U.S. patent application number 15/041536 was filed with the patent office on 2016-06-09 for predictive alert system for building energy management.
The applicant listed for this patent is UTILITY PROGRAMS AND METERING II, INC.. Invention is credited to Christopher John Angerame, David John Harroun.
Application Number | 20160161967 15/041536 |
Document ID | / |
Family ID | 51935895 |
Filed Date | 2016-06-09 |
United States Patent
Application |
20160161967 |
Kind Code |
A1 |
Angerame; Christopher John ;
et al. |
June 9, 2016 |
Predictive Alert System for Building Energy Management
Abstract
A system for managing energy use of buildings and the like uses
predicted meteorological data, for example from weather forecasts,
applied to models of the energy systems of the building to predict
future energy use and costs. This predicted information is used to
produce automatic alerts allowing future energy use to guide
current changes to reduce energy.
Inventors: |
Angerame; Christopher John;
(North Bergen, NJ) ; Harroun; David John; (Wales,
WI) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
UTILITY PROGRAMS AND METERING II, INC. |
New York |
NY |
US |
|
|
Family ID: |
51935895 |
Appl. No.: |
15/041536 |
Filed: |
February 11, 2016 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13899977 |
May 22, 2013 |
|
|
|
15041536 |
|
|
|
|
Current U.S.
Class: |
700/291 |
Current CPC
Class: |
G05B 2219/2642 20130101;
G06Q 50/06 20130101; G05B 13/04 20130101; F24D 19/1048 20130101;
G05B 2219/2639 20130101; G05D 23/1923 20130101; G05B 13/048
20130101; G05F 1/66 20130101; G05D 23/1904 20130101 |
International
Class: |
G05F 1/66 20060101
G05F001/66; G05B 13/04 20060101 G05B013/04 |
Claims
1. A building energy system comprising: a building including a
ventilation system communicating with heat exchangers for
controlling the temperature of circulated air in the ventilation
systems; a series of sensors providing first to data inputs of
real-time measurements of the ventilation system providing
parameters of the building energy system; an electronic computer
communicating with the series of sensors and executing a program
stored in a non-transient medium to: (a) receive first data inputs
of real-time measured parameters of the energy systems; (b) receive
second data inputs of current and predicted local meteorological
information; (c) apply predicted local meteorological information
and real-time measured parameters of the first and second inputs to
a model of the energy system to produce predicted parameters of the
energy system; (d) apply the predicted and measured parameters to
alert rules to generate alerts related to future avoidable energy
cost increases, and (e) output the alerts providing information
indicating a future avoidable cost increase and a prediction time
associated with a predicted time of the future avoidable energy
cost increase.
2. The apparatus of claim 1 wherein the information indicating a
future avoidable cost increase is a maximum for a current day
expressed in at least one of plant tonnage and plant
efficiency.
3. The apparatus of claim 1 wherein the model consists of
historical operating data of the energy system providing a
functional relationship between meteorological data and energy
usage.
4. The apparatus of claim 3 wherein the model includes at least two
submodels, a first submodel relating meteorological data to output
of the energy system and the second submodel relating output of the
energy system to energy use of the energy system and wherein each
model is obtained by a regression fit to historical data.
5. The apparatus of claim 4 wherein the first submodel provides an
output of cooling tons and the second submodel provides an output
of power usage.
6. The apparatus of claim 3 wherein the model further includes
operating rules describing a selection among different of
corresponding submodels associated with different configurations of
the energy system based on the meteorological data and reflecting
expected different configurations of the energy system in response
to different meteorological conditions.
7. The apparatus of claim 6 wherein the operating rules change
submodels according to different ranges of meteorological data.
8. The apparatus of claim 1 wherein the electronic computer further
executes the program to output to a user the predicted measured
parameters based on different operating rules to provide a
comparison of future performance under different operating
rules.
9. The apparatus of claim 1 wherein the electronic computer further
executes the program to compare predictions of the model to later
actual measurements of the energy system and to provide at least
one of an alert indicating disparity between the predictions and
the actual measurements, a selection of a different model, and a
cumulative ranking of models.
10. The system of claim 1 wherein execution of the program further
corrects the predicted local meteorological, information based on
previous comparisons of previous predicted local meteorological
information and actual meteorological information at a time
indicated by the previous predicted local meteorological
information.
11. The system of claim 1 wherein the electronic computer further
executes the program to repeatedly apply the predicted local
meteorological information and real-time measured parameters of the
first and second inputs to the model to produce new predicted
measured parameters of the energy system at multiple times
throughout a day.
