Efficient exploration of design spaces is highly sought after in engineering applications. A spectrum of tools has been proposed to deal with the computational difficulties associated with such problems. In the context of our case study, these tools can be broadly classified into optimization and supervised learning approaches. Optimization approaches, while successful, are inherently data inefficient, with evolutionary optimization-based methods being a good example. This inefficiency stems from data not being reused from previous design explorations. Alternately, supervised learning-based design paradigms are data efficient. However, the quality of ensuing solutions depends heavily on the quality of data available. Furthermore, it is difficult to incorporate physics models and domain knowledge aspects of design exploration into pure-learning-based methods. In this work, we formulate a reinforcement learning (RL)-based design framework that mitigates disadvantages of both approaches. Our framework simultaneously finds solutions that are more efficient compared with supervised learning approaches while using data more efficiently compared with genetic algorithm (GA)-based optimization approaches. We illustrate our framework on a problem of microfluidic device design for flow sculpting, and our results show that a single generic RL agent is capable of exploring the solution space to achieve multiple design objectives. Additionally, we demonstrate that the RL agent can be used to solve more complex problems using a targeted refinement step. Thus, we address the data efficiency limitation of optimization-based methods and the limited data problem of supervised learning-based methods. The versatility of our framework is illustrated by utilizing it to gain domain insights and to incorporate domain knowledge. We envision such RL frameworks to have an impact on design science.

Conceptual design of spatial compliant mechanisms with distinct input and output ports may be hard because of its complex interconnected topology, and is currently accomplished by computationally intensive automated techniques. This paper proposes a user insightful method for generating conceptual compliant topology solutions. The method builds on recent advances where the compliant mechanism deformation is represented as load flow in its constituent members. The nature of load flow enables functional decomposition of compliant mechanisms into maximally decoupled building blocks namely a Transmitter member and a Constraint member. The proposed design methodology seeks to synthesize spatial compliant designs by systematically combining transmitter-constraint members by first, identifying kinematically feasible transmitter load paths between input(s) and output(s), and then selecting appropriate constraints that enforce the load path. The paper proposes four design steps to generate feasible solutions and four additional guidelines to optimize load paths and constraint orientations. The method is applied with equal ease to three spatial complaint mechanism examples that belong to single-input single-output, multiple-input single output, and single-input multiple-output mechanisms.

High-performance cooling is often necessary for thermal management of high power density systems. However, human intuition and experience may not be adequate to identify optimal thermal management designs as systems increase in size and complexity. This article presents an architecture exploration framework for a class of single-phase cooling systems. This class is specified as architectures with multiple cold plates in series or parallel and a single fluid split and junction. Candidate architectures are represented using labeled rooted tree graphs. Dynamic models are automatically generated from these trees using a graph-based thermal modeling framework. Optimal performance is determined by solving an appropriate fluid flow distribution problem, handling temperature constraints in the presence of exogenous heat loads. Rigorous case studies are performed in simulation, with components subject to heterogeneous heat loads and temperature constraints. Results include optimization of thermal endurance for an enumerated set of 4051 architectures. The framework is also applied to identify cooling system architectures capable of steady-state operation under a given loading.

Customers post online reviews at any time. With the timestamp of online reviews, they can be regarded as a flow of information. With this characteristic, designers can capture the changes in customer feedback to help set up product improvement strategies. Here, we propose an approach for capturing changes in user expectation on product affordances based on the online reviews for two generations of products. First, the approach uses a rule-based natural language processing method to automatically identify and structure product affordances from review text. Then, inspired by the Kano model which classifies preferences of product attributes in five categories, conjoint analysis is used to quantitatively categorize the structured affordances. Finally, changes in user expectation can be found by applying the conjoint analysis on the online reviews posted for two successive generations of products. A case study based on the online reviews of Kindle e-readers downloaded from amazon.com shows that designers can use our proposed approach to evaluate their product improvement strategies for previous products and develop new product improvement strategies for future products.

This paper details a hybrid computational and analytical model to predict the performance of inline pressure compensating drip irrigation emitters. Pressure compensating emitters deliver a constant flow rate over a range of applied pressures to accurately meter water to crops. Flow rate is controlled within the emitter via a fixed resistance tortuous path, and a variable flow resistance composed of a flexible membrane that deflects under changes in pressure, restricting the flow path. A pressure resistance parameter was derived using an experimentally validated computational fluid dynamics (CFD) model to describe the flow behavior in tortuous paths. The bending mechanics of the membrane were modeled analytically and refined by deriving a correction factor using finite element analysis (FEA). A matrix formulation that calculates the force applied by a line or a patch load of any shape on a rectangular membrane, along which there is a prescribed deflection, was derived and was found to be accurate to be 1%. The combined hybrid computational–analytical model reduces the computational time of modeling emitters from hours to less than 30 min, dramatically lowering the time required to iterate and select optimal designs. The model was validated experimentally using three commercially available drip emitters and was accurate to within 12% of the experimental results.

