Condition, Logic, and Calculation Nodes

Add Node

The following example illustrates how you can use an Add node to add a constant value to a value that is defined by a predecessor node. Consider the following nodes and connection.

In this example, the Add node adds 1 to the value in the Alert Level field of a Health Indicator record. The following image shows the Properties window for the Add node.

Remainder and Subtract Nodes

Suppose that you want to add reading values to a Health Indicator, but you know that the accuracy of the reading values is +/-10%. Before adding the reading values to a Health Indicator, use a policy to round down the reading values.

Consider the following nodes and connections. In the policy shown in the following image, the Remainder node is used in conjunction with a Subtract node to round down to the nearest ten the last reading value associated with a Measurement Location.

The following image shows the Properties window of a Remainder node. You can use the Remainder node to calculate the amount remaining after the Last Reading value is divided by 10. For example, if the last reading value is 87, the result of the remainder node is 7 (87/10=8 with a remainder of 7).

The following image shows the Properties window of a Subtract node. You can then subtract the remainder from the reading value. The result of this calculation is the reading value rounded down to the nearest ten. Continuing with the previous example, the result of this node is 80 (87-7=80).

Finally, the result of the Subtract node (80) is added to a Health Indicator Value record that is linked to a Health Indicator.

And Node

The following example illustrates how you can use the And node to monitor policy execution. To simplify this example, only constant values are used in the policy model. Consider the following nodes and connections, which are shown after validation has been run.

In this example, you can see that each node executed successfully. The logical result of the And node is yes because all incoming logic paths result in a value of true passed to the And node (that is, the logical result of each preceding condition node matches the logic path of the corresponding connection). Therefore, policy execution continues past the And node.

Or Node

The following example illustrates how you can use the Or node to monitor policy execution. To simplify this example, only constant values are used in the policy model. Consider the following nodes and connections, which are shown after validation has been run.

In this example, you can see that each node executed successfully. The logical result of the Or node is yes because the logical result of at least one incoming logic path results in a value of true passed to the Or node (that is, the result of the 5 > 3 Condition node is yes, which matches the logic path of the corresponding connection). Therefore, policy execution continues past the Or node.

The following example illustrates how you can use an Average node to evaluate the readings that are associated with a Health Indicator record and return the average reading value. Consider the following nodes and connection.

As shown in the following image, you can use the Properties window for the Average node to select the Health Indicator node and the Readings option. The Value column of the Readings collection is also selected, indicating that the output of the node is the average of the values in the Value column.

Change the Output Value of a Node from One Value to Another

You can use the Case node in the policy model to change a node's output values so that they can be used in successor nodes to perform calculations or actions. For example, you can use the Case node to convert character-based values from a Health Indicator reading or an AMS Asset status message to numeric values that you can use in subsequent calculations.

Consider the following policy model in which reading values from a character-based Health Indicator record are converted to numeric values. The values are then added to a Health Indicator record that uses numeric values.

As shown in the following image, you can use the Properties window of the Case node to indicate that the output value of the Case node will differ depending on the reading values associated with the Character Health Indicator node.

In this example, one output, HI Value, is defined. For this output, if the value in the Last Char Reading Value field is:

• 1-OK, the output value of the Case node will be 100. This is an If Case.
• 2-Not OK, the output value will be 50. This an another If Case.
• Any other value, the output value will be 0. This is the Else Case.

These values represent the numeric equivalents of the character-based readings associated with the Character Health Indicator node.

As shown in the following image of the Properties window of the Add Value to Health Indicator node, you can add the converted values to a numeric-based health indicator by selecting the HI Value output from the Case node.

Combine Policy Logic after a Condition Node in the Policy Model

You can use the Case node to combine policy logic in situations when a Condition node's successor nodes are the same in each logic branch. Rather than creating separate logic paths with duplicate nodes, you can use a Casenode to define different outputs for each logical result of the condition. This is especially useful in large policy models with multiple logic branches and duplicate nodes.

The following image shows what a policy that uses Condition nodes looks like without Case nodes in the model.