12. The apparatus of claim 1 wherein the electronic computer
further executes the program to monitor the real-time measured
parameters for error selected from the group consisting of sensor
data that is missing, sensor data that is unvarying and, sensor
data that is out of range to provide an alert of such error.
13. The system of claim 1 wherein the alerts indicate a projected
change in energy cost categories according to a utility rate
table.
14. The apparatus of claim 1 wherein the alerts include cost values
indicating a cost significance of the alert based on a utility rate
table.
15. The apparatus of claim 1 wherein the electronic computer
further executes the stored program to output to a user the
predicted measured parameters.
16. The apparatus of claim 15 wherein the electronic computer
further executes the program to apply predicted local
meteorological information to a second model of the energy system
to produce predicted electrical and steam cogeneration wherein the
output from the second model provides an input to the first
model.
17. The system of claim 1 wherein the local meteorological data is
selected from the group consisting of wet bulb temperature of air
outside of a structure serviced by the energy system.
18. The system of claim 1 wherein the real-time measured parameters
are selected from the group consisting of cooling system output
tonnage, electrical power usage, and activation of particular
energy system components.
19. The system of claim 1 wherein the model of the energy system is
selected from the group consisting of models of energy systems of
furnaces, boilers, air-conditioning units, water chillers,
cogeneration facilities and combinations of these energy
systems.
20. A method of assisting in a management of energy systems for
plants and the like comprising the steps of: (a) receiving first
data inputs providing real-time measured parameters of the energy
systems: (b) receiving second data inputs providing predicted local
meteorological information; (c) applying predicted local
meteorological information and real-time measured parameters of the
first and second inputs to a model of the energy system to produce
predicted measured parameters of the energy system; (d) applying
the predicted measured parameters to alert rules to generate alerts
related to future avoidable energy cost increases; and (e)
outputting the alerts providing an indication of future avoidable
cost increase and a prediction time associated with a predicted
time of the future avoidable energy cost increase.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation on U.S. application Ser.
No. 13/899,977 filed May 22, 2013 and hereby incorporated by
reference.
BACKGROUND OF THE INVENTION
[0002] The present invention relates to energy management systems
for buildings and the like and in particular to an energy
management system that provides predictive alerts with respect to
energy usage.
[0003] Sophisticated management of energy usage of large buildings
and the like can provide substantial cost savings and environmental
benefits. To this end, computerized control systems may monitor
data related to the building environment, for example temperatures
and humidities, and provide near real-time control of energy
systems including heating, ventilation, and an conditioning
equipment, to balance energy usage with a desired building
environment.
[0004] Historical energy usage by a building may also be monitored
and reviewed to identify possible inefficiencies. Once these
inefficiencies are found, energy systems can be restructured and or
control algorithms or operating procedures used for the control of
those systems may be changed.
[0005] While significant advances have taken place with respect to
the control of building energy systems, the complexity of the
energy systems and the need to flexibly respond to unexpected
situations normally require input from a building engineer or the
like who may monitor the operation and performance of the equipment
on a day-to-day basis and make necessary changes and adjustments.
When the building engineer needs to intervene in the control of the
energy systems, there is often little warning and the engineer may
have limited ability to fully evaluate the necessary adjustments
before making them. Further, instantaneous change in the operating
state of the energy systems is normally not possible further
limiting the options available to the building engineer.
SUMMARY OF THE INVENTION
[0006] The present invention provides an energy monitoring system
that may produce predictive alerts to the building engineer or the
like such as may anticipate problems that may need response. This
predictive capacity allows adjustments to the energy management
equipment that may largely avoid problems before they occur with
reduced loss and operating efficiency.
[0007] Practical predictive alerts are provided by applying
meteorological forecasts to a model of the energy management
system. Meteorological forecasting is well-established and readily
obtained and represents a significant parameter affecting the
future operation of the energy management equipment.
[0008] Specifically then, the present invention provides an
apparatus for managing energy systems for buildings and the like
comprising an electronic computer executing a program stored in a
non-transient medium to receive, first, data inputs of real-time
measured parameters of the energy systems and, second, data inputs
of current and predicted local meteorological information. The
predicted local, meteorological information and real-time measured
parameters are applied to a model of the energy system to produce
predicted measured parameters of the energy system and the
predicted measured parameters are in turn applied to alert rules to
generate alerts related to future avoidable energy cost increases.
The alerts may provide a description of the future avoidable cost
increase and a prediction time associated with a predicted time of
the future avoidable energy cost increase are then output.
[0009] It is thus an object of at least one embodiment of the
invention to provide advance notice to building engineers or the
like by leveraging the power of sophisticated meteorological
forecasts in predicting energy usage.
