Effects of various information scenarios on layer-wise relevance propagation-based interpretable convolutional neural networks for air handling unit fault diagnosis

Deep learning(DL),especially convolutional neural networks(CNNs),has been widely applied in air handling unit(AHU)fault diagnosis(FD).However,its application faces two major challenges.Firstly,the accessibility of operational state variables for AHU systems is limited in practical,and the effectiveness and applicability of existing DL methods for diagnosis require further validation.Secondly,the interpretability performance of DL models under various information scenarios needs further exploration.To address these challenges,this study utilized publicly available ASHRAE RP-1312 AHU fault data and employed CNNs to construct three FD models under three various information scenarios.Furthermore,the layer-wise relevance propagation(LRP)method was used to interpret and explain the effects of these three various information scenarios on the CNN models.An R-threshold was proposed to systematically differentiate diagnostic criteria,which further elucidates the intrinsic reasons behind correct and incorrect decisions made by the models.The results showed that the CNN-based diagnostic models demonstrated good applicability under the three various information scenarios,with an average diagnostic accuracy of 98.55%.The LRP method provided good interpretation and explanation for understanding the decision mechanism of CNN models for the unlimited information scenarios.For the very limited information scenario,since the variables are restricted,although LRP can reveal key variables in the model’s decision-making process,these key variables have certain limitations in terms of data and physical explanations for further improving the model’s interpretation.Finally,an in-depth analysis of model parameters—such as the number of convolutional layers,learning rate,βparameters,and training set size—was conducted to examine their impact on the interpretative results.This study contributes to clarifying the effects of various information scenarios on the diagnostic performance and interpretability of LRP-based CNN models for AHU FD,which helps provide improved reliability of DL models in practical applications.

Daily power demand prediction for buildings at a large scale using a hybrid of physics-based model and generative adversarial network

Power demand prediction for buildings at a large scale is required for power grid operation.The bottom-up prediction method using physics-based models is popular,but has some limitations such as a heavy workload on model creation and long computing time.Top-down methods based on data driven models are fast,but less accurate.Considering the similarity of power demand patterns of single buildings and the superiority of generative adversarial network(GAN),this paper proposes a new method(E-GAN),which combines a physics-based model(EnergyPlus)and a data-driven model(GAN),to predict the daily power demand for buildings at a large scale.The new E-GAN method selects a small number of typical buildings and utilizes EnergyPlus models to predict their power demands.Utilizing the prediction for those typical buildings,the GAN then is adopted to forecast the power demands of a large number of buildings.To verify the proposed method,the E-GAN is used to predict 24-hour power demands for a set of residential buildings.The results show that(1)4.3%of physics-based models in each building category are required to ensure the prediction accuracy;(2)compared with the physics-based model,the E-GAN can predict power demand accurately with only 5%error(measured by mean absolute percentage error,MAPE)while using only approximately 9%of the computing time;and(3)compared with data-driven models(e.g.,support vector regression,extreme learning machine,and polynomial regression model),E-GAN demonstrates at least 60%reduction in prediction error measured by MAPE.

Adaptive Stabilization for a Class of Feedforward Systems with Zero-Dynamics

This paper investigates adaptive state feedback stabilization for a class of feedforward nonlinear systems with zero-dynamics, unknown linear growth rate and control coefficient. For design convenience, the state transformation is first introduced and the new system is obtained. Then, the estimation law is constructed for the unknown control coefficient, and the state feedback controller is proposed with a gain updated on-line. By appropriate choice of the estimation law for the control coefficient and the dynamic gain, the states of the closed-loop system are globally bounded, and the state of the original system converges to zero. Finally, a simulation example is given to illustrate the correctness of the theoretical results.

Development of High Temperature Acoustic Emission Sensing System Using Fiber Bragg Grating

In some applications in structural health monitoring （SHM）, the acoustic emission （AE） detection technology is used in the high temperature environment. In this paper, a high-temperature-resistant AE sensing system is developed based on the fiber Bragg grating （FBG） sensor. A novel high temperature FBG AE sensor is designed With a high signal-to-noise ratio （SNR） compared with the traditional FBG AE sensor. The output responses of the designed sensors with different sensing fiber lengths also are investigated both theoretically and experimentally. Excellent AE detection results are obtained using the proposed FBG AE sensing system over a temperature range from 25℃ to 200℃. The experimental results indicate that this FBG AE sensing system can well meet the application requirement in AE detecting areas at high temperature.

Tracking controller design for air handling units with uncertainties

As an important component of the heating,ventilating and air conditioning(HVAC)systems,air handling units(AHUs)are responsible for regulating indoor temperature and humidity.In this paper,a multivariable nonlinear dynamic model of the AHUs with unknown strength of the humidity source is considered,and an improved backstepping controller is proposed to realize the tracking objective of the indoor temperature,relative humidity and carbon dioxide concentration.Firstly,the original system is represented in simplified state space form,and then the state transformation is introduced with a gain to overcome the difficulty caused by the unknown strength of the humidity source.Then,the improved backstepping controller is designed in a step-by-step way.Moreover,the stability of the closed-loop system is analyzed in detail.Finally,we consider the case that the AHUs work in summer of Jinan,China,as an example.The simulation results show the effectiveness of the controller.Meanwhile,the performance of the improved backstepping controller are compared with that of the decoupled sliding mode and PID controllers.

An FBG impact location system based on broadband light source and improved TDoA algorithm

When a structure material is damaged by impact events, the reliability and lifetime of the material will be severely af- fected. So impact location is considered as the prime approach for structural health and damage monitoring. In this study, a novel fiber Bragg grating （FBG） impact location system based on broadband light source is designed, aiming at the shortcoming of existing location systems based on FBG. An improved localization algorithm based on the time difference of arrival （TDoA） is proposed for impact location. According to this algorithm, the impact position can be accurately predicted without wave velocity. Impact planar location experiments are carried out for verification of the FBG impact location system and algorithm on a 400 mmx400 mmx3 mm aluminum alloy plate. The resulted locating error shows high precision and good stability of the proposed system.