Patent Application
20120262472
The present disclosure pertains to presentation of information, and particularly to presentation of data. More particularly, the disclosure pertains to visualization of data.
The disclosure reveals an approach for visualization of time series data. The approach for conveying time-series data, whether ordinal, nominal, interval, or ratio, may be a “heatmap timeline”. Rather than use a spatial dimension indicate a datum value for each timestamp, the heatmap timeline may employ hue, saturation, or value of color, and/or pattern and/or shading, perhaps shown within a geometric shape, to indicate the datum value along a timeline. Specific values may be aggregated into one value indication for a certain portion of the timeline period. However, a tooltip may be pointed to a specific place on of the heatmap timeline to obtain a precise datum value at that place. More than one heatmap may be presented relative to one timeline in a display. Traditional line plots synchronized to the timeline may also be presented on the same display for comparison purposes. The heatmap timelines of various items of a hierarchical structure may be presented on a display. Items of the hierarchical structure may have markers allowing the items to be expanded to show heatmaps of components of the items. Information may be presented in a mosaic fashion with, for example, rows of blocks along a timeline. Each row may represent an item. Each block may have one of various colors indicating, for instance, a status of an item of the row with the respective block.
a and 7b are diagrams of sample views of a heating, ventilation and air conditioning equipment hierarchy and a building geometry hierarchy, respectively;
Commercial buildings, industrial plants, and other facilities are increasingly equipped with rich networks of sensors, allowing for collection and processing of large amounts of data. These data provide the opportunity for expert remote analysis through effective visualization techniques. For time-series data, analysts typically use line graphs, where data values are plotted over time. However, such plots are not necessarily effective for data of nominal measure type, and it may often be difficult for the analyst to visually aggregate ordinal data.
A technique for conveying time-series data, whether ordinal, nominal, interval, or ratio, may be a “heatmap timeline,” which is the subject of the present disclosure. Whereas line charts for time series data may use a vertical position to indicate the data value for each timestamp. The heatmap timeline may employ color hue, saturation, or value in order to indicate the data value.
Further precision may be provided in interactive computer environments through the use of “tooltips,” small pop-up windows that display the precise data value when a user points to an area of the heatmap timeline with an input device. The expected work flow may be that the analyst will first visually analyze the plot to identify areas of interest and then use the tooltip to discover the precise values.
Another aspect of the present timeline may be a coordination of the heatmap timeline and traditional line plots by timestamp. Multiple plots may be vertically stacked, or one or more line plots may be superimposed over a heatmap timeline plot. In this fashion, many variables may be displayed simultaneously and compared with one another.
These timelines may be viewed in static information graphics or alternatively integrated into interactive visualization environments.
To create an individual heatmap timeline plot, a data series may be required in which each data point is defined by a timestamp and value. The data value may be mapped to a color hue, saturation, shade, or value.
The heatmap timeline may have a fixed vertical height. Each data value may be represented by creating a rectangle where the height is equal to a fixed vertical height and the width is equal to the width of the overall chart, multiplied by the fractional part of the displayed timescale that the data value represents. In a case where a heatmap timeline is displayed on a computer screen, it may be common that the number of data values for display may exceed the number of screen pixels available to render the image. In this case, an aggregation strategy may be chosen to combine adjacent data values into a rectangle with a width of one or more pixels. Aggregation strategies may include, but are not limited to, the following items. For ordinal data, there may be a maximum, a minimum, a range, a sum, a weighted sum, a median, and an average. For nominal data, a function may be defined to choose a value judged most important or relevant.
Once both the rectangle width and color have been calculated, the rectangle may be rendered with its horizontal position located according to the associated timestamp's relative position along the entire displayed timeframe. This approach may be repeated for each data value or aggregation of data values to display virtually the entire selected timeframe.
As comparisons among different data views might often be essential to arriving at new insights, the present approach may provide for a coordinated display of heatmap timelines and trend data. A vertical line drawn through any stacked plots may intersect data values coordinated in time.
