System and method of predictive occupancy room conditioning   

This application is a continuation of U.S. patent application Ser. No. 12/284,795, filed Sep. 25, 2008, now pending, which application claims the benefit of U.S. Provisional Application No. 60/997,426, filed on Oct. 4, 2007, which application is incorporated herein by reference in its entirety.

BACKGROUND OF THE INVENTION

The technical field of this invention relates to zoned residential Heating Ventilation and Air Conditioning (HVAC) and lighting controls that employ embedded systems and wireless communications. HVAC systems use a large proportion of a building\'s energy usage and need to be optimized for both environmental and economic reasons.

There is a long history of invention and research associated with HVAC technology. Programmable Thermostats: Programmable setback thermostats that can be set by the occupant for set point, reset point and schedule have been available on the market for a number of years [EnergyStar]. In spite of favorable engineering analysis for prospective energy savings, field studies show that real-world field performance is no better than manual thermostats [Sachs] [Shiller] [Cross and Judd]. These studies have been suggested a number of causes including, dead band gap, difficulty of programming, and comfort issues, with the most likely cause being user overriding the energy saving features.

Residential and Light Commercial Zoning: Modern forced air zone control and VAV (Variable Air Volume) extends back decades. The most extensive use to date has been for commercial and industrial applications. There is substantial ongoing research, including that at the CEC as shown in “Advanced Variable Air Volume System Design Guide”[CEC] and Natural Resources. Canada\'s CANMET Energy Technology Centre (CETC) is currently involved in an ongoing residential zoned research project[CANMET Natural Resources Canada] with the company Ecologix [Ecologix Heating Technologies, Inc]. Home Comfort Zone teaches a basic version of zoned HVAC control[Alles]. Zone HVAC controls and systems are also taught in the patents of [Alles] [Girmado] [Parker] [Jackson] and [Nelson]. These systems, have several failures: 1) they are expensive to install, 2) are even more expensive to retrofit and 3) they make use of a barometric bypass damper that recycles conditioned overpressure air back into the return system causing reduced performance of the central HVAC plant.

Automated diagnostics, performance monitoring and continuous commissioning: The importance of initial commissioning and ongoing monitoring has long been established in buildings[ASHRAE][Bushby][SoCal-Edison]. According to Brambley, “Performance monitoring, automated fault detection and diagnosis, commissioning, optimal control and the use of development environments, design tools and trainers are complementary, and in some respects synergistic technologies that have strong potential to realize significant energy savings and other performance improvements in commercial buildings, including existing buildings. There is a significant body of previous R&D relating to these technologies that indicates their potential, both generically and for specific approaches and methods. In a significant number of cases, there is the opportunity to establish R&D programs and projects that leverage this existing work in order to move relatively quickly to tools that can be deployed in the marketplace.”

Actuated register/damper design: Actuated dampers and registers are very common in commercial and industrial HVAC installations and are occasionally seen in residential use. There are ongoing efforts to advance dampers such as pneumatic bladders [Alles], the technology offered by Zone Comfort as a retrofit option and the ratcheted Nitinol Shape Memory Alloy (SMA) based devices described by [Patel, et al].

Wired Network Technology: Wired electrical communication has been used for decades in residential HVAC controls. The most common use is the simple closure of a 24 VAC circuit to signal a central HVAC plant to turn on. Serial data links, Power Line Communication (PLC) and true networks are common in commercial HVAC application, but have seen limited use in residential HVAC. An example of a serial protocol is RS-232C. Examples of a of a wired network used in HVAC are BACNET and GANNET.

Wireless Technology: IEEE 802.15.4/ZigBee[IEEE][ZigBee Alliance] was essentially designed for sensor, command and control application such as residential HVAC. Other wireless technologies such as Z-Wave and Bluetooth are also in the marketplace.

Occupancy Sensors: Industrial and commercial HVAC systems have long used occupancy sensors and some limited use has been seen in residential settings [Seymour] [Simmions] [Keating] [Disser] [Bilger] [Gutta] [Gua] [Mozer].

