Predictive active control of building structures using LQR and artificial intelligence

This study presents a neural network-based model for predicting linear quadratic regulator(LQR)weighting matrices for achieving a target response reduction.Based on the expected weighting matrices,the LQR algorithm is used to determine the various responses of the structure.The responses are determined by numerically analyzing the governing equation of motion using the state-space approach.For training a neural network,four input parameters are considered:the time history of the ground motion,the percentage reduction in lateral displacement,lateral velocity,and lateral acceleration,Output parameters are LQR weighting matrices.To study the effectiveness of an LQR-based neural network(LQRNN),the actual percentage reduction in the responses obtained from using LQRNN is compared with the target percentage reductions.Furthermore,to investigate the efficacy of an active control system using LQRNN,the controlled responses of a system are compared to the corresponding uncontrolled responses.The trained neural network effectively predicts weighting parameters that can provide a percentage reduction in displacement,velocity,and acceleration close to the target percentage reduction.Based on the simulation study,it can be concluded that significant response reductions are observed in the active-controlled system using LQRNN.Moreover,the LQRNN algorithm can replace conventional LQR control with the use of an active control system.

Performance-based semi-active control algorithm for protecting base isolated buildings from near-fault earthquakes

Base isolated structures have been found to be at risk in near-fault regions as a result of long period pulses that may exist in near-source ground motions. Various control strategies, including passive, active and semi-active control systems, have been investigated to overcome this problem. This study focuses on the development of a semi-active control algorithm based on several performance levels anticipated from an isolated building during different levels of ground shaking corresponding to various earthquake hazard levels. The proposed performance-based algorithm is based on a modified version of the well-known semi-active skyhook control algorithm. The proposed control algorithm changes the control gain depending on the level of shaking imposed on the structure. The proposed control system has been evaluated using a series of analyses performed on a base isolated benchmark building subjected to seven pairs of scaled ground motion records. Simulation results show that the newly proposed algorithm is effective in improving the structural and nonstructural performance of the building for selected earthquakes.

Optimal Active Control of Wave-Induced Vibration for Offshore Platforms

An obvious motivation of this paper is to examine the effectiveness of the lateral vibration control of a jacket type offshore platform with an AMD control device, in conjunction with H-2 control algorithm, which is an optimal frequency domain control method based on minimization of H-2 norm of the system transfer function In this study, the offshore platform is modeled numerically by use of the finite element method, instead of a lumped mass model This structural model is later simplified to be single-degree-of-freedom (SDOF) system by extracting the first vibration mode of the structure. The corresponding 'generalized' wave force is determined based on an analytical approximation of the first mode shape function, the physical wave loading being calculated from the linearized Morison equation. This approach facilitates the filter design for the generalized force. Furthermore, the present paper also intends to make numerical comparison between H-2 active control and the corresponding passive control using a TMD with the same device parameters.

Active control of turbulence for drag reduction based on the detection of near-wall streamwise vortices by wall information

The spatial relations between the measurable wall quantities （streamwise shear stress τwx, spanwise shear stress τwz, and pressure fluctuations Pw） and the near-wall streamwise vortices （NWSV） are investigated via direct numerical simulation （DNS） databases of fully developed turbulent channel flow at a low Reynolds number. In the stan- dard turbulent channel flow, the results show that all the wall measurable variables are closely associated with the NWSV. But after applying a stochastic interference, the relation based on τwx breaks down while the correlations based on Pw and τwz are still robust. Hence, two wall flow quantities based on Pw and τwz are proposed to detect the NWSV. As an appli- cation, two new control schemes are developed to suppress the near-wall vortical structures using the actuation of wall blowing/suction and obtain 16 % and 11% drag reduction, respectively.

Design approaches for active,semi-active and passive control systems based on analysis of characteristics of active control force

In this paper, the characteristics of forces in active control systems connected to adjacent levels of a building are analyzed. The following characteristics are observed： （1） active control can provide significantly superior supplemental damping to a building, but causes a small frequency shift; （2） the linear quadratic regulator （LQR）-based control force is composed of an elastic restoring force component and a damping force component, where the damping force is almost identical to the total control force, however, the elastic restoring force is very small; and （3） the active control forces prevent mction most of the time during the entire control process. These three characteristics imply that active control systems connected to adjacent levels of a building behave like passive damping devices with adjustable parameters, namely damping characteristics in an active control, which is the mechanism used by semi-active control devices to reach similar performance as active control systems. Two indices are defined to quantify the damping characteristics of control forces in active control systems. These two indices can also be used to quantify the capacity of semi-active control to achieve the perfonrlance of active control. Based on the above observations, two principles are founded for optimization of parameters of semi-active control devices and passive dampers. The first is that the maximum output force of a semi-active or passive device to be designed is identical to an active device, called ＂design principle＂. The other is the response equivalent principle, which states that the response of a building with semi-active or passive devices is the same as with active devices when the same maximum output force is applied. The design procedure for semi-active control devices and passive dampers is described in detail. Finally, numerical simulations of two benchmark problems is conducted to demonstrate the damping characteristics of active control and investigate the capacity of semi-active control to achieve the same performance as active control.