Notice that the Add Value to Health Indicator node appears three times, and the Email Contact node appears twice.

The following image shows what the same policy looks like when you use Case nodes in the model.

In this model, each Case node takes the place of a logic branch. Rather than splitting the model into separate logic paths, the Case node supplies a specified output corresponding to the logical result of the preceding Conditionnode. Notice that in this model the Add Value to Health Indicator node and the Email Contact node appear only once.

The following image shows what the Properties window looks like for the first Case node in the preceding image.

In this example, if the logical result of the Condition node named Greater Than is:

• Yes, the output of the node will be 50.
• No, the output will be 0.

When Output 1 is selected as the input of a successor node, the GE Digital APM system will supply the corresponding value to the successor node.

The following example illustrates how you can use the Collection Filter node to filter query results such that only certain results are used in subsequent calculations. Consider the following nodes and connection.

In this example, a Query node, Policy Events Query, is connected to a Collection Filter node.

As shown in the following image, you can use the Properties window of the Collection Filter node to define multiple filters that are applied to the full table of results from the query. Specifically, the query results are filtered to include only the policy events that lasted longer than ten seconds, occurred within the last month, and have the name IOW Excursion.

The output of the Collection Filter node (the filtered results of the query), can then be used by successor nodes to perform a variety of calculations.

The following example illustrates how you can use a comparison node to compare the value in a field of a predecessor node to a constant value. Consider the following nodes and connections.

The following image shows what the Properties window looks like for the Greater Than or Equal To node.

Based on the connection properties, the following logic will be applied when the policy is executed:

• If the DA Reading Value is greater than or equal to 900, an email message will be sent. This logic is determined by the connection between the Greater Than or Equal To Condition node and the Email Analyst node. Because a logic path is not defined, a value of Yes is assumed but does not appear on the model.
• If the DA Reading Value is not greater than or equal to 900, a Policy Recommendation record will be created. This logic is determined by the No logic path specified for the connection to the Create Recommendation node.

Example:

The following example illustrates how the Convert Type node can be used to convert the text string output of a Text node to a time span value.

In this example, the Text node is configured to create a text string that indicates a time span, using the values represented by the Point Value nodes. Because the Subtract node cannot perform any calculation on a text string, the Convert Type node converts the string to a value of the Time Span data type. The converted time span value is then used in the Subtract node to calculate the required date.

The following image illustrates the Properties window of the Convert Type node described in this example.

The following example illustrates how you can use a Count node in the policy model to determine if any Measurement Location readings were recorded within the last week. Consider the following nodes and connections.

In this example, a Collection Filter node is used to filter the readings associated with the Measurement Location node to only the readings that occurred within the last week. The Count node is then used to calculate the number of readings that occurred. The following image shows what the Properties window for the Count node looks like.

Assign a default value if a value is null

The following example illustrates how you can use the Is Null node in a policy model to define a default value to be used in subsequent nodes in the scenario where the input value is null.

Consider the following nodes and connections.

As shown in the following image, you can use the Is Null node to check whether or not the Criticality field in the record associated with the AMS Asset node has a value. If there is no value, the Is Null node output will be 0.

Then, as shown in the following image of the Properties window for the Case node, you can now use the output of the Is Null node to map the numeric Criticality values to corresponding character values (HIGH, MEDIUM, and LOW), which are used to populate the Recommendation Priority field when the Create Recommendation node is executed.

Using the Is Null node in this way allows you to translate a null value to an actual value that can be used in subsequent calculations and actions.

Trigger an action if a field is null

The following example illustrates how you can use the Is Null node in a policy model to trigger some action if a particular field is null.

Consider the following nodes and connections.

As shown in the following image, you can use the Is Null node to check whether or not the Criticality field in the record associated with the AMS Asset node has a value. If there is no value, the Is Null node output will be -1. The value -1 was chosen in this case because it is a value that is invalid for criticality; therefore, it serves as an indication that the Criticality field does not contain a value.

