Systems and methods for data mining of energy consumption data

• US 8,989,910 B1
• Filed: 12/30/2011
• Issued: 03/24/2015
• Est. Priority Date: 12/31/2010
• Status: Active Grant

• Current Assignee: C3AI Incorporated
• Original Assignee: C3, Inc. (Charles & Colvard (HK) Ltd.)
• Inventors: Klots, Boris
• Primary Examiner(s): KASENGE, CHARLES R
• Application Number:
  US13/341,014
• Time in Patent Office: 1,180 Days
• Field of Search: 700/286, 700/291, 700/295, 702/60, 702/61, 702/179, 702/181, 713/340
• US Class Current: 700/291
• CPC Class Codes: G06Q 50/06 Energy or water supply
• CPC Class Codes: H02J 2203/20 Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
• CPC Class Codes: H02J 2310/54 according to a pre-established time schedule
• CPC Class Codes: H02J 3/144 Demand-response operation of the power transmission or distribution network
• CPC Class Codes: Y02B 70/3225 Demand response systems, e.g. load shedding, peak shaving
• CPC Class Codes: Y02E 60/00 Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
• CPC Class Codes: Y04S 10/50 Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
• CPC Class Codes: Y04S 20/222 Demand response systems, e.g. load shedding, peak shaving
• CPC Class Codes: Y04S 40/20 Information technology specific aspects, e.g. CAD, simulation, modelling, system security

Abstract

Methods, machine-readable media, apparatuses and systems are provided to collect and analyze data relating to energy consumption and factors affecting energy production. In various embodiments, such analysis can include time series analysis, multivariate correlation analysis, cluster analysis, anomaly detection methods, multi-dimensional density analysis and demand modeling.

Claims

1. A method comprising:

receiving, by a computing device, data corresponding to energy consumption for a time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing, by the computing device, a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying, by the computing device, an outlier in the combined distribution; and

generating, by the computing device, a representation of the outlier in the combined distribution,

wherein the outlier is identified using a K-nearest neighbor technique.

2. The method of claim 1 further comprising:

identifying, by the computing device, clusters in the combined distribution;

building, by the computing device, a model of previously established patterns of energy consumption based on the clusters;

receiving, by the computing device, data corresponding to energy consumption for a second time period;

receiving, by the computing device, data corresponding to the at least one factor that affects energy consumption for the second time period;

determining, by the computing device, that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating, by the computing device, a representation of the at least a portion of the data corresponding to energy consumption for the second time period.

3. The method of claim 2 wherein the clusters are identified using density estimations.

4. The method of claim 2 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

5. The method of claim 2 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

6. The method of claim 2 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

7. The method of claim 2 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

8. The method of claim 2 wherein the steps are practiced sequentially.

9. The method of claim 2 wherein the clusters are identified using methods of K-nearest neighbor estimation.

10. The method of claim 2 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption in the second time period represents data relating to a second set of equipment.

11. The method of claim 2 further comprising:

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

12. The method of claim 3 wherein the density estimations include non-parametric kernel density estimation.

13. The method of claim 1 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature weekends, holidays, seasons, and heating degree days.

14. The method of claim 1 wherein the outlier is further identified using at least one of regression analysis, density estimates, and proximity metrics.

15. A method comprising:

receiving, by a computing device, data corresponding to energy consumption for a time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the time period, wherein the at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing, by the computing device, a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying, by the computing device, an outlier in the combined distribution;

generating, by the computing device, a representation of the outlier in the combined distribution;

identifying, by the computing device, clusters in the combined distribution;

building, by the computing device, a model of previously established patterns of energy consumption based on the clusters;

receiving, by the computing device, data corresponding to energy consumption for a second time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the second time period;

determining, by the computing device, that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating, by the computing device, a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the clusters are identified using density estimations and wherein the density estimations include non-parametric kernel density estimation.

16. The method of claim 15 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, weekends, holidays, seasons, and heating degree days.