[0010] The model may consist of historical operating data of the
energy system providing a functional relationship between
meteorological data and energy usage.
[0011] It is thus an object of at least one embodiment of the
invention to provide a procedural method of producing a model that
can be implemented in a variety of complex energy systems.
[0012] The model may include at least two submodels, a first
submodel relating meteorological data to output of the energy
system and the second submodel relating output of the energy system
to energy use of the energy system and wherein each model is
obtained by a regression fit to historical data.
[0013] It is thus an object of at least one embodiment of the
invention to link readily measured parameters of the energy system
such as tonnage to energy use that will permit cost sensitive
analysis.
[0014] The model may further include operating rules describing a
selection among different corresponding submodels associated with
different configurations of the energy system based on the
meteorological data and reflecting expected different
configurations of the energy system in response to different
meteorological conditions.
[0015] It is thus an object of at least one embodiment of the
invention to incorporate into the model current best operating
procedures not easily captured by simple regression and the like
alone.
[0016] The operating rules may change sub-models according to
different ranges of meteorological data.
[0017] It is thus an object of at least one embodiment of the
invention to permit piecewise construction of a model that can work
under a variety of different environmental conditions.
[0018] The electronic computer may further execute the stored
program to output to a user the predicted measured parameters based
on different operating rules to provide a comparison of future
performance under different: operating rules.
[0019] It is thus an object of at least one embodiment of the
invention to employ the same mechanism for "what-if" type analysis
to analyze efficient operation of the energy systems.
[0020] The electronic computer may further execute the program to
compare predictions of the model to later actual measurements of
the energy system and to provide at least one of an alert
indicating disparity between the predictions and the actual
measurements, a selection of a different model, and a cumulative
ranking of models.
[0021] It is thus an object of at least one embodiment of the
invention to provide a check of model drift or inaccuracy that may
be formed on an automatic basis.
[0022] The execution of the stored program may further correct the
predicted local meteorological information based on previous
comparisons of previous predicted local meteorological information
and actual meteorological information at the time indicated by the
previous prediction.
[0023] It is thus an object of at least one embodiment of the
invention to provide an automatically adjusting model to improve
accuracy over time.
[0024] The electronic computer may further execute to repeatedly
apply the predicted local meteorological information and real-time
measured parameters of the first and second inputs to the model to
produce new predicted measured parameters of the energy system at
multiple times throughout a day.
[0025] It is thus an object of at least one embodiment of the
invention to eliminate cumulative prediction error by repeated
iteration using new data.
[0026] The electronic computer may further monitor the real-time
measured parameters for error selected from the group consisting of
sensor data that is missing, sensor data that is unvarying and
sensor data that is out of range to provide an alert of such
error.
[0027] It is thus an object of at least one embodiment of the
invention to provide a system that can detect within limits
erroneous inputs that may affect the modeling process.
[0028] The alerts may indicate a projected change in energy costs
categories according to a utility rate table.
[0029] It is thus an object of at least one embodiment of the
invention to incorporate often arcane utility schedules into the
modeling process and alert process.
[0030] The alerts may include cost values indicating a cost
significance of not responding to the alert based on a utility rate
table
[0031] It is thus an object of at least one embodiment of the
invention to provide a readily available reference for the
significance of the alert in terms of energy cost.
[0032] The electronic computer may further output to a user the
predicted measured parameters.
[0033] It is thus an object of at least one embodiment of the
invention to provide a general-purpose prediction tool for use in
managing energy systems.
[0034] The computer may apply predicted local meteorological
information to a second model of the energy system to produce
predicted electrical and steam cogeneration wherein the output from
the second model provides an input to the first model.
[0035] It is thus an object of at least one embodiment of the
invention to permit sophisticated cogeneration systems that may
fundamentally affect energy costs to be analyzed on a predicted
basis.
[0036] The meteorological data may be selected from the group
consisting of wet bulb temperature and dry bulb temperature of the
air outside of a structure serviced by the energy system.
[0037] It is thus an object of at least one embodiment of the
invention to provide models that may make use of objective and
readily obtainable forecast data.
[0038] The real-time measured parameters may be selected from the
group consisting of cooling system output tonnage, electrical power
usage, steam power usage, and activation of particular energy
system components.
[0039] It is thus an object of at least one embodiment of the
invention to use parameters that are typically available in
sophisticated energy management environments
[0040] The model of the energy system may be selected from the
group consisting of models of energy systems of furnaces, boilers,
air-conditioning units, chilled water chillers, and cogeneration
facilities and combinations of these energy systems.
