available data.     

Modeling and simulation of desiccant wheel for air conditioning

This paper presents the modeling of a desiccant wheel used for dehumidifying the ventilation air of an air-conditioning system. The simulation of the combined heat and mass transfer processes that occur in a solid desiccant wheel is carried out with MATLAB Simulink. Using the numerical method, the performance of an adiabatic rotary dehumidifier is parametrically studied, and the optimal rotational speed is determined by examining the outlet adsorption-side humidity profiles. The solutions of the simulation at different conditions used in air dehumidifier have been investigated according to the previous published studies. The model is validated through comparison the simulated results with the published actual values of an experimental work. This method is useful to study and modelling of solid desiccant dehumidification and cooling system. The modeling solutions are used to develop simple correlations for the outlet air conditions of humidity and temperature of air through the wheel as a function of the physically measurable input variables. These correlations will be used to simulate the desiccant cooling cycle in an HVAC system in order to define the year round efficiency.     

Desiccant HVAC system driven by a micro-CHP: Experimental analysis

In the Mediterranean area, there is increase in demand for summer cooling satisfied by electrically driven units in domestic and small commercial sectors; this involves electric peak loads and black-outs. Consequently, there is an increasing interest in small scale polygeneration systems fuelled by natural gas.In this paper, attention is paid to a test facility, located in Southern Italy, to carry out experimental analysis on a small scale polygeneration system based on a natural gas-fired Micro-CHP and a desiccant HVAC system. The MCHP provides thermal power, recovered from engine cooling and exhaust gas, for the regeneration of the desiccant wheel and electric power for the chiller, the auxiliaries and the external units (computers, lights, etc.). The HVAC system can also operate in traditional way, by interacting with electric grid and gas-fired boiler. An overview of the main experimental results is shown, considering both the desiccant wheel and the global polygeneration system.The experimental results confirm that the performances of the desiccant wheel are strongly influenced by outdoor thermal-hygrometric air properties and regeneration temperature. The polygeneration system guarantees primary energy savings up to 21.2% and greenhouse-gas emissions reductions up to 38.6% with respect to conventional HVAC systems based on separate energy “production”.     

CAST工艺的BP和RBF人工神经网络仿真模型

将误差反向传播前馈(BP)神经网络模型和径向基函数(RBF)神经网络模型应用到CAST工艺中,并采用多输入、双输出神经网络模拟处理过程中各变量之间的关系和预测出水水质.误差分析结果表明,训练阶段RBF神经网络模型的拟合精度比BP神经网络模型的高,但两者的预测精度相差不大;测试阶段BP神经网络模型和RBF神经网络模型预测出水COD的平均相对误差分别为6.35%、6.80%,预测出水TN的平均相对误差分别为7.19%、5.49%,均在8%以下,这说明两种神经网络模型均可用于模拟CAST污水处理工艺各变量之间的关系和预测出水水质,为污水厂的运行管理提供了理论依据.     

Condenser heat recovery with a PV/T air heating collector to regenerate desiccant for reducing energy use of an air conditioning room

This paper presents an experimental test along with procedures to investigate the validity of a developed simulation model in predicting the dynamic performance of a condenser heat recovery with a photovoltaic/thermal (PV/T) air heating collector to regenerate desiccant for reducing energy use of an air conditioning room under the prevailing meteorological conditions in tropical climates. The system consists of five main parts; namely, living space, desiccant dehumidification and regeneration unit, air conditioning system, PV/T collector, and air mixing unit. The comparisons between the experimental results and the simulated results using the same meteorological data of the experiment show that the prediction results simulated by the model agree satisfactorily with those observed from the experiments. The thermal energy generated by the system can produce warm dry air as high as 53 °C and 23% relative humidity. Additionally, electricity of about 6% of the daily total solar radiation can be obtained from the PV/T collector in the system. Moreover, the use of a hybrid PV/T air heater, incorporated with the heat recovered from the condenser to regenerate the desiccant for dehumidification, can save the energy use of the air conditioning system by approximately 18%.     

