Multi-step Input Research Articles

Low-cost data-driven estimation of indoor occupancy based on carbon dioxide (CO2) concentration: A multi-scenario case study

Occupancy levels significantly influence HVAC system operation, making accurate occupancy prediction essential for the advancement of Occupant-Centered HVAC control. This study aims to develop a simple and effective occupant prediction model in buildings using low-cost indoor environmental sensors and artificial intelligence technology. In-situ measurements were taken in two university classrooms in South Korea over a three-month period, collecting data on indoor and outdoor temperature, humidity, and CO2 levels. Five machine learning algorithms, including Linear Regression (LR), Random Forest (RF), Gradient Boosting Regression (GBR), Multi-Layer Perceptron (MLP), and Long Short-Term Memory neural networks (LSTM), were applied to compare models of indoor occupancy. The results demonstrate that, among the five machine learning models evaluated, the LSTM model outperforms the others, achieving an RMSE of 3.43. This result indicates a close match between predicted and actual indoor occupancy based on CO2 concentration. The integration of a multivariate multi-step input method further enhances its accuracy, making it suitable for a variety of real-world scenarios in indoor occupancy prediction. This study reveals that using processed data as input sources leads to improved prediction performance for indoor occupant states. Importantly, this work does not infringe on biometric information, such as human image privacy, and relies on minimal measurement data. Furthermore, it not only emphasizes the model's feasibility and practicality in predicting indoor occupancy but also its potential in HVAC system automation, building energy conservation, and indoor environmental management. This study offers guidance and support for the advancement of smart cities and intelligent buildings in the future.

A controllable neural network-based method for optimal energy management of fuel cell hybrid electric vehicles

Neural Networks (NNs) can be used for energy management of hybrid vehicles, but they are hard to tune in inference to adapt to different driving conditions. To make the NN-based energy management strategy more flexible, this paper proposes a controllable NN for optimal energy management of fuel cell hybrid electric vehicles. Inspired by the equivalent factor in the Equivalent Consumption Minimization Strategy (ECMS), we introduce an adjustable target variable for the final state as an input to the NN-based strategy. During training, classification and regression networks with single-step and multi-step inputs are considered. An efficient shooting method and an adaptive method are then introduced to realize the precise control of the final state and online parameter adaptation. Simulations of the proposed method and the benchmarking method are carried out in different battery discharge modes. Results demonstrate that the proposed shooting neural classifier can achieve 99.7% fuel optimality of dynamic programming in a similar computational time to the shooting ECMS, and the proposed adaptive neural classifier can adapt to different driving conditions and has better fuel economy than the adaptive ECMS.

Optimal Pneumatic Actuator Positioning and Dynamic Stability using Prescribed Performance Control with Particle Swarm Optimization: A Simulation Study

This paper introduces an optimal control strategy for pneumatic servo systems (PSS) positioning using Finite-time Prescribed Performance Control (FT-PPC) with Particle Swarm Optimization (PSO). Pneumatic servo systems are widely used in industrial automation, as well as medical and cybernetics systems that involve robotics applications. Precision in pneumatic control is crucial not only for the sake of efficiency but also safety. The primary goal of the proposed control strategy is to optimize the convergence rate and finite time of the prescribed performance function in error transformation of the FT-PPC, as well as the Proportional, Integral and Derivative (PID) controller as the inner-loop controller for this system. The study utilizes a dynamic model of a pneumatic proportional valve with a double-acting cylinder (PPVDC) as the targeted plant and performs simulations with a multi-step input trajectory. This offline tuning method is essential for such nonlinear systems to be safely optimized, avoiding major damage to the real-time fine-tuned works on the controller. The results demonstrate that the proposed control strategy surpasses the performance of FT-PPC with a PID controller alone, significantly improving the system's performance, including suppressing overshoot and oscillation in the responses. Further validation through the actual system of PPVDC using the fine-tuned values of FT-PPC and PID with PSO is a future task and more challenging to come, as hardware constraints may vary with different environments such as temperatures.

