SlideShare a Scribd company logo
Finite Impulse Response Estimation of Gas Furnace Data in IMPL
Industrial Modeling Framework (FIRE-GFD-IMF)
i n d u s t r IAL g o r i t h m s LLC. (IAL)
www.industrialgorithms.com
December 2014
Introduction to Finite Impulse Response Estimation, UOPSS and QLQP
Presented in this short document is a description of how to estimate deterministic and stochastic
non-parametric finite impulse response (FIR) models in IMPL applied to industrial gas furnace
data identical to that found in TSE-GFD-IMF using parametric transfer-functions. The
methodology of time-series analysis or system identification involves essentially three (3) stages
(Box and Jenkins, 1976): (1) model structure identification, (2) model parameter estimation and
(3) model checking and diagnostics. We do not address (1) which requires stationarity and
seasonality assessment/adjustment, auto-, cross- and partial-correlation, etc. to establish the
parametric transfer function polynomial degrees especially when we are using non-parametric
FIR estimation. Instead we focus only on the parameter estimation and diagnostics. These
types of parameter estimation problems involve dynamic and nonlinear relationships shown
below and we solve these using IMPL’s Sequential Equality-Constrained QP Engine (SECQPE)
and Supplemental Observability, Redundancy and Variability Estimator (SORVE). Other types
of non-parametric identification known as Subspace Identification (Qin, 2006) and can used to
estimate state-space models.
Figure 1 shows the gas furnace data example found in Series J of Box and Jenkins (1976)
where we depict the problem using signal processing constructs configured in our unit-
operation-port-state superstructure (UOPSS) (Kelly, 2004, 2005; Zyngier and Kelly, 2012).
Figure 1. Gas Furnace Data in a “Signal Processing” Based UOPSS Flowsheet.
The diamond shapes or objects are the sources and sinks known as perimeters where U1 and
Y are the input gas flowrate and CO2% in the effluent flue gas of the furnace respectively. The
rectangular shapes with the cross-hairs are continuous-process units where the
DeterministicModel (process) and StochasticModel (noise) are blackbox subtypes which allow
any number of process or operating conditions (and coefficients) to be attached with ad hoc
formulas or equations representing the non-parametric FIR. The Splicer is a signal processing
shape to add and/or subtract the inlet signals together producing a single output i.e., Z = Y – X1
in our case. The circles with and without cross-hairs are outlet and inlet port-states respectively.
Port-states are unambiguous interfaces between up and downstream unit-operations.
The deterministic FIR model in discrete-time or difference form (versus z-transform or
backwards-shift operator-based quotients of rational polynomials) is defined as follows:
X1,0 = G0*U1,0+G1*U1,1+G2*U1,2+G3*U1,3+G4*U1,4+G5*U1,5+G6*U1,6+G7*U1,7+G8*U1,8+G9*U1,9+G10*U1,10
where U1 is the exogenous input signal minus its mean of -0.057 at time-periods t-1, t-2, …, t-
10 and X1is the “deterministic state” at time-period t-0. The corresponding parameters or
coefficients are G0, G1, …, G10 are the FIR values for each time-shift or lag in the past where
G0 is typically set to zero (0.0) given discrete-time sampling and some of the initial coefficients
G1, etc. will also be effectively zero (0.0) given the dead-time or inherent delay in the system.
The static or steady-state gain of each input with respect to each output can be easily calculated
from the dynamic gains or FIR’s by taking their sum. During the initial part of the estimation
procedure, the number of FIR’s is typically set to some number greater than expected. Then,
using the parameter variances and the Student-t statistics (parameter confidence-intervals) the
actual number is reduced to hopefully avoid over-parameterization which is a well-known
disadvantage of FIR’s and non-parametric models in general.
Similarly, the stochastic FIR model or unmeasured noise disturbance model is also defined as
follows (Schoukens et. al. 2011):
A,0 = H0*Z,0+H1*Z,1+H2*Z,2+H3*Z,3+H4*Z,4+H5*Z,5+H6*Z,6+H7*Z,7+H8*Z,8+H9*Z,9+H10*Z,10
where A is the assumed white-noise input signal and Z is the “stochastic state” which is equal to
Z = Y – X1 and the time-series Y is also minus its mean of 53.51. The parameters H0, H1, …,
H10 and actually represent the “inverse” of stochastic noise model in non-parametric form. The
noise model is essentially a time- or frequency-dependent weighting filter and is very important
to ensure that the A-series is independent and identically distributed (MacGregor and Fogal,
1995; Shreesha and Gudi, 2004; Schoukens et. al. 2011) which is squared, summed and
minimized in the objective function of the prediction error or nonlinear least-squares regression.
Without this noise filter, the estimation is well-understood to be biased yielding inaccurate FIR
coefficients but unfortunately makes the estimator nonlinear since H and Z are variables.
From a quantity-logic-quality phenomena (QLQP) perspective, the time-series U1, Y, Z and A
found in Figure 1 are considered as flows or more appropriately signal-flows or data. However,
in our IML implementation found in Appendix A we collapse the three (3) continuous-processes
into one blackbox model as shown by the dotted-line box in Figure 1 where the flows of U1, Y, Z
and A are now considered as conditions and the G and H parameters are static coefficients in
the IMPL semantics.
Once the FIR parameters are known then these can be straightforwardly implemented into
advanced process controllers such as found in APC-IMF-Julia. It should also be stressed that
for multiple-input, multiple-output (MIMO) processes, the design of the input time-series plays
an important role in the success of regressing good and useful dynamic representations such as
transfer-function, state-space and FIR models where a novel design of the external excitation or
dither signals can be found in DSDP-CLE-IMF for open- and/or closed-loop identification.
Industrial Modeling Framework (IMF), IMPL and SSIIMPLE
To implement the mathematical formulation of this and other systems, IAL offers a unique
approach and is incorporated into our Industrial Modeling Programming Language we call IMPL.
IMPL has its own modeling language called IML (short for Industrial Modeling Language) which
is a flat or text-file interface as well as a set of API's which can be called from any computer
programming language such as C, C++, Fortran, C#, VBA, Java (SWIG), Python (CTYPES)
and/or Julia (CCALL) called IPL (short for Industrial Programming Language) to both build the
model and to view the solution. Models can be a mix of linear, mixed-integer and nonlinear
variables and constraints and are solved using a combination of LP, QP, MILP and NLP solvers
such as COINMP, GLPK, LPSOLVE, SCIP, CPLEX, GUROBI, LINDO, XPRESS, CONOPT,
IPOPT, KNITRO and WORHP as well as our own implementation of SLP called SLPQPE
(Successive Linear & Quadratic Programming Engine) which is a very competitive alternative to
the other nonlinear solvers and embeds all available LP and QP solvers.
In addition and specific to DRR problems, we also have a special solver called SECQPE
standing for Sequential Equality-Constrained QP Engine which computes the least-squares
solution and a post-solver called SORVE standing for Supplemental Observability, Redundancy
and Variability Estimator to estimate the usual DRR statistics. SECQPE also includes a
Levenberg-Marquardt regularization method for nonlinear data regression problems and can be
presolved using SLPQPE i.e., SLPQPE warm-starts SECQPE. SORVE is run after the
SECQPE solver and also computes the well-known "maximum-power" gross-error statistics
(measurement and nodal/constraint tests) to help locate outliers, defects and/or faults i.e., mal-
functions in the measurement system and mis-specifications in the logging system.
The underlying system architecture of IMPL is called SSIIMPLE (we hope literally) which is short
for Server, Solvers, Interfacer (IML), Interacter (IPL), Modeler, Presolver Libraries and
