The conventional statistical evaluations are limited in providing good predictions of the university educational quality. This paper presents an approach with both conventional statistical analysis and neural network modelling/prediction of students’ performance. Conventional statistical evaluations are used to identify the factors that likely affect the students’ performance. The neural network is modelled with 11 input variables, two layers of hidden neurons, and one output layer. Levenberg–Marquardt algorithm is employed as the backpropagation training rule. The performance of neural network model is evaluated through the error performance, regression, error histogram, confusion matrix and area under the receiver operating characteristics curve. Overall, the neural network model has achieved a good prediction accuracy of 84.8%, along with limitations.

We propose a method for detecting and forecasting events of high energy demand, which are managed at the national level in demand side response programmes, such as the UK Triads. The methodology consists of two stages: load forecasting with long short-term memory neural network and dynamic filtering of the potential highest electricity demand peaks by using the exponential moving average. The methodology is validated on real data of a UK building management system case study. We demonstrate successful forecasts of Triad events with RRMSE ≈ 2.2% and MAPE ≈ 1.6% and general applicability of the methodology for demand side response programme management, with reduction of energy consumption and indirect carbon emissions.

Fabric defects in the conventional manufacturing of acoustic panels are detected via manual visual inspections, which are prone to problems due to human errors. Implementing an automated fabric inspection system can improve productivity and increase product quality. In this work, advanced machine learning (ML) techniques for fabric defect detection are reviewed, and two deep learning (DL) models are developed using transfer learning based on pre-trained convolutional neural network (CNN) architectures. The dataset used for this work consists of 1800 images with six different classes, made up of one class of fabric in good condition and five classes of fabric defects. The model design process involves pre-processing of the images, modification of the neural network layers, as well as selection and optimisation of the network’s hyperparameters. The average accuracies of the two CNN models developed in this work, which used the GoogLeNet and the ResNet50 architectures, are 89.84% and 95.45%, respectively, showing statistically significant results. The interpretability of the models is discussed using the Grad-CAM technique. Relevant image acquisition hardware requirements are also put forward for integration with the detection software, which can enable successful deployment of the model for the automated fabric inspection.

Non-raster scanning is being widely used to increase the scanning speed of atomic force microscopes (AFMs). However, like any other AFM scanning techniques, non-raster scanning can also suffer from drift during the scanning process. This paper presents a simple novel method based on cross-correlation for the effective drift correction of non-raster Lissajous AFM scans with intersecting paths. The drift in x, y and z axes can be determined using the crossing points and repeated scans of the same features. In this paper, the general method is presented together with experimental results using Lissajous scans of two artifacts, which demonstrate that the drift in the three axes and the additional tilt can all be corrected resulting in significant improvement in the image obtained. Our drift correction method can overcome the limitations of the existing AFM drift correction methods which are usually based on extra scanning, and it exploits the crossing scan paths and four scanning cycles inherent in Lissajous scanning without using additional scans to effectively determine the three-dimensional drift using cross-correlation and least squares techniques. The drift correction method can be used with any AFMs capable of Lissajous scanning.

Post-occupancy evaluation (POE) is a method for measuring building performance in terms of meeting design intent, and identifying any gaps between actual and modelled performances in the of aspects indoor environmental quality (IEQ), occupant satisfaction, energy consumption, etc. In this paper, a structured POE assessment was conducted using the Building Use Survey (BUS) methodology in two higher educational buildings in the United Kingdom. This study aims to help close the building performance gap, optimize building operation systems, and improve occupants’ satisfaction levels. In this research, the questionnaire survey investigated the influences of environmental factors on user satisfaction from the main aspects of building overall design, thermal comfort, perceived control, indoor environment quality for noise, lighting, ventilation, and other non-environmental factors, such as the background information about age, gender, time in buildings, workgroup size, and so on. The results indicate that the occupant satisfaction with the main aspects of building overall design, indoor environment quality, and thermal comfort in summer and winter in both buildings is lower than the benchmark data. The feedback of this POE assessment has been reported to the building management team to allow managers to develop high-performance building operation guidelines to narrow down the performance gap and improve user satisfaction, comfort, and productivity.