Using Integration of Thermal-Balance Based and Data-Driven Models to Determine Single Duct Variable Air Volume System Cooling Baseline

2013 ◽  
Author(s):  
Gyujin Shim ◽  
Li Song ◽  
Gang Wang
Keyword(s):  
Physical Model ◽  
Least Square Method ◽  
Energy Performance ◽  
Least Square ◽  
Physical Models ◽  
Data Driven ◽  
Lumped Model ◽  
Variable Air Volume ◽  
Data Driven Approach ◽  
Cooling Coil

In order to use real-time energy measurements to identify system operation faults and inefficiencies, a cooling coil energy baseline is studied in an air-handling unit (AHU) through an integration of physical models and a data driven approach in this paper. A physical model for an AHU cooling coil energy consumption is first built to understand equipment mechanism and to determine the variables impacting cooling coil energy performance, and then the physical model is simplified into a lumped model by reducing the number of independent variables needed. Regression coefficients in the lumped model are determined statistically through searching optimal fit using the least square method with short periods of measured data. Experimental results on an operational AHU (8 ton) are presented to validate the effectiveness of this approach with statistical analysis. As a result of this experiment, the proposed cooling energy baselines at the cooling coil have ±20% errors at 99.7% confidence. Six-day data for obtaining baseline is preferred since it shows similar results as 12-day.

scholarly journals A Data-Driven Approach to Trip Generation Modeling for Urban Residents and Non-local Travelers

2020 ◽  
Vol 12 (18) ◽  
pp. 7688
Author(s):  
Fan Yang ◽  
Linchao Li ◽  
Fan Ding ◽  
Huachun Tan ◽  
Bin Ran
Keyword(s):  
Cellular Phone ◽  
Test Methods ◽  
Least Square ◽  
Data Driven ◽  
Urban Residents ◽  
Trip Generation ◽  
Explanatory Variables ◽  
Ordinary Least Square ◽  
Data Driven Approach ◽  
Non Local

Trip generation modeling is essential in transportation planning activities. Previous modeling methods that depend on traditional data collection methods are inefficient and expensive. This paper proposed a novel data-driven trip generation modeling method for urban residents and non-local travelers utilizing location-based social network (LBSN) data and cellular phone data and conducted a case study in Nanjing, China. First, the point of interest (POI) data of the LBSN were classified into various categories by the service type, then, four features of each category including the number of users, number of POIs, number of check-ins, and number of photos were aggregated by traffic analysis zones to be used as explanatory variables for the trip generation models. We used a random tree regression method to select the most important features as the model inputs, and the trip models were established based on the ordinary least square model. Then, an exploratory approach was used to test the performance of each combination of the variables with various test methods to identify the best model for residents’ and travelers’ trip generation functions. The results suggest land use compositions have significant impact on trip generations, and the trip generation patterns are different between urban residents and non-local travelers.

Automatic Reverse Engineering Based on Reconstructing Measurement Data in 3D-Lattice

2012 ◽  
Vol 523-524 ◽  
pp. 414-419
Author(s):  
Kiyomoto Tsushima ◽  
Hideki Aoyama
Keyword(s):  
Reverse Engineering ◽  
Mathematical Models ◽  
Measurement Data ◽  
Least Square Method ◽  
Lattice Points ◽  
Least Square ◽  
Physical Models ◽  
Engineering System ◽  
Surface Data ◽  
Data Density

Reverse engineering systems are used to construct mathematical models of physical models such as clay model based on measurement data. In this study, we proposed a reverse engineering method which can construct high quality surface data automatically. This method consists of the following steps; The first globally and regionally smooths measured data based on the target shape by fitting quadric surface to measurement data. The second defines quadric surfaces and converts measurement points into 3D lattice points to obtain uniform measurement data density. As the positions of measurement data are converted from coordinate values into 3D lattice points, it is easier to find neighboring points and clarify neighboring relations between surfaces. The third acquires segment measurement data based on maximum curvatures and normals at each point. The last defines NURBS surfaces for each segment using the least square method to average positional errors. In order to validate the effectiveness of the proposed method, we developed a reverse engineering system and constructed mathematical models through basic experiments using clay car model measurement data.

Hybrid Physics-Constrained and Data-Driven Approach for Interwell Saturation Estimation from Well Logs

2021 ◽  
Author(s):  
Dmitry Kovalev ◽  
Sergey Safonov ◽  
Klemens Katterbauer ◽  
Alberto Marsala
Keyword(s):  
Water Saturation ◽  
Physical Models ◽  
Data Driven ◽  
Well Log ◽  
Points Of Interest ◽  
Step Model ◽  
Primary Fluid ◽  
Data Driven Approach ◽  
Aggregated Model ◽  
Well Log Analysis

Abstract Combining physics-based models for well log analysis with artificial intelligence (AI) advanced algorithms is crucial for wellbore studies. Data-driven methods do not generalize well and lack theoretical knowledge accumulated in the field. Estimating well saturation significantly improves if predictions from physical models are used to constrain data-driven algorithms in outlined primary fluid channels and other important points of interest. Saturation propagations in the reservoirs interwell region also generalize better under using combination of models. This work addresses combined usage of theoretical and data-driven models by aggregating them into single hybrid model. Multiple physical and data-driven models are under study, their parameters are optimized using observations. Weighted sum is used to predict water saturation at every point with weights being recomputed at each step. Model outputs are compared in terms accuracy and cumulative loss. A synthesized reservoir box model encompassing volumetric interwell porosity, resistivity and saturation data is used for the validation of the algorithms. Aggregated model for estimating interwell saturation shows improved prediction accuracy compared both to physics-based or data-driven approaches separately.