Executable. The Server, Solvers, Presolver and Executable are primarily model or problem-
independent whereas the Interfacer, Interacter and Modeler are typically domain-specific i.e.,
model or problem-dependent. Fortunately, for most industrial planning, scheduling,
optimization, control and monitoring problems found in the process industries, IMPL's standard
Interfacer, Interacter and Modeler are well-suited and comprehensive to model the most difficult
of production and process complexities allowing for the formulations of straightforward
coefficient equations, ubiquitous conservation laws, rigorous constitutive relations, empirical
correlative expressions and other necessary side constraints.
User, custom, adhoc or external constraints can be augmented or appended to IMPL when
necessary in several ways. For MILP or logistics problems we offer user-defined constraints
configurable from the IML file or the IPL code where the variables and constraints are
referenced using unit-operation-port-state names and the quantity-logic variable types. It is also
possible to import a foreign *.ILP file (row-based MPS file) which can be generated by any
algebraic modeling language or matrix generator. This file is read just prior to generating the
matrix and before exporting to the LP, QP or MILP solver. For NLP or quality problems we offer
user-defined formula configuration in the IML file and single-value and multi-value function
blocks writable in C, C++ or Fortran. The nonlinear formulas may include intrinsic functions
such as EXP, LN, LOG, SIN, COS, TAN, MIN, MAX, IF, NOT, EQ, NE, LE, LT, GE, GT and CIP,
LIP, SIP and KIP (constant, linear and monotonic spline interpolations) as well as user-written
extrinsic functions (XFCN). It is also possible to import another type of foreign file called the
*.INL file where both linear and nonlinear constraints can be added easily using new or existing
IMPL variables.
Industrial modeling frameworks or IMF's are intended to provide a jump-start to an industrial
project implementation i.e., a pre-project if you will, whereby pre-configured IML files and/or IPL
code are available specific to your problem at hand. The IML files and/or IPL code can be
easily enhanced, extended, customized, modified, etc. to meet the diverse needs of your project
and as it evolves over time and use. IMF's also provide graphical user interface prototypes for
drawing the flowsheet as in Figure 1 and typical Gantt charts and trend plots to view the solution
of quantity, logic and quality time-profiles. Current developments use Python 2.3 and 2.7
integrated with open-source Gnome Dia and Matplotlib modules respectively but other
prototypes embedded within Microsoft Excel/VBA for example can be created in a
straightforward manner.
However, the primary purpose of the IMF's is to provide a timely, cost-effective, manageable
and maintainable deployment of IMPL to formulate and optimize complex industrial
manufacturing systems in either off-line or on-line environments. Using IMPL alone would be
somewhat similar (but not as bad) to learning the syntax and semantics of an AML as well as
having to code all of the necessary mathematical representations of the problem including the
details of digitizing your data into time-points and periods, demarcating past, present and future
time-horizons, defining sets, index-sets, compound-sets to traverse the network or topology,
calculating independent and dependent parameters to be used as coefficients and bounds and
finally creating all of the necessary variables and constraints to model the complex details of
logistics (discrete) and quality (nonlinear) industrial optimization problems. Instead, IMF's and
IMPL provide, in our opinion, a more elegant and structured approach to industrial modeling and
solving so that you can capture the benefits of advanced decision-making faster, better and
cheaper.
Finite Impulse Response Estimation of Gas Furnace Data Synopsis
After iterating using SECQPE several times and setting certain G and H coefficients to zero
(0.0) depending on their reported confidence-intervals from SORVE, which is the typical
protocol especially with non-parametric estimation, their values with two (2) times their
standard-error are shown below:
G0 = 0.0
G1 = 0.0
G2 = 0.0
G3 = -0.534 +/- 0.15
G4 = -0.667 +/- 0.16
G5 = -0.861 +/- 0.16
G6 = -0.496 +/- 0.16
G7 = -0.260 +/- 0.13
G8 = -0.123 +/- 0.10
G9 = 0.0
G10 = 0.0
H0 = 1.0
H1 = -1.522 +/- 0.10
H2 = 0.613 +/- 0.10
H5 = 0.0
H6 = 0.0
H7 = 0.0
H8 = 0.0
H9 = 0.0
H10 = 0.0
The objective function value computed is 16.9 in twelve (12) iterations of SECQPE. The
reported standard-error of the residuals is 16.9/296 = 0.0571 which approximates the standard-
deviation of the (hopefully) white-noise residuals of time-series A. The absolute values for H1
and H2 are almost identical to those found in Box and Jenkins (1976) which is consistent with
the fact that they also used an auto-regressive (AR) noise model. In addition, no significant
auto-correlation of the residuals (our time-series A) was detected confirming that the estimation
should be unbiased.
The static gain (i.e., the first-order partial derivative of how U1 affects Y) using the truncated
impulse response G is -2.941 +/- 1.02 which is close to the steady-state gain reported in TSE-
GFD-IMF of (-0.53-0.37-0.51)/(1-0.57-0.01) = -3.357 by setting the backwards shift operator (z^-
1) to unity (1.0) where the dead-times are identical to three (3) time-periods. Although there is
over a 10% difference between the two static gain estimates, this is not uncommon when fitting
its value from passive or happenstance data which may include some form of feedback (closed-
loop interactions) as opposed to a well-designed, open/closed-loop PRBS/GBNS input/dither
signal (DSDP-CLE-IMF).
In summary, we have highlighted the application of finite impulse response estimation (FIRE)
using the industrial gas furnace data (Series J) from Box and Jenkins (1976) for both the
deterministic and stochastic terms. The model was formulated in IMPL and solved successfully
using its SECQPE and SORVE and can also be used to estimate static gains of the system
which would be useful in further steady-state process optimization (active) and/or process
monitoring (passive) applications.
References
Box, G.E.P., Jenkins, G.M., “Time-series analysis: forecasting and control”, revised edition,
Holden Day, Oakland, CA, 389-400 and Series J. (1976).
MacGregor, J.F., and Fogal, D.T., “Closed-loop identification: the role of the noise model and
prefilters”, Journal of Process Control, 5, 163, 171, (1995).
Shreesha, C., Gudi, R.D., “Analysis of pre-filter based closed-loop control-relevant identification
methodologies”, Canadian Journal of Chemical Engineering, 82, (2004).
Kelly, J.D., "Production modeling for multimodal operations", Chemical Engineering Progress,
February, 44, (2004).
Kelly, J.D., "The unit-operation-stock superstructure (UOSS) and the quantity-logic-quality
paradigm (QLQP) for production scheduling in the process industries", In: MISTA 2005
Conference Proceedings, 327, (2005).
Qin, J.S., “An overview of subspace identification”, Computers and Chemical Engineering,
1502-1513, (2006).
Kelly, J.D., Zyngier, D., "A new and improved MILP formulation to optimize observability,
redundancy and precision for sensor network problems", American Institute of Chemical
Engineering Journal, 54, 1282, (2008).
Schoukens, J., Rolain, Y., Vandersteen, G., Pintelon, R., “User friendly Box-Jenkins
identification using nonparametric noise models, 50th
IEEE Conference on Decision and Control
European Control Conference (CDC-ECC), Orlando, Florida, USA, December, (2011).
IAL, “Time series estimation of gas furnace data industrial modeling framework (TSE-GFD-
IMF)”, Slideshare, August, 2014.
IAL, “Advanced process control (APC) industrial modeling framework in the Julia programming
language (APC-IMF-Julia)”, Slideshare, October, 2014.
IAL, “Dither signal design problem for closed-loop estimation industrial modeling framework
(DSDP-CLE-IMF)”, Slideshare, December, 2014.
Appendix A – FIRE-GFD-IMF.IML File
i M P l (c)
