IU Indianapolis ScholarWorks :: Browsing by Author "Razban, Ali"

Browsing by Author "Razban, Ali"

Now showing 1 - 10 of 50

A Bi-Level Data-Driven Framework for Fault-Detection and Diagnosis of HVAC Systems feature explainability
(Elsevier, 2022-07) Movahed, Paria; Taheri, Saman; Razban, Ali; Mechanical Engineering, School of Engineering and Technology
Machine learning methods have lately received considerable interest for fault detection diagnostic (FDD) analysis of heating, ventilation, and air conditioning (HVAC) systems due to their high detection accuracy. Meanwhile, HVAC malfunctions are regarded as rare occurrences, hence normal operating data samples are much more accessible than data samples in faulty and malfunctioning conditions. The dominating frequency of normal operation in HVAC datasets have also led to heavily biased classification algorithms within the literature. Moreover, the focus of previous literature has been on increasing accuracy of the models while this leads to a high number of false positives (misleading alarms) in the system. To enhance the performance of diagnostic procedures and fill the mentioned gaps, this study proposes a novel data-driven framework. A bi-level machine learning framework is developed for diagnosing faults in air handling units and rooftop units based on principal component analysis (PCA), time series anomaly detection, and random forest (RF). It is shown that PCA can reduce the dataset dimension with one principal component accounting for 95% of data variance. Also, the random forest could classify the faults with 89% precision for single zone AHU, 85% precision for RTU, and 79% for multi-zone AHU.
A Control Oriented Soot Prediction Model for Diesel Engines Using an Integrated Approach
(American Society of Mechanical Engineers, 2021-11-01) Shewale, Mahesh S.; Razban, Ali; Mechanical Engineering, School of Engineering and Technology
Diesel engines have been used in many vehicles and power generation units since a long time due to their less fuel consumption and high trustworthiness. With reference to upcoming emission norms, various engine out emissions have proved to be causing adverse effect on human health and environment. Soot, or particulate matter is one of the major pollutants in diesel engine out emissions and causes various lung related issues. There have been efforts to reduce the amount of soot generated using after-treatment devices like diesel particulate filter (DPF) to filter out particles and get clean tailpipe emissions. These technologies increase load on the system and involves additional maintenance. Also, deposition-based soot sensors have been found to be inoperative in certain scenarios like cold start conditions. In this research work, an effort has been made to develop a phenomenological model that predicts soot mass generated in a Cummins 6.7L diesel engine. The model uses in-cylinder conditions such as pressure, bulk mean temperature, fuel mass flow rate and injector orifice diameter. The difference between soot mass formed and oxidized yields the net amount of soot generated at engine out end. Furthermore, the generated soot mass is compared with benchmark results for specific load conditions and appropriate controller is designed to minimize this tradeoff. The control parameter being used here is fuel rail pressure, which controls the lift-off length, and ultimately equivalence ratio, which predicts mass of soot, generated in formation phase. The presented method shows a prediction error ranging from 5–20%, which is significantly reduced to 2% using a PID controller. The approach presented in this research work is generic and can be operated as stand-alone system or an integrated subsystem in a higher order control architecture.