Fault Diagnostic of Electro-Hydraulic Servo Valve

2012 ◽  
Vol 619 ◽  
pp. 259-263
Author(s):  
Hai Yang Pan ◽  
Shu Nan Liu ◽  
Yong Ming Yao ◽  
Yi Lin Jiang
Keyword(s):  
Physical Model ◽  
Failure Modes ◽  
Health Management ◽  
Data Driven ◽  
Servo Valve ◽  
Data Driven Approach ◽  
Hydraulic Servo ◽  
Simulation Results ◽  
Fault Diagnostic ◽  
Prognostic And Health Management

In order to achieve the fault diagnostic and degradation assessment of the electro-hydraulic servo valve, a system of prognostic and health management based on the data-driven approach for the electro-hydraulic servo valve is presented. FMMEA performed in this study considered the degree of sub-components in electro-hydraulic servo valve. In order to use only five parameters to assess the cause of degradation, a physical model of the EH servo valve was built up to simulate the failure modes. The simulation results are very useful because the methods can be applied to assess the cause of degradation such as leakage, jamming, clogging and so on.

Linearizing Power Flow Model: A Hybrid Physical Model-Driven and Data-Driven Approach

2020 ◽  
Vol 35 (3) ◽  
pp. 2475-2478 ◽  
Author(s):  
Yi Tan ◽  
Yuanyang Chen ◽  
Yong Li ◽  
Yijia Cao
Keyword(s):  
Physical Model ◽  
Power Flow ◽  
Flow Model ◽  
Data Driven ◽  
Model Driven ◽  
Data Driven Approach

scholarly journals Augmenting physical models with deep networks for complex dynamics forecasting*

2021 ◽  
Vol 2021 (12) ◽  
pp. 124012
Author(s):  
Yuan Yin ◽  
Vincent Le Guen ◽  
Jérémie Dona ◽  
Emmanuel de Bézenac ◽  
Ibrahim Ayed ◽  
...  
Keyword(s):  
Physical Model ◽  
Wave Equations ◽  
Complex Dynamics ◽  
Reaction Diffusion ◽  
Reaction Diffusion Equations ◽  
Physical Models ◽  
Data Driven ◽  
Physical Parameters ◽  
Learning Problem ◽  
Scientific Fields

Abstract Forecasting complex dynamical phenomena in settings where only partial knowledge of their dynamics is available is a prevalent problem across various scientific fields. While purely data-driven approaches are arguably insufficient in this context, standard physical modeling-based approaches tend to be over-simplistic, inducing non-negligible errors. In this work, we introduce the APHYNITY framework, a principled approach for augmenting incomplete physical dynamics described by differential equations with deep data-driven models. It consists of decomposing the dynamics into two components: a physical component accounting for the dynamics for which we have some prior knowledge, and a data-driven component accounting for errors of the physical model. The learning problem is carefully formulated such that the physical model explains as much of the data as possible, while the data-driven component only describes information that cannot be captured by the physical model; no more, no less. This not only provides the existence and uniqueness for this decomposition, but also ensures interpretability and benefit generalization. Experiments made on three important use cases, each representative of a different family of phenomena, i.e. reaction–diffusion equations, wave equations and the non-linear damped pendulum, show that APHYNITY can efficiently leverage approximate physical models to accurately forecast the evolution of the system and correctly identify relevant physical parameters. The code is available at https://github.com/yuan-yin/APHYNITY.

Data-driven rational feedforward tuning: With application to an ultraprecision wafer stage

2019 ◽  
Vol 234 (6) ◽  
pp. 748-758 ◽  
Author(s):  
Weicai Huang ◽  
Kaiming Yang ◽  
Yu Zhu ◽  
Xin Li ◽  
Haihua Mu ◽  
...  
Keyword(s):  
High Performance ◽  
Optimal Parameter ◽  
Least Square Method ◽  
Parameter Tuning ◽  
Basis Functions ◽  
Least Square ◽  
Data Driven ◽  
Estimation Accuracy ◽  
Rational Basis ◽  
Noisy Condition

Rational basis functions are introduced into iterative learning control to enhance the flexibility towards nonrepeating tasks. At present, the application of rational basis functions either suffers from nonconvex optimization problem or requires the predefinition of poles, which restricts the achievable performance. In this article, a new data-driven rational feedforward tuning approach is developed, in which convex optimization is realized without predefining the poles. Specifically, the optimal parameter which eliminates the reference-induced error is directly solved using the least square method. No parametric model is involved in the parameter tuning process and the optimal parameter is estimated using the measured data. In the noisy condition, it is proved that the estimated optimal parameter is unbiased and the estimation accuracy in terms of variance is analysed. The performance of the proposed approach is tested on an ultraprecision wafer stage. The experimental results confirm that high performance is achieved using the proposed approach.

Sign in / Sign up

Export Citation Format

Share Document