A fusion data preprocessing method and its application in complex industrial power consumption prediction

Data-driven prediction methods often encounter problems in industrial applications due to noise, data redundancy, and insufficient labeled data. Preprocessing the data is required to improve prediction accuracy. At present, most data preprocessing methods are independent of each other and the correlation between methods is relatively small, resulting in repeated redundancy and reduced efficiency. To address this problem, this work proposes a fusion data preprocessing method based on a stacked denoising auto-encoders network (SDAE) and generative adversarial networks (GANs). This method reconstructs the noise reduction link in the SDAE network, improves the noise reduction effect of the model and extracts the main features of the data; the GANs is used to avoid overfitting caused by insufficient training data. Finally, the proposed method is applied to power consumption prediction in the metallurgical industry. Experiments show that the proposed method effectively reduces the model calculation time and improves the overall prediction accuracy.     

分数阶Buck变换器的混沌分析与控制

基于电感与电容的分数阶事实,在Matlab/Simulink下建立了电压控制型Buck变换器的分数阶数学仿真模型及对应的电路仿真模型,分析验证了模型的正确性。基于建立的分数阶数学仿真模型,少有文献研究的以参考电压作为混沌控制变量的Buck变换器混沌行为,绘制了以参考电压为混沌变量的Buck变换器V-I混沌相图与对应的分岔图,得出较小的参考电压易引起Buck变换器进入混沌状态。最后,采用参数扰动法实现了混沌控制,得到了新的分岔图,与混沌控制前的分岔图形成了鲜明的对比。同时,通过选定特定的参考电压值,经分析混沌控制前后电感电流的时域波形图,验证了该控制算法的有效性。     

基于SPCE061A单片机的MGLS12864液晶显示模块的应用设计

通过对点阵式液晶显示模块MGLS12864结构的分析,结合16位单片机SPCE061A的特点,设计了MGLS12864与SPCE061A单片机的硬件接口电路和显示程序流程图。实验证明,系统具有良好的显示效果,达到了设计要求。     

基于细胞免疫的红外图像分割算法及FPGA实现

在工业生产中,高温蒸汽管道通常被用于输送高温蒸汽、高温工业废水等,但为了安全,管道通常被放置在环境比较复杂的地方,不利于工人对管道的检测与维护。快速地定位复杂背景下蒸汽管道的位置并对周围环境进行区分,已经成了一个亟待解决的问题。由于最大类间方差（Otsu）算法不能满足上述要求,基于细胞免疫机制提出了一种改进的Otsu算法,该算法通过红外图像中管道以及复杂背景的特征,能够计算出两个不同的阈值并将其分别用于图像中管道的提取与复杂背景的区分。借助QuartusⅡ软件搭建了基于FPGA的软硬件系统平台,实现了数据通信传输,并对改进的Otsu算法进行验证。实验结果表明,该算法应用在红外管道图像中能取得较好的效果。与几种边缘检测算子和经典Otsu算法相比,无论是在管道的分割,还是复杂背景的区分,本文算法都具有较高的真阳率(True Positive Rate, TPR)和较低的假阳率(False Positive Rate, FPR)。     

基于图模型的小尺寸飞行器地面试验中位姿估计方法

在多飞行器地面试验位姿估计中,由于跟踪算法导致的跟踪轨迹不连续会使得位姿估计产生累积误差,为了实现位姿的精确估计,提出了一种基于图模型的全局位姿估计非线性优化方法。首先,建立了一个飞行器地面视觉位姿估计系统。然后根据飞行器上特征点的数目提出了一种向量交叉式的飞行器位姿解算方法,求解得到数据已关联飞行器位姿估计值。利用中介坐标系法求解得到轨迹段初始位姿节点在测量坐标系下的值,最后,在图模型基础上下,对整个量测过程中飞行器的位姿估计结果进行非线性全局优化减小线性算法的累积误差,并通过仿真与实际实验对飞行器位姿估计算法的可行性与精度进行验证。实验结果表明:在测量范围为6 000 mm×6 000 mm×3 000 mm的范围内,飞行器尺寸约为400 mm,特征点三维定位精度为2.9 mm的条件下,基于非线性优化的飞行器位姿估计算法的理论精度分别可达0.5°(3σ)与3 mm (3σ),实际绝对测量精度分别可达1.3°(3σ)与4 mm (3σ),基本满足地面试验对多飞行器编队算法开发以及制导控制系统性能长时间评估稳定可靠、精度高、抗干扰能力强等要求。     