Controllable wave propagation in a weakly nonlinear monoatomic lattice chain with nonlocal interaction and active control

The one-dimensional monoatomic lattice chain connected by nonlinear springs is investigated, and the asymptotic solution is obtained through the Lindstedt-Poincar′e perturbation method. The dispersion relation is derived with the consideration of both the nonlocal and the active control effects. The numerical results show that the nonlocal effect can effectively enhance the frequency in the middle part of the dispersion curve.When the nonlocal effect is strong enough, zero and negative group velocities will be evoked at different points along the dispersion curve, which will provide different ways of transporting energy including the forward-propagation, localization, and backwardpropagation of wavepackets related to the phase velocity. Both the nonlinear effect and the active control can enhance the frequency, but neither of them is able to produce zero or negative group velocities. Specifically, the active control enhances the frequency of the dispersion curve including the point at which the reduced wave number equals zero, and therefore gives birth to a nonzero cutoff frequency and a band gap in the low frequency range. With a combinational adjustment of all these effects, the wave propagation behaviors can be comprehensively controlled, and energy transferring can be readily manipulated in various ways.

Neural Network-Based Active Control for Offshore Platforms

A new active control scheme, based on neural network, for the suppression of oscillation in multiple-degree-of-freedom (MDOF) offshore platforms, is studied in this paper. With the main advantages of neural network, i.e. the inherent robustness, fault tolerance, and generalized capability of its parallel massive interconnection structure, the active structural control of offshore platforms under random waves is accomplished by use of the BP neural network model. The neural network is trained offline with the data generated from numerical analysis, and it simulates the process of Classical Linear Quadratic Regular Control for the platform under random waves. After the learning phase, the trained network has learned about the nonlinear dynamic behavior of the active control system, and is capable of predicting the active control forces of the next time steps. The results obtained show that the active control is feasible and effective, and it finally overcomes time delay owing to the robustness, fault tolerance, and generalized capability of artificial neural network.

INVESTIGATION OF SIMULATION EXPERIMENT ON ACTIVE CONTROL OF TORSIONAL VIBRATION IN A TURBOGENERATOR SHAFT SYSTEM

According to the theoretical analysis and calculation on the base ofcontinuous mass system, the simulation experimental investigation on active control of torsionalvibration in a turbogenerator shaft system is conducted. A test installation with the excitation ofgenerator motor and multi-stepped shaft system is established to simulate the torsional vibration ofa turbogenerator rotor shaft system, and to examine the active control strategy presented. Someuseful results are reached in the experimental study.

Active control of wall-bounded turbulence for drag reduction with piezoelectric oscillators

An experimental investigation was performed for active control of coherent structure bursting in the near-wall region of the turbulent boundary layer. By means of synchronous and asynchronous vibrations with double piezoelectric vibrators, the influence of periodic vibration of the double piezoelectric vibrators on the mean velocity profile, drag reduction rate, and coherent structure bursting is analyzed at Reo = 2766. The case with 100 V/160 Hz-ASYN is superior to other conditions in the experiment and a relative drag reduction rate of 18.54% is exciting. Asynchronous vibration is more effective than synchronous vibration in drag reduction at the same voltage and frequency. In all controlled cases, coherent structures at large scales are regulated while the small-scale structures are stimulated. The fluctuating velocity increases significantly. A periodic regulating effect on the coherent structure can be seen in the ASYN control conditions at the frequency of 160 Hz.

ACTIVE CONTROL OF FLEXIBLE BEAMS SUBJECTED TO MULTI-DISTURBANCE USING ONE CONTROLLER

The active control of vibration for a beam subjected to multi-disturbances is investigated based on wave propagating suppression. In this control system, there are the same numbers of the sensors, the signal inputting to the controller and the disturbances, but there is only one controller. It is a local control system, the system parameters depend only on the characteristics of the structure bounded by the sensors and the controller, and we need not take into account. the boundary conditions and the properties of structures outside of this held. The system is efficient when a structure vibrates in middle and high frequency regions. Some control design rules are developed from the calculation results.