Copyright and Property of i n d u s t r I A L g o r i t h m s LLC.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Calculation Data (Parameters)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&sCalc,@sValue
START,0.0
BEGIN,7.0
END,296.0
PERIOD,1.0
SE,1.0 !0.0571 != 16.9/296
LRGBND,1d+2
gbnd,1d+2
hbnd,1d+2
&sCalc,@sValue
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Chronological Data (Periods)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
@rPastTHD,@rFutureTHD,@rTPD
START,END,PERIOD
@rPastTHD,@rFutureTHD,@rTPD
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Constant Data (Parameters)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&sData,@sValue
u1,-0.052
,0.057
,0.235
,0.396
,0.43
,0.498
,0.518
,0.405
,0.184
,-0.123
,-0.531
,-0.998
,-1.364
,-1.463
,-1.245
,-0.757
,-0.418
,-0.136
,0.145
,0.492
,0.828
,0.923
,0.932
,0.948
,1.044
,1.32
,1.832
,2.033
,1.991
,1.923
,1.889
,1.824
,1.665
,1.322
,0.847
,0.417
,0.172
,0.145
,0.388
,0.702
,1.017
,1.466
,2.727
,2.891
,2.869
,2.54
,1.986
,1.542
,1.271
,1.296
,1.665
,1.962
,2.08
,1.872
,0.592
,0.179
,0.066
,0.221
,0.728
,1.076
,1.203
,1.212
,1.169
,1.178
,1.28
,1.314
,1.214
,0.97
,0.677
,0.312
,-0.223
,-1.023
,-1.494
,-1.742
,-1.768
,-1.399
,-0.887
,-0.513
,-0.374
,-0.52
,-0.903
,-1.559
,-1.818
,-1.834
,-1.689
,-1.417
,-1.144
,-0.87
,-0.467
,0.097
,0.845
,1
,0.987
,1.063
,1.194
,1.255
,1.111
,0.652
,-0.023
,-0.257
,-0.231
,-0.096
,-0.052
,-0.13
,-0.198
,-0.172
,0.05
,0.311
,0.387
,0.159
,-0.366
,-1.082
,-2.218
,-2.537
,-2.659
,-2.453
,-1.733
,-1.289
,-1.024
,-0.853
,-0.819
,-0.828
,-0.743
,-0.487
,-0.359
,-0.214
,0.057
,0.46
,0.898
,1.342
,1.664
,1.803
,1.74
,1.542
,1.05
,0.705
,0.634
,0.634
,0.689
,0.804
,0.957
,1.05
,1.025
,0.847
,0.456
,-0.104
,-0.496
,-0.546
,-0.367
,-0.137
,0.008
,0.117
,0.218
,0.358
,0.574
,0.623
,0.617
,0.63
,0.649
,0.728
,0.99
,1.394
,1.517
,1.41
,0.829
,0.275
,-0.18
,-0.657
,-1.042
,-1.212
,-1.118
,-0.619
,0.09
,0.613
,0.7
,0.541
,0.166
,-0.253
,-0.64
,-0.99
,-1.161
,-1.126
,-0.816
,-0.279
,0.12
,0.141
,0.057
,0.058
,0.266
,0.613
,0.839
,0.915
,0.975
,0.919
,0.473
,-0.279
,-0.902
,-1.756
,-2.321
,-2.442
,-2.416
,-2.273
,-1.996
,-1.682
,-1.204
,-0.512
,-0.08
,0.033
,0.007
,-0.078
,-0.219
,-0.477
,-0.814
,-1.186
,-1.382
,-1.365
,-1.118
,-0.756
,-0.577
,-0.525
,-0.568
,-0.656
,-0.791
,-0.982
,-1.289
,-1.571
,-1.562
,-1.092
,-0.431
,-0.103
,0.05
,-0.035
,-0.563
,-1.029
,-1.468
,-1.801
,-1.972
,-1.967
,-1.904
,-1.895
,-1.737
,-1.245
,-0.973
,-0.861
,-0.741
,-0.81
,-0.99
,-1.066
,-0.819
,-0.338
,0.242
,0.719
,0.766
,0.662
,0.558
,0.66
,1
,1.28
,1.306
,0.881
,0.159
,0.082
,0.439
,0.979
,1.089
,0.923
,0.584
,0.15
,-0.401
,-0.691
,-0.89
,-0.972
,-0.871
,-0.588
,-0.367
,-0.219
,-0.101
,0.024
,0.159
,0.308
,0.337
,0.057
,-0.436
,-0.702
,-0.767
,-0.683
,-0.471
,-0.147
,0.091
,0.261
,0.31
,0.252
,0.188
,0.074
,-0.125
,-0.205
y,0.29
,0.09
,-0.01
,-0.01
,-0.11
,-0.41
,-0.81
,-1.11
,-1.31
,-1.51
,-1.51
,-1.11
,-0.51
,0.49
,1.39
,2.49
,3.29
,3.29
,2.89
,2.19
,1.49
,0.79
,-0.31
,-1.21
,-1.91
,-2.31
,-2.71
,-3.01
,-3.51
,-4.31
,-5.11
,-5.61
,-5.91
,-6.01
,-6.01
,-5.91
,-5.41
,-4.51
,-3.51
,-2.41
,-1.71
,-1.61
,-1.81
,-2.31
,-3.51
,-5.21
,-6.51
,-7.71
,-7.91
,-7.51
,-6.61
,-5.71
,-5.31
,-5.21
,-5.61
,-6.31
,-6.31
,-5.41
,-4.11
,-2.91
,-2.01
,-1.91
,-2.31
,-3.01
,-3.41
,-3.71
,-3.91
,-4.11
,-4.21
,-4.31
,-4.21
,-3.81
,-3.21
,-2.21
,-0.71
,0.89
,2.49
,3.39
,3.99
,3.79
,3.09
,2.49
,1.89
,1.89
,2.89
,3.69
,4.49
,4.89
,4.89
,4.59
,4.19
,3.49
,2.49
,1.19
,-0.31
,-1.41
,-1.91
,-2.51
,-3.01
,-3.11
,-2.51
,-1.71
,-1.11
,-0.51
,-0.11
,0.09
,0.19
,0.29
,0.29
,0.29
,-0.21
,-0.51
,-0.61
,-0.11
,1.09
,2.89
,4.49
,5.89
,6.69
,6.49
,5.89
,4.89
,4.09
,3.39
,2.89
,2.49
,2.19
,1.79
,1.49
,0.89
,0.19
,-0.71
,-1.91
,-2.91
,-4.11
,-4.71
,-5.01
,-4.81
,-4.31
,-3.71
,-3.11
,-2.81
,-2.61
,-2.81
,-3.01
,-3.11
,-3.31
,-3.11
,-2.31
,-1.21
,-0.31
,0.39
,0.59
,0.49
,0.09
,-0.31
,-0.51
,-0.71
,-1.21
,-1.61
,-1.91
,-1.91
,-2.11
,-2.31
,-2.81
,-3.51
,-4.11
,-4.21
,-3.81
,-2.91
,-1.71
,-0.51
,0.49
,1.79
,2.39
,2.39
,1.09
,-0.01
,-1.11
,-1.41
,-1.21
,-0.51
,0.29
,1.09
,1.89
,2.39
,2.39
,1.69
,0.89
,0.19
,0.09
,0.09
,-0.31
,-1.01
,-1.51
,-2.11
,-2.51
,-2.61
,-1.11
,-0.01
,2.09
,4.49
,5.99
,6.49
,6.89
,6.99
,6.69
,6.19
,5.49
,4.09
,2.89
,1.69
,0.99
,0.59
,0.59
,0.89
,1.99
,2.69
,3.49
,3.79
,3.89
,3.49
,2.89
,2.39
,1.99
,1.79
,1.69
,1.89
,2.49
,2.99
,3.59
,3.79
,3.29
,2.09
,1.49
,0.59
,0.79
,1.79
,2.89
,3.69
,4.29
,4.79
,5.09
,5.29
,5.29
,5.09
,4.49
,3.89
,3.49
,2.89
,2.79
,2.89
,2.89
,2.49
,1.69
,0.49
,-0.51
,-1.51
,-1.91
,-1.91
,-2.41
,-3.11
,-3.51
,-3.51
,-1.51
,0.49
,1.59
,0.99
,-0.71
,-2.11
,-2.71
,-2.31
,-1.51
,-0.71
,0.29
,0.99
,1.39
,1.39
,1.29
,0.89
,0.19
,-0.21
,-0.71
,-0.91
,-0.91
,-0.51
,0.79
,2.49
,3.49
,4.49
,5.09
,4.99
,4.79
,4.29
,3.79
,3.49
&sData,@sValue
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Construction Data (Pointers)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&sUnit,&sOperation,@sType,@sSubtype,@sUse
BLACKBOX,,processc,blackbox,,
&sUnit,&sOperation,@sType,@sSubtype,@sUse
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Condition Data (Properties)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&sCondition
Y
U1
X1
Z
A
&sCondition
&sCoefficient,@sType,@sPath_Name,@sLibrary_Name,@sFunction_Name,@iNumber_Conditions,@rPerturb_Size,@sCondition_Names
g0,static
g1,static
g2,static
g3,static
g4,static
g5,static
g6,static
g7,static
g8,static
g9,static
g10,static
h0,static
h1,static
h2,static
h3,static
h4,static
h5,static
h6,static
h7,static
h8,static
h9,static
h10,static
&sCoefficient,@sType,@sPath_Name,@sLibrary_Name,@sFunction_Name,@iNumber_Conditions,@rPerturb_Size,@sCondition_Names
&sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target
BLACKBOX,,Y,-LRGBND,LRGBND,
BLACKBOX,,U1,-LRGBND,LRGBND,
BLACKBOX,,X1,-LRGBND,LRGBND,
BLACKBOX,,Z,-LRGBND,LRGBND,
BLACKBOX,,A,-LRGBND,LRGBND,
&sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target
&sUnit,&sOperation,&sCoefficient,@rCoefficient_Lower,@rCoefficient_Upper,@rCoefficient_Target
BLACKBOX,,g0, 0.0,0.0,
BLACKBOX,,g1, -0*gbnd,0*gbnd,
BLACKBOX,,g2, -0*gbnd,0*gbnd,
BLACKBOX,,g3, -gbnd,gbnd,
BLACKBOX,,g4, -gbnd,gbnd,
BLACKBOX,,g5, -gbnd,gbnd,
BLACKBOX,,g6, -gbnd,gbnd,
BLACKBOX,,g7, -gbnd,gbnd,
BLACKBOX,,g8, -gbnd,gbnd,
BLACKBOX,,g9, -0*gbnd,0*gbnd,
BLACKBOX,,g10,-0*gbnd,0*gbnd,
BLACKBOX,,h0, 1.0,1.0,
BLACKBOX,,h1, -hbnd,hbnd,
BLACKBOX,,h2, -hbnd,hbnd,
BLACKBOX,,h3, -0*hbnd,0*hbnd,
BLACKBOX,,h4, -0*hbnd,0*hbnd,
BLACKBOX,,h5, -0*hbnd,0*hbnd,
BLACKBOX,,h6, -0*hbnd,0*hbnd,
BLACKBOX,,h7, -0*hbnd,0*hbnd,
BLACKBOX,,h8, -0*hbnd,0*hbnd,
BLACKBOX,,h9, -0*hbnd,0*hbnd,
BLACKBOX,,h10,-0*hbnd,0*hbnd,
&sUnit,&sOperation,&sCoefficient,@rCoefficient_Lower,@rCoefficient_Upper,@rCoefficient_Target
Conditions-&sMacro,@sValue
X1, g0*U1[0]+g1*U1[1]+g2*U1[2]+g3*U1[3]+g4*U1[4]+g5*U1[5]+g6*U1[6]+g7*U1[7]+g8*U1[8]+g9*U1[9]+g10*U1[10]
Z, Y - X1
A, h0*Z[0]+h1*Z[1]+h2*Z[2]+h3*Z[3]+h4*Z[4]+h5*Z[5]+h6*Z[6]+h7*Z[7]+h8*Z[8]+h9*Z[9]+h10*Z[10]
Conditions-&sMacro,@sValue
ConditionsUOCondition-&sUnit,&sOperation,&sCondition,@sType,@rValue,@sValue
BLACKBOX,,X1,?,3,X1
BLACKBOX,,Z,?,3,Z
BLACKBOX,,A,?,3,A
ConditionsUOCondition-&sUnit,&sOperation,&sCondition,@sType,@rValue,@sValue
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Cost Data (Pricing)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&sUnit,&sOperation,&sCondition,@rConditionPro_Weight,@rConditionPer1_Weight,@rConditionPer2_Weight,@rConditionPen_Weight
BLACKBOX,,A,,,1.0/SE,
&sUnit,&sOperation,&sCondition,@rConditionPro_Weight,@rConditionPer1_Weight,@rConditionPer2_Weight,@rConditionPen_Weight
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Command Data (Future Provisos)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&sUnit,&sOperation,@rSetup_Lower,@rSetup_Upper,@rBegin_Time,@rEnd_Time
BLACKBOX,,1,1,START,END
&sUnit,&sOperation,@rSetup_Lower,@rSetup_Upper,@rBegin_Time,@rEnd_Time
&sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target,@rBegin_Time,@rEnd_Time
BLACKBOX,,U1,u1,u1,,START,PERIOD
BLACKBOX,,Y,y,y,,START,PERIOD
BLACKBOX,,A,-1000,1000,,START,BEGIN
,,A,-1000,1000,0.0,BEGIN,END
&sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target,@rBegin_Time,@rEnd_Time