The Equal condition node in the policy evaluates whether the value from the Is Null node is equal to -1. If it is, this indicates that the Criticality field contains no value and an email alert is sent.

The following example illustrates how you can use a Last node to retrieve the most recent reading associated with a Measurement Location record and use this value in a subsequent calculation. Consider the following nodes and connections.

As shown in the following image, you can use the Properties window for the Last node to select the Measurement Location node and the Readings option. When this option is selected, the Last node evaluates the readings that are associated with the Measurement Location record and returns the last row in the collection, which contains the most recent reading.

You can use the most recent reading in various subsequent calculations. In this example, the Greater Than node evaluates the value of the most recent reading to determine whether or not it is greater than 1,000.

The following example illustrates how you can use the Math node to evaluate a Math expression.

Consider the following policy model that is designed to provide an email alert when an equipment has exhausted the threshold percentage of its life expectancy.

The threshold percentage of life used for the equipment is stored as a Technical Characteristic. The life data of the equipment (for example, hours for turbine, starts for crane, take-offs for aircraft, tonnes moved for mining equipment), is stored as Measurement Location readings.

The percentage of expected life of the equipment is calculated in the Math node using the Cumulative Distribution Function (CDF) of the Weibull Distribution, which is a mathematical function for analyzing life data.

The Math expression used for this calculation is as follows:

100*(1-e^(-((Time/Eta)^Beta)))

Where:

• e is a mathematical constant.
• Time is the life used.
• Eta and Beta are Weibull distribution parameters.

The following image shows the Properties window of the Math node:

To easily identify the Measurement Location, Technical Characteristic, and Weibull Distribution parameters associated with the equipment, the Equipment Entity input node is specified as the primary node, indicating that the record specified for this node is the primary record to which all the other input records (represented by other Input nodes) will be linked.

The Weibull Distribution Entity input node provides the distribution parameters (that is, Eta and Beta), which were determined in the GE Digital APM Reliability Analytics module by analyzing the failure data of a similar equipment.

The Life Used Measurement Location input node provides the readings to the Recent Reading Last calculation node, which retrieves the latest measurement reading (that is, Time).

The resulting value of the Math expression is passed to the Cumulative Probability Return Value action node, which will include the value in the policy execution history.

The Life Used Alert Level Entity input node provides the Technical Characteristic value to a comparison condition node, which compares it with the output of the Weibull CDF Math calculation node.

When the Weibull CDF value exceeds the Technical Characteristic value (that is, the threshold percentage) of the equipment, the message specified in the Email Text calculation node is sent via email to the responsible user specified in the Email Contact action node.

The following image shows the Execution Details window of the Math node:

The following example illustrates how you can use a Max node to determine which row of a query has the largest value in certain field and then use a successor node to evaluate a different field in the same row. Consider the following nodes and connections.

As shown in the following image, you can use the Properties window for the Longest Duration Max node to select the Query node and the Result Set option. The Duration (seconds) column of the Result Set is also selected, indicating that the query row with the longest duration will be returned.

As shown in the following image, you can then use the Properties window for the Equal node to determine if the Severity field in the query row with the longest duration contains the value Warning.

The following example illustrates how you can use the Round node in a policy model to round the value of an OPC Tag reading, represented by the Pressure node, to the nearest ten. Consider the following nodes and connection.

As shown in the following image, you can use the Properties window for the Round node to specify the type of rounding which you want to use. In this example, Precision rounding is selected as the type of rounding and 1 is specified as the number of digits from the input value that will be kept. This means that if, for example, the input to the node is 68, the output value will be 70.

The Round node in this example could be connected to various successor nodes in order to facilitate calculations. You could, for example, connect the Round node to a Case node, and then define a specific case to be executed based on the rounded input (that is, 10, 20, 30, and so on). Using the Round node in this way facilitates the execution of the Case node by limiting the number of possible inputs.

R Script Using Multiple Inputs from Policy

The following example illustrates how you can use an R Script node to evaluate certain values in other nodes and return a particular value. Consider the following nodes and connections.

