Background. One of the challenges about smart control valves system is processing and analyzing sensors data to extract useful information. These types of information can be used to detect the deviating behaviors which can be an indication of faults and issues in the system. Outlier detection is a process in which we try to find these deviating behaviors that occur in the system.Objectives. First, perform a literature review to get an insight about the machine learning (ML) and data mining (DM) techniques that can be applied to extract patternfrom time-series data. Next, model the operating behaviors of the control valve system using appropriate machine learning and data mining techniques. Finally,evaluate the proposed behavioral models on real world data.Methods. To have a better understanding of the different ML and DM techniques for extracting patterns from time-series data and fault detection and diagnosis of building systems, literature review is conducted. Later on, an unsupervised learning approach is proposed for modeling the typical operating behaviors and detecting the deviating operating behaviors of the control valve system. Additionally, the proposed method provides supplementary information for domain experts to help them in their analysis.Results. The outcome from modeling and monitoring the operating behaviors ofthe control valve system are analyzed. The evaluation of the results by the domain experts indicates that the method is capable of detecting deviating or unseen operating behaviors of the system. Moreover, the proposed method provides additional useful information to have a better understanding of the obtained results.Conclusions. The main goal in this study was achieved by proposing a method that can model the typical operating behaviors of the control valve system. The generated model can be used to monitor the newly arrived daily measurements and detect the deviating or unseen operating behaviors of the control valve system. Also, it provides supplementary information that can help domain experts to facilitate and reduce the time of analysis.

Characteristics of heating, ventilation, air conditioning and refrigeration (HVAC&R) equipment sounds and people's responses to them were studied in order to develop models to predict annoyance from recordings of the sound. These models are intended to address shortcomings of currently used methods for HVAC&R product sound assessment. Coupled with sound prediction models, the annoyance models will be used to monitor and guide improvements to HVAC&R equipment sound quality throughout the product design process: from virtual early design, through to the prototyping and product refinement stages. Responses to residential and refrigerated truck product noise was studied; both produce broadband random noise and families of harmonics related to rotating and reciprocating components within the system. Tests were conducted to determine how people describe HVAC&R equipment sounds; how their descriptions relate to sound characteristics and overall assessments; and to develop models that relate predicted strengths of sound characteristics to the overall assessment. Annoyance models were developed for each types of product. Loudness and spectral balance metrics are included in models for both types of product. Inclusion of a tonalness metric improved models for residential units, and roughness and impulsiveness metrics improved models for refrigerated truck units. The models developed were used to predict responses in the other tests and there was good agreement between predicted and measured responses. An illustration of the use of the annoyance models, in conjunction with sound visualization and signal modification techniques, to guide improvements to product sound quality is given.

<p>Most air conditioning and refrigeration systems that employ the vapor compression cycle rely on oil circulating with refrigerant to lubricate the bearings and other contact surfaces in the compressor. The lubricant acts as a sealant to reduce leakage losses during the compression process and it also helps to absorb some of the excess heat that is generated in the compression chamber. However, this oil circulation results in oil retention in various other components outside the compressor depending on the physical interaction between lubricant and refrigerant and their transport properties. Other factors, such as the geometry and orientation of connecting lines, and the system operating conditions, such as refrigerant flow rate and oil circulation ratio, also impact the oil retention. Because of oil retention, the oil level in the compressor reduces, which may ultimately affect its efficiency and life span. In addition, the effectiveness of heat exchangers (evaporators and condensers) decreases. The current line sizing rules reported in the ASHRAE Handbook on Refrigeration have only limited consideration of the effects of oil in the system. With the increasing development of variable-speed systems as well as future use of newer HFO refrigerants, there is a need in the industry for upgrading the line sizing recommendations, especially the connecting gas lines of unitary split systems, which consider the effects of oil retention. To develop these rules, measuring oil retention at different operating conditions is important. A test setup has been built to measure oil retention in horizontal and vertical lines of different diameters at different refrigerant and oil flow conditions. Based on the collected data, a physics-based semi-empirical model is developed which can predict oil retention in gas lines for some of the commonly used refrigerant-lubricant combinations in the HVAC&R industry.</p><p>Oil Circulation Ratio (OCR) is one of the input parameters to the model which predicts oil retention. A non-invasive, in-situ method to measure OCR in real time, which involves minimal human intervention, is developed. This method is based on oil separation and is implemented on the suction line. The approach has been validated with two different methods, one of which is an ASHRAE standard. The results of the study offer clear evidence that the method is as accurate as a standard method and it involves less human intervention as the measurement process is automated.</p>

