Announcement

Collapse
No announcement yet.

How to create Abnormal Condition Summary

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    How to create Abnormal Condition Summary

    I am wondering if there is an existing method for the end user to create a summary of devices that are in an abnormal state? In industrial control systems such as SCADA, you typically find this functionality built-in. The user must first edit the data points(devices/features) to flag their available status or values to be abnormal. For instance, a status of ON may be Normal while OFF is considered the Abnormal condition. The same flags would be assigned to values or ranges of values. This flag is maintained in the point(device) database. The operator can call up the Abnormal Summary listing to quickly see what features if any are in the Abnormal state, a very useful tool to assess the overall system state.

    If such a thing is not available it would be great to add to the HS built-in feature road map. The device database would need to have a field added for the abnormal flag in the feature status/graphics section. A new live display page would need to be available to show all the features in abnormal state.

    Thoughts anyone?

    #2
    While certain aspects of industrial control systems are very analogous to home automation, like state machine design/logic, IMO what you are proposing is not. There are very few conditions in a home which should be considered "abnormal", the different states of various devices such as lighting controls, motion sensors, etc is all part of daily life and normal operating conditions. It's not like you are operating a piece of machinery near its safe operating conditions in a factory.

    That being said, you can easily do what you want by creating virtual devices for the "real" devices you want to monitor. Each virtual device would be set to "abnormal" based on whatever criteria you setup in an event to monitor the real device. You can then setup a view for just these virtual device monitors and bring them up in a web view.

    Comment


      #3
      Originally posted by TC1 View Post
      While certain aspects of industrial control systems are very analogous to home automation, like state machine design/logic, IMO what you are proposing is not. There are very few conditions in a home which should be considered "abnormal", the different states of various devices such as lighting controls, motion sensors, etc is all part of daily life and normal operating conditions. It's not like you are operating a piece of machinery near its safe operating conditions in a factory.

      That being said, you can easily do what you want by creating virtual devices for the "real" devices you want to monitor. Each virtual device would be set to "abnormal" based on whatever criteria you setup in an event to monitor the real device. You can then setup a view for just these virtual device monitors and bring them up in a web view.
      While I agree not every device in HS is critical typically and would not warrant this type of treatment, there are ones that would be nice to handle with this. I'm thinking battery levels, doors left open, ups on battery, internet trouble, etc.

      Setting up virtual devices and events for each of these would be tedious to maintain. I always favor the native functions if it makes more sense than a workaround. I'm not even sure how to create a special view for these as you mentioned, but I am pretty green on HS, maybe you can explain?

      Comment


        #4
        Originally posted by davidcald View Post

        While I agree not every device in HS is critical typically and would not warrant this type of treatment, there are ones that would be nice to handle with this. I'm thinking battery levels, doors left open, ups on battery, internet trouble, etc.

        .... I am pretty green on HS, maybe you can explain?
        Agree with all your points regarding battery levels, etc. And your point about being new to HS is noted and mostly likely a cause of many of your questions/suggestions

        HS is designed to be extensible with the plugin architecture, plugins are just as native to HS as any other so-called built-in functionality. The owners designed it this way to be communication agnostic and allow the user community to introduce features at a faster pace than the company.

        Many of the use-cases you are looking for such as battery device levels, UPS monitoring, etc are already covered by many plugins, some of them free. For battery level monitoring many of us use https://shop.homeseer.com/collection...lug-in-for-hs3

        Below is that "at a glance" type of dashboard you were looking for and I have this report emailed to me whenever a battery level goes below whatever threshold I define for that device-

        Click image for larger version

Name:	Capture.PNG
Views:	132
Size:	65.8 KB
ID:	1517206

        Comment


          #5
          Originally posted by TC1 View Post

          Agree with all your points regarding battery levels, etc. And your point about being new to HS is noted and mostly likely a cause of many of your questions/suggestions

          HS is designed to be extensible with the plugin architecture, plugins are just as native to HS as any other so-called built-in functionality. The owners designed it this way to be communication agnostic and allow the user community to introduce features at a faster pace than the company.

          Many of the use-cases you are looking for such as battery device levels, UPS monitoring, etc are already covered by many plugins, some of them free. For battery level monitoring many of us use https://shop.homeseer.com/collection...lug-in-for-hs3

          Below is that "at a glance" type of dashboard you were looking for and I have this report emailed to me whenever a battery level goes below whatever threshold I define for that device-
          I would be happy with a purpose built plugin to accomplish what I'm looking for. I have been using several good plugins including one for APC, and the SDJ-Health one you included for batteries, I get the same report . My problem is while these are excellent plugins, they are still all separate and serve their specific purposes. Nowhere can I find something that recognizes a feature in abnormal state, whatever that is that I designate, over the entire database. I don't want to go through the whole devices page scanning for these abnormal items via multiple plugins with their unique UI. SDJ-Health is probably the closest to what I would like but would have to be extended. Its General monitoring, which I use, seems to almost achieve this function indirectly by watching log entries and inactivity timeouts. But it is mostly concerning the condition of the "interface" not specific features. So it doesn't cover the case where a feature value of 0-5 is OK/Normal but >5 is Abnormal.