More Related Content

What's hot

The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...
The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...
The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...
IJERA Editor
 
Rule Evaluation on a Motorola SIMD
Rule Evaluation on a Motorola SIMDRule Evaluation on a Motorola SIMD
Rule Evaluation on a Motorola SIMD
Meltin Bell
 
Capital Investment Industrial Modeling Framework - IMPRESS
Capital Investment Industrial Modeling Framework - IMPRESSCapital Investment Industrial Modeling Framework - IMPRESS
Capital Investment Industrial Modeling Framework - IMPRESSAlkis Vazacopoulos
 
Bt0066 database management system2
Bt0066 database management system2Bt0066 database management system2
Bt0066 database management system2
Techglyphs
 
Lesson 4 ar-ma
Lesson 4 ar-maLesson 4 ar-ma
Lesson 4 ar-ma
ankit_ppt
 
Basic Coding In VHDL COding
Basic Coding In VHDL COdingBasic Coding In VHDL COding
Basic Coding In VHDL COding
anna university
 
Maritime Industrial Modeling Framework - IMPRESS
Maritime Industrial Modeling Framework - IMPRESSMaritime Industrial Modeling Framework - IMPRESS
Maritime Industrial Modeling Framework - IMPRESSAlkis Vazacopoulos
 
Mini Project- ROM Based Sine Wave Generator
Mini Project- ROM Based Sine Wave GeneratorMini Project- ROM Based Sine Wave Generator
A novel methodology for test scenario generation based on control flow analys...
A novel methodology for test scenario generation based on control flow analys...A novel methodology for test scenario generation based on control flow analys...
A novel methodology for test scenario generation based on control flow analys...
eSAT Publishing House
 
Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...
Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...
Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...
IRJET Journal
 
Programmable logic array
Programmable logic arrayProgrammable logic array
Programmable logic array
Huba Akhtar
 
Pattern Matching using Computational and Automata Theory
Pattern Matching using Computational and Automata TheoryPattern Matching using Computational and Automata Theory
Pattern Matching using Computational and Automata Theory
IRJET Journal
 
5 the relational algebra and calculus
5 the relational algebra and calculus5 the relational algebra and calculus
5 the relational algebra and calculusKumar
 
Algorithm manual
Algorithm manualAlgorithm manual
Algorithm manual
EfendyMasuli
 
On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...
On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...
On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...Amir Ziai
 
An Introduction to the SOLID Principles
An Introduction to the SOLID PrinciplesAn Introduction to the SOLID Principles
An Introduction to the SOLID Principles
Attila Bertók
 

What's hot (20)

The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...
The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...
The Haar-Recursive Transform and Its Consequence to the Walsh-Paley Spectrum ...
 
Rule Evaluation on a Motorola SIMD
Rule Evaluation on a Motorola SIMDRule Evaluation on a Motorola SIMD
Rule Evaluation on a Motorola SIMD
 
AR model
AR modelAR model
AR model
 
Capital Investment Industrial Modeling Framework - IMPRESS
Capital Investment Industrial Modeling Framework - IMPRESSCapital Investment Industrial Modeling Framework - IMPRESS
Capital Investment Industrial Modeling Framework - IMPRESS
 
Bt0066 database management system2
Bt0066 database management system2Bt0066 database management system2
Bt0066 database management system2
 
Lesson 4 ar-ma
Lesson 4 ar-maLesson 4 ar-ma
Lesson 4 ar-ma
 
Basic Coding In VHDL COding
Basic Coding In VHDL COdingBasic Coding In VHDL COding
Basic Coding In VHDL COding
 
Lecture03(c expressions & operators)
Lecture03(c expressions & operators)Lecture03(c expressions & operators)
Lecture03(c expressions & operators)
 
Maritime Industrial Modeling Framework - IMPRESS
Maritime Industrial Modeling Framework - IMPRESSMaritime Industrial Modeling Framework - IMPRESS
Maritime Industrial Modeling Framework - IMPRESS
 
Mini Project- ROM Based Sine Wave Generator
Mini Project- ROM Based Sine Wave GeneratorMini Project- ROM Based Sine Wave Generator
Mini Project- ROM Based Sine Wave Generator
 
SECh910
SECh910SECh910
SECh910
 
A novel methodology for test scenario generation based on control flow analys...
A novel methodology for test scenario generation based on control flow analys...A novel methodology for test scenario generation based on control flow analys...
A novel methodology for test scenario generation based on control flow analys...
 
Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...
Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...
Performance Analysis, Designing and Testing 512 Bit Sram Memory Chip Using Xi...
 
Ch10
Ch10Ch10
Ch10
 
Programmable logic array
Programmable logic arrayProgrammable logic array
Programmable logic array
 
Pattern Matching using Computational and Automata Theory
Pattern Matching using Computational and Automata TheoryPattern Matching using Computational and Automata Theory
Pattern Matching using Computational and Automata Theory
 
5 the relational algebra and calculus
5 the relational algebra and calculus5 the relational algebra and calculus
5 the relational algebra and calculus
 
Algorithm manual
Algorithm manualAlgorithm manual
Algorithm manual
 
On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...
On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...
On the Performance of the Pareto Set Pursuing (PSP) Method for Mixed-Variable...
 
An Introduction to the SOLID Principles
An Introduction to the SOLID PrinciplesAn Introduction to the SOLID Principles
An Introduction to the SOLID Principles
 

Viewers also liked

The basics of advertising online
The basics of advertising onlineThe basics of advertising online
The basics of advertising onlinesglte
 
Faq onlinestudents fa10
Faq onlinestudents fa10Faq onlinestudents fa10
Faq onlinestudents fa10
Suse1
 
Cassie Hilaire
Cassie HilaireCassie Hilaire
Cassie Hilairecassieh939
 
Argentina powerpoint1
Argentina powerpoint1Argentina powerpoint1
Argentina powerpoint1
Jenifer Carter
 
Mas
MasMas
Social media presentation from aravind
Social media presentation from aravindSocial media presentation from aravind
Social media presentation from aravind
Aravind Kotagiri
 
Animalsoundgame
AnimalsoundgameAnimalsoundgame
Animalsoundgame
Deegishuree
 
peer division multiplexing
peer division multiplexingpeer division multiplexing
peer division multiplexingajayj251
 
Q6
Q6Q6
мосты
мостымосты
мосты
Andzhey Arshavskiy
 
Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3
Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3
Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3
08600 Vasilkov
 
Hoi dap ve quan ly moi truong
Hoi dap ve quan ly moi truongHoi dap ve quan ly moi truong
Hoi dap ve quan ly moi truong
lethao1491
 
Development of Communicative Competence Analysis
Development of Communicative Competence AnalysisDevelopment of Communicative Competence Analysis
Development of Communicative Competence Analysis
Joan Cahuich Perez
 
Impl installation manual
Impl installation manualImpl installation manual
Impl installation manual
Alkis Vazacopoulos
 
Alfabetización tecnológica, acceso y uso de internet
Alfabetización tecnológica, acceso y uso de internetAlfabetización tecnológica, acceso y uso de internet
Alfabetización tecnológica, acceso y uso de internet
Joan Cahuich Perez
 
Professionalism
ProfessionalismProfessionalism
ProfessionalismAlicecen
 

Viewers also liked (20)

The basics of advertising online
The basics of advertising onlineThe basics of advertising online
The basics of advertising online
 
Proiect de lege PLx 698/11
Proiect de lege PLx 698/11Proiect de lege PLx 698/11
Proiect de lege PLx 698/11
 
Faq onlinestudents fa10
Faq onlinestudents fa10Faq onlinestudents fa10
Faq onlinestudents fa10
 
Cassie Hilaire
Cassie HilaireCassie Hilaire
Cassie Hilaire
 
Argentina powerpoint1
Argentina powerpoint1Argentina powerpoint1
Argentina powerpoint1
 
Mas
MasMas
Mas
 
Social media presentation from aravind
Social media presentation from aravindSocial media presentation from aravind
Social media presentation from aravind
 
Animalsoundgame
AnimalsoundgameAnimalsoundgame
Animalsoundgame
 
peer division multiplexing
peer division multiplexingpeer division multiplexing
peer division multiplexing
 
Q6
Q6Q6
Q6
 
 
Violencedomestique
ViolencedomestiqueViolencedomestique
Violencedomestique
 
мосты
мостымосты
мосты
 
Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3
Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3
Презентація команди Васильківської ЗОШ І-ІІІ ступенів №3
 
Hoi dap ve quan ly moi truong
Hoi dap ve quan ly moi truongHoi dap ve quan ly moi truong
Hoi dap ve quan ly moi truong
 
Development of Communicative Competence Analysis
Development of Communicative Competence AnalysisDevelopment of Communicative Competence Analysis
Development of Communicative Competence Analysis
 
Impl installation manual
Impl installation manualImpl installation manual
Impl installation manual
 
Alfabetización tecnológica, acceso y uso de internet
Alfabetización tecnológica, acceso y uso de internetAlfabetización tecnológica, acceso y uso de internet
Alfabetización tecnológica, acceso y uso de internet
 
Professionalism
ProfessionalismProfessionalism
Professionalism
 
統計第8回
統計第8回統計第8回
統計第8回
 

Similar to Finite Impulse Response Estimation of Gas Furnace Data in IMPL Industrial Modeling Framework (FIRE-GFD-IMF)

Advanced Production Accounting of a Flotation Plant
Advanced Production Accounting of a Flotation PlantAdvanced Production Accounting of a Flotation Plant
Advanced Production Accounting of a Flotation PlantAlkis Vazacopoulos
 
Advanced Production Accounting
Advanced Production AccountingAdvanced Production Accounting
Advanced Production AccountingAlkis Vazacopoulos
 
Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...
Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...
Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...
Alkis Vazacopoulos
 
Planuling & Phasing
Planuling & PhasingPlanuling & Phasing
Planuling & Phasing
Alkis Vazacopoulos
 
Advanced Production Control Using Julia & IMPL
Advanced Production Control Using Julia & IMPLAdvanced Production Control Using Julia & IMPL
Advanced Production Control Using Julia & IMPL
Alkis Vazacopoulos
 
