基于稳态ANN模型和模糊PD控制的变速直膨式空调系统温湿度同时控制方法

直膨式空调系统温湿度控制过程高度耦合,造成传统方法下室内空气温湿度的同时精确控制较难实现。本研究基于模糊PD控制逻辑,利用稳态ANN模型建立新型温湿度同时控制算法,根据实时温湿度的控制误差计算所需的显、潜冷量,输出风机、压缩机转速,实现温湿度的同时控制。针对建立的新型控制算法,进行了控制性能验证实验,命令跟随实验结果表明,在新型控制算法的控制下,空气干球与湿球温度设定值改变后在720 s内被稳定在新的设定值,误差在±0.2℃以内;负荷干扰实验结果表明,在有负荷扰动的条件下,控制器在干湿球温度偏离设定值0.5℃后迅速响应,并在600 s内将干湿球温度控制到设定值,波动不超过0.2℃。因此本文建立的新型控制方法可以实现使用变速直膨式系统进行室内空气温湿度同时控制。     

太阳能空调相变蓄冷理论及实验研究

针对一种自制的能够与太阳能空调系统匹配的相变蓄冷材料,建立了蓄冷球蓄\释冷过程数学模型,得到稳态及非稳态工况下蓄冷球内温度分布、蓄\释冷量、蓄\释冷速率的变化规律及影响因素。同时,在相应工况下对单个蓄冷球进行蓄\释冷循环实验,验证理论结果。研究表明,自制蓄冷球能够在170min完成相变。缩小球径、降低冷冻水温度、增大球壁热导率及减小球壁厚度均可缩短蓄冷时间。稳态运行工况下,蓄冷球的蓄\释冷量分别为17.30kJ和16.46kJ;太阳能空调非稳态运行工况下,蓄冷球在165min完成相变,蓄冷量为16.34kJ。     

动态膜制备过程的理论分析及实验验证

基于大量实验结果,对动态膜的制备过程进行了深入分析,考虑到制膜时膜孔堵塞因素,提出了更切合实际的动态膜形成过程数学模型,并对该模型进行了实验验证.以1 250目和6 000目的高岭土为动态膜材料,分别在不同的流量、压力和涂膜液浓度下进行涂膜实验,结果表明:当系统常数k0分别取2和10时,实验值与计算值在非稳态和稳态情况下均比较接近.与计算值不同的是,实验中膜通量进入稳态后仍随时间持续下降.     

多稳态电磁俘能系统的非线性动力学实验研究

为克服共振式俘能器工作频带窄和压电式俘能器输出电流低等问题,设计了多稳态电磁式振动俘能系统。建立了系统的分数阶阻尼模型,通过实验揭示了多稳态电磁俘能系统的动态分岔、势能阱逃逸、高能态轨道和混沌运动等非线性行为。结果表明:采用庞加莱截面频闪采样算法和分岔图可有效刻画系统的非线性振动特性;利用多稳态电磁式俘能系统的非线性振动特性可显著增加输出电流并拓宽系统的有效工作频带。     

空气制冷机非稳态降温特性研究

以逆布雷顿回冷循环为实验流程，对实验工况下的空气制冷机非稳态降温特性进行了实验研究。建立了回冷空气制冷机非稳态数学模型，经与实验结果对比，验证了所建数学模型的合理性及实用性。研究结果为以后小型空气制冷机系统的开发和应用奠定了良好的基础。     

基于自适应粒子群优化的非稳态自动平衡控制算法研究

自动平衡作为新兴的动平衡技术,可实现在线振动抑制。针对非稳态过程提出一种基于自适应粒子群优化(APSO)的自动平衡控制算法,结合BP神经网络,实现一次启停机跟踪后系统的自动平衡控制,在变速模拟试验台上进行了验证。该算法以启停机过程中配重块的调整参数及工作转速作为神经网络的输入,以系统残余振动值作为网络输出,建立输入输出量间的神经网络。将网络培训后输出残余振动预测值作为目标函数,采用粒子群优化对目标函数值最小时的配重块的调整参数进行寻优。仿真及试验结果表明,APSO-BP方法在稳态与非稳态状态下皆可完成系统的自动配平,该控制策略在悬臂试验台上进行测试:变速过程中,不平衡振动幅值在14 s内下降约75%。     