17. The method of claim 15 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

18. The method of claim 15 wherein the clusters are identified using methods of K-nearest neighbor estimation.

19. The method of claim 15 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

20. The method of claim 15 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

21. The method of claim 15 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

22. The method of claim 15 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

23. The method of claim 15 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

24. The method of claim 15 wherein the steps are practiced sequentially.

25. The method of claim 15 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

26. A method comprising:

receiving, by a computing device, data corresponding to energy consumption for a time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the time period, wherein the at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing, by the computing device, a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying, by the computing device, an outlier in the combined distribution;

generating, by the computing device, a representation of the outlier in the combined distribution;

identifying, by the computing device, clusters in the combined distribution;

building, by the computing device, a model of previously established patterns of energy consumption based on the clusters;

receiving, by the computing device, data corresponding to energy consumption for a second time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the second time period;

determining, by the computing device, that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating, by the computing device, a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the clusters are identified using methods of K-nearest neighbor estimation.

27. The method of claim 26 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, weekends, holidays, seasons, and heating degree days.

28. The method of claim 26 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

29. The method of claim 26 wherein the clusters are identified using density estimations.

30. The method of claim 26 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

31. The method of claim 26 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

32. The method of claim 26 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

33. The method of claim 26 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

34. The method of claim 26 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

35. The method of claim 26 wherein the steps are practiced sequentially.

36. The method of claim 26 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

37. A method comprising:

receiving, by a computing device, data corresponding to energy consumption for a time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing, by the computing device, a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying, by the computing device, an outlier in the combined distribution;

generating, by the computing device, a representation of the outlier in the combined distribution;

identifying, by the computing device, clusters in the combined distribution;

building, by the computing device, a model of previously established patterns of energy consumption based on the clusters;

receiving, by the computing device, data corresponding to energy consumption for a second time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the second time period;

determining, by the computing device, that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating, by the computing device, a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

38. The method of claim 37 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, weekends, holidays, seasons, and heating degree days.

39. The method of claim 37 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

40. The method of claim 37 wherein the clusters are identified using density estimations.

41. The method of claim 37 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

42. The method of claim 37 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

43. The method of claim 37 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

44. The method of claim 37 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

45. The method of claim 37 wherein the steps are practiced sequentially.

46. The method of claim 37 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

47. A method comprising:

receiving, by a computing device, data corresponding to energy consumption for a time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing, by the computing device, a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying, by the computing device, an outlier in the combined distribution;

generating, by the computing device, a representation of the outlier in the combined distribution;

identifying, by the computing device, clusters in the combined distribution;

building, by the computing device, a model of previously established patterns of energy consumption based on the clusters;

receiving, by the computing device, data corresponding to energy consumption for a second time period;

receiving, by the computing device, data corresponding to at least one factor that affects energy consumption for the second time period;

determining, by the computing device, that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption;

generating, by the computing device, a representation of the at least a portion of the data corresponding to energy consumption for the second time period;

building, by the computing device, a demand model of energy consumption per cluster;

forecasting, by the computing device, energy consumption for a third time period using the demand model; and

generating, by the computing device, a representation of the forecasted energy consumption for the third time period.

48. The method of claim 47 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, weekends, holidays, seasons, and heating degree days.

49. The method of claim 47 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

50. The method of claim 47 wherein the clusters are identified using density estimations.

51. The method of claim 47 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

52. The method of claim 47 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

53. The method of claim 47 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

54. The method of claim 47 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

55. The method of claim 47 wherein the steps are practiced sequentially.

56. A system comprising:

at least one processor; and

a memory storing instructions configured to instruct the at least one processor to perform;

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution; and

generating a representation of the outlier in the combined distribution,

wherein the outlier is identified using a K-nearest neighbor technique.

57. The system of claim 56 further comprising:

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to the at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period.

58. The system of claim 57 wherein the clusters are identified using density estimations.

59. The system of claim 58 wherein the density estimations include non-parametric kernel density estimation.

60. The system of claim 57 wherein the clusters are identified using methods of K-nearest neighbor estimation.

61. The system of claim 57 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption in the second time period represents data relating to a second set of equipment.

62. The system of claim 57 further comprising:

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

63. The system of claim 57 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

64. The system of claim 57 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

65. The system of claim 57 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

66. The system of claim 57 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

67. The system of claim 57 wherein the steps are practiced sequentially.

68. The system of claim 56 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

69. The system of claim 56 wherein the outlier is further identified using at least one of regression analysis, density estimates, and proximity metrics.