Accurate location of tumor in head and neck cancer radiotherapy treatment with respect to machine isocentre
(2017-05) Tangirala, Deepak Kumar; Razban, Ali; Chen, Jie; Tovar, Andres
Radiation Therapy has been one of the most common techniques to treat various types of cancers, in particular is Head and Neck Cancer (HNC) which accounts for three percent of all cancers in the United States. During the treatment procedure, the patient is immobilized using immobilization devices such as the full head face mask, bite blocks, stereotactic frame, etc. to get accurate location of tumor. The disadvantage of these devices is that they are very uncomfortable to the patient especially people suffering from Post-Traumatic Stress Disorder (PTSD) and claustrophobia who cannot wear any confined masked system such as the full head mask or bite block during the treatment procedure. To mitigate this problem, there has been a lot of research in modifying such immobilizing devices without neglecting the accurate location of tumor. To this end, the research presented in this thesis focuses on developing a mask less system with accurately locating the position of tumor using the technique of coordinate transformation at the same time fulfilling the three important characteristics: • Comfort • Accuracy • Low price Such a system is comfortable to the patient because no confining mask system is used and we choose minimal contact points on the patient for fixing the patient. Traditionally, such type of cancer treatment is carried out in two stages: Diagnosis stage, which identifies the location of the tumor and the external markers and the Treatment stage where the tumor is treated with immobilization device being common in both the stages. In the new system, the immobilization devices vary at the two stages. The head position is monitored by using pressure sensor assembly where spring and pressure sensor setup detects the amount and direction of head deviation. We also prepare a customized 3D printed nose bridge part for extra referencing in the treatment room. Also, it is important that we use material for our immobilization devices which does not contain any metal and MRI compatible. Once the patient lies down on the treatment couch and is immobilized using the immobilization devices, then tumor location is calculated using the theory of coordinate transformation and transformation matrix in the Diagnosis and Treatment Stage. To validate the system, simulation of immobilization devices used in the new design was carried out using ANSYS Workbench 15.0 and LS-Dyna software’s Explicit Dynamics method. The simulation for the head-fixing device showed a deflection of ±0.1974 mm with respect to machine isocenter with a load of 60 N, which is lower than the customer requirement of ±3 mm with respect to machine isocenter of head deviation. The material used for the external markers for patient positioning was selected to be polyetheretherketone (PEEK) which is a radiolucent and widely used MRI compatible material. The system also takes into consideration the effect of weight loss, which is one of the drawbacks of the current systems. Although still in the development stage, this mask less system holds to be the next new variety of immobilization devices that are comfortable to the patient and less expensive to be implemented in future cancer treatment practices.
Air Compressor Load Forecasting using Artificial Neural Network
(Elsevier, 2021-04) Wu, Da-Chun; Bahrami Asl, Babak; Razban, Ali; Chen, Jie; Mechanical and Energy Engineering, School of Engineering and Technology
Air compressor systems are responsible for approximately 10% of the electricity consumed in United States and European Union industry. As many researches have proven the effectiveness of using Artificial Neural Network in air compressor performance prediction, there is still a need to forecast the air compressor electrical load profile. The objective of this study is to predict compressed air systems' electrical load profile, which is valuable to industry practitioners as well as software providers in developing better practice and tools for load management and look-ahead scheduling programs. Two artificial neural networks, Two-Layer Feed-Forward Neural Network and Long Short-Term Memory were used to predict an air compressors electrical load. Compressors with three different control mechanisms are evaluated with a total number of 11,874 observations. The forecasts were validated using out-of-sample datasets with 5-fold cross-validation. Models produced average coefficient of determination values from 0.24 to 0.94, average root-mean-square errors from 0.05 kW - 5.83 kW, and mean absolute scaled errors from 0.20 to 1.33. The results indicate that both artificial neural networks yield good results for compressors using variable speed drive (average R2 = 0.8 and no naïve forecasting), only the long short-term memory model gives acceptable results for compressors using on/off control (average R2 = 0.82 and no naïve forecasting), and no satisfactory results are obtained for load/unload type air compressors (models constituting naïve forecasting).
Analysis of magnetic flux in magneto-rheological damper
(IOP, 2019-07) Purandare, Snehal; Zambare, Hrishikesh; Razban, Ali; Mechanical and Energy Engineering, School of Engineering and Technology
Magnetorheological materials are a class of smart substances whose rheological properties can rapidly be varied by application of a magnetic field. The proposed damper consists of an electromagnet and a piston immersed in MR fluid. When current is applied to the electromagnet, the MR fluid solidifies as its yield stress varies in response to the applied magnetic field. Hence, the generation of a magnetic field is an important phenomenon in MR damper. In this research, the magnetic field generated in the damper was analyzed by applying finite element method using COMSOL Multiphysics and was validated using magnetic circuit theory. A quasi-static, 2D—Axisymmetric model was developed using parametric study by varying current from 0–3 A and the magnetic flux density change generated in the fluid flow gap of MR fluid due to external applied current was evaluated. According to the analytical calculations magnetic flux density generated at MR fluid gap was 0.64 Tesla and when calculated using FEA magnetic flux density generated was 0.61 Tesla for 1A current. There is a difference of 4.8% in the simulated results and analytically calculated results of automotive MR damper due to non linear BH curve consideration in Finite element analysis over linear consideration of BH relation in magnetic circuit theory.