As shown in the following image, you can use the Properties window for the R Script node to select a particular R script that was created using the R script editor. The R script in this example requires three input values: LinerBaseDF, LiningConditionAdjustment, and OnlineMonitoringAdjustment. These input values correspond to variables in the specified R script.

You can see that output values from the predecessor Input nodes have been specified to supply the values required by the R script. When this policy is executed, these values will be used to calculate the result of the R script. The Return Value node then returns the value calculated by the R Script node.

Example: Calculating Average and Standard Deviation of a Reading Collection

The following example illustrates how you can use a policy to execute an R script that calculates statistical information regarding readings related to a Measurement Location, and then send that information via email message to a designated recipient.

Consider the following R script and corresponding parameters.

This R script calculates the average reading value, count of readings, standard deviation, and most recent reading date of input values provided as a Data Frame.

The following policy is used to provide the input to the R script and to use the results of the R script in subsequent operations.

This policy first filters Measurement Location readings to only those readings recorded within the last 30 days. Then, the filtered readings are sent as a Data Frame input to the R script.

The following image shows the Properties window for the R Script node.

Next, the calculated results returned by the R script are used in a Text node to construct a message that will ultimately be sent via email to a responsible user.

The following image shows the Properties window for the Text node.

The following image shows the full text pattern specified in the Text node.

The following example illustrates how you can use a Sum node in a policy model to determine the total value of the Measurement Location readings that were taken within the last week. Consider the following nodes and connections.

In this example, the Collection Filter node is used to filter the readings associated with the Measurement Location node to only the readings that were recorded within the last week. The Sum node is then used to calculate the total value of these readings. The following image shows what the Properties window looks like for the Sum node.

.NET String.Format Expressions

The Text node supports .NET String.Format expressions, including the formatting capabilities supported by String.Format. The following table shows various example expressions and the resulting strings.

The following example illustrates how you can use the Text node in a policy to create a custom email message using various values from the policy's execution.

In this policy, the last reading related to a particular Measurement Location record is monitored. If the reading value exceeds the corresponding Upper Level 3 limit and no recommendations have been created since the reading was recorded, an email message is sent. The body of the email message includes a custom text message created using the Text node.

The following image shows the Properties window for the Text node.

You can see that various values from the policy execution are included within the text, thereby giving the recipient of the email further details and context surrounding the conditions that caused the message to be sent.

Assume that the policy returns the following values:

• Measurement Location - Upper Level 3 Numeric Value: 90.555
• Measurement Location - Checkpoint ID: chk-005
• Last Reading - Timestamp: 3/9/2016 12:27 PM

In this case, the output of the string node will contain the following text: Latest Reading for Measurement Location chk-005 taken on Wednesday, March 9, 2015 exceeds the Upper Level 3 limit of 90.55. No Recommendation has been created for this ML on or after Wednesday, March 9, 2015.

The following example illustrates how you can use a Threshold Statistics node with Condition nodes to evaluate how many times and for how long reading values exceeded a defined threshold. Consider the following nodes and connections.

In this model, the Temperature OPC Tag node is connected to the Above 600 F Threshold Statistics node, which is also connected to two Greater Than nodes.

As shown in the following image, you can use the Properties window for the Threshold Statistics node to:

• Indicate that you want to evaluate the collection of HDA Readings against some threshold.
• Identify the Timestamp and Value columns of the HDA Readings collection as the columns that store the timestamps and values that you want to evaluate.
• Indicate that the values in the Value column will be evaluated to determine whether or not they are greater than or equal to 600.

As shown in the following image, you can use the Properties window for the Count > 1Greater Than node to select the input Count from the Threshold Statistics node.

As shown in the following image, you can use the Properties window for the Accumulated Time > 3 Days Greater Than node to select the input Accumulated Timefrom the Threshold Statistics node.

When these options are selected, the condition nodes evaluate the number of times and amount of time, respectively, over which the reading values recorded in the process historian exceeded the threshold of 600 degrees Fahrenheit.