The continued development of electro-hydrostatic actuators (EHAs) in aerospace applications has put forward an increasing demand upon EHA pumps for their high power density. Besides raising the delivery pressure, increasing the rotational speed is another effective way to achieve high power density of the pump, especially when the delivery pressure is limited by the strength of materials. However, high-speed operating conditions can lead to several challenges to the pump design. This paper reviews the current challenges including the cavitation, flow and pressure ripples, tilting motion of rotating group and heat problem, associated with a high-speed rotation. In addition, potential solutions to the challenges are summarized, and their advantages and limitations are analyzed in detail. Finally, future research trends in EHA pumps are suggested. It is hoped that this review can provide a full understanding of the speed limitations for EHA pumps and offer possible solutions to overcome them.

In the last few decades, power-split infinitely variable transmission (IVT) systems have attracted considerable attention as they ensure high driving comfort with high total efficiencies, especially in off-highway vehicles and agricultural machines. In this study, a novel power-split-input-coupled IVT system is developed. The effects of various dynamic parameters such as power flow and Willis transmission ratio on the mechanical efficiency of the systems are investigated. Kinematic analysis of the new system has been carried out. In addition power flow equations are derived as functions of the power that flows through the infinitely variable unit (IVU). The results indicate that the main parameters, which are strictly related to mechanical efficiency are the power and torque flows through the IVU.

Complex engineered systems can carry risk of high failure consequences, and as a result, resilience—the ability to avoid or quickly recover from faults—is desirable. Ideally, resilience should be designed-in as early in the design process as possible so that designers can best leverage the ability to explore the design space. Toward this end, previous work has developed functional modeling languages which represent the functions which must be performed by a system and function-based fault modeling frameworks have been developed to predict the resulting fault propagation behavior of a given functional model. However, little has been done to formally optimize or compare designs based on these predictions, partially because the effects of these models have not been quantified into an objective function to optimize. The work described herein closes this gap by introducing the resilience-informed scenario cost sum (RISCS), a scoring function which integrates with a fault scenario-based simulation, to enable the optimization and evaluation of functional model resilience. The scoring function accomplishes this by quantifying the expected cost of a design's fault response using probability information, and combining this cost with design and operational costs such that it may be parameterized in terms of designer-specified resilient features. The usefulness and limitations of using this approach in a general optimization and concept selection framework are discussed in general, and demonstrated on a monopropellant system design problem. Using RISCS as an objective for optimization, the algorithm selects the set of resilient features which provides the optimal trade-off between design cost and risk. For concept selection, RISCS is used to judge whether resilient concept variants justify their design costs and make direct comparisons between different model structures.

This paper describes a unique design concept that is capable of electronically controlling the flow delivered by an external gear pump (EGP). The principle used for varying the flow relies on the variable timing concept which has been previously demonstrated by the author's research team for EGP's operating at high pressures (HPs) (p > 100 bar). This principle permits to vary the flow within a certain range, without introducing additional sources of power loss. In this paper, the above concept has been applied to formulate a design for a variable delivery EGP for low pressure (LP) applications (p < 30 bar), suitable for direct electric actuation. Specific design principles for the gear and the flow variation mechanisms are introduced to limit the force required by the electric actuation, and for maximizing the flow variation range. Also, the low target pressure allows the variable timing principle to be realized with an asymmetric solution, with only one variable timing element present at one side of the gears. A detailed analysis concerning the relationship between the electrically commanded position of the flow varying element and the theoretical flow delivered by the pump is also presented. This analysis is used to formulate analytical expressions for the instantaneous flow rate and the flow nonuniformity of the pump. The paper details the design principle of the proposed variable flow pump and describes the multi-objective optimization approach used for sizing the gears and flow variation mechanism. The paper also discusses the experimental activity performed on a prototype of the proposed unit, able to achieve a flow variation of 31%.

A material's design revolution is underway with a focus to design the material microstructure and processing paths to achieve certain performance requirements of products. A host of manufacturing processes are involved in producing a product. The processing carried out in each process influences its final properties. To couple the material processing-structure-property-performance (PSPP) spaces, models of specific manufacturing processes must be enhanced and integrated using multiscale modeling techniques (vertical integration) and then the input and output of the various manufacturing processes must be integrated to facilitate the flow of information from one process to another (horizontal integration). Together vertical and horizontal integration allows for the decision-based design exploration of the manufacturing process chain in an inverse manner to realize the end product. In this paper, we present an inverse method to achieve the integrated design exploration of materials, products, and manufacturing processes through the vertical and horizontal integration of models. The method is supported by the concept exploration framework (CEF) to systematically explore design alternatives and generate satisficing design solutions. The efficacy of the method is illustrated for a hot rod rolling (HRR) and cooling process chain problem by exploring the processing paths and microstructure in an inverse manner to produce a rod with specific mechanical properties. The proposed method and the exploration framework are generic and support the integrated decision-based design exploration of a process chain to realize an end product by tailoring material microstructures and processing paths.