Strategies may be used to convey the value of the data to the analyst. One strategy may be to communicate the meaning of the hue, saturation, or value of color; a legend may be created where the distinct colors or color gradients are displayed, with labels indicating the data values associated with the distinct color, or the values at the extremities of the color gradients. Colors in a non-color layout may be represented by shade, symbols, patterns, and/or other grayscale or black and white techniques. A variety of symbols may be used, including but not limited to circles, squares, triangles, diamonds, stars, and so forth. Another strategy may be to allow for precise communication of the data value, and to allow for comparison of values among multiple data series; here, a tooltip may be used. The tooltip may be a small window that pops up when the user uses an input device to place a pointer on the heatmap timeline. Similar windows may appear on each heatmap timeline and line plot, so that the user can compare the precise values of many data types synchronized in time.
When data aggregation strategies have been used such that each rectangle in the heatmap timeline plot represents more than one data value, the tooltips may also be used to express the full timeframe represented by the rectangle, the aggregation strategy used, and/or the values of virtually all underlying data points aggregated into the chosen rectangle.
Often, analyzing data over differing time periods may reveal different insights. For example, data that have a daily cycle may be best viewed one week at a time to identify patterns, whereas data cycling hourly may be best be viewed one day at a time. So that an analyst may select an appropriate view of the data, the heatmap timeline is created based on a selected first and last timestamp. Even if the full data series extends beyond the selected beginning and ending timestamp, the heatmap timeline may be created based on the selected start and end time. When a new time interval is selected, aggregated performance numbers may be recomputed for the selected timeframe. In areas where data is not present or valid, the data may be conveyed using an additional color.
A heatmap timeline for time series data may be displayed as a graph 20 in
Coordinated tooltips such as tooltip 17 may be used to display more precise data values (e.g., 71.96) than otherwise indicated by the heatmap at a particular location such as in, for example, row 11. Information in row 12 for instance may indicate a condition such as a match as indicated by a tooltip 21. Data may be of a nominal type with each value mapped to a different color as indicated by a tooltip 22 which may reveal cooling and ventilation in row 13. Row 14 shows traditional line plots 43, 44 and 45 which may be synchronized with heatmap timeline plots in rows 11-13. Tooltips 23, 24 and 25 may identify the plots 43, 44 and 45 in row 14 as “Ucc: 1”, “Uc: 0.095” and Uhc: 0″, respectively. Plots 43, 44 and 45 may have their lines in color such as, for example, blue, red and purple, respectively. A tooltip 26 may indicate a specific time at a particular location in one or more of the rows 11-14 along an X-axis of row 15. The X-axis may be used to display a synchronized timescale for heatmap timeline plots and line plots.
A color of each field may correspond to a specific state. A state may indicate an operating mode, a fault, and so forth. For an illustrative example, yellow may indicate that the equipment is off. Green may indicate that the equipment is running. Blue may indicate that the equipment is in a specific mode. Red may indicate a faulty state of the equipment, such as an AHU-2 in row 32 at times 14:00 and 15:00, and thus require investigation of the state of the AHU-2. In plot 28, yellow may be indicated by “Y”, green may be indicated by “G”, blue may be indicated by “B”, and red may be indicated by “R”. A legend in
Comm. failure (H) may appear to have a problem with its row being mostly red over nearly the entire time represented by X-axis 122. The other components appear have corresponding rows of green over virtually all of the time represented by X-axis 122. An exception may be indication by a short term of red as indicated by symbol 123 relative to a stuck heating valve and a leaking heating valve.
In the graph 50, there may be, for example, 24 rows of vertical lines or stripes of various colors indicating conditions at certain times as noted with an X-axis 77. The X-axis may have labels indicating times, for example, from “Jul21 11:15” to “Apr19 06:00”, with in-between times listed on the axis. The time increments may be determined in accordance with needs or desires of the user or users of the graph. The rows may provide conditions of various kinds of items, as listed along a Y-axis 78 at a left portion of graph 50.
A top row 51 is, for instance, labeled “AHU Aggregated Fault Status”, which may indicate a top level maximum aggregated fault value. Row 53 may indicate a detected mode and row 54 may indicated an expected mode. Row 52 may indicate an AHU mode comparison. For the same time slot, if the detected and expect modes, in row 53 and 54 respectively, are significantly different in condition, according their color lines or stripes, then a line or stripe for that time slot for the AHU mode comparison in row 52 may indicate a poor or faulty condition, despite whether the colors in the stripes or lines of both rows represent a faulty or good condition, or a condition in between the faulty and good conditions. For the same time slot, if the detected and expected modes, in rows 53 and 54 respectively, are significantly the same in condition, according their color lines or stripes, then a line or stripe for that time slot for the AHU mode comparison in row 52 may indicate a good condition or good comparison, despite whether the colors in the stripes or lines of both rows represent a faulty or good condition, or a condition between the faulty and good conditions. The AHU mode comparison in row 52 may be regarded as a control inefficiency monitor.
