# US 5,708,590 A · Method and apparatus for real time recursive parameter energy management system

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 5,708,590 A
- Document ID: US5708590A
- Status: Granted
- Kind: A
- Application number: 08/761,033
- Filed: 1996-12-05
- Granted: 1998-01-13
- Earliest claimed: 1995-04-17
- Assignees: Siemens Energy & Automation, Inc.
- Inventors: Ilya Slutsker; Sasan Mokhtari; Kevin Clements
- Web page: https://patentgrep.com/patent/US5708590A

## CPC

- G01R 21/133: by using digital technique
- H02J 3/0014: for preventing or reducing power oscillations in networks
- G01R: MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES (indicating correct tuning of resonant circuits H03J3/12)
- H02J: ELECTRIC POWER NETWORKS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY

## Abstract

A new method and apparatus for recursive parameter estimation which closely approximates actual values is taught. Briefly stated, the within invention estimates impedance parameters of network branches in both on-line and off-line modes, providing accurate estimation of branch parameters in the presence of noise in measurements and rejecting gross measurement errors, by use of a Kalman type Markov filter scheme using maximum likelihood estimation.

## Claims

### Claim 1

An energy management system for recursive parameter energy management control, comprising:
control means for changing the settings of a power grid or subsystem (grid);
measurement means for measuring the status and settings of said (grid) and producing representative grid data thereof;
a computer having memory means for storing said grid data and further having a program means for analyzing said grid data and producing values representative of the impedance parameters of said grid, said program means including means for separating said grid data into a plurality of sub-grid data sets, each of said sub-grid data sets containing a predetermined number of unknown grid impedance parameters and means for recursively modeling each of said sub-grid data sets into Markov processes; and
comparing the impedance parameters of said grid which were derived from Markov processes recursively modeling said sub-grid data sets a predetermined grid impedance parameters.

### Claim 2 (depends on 1)

An energy management system according to claim 1 wherein said control means changes said settings of said grid in response to said comparison of said derived impedance parameters with said predetermined grid impedance parameters.

### Claim 3 (depends on 1)

An energy management system according to claim 1 wherein said grid data is comprised of at least voltage magnitudes and angles and wherein said program means treats said voltage magnitudes and angles as noncorrelated in time where X(t.sub.i +1) does not depend on X(t.sub.i).

### Claim 4 (depends on 3)

An energy management system according to claim 3 wherein the model for X is such that it is a discrete time white noise random process obeying the difference equation:
X(t.sub.i -1)=X(t.sub.i)+.omega.(t.sub.i)
where .omega.(t.sub.i) is discrete time gaussian white noise.

### Claim 5 (depends on 1)

An energy management system according to claim 1 wherein unknown grid impedance parameters are modeled to said Markov processes according to:
P(t.sub.i +1)=P(t.sub.i)+.nu.(t.sub.i)
where .nu.(ti) is discrete time gaussian white noise.