Occupancy Prediction: While scheduled occupancy has been wide spread for both residential and commercial use (see programmable thermostats above), sensor based predictive occupancy has not seen commercial success. Mozer teaches a concept of a Neurothermostat in his prototype “Adaptive Control of Home Environment,” system that includes HVAC, domestic hot water and lighting control. The Neurothermstat makes use of a PC based neural network and X10 sensors and controls. Mozer reports that the X10 communication protocol adds too much latency for his application. The Mozer design uses standard neuro-network train with energy use and occupancy error as feed back values.

SUMMARY

Modern life has patterns centered around work and play. The system takes advantage of these realities in the home and predicts where energy is needed for maximum comfort and efficiency. Using integrated sensors, the embodiment learns the rhythms of the homeowners, what time they wake on a workday, what time they use the kitchen or what time they go to bed; and heats or cools individual rooms before they enter to the desired temperature. By learning how a family uses a home, the system greatly reduces the energy used in areas that are vacant, resulting in maximum comfort and efficiency. When a room is entered unexpectedly, the system rapidly brings the room to the desired temperature as the system focuses resources on real use. A combination of home zone controls, sensors and advanced learning software provide homeowners with a highly cost effective means of increasing comfort and reducing greenhouse emissions.

The system is intended to be very low cost in mass production and is designed for optional installation by the homeowner. The potential energy savings from advanced residential predictive HVAC zone controls is substantial. A first order analysis suggests that a savings of about 50% of HVAC energy usage versus a fixed manual thermostat can be expected, depending on resident usage patterns and climate. If, for example, just 10% of California homes were to implement a system that was able to achieve just a 10% air conditioning electricity savings, California could see a reduction of 111.54 mega Watts of peak demand as demonstrated by [Cal Energy Peak Loads]

Existing programmable EnergyStar thermostats\' failure to perform [EnergyStar] in the field as well as anticipated is largely due to their complex user interface. The proposed system does not require complex user programming, but simply learns when a resident occupies a room and with a simple up/down button their preferred temperature setting. By having the room properly conditioned (heated or cooled) before a resident enters a room the system delivers a much higher acceptance of temperature setback than EnergyStar systems have been able to achieve. The proposed system includes wireless duct register/damper units (DRUs) that control airflow into a room and wireless wall sensor units (WSU) that measure room temperature and occupancy. Each existing HVAC register is replaced with the new design and a wall sensor unit is placed in each room. In retrofits the existing HVAC thermostat is replaced with a special version of the wall sensor that can also control the central heat pump or furnace. Once installed the system will record the occupancy of each room and the preferred temperature. The system then reviews occupancy data and predicts if a room will be occupied and conditions the room appropriately. If an occupant unexpectedly enters a room the occupancy sensors detect entrance and focus the HVAC system onto that room to quickly condition it.

The system has great potential to integrate into other long term energy efficiency efforts beyond immediate energy savings. This smart residential HVAC system can easily be integrated with fresh air economizers such as the one demonstrated in the California Department of Energy\'s PIER Night Breeze project and indeed extended by opportunistically over cooling expected unoccupied rooms when cool outside air is available. The system also easily integrates future opportunities, includeing smart grid interfaces for on-demand load shedding, dynamic cost response changes to set points and time of day load shifting.

Advantages of the invention include a low manufacturing cost, a low installation cost, very high energy use efficiency, high user acceptance, high comfort level and a low error rate.

FIG. 1 illustrates a non-limiting example of components for a single zone that includes a wall sensor unit and one or more automated dampers units.

FIG. 2 illustrates a non-limiting example of a block diagram of certain electronic and mechanical components that comprise an example installation.

FIG. 3 illustrates a non-limiting example of a block diagram of certain electronic components that comprise a typical wall sensor unit.

FIG. 4 illustrates a non-limiting example of a block diagram of certain electronic components that comprise a typical damper/register unit.

FIG. 5 illustrates a non-limiting example of certain data structures.