70. A system comprising:

at least one processor; and

a memory storing instructions configured to instruct the at least one processor to perform;

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the clusters are identified using density estimations and wherein the density estimations include non-parametric kernel density estimation.

71. The system of claim 70 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

72. The system of claim 70 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

73. The system of claim 70 wherein the clusters are identified using methods of K-nearest neighbor estimation.

74. The system of claim 70 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

75. The system of claim 70 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

76. The system of claim 70 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

77. The system of claim 70 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

78. The system of claim 70 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

79. The system of claim 70 wherein the steps are practiced sequentially.

80. The system of claim 70 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

81. A system comprising:

at least one processor; and

a memory storing instructions configured to instruct the at least one processor to perform;

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the clusters are identified using methods of K-nearest neighbor estimation.

82. The system of claim 81 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

83. The system of claim 81 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

84. The system of claim 81 wherein the clusters are identified using density estimations.

85. The system of claim 81 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

86. The system of claim 81 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

87. The system of claim 81 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

88. The system of claim 81 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

89. The system of claim 81 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

90. The system of claim 81 wherein the steps are practiced sequentially.

91. The system of claim 81 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

92. A system comprising:

at least one processor; and

a memory storing instructions configured to instruct the at least one processor to perform;

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

93. The system of claim 92 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

94. The system of claim 92 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

95. The system of claim 92 wherein the clusters are identified using density estimations.

96. The system of claim 92 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

97. The system of claim 92 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

98. The system of claim 92 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

99. The system of claim 92 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

100. The system of claim 92 wherein the steps are practiced sequentially.

101. The system of claim 92 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

102. A system comprising:

at least one processor; and

a memory storing instructions configured to instruct the at least one processor to perform;

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption;

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

103. The system of claim 102 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

104. The system of claim 102 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

105. The system of claim 102 wherein the clusters are identified using density estimations.

106. The system of claim 102 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

107. The system of claim 102 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

108. The system of claim 102 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

109. The system of claim 102 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

110. The system of claim 102 wherein the steps are practiced sequentially.

111. A non-transitory computer storage medium storing computer executable instructions that, when executed, cause a computer system to perform a computer implemented method comprising:

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution; and

generating a representation of the outlier in the combined distribution,

wherein the outlier is identified using a K-nearest neighbor technique.

112. The non-transitory computer storage medium of claim 111 further comprising:

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to the at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period.

113. The non-transitory computer storage medium of claim 112 wherein the clusters are identified using density estimations.

114. The non-transitory computer storage medium of claim 113 wherein the density estimations include non-parametric kernel density estimation.

115. The non-transitory computer storage medium of claim 112 wherein the clusters are identified using methods of K-nearest neighbor estimation.

116. The non-transitory computer storage medium of claim 112 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption in the second time period represents data relating to a second set of equipment.

117. The non-transitory computer storage medium of claim 112 further comprising:

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

118. The non-transitory computer storage medium of claim 112 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

119. The non-transitory computer storage medium of claim 112 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

120. The non-transitory computer storage medium of claim 112 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

121. The non-transitory computer storage medium of claim 112 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

122. The non-transitory computer storage medium of claim 112 wherein the steps are practiced sequentially.

123. The non-transitory computer storage medium of claim 111 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

124. The non-transitory computer storage medium of claim 111 wherein the outlier is further identified using at least one of regression analysis, density estimates, and proximity metrics.

125. A non-transitory computer storage medium storing computer executable instructions that, when executed, cause a computer system to perform a computer implemented method comprising:

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the clusters are identified using density estimations and wherein the density estimations include non-parametric kernel density estimation.

126. The non-transitory computer storage medium of claim 125 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

127. The non-transitory computer storage medium of claim 125 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

128. The non-transitory computer storage medium of claim 125 wherein the clusters are identified using methods of K-nearest neighbor estimation.

129. The non-transitory computer storage medium of claim 125 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

130. The non-transitory computer storage medium of claim 125 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

131. The non-transitory computer storage medium of claim 125 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

132. The non-transitory computer storage medium of claim 125 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

133. The non-transitory computer storage medium of claim 125 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

134. The non-transitory computer storage medium of claim 125 wherein the steps are practiced sequentially.

135. The system of claim 125 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

136. A non-transitory computer storage medium storing computer executable instructions that, when executed, cause a computer system to perform a computer implemented method comprising:

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the clusters are identified using methods of K-nearest neighbor estimation.