[0041] It is thus an object of at least one embodiment of the
invention to provide a predictive alert system applicable to a wide
variety of significant energy systems.
[0042] These particular objects and advantages may apply to only
some embodiments falling within the claims and thus do not define
the scope of the invention.
BRIEF DESCRIPTION OF THE FIGURES
[0043] FIG. 1 is a simplified block, diagram of a building having a
heating and cooling energy system as controlled by a building
controller and as monitored by a computer implementing the present
invention;
[0044] FIG. 2 is a diagram of the functional blocks of the program
implemented by the computer of FIG. 1 showing a set of models that
may receive predicted weather data to provide predicted energy
consumption and further showing a rate table for converting energy
consumption into cost, an alert generator monitoring the outputs of
the models, and a model monitor monitoring and correcting model
accuracy;
[0045] FIG. 3 is a simplified representation of a multidimensional
acquisition of historical energy data as may be abstracted into
multiple submodels by regression analysis;
[0046] FIG. 4 is a modeling data flow diagram showing the
application of predicted meteorological data to a composite model
made of the multiple submodel components which may provide results
to a second submodel and in turn provide results to a rate table
for energy cost estimation;
[0047] FIG. 5 is a fragmentary block diagram of the modeling system
of FIG. 2 showing an implementation of a model corrector comparing
predictions from multiple models against actual data to rank and
select among models;
[0048] FIG. 6 is a simplified representation of a rate table
showing rate tiers and showing one set of alert rules for the alert
generator generating alerts based on transition through rate
tiers;
[0049] FIG. 7 is an operating rule table working as part of the
modeling system to capture standard operating procedures with
respect to equipment operation;
[0050] FIG. 8 is a graph of output prediction at multiple times
showing correction of each prediction with actual operating data in
an iterative process;
[0051] FIG. 9 is a simplified output chart showing energy
consumption under a "what if" scenario;
[0052] FIG. 10 is a fragmentary view of FIG. 2 having added
cogeneration models for use in providing great information; and
[0053] FIG. 11 is a display screen of a computer implementing the
present invention as shown in FIG. 1, the display providing a
comprehensive view of the results of the modeling process.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
[0054] Referring now to FIG. 1, a building 10 such as a high-rise
office building, factory, server farm or the like may provide for a
ventilation system 12, for example, having an air intake 14
receiving outside and recycled inside air and passing it in
different ventilation ducts. The ventilation ducts may pass through
heat exchangers 16 that may heat or cool the air in each of the
ventilation ducts for distribution to the interior of the building
10, for example, under the control of multiple fans 18 and manual
or motorized duct valves (not shown).
[0055] The heat exchangers 16 may receive a source of heated water
20 or chilled water 22 (for example through separate exchange
elements) for heating or cooling under the control of one or more
valves 24. The heated water 20 may, for example, be pumped by a
pump 26 from a boiler 28, the latter receiving a source of fuel 30
such as natural gas, oil, or the like through a fuel meter 32. The
chilled water 22 may come from a water chiller 34 fed by compressor
units of an air-conditioning unit 36 (holding a thermostatically
controlled compressor), the latter exhausting heat through cooling
towers 38 or the like. The heated water 20 or chilled water 22 may
have its temperatures monitored by one or more temperature sensors
25.
[0056] The various electrical equipment including the
air-conditioning units 36 may receive electrical service 40 from
electrical service 44 monitored by electrical meter 42. Each of
these electronically actuable or electronically readable elements
and their associated equipment together comprise an energy system
45 for providing control of energy (e.g. heating and cooling) to
the building.
[0057] A building controller 46 in communication with the above
sensors and actuators provides control of the same on an automatic
or semiautomatic basis. The building controller 46 may include an
electronic computer having a processor 48 and memory 50, the latter
holding a stored control program held in non-transitory media such
as electronic disk or electronic memory. The controller 46 may
communicate with a control console 52, for example, providing a
graphic display terminal and user input device such as a keyboard
or the like.
[0058] The building controller 46 may receive electronic signals
from each of the above-described components or manually entered
data read from such components to obtain an understanding of the
building state 86. In this regard, the building controller 46 may
monitor the temperatures of the chilled water 22 and heated water
20 through temperature sensors 25 and may monitor interior space
temperatures and humidity through in-room monitors 27. The
controller 46 may also monitor fuel usage through meters 32 and 42
although this may not be used in its control capacity. This
building state 86 is input to a control algorithm executing on the
building controller 46 to provide for real-time control of the
energy system 45.