FIG. 6 illustrates a non-limiting example of a weighting summation function for a Decaying Occupancy Temporal (DOT) Algorithm

FIG. 7 illustrates a non-limiting example of a weighting function.

FIG. 8 illustrates a non-limiting example of a flowchart for evaluating a weighing. function.

FIG. 9 illustrates a non-limiting example of a block diagram showing the triggering of multiple zones by an occupant.

FIG. 10 illustrates a non-limiting example of communication paths between certain electronic components that comprise an example installation.

FIG. 11 illustrates a non-limiting example of a flowchart for opening dampers and controlling the central HVAC plant.

FIG. 12 illustrates a non-limiting exploded view of the preferred embodiment of a Damper/Register Unit (RDU)

FIG. 13 illustrates a non-limiting example of a side sectional view of certain electrical and mechanical components that comprise the actuator mechanism the DRU in FIG. 12

FIG. 14 illustrates a non-limiting example of a side sectional view of certain electrical and mechanical components that comprise the actuator mechanism the DRU in FIG. 12 in the closed position.

FIG. 15 illustrates a non-limiting example of a side sectional view of certain electrical and mechanical components that comprise the actuator mechanism the DRU in FIG. 12 in the open position.

FIG. 16 illustrates a non-limiting example of a top sectional view of certain electrical, electronic and mechanical components that comprise the actuator mechanism the DRU in FIG. 12

FIG. 17 illustrates a non-limiting example of a side sectional view of certain electrical and mechanical components that comprise the actuator mechanism the DRU in FIG. 12 showing the damper plate in the closed position and certain surface detail.

FIG. 18 illustrates a non-limiting example of an alternative embodiment of a Wall Sensor Unit (WSU) that includes buttons for controlling lighting.

DETAILED DESCRIPTION

Referring in particular to FIG. 1, Wall Sensor Unit, 100 is representative of a device that includes a display sub system such as a LCD 102, one or more button, 103, an environmental sensor subsystem composed of temperature and humidity sensors, 105, an acoustic sensor, 106, a motion sensor, 107, and a light sensor, 109. Also in FIG. 1 is a representation of a register/damper unit, 101, that acts as a terminal unit and includes vent holes, 110, a duct skirt, 111, and one or more optional mounting holes for screws, 112.

FIG. 2 shows block drawing of an embodiment of the system with four zones, 97, with each zone having a Wall Sensor Unit, 100, a register/damper unit, 101, supply duct work 89, and a furnace or other HVAC plant, 99. Not shown in the drawing is a return duct to the HVAC plant, 99. A wired connection from a single wall sensor unit is shown to the HVAC plant, 93 or as an alternate embodiment a system controller 92. A duct pressure or flow sensor is shown 94. A fresh-air economizer is shown, 91, in the return air duct with a wireless control unit 92, controlling the flow direction. An outside air enthalpy sensor is shown, 90.

FIG. 3 shows the electronic components of a preferred embodiment of the wall senor unit shown in FIG. 1 that includes a display system sub system such as a LCD 102, one or more buttons, 103, an environmental sensor subsystem composed of temperature and humidity sensors, 105, an acoustic sensor, 106, a motion sensor, 107, a light sensor, 109, a RF antenna, 113, one ore more batteries, 114, one or more persistent data storage devices such as FLASH, battery backed SRAM, a hard disk drive or other persistent data storage device for holding the program and data, 115, a relay, solid state relay, direct logic level output or other digital interface for signaling HVAC devices, 116, analog signal conditioning and buffering circuits such as a transistor or MOSFT, 116, for controlling the HVAC interface, 116, the connection points such as a screw or push terminal to the HVAC equipment, 119, and one or more CPUs that support a wireless communication protocol, such as a ZigBee, and can support the basic control and I/O functions of an embedded processor.