Analysis of magnetic flux in magneto-rheological damper
(MDPI, 2019-07) Purandare, Snehal; Zambare, Hrishikesh; Razban, Ali; Mechanical and Energy Engineering, School of Engineering and Technology
Magnetorheological materials are a class of smart substances whose rheological properties can rapidly be varied by application of a magnetic field. The proposed damper consists of an electromagnet and a piston immersed in MR fluid. When current is applied to the electromagnet, the MR fluid solidifies as its yield stress varies in response to the applied magnetic field. Hence, the generation of a magnetic field is an important phenomenon in MR damper. In this research, the magnetic field generated in the damper was analyzed by applying finite element method using COMSOL Multiphysics and was validated using magnetic circuit theory. A quasi-static, 2D—Axisymmetric model was developed using parametric study by varying current from 0–3 A and the magnetic flux density change generated in the fluid flow gap of MR fluid due to external applied current was evaluated. According to the analytical calculations magnetic flux density generated at MR fluid gap was 0.64 Tesla and when calculated using FEA magnetic flux density generated was 0.61 Tesla for 1A current. There is a difference of 4.8% in the simulated results and analytically calculated results of automotive MR damper due to non linear BH curve consideration in Finite element analysis over linear consideration of BH relation in magnetic circuit theory.
Analyzing Compressed Air Demand Trends to Develop a Method to Calculate Leaks in a Compressed Air Line Using Time Series Pressure Measurements
(2022-05) Daniel, Ebin John; Razban, Ali; Goodman, David; Chen, Jie
Compressed air is a powerful source of stored energy and is used in a variety of applications varying from painting to pressing, making it a versatile tool for manufacturers. Due to the high cost and energy consumption associated with producing compressed air and it’s use within industrial manufacturing, it is often referred to as a fourth utility behind electricity, natural gas, and water. This is the reason why air compressors and associated equipment are often the focus for improvements in the eyes of manufacturing plant managers. As compressed air can be used in multiple ways, the methods used to extract and transfer the energy from this source vary as well. Compressed air can flow through different types of piping, such as aluminum, Polyvinyl Chloride (PVC), rubber, etc. with varying hydraulic diameters, and through different fittings such as 90-degree elbows, T-junctions, valves, etc. which can cause one of the major concerns related to managing the energy consumption of an air compressor, and that is the waste of air through leaks. Air leaks make up a considerable portion of the energy that is wasted in a compressed air system, as they cause a multitude of problems that the compressor will have to make up for to maintain the steady operation of the pneumatic devices on the manufacturing floor that rely on compressed air for their application. When air leaks are formed within the compressed air piping network, they act as continuous consumers and cause not only the siphoning off of said compressed air, put also reduce the pressure that is needed within the pipes. The air compressors will have to work harder to compensate for the losses in the pressure and the amount of air itself, causing an overconsumption of energy and power. Overworking the air compressor also causes the internal equipment to be stretched beyond its capabilities, especially if they are already running at full loads, reducing their total lifespans considerably. In addition, if there are multiple leaks close to the pneumatic devices on the manufacturing floor, the immediate loss in pressure and air can cause the devices to operate inefficiently and thus cause a reduction in production. This will all cumulatively impact the manufacturer considerably when it comes to energy consumption and profits. There are multiple methods of air leak detection and accounting that currently exist so as to understand their impact on the compressed air systems. The methods are usually conducted when the air compressors are running but during the time when there is no, or minimal, active consumption of the air by the pneumatic devices on the