Impl reference manual_for_quantities
Impl reference manual_for_quantitiesImpl reference manual_for_quantities
Impl reference manual_for_quantities
Alkis Vazacopoulos
 
A study of the Behavior of Floating-Point Errors
A study of the Behavior of Floating-Point ErrorsA study of the Behavior of Floating-Point Errors
A study of the Behavior of Floating-Point Errors
ijpla
 
Impl reference manual_for_logic_logistics
Impl reference manual_for_logic_logisticsImpl reference manual_for_logic_logistics
Impl reference manual_for_logic_logistics
Alkis Vazacopoulos
 
Ial impl-imf-book-1-0
Ial impl-imf-book-1-0Ial impl-imf-book-1-0
Ial impl-imf-book-1-0
Alkis Vazacopoulos
 
IRJET- Design of Photovoltaic System using Fuzzy Logic Controller
IRJET- Design of Photovoltaic System using Fuzzy Logic ControllerIRJET- Design of Photovoltaic System using Fuzzy Logic Controller
IRJET- Design of Photovoltaic System using Fuzzy Logic Controller
IRJET Journal
 
Pipeline optimization Industrial Modeling Framework
Pipeline optimization Industrial Modeling FrameworkPipeline optimization Industrial Modeling Framework
Pipeline optimization Industrial Modeling Framework
Alkis Vazacopoulos
 
IMPL Data Analysis
IMPL Data AnalysisIMPL Data Analysis
IMPL Data Analysis
Alkis Vazacopoulos
 
An Approach To Verilog-VHDL Interoperability For Synchronous Designs
An Approach To Verilog-VHDL Interoperability For Synchronous DesignsAn Approach To Verilog-VHDL Interoperability For Synchronous Designs
An Approach To Verilog-VHDL Interoperability For Synchronous Designs
Dawn Cook
 
Gene's law
Gene's lawGene's law
Gene's law
Hoopeer Hoopeer
 
E121 gt2014 26029
E121 gt2014 26029E121 gt2014 26029
E121 gt2014 26029
Ivan Gonzalez Castillo
 
Design of multiloop controller for multivariable system using coefficient 2
Design of multiloop controller for multivariable system using coefficient 2Design of multiloop controller for multivariable system using coefficient 2
Design of multiloop controller for multivariable system using coefficient 2IAEME Publication
 
FUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINES
FUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINESFUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINES
FUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINES
IAEME Publication
 
IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...
IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...
IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...
IRJET Journal
 
Enhance interval width of crime forecasting with ARIMA model-fuzzy alpha cut
Enhance interval width of crime forecasting with ARIMA model-fuzzy alpha cutEnhance interval width of crime forecasting with ARIMA model-fuzzy alpha cut
Enhance interval width of crime forecasting with ARIMA model-fuzzy alpha cut
TELKOMNIKA JOURNAL
 
Speed Control of DC Motor using PID FUZZY Controller.
Speed Control of DC Motor using PID FUZZY Controller.Speed Control of DC Motor using PID FUZZY Controller.
Speed Control of DC Motor using PID FUZZY Controller.
Binod kafle
 

Similar to Finite Impulse Response Estimation of Gas Furnace Data in IMPL Industrial Modeling Framework (FIRE-GFD-IMF) (20)

Advanced Production Accounting of a Flotation Plant
Advanced Production Accounting of a Flotation PlantAdvanced Production Accounting of a Flotation Plant
Advanced Production Accounting of a Flotation Plant
 
Advanced Production Accounting
Advanced Production AccountingAdvanced Production Accounting
Advanced Production Accounting
 
Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...
Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...
Advanced Process Monitoring for Startups, Shutdowns & Switchovers Industrial ...
 
Planuling & Phasing
Planuling & PhasingPlanuling & Phasing
Planuling & Phasing
 
Advanced Production Control Using Julia & IMPL
Advanced Production Control Using Julia & IMPLAdvanced Production Control Using Julia & IMPL
Advanced Production Control Using Julia & IMPL
 
Impl reference manual_for_quantities
Impl reference manual_for_quantitiesImpl reference manual_for_quantities
Impl reference manual_for_quantities
 
A study of the Behavior of Floating-Point Errors
A study of the Behavior of Floating-Point ErrorsA study of the Behavior of Floating-Point Errors
A study of the Behavior of Floating-Point Errors
 
Impl reference manual_for_logic_logistics
Impl reference manual_for_logic_logisticsImpl reference manual_for_logic_logistics
Impl reference manual_for_logic_logistics
 
Ial impl-imf-book-1-0
Ial impl-imf-book-1-0Ial impl-imf-book-1-0
Ial impl-imf-book-1-0
 
IRJET- Design of Photovoltaic System using Fuzzy Logic Controller
IRJET- Design of Photovoltaic System using Fuzzy Logic ControllerIRJET- Design of Photovoltaic System using Fuzzy Logic Controller
IRJET- Design of Photovoltaic System using Fuzzy Logic Controller
 
Pipeline optimization Industrial Modeling Framework
Pipeline optimization Industrial Modeling FrameworkPipeline optimization Industrial Modeling Framework
Pipeline optimization Industrial Modeling Framework
 
IMPL Data Analysis
IMPL Data AnalysisIMPL Data Analysis
IMPL Data Analysis
 
An Approach To Verilog-VHDL Interoperability For Synchronous Designs
An Approach To Verilog-VHDL Interoperability For Synchronous DesignsAn Approach To Verilog-VHDL Interoperability For Synchronous Designs
An Approach To Verilog-VHDL Interoperability For Synchronous Designs
 
Gene's law
Gene's lawGene's law
Gene's law
 
E121 gt2014 26029
E121 gt2014 26029E121 gt2014 26029
E121 gt2014 26029
 
Design of multiloop controller for multivariable system using coefficient 2
Design of multiloop controller for multivariable system using coefficient 2Design of multiloop controller for multivariable system using coefficient 2
Design of multiloop controller for multivariable system using coefficient 2
 
FUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINES
FUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINESFUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINES
FUZZY LOGIC CONTROL DESIGN FOR ELECTRICAL MACHINES
 
IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...
IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...
IRJET- A Comparative Forecasting Analysis of ARIMA Model Vs Random Forest Alg...
 
Enhance interval width of crime forecasting with ARIMA model-fuzzy alpha cut
Enhance interval width of crime forecasting with ARIMA model-fuzzy alpha cutEnhance interval width of crime forecasting with ARIMA model-fuzzy alpha cut
Enhance interval width of crime forecasting with ARIMA model-fuzzy alpha cut
 
Speed Control of DC Motor using PID FUZZY Controller.
Speed Control of DC Motor using PID FUZZY Controller.Speed Control of DC Motor using PID FUZZY Controller.
Speed Control of DC Motor using PID FUZZY Controller.
 

More from Alkis Vazacopoulos

Automatic Fine-tuning Xpress-MP to Solve MIP
Automatic Fine-tuning Xpress-MP to Solve MIPAutomatic Fine-tuning Xpress-MP to Solve MIP
Automatic Fine-tuning Xpress-MP to Solve MIP
Alkis Vazacopoulos
 
Data mining 2004
Data mining 2004Data mining 2004
Data mining 2004
Alkis Vazacopoulos
 
Amazing results with ODH|CPLEX
Amazing results with ODH|CPLEXAmazing results with ODH|CPLEX
Amazing results with ODH|CPLEX
Alkis Vazacopoulos
 
Bia project poster fantasy football
Bia project poster  fantasy football Bia project poster  fantasy football
Bia project poster fantasy football
Alkis Vazacopoulos
 
NFL Game schedule optimization
NFL Game schedule optimization NFL Game schedule optimization
NFL Game schedule optimization
Alkis Vazacopoulos
 
2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...
2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...
2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...
Alkis Vazacopoulos
 
Posters 2017
Posters 2017Posters 2017
Posters 2017
Alkis Vazacopoulos
 
Very largeoptimizationparallel
Very largeoptimizationparallelVery largeoptimizationparallel
Very largeoptimizationparallel
Alkis Vazacopoulos
 
Retail Pricing Optimization
Retail Pricing Optimization Retail Pricing Optimization
Retail Pricing Optimization
Alkis Vazacopoulos
 
Optimization Direct: Introduction and recent case studies
Optimization Direct: Introduction and recent case studiesOptimization Direct: Introduction and recent case studies
Optimization Direct: Introduction and recent case studies
Alkis Vazacopoulos
 
Informs 2016 Solving Planning and Scheduling Problems with CPLEX
Informs 2016 Solving Planning and Scheduling Problems with CPLEX Informs 2016 Solving Planning and Scheduling Problems with CPLEX
Informs 2016 Solving Planning and Scheduling Problems with CPLEX
Alkis Vazacopoulos
 
ODHeuristics
ODHeuristicsODHeuristics
ODHeuristics
Alkis Vazacopoulos
 
Missing-Value Handling in Dynamic Model Estimation using IMPL
Missing-Value Handling in Dynamic Model Estimation using IMPL Missing-Value Handling in Dynamic Model Estimation using IMPL
Missing-Value Handling in Dynamic Model Estimation using IMPL
Alkis Vazacopoulos
 
Industrial Modeling Service (IMS-IMPL)
Industrial Modeling Service (IMS-IMPL)Industrial Modeling Service (IMS-IMPL)
Industrial Modeling Service (IMS-IMPL)
Alkis Vazacopoulos
 
Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...
Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...
Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...Alkis Vazacopoulos
 
Xmr im
Xmr imXmr im
Distillation Curve Optimization Using Monotonic Interpolation
Distillation Curve Optimization Using Monotonic InterpolationDistillation Curve Optimization Using Monotonic Interpolation
Distillation Curve Optimization Using Monotonic Interpolation
Alkis Vazacopoulos
 
Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...
Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...
Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...
Alkis Vazacopoulos
 
Hybrid Dynamic Simulation (HDS) Industrial Modeling Framework (HDS-IMF)
Hybrid Dynamic Simulation (HDS)  Industrial Modeling Framework (HDS-IMF)Hybrid Dynamic Simulation (HDS)  Industrial Modeling Framework (HDS-IMF)
Hybrid Dynamic Simulation (HDS) Industrial Modeling Framework (HDS-IMF)Alkis Vazacopoulos
 

More from Alkis Vazacopoulos (20)

Automatic Fine-tuning Xpress-MP to Solve MIP
Automatic Fine-tuning Xpress-MP to Solve MIPAutomatic Fine-tuning Xpress-MP to Solve MIP
Automatic Fine-tuning Xpress-MP to Solve MIP
 
Data mining 2004
Data mining 2004Data mining 2004
Data mining 2004
 
Amazing results with ODH|CPLEX
Amazing results with ODH|CPLEXAmazing results with ODH|CPLEX
Amazing results with ODH|CPLEX
 
Bia project poster fantasy football
Bia project poster  fantasy football Bia project poster  fantasy football
Bia project poster fantasy football
 
NFL Game schedule optimization
NFL Game schedule optimization NFL Game schedule optimization
NFL Game schedule optimization
 
2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...
2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...
2017 Business Intelligence & Analytics Corporate Event Stevens Institute of T...
 