FIG. 4 shows the electronic components of a preferred embodiment of the damper register unit shown in FIG. 1 that includes an actuator such as a DC motor, a solenoid, a shape memory alloy device, 121, a RF antenna, 113, one or more batteries, 114, one or more persistent data storage devices such as FLASH, battery backed SRAM, a hard disk drive or other persistent data storage device for holding the program and data, 115, a pair of analog signal conditioning and buffering circuits such as a transistor or MOSFT, 120, for controlling the actuator, 121 and one or more CPUs that support a wireless communication protocol, such as a ZigBee, and can support the basic control and I/O functions of an embedded processor.

FIG. 5 shows an embodiment of of a data structure for storing occupancy events. Each bit stores a history for a 15 minute block of time with each byte representing 8 15 minute time blocks or 2 hours of time. A value of 1 in a bit represents occupancy during that block of time and a 0 represents non-occupancy. The bytes are arranged continuously in chronological order with 17,531 bytes used to represent 4 calender years, 214. This calendar includes a leap year day. The calendar is arranged starting with January 1 in the first byte and with leap year as the last year. This calendar is stored in persistent memory for each zone such as is shown in 115 in FIG. 3. Occupancy is recorded in 15 minute increments and 8 samples (2 hours) are combined into a byte. A day requires 12 bytes (4 samples/hour, 24 hours per day=4*24=96 samples/day/8 bits=12 bytes). On a bit level data is recoded working from the MSb to the Lsb; that is, for the first sample of leap year the datum would occupy bit 7 of byte 0, the second sample would occupy bit 6 of byte 0, the eighth sample would occupy bit 0 of byte 0 and the 9th sample would occupy bit 7 of byte 1. Temperature Set/Reset Point Record Structure, 215, data is recorded as unsigned 8 bit data in 0.5 deg. F with 0 representing a null record. Therefore, a record of 1 represents 0.5 F, 2-117, 144=72F. The data is recored linearly every 15 minutes starting with 00:00h Sunday and ending with 23:45 Saturday and wrapped as needed. A null (0x00) value indicates that no value has been entered yet and the last valid value should be used, if none use the building default. A total of 672 bytes are reserved for this use. Error Log Record Structure, 216, The error log indicates if occupancy was not as calculated. It is recorded in a format similar to occupancy, but is only recorded for the previous 4 weeks in 2688 bytes. FIG. 6 indicates the function to parse & weigh likelihood of occupancy. As indicated in FIG. 7 this function is called with a string command where the first 16 bit unsigned integer (uintl6), 210, is the number of weight code value pairs following (excluding the size of the length value). Following the length value are pairs of 8 bit unsigned integers (uint8s), 211, that indicate the weight function shown as a hexadecimal number, 212, and weight value shown as a hexadecimal number, 213. The weight value, 213, is optimized by a genetic algorithm working in the background using the history of occupancy, 214, and the error log, 215, to evaluate the the fitness of the existing population (weight values) and the new population.

FIG. 8 is a flow chart of one embodiment of an algorithm for calculating a likelihood of an occupancy occurring during a given time slot. The input values are the initial date-time, DATE, and offset between previous dates-times, Offset, initial values are initiated to 0 with j used as a iteration counter and X used as a working value to build the result, 225, the value X is shifted left one bit, 226, the number of iterations is indicated by n, the boolean that represents the date-time indicated by Date is located in the data base, 228, and added to the working value X, 227, the date-time is decremented by the value indicated by the input, Offset, and wrapped around to the far end of the database if the date-time is less than the start of the database, 229, the iteration counter j is incremented, 230, if the iteration counter is greater than zero then we branch to 226, but when the iteration counter reaches zero the function terminates by returning the working value X, 231. Typical values for n would be 4 and for Offset would be 15 minutes, hour, day, month or year. Additionally, the Offset can be date logical, for example, decrementing by weekday such that that decrementing from a Monday would result in the previous Friday or decrementing by weekend-day such that that decrementing from a Sunday would result in the previous Saturday.

FIG. 9 represents a pattern of motion by an occupant. At the building level the system looks for multi-room patterns of usage. The system coordinator, 95 or 100, runs a background process to poll the occupancy records of each wall sensor unit for the previous 28 days looking for reproduced sequences of events that occur at approximately the same time. The system coordinator builds, maintains and stores a list of these events such that if an occupancy event occurs the subsequent zones are conditioned to the set point temperature.