Rows 55, 61, 67 and 72 may represent illustrative examples which include a high relevance fault such as a stuck heating valve, a stuck cooling valve, a leaking heating valve and a comm. failure (cooling). Relative to the stuck heating valve of row 55, the contributing symptoms may be +SH03, +SH08, +SC04, +SC08 and +SC10 of rows 56, 57, 58, 59 and 60, respectively. Relative to the stuck cooling valve of row 61, the contributing symptoms may be +SH08, +SH10, +SC03, +SC04 and +SC08 of rows 62, 63, 64, 65 and 66, respectively. Relative to the leaking heating valve of row 67, the contributing symptoms may be +SH03, +SH04 and +SC10 of rows 68, 69 and 70, respectively. A canceling symptom relative to the leaking heating valve may be −SH01 of row 71. Relative to the comm. failure (cooling) of row 72, a contributing symptom may be +SH01 of row 73. A canceling symptom relative to the comm. failure (cooling) may be −SH09 of row 74.
Rows 11-14 may be referred to heatmap timelines in
An index or navigation tree 85 at the left side of screen print 81 shows “AthleticAHU06(74.3%), “Control Inefficiencies”, Detected Modes” and “Trend Data”, being selected with checkmarks, may be revealed in graph 82. Also selected with checkmarks, there may be items “CFBAHU3(82.71%)”, “CFBAHU5(64.43%)” and “MonahanAHU1(63.75%)”. Row 16 may show values for CFBAHU3 at the bottom of screen print 81. One may with a bar 86 scroll down to rows of values for CFBAHU5 and MonahanAHU1. Other items which might be selected in tree 85 may incorporate heating coil faults, cooling coil faults, common faults and data cleaning as examples.
Building geometry 92 of building 1 may incorporate a basement 96, a first floor 97 and a second floor 98. The geometry may incorporate more or less floors, or other types of areas defined in different terms. The VAV-01 may serve, for example, zones B-01, 1-01 and 2-01 of the basement 96, first floor 97 and second floor 98, respectively. The VAV-02 may serve, for example, zones B-02, 1-02 and 2-03 of the basement 96, first floor 97 and second floor 98, respectively. The VAV-03 may serve, for example, zones B-03, 1-03 and 2-03 of the basement 96, first floor 97 and second floor 98, respectively. There may be other hierarchy structure configurations for HVAC equipment and building geometry that may be implemented. A configuration may be designed in response to desired extensions for building optimization analytics, viewing and analysis of conditions, and results of HVAC equipment 91 and building geometry 92. There may be links between the VAVs and served zones that can tie faults and other diagnostics from equipment to zones. There may be a drill-down from an overall perspective to different analysis levels.
a and 7b are diagrams of sample hierarchical views 101 and 102, respectively. A view may be indexed by an HVAC equipment 91 hierarchy or a building geometry 92 hierarchy, as represented by the sample hierarchical views 101 and 102, respectively. Both views may indicate an aggregated fault status. The rows 103-106 and 108-116 may be in the same format as the rows of information with similar coding and time scale as view 50 of
In views 101 and 102, there may be an interactive drill-down by expanding [+] markers in the index which may be linked to an index or navigation tree. For instance in graph 101, [−]Bldg-01 may be [+]Bldg-01 at row 103 without rows 104, 105 and 106. If [+]Bldg-01 is clicked on (i.e., the [+] marker being expanded), then rows 104, 105 and 106 may appear representing information (e.g., aggregated fault status) about AHUs in Building 01. Particularly, rows 104, 105 and 106 may represent information about AHU-01, AHU-02 and AHU-03, respectively. The ledgers in the hierarchical view or graph may represent the AHUs with the notation [+]AHU-01, [+]AHU-02 and [+]AHU-03, respectively, at rows 104, 105 and 106. Interactive drill-down may be achieved by clicking on or expanding a [+] marker of one of the AHUs. For example, if [+]AHU-01 were instead [+]AHU06 and clicked on, one may get a drill down from this AHU in view 101 under a [−]AHU06 which would look like chart, graph or view 50 of AHU aggregated fault status at row 51 along with rows 51-74, as shown in
b is a diagram of a hierarchical view 102 of the building 1 indexed by geometry 92. Row 108 may indicate an aggregated fault status of building 1. Row 108 may be labeled as [−]Bldg-01 which has a drill down from an expansion of a [+] marker in a previous label [+]Bldg-01 where rows 109-116 were absent. A drill down from [−]Bldg-01 may result in rows 109, 110 and 111 for [+]Basement, [+]1st Floor and [+]2nd Floor. A drill down from [−]2nd Floor may result in rows 112, 115 and 116 for zones [+]2-01, [+]2-02 and [+]2-03, respectively. A drill down from, for example, [−]2-01 may result in a [+]Fault 1 with a row 113. A click on [+]Fault 1 may result in drill down from [−]Fault 1 at row 113 having a symptom 1 at row 114.