[0059] In response to the control algorithm, the building
controller 46 may control the fans 18 and associated duct valves,
the activation of valves 24 for controlling chilled and heated
water, the pump 26 activation of the boiler 28 and motors and
valves of the individual air-conditioning unit 36. Secondary
control loops may be established, for example, between fans of the
cooling towers 38 and the air-conditioning units 36, for example,
by local controllers which may be practically considered a portion
of the building controller 46. Building controllers 46 for
controlling energy systems 45 are generally understood in the art
and may use a variety of different control algorithms, for example,
implementing all or portions of proportional, integral, and
derivative control strategies, for example.
[0060] The present invention supplements the control of the energy
system 45 implemented by the building controller 46 and may, in
fact, be executed on the building controller 46 but more typically
will be implemented on a second computer 60. The second computer 60
may also have a processor 62 and a memory 63 holding various data
files and a program implementing the present invention as will be
discussed further below. Computers 60 may communicate with standard
human machine interface (HMI) elements including a display 64 and
keyboard 66 for presenting information to a human operator and
receiving commands from the operator, respectively. In one
embodiment, the computer 60 may be directly connected to each of
the energy system 45 components monitored or actuable by the
controller 46 so as to be able to obtain information about the
current building state 86 of the building 10 and its energy system
45.
[0061] The computer 60 may also receive current meteorological data
68, for example, from local weather instrumentation 71, the latter,
for example, providing equipment that provides local wet bulb and
dry bulb information proximate to the building 10. Ideally, this
information is directly obtained with respect to the local building
environment but may also be derived from general-purpose weather
sources interpolated to the building region. The computer 60 may
also be connected, for example, through the Internet 67 with a
source of predicted meteorological data 80, for example, from a
commercial weather prediction service. In one embodiment, predicted
data may be from a government source such as the National Oceanic
and Atmospheric Administration (NOAA) and a commercial source such
as Weather Underground combined and localized to the building
10.
[0062] It will be appreciated that the computer 60 need not
necessarily be on site and that this building state 86, predicted
meteorological data 80, and current meteorological data 68 may be
relayed, for example, over the Internet or the like and further
that the tasks that will now be described with respect to computer
60 may be distributed over multiple computers if desired. It will
be further appreciated that some of this data may be collected
manually.
[0063] Generally, the memory 63 of the computer 60 will hold an
operating program 70 whose operation will be described below
together with multiple submodels 112 that may be assembled into a
building model 84 by using stored operating rules 74. Generally,
the submodels 112 provide a mapping between the building state 86
and energy usage by the building 10 under a predefined set of
operating conditions. The operating rules 74 describe those
operating conditions that capture general strategies for the
control of the energy system 45, for example, providing a priority
in usage of different air-conditioning units 36 or choice between
different fuels, or advanced cooling or heating based on
temperature or temperature trends.
[0064] The memory 63 will also hold various other data files 76
including rate table information, model accuracy information, and
correction factors, all as will be described below.
[0065] Referring now to FIGS. 2 and 8, the program 70 operates to
receive predicted meteorological data 80, for example, predicting
the weather twelve hours in advance at one hour increments
(t.sub.0-t.sub.4). This information is processed by a weather
correction function 82 which will be described below and then
provided to a building model 84. The building model 84 also
receives the current building state 86 representing generally the
electronic state of the building 10 reflected in the various
electrically actuated elements described above such as the pumps 26
and valves 24 and data from the various temperature sensors 25 and
meters 42, for example.
[0066] For example, current building information at time to may be
received by the building model 84. Using the current building state
86 and first increment of predicted meteorological data 80 (for
example one hour hence at time t.sub.1) the building model 84
produces a predicted energy usage 88 for each of one hour into the
future at time t.sub.1 and the succeeding four hours in the future
t.sub.4 as shown by predicted energy usage 88a in FIG. 8. This
prediction may be made recursively, with predictions for each
succeeding hour being used as initial conditions for the
predictions for the next hour.
[0067] As time passes, for example, so that the current time is now
time t.sub.1, predictions for additional hours into the future may
be refreshed recursively using current meteorological data 68 for
time t.sub.1, actual building state 86 for time t.sub.1 and updated
predicted meteorological data 80, for example, to produce
predictions shown by predicted energy usage 88b.
[0068] Alerts may be generated for any time throughout the
prediction span (initially (t.sub.0-t.sub.4)) by the alert
generator 92 monitoring the most current predicted energy usage 88.
Generally the alert generator 92 incorporates a set of Boolean
and/or arithmetic conditions, for example that may be implemented
in a script, to monitor the prediction data and provide warnings to
the operator based on those Boolean and/or arithmetic conditions.