<p>In the heating, ventilation, air conditioning and refrigeration (HVAC&R) industry, bonding and joining play an important role in the manufacturing and assembly process, which is critical to the cost, safety, reliability, and design freedom of systems. The goal of this thesis is to understand and evaluate the usage of adhesive bonds in the manufacture of HVAC&R systems, specifically in regards to leakage/reliability characterization and stress analysis under loading.</p> <p>The bonding performance under static loading is first studied using a commercial epoxy adhesive product. In addition to the traditional surface preparation methods of mechanical and chemical etching, a novel laser-interference surface structuring preparation technique was utilized to improve bonding performance. Laser interference structuring uses a ND:YAG laser beam that is split into two beams that are re-directed to overlap on the same area of a copper alloy. A structuring pattern near the interference structuring limits is achieved due to the phase shift between the beams that is imparted as they are re-directed. Two different laser structuring methods were tested: spot-by-spot and laser raster. Different structuring parameters were varied including the laser spot size and pulses per spot (2, 4, 6, 8, 10, 12 pulses/spot) for the spot-by-spot method, and raster speed (2, 4, 6, 8, 10, 12 mm/s) for laser raster method. The microstructure morphology and surface profile after processing were characterized using the scanning electron microscopy (SEM) and profilometry for all surfaces. It was found that the laser-interference structuring removed the surface contaminants efficiently and formed dot- or net-shaped structures on the surface. This indicates that melting, vaporization, and solidification of the metal happened differently. Due to the much higher speed of the laser raster method, considering practical industrial applications, it is selected for additional investigation for shear strength improvement. The shear strength is measured by a single lap shear test which pulls apart adhesively bonded single lap joint specimen under shear loading using a mechanical tester.</p> <p>Based on the surface profiles, three different laser raster speeds of 2 mm/s, 6 mm/s and 12 mm/s were selected for the manufacture of single lap joint specimens for comparison with the traditional surface preparation methods. The shear lap strength and displacement at maximum load were obtained for the specimens. The laser raster at 6 mm/s increased these values by approximately 11.0% and 25.1%, respectively, while the 12 mm/s condition had an increase of 16.8% and 43.8%, compared with the baseline traditional surface preparation method. It is concluded that laser structuring can enhance the single lap shear joint bonding performance. Within the tested laser processing parameters, a higher laser raster speed results in a larger enhancement. </p> <p>In addition to the static loading test with epoxy adhesive, different adhesive formulations are investigated and developed by the collaborating adhesive manufacturer to determine their suitability for use under the temperature and pressure conditions in HVAC&R systems. Reliability, especially fatigue failure, is another major concern because the strength of the adhesive joints is sufficient for HVAC&R applications. Two primary types of fatigue may happen in practical applications: thermal fatigue and vibration fatigue. Two test facilities were designed and built to test the adhesive performance and understand the failure mechanisms. For the thermal fatigue testing, a novel pressure and temperature cyclic (PTC) test stand was designed to simulate the pressure and temperature changes that may occur in HVAC&R systems. The test stand was designed to switch between hot high-pressure gas and cold low-pressure gas by using a compressor with hot gas bypass setup. For the vibration testing, a standard industrial shaker was used to provide the required vibration at a given displacement and frequency with a specially designed fixture for the tested joints. In both tests, adhesive joints were tested in parallel with brazed joints, undergoing extreme thermal and vibration loading conditions. All the samples were leak-checked before and after the testing, which were found to be leak-free after the testing, indicating that they pass the required qualification test according to available standards. It is confirmed that adhesive joints can be a potential alternative when dealing with thermal and vibration fatigue in the common working conditions of HVAC&R systems.</p> <p>The qualification testing is specific to the required loading conditions, such as pressure and temperature variations, and limited to certain tube sizes. An analytical model is developed to allow for design and evaluation across various operating conditions. The model aims to predict the adhesive stress and strain fields of in tube-to-tube joints based on the geometric parameters, material properties, and the loading conditions. In particular, the model uniquely considers the influence of thermal expansion and contraction in the joint, which is necessary for the periodically changing temperatures in HVAC&R systems. It is numerically solved using Mathematica and validated against the published data in the literature. The exact same solutions are achieved using the reported data in the literature, under simplified conditions without any temperature change involved. The validated model is then used in parametric studies to investigate the influence of geometric sizes and temperature change. Several conclusions are made about the trend of stress changes as well as the maximum stress, which provide insight from a perspective of general design guidance. Adhesive bonding length should be selected such that the maximum stress is smaller than the allowed material strength for both normal and shear stress. Adhesive thickness has less impact in the parametric range considered and is nevertheless usually dictated by the manufacture recommendation in view of other practical considerations. In regard to the thermal stresses, it is found that in practical HVAC&R working environment, the temperature-induced thermal stress dominates the stress fields and leads to significant change in the stress distribution across the adhesive layer. If a temperature change is present, the combination of all possible loading and temperature change should be analyzed to find the most extreme loading condition. This work demonstrates the first stress and strain analysis of tube-to-tube adhesive joints considering the working conditions of HVAC&R applications involving temperature cycling. All of these results provide a detailed guidance for use of adhesive joints across different application or locations in HVAC&R systems. The model can be also used as a framework to evaluate and compare the performance of different adhesives, as long as the adhesive properties are available.</p> <p>Lastly, it is also essential to demonstrate the application of these joints in real HVAC&R systems. A proof-of-concept test was done to demonstrate that the use of adhesive joints in a real system would cause no change in operation or leakage. A commercial heat pump dryer system was used to perform the testing at the Ray W. Herrick Laboratories. Two adhesive joints were installed to replace the brazing joints at the compressor inlet and outlet, where the most extreme temperature and pressure conditions are present. Results show that the system operates without any change in performance and experience no obvious leakage after more than 50 hours of testing over 6 months. </p> <p>This work explores the feasibility and reliability of adhesive bonding of copper for HVAC&R applications. The bonding strength of adhesive was studied and tested with both traditional surface preparation and advanced laser-interference structuring technique. The results show that for the tested structural epoxy adhesive, the bonding strength is large enough considering the internal pressure in the tube and the laser structure technique can increase the shear strength. </p> <p>The long-term reliability with respect to thermal, stress and vibration fatigue are then experimentally investigated and the adhesive joints pass the qualifications tests required by the standard. Further modeling work for predicting the stress distribution in adhesively bonded joints is developed to understand the influence on geometric parameters and temperature change. The adhesive length can influence the stress distribution significantly and temperature-induced stress dominates the stress distribution under the HVAC&R loading conditions. Further material characterization is needed for crack propagation or detailed fatigue analysis, which is highly dependent on the adhesive formula, working environment and loading conditions, which can be performed with a more specific targeted application. The experimental and modeling work in this thesis provides a foundation for adhesives to be applied in HVAC&R applications and a framework to further develop, optimize, and utilize adhesive joining in HVAC&R applications. </p>