A stick-slip piezoelectric actuator with an integrated sensing unit for the measurement and active control of the contact force

Output characteristics of the stick-slip piezoelectric actuators are intrinsically affected by the contact force between the driving foot and the mover. Therefore, it would be quite meaningful to measure and control the contact force. In this study, by integrating a group of strain gauges into the driving compliant mechanism, a stick-slip piezoelectric actuator with measured contact force was proposed. The main structural parameters of the driving compliant mechanism were designed and analyzed. Through experiments, the sensing unit was calibrated, and the results showed that it had good linearity, and the noise level was 3.5 mN. The contact force was measured by the sensing unit, and accordingly, the relationship between the contact force and the stepping characteristic was explored. It was shown that the change in contact force would lead to the evolution in stepping characteristic, which finally resulted in the motion nonlinearity of the actuator in large stroke output. Accordingly, by actively controlling the contact force, a method was proposed to tune the stepping characteristic. The comparative results demonstrated that this method could significantly improve the motion linearity of the actuator in large stroke output under various preloads and even very rough surface.     

基于Raspberry PI为结点的天气云测量网络实现

本论文提出了一种天气信息云测量网络设想,基于此设想设计并实现了一种典型的天气云测量结点。根据网络中测量结点需要具有对多种数据综合采集、数据传输和远程控制能力的要求,提出了云端天气测量平台的总体方案,构建了基于3G网络的通信协议,完成了包含有多种天气测量传感器的结点硬件和软件设计并进行了测试试验。试验表明所设计系统满足云端天气测量平台的要求,具有集成便携、鲁棒性强、实时性好、传输能力好的特点。     

An efficient reversible data hiding scheme using prediction and optimal side information selection

In this paper, we present an efficient histogram shifting (HS) based reversible data hiding scheme for copyright protection of multimedia. Firstly, an improved HS based multi-layer embedding process for rhombus prediction is employed by introducing a control parameter to explore the correlation of prediction errors. A rate-distortion model for HS embedding is then developed for optimal side information selection, which is especially suitable for low payload reversible data hiding when only a single layer embedding is required. Finally, a modified location map is constructed to facilitate the compression of location map and further increase the embedding capacity. Compared with similar schemes, experimental results demonstrate the superior performance of the proposed scheme in the terms of embedding capacity and stego-image quality.     

Content-based image retrieval using local visual attention feature

Content-based image retrieval (CBIR) has been an active research topic in the last decade. As one of the promising approaches, salient point based image retrieval has attracted many researchers. However, the related work is usually very time consuming, and some salient points always may not represent the most interesting subset of points for image indexing. Based on fast and performant salient point detector, and the salient point expansion, a novel content-based image retrieval using local visual attention feature is proposed in this paper. Firstly, the salient image points are extracted by using the fast and performant SURF (Speeded-Up Robust Features) detector. Then, the visually significant image points around salient points can be obtained according to the salient point expansion. Finally, the local visual attention feature of visually significant image points, including the weighted color histogram and spatial distribution entropy, are extracted, and the similarity between color images is computed by using the local visual attention feature. Experimental results, including comparisons with the state-of-the-art retrieval systems, demonstrate the effectiveness of our proposal.     

Pilot-aided sampling frequency offset estimation and compensation using DSP technique in DD-OOFDM systems

To improve the outage performance of an optical orthogonal frequency-division multiplexing (OFDM) system under the frequency offset between the sampling clocks in the transmitter and receiver, a pilot-aided sampling frequency offset (SFO) estimation and compensation scheme for the optical OFDM system with intensity-modulation and direct-detection (DD-OOFDM) is experimentally demonstrated. The experimental and simulated results show that the scheme can work effectively even with large sampling frequency offsets. In addition, it can achieve a good bit error rate (BER) performance without the sampling clock frequency synchronization in the receiver.