For example, a simple alert may generate a warning if energy
consumption values are expected to be above a certain value. The
alert may identify the particular condition and the time at which
the condition will occur. These alerts may be displayed on the
display 64 or sent by e-mail or conveyed in other manners to a
human operator. Any previous alerts based on earlier predicted
energy usage 88 are retracted but may be stored in an historical
log table.
[0069] The predicted energy usage 88 will be provided to a rate
table 90 providing current energy usage rates, for example,
kilowatts of electricity or BTUs of natural gas or the like, into
predicted energy costs 91. This predicted energy cost 91 may be
displayed to the user as will be described below.
[0070] The alert generator 92 may also monitor predicted energy
cost 91 to provide alerts based on the predicted energy cost 91 in
the same manner as described above with respect to predicted energy
use 88. Referring momentarily to FIG. 6, in a typical rate table 90
providing electrical rates, the cost of each kilowatt of electrical
power may change depending on the total number of kilowatts
consumed according multiple different kilowatt tiers 95 providing
abrupt price changes between those tiers 95 increasing
disproportionately as one rises through the tiers 95. This price
structure reflects the high marginal cost of additional electrical
generation capacity under peak demands.
[0071] In addition, the predicted energy cost 91 may reflect a
"demand" agreement between the electrical utility and a commercial
user of the building 10. A demand agreement is a contract between
the user and the electrical utility. Under this contract, when the
electrical utility needs additional power, it may contact its
customers subject to the demand agreement with a request that the
customer reduce its energy usage in the future for a predetermined
time. This agreement provides advantageous prices but contains
penalties when those requests are not met. Those penalties are
incorporated into the predicted energy cost 91.
[0072] In addition, the rate table 90 may provide for "real time"
utility pricing in which the utility transmits changing cost per
kilowatt and per kilowatt hour for the next 24 hours. That is the
rate table 90 may be arbitrarily dynamic.
[0073] An example alert generated by the alert generator 92 may
compare predicted energy cost 91 against the defined ranges of the
tiers 95 to provide an alert at times 97 where the projected cost
will rise to the next tier 95. This allows the building engineer,
for example, to take steps before that crossing time to decrease
electrical consumption. Other alerts may include, for example, the
use of particular pieces of equipment representing known
inefficient operating modes of the equipment, energy usage at
particular times of the day beyond a certain amount, and any
logical or arithmetic combination of the monitored parameters
representing the building state 86.
[0074] A model monitor 96 may also receive the predicted energy
usage 88 and actual energy usage 99, the latter derived from the
building state 86, and may compare the two for the same time period
by storing and holding the predicted energy usage 88 values so that
they may be compared with the actual energy usage 99. This
comparison allows assessment of the accuracy of the prediction by
the building model 84 after the fact. As will be discussed below,
the outcome of the comparison may be used to create a model failure
report 98 that may be sent to the alert generator 92 used to
generate an alert 94 in the event that the modeling is beyond
certain pre-established ranges of accuracy.
[0075] Referring still to FIG. 2, the building model 84 may be
selected from multiple alternative models 100 typically each
founded on use of a set of different operating rules 74 as will be
discussed below. In this case, the model monitor 96, as well as
providing warnings about model inaccuracy, may control selection
among different alternative models 100 used for the building model
84 or may correct given models used for the building model 84 by
selection/correction command 101 provided from the model monitor 96
to the to the building model 84.
[0076] In a similar way, the model monitor 96 may also serve a
secondary duty to monitor predicted meteorological data 80 and
current meteorological data 68 to make corrections to the predicted
weather data implemented by weather correction function 82. In one
embodiment, these corrections may ascertain a difference between
predicted meteorological values such as wet bulb or dry bulb
temperature, and the actual values that come to pass, and may
establish a correction factor or correction offset based on an
historical rolling average of these errors. Thus, for example, if
the projected dry bulb temperature is always two degrees or one
percent low, a suitable correction offset of two degrees or factor
of one percent may be implemented.
[0077] As shown in FIG. 9, the availability of different competing
alternative models 100 allows the simultaneous generation of
different predicted energy costs 91a and 91b each associated with a
different alternative model 100. These different predicted energy
costs 91a and 91b may be provided to the user to compare various
possible operating modes to guide both current decisions in
managing the energy systems 45 of the building 10 and to change the
operating rules 74 being used or adjust the operating rules 74
going forward. As depicted, the predicted energy cost 91b
associated with a second set of operating rules 74 provides
improved energy efficiency over the predicted energy cost 91a
associated with the first set of operating rules 74.
[0078] As shown in FIG. 5, in an alternative embodiment, the model
monitor 96 may simultaneously monitor the outputs of multiple
alternative models 100 with respect to the predicted values and
actual values ultimately achieved in the same manner as described
above with detecting model errors. The success of each model may be
logged in a matrix 85, for example, describing the particular model
and the particular conditions under which the prediction was made.
