DATA ANALYSIS APPLICATIONS

• US 20110125461A1
• Filed: 02/04/2011
• Published: 05/26/2011
• Est. Priority Date: 11/17/2005
• Status: Active Grant

• Current Assignee: Florida Power & Light Company (NextEra Energy, Inc.)
• Original Assignee: Florida Power & Light Company (NextEra Energy, Inc.)
• Inventors: Moncrief, Gary JR., Chaney, Jennifer D., Smith, Donville, More, Eduardo J., Suero, Marcos
• Granted Patent: US 8,170,835 B2
• Time in Patent Office: Days
• US Class Current: 702/179
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs

Abstract

A system and method of monitoring equipment performance and predicting failures. The system can include a data historian that stores data for a piece of equipment and designates the data to tags. The tags can correspond to sensors that gather the data from the piece of equipment. A matrix model builder application can allow a user to generate regression models for various time spans to determine whether new data is within a normal operating range.

Claims

1-70. -70. (canceled)

71. A method of monitoring equipment performance and predicting failures, the method comprising:

receiving historical data for a piece of equipment;

designating the historical data with at least two input tags;

selecting a first time span for the at least two input tags from the historical data;

generating a regression model including an expected output tag for the first time span of the at least two input tags;

gathering new data from the piece of equipment corresponding to the at least two input tags;

selecting a second time span for the at least two input tags from the new data; and

comparing the second time span with the new data to the expected output tag to determine whether the new data is within a normal operating range.

72. The method of claim 71 and further comprising generating a polynomial regression model including at least one of a linear regression model, a quadratic regression model, and a cubic regression model.

73. The method of claim 71 and further comprising generating a regression model including a standard deviation of the data and an equation of a surface of the data.

74. The method of claim 71 and further comprising selecting and plotting upper and lower bounds for the normal operating range for the piece of equipment.

75. The method of claim 74 and further comprising displaying a percentage of the new data that is included within the upper and lower bounds.

76. The method of claim 71 and further comprising plotting at least one of the first time span of historical data, the second time span of new data, upper bounds, and lower bounds.

77. The method of claim 76 and further comprising plotting and graphically differentiating the historical data and the new data.

78. The method of claim 71 and further comprising designating the historical data and the new data with a plurality of tags.

79. The method of claim 71 and further comprising receiving historical data for a piece of equipment from a data historian.

80. The method of claim 71 and further comprising generating a regression model for the at least two input tags, the regression model providing an expected output tag for a normal operating range for the piece of equipment.

81. The method of claim 71 and further comprising altering the first time span of the historical data and generating a new regression model.

82. The method of claim 71 and further comprising plotting at least one of the first time span of the historical data and the second time span of the new data on at least one of a two dimensional plot and a three-dimensional plot.

83. The method of claim 71 and further comprising altering the at least two input tags to generate a new regression model with a new expected output tag.

84. The method of claim 71 and further comprising generating a plurality of regression models and choosing one regression model to apply to the second time span of the new data.

85. The method of claim 71 and further comprising generating and saving a plurality of regression models for characterizing normal performance for the piece of equipment.

86. The method of claim 71 and further comprising loading the regression model into an equation processor, retrieving the new data, calculating an actual output using the regression model and the equation processor, and writing the actual output to a database.

87. The method of claim 86 and further comprising at least one of monitoring the actual output in real-time and comparing the actual output to the expected output tag in real-time.

88. The method of claim 71 and further comprising initializing an alarm signal when an actual output of the new data varies from the expected output tag.

89. The method of claim 88 and further comprising transmitting an alarm notification to at least one user based on the alarm signal.

90. The method of claim 88 and further comprising color-coding a parameter block displayed on an alarm screen based on a severity of the alarm signal.

91. The method of claim 90 and further comprising displaying a plurality of color-coded parameter blocks corresponding to a plurality of parameters on the alarm screen.

92. The method of claim 91 and further comprising selecting one of the plurality of color-coded parameter blocks in order to alter alarm settings for one of the plurality of parameters.

93. The method of claim 88 and further comprising manually specifying at least one of temporary and permanent alarm settings.

94. The method of claim 88 and further comprising automatically specifying alarm bounds by applying a set threshold at least one of above and below a startup value.

95. The method of claim 71 and further comprising viewing a trend of at least one of the new data and the historical data by plotting data for a plurality of time ranges on a single plot.

96. The method of claim 71 and further comprising providing a statistical summary of a plurality of tags from at least one of the new data and the historical data.

97. The method of claim 96, wherein the statistical summary comprises at least one of a capability chart, a box plot, and a histogram.

98. The method of claim 97, wherein the capability chart comprises a grid of color-coded parameters grouped by a plurality of categories, each one of the color-coded parameters including at least one of a process capability value, a mean value, and a standard deviation value.

99. The method of claim 71 and further comprising gathering new data from at least one of a turbine, a compressor, a combustion chamber, an air preheater, a boiler feed pump, a heating element, and a cooling tower.

100. The method of claim 71 and further comprising generating a plurality of regression models and sorting the plurality of regression models by at least one of plant, system, and contact person.

101. The method of claim 71 and further comprising comparing an actual output of the new data to the expected output tag and generating at least one of a deviation scaled plot and an actual value scaled plot.

102. The method of claim 71 and further comprising generating a crew capability report to compare performance among various employee crews.

103. The method of claim 71 and further comprising tracking deviations between actual outputs of the new data and expected output tags by filtering and normalizing multiple tags and by displaying multiple charts on a single screen.

104. The method of claim 71 and further comprising tracking deviations between actual outputs of the new data and expected output tags by filtering and normalizing multiple tags and by generating at least one of a polar plot, an Xbar-R chart, and a histogram.

105. The method of claim 71 and further comprising monitoring at least one of quantity and duration of excursions of an actual output of the new data from the expected output tag.

106. The method of claim 105 and further comprising generating an excursion notification when at least one of the quantity and the duration of the excursion exceeds a threshold.

107. The method of claim 95, wherein the statistical summary comprises at least one of a capability chart, a box plot, and a histogram.

108. The method of claim 107, wherein the capability chart comprises a grid of color-coded parameters grouped by a plurality of categories, each one of the color-coded parameters including at least one of a process capability value, a mean value, and a standard deviation value.