旋转偏心质量块式消振电力作动器建模与控制EI北大核心CSCD

直升机等非固定翼飞行器在飞行状态时由桨叶旋转所产生的周期性低频振动会通过刚性机体传递至驾驶舱、航空发动机以及起落架等部位,会造成机体的持续振动,严重时会影响驾驶员的生命安全。提出了旋转偏心质量块式消振电力作动器,并开展了控制方法研究。从理论上推导了旋转偏心质量块式消振电力作动器的输出力模型以及负载转矩模型;提出基于双电机并行独立控制的电力作动器输出力伺服控制策略,在复频域进行了稳定性分析,并针对正弦波非线性负载扰动带来的转矩脉动问题,对双电机并行独立控制策略展开了优化设计,通过负载前馈控制使系统具备良好的鲁棒性和抗干扰性;研制了重量为14 kg的实验样机并完成了优化控制策略前后电力作动器稳态、动态性能的对比验证实验。结果表明,优化控制策略下的作动器输出力动稳态性能满足各项技术指标要求。     

多稳态压电振动能量采集器的非线性动力学特性及其实验研究

为了提高非线性双稳态压电振动能量采集器的输出性能,提出了一种基于磁-机-压电耦合的非线性多稳态振动能量采集器,通过在双稳态压电振动能量采集器模型基础上增加一对外部磁铁,构造了具有四个稳态的非线性压电振动能量采集器。利用磁偶极子理论建立了采集器悬臂梁末端磁铁与外部磁铁之间的非线性磁力模型;利用Hamilton原理和Raleigh-Ritz方法建立了四稳态压电振动能量采集系统的分布参数机电耦合动力学模型;仿真分析了磁铁水平间距和外部磁铁间距等参数对系统非线性磁力、非线性分岔特性和动力学特性的影响。制作了四稳态压电振动能量采集器原理样机,搭建了样机性能测试平台,实验结果与仿真结果具有较好的吻合度。研究结果表明四稳态压电振动能量采集器可以在低激励水平作用下显著提高能量收集效率,且具有较宽的工作频带。     

双馈变速抽蓄机组参与平抑风电功率波动研究

由于风电功率存在间歇性、波动性,大规模风电并网给电网安全稳定运行带来了挑战。为此,提出了基于双馈变速抽蓄机组的风电功率波动平抑控制策略。首先,基于DIgSILENT/PowerFactory软件,建立了双馈变速抽蓄机组发电、抽水状态下的仿真模型,验证其快速功率控制能力。其次,基于低通滤波原理,结合风电预测功率,建立双馈变速抽蓄机组在发电、抽水2种工况下参与平抑风电功率波动的控制模型。最后,通过仿真算例进行验证,结果表明：双馈变速抽蓄机组具有快速功率响应特性;在发电和抽水2种工况下,均可有效平抑风电功率波动,减小常规机组调节压力,改善地区电网稳定性。     

直膨式太阳能热泵柔性设计

本文引入柔性理论,提出了一种直膨式太阳能热泵系统的柔性设计方案。由于柔性空间内存在一个虚拟工况点,使柔性空间内工况点都能实现稳定的换热效果。实验验证模拟的可靠性,再利用控制环境变量得到影响系统COP的环境因素权重:辐照强度占52.2%,环境温度占34.7%,风速占13.1%。根据权重构建柔性空间,并对落入柔性区间内工况点进行验证。通过对25套系统分别在全年具有代表性天气工况下的运行模拟,得到虚拟工况点最优组合:辐照强度、环境温度、风速分别为559.97 W/m~2、21.6℃、2.89 m/s。模拟结果表明以虚拟工况点最优组合作为依据设计稳定性优于传统静态设计。     

Predicting of mechanical properties of Fe–Mn–(Al, Si) TRIP/TWIP steels using neural network modeling

In this work, an artificial neural network (ANN) model was established in order to predict the mechanical properties of transformation induced plasticity/twinning induced plasticity (TRIP/TWIP) steels. The model developed in this study was consider the contents of Mn (15–30 wt%), Si (2–4 wt%) and Al (2–4 wt%) as inputs, while, the total elongation, yield strength and tensile strength are presented as outputs. The optimal ANN architecture and training algorithm were determined. Comparing the predicted values by ANN with the experimental data indicates that trained neural network model provides accurate results.     

