Distributed Sensing. Luca Benini, Davide Brunelli. Support from Artist-Design NoE & Telecom Italia

Transcription

1 Energy-Neutral Distributed Sensing Luca Benini, Davide Brunelli DEIS Università di Bologna Support from Artist-Design NoE & Telecom Italia

2 Building automation today Power grid distributed only to appliances All Sensors, controls and appliances connected by a single, low power bus (e.g. Knx, lonworks, BACnet, chorus,..) Modular, scalable and safe Deployment cost is high Retrofitting existing From GEWISS buildings is often impossible Modification of an existing layout is expensive and labor-intensive

3 Energy-Neutral Sensing & Control No power and data cable for sensors and controls Easy to install (in the optimal position) Retrofitting is feasible and inexpensive Battery-powered operation maintenance Energy-neutral harvest energy from the surrounding environment E(from power grid) = 0 Photovoltaic (outdoor, indoor) Inductive coupling RF energy Air flow Motion, vibration

4 Quantifying the challenge Harvesters Average Power 1W Consumers 1 cm 2 a-si PV in sun lit airplane pax window 1 in 2 TEG on crease beam 100 mw Zigbee mesh network node (w/ Rx from wireless sensor) Chipcon CC2500 radio (Tx mode) 10 mw Wireless dimming window TEG stringer clip 6 mm 2 TEG on hydraulic line TI MSP430 microprocessor (awake) Large inductive vibe harvesters 1 mw Push button harvester mw node: many on the market today! Wireless 1 Hz 1 cm 2 a-si PV in blue sky 1 cm 2 a-si PV in cabin lighting g Small piezo beam vibe harvesters 100 µw GSE monitoring sensor so (log data every 10sec, Tx 2X per day) 10 µw 1 µw Chipcon CC2500 radio (asleep) TI MSP430 microprocessor (asleep) 2.8 hrs interval

5 Sensor Node Evolution Non-E World 20 W 20 W 40 W 20 W 80 Mops 2nJ/b Sensor CE-ADC Processor PicoRadio DSP&storage Security MAC Power Mgr Energy Harvester Ambient energy RF Avg. Power <10kb/s 1% Objective: 100 µw Avg Energy neutrality becomes easy

6 High-efficiency PV harvesting 2000 Problems: Maximum Power point (MPP) Tracking MPP (MPPT) Low Power Budget (mw) Vc ( mv) J 15.7J 500 Goal: reduce PV-cell size 0 reduce storage device size increase the autonomy P-V chart T (s) 250 P[uW] P [mw] ,5 1 1,5 2 2,5 3 3,5 4 4,5 V[Volt] V Research supported by a grant of Telecom Italia

7 mw-level MPP tracking Vsolar Vctrl Vlow crossing switch off Vhigh crossing switch on Vsolar P V high Controlled variable V low, V high duty cycle V low Vsolar Online control for tracking PV curve variations with incident light, temperature

8 MPP tracking with a pilot PV cell Fractional Open Circuit Voltage technique V MPP (T,L) K FOCV V OC (T,L) K FOCV K T V T (T, L) V T V low, V high generation Comparator + driver Use V T from a micro PV cell to generate V low, V high V OC V MPP V T Minimize the cost of MPPT + tracker is unregulated Highest efficiency reported (until 2008)

9 Sub-mW harvesting WSN HW support a wide voltage supply range (usually between 1V and 4V ) Tmote Sky 2,1 3,6 V TinyNode 584 2,4 3,6 V TI Node 1,8 3,6V [µsolar scavenger 10mm 2 PV surface: UNIBO 07] Powering sensor nodes with unregulated and variable voltage supply from the solar cell adaptive Active-Recovery DC Minimize the energy used for DC/DC or linear regulation Automatically adapt duty-cycle with analog thresholds (comparators) on voltage supply Optimize thresholds for MPP in low-lighting condition (no tracking at high lighting as energy is over-abundant) Indoor PV powering is feasible!

10 Inductive Harvester Inductively powerered WSN Node + + Energy harvesting exploiting the EM field from AC electric current during idle (no measurement) times Fully energy-neutral neutral solution Research supported by a grant of Telecom Italia

11 Radio Frequency Harvester Energy harvested from RF waves, generated by a transmitter (wireless power transmission) Store the energy with supercapacitor like energy buffer RFID transmitter 868 MHz

12 Power Transfer Efficiency WISP WISP Power Cast Lessons Learned: Power levels are low (tens of µw) Advanced RF & Antenna design is needed

13 Windmill Harvester In-runner brushless higher number of turns higher output voltage Up to 10mW output Power Output power (mw) 10 Velocità Speed 11 8 Velocità Speed 22 Velocità Speed Load (Ω) MPPT with constant load at any wind speed.

14 Kinetic Harvesters Piezoelectric Size Weight Energy buffer Mean power (benchmark 2 Hz) Energy (1 min.) 9,8 x 5,7 x 3 cm ~120 g 4,7 μf 18 μw 1,1 mj Electromechanical Size Weight Energy buffer Mean power (benchmark 2 Hz) Energy (1 min.) 6,5 x 2,5 x 2,5 cm ~80 g 4700 μf 206 μw 12,4 mj Motion frequency spectrum is a key design parameter! Research supported by a grant of Telecom Italia

15 Market Outlook PV: quite mature, with many products Flexible PV materials are interesting e.g. Solution provides (Piezo, kinetic, solar) (EM kinetic) (thermal) (RF transmission) (Piezo) and many others [Yole 09]

16 Energy Storage Technologies Options Secondary batteries Capacitors Flywheels Fuel cell Tradeoffs Batteries Mature technology, high energy density, less efficient, limited to few hundred full recharging cycles (significantly more shallow cycles) Ultracapacitors (up to hundreds of Farads) Virtually infinite recharge cycles, higher leakage current (goes up with size) Configurations: Battery-only, Capacitor-only, Tiered Capacitor+Battery

17 Energy Neutral System Design Evolution of design techniques and tools Low Power Design Energy Aware Design Battery Aware Design Energy Neutral design Hardware Design Conversion efficiency Impedance Matching Maximum power transferred Software Design Scheduling algorithm Adaptive duty cycle Energy prediction algorithm

18 Harvester-friendly Circuits Vgen [Amirtharajah06] Vrct 1 2 ENERGY SELF-TIMED RECOVERY LOGIC LOGIC Logic styles for AC power supplies: self-timed circuits and energy recovery ULP CMOS to avoid charge pumps

19 Managing Harvested Energy It is different from battery energy Supply varies in time Sometimes is scarce, sometime over-abundant Supply varies in space Different nodes get different energy: need (dynamic) load migration Supply is repetitive (does not die out) & predictable Opportunity for predictive & adaptive management techniques

20 Summary Energy neutral sensors (actuators) are promising i for many smart & efficient energy applications Energy Harvesting and power storage devices are key enablers All system stem components need to be considered: not only a HW design challenge Distributed energy management is the frontier of system research