Modeling and simulation for additive manufacturing (AM) is commonly used in industry. Nevertheless, a central issue remaining is the integration of different models focusing on different objectives and targeting different levels of details. The objective of this work is to increase the prediction capability of characteristics and performances of additively manufactured parts and to co-design parts and processes. The paper contributes to this field of research by integrating part's performance model and additive technology process model into a single early integrated model. The paper uses the dimensional analysis conceptual modeling (DACM) framework in an AM perspective to generate causal graphs integrating the AM equipment and the part to be printed. DACM offers the possibility of integrating existing knowledge in the model. The framework supported by a computer tool produces a set of governing equations representing the relationships among the influencing variables of the integrated model. The systematic identification of the weaknesses and contradictions in the system and qualitative simulation of the system are some of the potential uses of the model. Ultimately, it is a way to create better designs of machines and parts, to control and qualify the manufacturing process, and to control three-dimensional (3D) printing processes. The DACM framework is tested on two cases of a 3D printer using the fused filament fabrication (FFF) powder bed fusion. The analysis, applied to the global system formed of the 3D printer and the part, illustrates the existence of contradictions. The analysis supports the early redesign of both parts and AM process (equipment) and later optimization of the control parameters.

In a planetary gear train (PGT), the power loss by tooth friction is a function of the potential power developed within the gear train elements rather than that being transmitted through it. In the present work, we focus on the operating conditions of two-degree-of-freedom (two-DOF) PGTs. Any operating condition induces its own internal power flow pattern; this implies that tooth friction loss depends on the mechanism of power loss developed in the gearing that differs from one case to another over the entire range of operating conditions. The approach adopted in this paper stems from a unification of the kinematics and tooth friction losses of PGTs and is based on potential powers and power ratios. The range of applicability of the power relations is investigated and clearly defined, and tooth friction loss formulas obtained by their use are tabulated. A short comparison with formulas currently available in the literature is also made. The simplicity of the proposed method for analyzing two-input or two-output planetary gear trains is helpful in the design, optimization, and control of hybrid transmissions. It assists particularly in choosing correctly the appropriate operating conditions to the involved application.

This study presents the design and validation of on-line pressure-compensating (PC) drip irrigation emitters with a substantially lower minimum compensating inlet pressure (MCIP) than commercially available products. A reduced MCIP, or activation pressure, results in a drip irrigation system that can operate at a reduced pumping pressure, has lower power and energy requirements, requires a lower initial capital cost, and facilitates solar-powered irrigation systems. The technology presented herein can help spread drip irrigation to remote regions and contribute to reducing poverty, particularly in developing countries. The activation pressures of drip emitters at three flow rates were minimized using a genetic algorithm (GA)-based optimization method coupled with a recently published fluid–structure interaction analytical model of on-line PC drip emitter performance. The optimization took into account manufacturing constraints and the need to economically retrofit existing machines to manufacture new emitters. Optimized PC drip emitter designs with flow rates of 3.3, 4.2, and 8.2 lph were validated using precision machined prototype emitters. The activation pressure for all was ≤0.2 bar, which is as low as 16.7% that of commercial products. A limited production run of injection molded 8.2 lph dripper prototypes demonstrated they could be made with conventional manufacturing techniques. These drippers had an activation pressure of 0.15 bar. A cost analysis showed that low MCIP drip emitters can reduce the cost of solar-powered drip irrigation systems by up to 40%.

For decomposition and integration of systems, one needs extensive knowledge of system structure. A design structure matrix (DSM) model provides a simple, compact, and visual representation of dependencies between system elements. By permuting the rows and columns of a DSM using a clustering algorithm, the underlying structure of a system can be revealed. In this paper, we present a new DSM clustering algorithm based upon Markov clustering, that is able to cope with the presence of “bus” elements, returns multilevel clusters, is capable of clustering weighted, directed, and undirected DSMs, and allows the user to control the cluster results by tuning only three input parameters. Comparison with two algorithms from the literature shows that the proposed algorithm provides clusterings of similar quality at the expense of less central processing unit (CPU) time.