In this way increased experience with the models 100 may be used to
select better models 100 over time.
[0079] Referring now to FIGS. 2 and 3, the alternative models 100
implemented as the building models may be constructed from
historical measures of the actual operation of the energy system 45
as captured in one or more submodels 112. Generally, this
historical data captured by the submodels 112 will have multiple
dimensions, but in a simplified example, a particular model output
value may indicate thermal demand 102 for a particular element of
the energy system 45 (for example cooling tons for a particular
chiller) as a function of input arguments, for example dry bulb
temperature 104 and time 106 (e.g. time of day, time of year) This
large data set collected over time may be reduced by normalizing it
to particular seasons and times of day, for example, data for a
constant time of day for a particular season may be collapsed to a
single normalized energy usage value by removing a seasonal factor
(for example, established by regression analysis) and ensemble
averaging over many days of that season. In this way, a scatterplot
108 of points relating dry bulb temperature 104 to thermal demand
102 may be generated that may be corrected to particular times of
day and seasons by a simple factor of polynomial adjustment. The
scatterplot 108 may then be reduced to a functional curve 110 (for
example by linear or polynomial regression) which provides a
submodel 112 relating wet bulb temperature 104 to energy usage for
the building 10. Multiple functional curves 110 of different
arguments and values provide partial submodels 112 relating
meteorological data to energy consumption for multiple subsystems
of the building, for example an individual air-conditioning unit
36.
[0080] Referring now to FIG. 4, the various partial submodels 112
may then be assembled into a composite model 114 by combining the
partial submodels 112 together according to operating rules 74, for
example, representing current standard operating procedure executed
by either the controller 46 or the controller 46 in conjunction
with a building engineer. Referring now to FIG. 7, for example,
each operating rule 74 may define a priority of equipment usage
representing the output control values 117, in this case which
chillers to use first, as additional capacity is required as
reflected by input arguments 115 of wet bulb temperature. The
operating rules 74 in this case generally reflect the switching on
and off of systems at different times or different conditions and
may be implemented by a set of Boolean or arithmetical conditions
in a script for the like or a tabular form as shown.
[0081] Most generally, the operating rules 74 capture general
strategies for the control of the energy systems 45 of the building
10, for example providing a priority in usage of any of the
components of the energy system 45, choice between different fuels,
decisions about advanced cooling or heating based on temperature or
temperature trends, or the like.
[0082] Referring now to FIG. 4, in one embodiment, the operating
rules 74 may piece together or otherwise combine the submodels 112
based on operating rule determinations of which equipment and which
submodels 112 are relevant. In one example, a composite model 114
may be assembled from portions of different submodels 112 according
to operating rules 74 selecting different elements of equipment
(associated with different submodels 112) as a function of
meteorological data 116 or may be driven by any of the other
building states 86. As the different pieces of equipment are used,
different submodels are selected.
[0083] In this example, a composite model 114 may provide a mapping
between meteorological data 116 such as wet bulb temperature, and
energy consumption based on a first set of operating conditions. A
second submodel 112 may then be used to convert thermal energy load
(for example tonnage) to a measure of energy rate (for example
kilowatts) that may be converted to cost using a rate table 90.
[0084] In a more complete example, meteorological parameters 116
such as wet bulb temperature may be converted by a composite model
114 to plant tonnage, plant tonnage may be related by a submodel
112 to chilled water flow, and plant tonnage and chilled water flow
may he converted to total plant consumption in kilowatt hours. The
particular combination of composite models 114 and submodels 112
and their arguments and values will be dependent on the
interconnection of the various elements of the energy system 45
particular to a given building.
[0085] Referring now to FIG. 10, it will be appreciated that the
present invention may be applied to a cogeneration system in which
one of the elements of the energy system 45 generates electricity,
for example, by using the energy used in the boiler 28 to generate
steam to run a steam turbine or the like as well as to heat water.
An alternative example of a cogeneration system may be one which
uses natural gas from a utility to generate electricity on site and
in which the generation of electricity has a byproduct of
generating steam that can be used by a steam chiller (that is, a
chiller that receives its energy in the form of heated steam). A
separate cogeneration model 200 may be produced for such
cogeneration systems using the principles described above with
respect to the building model 84. The cogeneration model 200 may
also receive predicted meteorological data 80 and building state 86
and a predicted energy usage 88 from the model 84 to provide an
electrical availability signal 210. This projected available
electrical energy 210 may then be used to change the effective rate
table 90 (for example by subtracting from the predicted energy
usage 88) to indicate a cost of electrical power assuming that the
cogeneration system substitutes some electrical power for that
normally obtained by the metered public electrical grid.