FIG. 10 represents the data flow in an installation. The damper/registers, 101-A, 101-B, 101-C, 101-D, furnace sensors/controls, 92, 94, 95 and outside sensors 90, act as limited function end device, the wall sensor units, 100-A, 100-B and 100-C all act as at least a routers. In the displayed configuration either wall sensor 100A or furnace control 95 would act as the network coordinator. In a retrofit where an adequate existing control wire, 93, exists from the furnace/AC to a suitable location on the wall, a furnace control unit would not be needed. The arrows show how a network can form and how data can be routed through different nodes. These communication routes would be different according to the number of nodes required per building, the locations of the nodes and the RF characteristics of each building. The communication routes will change over time because of changes in the RF characteristics of the building. A link to a PC or other networking device, 91, allows for a web interface for displaying to the homeowner the current status of the system and to allow for dynamic load shedding and setback/set point changes, support of real-time metering and smart grid functions. As an alternative embodiment, the wireless communication between all or come of the wireless links can be replaced with a wired link such as CANBUS.

FIG. 11 represents the process that the coordinator goes through when a conditioning request arrives from a wall sensor or from the coordinator itself. When a request for conditioning arrives at the coordinator each of the other wall sensor units are required to determine their likelihood of requiring conditioning. Enough damper/registers are opened according to the likelihood of each zone requiring conditioning until enough open cross section is provided to allow for efficient energy transfer from the central plant, to keep noise to a minimum and to prevent damage to the duct work.

FIG. 12 represents an exploded view of the preferred embodiment of the damper/register. A top cover, 300, that has a plurality of vent holes for the passage of air, 303, and a plurality of screw mount holes, 304. A functional unit, 301, that has a plurality of vent holes for the passage of air, 303, that attaches to the bottom of the top cover has a duct skirt, 306 and a top mount, 305. The damper plate, 302, that has a plurality of vent holes for the passage of air, 303, and push tabs, 308 and 316.

FIG. 13 represents the mechanism that opens and closes the register/damper in FIG. 12. There are two metal pushers, 309 and 317, that pivot on studs 311 and 321 respectively. Two Shape Memory Alloy (SMA) wires, 310 and 320, are attached to the respective metal pushers and anchor points, 318 and 312. The SMA wires have lead wires, 319, 313, respectively. There is a metallic spring 315, that is connected to each of the pushers on the opposite side of the pivot from where the SMA wires are attached with a lead wire, 314, attached to the spring. A SMA wire contracts when heated and in this embodiment a current is run through the SMA wire lead, 319 or 313 through the respective SMA wire and pusher to the spring, 315 and out to the spring wire lead, 314. The relationship between the pushers and the damper plate, 302, and push tabs 308 and 316 is shown with a partial display of the damper plate. The damper plate in this figure is shown in an intermediate open/close position.

FIG. 14 represents the same components as shown in FIG. 13, but shows SMA wire 310 actively contracting and damper plate 302 in the open position.

FIG. 15 represents the same components as shown in FIG. 13, but shows SMA wire 320 actively contracting and damper plate 302 in the closed position.

FIG. 16 is a section view from above of the main register/damper piece, 301 at the level of the section shows the duct skirt, 306, from FIG. 13. Included in this figure are batteries, 322, the pivots 311 and 321, the anchor points, 312 and 318, the SMA wires, 319 and 320, the printed circuit board for the electronics, 323 and an internal mounting support 324. Not seen at this section level is the spring, 315 and the lead wires, 314, 313 and 319.

FIG. 17 represents a section view from the side of the register/damper from FIG. 12. The top cover, 300, the functional unit 301, and skirt 306, and the damper plate, 302, with push tabs 308 and 316, the batteries 322 and printed circuit board 323. The damper plate is shown in the closed position. The surfaces of the damper plate and functional unit that meet are roughed sufficiently to prevent uncommanded motion of the damper plate as indicated in the enlarged view indicated by 324.