Posters 2017
Posters 2017Posters 2017
Posters 2017
 
Very largeoptimizationparallel
Very largeoptimizationparallelVery largeoptimizationparallel
Very largeoptimizationparallel
 
Retail Pricing Optimization
Retail Pricing Optimization Retail Pricing Optimization
Retail Pricing Optimization
 
Optimization Direct: Introduction and recent case studies
Optimization Direct: Introduction and recent case studiesOptimization Direct: Introduction and recent case studies
Optimization Direct: Introduction and recent case studies
 
Informs 2016 Solving Planning and Scheduling Problems with CPLEX
Informs 2016 Solving Planning and Scheduling Problems with CPLEX Informs 2016 Solving Planning and Scheduling Problems with CPLEX
Informs 2016 Solving Planning and Scheduling Problems with CPLEX
 
ODHeuristics
ODHeuristicsODHeuristics
ODHeuristics
 
Missing-Value Handling in Dynamic Model Estimation using IMPL
Missing-Value Handling in Dynamic Model Estimation using IMPL Missing-Value Handling in Dynamic Model Estimation using IMPL
Missing-Value Handling in Dynamic Model Estimation using IMPL
 
Industrial Modeling Service (IMS-IMPL)
Industrial Modeling Service (IMS-IMPL)Industrial Modeling Service (IMS-IMPL)
Industrial Modeling Service (IMS-IMPL)
 
Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...
Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...
Dither Signal Design Problem (DSDP) for Closed-Loop Estimation Industrial Mod...
 
Xmr im
Xmr imXmr im
Xmr im
 
Distillation Curve Optimization Using Monotonic Interpolation
Distillation Curve Optimization Using Monotonic InterpolationDistillation Curve Optimization Using Monotonic Interpolation
Distillation Curve Optimization Using Monotonic Interpolation
 
Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...
Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...
Multi-Utility Scheduling Optimization (MUSO) Industrial Modeling Framework (M...
 
Hybrid Dynamic Simulation (HDS) Industrial Modeling Framework (HDS-IMF)
Hybrid Dynamic Simulation (HDS)  Industrial Modeling Framework (HDS-IMF)Hybrid Dynamic Simulation (HDS)  Industrial Modeling Framework (HDS-IMF)
Hybrid Dynamic Simulation (HDS) Industrial Modeling Framework (HDS-IMF)
 
Benefits of using IMPL
Benefits of using IMPLBenefits of using IMPL
Benefits of using IMPL
 

Recently uploaded

一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
ewymefz
 
一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
ewymefz
 
一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单
enxupq
 
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdfSample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Linda486226
 
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
u86oixdj
 
一比一原版(BU毕业证)波士顿大学毕业证成绩单
一比一原版(BU毕业证)波士顿大学毕业证成绩单一比一原版(BU毕业证)波士顿大学毕业证成绩单
一比一原版(BU毕业证)波士顿大学毕业证成绩单
ewymefz
 
做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样
做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样
做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样
axoqas
 
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
Tiktokethiodaily
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
nscud
 
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
vcaxypu
 
一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理
一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理
一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理
mbawufebxi
 
Machine learning and optimization techniques for electrical drives.pptx
Machine learning and optimization techniques for electrical drives.pptxMachine learning and optimization techniques for electrical drives.pptx
Machine learning and optimization techniques for electrical drives.pptx
balafet
 
一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单
一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单
一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单
ewymefz
 
一比一原版(CU毕业证)卡尔顿大学毕业证成绩单
一比一原版(CU毕业证)卡尔顿大学毕业证成绩单一比一原版(CU毕业证)卡尔顿大学毕业证成绩单
一比一原版(CU毕业证)卡尔顿大学毕业证成绩单
yhkoc
 
SOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape ReportSOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape Report
SOCRadar
 
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
NABLAS株式会社
 
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
ukgaet
 
Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)
TravisMalana
 
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
AbhimanyuSinha9
 
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
slg6lamcq
 

Recently uploaded (20)

一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
一比一原版(UPenn毕业证)宾夕法尼亚大学毕业证成绩单
 
一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
 
一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单
 
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdfSample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
 
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
 
一比一原版(BU毕业证)波士顿大学毕业证成绩单
一比一原版(BU毕业证)波士顿大学毕业证成绩单一比一原版(BU毕业证)波士顿大学毕业证成绩单
一比一原版(BU毕业证)波士顿大学毕业证成绩单
 
做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样
做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样
做(mqu毕业证书)麦考瑞大学毕业证硕士文凭证书学费发票原版一模一样
 
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
 
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
 
一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理
一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理
一比一原版(Bradford毕业证书)布拉德福德大学毕业证如何办理
 
Machine learning and optimization techniques for electrical drives.pptx
Machine learning and optimization techniques for electrical drives.pptxMachine learning and optimization techniques for electrical drives.pptx
Machine learning and optimization techniques for electrical drives.pptx
 
一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单
一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单
一比一原版(UofM毕业证)明尼苏达大学毕业证成绩单
 
一比一原版(CU毕业证)卡尔顿大学毕业证成绩单
一比一原版(CU毕业证)卡尔顿大学毕业证成绩单一比一原版(CU毕业证)卡尔顿大学毕业证成绩单
一比一原版(CU毕业证)卡尔顿大学毕业证成绩单
 
SOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape ReportSOCRadar Germany 2024 Threat Landscape Report
SOCRadar Germany 2024 Threat Landscape Report
 
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
 
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
 
Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)Malana- Gimlet Market Analysis (Portfolio 2)
Malana- Gimlet Market Analysis (Portfolio 2)
 
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...Best best suvichar in gujarati english meaning of this sentence as Silk road ...
Best best suvichar in gujarati english meaning of this sentence as Silk road ...
 
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
 

Finite Impulse Response Estimation of Gas Furnace Data in IMPL Industrial Modeling Framework (FIRE-GFD-IMF)