A short- and long-term prognostic associating with remaining useful life estimation for proton exchange membrane fuel cell

Proton exchange membrane fuel cell (PEMFC), as a promising power source, provides a feasible solution for clean and low-carbon energy systems. The durability problem restricts PEMFC application in some scenarios, which can be improved by the prognostic technology indirectly. This paper aims to develop a data-based method to implement the short-term and long-term prognostic simultaneously, and the developed long-term prognostic can be performed without future operation information. First, the short-term prognostics of five multi-step ahead forecasting strategies are proposed and compared based on a long short-term memory (LSTM) network. Results show that the multi-step input and multi-step output (MIMO) with LSTM strategy has a better performance in the short-term prognostics under the test conditions of the stationary and dynamic current. Then, the hyper-parameters of the prediction model are determined by an evolutionary algorithm. Furthermore, in the long-term prognostics regime, the variable-step long-term method is proposed and rectified by the short-term prognostics. Finally, the developed remaining useful life (RUL) prediction is compared with a model-based extended Kalman filter. The average root mean square error results for the short-term prognostics of two conditions are 0.00532 and 0.00538, respectively. The RUL estimations of two PEMFCs named FC1 and FC2 are given with 95% and 90% confidence intervals, respectively. Consequently, the proposed method can achieve acceptable accuracies in the short-term prognostic, the long-term prognostic, and the RUL prediction.

An occupant-centric air-conditioning system for occupant thermal preference recognition control in personal micro-environment

Thermal comfort is one of the most important factors of indoor environment quality, affecting occupants’ well-being and work efficiency. With the advent of smart control technology, personalized and intelligent air conditioners have been promoted for occupant-centric intelligent air-conditioning control. Based on commonly used air-conditioning (AC), this paper quantitatively describes the method for occupant thermal preference adaptation, and proposes a rule-based classification method of occupant thermal preference recognition. With the quantitative description and classification of the occupant thermal preference, this paper proposes a multi-step input control method for an occupant-centric fan-coil system. This method provides an indoor thermal environment that fulfills the demands of different preferences and is easy to implement with existing air-conditioning control systems without additional sensors. To perform an application-oriented, closed-loop research of the proposed control method, two prediction models of occupant thermal preferences are developed based on an occupant behavior dataset and they could be used as the well initialized models for future online tunning by continually accumulated dataset. Moreover, aiming for a practical operation guide for conventional occupant-centric air-conditioning systems, this paper validates the effectiveness and accuracy of the proposed multi-step input control method, integrated with occupant thermal preference recognition. This was done by using Programmable Logic Controller (PLC) control experiments and Simulink simulations of an actual personal office room, equipped with a fan-coil unit (FCU) in Shanghai. The research results indicate that dynamic indoor air temperature response with different air-conditioning control modes can meet the control needs of different occupant thermal preference patterns.

Multi-Axis Inputs for Identification of a Reconfigurable Fixed-Wing UAV

Designing a reconfiguration system for an aircraft requires a good mathematical model of the object. An accurate model describing the aircraft dynamics can be obtained from system identification. In this case, special maneuvers for parameter estimation must be designed, as the reconfiguration algorithm may require to use flight controls separately, even if they usually work in pairs. The simultaneous multi-axis multi-step input design for reconfigurable fixed-wing aircraft system identification is presented in this paper. D-optimality criterion and genetic algorithm were used to design the flight controls deflections. The aircraft model was excited with those inputs and its outputs were recorded. These data were used to estimate stability and control derivatives by using the maximum likelihood principle. Visual match between registered and identified outputs as well as relative standard deviations were used to validate the outcomes. The system was also excited with simultaneous multisine inputs and its stability and control derivatives were estimated with the same approach as earlier in order to assess the multi-step design.