manufacturing floor. This time period is usually called non-production hours and generally occur during breaks or between employee shift changes. This time is specifically chosen so that the only air consumption within the piping is that of the leaks and thus, the majority of the energy and power consumed during this time is noted to be used to feed the air leaks. The collected data is then used to extrapolate and calculate the energy and power consumed by these leaks for the rest of the year. There are, however, a few problems that arise when using such a method to understand the effects of the leaks in the system throughout the year. One of the issues is that it is assumed that the air and pressure lost through the found leaks are constant even during the production hours i.e. the hours that there is active air consumption by the pneumatic devices on the floor, which may not be the case due to the increased air flow rates and varying pressure within the line which can cause an increase in the amount of air lost through the same orifices that was initially detected. Another challenge that arises with using only the data collected during a single non-production time period is that there may be additional air leaks that may be created later on, and the energy and power lost due to the newer air leaks would remain unaccounted for. As the initial estimates will not include the additional losses, the effects of the air leaks may be underestimated by the plant managers. To combat said issues, a continuous method of air leak analyses will be required so as to monitor the air compressors’ efficiency in relation to the air leaks in real time. By studying a model that includes both the production, and non-production hours when accounting for the leaks, it was observed that there was a 50.33% increase in the energy losses, and a 82.90% increase in the demand losses that were estimated when the effects of the air leaks were observed continuously and in real time. A real time monitoring system can provide an in-depth understanding of the compressed air system and its efficiency. Managing leaks within a compressed air system can be challenging especially when the amount of energy wasted through these leaks are unaccounted for. The main goal of this research was to find a nonintrusive way to calculate the amount of air as well as energy lost due to these leaks using time series pressure measurements. Previous studies have shown a strong relationship between the pressure difference, and the use of air within pneumatic lines, this correlation along with other factors has been exploited in this research to find a novel and viable method of leak accounting to develop a Continuous Air Leak Monitoring (CALM) system.
ARC algorithm: A novel approach to forecast and manage daily electrical maximum demand
(Elsevier, 2018-07) Wu, Da-Chun; Amini, Amin; Razban, Ali; Chen, Jie; Mechanical Engineering, School of Engineering and Technology
This paper proposes an innovative algorithm for predicting short-term electrical maximum demand by using historical demand data. The ability to recognize in peak demand pattern for commercial or industrial customers would propose numerous direct and indirect benefits to the customers and utility providers in terms of demand reduction, cost control, and system stability. Prior works in electrical maximum demand forecasting have been mainly focused on seasonal effects, which is not a feasible approach for industrial manufacturing facilities in short-term load forecasting. The proposed algorithm, denoted as the Adaptive Rate of Change (ARC), determines the logarithmic rate-of-change in load profile prior to a peak by postulating the demand curve as a stochastic, mean-reverting process. The rationale behind this analysis, is that the energy efficient program requires not only demand estimation but also to warn the user of imminent maximum peak occurrence. This paper analyzes demand trend data and incorporates stochastic model and mean reverting half-life to develop an electrical maximum demand forecasting algorithm, which is statistically evaluated by cross-table and F-score for three different manufacturing facilities. The aggregate results show an overall accuracy of 0.91 and a F-score of 0.43, which indicates that the algorithm is effective predicting peak demand in predicting peak demand.