[0086] Referring now to FIGS. 1 and 11, a screen output 220 on the
computer display 64 (shown in FIG. 1) may provide for real-time
alerts to the user. In one embodiment, the screen output 22
provides for a number of different fields including current and
predicted meteorological data 222, current and projected peak
operating parameters 223 (derived from the current and predicted
building state 86), current and suggested operating parameters 224
(derived from the current building state and the modeling process
of the present invention), a plot of historic and forecasted data
226, and a data dashboard 228 providing other selected information
from the building state 86.
[0087] The current and predicted meteorological data 222 may simply
extract useful information from the predicted meteorological data
80 such as current, forecast, and maximum outside air temperature
and current, forecast and maximum outside air wet bulb temperature,
including a time at which the maximum will occur. This data helps
provide a reference for the other data of the screen output
220.
[0088] The current and peak projected operating parameters 223, in
a manner roughly analogous to the current and predicted
meteorological data 222, provides selected operating parameters of
the plant, for example, plant tonnage and plant efficiency, as
derived from the building state data 86 for a current time (for
example, 8:30 am, as illustrated in FIG. 11) and as forecast one
hour into the future using the modeling system of the present
invention, and forecast to a maximum value for that day further
indicating the time of day when that maximum will occur. This
snapshot into the future provides an alert that improves a
manager's ability to react in advance to trends in energy usage and
efficiency. It will be appreciated that the maximum refers to a
maximum of energy usage and may be expressed as plant tonnage,
which increases with increased energy usage, or plant efficiency
which decreases with increased energy usage or other similar
measures.
[0089] The current and suggested operating parameters 224 provide
information for selected monitored equipment of the energy system
45 indicating its current state, for example, showing which
chillers are on and which chillers are off, the realized efficiency
tonnage and percent load of the active equipment, and in the final
column the recommended change in settings for this equipment based
on the modeling provided by the present invention and discussed
above. This display may be used to change the settings for
controlling the building according to the recommendations of the
present invention.
[0090] The plot of forecasted data 226 may plot particular
parameters throughout the day in the form of a bar chart providing,
for example, the intended setpoint value 230 and the actual value
232 for different parameters (for example, chiller electric
demand). An absolute difference and percent difference at each time
may also be indicated.
[0091] It will be appreciated that any of these values described
above may also be converted to dollar or money values through the
use of the rate table 90 discussed above.
[0092] Certain terminology is used herein for purposes of reference
only, and thus is not intended to be limiting. For example, terms
such as "upper", "lower", "above", and "below" refer to directions
in the drawings to which reference is made. Terms such as "front",
"back", "rear", "bottom" and "side", describe the orientation of
portions of the component within a consistent but arbitrary frame
of reference which is made clear by reference to the text and the
associated drawings describing the component under discussion. Such
terminology may include the words specifically mentioned above,
derivatives thereof, and words of similar import. Similarly, the
terms "first", "second" and other such numerical terms referring to
structures do not imply a sequence or order unless clearly
indicated by the context.
[0093] When introducing elements or features of the present
disclosure and the exemplary embodiments, the articles "a", "an",
"the" and "said" are intended to mean that there are one or more of
such elements or features. The terms "comprising", "including" and
"having" are intended to be inclusive and mean that there may be
additional elements or features other than those specifically
noted. It is further to be understood that the method steps,
processes, and operations described herein are not to be construed
as necessarily requiring their performance in the particular order
discussed or illustrated, unless specifically identified as an
order of performance. It is also to be understood that additional
or alternative steps may be employed.
[0094] References to "a computer" or "a processor" can be
understood to include one or more microprocessors that can
communicate in a stand-alone and/or a distributed environment(s),
and can thus be configured to communicate via wired or wireless
communications with other processors, where such one or more
processor can be configured to operate on one or more
processor-controlled devices that can be similar or different
devices. Furthermore, references to memory, unless otherwise
specified, can include one or more processor-readable and
accessible memory elements and/or components that can be internal
to the processor-controlled device, external to the
processor-controlled device, and can be accessed via a wired or
wireless network.
[0095] It is specifically intended that the present invention not
be limited to the embodiments and illustrations contained herein
and the claims should be understood to include modified forms of
those embodiments including portions of the embodiments and
combinations of elements of different embodiments as come within
the scope of the following claims. All of the publications
described herein, including patents and non-patent publications are
hereby incorporated herein by reference in their entireties.
* * * * *