Seminar Course 392 N Spring 2011 Lecture

Seminar Course 392 N ● Spring 2011 Lecture 3 Intelligent Energy Systems: Control and Monitoring Basics Dimitry Gorinevsky ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 1

Traditional Grid • Worlds Largest Machine! – 3300 utilities – 15, 000 generators, 14, 000 TX substations – 211, 000 mi of HV lines (>230 k. V) • A variety of interacting control systems ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 2 2

Smart Energy Grid Intelligent Energy Network Source IPS energy subnet Load IPS Intelligent Power Switch Generation Transmission Distribution Load Conventional Electric Grid Conventional Internet ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 3

Intelligent Energy Applications Tablet Computer Smart phone Communications Internet Energy Application Presentation Layer Application Logic Business Logic (Intelligent Functions) Database ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 4

Control Function • Control function in a systems perspective ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 5

Analysis of Control Function • Control analysis perspective • Goal: verification of control logic – Simulation of the closed-loop behavior – Theoretical analysis ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 6

Key Control Methods • Control Methods – Design patterns – Analysis templates • • • P (proportional) control I (integral) control Switching control Optimization Cascaded control design ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 7

Generation Frequency Control • Example control command Controller sensor measurements Turbine /Generator disturbance Load ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 8

Generation Frequency Control • Simplified classic grid frequency control model – Dynamics and Control of Electric Power Systems, G. Andersson, ETH Zurich, 2010 http: //www. eeh. ee. ethz. ch/en/eeh/education/courses/viewcourse/227 -0528 -00 l. html Swing equation: ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 9

P-control • P (proportional) feedback control • Closed –loop dynamics x+ 0 frequency droop Step response • Steady state error u 1 0. 8 0. 6 0. 4 0. 2 0 x(t) 0 2 4 frequency droop ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 10

AGC Control Example • AGC = Automated Generation Control • AGC frequency control generation command AGC frequency measurement disturbance ee 392 n - Spring 2011 Stanford Intelligent Energy Systems Load 11

AGC Frequency Control • Frequency control model – x is frequency error – cl is frequency droop for load l – u is the generation command • Control logic – I (integral) feedback control • This is simplified analysis ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 12

P and I control • P control of an integrator d b x -kp • I control of a gain system. The same feedback loop cl x g -k. I ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 13

Cascade (Nested) Loops • Inner loop has faster time scale than outer loop • In the outer loop time scale, consider the inner loop as a gain system that follows its setpoint input outer loop setpoint (command) - Outer Loop Control inner loop setpoint - Plant output outer loop ee 392 n - Spring 2011 Stanford Inner Loop Control Intelligent Energy Systems inner loop 14

Switching (On-Off) Control • State machine model – Hides the continuous-time dynamics – Continuous-time conditions for switching • Simulation analysis – Stateflow by Mathworks off setpoint ee 392 n - Spring 2011 Stanford on passive cooling furnace heating Intelligent Energy Systems 15

Optimization-based Control • Is used in many energy applications, e. g. , EMS • Typically, LP or QP problem is solved – Embedded logic: at each step get new data and compute new solution Optimization Problem Formulation Measured Data Sensors ee 392 n - Spring 2011 Stanford Embedded Optimizer Solver Plant Intelligent Energy Systems Control Variables Actuators 16

Cascade (Hierarchical) Control • Hierarchical decomposition – Cascade loop design – Time scale separation ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 17

Hierarchical Control Examples • Frequency control – I (AGC) P (Generator) • ADR – Automated Demand Response – Optimization Switching • Energy flow control in EMS – Optimization PI • Building control: – PI Switching – Optimization ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 18

Power Generation Time Scales • Power generation and distribution • Energy supply side 1/10 100 Power Supply Scheduling 1000 Time (s) http: //www. eeh. ee. ethz. ch/en/eeh/education/courses/viewcourse/227 -0528 -00 l. html ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 19

Power Demand Time Scales • Power consumption – DR, Homes, Buildings, Plants • Demand side Enterprise Demand Scheduling Building HVAC Home Thermostat Demand Response 100 ee 392 n - Spring 2011 Stanford 1, 000 10, 000 Intelligent Energy Systems Time (s) 20

Research Topics: Control • Potential topics for the term paper. • Distribution system control and optimization – Voltage and frequency stability – Distributed control for Distributed Generation – Distribution Management System: energy optimization, DR ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 21

Monitoring & Decision Support • Open-loop functions - Data presentation to a user ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 22

Monitoring Goals • Situational awareness – Anomaly detection – State estimation • Health management – Fault isolation – Condition based maintenances ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 23

Condition Based Maintenance • CBM+ Initiative ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 24

SPC: Shewhart Control Chart W. Shewhart, Bell Labs, 1924 Statistical Process Control (SPC) UCL = mean + 3· LCL = mean - 3· Upper Control Limit quality variable • • mean 3 ee 392 n - Spring 2011 Stanford 6 9 sample 12 15 Walter Shewhart (1891 -1967) Lower Control Limit Intelligent Energy Systems 25

Multivariable SPC • Two correlated univariate processes y 1(t) and y 2(t) cov(y 1, y 2) = Q, Q-1= LTL • Uncorrelated linear combinations z(t) = L·[y(t)- ] • Declare fault (anomaly) if ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 26

Multivariate SPC - Hotelling's 2 T • Empirical parameter estimates • Hotelling's T 2 statistics is Harold Hotelling (1895 -1973) • T 2 can be trended as a univariate SPC variable ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 27

Advanced Monitoring Methods • Estimation is dual to control – SPC is a counterpart of switching control • Predictive estimation – forecasting, prognostics – Feedback update of estimates (P feedback EWMA) • Cascaded design – Hierarchy of monitoring loops at different time scales • Optimization-based methods – Optimal estimation ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 28

Research Topics: Monitoring • Potential topics for the term paper. • Asset monitoring – Transformers • Electric power circuit state monitoring – Using phasor measurements – Next chart ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 29

Electric Power Circuit Monitoring model Optimization Problem Electric Power System State estimate • Fault isolation Measurements: • Currents • Voltages • Breakers, relays ee 392 n - Spring 2011 Stanford ACC, 2009 Intelligent Energy Systems 30

End of Lecture 3 ee 392 n - Spring 2011 Stanford Intelligent Energy Systems 31