Spinning gasodynamic projectile system identification experiment design

PurposeThe purpose of this paper is to present the methodology that was used to design a system identification experiment of a generic spinning gasodynamic projectile. For this object, because the high-speed spinning motion, it was not possible to excite the aircraft motion along body axes independently. Moreover, it was not possible to apply simultaneous multi-axes excitations because of the short time in which system identification experiments can be performed (multi-step inputs) or because it is not possible to excite the aircraft with a complex input (multi-sine signals) because of the impulse gasodynamic engines (lateral thrusters) usage.Design/methodology/approachA linear projectile model was used to obtain information about identifiability regions of stability and control derivatives. On this basis various sets of lateral thrusters’ launching sequences, imitating continuous multi-step inputs were used to excite the nonlinear projectile model. Subsequently, the nonlinear model for each excitation set was identified from frequency responses, and the results were assessed. For comparison, the same approach was used for the same projectile exited with aerodynamic controls.FindingsIt was found possible to design launching sequences of lateral thrusters that imitate continuous multi-step input and allow to obtain accurate system identification results in specified frequency range.Practical implicationsThe designed experiment can be used during polygonal shooting to obtain a true projectile aerodynamic model.Originality/valueThe paper proposes a novel approach to gasodynamic projectiles system identification and can be easily applied for similar cases.

Design of a Command-Shaping Scheme for Mitigating Residual Vibrations in Dielectric Elastomer Actuators

Abstract Dielectric elastomers (DEs) are a class of highly deformable electroactive polymers (EAPs) employed for electromechanical transduction technology. When electrostatically actuated dielectric elastomer actuators (DEAs) are subjected to an input signal comprising multiple Heaviside voltage steps, the emerging inherent residual vibrations may limit their motion accuracy in practical applications. In this paper, the systematic development of a command-shaping scheme is proposed for controlling residual vibrations in an electrically driven planar DEA. The proposed scheme relies on invoking the force balance at the point of maximum lateral stretch in an oscillation cycle to bring the actuator to a stagnation state followed by the application of an additional electric input signal of predetermined magnitude at a specific time. The underlying concept of the proposed control scheme is articulated for a single Heaviside step input-driven actuator and further extended to the actuator subjected to the multistep input signal. The equation governing the dynamic motion of the actuator is derived using the principle of virtual work. The devised dynamic model of the actuator incorporates the effects of strain stiffening of elastomer and viscous energy dissipation. The nonlinear dynamic governing equation is solved using matlab ode solver for extracting the dynamic response of the actuator. The applicability of the devised command-shaping control scheme is illustrated by taking a wide range of parameters including variations in the extent of equilibrium state sequences, damping, and polymer chain extensibility. The proposed scheme is found to be adaptable in controlling the vibrations of the actuator for any desired equilibrium state. The results presented in this paper can find its potential application in the design of an open-loop control system for DEAs.

D-Optimal Simultaneous Multistep Excitations for Aircraft Parameter Estimation

In this study, a numerical procedure for designing an aircraft system-identification maneuver with simultaneous flight-control deflections is described. The elevator, aileron, and rudder multistep excitations that maximize the D-optimality criterion were found by a genetic algorithm. Like for any dynamic system, this input-design process would require determining optimal time-dependent functions resulting in infinite-dimensional optimization. To limit the search-space dimensions, the switching points of the inputs were selected on the basis of the discrete wavelet transform. In the investigation, the design was compared with the maneuvers that used harmonically related multisines and typical multistep inputs (3-2-1-1 on elevator, 1-2-1 on aileron, rudder doublet) for simultaneous flight-control deflections. The case of typical excitations applied one at a time was examined as well. Through simulations, it was found that the maneuvers with simultaneous flight-control deflections that were based on the D-optimality criterion or harmonically related multisine signals are better in terms of accuracy, aircraft response, and resulting force than the designs based on the typical excitations.