Drill down may be taken to be an arbitrary depth (e.g., based on a number of hierarchical levels in the actual system). However, something that is hierarchical may not necessarily need to be explained as multiple layers of the hierarchy.
In
In chart 133 of
Similarly as in
To recap, an approach 150, in a diagram of
Approach 150 may further incorporate providing a line plot that represents values of the information corresponding to the one or more designations. The line plot may be synchronized with the values of the information in accordance with the timeline. The line plot may be superimposed over a respective row of information. There may be an axis for determining values from the line plot. The approach may also incorporate obtaining a precise value of the information at a particular place relative to the timeline. The precise value of the information may be provided by a tooltip having a pop-up window that displays the precise value of the information.
A system 160, in a diagram of
The symbols may be arranged in a row proximate to a designation of the listing of one or more designations having the data values represented by the symbols. The symbols may incorporate rectangles, as examples, situated proximate to one another in a row. Each rectangle may have a start and end time with a particular data value represented by the hue, saturation and/or value of color, or pattern and/or shading within a respective rectangle corresponding to the particular data value. The start and end times may be indicated by ends of a rectangle relative to the timeline.
System 160 may also have one or more line plots of data values of the one or more designations having the data values. The line plots may be time synchronized with the start and end times of the rectangles.
In system 160, a symbol may represent an aggregation of a plurality of data values. Aggregation strategies may, for instance, include, but not be limited to, items noted herein. For ordinal data, there may be a maximum, a minimum, a range, a sum, a weighted sum, a median, and an average. For nominal data, a function may be defined to choose a value judged most important or relevant.
System 160 may have one or more tooltips which may be placed on a symbol to obtain a particular data value at a certain time on the timeline.
The data values of one or more designations may indicate an aggregated status for various times along the timeline. Some of the one or more designations have expansion markers which may be activated for a drill-down of a hierarchy of each of the some of the one or more designations. The drill down may result in one or more components of the hierarchy. The one or more components may have data values represented by an addition of one or more heatmap timelines. The drill down may be taken to be to an arbitrary depth (e.g., based on a number of hierarchical levels in the actual system). It may be noted that something that is hierarchical may not necessarily have to be explained as multiple levels or layers of the hierarchy.
In system 160, format 164 may incorporate a mosaic. The mosaic may have a timeline and one or more rows of symbols or items, such as blocks used merely as illustrative examples of the symbols or items, parallel to the timeline. Each of the one or more rows of the blocks may be associated with one or more pieces of equipment, or one or more components of equipment. Each block of the one or more rows of blocks may also be associated with a time increment. Each block may represent a data value with a hue, saturation and/or value of a color, or pattern and/or shading. The data value may be a status or property of one or more pieces of equipment or one or more components of the one or more pieces of equipment.
An approach 170, in a diagram of
In the present specification, some of the matter may be of a hypothetical or prophetic nature although stated in another manner or tense.
Although the present system and/or approach has been described with respect to at least one illustrative example, many variations and modifications will become apparent to those skilled in the art upon reading the specification. It is therefore the intention that the appended claims be interpreted as broadly as possible in view of the prior art to include all such variations and modifications.