Multivariable control-oriented modeling of a direct expansion (DX) air conditioning (A/C) system

A dynamic mathematical model for a DX A/C system has been developed. The dynamic model, written in state-space representation which was suitable for designing multivariable control, was linearized at steady state operating points. The linearized model has been validated by comparing the model simulation results with the experimental data obtained from an experimental DX A/C system. The simulated results agreed well with the experimental data, suggesting that the model developed was able to capture the transient characteristics of the DX A/C system modeled. It is expected that the model developed can be useful in designing a multi-input multi-output (MIMO) controller to simultaneously control indoor air temperature and humidity in a space served by a DX A/C system.     

An Artificial Neural Network (ANN) Model for Predicting Instability Regimes in Copper-Aluminum Alloys

Materials workability is one of the important aspects for any process design to achieve quality products. Identifying optimum process parameters like temperature, strain rate, and strain are normally done by trial and error. In recent years, processing maps are used in choosing these parameters for hot working of materials. Identification of these parameters requires certain high-level expertise as well as detailed microstructural evidences. In this study, using the available copper-aluminum alloy data, an Artificial Neural Network (ANN) model has been developed to classify the hot-working process parameters, like temperature, strain rate, flow stress for instability regime, directly from the corrected flow stress data without applying the Dynamic Materials Model (DMM). This model uses four compositions of Cu-Al system, ranging from 0.5% to 6% Aluminum. Details about the ANN architecture, and the training and testing of these models are explained. The results obtained using the ANN model are compared and validated with those obtained from the processing maps using DMM. It is further shown that even with smaller data set the development of an ANN model is possible as long as the data has some pattern in it.     

大空间建筑下送风分层空调垂直温度分布研究

本文对某一下送下回的实际大空间建筑建立了求解室内空气垂直温度分布的多节点模型,对该建筑进行了5个不同送风量(1.5×10^4~2.5×10^4 m^3/h)和室外气象参数(28~34℃)的热环境实验。实验结果表明:在各工况下垂直空气温度分布各温度的模型计算值与实测值的最大相对误差分别为6%、7%、14%、-6%和15%,各工况标准方差平均值为2.05℃。同时,本文利用所建多节点模型对大空间建筑下送风分层空调热环境受空调送风量、室外气温、太阳辐射影响的特性进行了分析,绘制了随室内外环境参数变化来调节空调送风量的曲线图。     

Artificial neural networks as alternative tool for minimizing error predictions in manufacturing ultradeformable nanoliposome formulations

This work was aimed at determining the feasibility of artificial neural networks (ANN) by implementing backpropagation algorithms with default settings to generate better predictive models than multiple linear regression (MLR) analysis. The study was hypothesized on timolol-loaded liposomes. As tutorial data for ANN, causal factors were used, which were fed into the computer program. The number of training cycles has been identified in order to optimize the performance of the ANN. The optimization was performed by minimizing the error between the predicted and real response values in the training step. The results showed that training was stopped at 10?000 training cycles with 80% of the pattern values, because at this point the ANN generalizes better. Minimum validation error was achieved at 12 hidden neurons in a single layer. MLR has great prediction ability, with errors between predicted and real values lower than 1% in some of the parameters evaluated. Thus, the performance of this model was compared to that of the MLR using a factorial design. Optimal formulations were identified by minimizing the distance among measured and theoretical parameters, by estimating the prediction errors. Results indicate that the ANN shows much better predictive ability than the MLR model. These findings demonstrate the increased efficiency of the combination of ANN and design of experiments, compared to the conventional MLR modeling techniques.     

