Explainability and Transparency of Classifiers for Air-Handling Unit Faults Using Explainable Artificial Intelligence (XAI)

In recent years, explainable artificial intelligence (XAI) techniques have been developed to improve the explainability, trust and transparency of machine learning models. This work presents a method that explains the outputs of an air-handling unit (AHU) faults classifier using a modified XAI technique, such that non-AI expert end-users who require justification for the diagnosis output can easily understand the reasoning behind the decision. The method operates as follows: First, an XGBoost algorithm is used to detect and classify potential faults in the heating and cooling coil valves, sensors, and the heat recovery of an air-handling unit. Second, an XAI-based SHAP technique is used to provide explanations, with a focus on the end-users, who are HVAC engineers. Then, relevant features are chosen based on user-selected feature sets and features with high attribution scores. Finally, a sliding window system is used to visualize the short history of these relevant features and provide explanations for the diagnosed faults in the observed time period. This study aimed to provide information not only about what occurs at the time of fault appearance, but also about how the fault occurred. Finally, the resulting explanations are evaluated by seven HVAC expert engineers. The proposed approach is validated using real data collected from a shopping mall.

Real-Time Regulation of Beam-Based Feedback: Implementing an FPGA Solution for a Continuous Wave Linear Accelerator

Control applications targeting fast industrial processes rely on real-time feasible implementations. One of such applications is the stabilization of an electron bunch arrival time in the context of a linear accelerator. In the past, only the electric field accelerating the electron bunches was actively controlled in order to implicitly stabilize the accelerated electron beam. Nowadays, beam properties are specifically measured at a target position and then stabilized by a dedicated feedback loop acting on the accelerating structures. This dedicated loop is usually referred to as a beam-based feedback (BBF). Following this, the control system of the electron linear accelerator for beams with high brilliance and low emittance (ELBE) is planned to be upgraded by the BBF, and the problem of implementing a designed control algorithm becomes highly relevant. In this work, we propose a real-time feasible implementation of a high-order H2 regulator based on a field-programmable gate array (FPGA). By presenting simulation and synthesis results made in hardware description language (HDL) VHDL, we show that the proposed digital solution is fast enough to cover the bunch repetition rates frequently used at ELBE, such as 100 kHz. Finally, we verify the implementation by using a dedicated FPGA testbench.

Bridging the Gap in Technology Transfer for Advanced Process Control with Industrial Applications

In the present paper, a software framework comprising the implementation of Model Predictive Control—a popular industrial control method—is presented. The framework is versatile and can be run on a variety of target systems including programmable logic controllers and distributed control system implementations. However, the main attractive property of the framework stems from the goal of achieving smooth technology transfer from the academic setting to real industrial applications. Technology transfer is, in general, difficult to achieve, because of the apparent disconnect between academic studies and actual industry. The proposed software framework aims at bridging this gap for model predictive control—a powerful control technique which can result in substantial performance improvement of industrial control loops, thus adhering to modern trends for reducing energy waste and fulfilling sustainable development goals. In the paper, the proposed solution is motivated and described, and experimental evidence of its successful deployment is provided using a real industrial plant.

An Evolutionary Field Theorem: Evolutionary Field Optimization in Training of Power-Weighted Multiplicative Neurons for Nitrogen Oxides-Sensitive Electronic Nose Applications

Neuroevolutionary machine learning is an emerging topic in the evolutionary computation field and enables practical modeling solutions for data-driven engineering applications. Contributions of this study to the neuroevolutionary machine learning area are twofold: firstly, this study presents an evolutionary field theorem of search agents and suggests an algorithm for Evolutionary Field Optimization with Geometric Strategies (EFO-GS) on the basis of the evolutionary field theorem. The proposed EFO-GS algorithm benefits from a field-adapted differential crossover mechanism, a field-aware metamutation process to improve the evolutionary search quality. Secondly, the multiplicative neuron model is modified to develop Power-Weighted Multiplicative (PWM) neural models. The modified PWM neuron model involves the power-weighted multiplicative units similar to dendritic branches of biological neurons, and this neuron model can better represent polynomial nonlinearity and they can operate in the real-valued neuron mode, complex-valued neuron mode, and the mixed-mode. In this study, the EFO-GS algorithm is used for the training of the PWM neuron models to perform an efficient neuroevolutionary computation. Authors implement the proposed PWM neural processing with the EFO-GS in an electronic nose application to accurately estimate Nitrogen Oxides (NO_x) pollutant concentrations from low-cost multi-sensor array measurements and demonstrate improvements in estimation performance.

Evaluation of Deep Neural Network Compression Methods for Edge Devices Using Weighted Score-Based Ranking Scheme

The demand for object detection capability in edge computing systems has surged. As such, the need for lightweight Convolutional Neural Network (CNN)-based object detection models has become a focal point. Current models are large in memory and deployment in edge devices is demanding. This shows that the models need to be optimized for the hardware without performance degradation. There exist several model compression methods; however, determining the most efficient method is of major concern. Our goal was to rank the performance of these methods using our application as a case study. We aimed to develop a real-time vehicle tracking system for cargo ships. To address this, we developed a weighted score-based ranking scheme that utilizes the model performance metrics. We demonstrated the effectiveness of this method by applying it on the baseline, compressed, and micro-CNN models trained on our dataset. The result showed that quantization is the most efficient compression method for the application, having the highest rank, with an average weighted score of 9.00, followed by binarization, having an average weighted score of 8.07. Our proposed method is extendable and can be used as a framework for the selection of suitable model compression methods for edge devices in different applications.

Incorporation of fractional-order dynamics into an existing PI/PID DC motor control loop

The problem of changing the dynamics of an existing DC motor control system without the need of making internal changes is considered in the paper. In particular, this paper presents a method for incorporating fractional-order dynamics in an existing DC motor control system with internal PI or PID controller, through the addition of an external controller into the system and by tapping its original input and output signals. Experimental results based on the control of a real test plant from MATLAB/Simulink environment are presented, indicating the validity of the proposed approach.