This paper presents a method to automatically extract function knowledge from natural language text. The extraction method uses syntactic rules to acquire subject-verb-object (SVO) triplets from parsed text. Then, the functional basis taxonomy, WordNet, and word2vec are utilized to classify the triplets as artifact-function-energy flow knowledge. For evaluation, the function definitions associated with 30 most frequent artifacts compiled in a human-constructed knowledge base, Oregon State University's design repository (DR), were compared to the definitions identified by extraction the method from 4953 Wikipedia pages classified under the category “Machines.” The method found function definitions for 66% of the test artifacts. For those artifacts found, 50% of the function definitions identified were compiled in the DR. In addition, 75% of the most frequent function definitions found by the method were also defined in the DR. The results demonstrate the potential of the current work in enabling automated construction of function knowledge repositories.

Modular design is an effective approach to shorten lead-time and reduce cost for development of complex products and systems (CoPS). Because the physical details of the product are not available at the conceptual design stage, considerations in the downstream product development phases such as manufacturing and assembly cannot be used for partition of modules at the conceptual design stage. Since design solution at the conceptual design stage can be modeled by functions and relationships among these functions such as function flows including information flows, material flows, and energy flows, a novel approach is introduced in this research for function module partition of CoPS through community detection using weighted and directed complex networks (WDCN). First, the function structure is obtained and mapped into a weighted and directed complex network. Based on the similarity between behaviors of communities in WDCN and behaviors of modules in CoPS, a LinkRank-based community detection approach is employed for function module partition through optimization with simulated annealing. The function module partition for the power mechanism in a large tonnage crawler crane is conducted as a case study to demonstrate the effectiveness of the developed approach.

This paper proposes a new Starling resistor architecture to control flow limitation in flexible tubes by introducing a needle valve to restrict inlet flow. The new architecture is able to separately control the activation pressure and the flow rate: The tube geometry determines the activation pressure and the needle valve determines the flow rate. A series of experiments were performed to quantify the needle valve and the tube geometry's effect on flow limitation. The examined factors include the inner diameter, the length, and the wall thickness. A lumped-parameter model was developed to capture the magnitude and trend of the flow limitation, which was able to satisfactorily predict Starling resistor behavior observed in our experiments.

Applying previous solutions to solve new problems is a core aspect of design, and designers routinely use informal analogies to solve a wide variety of design problems. However, when the goal is to consider a large quantity and variety of creative solutions, relying on informal analogy recall may limit the analogy and solution breadth. This paper reports on work to identify the analogy connections that designers make during concept generation such that computational support can be employed to intentionally retrieve analogical solutions from existing systems. A study of the types of similarity that are commonly used to draw design analogies, and whether some types of similarity are used more frequently in compound analogy versus single analogy, was designed and implemented. The experiment consists of a design task and a follow up interview. Ten mechanical engineering graduate students specializing in design participated. Eight different types of analogical similarity are observed, and each type is equally likely to be used to form either single or compound analogies. Notably, the flow behavior was a commonly observed type of abstract similarity that helped designers notice connections across domains, suggesting the value of capturing and retrieving (computationally) flow behavior abstractions for the purpose of relating systems analogically.

A system must continue to meet stakeholder needs throughout its service life to maintain value. Excess that is embedded into components during the design phase can enable in-service system evolution when new or changed requirements are introduced. However, while the concept of excess has been established in the literature, it is not clear how to identify and quantify the set of excesses in a particular design. This paper uses component properties and functional flow information to map and quantify the excess that exists within a system. Understanding the functional flow relationships between components allows for the bottlenecks at component interfaces to be identified. Those flows that do not limit the potential evolvability of a system can be removed from consideration, allowing for critical interface parameters to be highlighted and their capabilities quantified. The method is demonstrated on a consumer heat gun, where quantifying the excess within components allows for a reduced map to be created with irrelevant flows removed. Finally, changes to the system are explored to demonstrate how knowledge of component excess can be used to initially validate a proposed evolution.

The concept of potential power efficiency is introduced as the efficiency of an epicyclic gear train (EGT) measured in any moving reference frame. The conventional efficiency can be computed in a carrier-moving reference frame in which the gear carrier appears relatively fixed. In principle, by attaching the reference frame to an appropriate link, torques can be calculated with respect to each input, output, or (relatively) fixed link in the EGT. Once the power flow direction is obtained from the potential power ratio, the torque ratios are obtained from the potential power efficiencies, the particular expression of the efficiency of the EGT is found in a simple manner. A systematic methodology for the efficiency analysis of one and two degree-of-freedom (DOF) EGTs is described, and 14 ready-to-use efficiency formulas are derived for 2DOF gear pair entities (GPEs). This paper includes also a discussion on the redundancy of the efficiency formulas used for 1DOF GPEs. An incomplete in the efficiency formulas in previous literature, which make them susceptible to wrong application, is brought to light.