137. The non-transitory computer storage medium of claim 136 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

138. The non-transitory computer storage medium of claim 136 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

139. The non-transitory computer storage medium of claim 136 wherein the clusters are identified using density estimations.

140. The non-transitory computer storage medium of claim 136 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

141. The non-transitory computer storage medium of claim 136 wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

142. The non-transitory computer storage medium of claim 136 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

143. The non-transitory computer storage medium of claim 136 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

144. The non-transitory computer storage medium of claim 136 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

145. The non-transitory computer storage medium of claim 136 wherein the steps are practiced sequentially.

146. The non-transitory computer storage medium of claim 136 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

147. A non-transitory computer storage medium storing computer executable instructions that, when executed, cause a computer system to perform a computer implemented method comprising:

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption; and

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period,

wherein the data corresponding to energy consumption for the time period represents data relating to a first set of equipment and the data corresponding to energy consumption for the second time period represents data relating to a second set of equipment.

148. The non-transitory computer storage medium of claim 147 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

149. The non-transitory computer storage medium of claim 147 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

150. The non-transitory computer storage medium of claim 147 wherein the clusters are identified using density estimations.

151. The non-transitory computer storage medium of claim 147 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

152. The non-transitory computer storage medium of claim 147 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

153. The non-transitory computer storage medium of claim 147 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

154. The non-transitory computer storage medium of claim 147 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

155. The non-transitory computer storage medium of claim 147 wherein the steps are practiced sequentially.

156. The non-transitory computer storage medium of claim 147 further comprising:

identifying a plurality of clusters in the combined distribution;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

157. A non-transitory computer storage medium storing computer executable instructions that, when executed, cause a computer system to perform a computer implemented method comprising:

receiving data corresponding to energy consumption for a time period;

receiving data corresponding to at least one factor that affects energy consumption for the time period, wherein the data corresponding to at least one factor that affects energy consumption includes at least one of production levels, working shifts, idle time, and repair periods;

analyzing a combined distribution of the data corresponding to energy consumption and the data corresponding to at least one factor that affects energy consumption;

identifying an outlier in the combined distribution;

generating a representation of the outlier in the combined distribution;

identifying clusters in the combined distribution;

building a model of previously established patterns of energy consumption based on the clusters;

receiving data corresponding to energy consumption for a second time period;

receiving data corresponding to at least one factor that affects energy consumption for the second time period;

determining that at least a portion of at least one of the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period deviates from the previously established patterns of energy consumption;

generating a representation of the at least a portion of the data corresponding to energy consumption for the second time period;

building a demand model of energy consumption per cluster;

forecasting energy consumption for a third time period using the demand model; and

generating a representation of the forecasted energy consumption for the third time period.

158. The non-transitory computer storage medium of claim 157 wherein the at least one factor that affects energy consumption for the time period includes at least one of ambient temperature, production levels, working shifts, idle time, weekends, holidays, repair periods, seasons, weather and heating degree days.

159. The non-transitory computer storage medium of claim 157 wherein the outlier is identified using at least one of regression analysis, density estimates, proximity metrics, and a K-nearest neighbor technique.

160. The non-transitory computer storage medium of claim 157 wherein the clusters are identified using density estimations.

161. The non-transitory computer storage medium of claim 157 wherein the data corresponding to energy consumption for the time period and the data corresponding to the at least one factor that affects energy consumption for the time period represent data relating to a first physical location and the data corresponding to energy consumption for the second time period and the data corresponding to the at least one factor that affects energy consumption for the second time period represent data relating to a second physical location.

162. The non-transitory computer storage medium of claim 157 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a production inefficiency.

163. The non-transitory computer storage medium of claim 157 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents a malfunction in equipment.

164. The non-transitory computer storage medium of claim 157 wherein the at least a portion of the data corresponding to energy consumption for the second time period that deviates from the previously established patterns of energy consumption represents outdated equipment.

165. The non-transitory computer storage medium of claim 157 wherein the steps are practiced sequentially.