Active control of flow past an elliptic cylinder using an artificial neural network trained by deep reinforcement learning

The active control of flow past an elliptical cylinder using the deep reinforcement learning(DRL)method is conducted.The axis ratio of the elliptical cylinderΓvaries from 1.2 to 2.0,and four angles of attackα=0°,15°,30°,and 45°are taken into consideration for a fixed Reynolds number Re=100.The mass flow rates of two synthetic jets imposed on different positions of the cylinderθ1andθ2are trained to control the flow.The optimal jet placement that achieves the highest drag reduction is determined for each case.For a low axis ratio ellipse,i.e.,Γ=1.2,the controlled results atα=0°are similar to those for a circular cylinder with control jets applied atθ1=90°andθ2=270°.It is found that either applying the jets asymmetrically or increasing the angle of attack can achieve a higher drag reduction rate,which,however,is accompanied by increased fluctuation.The control jets elongate the vortex shedding,and reduce the pressure drop.Meanwhile,the flow topology is modified at a high angle of attack.For an ellipse with a relatively higher axis ratio,i.e.,Γ1.6,the drag reduction is achieved for all the angles of attack studied.The larger the angle of attack is,the higher the drag reduction ratio is.The increased fluctuation in the drag coefficient under control is encountered,regardless of the position of the control jets.The control jets modify the flow topology by inducing an external vortex near the wall,causing the drag reduction.The results suggest that the DRL can learn an active control strategy for the present configuration.

PASSIVE-ACTIVE CONTROL OF POWER FLOW IN AN ISOLATION SYSTEM MOUNTED ON FLEXIBLE FOUNDATIONS

A general model of flexible isolation systems which involves both the passive and active control factors is established by inserting actuators into an passive isolation system. And the power flow transmission function in such a system as with multi disturbance, multi mounts, passive isolators and actuators is deduced. By means of the numerical simulation method, the influence of actuators on power flow transmission characteristic is studied. And as a conclusion, the passive active synthetic control strategy of power flow is summarized.

Instability of cylinder wake under open-loop active control

Instability of a wake controlled by a streamwise Lorentz force is investigated through a Floquet stability analysis. The streamwise Lorentz force, which is a two-dimensional control input created by an electromagnetic actuator located on the cylinder surface,adjusts the base flow to affect the three-dimensional wake instability and achieve wake stabilization and transition delay. The instability mode at a Reynolds number Re = 300 can be transformed from B to A with N = 1.0, where N is an interaction number representing the strength of the Lorentz force relative to the inertial force in the fluid. The wake flow is Floquet stable when N increases to 1.3. The spanwise perturbation wavelengths are 3.926 D and 0.822 D in the modes A and B, respectively, where D is the cylinder diameter. In addition, the oscillating amplitudes of drag and lift are reduced with the increase in the interaction number. Particle tracing is used to explore the essential physical mechanism for mode transformation. The path lines show that suppression of flow separation hinders the fluid deformation and rotation, leading to the decrease in elliptic and hyperbolic instability regions, which is the material cause of mode transformation.All of the results indicate that wake stabilization and transition delay can be achieved under open-loop active control via the streamwise Lorentz force.

An Active Control Study of the 1Cr18Ni9 Steel Plate Multipoint Excitation

A multipoint excitation experiment system of the 1Cr18Ni9 steel plate-pipe network as the main vibration body was established,which used transient flow produced by the wave exciter to excite the 1Cr18Ni9 steel plate-pipe network to vibrate.The experimental results show that the 1Cr18Ni9 steel plate vibrates in three dimensions.The vibrations of the particles of the steel plate distributing along the pipe＇s axial have phase differences and take place near the cylinder side first.Therefore,it is a multipoint vibration mode.The amplitude of the 1Cr18Ni9 steel plate increases as the system pressure increases and decreases as the wave exciter＇s frequency decreases.The vibration intensity of the particles of the 1Cr18Ni9 steel plate close to the cylinder is the strongest and the amplitude is the highest in Y direction at a given system frequency.The vibration intensity of the particles close to the wave exciter is the strongest and the amplitude is the smallest in Z direction at a given system pressure.