FIG. 18 shows an alternative embodiment of the wall sensor unit, 100 from FIG. 1. As in FIG. 1 this embodiment includes a display sub system such as a LCD 102, one or more button, 103, an environmental sensor subsystem composed of temperature and humidity sensors, 105, an acoustic sensor, 106, a motion sensor, 107, and a light sensor, 109. Also included in this embodiment are the addition of three momentary switches 351, 352, 353 that are used for raising, toggling and dimming respectively a remote ZigBee light. The form factor is able to fit into a standard dual position modem style mud plate, 354 and standard two position junction box.

DETAILED DESCRIPTION

In the following description, for purposes of explanation and non-limitation, specific details are set forth, such as particular nodes, functional entities, techniques, protocols, standards, etc. in order to provide an understanding of the described technology. It will be apparent to one skilled in the art that other embodiments may be practiced apart from the specific details disclosed below. In other instances, detailed descriptions of well-known methods, devices, techniques, etc. are omitted so as not to obscure the description with unnecessary detail. Individual function blocks are shown in the figures. Those skilled in the art will appreciate that the functions of those blocks may be implemented using individual hardware circuits, using software programs and data in conjunction with a suitably programmed microprocessor or general purpose computer, using applications specific integrated circuitry (ASIC), and/or using one or more digital signal processors (DSPs). Generally speaking, the systems, methods, and techniques described herein may be implemented in digital electronic circuitry, computer hardware, firmware, software, or in combinations of these elements. Apparatus embodying these techniques may include appropriate input and output devices, a computer processor, and a computer program product tangibly embodied in a machinereadable storage device for execution by a programmable processor. A process embodying these techniques may be performed by a programmable processor executing a program or script of instructions to perform desired functions by operating on input data and generating appropriate output. The techniques may be implemented in one or more computer programs or scripts that are executable on a programmable system including at least one programmable processor coupled to receive data and instructions from, and to transmit data and instructions to, a data storage system, at least one input device, and at least one output device. Suitable processors include, by way of example, both general and special purpose microprocessors. Generally, a processor will receive instructions and data from a read-only memory and/or a random access memory. Storage devices suitable for tangibly embodying computer program instructions and data include all forms of volatile and non-volatile memory, including by way of example semiconductor memory devices, such as Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and Compact Disc Read-Only Memory (CD-ROM). Any of the foregoing may be supplemented by, or incorporated in, specially-designed ASICs (application-specific integrated circuits). The computer program instructions or scripts may also be provided as data signals embodied in a carrier wave or other propagation medium via a communication link (e.g., a modem or wired or wireless network connection).

This invention provides highly efficient, low cost and easy to install zoned HVAC controls for residential use. The system includes Wall Sensor Units (WSU) that superficially appear to be programmable thermostats, but include temperature/humidity, occupancy sensors, advanced software, processor, FLASH memory for recording occupancy information and a ZigBee wireless network capability. The system also includes ZigBee based wireless battery operated Damper/Register Units (DRUs) that are actuated by Shape Memory Alloy (SMA) wires. The preferred embodiment for the DRU includes two SMAs that when commanded each actuate a pusher that in turn each each presses on a push tab on the damper plate, and a single shared return spring. Also included in the system are additional components including ZigBee wireless outside air enthalpy sensors, ZigBee wireless duct pressure/flow sensors and HVAC plant controller where needed by the installation. Each of the components have an embedded MCU and ZigBee support. The system is intended to support zoned control and as each building the system is installed into is likely to be different the number and configuration of each installation will be different. Residential HVAC zones typical are based on a room that has a door that closes. There are some places where rooms are connected without doors that close such as is often found between kitchens and living rooms in open plan homes. In these cases where the air between rooms communicates well they may be treated as one zone. There are special cases in very large or unique rooms where the temperate difference between different places in the room can become uncomfortable, such as a long narrow room with large window only at one end, can be broken into more than one zone if there are HVAC supplies at each end that can be individually controlled. In each of these zones a WSU is placed on the wall and at least one DRU is placed at the exit of the HVAC source duct into the zone. If there is more that one HVAC outlet in a zone then each HVAC outlet has its own DRU. An exception to this is the case of a home, such as a studio, that is functionally a single zone then DRUs are not needed. The WSU in a zone controls each DRU in a zone via wireless ZigBee communication. Most communication between the WSU and DRU is from the WSU to the DRU ordering the DRU to open or close. The DRU will communicate back to the WSU open/close state confirmation, temperature, low battery and fault data. The DRUs act as a ZigBee End Device (ZED). Most WSU\'s function as ZigBee Router (ZR), but one WSU will act as a ZigBee coordinator (ZC) unless a HVAC central controller is present in which case the HVAC central controller acts as the ZC.