An Automated Grid-Based Robotic Alignment System for Pick and Place Applications
(2013-12) Bearden, Lukas R.; Razban, Ali; Wasfy, Tamer; Li, Lingxi; Anwar, Sohel
This thesis proposes an automated grid-based alignment system utilizing lasers and an array of light-detecting photodiodes. The intent is to create an inexpensive and scalable alignment system for pick-and-place robotic systems. The system utilizes the transformation matrix, geometry, and trigonometry to determine the movements to align the robot with a grid-based array of photodiodes. The alignment system consists of a sending unit utilizing lasers, a receiving module consisting of photodiodes, a data acquisition unit, a computer-based control system, and the robot being aligned. The control system computes the robot movements needed to position the lasers based on the laser positions detected by the photodiodes. A transformation matrix converts movements from the coordinate system of the grid formed by the photodiodes to the coordinate system of the robot. The photodiode grid can detect a single laser spot and move it to any part of the grid, or it can detect up to four laser spots and use their relative positions to determine rotational misalignment of the robot. Testing the alignment consists of detecting the position of a single laser at individual points in a distinct pattern on the grid array of photodiodes, and running the entire alignment process multiple times starting with different misalignment cases. The first test provides a measure of the position detection accuracy of the system, while the second test demonstrates the alignment accuracy and repeatability of the system. The system detects the position of a single laser or multiple lasers by using a method similar to a center-of-gravity calculation. The intensity of each photodiode is multiplied by the X-position of that photodiode. The summed result from each photodiode intensity and position product is divided by the summed value of all of the photodiode intensities to get the X-position of the laser. The same thing is done with the Y-values to get the Y-position of the laser. Results show that with this method the system can read a single laser position value with a resolution of 0.1mm, and with a maximum X-error of 2.9mm and Y-error of 2.0mm. It takes approximately 1.5 seconds to process the reading. The alignment procedure calculates the initial misalignment between the robot and the grid of photodiodes by moving the robot to two distinct points along the robot’s X-axis so that only one laser is over the grid. Using these two detected points, a movement trajectory is generated to move that laser to the X = 0, Y = 0 position on the grid. In the process, this moves the other three lasers over the grid, allowing the system to detect the positions of four lasers and uses the positions to determine the rotational and translational offset needed to align the lasers to the grid of photodiodes. This step is run in a feedback loop to update the adjustment until it is within a permissible error value. The desired result for the complete alignment is a robot manipulator positioning within ±0.5mm along the X and Y-axes. The system shows a maximum error of 0.2mm in the X-direction and 0.5mm in the Y-direction with a run-time of approximately 4 to 5 minutes per alignment. If the permissible error value of the final alignment is tripled the alignment time goes down to 1 to 1.5 minutes and the maximum error goes up to 1.4mm in both the X and Y-directions. The run time of the alignment decreases because the system runs fewer alignment iterations.
Benchmarking Tool Development for Commercial Buildings' Energy Consumption Using Machine Learning
(2024-05) Hosseini, Paniz; Razban, Ali; Chen, Jie; Goodman, David
This thesis investigates approaches to classify and anticipate the energy consumption of commercial office buildings using external and performance benchmarking to reduce the energy consumption. External benchmarking in the context of building energy consumption considers the influence of climate zones that significantly impact a building's energy needs. Performance benchmarking recognizes that different types of commercial buildings have distinct energy consumption patterns. Benchmarks are established separately for each building type to provide relevant comparisons. The first part of this thesis is about providing a benchmarking baseline for buildings to show their consumption levels. This involves simulating the buildings based on standards and developing a model based on real-time results. Software tools like Open Studio and Energy Plus were utilized to simulate buildings representative of different-sized structures to organize the benchmark energy consumption baseline. These simulations accounted for two opposing climate zones—one cool and humid and one hot and dry. To ensure the authenticity of the simulation, details, which are the building envelope, operational hours, and HVAC systems, were matched with ASHRAE standards. Secondly, the neural network machine learning model is needed to predict the consumption of the buildings based on the trend data came out of simulation part, by training a comprehensive set of environmental characteristics, including ambient temperature, relative humidity, solar radiation, wind speed, and the specific HVAC (Heating, Ventilation, and Air Conditioning) load data for both heating and cooling of the building. The model's exceptional accuracy rating of 99.54% attained across all, which comes from the accuracy of training, validation, and test about 99.6%, 99.12%, and 99.42%, respectively, and shows the accuracy of the predicted energy consumption of the building. The validation check test confirms that the achieved accuracy represents the optimal performance of the model. A parametric study is done to show the dependency of energy consumption on the input, including the weather data and size of the building, which comes from the output data of machine learning, revealing the reliability of the trained model. Establishing a Graphic User Interface (GUI) enhances accessibility and interaction for users. In this thesis, we have successfully developed a tool that predicts the energy consumption of office buildings with an impressive accuracy of 99.54%. Our investigation shows that temperature, humidity, solar radiation, wind speed, and the building's size have varying impacts on energy use. Wind speed is the least influential component for low-rise buildings but can have a more substantial effect on high-rise structures.

Browsing by Author "Razban, Ali"

Results Per Page

Sort Options