The effect of the input load current changed to a 1.2kW PEMFC performance

This paper focuses on the performance and dynamic behavior of proton exchange membrane fuel cell (PEMFC) with 1.2 kW under varies of an input load current. The mathematical model of PEMFC is developed based on energy, mass, and electrochemical equation. The some parameters in the model are kept constant. The performance and dynamic behaviors of PEMFC are determined by applied the variety of input load current with different types of input load current, such as saw-tooth, step, step input with dead time, and multi-step input load current. The results of model are compared to the experimental results. The results of the model show that the output cell voltages are suddenly dropped when the input load current have been increased and output cell voltage increased when decreased the input load current. This mathematical model can predict the dynamic behavior of the PEMFC when the input load currents have been changed. The experimental setups use 1.2 kW of PEMFC with connected to electronic load device and can be operated the input load current from 0A to 500A. Hydrogen is used as a fuel at the anode side and air is used at the cathode side. The operating temperature is kept to be constant at room temperature. The experiments show that the output cell voltage is decreased when the input load current increased. There is some slightly different between the model and experiment results when compared at the same conditions. This is because of in the model we kept some parameter constant. The water generated during the process is tested under varying input load current from 1A to 30A. The dynamic behavior of the PEMFC under different types of input load current shows very well.

자동화 비행시험기법에 의한 소형 무인헬리콥터의 파라메터 추정

In this paper dynamic modeling parameters were estimated using a frequency domain estimation method. A systematic flight test method was employed using preprogrammed multistep excitation of the swashplate control input. In addition when one axis is excited, the autopilot is engaged in the other axis, thereby obtaining high-quality flight data. A dynamic model was derived for a small scale unmanned helicopter (CNUHELI-020, developed by Chungnam National University) equipped with a Bell-Hiller stabilizer bar. Six degree of freedom equations of motion were derived using the total forces and moments acting on the small scale helicopter. The dynamics of the main rotor is simplified by the first order tip-path plane, and the aerodynamic effects of fuselage, tail rotor, engine, and horizontal/vertical stabilizer were considered. Trim analysis and linearized model were used as a basic model for the parameter estimation. Doublet and multistep inputs are used to excite dynamic motions of the helicopter. The system and input matrices were estimated in the frequency domain using the equation error method in order to match the data of flight test with those of the dynamic modeling. The dynamic modeling and the flight test show similar time responses, which validates the consequence of analytic modeling and the procedures of parameter estimation.

System identification and stability evaluation of an unmanned aerial vehicle from automated flight tests

This paper presents a consequence of the systematic approach to identify the aerodynamic parameters of an unmanned aerial vehicle (UAV) equipped with the automatic flight control system. A 3-2-1-1 excitation is applied for the longitudinal mode while a multi-step input is applied for lateral/directional excitation. Optimal time step for excitation is sought to provide the broad input bandwidth. A fully automated programmed flight test method provides highquality flight data for system identification using the flight control computer with longitudinal and lateral/directional autopilots, which enable the separation of each motion during the flight test. The accuracy of the longitudinal system identification is improved by an additional use of the closed-loop flight test data. A constrained optimization scheme is applied to estimate the aerodynamic coefficients that best describe the time response of the vehicle. An appropriate weighting function is introduced to balance the flight modes. As a result, concurrent system models are obtained for a wide envelope of both longitudinal and lateral/directional flight maneuvers while maintaining the physical meanings of each parameter.

무인항공기의 시스템 식별을 위한 비행시험기법

In this paper, a flight test method is described for the system identification of the unmanned aerial vehicle equipped with an automatic flight control system. Multistep inputs are applied for both longitudinal mode and lateral/directional excitation. Optimal time step for excitation is sought to provide the broad input bandwidth. A programmed mode flight test method provides high-quality flight data for system identification using the flight control computer with the longitudinal and lateral/directional autopilot which enables the separation of each motion during the flight test. In addition, exact actuating input that is almost equivalent to the designed one guarantees the highest input frequency attainable. Several repetitive flight tests were implemented in the calm air in order to extract the consistent system model for the air vehicle. The enhanced airborne data acquisition system endowed the high-quality flight data for the system identification. The flight data were effectively used to the system identification of the unmanned aerial vehicle.