  • 1. Finite Impulse Response Estimation of Gas Furnace Data in IMPL Industrial Modeling Framework (FIRE-GFD-IMF) i n d u s t r IAL g o r i t h m s LLC. (IAL) www.industrialgorithms.com December 2014 Introduction to Finite Impulse Response Estimation, UOPSS and QLQP Presented in this short document is a description of how to estimate deterministic and stochastic non-parametric finite impulse response (FIR) models in IMPL applied to industrial gas furnace data identical to that found in TSE-GFD-IMF using parametric transfer-functions. The methodology of time-series analysis or system identification involves essentially three (3) stages (Box and Jenkins, 1976): (1) model structure identification, (2) model parameter estimation and (3) model checking and diagnostics. We do not address (1) which requires stationarity and seasonality assessment/adjustment, auto-, cross- and partial-correlation, etc. to establish the parametric transfer function polynomial degrees especially when we are using non-parametric FIR estimation. Instead we focus only on the parameter estimation and diagnostics. These types of parameter estimation problems involve dynamic and nonlinear relationships shown below and we solve these using IMPL’s Sequential Equality-Constrained QP Engine (SECQPE) and Supplemental Observability, Redundancy and Variability Estimator (SORVE). Other types of non-parametric identification known as Subspace Identification (Qin, 2006) and can used to estimate state-space models. Figure 1 shows the gas furnace data example found in Series J of Box and Jenkins (1976) where we depict the problem using signal processing constructs configured in our unit- operation-port-state superstructure (UOPSS) (Kelly, 2004, 2005; Zyngier and Kelly, 2012). Figure 1. Gas Furnace Data in a “Signal Processing” Based UOPSS Flowsheet. The diamond shapes or objects are the sources and sinks known as perimeters where U1 and Y are the input gas flowrate and CO2% in the effluent flue gas of the furnace respectively. The rectangular shapes with the cross-hairs are continuous-process units where the DeterministicModel (process) and StochasticModel (noise) are blackbox subtypes which allow any number of process or operating conditions (and coefficients) to be attached with ad hoc formulas or equations representing the non-parametric FIR. The Splicer is a signal processing shape to add and/or subtract the inlet signals together producing a single output i.e., Z = Y – X1
  • 2. in our case. The circles with and without cross-hairs are outlet and inlet port-states respectively. Port-states are unambiguous interfaces between up and downstream unit-operations. The deterministic FIR model in discrete-time or difference form (versus z-transform or backwards-shift operator-based quotients of rational polynomials) is defined as follows: X1,0 = G0*U1,0+G1*U1,1+G2*U1,2+G3*U1,3+G4*U1,4+G5*U1,5+G6*U1,6+G7*U1,7+G8*U1,8+G9*U1,9+G10*U1,10 where U1 is the exogenous input signal minus its mean of -0.057 at time-periods t-1, t-2, …, t- 10 and X1is the “deterministic state” at time-period t-0. The corresponding parameters or coefficients are G0, G1, …, G10 are the FIR values for each time-shift or lag in the past where G0 is typically set to zero (0.0) given discrete-time sampling and some of the initial coefficients G1, etc. will also be effectively zero (0.0) given the dead-time or inherent delay in the system. The static or steady-state gain of each input with respect to each output can be easily calculated from the dynamic gains or FIR’s by taking their sum. During the initial part of the estimation procedure, the number of FIR’s is typically set to some number greater than expected. Then, using the parameter variances and the Student-t statistics (parameter confidence-intervals) the actual number is reduced to hopefully avoid over-parameterization which is a well-known disadvantage of FIR’s and non-parametric models in general. Similarly, the stochastic FIR model or unmeasured noise disturbance model is also defined as follows (Schoukens et. al. 2011): A,0 = H0*Z,0+H1*Z,1+H2*Z,2+H3*Z,3+H4*Z,4+H5*Z,5+H6*Z,6+H7*Z,7+H8*Z,8+H9*Z,9+H10*Z,10 where A is the assumed white-noise input signal and Z is the “stochastic state” which is equal to Z = Y – X1 and the time-series Y is also minus its mean of 53.51. The parameters H0, H1, …, H10 and actually represent the “inverse” of stochastic noise model in non-parametric form. The noise model is essentially a time- or frequency-dependent weighting filter and is very important to ensure that the A-series is independent and identically distributed (MacGregor and Fogal, 1995; Shreesha and Gudi, 2004; Schoukens et. al. 2011) which is squared, summed and minimized in the objective function of the prediction error or nonlinear least-squares regression. Without this noise filter, the estimation is well-understood to be biased yielding inaccurate FIR coefficients but unfortunately makes the estimator nonlinear since H and Z are variables. From a quantity-logic-quality phenomena (QLQP) perspective, the time-series U1, Y, Z and A found in Figure 1 are considered as flows or more appropriately signal-flows or data. However, in our IML implementation found in Appendix A we collapse the three (3) continuous-processes into one blackbox model as shown by the dotted-line box in Figure 1 where the flows of U1, Y, Z and A are now considered as conditions and the G and H parameters are static coefficients in the IMPL semantics. Once the FIR parameters are known then these can be straightforwardly implemented into advanced process controllers such as found in APC-IMF-Julia. It should also be stressed that for multiple-input, multiple-output (MIMO) processes, the design of the input time-series plays an important role in the success of regressing good and useful dynamic representations such as transfer-function, state-space and FIR models where a novel design of the external excitation or dither signals can be found in DSDP-CLE-IMF for open- and/or closed-loop identification. Industrial Modeling Framework (IMF), IMPL and SSIIMPLE
  • 3. To implement the mathematical formulation of this and other systems, IAL offers a unique approach and is incorporated into our Industrial Modeling Programming Language we call IMPL. IMPL has its own modeling language called IML (short for Industrial Modeling Language) which is a flat or text-file interface as well as a set of API's which can be called from any computer programming language such as C, C++, Fortran, C#, VBA, Java (SWIG), Python (CTYPES) and/or Julia (CCALL) called IPL (short for Industrial Programming Language) to both build the model and to view the solution. Models can be a mix of linear, mixed-integer and nonlinear variables and constraints and are solved using a combination of LP, QP, MILP and NLP solvers such as COINMP, GLPK, LPSOLVE, SCIP, CPLEX, GUROBI, LINDO, XPRESS, CONOPT, IPOPT, KNITRO and WORHP as well as our own implementation of SLP called SLPQPE (Successive Linear & Quadratic Programming Engine) which is a very competitive alternative to the other nonlinear solvers and embeds all available LP and QP solvers. In addition and specific to DRR problems, we also have a special solver called SECQPE standing for Sequential Equality-Constrained QP Engine which computes the least-squares solution and a post-solver called SORVE standing for Supplemental Observability, Redundancy and Variability Estimator to estimate the usual DRR statistics. SECQPE also includes a Levenberg-Marquardt regularization method for nonlinear data regression problems and can be presolved using SLPQPE i.e., SLPQPE warm-starts SECQPE. SORVE is run after the SECQPE solver and also computes the well-known "maximum-power" gross-error statistics (measurement and nodal/constraint tests) to help locate outliers, defects and/or faults i.e., mal- functions in the measurement system and mis-specifications in the logging system. The underlying system architecture of IMPL is called SSIIMPLE (we hope literally) which is short for Server, Solvers, Interfacer (IML), Interacter (IPL), Modeler, Presolver Libraries and Executable. The Server, Solvers, Presolver and Executable are primarily model or problem- independent whereas the Interfacer, Interacter and Modeler are typically domain-specific i.e., model or problem-dependent. Fortunately, for most industrial planning, scheduling, optimization, control and monitoring problems found in the process industries, IMPL's standard Interfacer, Interacter and Modeler are well-suited and comprehensive to model the most difficult of production and process complexities allowing for the formulations of straightforward coefficient equations, ubiquitous conservation laws, rigorous constitutive relations, empirical correlative expressions and other necessary side constraints. User, custom, adhoc or external constraints can be augmented or appended to IMPL when necessary in several ways. For MILP or logistics problems we offer user-defined constraints configurable from the IML file or the IPL code where the variables and constraints are referenced using unit-operation-port-state names and the quantity-logic variable types. It is also possible to import a foreign *.ILP file (row-based MPS file) which can be generated by any algebraic modeling language or matrix generator. This file is read just prior to generating the matrix and before exporting to the LP, QP or MILP solver. For NLP or quality problems we offer user-defined formula configuration in the IML file and single-value and multi-value function blocks writable in C, C++ or Fortran. The nonlinear formulas may include intrinsic functions such as EXP, LN, LOG, SIN, COS, TAN, MIN, MAX, IF, NOT, EQ, NE, LE, LT, GE, GT and CIP, LIP, SIP and KIP (constant, linear and monotonic spline interpolations) as well as user-written extrinsic functions (XFCN). It is also possible to import another type of foreign file called the *.INL file where both linear and nonlinear constraints can be added easily using new or existing IMPL variables. Industrial modeling frameworks or IMF's are intended to provide a jump-start to an industrial project implementation i.e., a pre-project if you will, whereby pre-configured IML files and/or IPL