Real-time neural inverse optimal control for indoor air temperature and humidity in a direct expansion (DX) air conditioning (A/C) system

A real-time neural inverse optimal control for the simultaneous control of indoor air temperature and humidity using a direct expansion (DX) air conditioning (A/C) system has been developed and the development results are reported in this paper. A recurrent high order neural network (RHONN) was used to identify the plant model of an experimental DX A/C system. Based on this model, a discrete-time inverse optimal control strategy was developed and implemented to an experimental DX A/C system for simultaneously controlling indoor air temperature and humidity. The neural network learning was on-line performed by extended Kalman filtering (EKF). This control scheme was experimentally tested via implementation in real time using an experimental DX A/C system. The obtained results for trajectory tracking illustrated the effectiveness of the proposed control scheme.     

Predicting Interfacial Loads between the Prosthetic Socket and the Residual Limb for Below-Knee Amputees – A Case Study

Abstract:  In this study, an artificial neural network (ANN) was deployed as a tool to determine the internal loads between the residual limb and prosthetic socket for below-knee amputees. This was achieved by using simulated load data to validate the ANN and captured clinical load data to predict the internal loads at the residual limb–socket interface. Load/pressure was applied to 16 regions of the socket, using loading pads in conjunction with a load applicator, and surface strains were collected using 15 strain gauge rosettes. A super-position program was utilised to generate training and testing patterns from the original load/strain data collected. Using this data, a back-propagation ANN, developed at the University of the West of England, was trained. The input to the trained network was the surface strains and the output the internal loads/pressure. The system was validated and the mean square error (MSE) of the system was found to be 8.8% for 1000 training patterns and 8.9% for 50 testing patterns, which was deemed an acceptable error. Finally, the validated system was used to predict pressure-sensitive/-tolerant regions at the limb–socket interface with great success.     

Artificial neural network (ANN) approach for modeling of Pb(II) adsorption from aqueous solution by Antep pistachio (Pistacia Vera L.) shells

A three-layer artificial neural network (ANN) model was developed to predict the efficiency of Pb(II) ions removal from aqueous solution by Antep pistachio (Pistacia Vera L.) shells based on 66 experimental sets obtained in a laboratory batch study. The effect of operational parameters such as adsorbent dosage, initial concentration of Pb(II) ions, initial pH, operating temperature, and contact time were studied to optimise the conditions for maximum removal of Pb(II) ions. On the basis of batch test results, optimal operating conditions were determined to be an initial pH of 5.5, an adsorbent dosage of 1.0 g, an initial Pb(II) concentration of 30 ppm, and a temperature of 30 degrees C. Experimental results showed that a contact time of 45 min was generally sufficient to achieve equilibrium. After backpropagation (BP) training combined with principal component analysis (PCA), the ANN model was able to predict adsorption efficiency with a tangent sigmoid transfer function (tansig) at hidden layer with 11 neurons and a linear transfer function (purelin) at output layer. The Levenberg-Marquardt algorithm (LMA) was found as the best of 11 BP algorithms with a minimum mean squared error (MSE) of 0.000227875. The linear regression between the network outputs and the corresponding targets were proven to be satisfactory with a correlation coefficient of about 0.936 for five model variables used in this study.     

基于神经网络预测ORC系统的最佳蒸发温度

有机朗肯循环系统(ORC)的蒸发温度是决定系统净发电量的关键参数。采用热力学的方法建立数值模型,计算了不同热源温度、冷凝温度及蒸发器夹点温差下的最佳蒸发温度。以此为样本,对神经网络模型进行训练,研究神经网络对ORC系统最佳蒸发温度的预测效果。结果表明,训练速率为0.4、隐层神经元数目为5、训练函数为“trainlm”时,神经网络的预测精度最高。采用两种方式对神经网络的预测结果进行验证,具体为:(1)以9:1比例划分训练集和验证集,(2)系统输入参数取值范围内随机生成100组数据作为验证集。两种验证方式的结果均显示,神经网络对ORC蒸发温度的预测值与数值模拟值较为接近,误差均在2%范围内,表明神经网络模型可以较好的预测ORC最佳蒸发温度,可以为ORC系统的运行参数优化提供参考。