          Comment


            #6
            I understand what you are driving at, but I can't justify (at least inside my head, lol) this being an universally needed feature when what you are asking can be accomplished through various methods (events, virtual devices, plugins, scripts, etc).

            Maybe others have a different opinion than mine and would welcome to hear their thoughts.

            If you feel you really want this feature you might want to open a feature request on the HS support page.

            Comment


              #7
              Originally posted by davidcald View Post

              I would be happy with a purpose built plugin to accomplish what I'm looking for. I have been using several good plugins including one for APC, and the SDJ-Health one you included for batteries, I get the same report . My problem is while these are excellent plugins, they are still all separate and serve their specific purposes. Nowhere can I find something that recognizes a feature in abnormal state, whatever that is that I designate, over the entire database. I don't want to go through the whole devices page scanning for these abnormal items via multiple plugins with their unique UI. SDJ-Health is probably the closest to what I would like but would have to be extended. Its General monitoring, which I use, seems to almost achieve this function indirectly by watching log entries and inactivity timeouts. But it is mostly concerning the condition of the "interface" not specific features. So it doesn't cover the case where a feature value of 0-5 is OK/Normal but >5 is Abnormal.
              Not the complete solution you're looking for, but EasyTrigger could make event-driven monitoring a lot simpler than one event per device suggested earlier in this thread. Create ET groups of similar devices/features and use an event to monitor value. Using your example above, ET can readily monitor a group and fire actions if any in the group have their value set >5.

              I do this for a number of device types such as refrigerator/freezer temps and door locks. The events fire alert emails with current status/values, links to trend data charts, etc.
              -Wade

              Comment


                #8
                While not exactly the same as the OP question I have a very similar situation. I have a few fringe battery Z-Wave devices that can receive their commands but are sometimes unable to send a status back to HomeSeer. This puts the device in an Unknown condition with a value of 0. The value of zero is already part of a Status/Graphics pair so I can't trigger on that as being an error condition. Like the OP I would like to query the entire database to see if any devices have a status of "Unknown".

                I have excluded / re-included in place to no avail. Have tried different sensors as well.
                The Devices have been Optimized / Full Optimized without change.
                If they are in an Unknown condition, usually a simple Polling of the Device is able to get its status updated.

                Anyway to scan for this? The same answer would probably work for the OP.

                If EasyTrigger could do a "If any Device in Group has a Status of" it would be simple then.

                Roger D

                Comment


                  #9
                  Originally posted by Roger D View Post
                  While not exactly the same as the OP question I have a very similar situation. I have a few fringe battery Z-Wave devices that can receive their commands but are sometimes unable to send a status back to HomeSeer. This puts the device in an Unknown condition with a value of 0. The value of zero is already part of a Status/Graphics pair so I can't trigger on that as being an error condition. Like the OP I would like to query the entire database to see if any devices have a status of "Unknown".

                  I have excluded / re-included in place to no avail. Have tried different sensors as well.
                  The Devices have been Optimized / Full Optimized without change.
                  If they are in an Unknown condition, usually a simple Polling of the Device is able to get its status updated.

                  Anyway to scan for this? The same answer would probably work for the OP.

                  If EasyTrigger could do a "If any Device in Group has a Status of" it would be simple then.

                  Roger D
                  Hi Roger,

                  The General Devices section of SDJ-Health can scan for Invalid State. Not the 'abnormal' state that the op was asking about but it might fit your requirement.

                  Click image for larger version

Name:	Snip-1.JPG
Views:	98
Size:	75.2 KB
ID:	1518802

                  Click image for larger version

Name:	Snip-2.JPG
Views:	93
Size:	39.4 KB
ID:	1518803
                  Steve

                  Comment


                    #10
                    Originally posted by Roger D View Post
                    While not exactly the same as the OP question I have a very similar situation. I have a few fringe battery Z-Wave devices that can receive their commands but are sometimes unable to send a status back to HomeSeer. This puts the device in an Unknown condition with a value of 0. The value of zero is already part of a Status/Graphics pair so I can't trigger on that as being an error condition. Like the OP I would like to query the entire database to see if any devices have a status of "Unknown".

                    I have excluded / re-included in place to no avail. Have tried different sensors as well.
                    The Devices have been Optimized / Full Optimized without change.
                    If they are in an Unknown condition, usually a simple Polling of the Device is able to get its status updated.

                    Anyway to scan for this? The same answer would probably work for the OP.

                    If EasyTrigger could do a "If any Device in Group has a Status of" it would be simple then.

                    Roger D
                    Do you use my Device Viewer and Documenter package? That has an option to search for invalid devices.

                    Click image for larger version

Name:	Capture.PNG
Views:	116
Size:	107.5 KB
ID:	1518805
                    Jon

                    Comment

                    Working...
                    X