  • 4. code are available specific to your problem at hand. The IML files and/or IPL code can be easily enhanced, extended, customized, modified, etc. to meet the diverse needs of your project and as it evolves over time and use. IMF's also provide graphical user interface prototypes for drawing the flowsheet as in Figure 1 and typical Gantt charts and trend plots to view the solution of quantity, logic and quality time-profiles. Current developments use Python 2.3 and 2.7 integrated with open-source Gnome Dia and Matplotlib modules respectively but other prototypes embedded within Microsoft Excel/VBA for example can be created in a straightforward manner. However, the primary purpose of the IMF's is to provide a timely, cost-effective, manageable and maintainable deployment of IMPL to formulate and optimize complex industrial manufacturing systems in either off-line or on-line environments. Using IMPL alone would be somewhat similar (but not as bad) to learning the syntax and semantics of an AML as well as having to code all of the necessary mathematical representations of the problem including the details of digitizing your data into time-points and periods, demarcating past, present and future time-horizons, defining sets, index-sets, compound-sets to traverse the network or topology, calculating independent and dependent parameters to be used as coefficients and bounds and finally creating all of the necessary variables and constraints to model the complex details of logistics (discrete) and quality (nonlinear) industrial optimization problems. Instead, IMF's and IMPL provide, in our opinion, a more elegant and structured approach to industrial modeling and solving so that you can capture the benefits of advanced decision-making faster, better and cheaper. Finite Impulse Response Estimation of Gas Furnace Data Synopsis After iterating using SECQPE several times and setting certain G and H coefficients to zero (0.0) depending on their reported confidence-intervals from SORVE, which is the typical protocol especially with non-parametric estimation, their values with two (2) times their standard-error are shown below: G0 = 0.0 G1 = 0.0 G2 = 0.0 G3 = -0.534 +/- 0.15 G4 = -0.667 +/- 0.16 G5 = -0.861 +/- 0.16 G6 = -0.496 +/- 0.16 G7 = -0.260 +/- 0.13 G8 = -0.123 +/- 0.10 G9 = 0.0 G10 = 0.0 H0 = 1.0 H1 = -1.522 +/- 0.10 H2 = 0.613 +/- 0.10 H5 = 0.0 H6 = 0.0 H7 = 0.0 H8 = 0.0 H9 = 0.0 H10 = 0.0 The objective function value computed is 16.9 in twelve (12) iterations of SECQPE. The reported standard-error of the residuals is 16.9/296 = 0.0571 which approximates the standard- deviation of the (hopefully) white-noise residuals of time-series A. The absolute values for H1 and H2 are almost identical to those found in Box and Jenkins (1976) which is consistent with the fact that they also used an auto-regressive (AR) noise model. In addition, no significant
  • 5. auto-correlation of the residuals (our time-series A) was detected confirming that the estimation should be unbiased. The static gain (i.e., the first-order partial derivative of how U1 affects Y) using the truncated impulse response G is -2.941 +/- 1.02 which is close to the steady-state gain reported in TSE- GFD-IMF of (-0.53-0.37-0.51)/(1-0.57-0.01) = -3.357 by setting the backwards shift operator (z^- 1) to unity (1.0) where the dead-times are identical to three (3) time-periods. Although there is over a 10% difference between the two static gain estimates, this is not uncommon when fitting its value from passive or happenstance data which may include some form of feedback (closed- loop interactions) as opposed to a well-designed, open/closed-loop PRBS/GBNS input/dither signal (DSDP-CLE-IMF). In summary, we have highlighted the application of finite impulse response estimation (FIRE) using the industrial gas furnace data (Series J) from Box and Jenkins (1976) for both the deterministic and stochastic terms. The model was formulated in IMPL and solved successfully using its SECQPE and SORVE and can also be used to estimate static gains of the system which would be useful in further steady-state process optimization (active) and/or process monitoring (passive) applications. References Box, G.E.P., Jenkins, G.M., “Time-series analysis: forecasting and control”, revised edition, Holden Day, Oakland, CA, 389-400 and Series J. (1976). MacGregor, J.F., and Fogal, D.T., “Closed-loop identification: the role of the noise model and prefilters”, Journal of Process Control, 5, 163, 171, (1995). Shreesha, C., Gudi, R.D., “Analysis of pre-filter based closed-loop control-relevant identification methodologies”, Canadian Journal of Chemical Engineering, 82, (2004). Kelly, J.D., "Production modeling for multimodal operations", Chemical Engineering Progress, February, 44, (2004). Kelly, J.D., "The unit-operation-stock superstructure (UOSS) and the quantity-logic-quality paradigm (QLQP) for production scheduling in the process industries", In: MISTA 2005 Conference Proceedings, 327, (2005). Qin, J.S., “An overview of subspace identification”, Computers and Chemical Engineering, 1502-1513, (2006). Kelly, J.D., Zyngier, D., "A new and improved MILP formulation to optimize observability, redundancy and precision for sensor network problems", American Institute of Chemical Engineering Journal, 54, 1282, (2008). Schoukens, J., Rolain, Y., Vandersteen, G., Pintelon, R., “User friendly Box-Jenkins identification using nonparametric noise models, 50th IEEE Conference on Decision and Control European Control Conference (CDC-ECC), Orlando, Florida, USA, December, (2011). IAL, “Time series estimation of gas furnace data industrial modeling framework (TSE-GFD- IMF)”, Slideshare, August, 2014.
  • 6. IAL, “Advanced process control (APC) industrial modeling framework in the Julia programming language (APC-IMF-Julia)”, Slideshare, October, 2014. IAL, “Dither signal design problem for closed-loop estimation industrial modeling framework (DSDP-CLE-IMF)”, Slideshare, December, 2014. Appendix A – FIRE-GFD-IMF.IML File i M P l (c) Copyright and Property of i n d u s t r I A L g o r i t h m s LLC. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Calculation Data (Parameters) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! &sCalc,@sValue START,0.0 BEGIN,7.0 END,296.0 PERIOD,1.0 SE,1.0 !0.0571 != 16.9/296 LRGBND,1d+2 gbnd,1d+2 hbnd,1d+2 &sCalc,@sValue !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Chronological Data (Periods) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! @rPastTHD,@rFutureTHD,@rTPD START,END,PERIOD @rPastTHD,@rFutureTHD,@rTPD !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Constant Data (Parameters) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! &sData,@sValue u1,-0.052 ,0.057 ,0.235 ,0.396 ,0.43 ,0.498 ,0.518 ,0.405 ,0.184 ,-0.123 ,-0.531 ,-0.998 ,-1.364 ,-1.463 ,-1.245 ,-0.757 ,-0.418 ,-0.136 ,0.145 ,0.492 ,0.828 ,0.923 ,0.932 ,0.948 ,1.044 ,1.32 ,1.832 ,2.033 ,1.991 ,1.923 ,1.889 ,1.824 ,1.665 ,1.322 ,0.847 ,0.417 ,0.172 ,0.145 ,0.388 ,0.702 ,1.017 ,1.466 ,2.727 ,2.891 ,2.869 ,2.54
  • 12. ,1.79 ,1.69 ,1.89 ,2.49 ,2.99 ,3.59 ,3.79 ,3.29 ,2.09 ,1.49 ,0.59 ,0.79 ,1.79 ,2.89 ,3.69 ,4.29 ,4.79 ,5.09 ,5.29 ,5.29 ,5.09 ,4.49 ,3.89 ,3.49 ,2.89 ,2.79 ,2.89 ,2.89 ,2.49 ,1.69 ,0.49 ,-0.51 ,-1.51 ,-1.91 ,-1.91 ,-2.41 ,-3.11 ,-3.51 ,-3.51 ,-1.51 ,0.49 ,1.59 ,0.99 ,-0.71 ,-2.11 ,-2.71 ,-2.31 ,-1.51 ,-0.71 ,0.29 ,0.99 ,1.39 ,1.39 ,1.29 ,0.89 ,0.19 ,-0.21 ,-0.71 ,-0.91 ,-0.91 ,-0.51 ,0.79 ,2.49 ,3.49 ,4.49 ,5.09 ,4.99 ,4.79 ,4.29 ,3.79 ,3.49 &sData,@sValue !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Construction Data (Pointers) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! &sUnit,&sOperation,@sType,@sSubtype,@sUse BLACKBOX,,processc,blackbox,, &sUnit,&sOperation,@sType,@sSubtype,@sUse !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Condition Data (Properties) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! &sCondition Y U1 X1 Z A &sCondition &sCoefficient,@sType,@sPath_Name,@sLibrary_Name,@sFunction_Name,@iNumber_Conditions,@rPerturb_Size,@sCondition_Names g0,static
  • 13. g1,static g2,static g3,static g4,static g5,static g6,static g7,static g8,static g9,static g10,static h0,static h1,static h2,static h3,static h4,static h5,static h6,static h7,static h8,static h9,static h10,static &sCoefficient,@sType,@sPath_Name,@sLibrary_Name,@sFunction_Name,@iNumber_Conditions,@rPerturb_Size,@sCondition_Names &sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target BLACKBOX,,Y,-LRGBND,LRGBND, BLACKBOX,,U1,-LRGBND,LRGBND, BLACKBOX,,X1,-LRGBND,LRGBND, BLACKBOX,,Z,-LRGBND,LRGBND, BLACKBOX,,A,-LRGBND,LRGBND, &sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target &sUnit,&sOperation,&sCoefficient,@rCoefficient_Lower,@rCoefficient_Upper,@rCoefficient_Target BLACKBOX,,g0, 0.0,0.0, BLACKBOX,,g1, -0*gbnd,0*gbnd, BLACKBOX,,g2, -0*gbnd,0*gbnd, BLACKBOX,,g3, -gbnd,gbnd, BLACKBOX,,g4, -gbnd,gbnd, BLACKBOX,,g5, -gbnd,gbnd, BLACKBOX,,g6, -gbnd,gbnd, BLACKBOX,,g7, -gbnd,gbnd, BLACKBOX,,g8, -gbnd,gbnd, BLACKBOX,,g9, -0*gbnd,0*gbnd, BLACKBOX,,g10,-0*gbnd,0*gbnd, BLACKBOX,,h0, 1.0,1.0, BLACKBOX,,h1, -hbnd,hbnd, BLACKBOX,,h2, -hbnd,hbnd, BLACKBOX,,h3, -0*hbnd,0*hbnd, BLACKBOX,,h4, -0*hbnd,0*hbnd, BLACKBOX,,h5, -0*hbnd,0*hbnd, BLACKBOX,,h6, -0*hbnd,0*hbnd, BLACKBOX,,h7, -0*hbnd,0*hbnd, BLACKBOX,,h8, -0*hbnd,0*hbnd, BLACKBOX,,h9, -0*hbnd,0*hbnd, BLACKBOX,,h10,-0*hbnd,0*hbnd, &sUnit,&sOperation,&sCoefficient,@rCoefficient_Lower,@rCoefficient_Upper,@rCoefficient_Target Conditions-&sMacro,@sValue X1, g0*U1[0]+g1*U1[1]+g2*U1[2]+g3*U1[3]+g4*U1[4]+g5*U1[5]+g6*U1[6]+g7*U1[7]+g8*U1[8]+g9*U1[9]+g10*U1[10] Z, Y - X1 A, h0*Z[0]+h1*Z[1]+h2*Z[2]+h3*Z[3]+h4*Z[4]+h5*Z[5]+h6*Z[6]+h7*Z[7]+h8*Z[8]+h9*Z[9]+h10*Z[10] Conditions-&sMacro,@sValue ConditionsUOCondition-&sUnit,&sOperation,&sCondition,@sType,@rValue,@sValue BLACKBOX,,X1,?,3,X1 BLACKBOX,,Z,?,3,Z BLACKBOX,,A,?,3,A ConditionsUOCondition-&sUnit,&sOperation,&sCondition,@sType,@rValue,@sValue !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Cost Data (Pricing) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! &sUnit,&sOperation,&sCondition,@rConditionPro_Weight,@rConditionPer1_Weight,@rConditionPer2_Weight,@rConditionPen_Weight BLACKBOX,,A,,,1.0/SE, &sUnit,&sOperation,&sCondition,@rConditionPro_Weight,@rConditionPer1_Weight,@rConditionPer2_Weight,@rConditionPen_Weight !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! Command Data (Future Provisos) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! &sUnit,&sOperation,@rSetup_Lower,@rSetup_Upper,@rBegin_Time,@rEnd_Time BLACKBOX,,1,1,START,END &sUnit,&sOperation,@rSetup_Lower,@rSetup_Upper,@rBegin_Time,@rEnd_Time &sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target,@rBegin_Time,@rEnd_Time BLACKBOX,,U1,u1,u1,,START,PERIOD BLACKBOX,,Y,y,y,,START,PERIOD BLACKBOX,,A,-1000,1000,,START,BEGIN ,,A,-1000,1000,0.0,BEGIN,END &sUnit,&sOperation,&sCondition,@rCondition_Lower,@rCondition_Upper,@rCondition_Target,@rBegin_Time,@rEnd_Time