Power plant early warning device and method employing multiple prediction model

Patent metadata

Field	Value
Publication number	US-11960271-B2
Application number	US-201917279118-A
Country	US
Kind code	B2
Filing date	Oct 10, 2019
Priority date	Oct 12, 2018
Publication date	Apr 16, 2024
Grant date	Apr 16, 2024

Abstract

The present invention relates to a power plant early warning device and method which employ a multiple prediction model, and includes: a plurality of prediction models which receive information about the operation states of equipment in a power plant, and output prediction values and the reliabilities of the prediction values; a reliability analysis module which quantizes and analyzes the reliabilities output from the plurality of prediction models to determine the ranks of the plurality of prediction models, and calculates a final prediction value; a comparison module which compares the final prediction value with an actual measurement value, and outputs a residual; and a determination module which analyzes the residual to determine whether there are defects in the operating states of the equipment in the power plant.

First claim

The invention claimed is:

1. An apparatus comprising a power plant early warning device for installation in a power plant, the power plant early warning device being configured for determining whether there exist defects in operation states of equipment in the power plant, wherein the power plant early warning device comprises a signal measurement unit that receives monitoring signals that are related to operating states of equipment in the power plant, the operating states including one of pressure, temperature, and flow data, the signals being analog signals, wherein the signal measurement unit is attached to the equipment in the power plant to measure and transmit operating states of the equipment during each of a plurality of transmission intervals and wherein the signal measurement unit carries out analog-to-digital conversion to transmit a digital signal, wherein the power plant early warning device is configured to receive information on operating states of equipment of a power plant from the signal measurement unit and to output at least three prediction values and at least three reliabilities, each of which corresponds to one of the prediction values, to quantify and analyze the reliabilities to determine at least three ranks, each of which corresponds to one of the three prediction values, to calculate a final prediction value by applying weights according to the ranks of the prediction values, to compare the final prediction value with an actual measurement value provided by the signal measurement unit, to output a residual, to analyze the residual to determine whether a defect exists in an operating state of the equipment of the power plant, and to cause a warning to occur when the residual is outside of a normal range.

2. The apparatus of claim 1 , wherein among the ranks of the prediction values is a highest rank and wherein applying weights according to the ranks of the prediction values comprises using the highest rank when calculating the final prediction value.

3. The apparatus of claim 1 , wherein the final prediction value is calculated using an arithmetic mean of two highest ranking prediction values.

4. The apparatus of claim 1 , wherein all the prediction values are determined by using a particular algorithm and wherein the algorithm is applied differently for each operating state.

5. The apparatus of claim 1 , wherein the prediction values are determined using different algorithms.

6. The apparatus of claim 1 , further comprising the power plant.

Assignees

• Korea Hydro & Nuclear Power Co

Inventors

• Min Ji Ho
• Kim Dae Woong
• Lee Do Hwan
• Kim Yun Goo

Classifications

• G05B23/0254 Primary

  based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks · CPC title

• G05B23/0283

  Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] (preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06) · CPC title

• H02J3/004

  Generation forecast, e.g. methods or systems for forecasting future energy generation · CPC title

• G08B29/183 Primary

  Single detectors using dual technologies (G08B13/1672, G08B13/2448, G08B13/2494 take precedence) · CPC title

• G06Q10/04

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title