Experimental study on dehumidifier and regenerator of liquid desiccant cooling air conditioning system

A new type of air conditioning system, the liquid desiccant evaporation cooling air conditioning system (LDCS) is introduced in this paper. Desiccant evaporation cooling technology is environmental friendly and can be used to condition the indoor environment of buildings. Unlike conventional air conditioning systems, the system can be driven by low-grade heat sources such as solar energy and industrial waste heat with temperatures between 60 and 80 °C. In this paper, a LDCS, as well as a packed tower for the regenerator and dehumidifier is described. The effects of heating source temperature, air temperature and humidity, desiccant solution temperature and desiccant solution concentration on the rates of dehumidification and regeneration are discussed. Based on the experimental results, mass transfer coefficients of the regeneration process were experimentally obtained. The results showed that the mean mass transfer coefficient of the packing regenerator was 4 g/(m² s). In the experiments of dehumidification, it was found that there was maximal tower efficiency with the suitable inlet humidity of the indoor air. The effective curves of heating temperature on the outlet parameters of the regenerator were obtained. The relationships of regeneration mass transfer coefficient as a function of heating temperature and desiccant concentration are introduced.     

Exergetic modeling and experimental performance assessment of a novel desiccant cooling system

New approaches to space conditioning of buildings are required to resolve economic, environmental, and regulatory issues. One of the alternative systems that is brought to agenda is the desiccant cooling systems, which may provide important advantages in solving air conditioning problems. This study deals with the performance analysis and evaluation of a novel desiccant cooling system using exergy analysis method. The system was designed, constructed and tested in Cukurova University, Adana, Turkey and has been successfully operated since 2008. This system consists of a desiccant wheel, heat exchangers, fans, evaporative cooler, electric heater unit and refrigeration unit. The exergy transports between the components and the destructions in each of the components of the desiccant cooling system are determined for the average measured parameters obtained from the experimental results. Exergy efficiencies of the system components are determined in an attempt to assess their individual performances and the potential for improvements is also presented. The exergetic efficiency values for the whole system on the exergetic product/fuel basis are calculated to range from round 32% to 10% at the varying dead (reference) state temperatures of 0-30 °C.     

Forecasting nitrogen dioxide concentration in ambient air using artificial Neural‐networks

Artificial Neural‐networks is employed to predict the nitrogen dioxide concentration during November, 1997 to February, 1998 at three sites each representing industrial, commercial and residential activity respectively in Mumbai. The application of the Multilayer Recurrent network with back‐propagation learning algorithm is reported in the prediction at three sites using the meteorological variables at one site. The generalization ability of the model is confirmed by root mean square error and correlation between observed and predicted concentrations. The evaluation of model results shows that the degree of success in forecasting NO₂ concentration is promising.     

Study on a novel thermally driven air conditioning system with desiccant dehumidification and regenerative evaporative cooling

Existing desiccant cooling systems reduce the temperature of process air either by adopting evaporative coolers or incorporating vapor compression systems. While the former is restricted by inaccurate control, the latter still consumes certain quantity of electric power. To solve this problem, a thermally driven air conditioning system, which combines the technologies of rotary desiccant dehumidification and regenerative evaporative cooling, has been proposed and investigated. In addition to dehumidification, the system is capable of producing chilled water, thereby realizing separate temperature and humidity control without increasing electrical load. To find out the characteristics of produced chilled water and evaluate the feasibility and energy saving potential of this novel system, a mathematical model has been developed. Case studies have been conducted under Air conditioning and Refrigeration Institute (ARI) summer, ARI humid and Shanghai summer conditions. It is found that the system can achieve a thermal COP higher than 1.0 and an electric COP about 8.0. The temperature of chilled water produced by the system is around 14–20 °C. This chilled water can be used with capillary tube mats for radiant cooling. It is suggested that the system can also be designed as a standalone chilled water plant. As a desiccant dehumidification-based chilled water producing technology, this would expand desiccant cooling to a boarder niche application. The effects of chilled water flow rate, air distribution ratio, inlet air conditions and regeneration temperature have been analyzed in detail. Reachable handling regions, which will be helpful to system design and optimization, have been obtained.     

无黏性管涌型土的BP神经网络判别法研究

运用基于Matlab的神经网络工具箱(NNT)建立无黏性管涌型土的预测模型,并分别以试验数据为研究对象,运用BP神经网络模型对该段土样的渗透破坏形式进行预测。预测结果与规范及试验所得结果相同,说明了基于Matlab的BP神经网络运用于无黏性管涌型土预测的可行性和应用价值。     

高速公路沉降预测神经网络法应用研究

提出了改进的BP神经网络模型 ,把它应用到软基高速公路的沉降预测中 ,提出了两种构造神经网络训练样本基本思路 ,并分别进行了计算和对比 ,指出了各自的优、缺点。结果表明改进的BP网络模型比较稳健、收敛快 ,而且根据时间与对应的沉降量形成的样本训练的网络预测出的沉降误差小、精度高。