Active control scheme for improving mass resolution of film bulk acoustic resonators

High mass resolution of sensors based on film bulk acoustic resonators （FBARs） is required for the detection of small molecules with the low concentration. An active control scheme is presented to improve the mass resolution of the FBAR sen- sors by adding a feedback voltage onto the driving voltage between two electrodes of the FBAR sensors, The feedback voltage is obtained by giving a constant gain and a constant phase shift to the current on the electrodes of the FBAR sensors. The acoustic energy produced by the feedback voltage partly compensates the acoustic energy loss due to the material damping and the acoustic scattering, and thus improves the quality factor and the mass resolution of the FBAR sensors. An explicit expression relating to the impedance and the frequency for an FBAR sensor with the active control is derived based on the continuum theory by neglecting the influence of the electrodes. Numerical simulations show that the impedance of the FBAR sensor strongly depends on the gain and the phase shift of the feedback voltage, and the mass resolution of the FBAR sensor can greatly be improved when the appropriate gain and the phase shift of the feedback voltage are used. The active control scheme also provides an effective solution to improve the resolution of the quartz crystal microbalance （QCM）.

Nonlinear active control of damaged piezoelectric smart laminated plates and damage detection

Considering mass and stiffness of piezoelectric layers and damage effects of composite layers, nonlinear dynamic equations of damaged piezoelectric smart laminated plates are derived. The derivation is based on the Hamilton＇s principle, the higher- order shear deformation plate theory, von Karman type geometrically nonlinear straindisplacement relations, and the strain energy equivalence theory. A negative velocity feedback control algorithm coupling the direct and converse piezoelectric effects is used to realize the active control and damage detection with a closed control loop. Simply supported rectangular laminated plates with immovable edges are used in numerical computation. Influence of the piezoelectric layers＇ location on the vibration control is in- vestigated. In addition, effects of the degree and location of damage on the sensor output voltage are discussed. A method for damage detection is introduced.

HELICOPTER VIBRATION REDUCTION TECHNIQUE WITH HARMONIC ACTIVE CONTROL

In forward flight, the helicopter performance is seriously affected by the airframe vibration caused by asymmetrical airflow through the rotor disk. so the higher harmonic rotor blade pitch control technique is becoming an active subject in the 1980's, which is one of the active control techniques of helicopter vibration reduction and whose effectivity has been demonstrated by flight testing. In the view of control theory, this paper interprets the basic principles of higher harmonic control(HHC) and proposes mathematical models of the rotor and of airframe vibration, as well as methods for identification of the parameters. A higher harmonic controller is designed and the results of numerical simulation show that the vibration level is reduced effectively.

Parametric vibration stability and active control of nonlinear beams

The vibration stability and the active control of the parametrically excited nonlinear beam structures are studied by using the piezoelectric material. The velocity feedback control algorithm is used to obtain the active damping. The cubic aonlineax equation of motion with damping is established by employing Hamilton＇s principle. The multiple-scale method is used to solve the equation of motion, and the stable region is obtained. The effects of the control gain and the amplitude of the external force on the stable region and the amplitude-frequency curve axe analyzed numerically. From the numerical results, it is seen that, with the increase in the feedback control gain, the axial force, to which the structure can be subjected, is increased, and in a certain scope, the structural active damping ratio is also increased. With the increase in the control gain, the response amplitude decreases gradually, but the required control voltage exists a peak value.

Model reduction and active control of flexible beam using internal balance technique

The internal balance technique is effective for the model reduction in flexible structures, especially the ones with dense frequencies. However, due to the difficulty in extracting the internal balance modal coordinates from the physical sensor readings, research on this topic has been mostly theoretical so far, and little has been done in experiments or engineering applications. This paper studies the internal balance method theoretically as well as experimentally and designs an active controller based on the reduction model. The research works on a digital signal processor （DSP） TMS320F2812- based experiment system with a flexible beam and proposes an approximate approach to access the internal balance modal coordinates. The simulation and test results have shown that the proposed approach is feasible and effective, and the designed controller is successful in restraining the beam vibration.

Effect of high-or low-speed fluctuations on the small-scale bursting events in an active control experiment

Active control of a fully developed turbulence boundary layer(TBL) over a flat plate has been investigated with a statistical view. The piezoelectric(PZT) oscillator is employed to produce periodic input into the inner region of the TBL.A wall probe is fixed upstream of the oscillator to identify the high-or low-speed fluctuations as the detecting signals.Then, the impact of the detecting signals on the small-scale bursting process is investigated based on the data acquired by the traversing probe downstream of the oscillator. The results indicate that the small-scale bursting intensity is restrained more apparently at high-speed detecting fluctuations but less impacted at low-speed detecting fluctuations. Furthermore, the perturbed-scale fluctuations arrange the small-scale bursting process in the near-wall region. The detecting signals have an obvious impact on this arrangement, especially the high-intensity regions of the small-scale bursting events: the vibration enhances the intensity at high-speed detecting signals but weakens it at low-speed detecting signals in these regions, which gives a direct evidence on how detecting signals interfering the small-scale bursting process.