Energy harvesting has become an increasingly important topic thanks to the advantages in renewability and environmental friendliness. In this paper, a comprehensive study on contemporary portable wind energy harvesters has been conducted. The electrical power generation methods of portable wind energy harvesters are surveyed in three major groups, piezoelectric-, electromagnetic-, and electrostatic-based generators. The paper also takes another view of this area by gauging the required mechanisms for trapping wind flow from ambient environment. In this regard, rotational and aeroelastic mechanisms are analyzed for the portable wind energy harvesting devices. The comparison between both mechanisms shows that the aeroelastic mechanism has promising potential in producing an energy harvester in smaller scale although how to maintain the resonator perpendicular to wind flow for collecting the maximum vibration is still a major challenge to overcome for this mechanism. Furthermore, this paper categorizes the previously published portable wind energy harvesters to macro and micro scales in terms of their physical dimensions. The power management systems are also surveyed to explore the possibility of improving energy conversion efficiency. Finally some insights and research trends are pointed out based on an overall analysis of the previously published works along the historical timeline.

The conventional micro-electromechanical system (MEMS) energy harvesters can only generate voltage disadvantageously in a narrow bandwidth at higher frequencies. In this paper, we propose a piezoelectric MEMS harvester with the capability of vibrating in multi-degrees-of-freedom, whose operational bandwidth is enhanced by taking the advantage of both the multimodal and nonlinear mechanisms. The proposed harvester has a symmetric structure with a doubly-clamped configuration enclosing three proof masses in distinct locations. Thanks to the uniform mass distribution, the energy harvesting efficiency can be considerably enhanced. To determine the optimum geometry for the preferred nonlinear behavior, we propose an automated design and optimization methodology based on the genetic algorithm (GA). By using the micromachining process, our optimized harvester with a total volume of 4.1 mm ³ has been fabricated and measured. The prototype measurements demonstrate that our proposed piezoelectric MEMS harvester is able to generate voltage with a frequency bandwidth of 59 Hz, ranging from 227 to 286 Hz. When the device operates at its second-mode resonant frequency, the nonlinear behavior can be obtained with an extremely small magnitude of the base excitation (i.e., 0.2 m/s ² ). Its normalized power density (NPD) of 595.12 (μW·cm ^-3 ·m ^-2 ·s ⁴ ) is found to be superior to any previously reported wideband aluminum nitride (AlN) piezoelectric MEMS harvesters in the literature.

Low-power electronic applications are normally powered by batteries, which have to deal with stringent lifetime and size constraints. To enhance operational autonomy, energy harvesting from ambient vibration by microelectromechanical systems (MEMS) has been identified as a vivid solution to this universal problem. This paper proposes an automated design and optimization methodology with minimum human efforts for MEMS-based piezoelectric energy harvesters. The analytic equations for estimating the harvested voltage by the unimorph piezoelectric energy harvesters are presented with their accuracy validated by using finite element method (FEM) simulation and prototype measurement. Thanks to their high accuracy, we use these analytic equations as fitness functions of genetic algorithm (GA), an evolutionary computation method for optimization problems by mimicking biological evolution. Our experimental results show that GA is capable of optimizing multiple physical parameters of piezoelectric energy harvesters to considerably enhance the output voltage. This harvesting efficiency improvement is also desirably coupled with physical size reduction as preferred for the MEMS design process. To demonstrate capability of the proposed optimization method, we have also included a commercial optimization product (i.e., COMSOL optimization module) in our comparison study. The experiments show that our proposed GA-based optimization methodology offers higher effectiveness in the magnitude improvement of harvested voltage along with less runtime compared to the other optimization approaches. Furthermore, the effects of geometry optimization on mechanical and electrical properties (e.g., resonant frequency, stiffness, and internal impedance) are also studied and an effective solution to producing maximum power from unimorph piezoelectric harvesters is proposed.

This article presents a novel deep-reinforcement-learning-based method for topology synthesis of analog-integrated circuits, especially operational amplifiers (OpAmps). It behaves like a human designer, who learns from trials, derives design knowledge and experience, and evolves gradually to finally figure out optimal manners to construct proper circuit topologies that meet design specifications. Essential design rules are defined and applied to set up the specialized environment for reinforcement learning in order to reasonably construct circuit topologies with building blocks as the basic components. Our proposed method can not only handle large-size circuit designs but also generate creative circuit topologies. The produced circuit topologies are verified by the simulation-in-loop sizing. In order to improve the evaluation efficiency, hash table and symbolic analysis techniques are utilized to significantly reduce the number of the produced topologies to be sized during the synthesis process. Compared with the state-of-the-art approaches, our proposed method significantly improves the synthesis efficiency by consuming only several hours on average to produce a trustworthy solution. Our experimental results demonstrate its sound efficiency, strong reliability, and wide applicability.

This paper presents an analog integrated circuit automated topology synthesis framework, where circuit topology synthesis can be efficiently realized by encoding circuit topology generation process as tree structure construction. Then the tree structures are decoded into circuit topologies. Our proposed method can not only handle large circuit designs, but also generate creative topologies. To ensure only unique circuit topologies to be generated, two levels of isomorphism checks are performed at both tree structure level and circuit topology level. Then the generated un-sized circuit topologies are efficiently evaluated through a new method, which integrates topological symbolic analysis with gm/ID methodology and curve-fitting technique. Along with small-signal analysis, both linear and nonlinear programming techniques are utilized for topology feasibility checking. With only a small number of circuit topologies through the fast evaluation stage towards the subsequent detailed sizing and further evaluation, the efficiency of the whole circuit synthesis process can be significantly improved. The experimental results demonstrate high efficiency, strong reliability, and wide applicability of our proposed methods.

High operational frequency is one of the major limitations of the conventional MEMS vibration energy harvesters. In this work, we present a piezoelectric MEMS energy harvester with the capability of operating at a low resonant frequency (i.e., less than 200 Hz). The proposed harvester has a symmetric serpentine structure with a doubly clamped configuration comprising several proof masses at the junctions. In order to facilitate the design process and determine the optimum physical dimensions, an artificial neural network is used to model the design. In the first step, a dataset with 108 samples is generated by finite element modeling (FEM) to train a deep neural network. The validation results indicate that the trained deep neural network model can achieve around 90% estimation accuracy of device features, such as resonant frequency and harvested voltage. Next, this trained model is integrated with genetic algorithm as a performance estimator to optimize the geometry of the harvester to lower the resonant frequency and improve the harvested voltage. An optimized harvester with a total area of 8.7 mm ² has been fabricated through a standard micromachining process. Our measurement results confirm that the proposed deep-learning-based method can help reach the balanced summit of both higher power density and lower resonant frequency among the published works. Our prototype device features the first resonant frequency of 121.7 Hz and the harvested power of $0.73~\mu \text{W}$ under 0.1g input acceleration.