The WSU redundant occupancy sensors are able determine if the room is occupied. Redundant senors are used because often one time of sensor will not be able to determine if a human is present. For example, PIR sensors can only detect humans when they move and microphones can only detect their presence when they make noise. Conversely, occupancy sensor often have false positives. By combining the inputs from multiple sensor and filtering, a much more accurate picture of occupancy can be obtained. Each WSU records the occupancy of its zone in its FLASH memory in 15 minute increments with a rolling record for 4 years. The occupancy record is used to predict future occupancy. The occupancy record is examined using a Decaying Occupancy Temporal (DOT) Algorithm by looking at periodic occupancy records with decaying impact from older records. Occupancy is recorded as 1 if occupied and 0 in unoccupied. For example if looking at the past four days then eight times the occupancy value the same time one day before the target record is added with four times the occupancy value from two days before plus two times the value of occupancy record from three days before plus the occupancy value from four days before. This algorithm can be used over different time intervals such as past n 15 minutes, past n hours, past n days, past n months, past n years or time-logical intervals such as past n week days, past n weekend days, same day of month past n months. These various occupancy interval measurements are weighted and summed and compared against a user configurable economy/comfort threshold. The weighing values for the various occupancy interval measurements are determined by a genetic algorithm running in the background that uses the uses the weighing values as the genetic representation and performance with recent occupancy history as a fitness function to select the weighting values. Each WSU also keeps a record of set points, reset points and ventilation rates in FLASH in 15 minute increments based on a one week calendar that can be altered by the occupants.

If the occupancy sensors or thermal sensors in a WSU are triggered a request is sent to the ZC. Every 15 minutes the ZC sends a command to the WSUs to execute the occupancy prediction algorithm and returns any conditioning results to the ZC. If there is a conditioning request pending the ZC opens a sufficient number of DRUs to keep the pressure of the ducts low enough to maintain a high enough air flow for heat transfer, HVAC central plant safety and to prevent damage to the duct system. The DRUs to open is determined by current HVAC needs and by near future HVAC needs by the ZC querying the WSUs that will run their respective prediction algorithms for current and near future occupancy as well as distance from current set/setback point. From these values, the ZC opens the next most likely DRUs to open until enough are opened to provide adequate air flow in the duct network and to also provide minimum building ventilation. The WSU contain a full set of algorithms for setback, pre-cooling, fresh air economizers, load shedding, economic/comfort trade-offs and time of day meters optimization. The ZC also contains a full set of algorithms for controlling all standard HVAC plant types as well as relays for signaling the central HVAC plant.

The various embodiments described above can be combined to provide further embodiments. All of the U.S. patents, U.S. patent application publications, U.S. patent applications, foreign patents, foreign patent applications and non-patent publications referred to in this specification and/or listed in the Application Data Sheet are incorporated herein by reference, in their entirety. Aspects of the embodiments can be modified, if necessary to employ concepts of the various patents, applications and publications to provide yet further embodiments.

These and other changes can be made to the embodiments in light of the above-detailed description. In general, in the following claims, the terms used should not be construed to limit the claims to the specific embodiments disclosed in the specification and the claims, but should be construed to include all possible embodiments along with the full scope of equivalents to which such claims are entitled. Accordingly, the claims are not limited by the disclosure.