Python script for live playback of simulated factory event data through OPC UA, operating as either a server or a client.
Requires opcua-asyncio (https://github.com/FreeOpcUa/opcua-asyncio)
pip install asyncua
python run_sim.py -p basic
Further command line option information:
usage: python run_sim.py [-h] [-p PRESET] [-t] [-c] [-e ENDPOINT] [-n NAME]
[-f FEED_FILE] [-s TIMESTAMP] [-b TIMESTAMP] [-x SPEED]
Provides live OPC UA data from a CSV feed of simulated factory events
options:
-h, --help show this help message and exit
-p PRESET Name of base configuration preset to use from presets.cfg
override preset settings:
-t, --tmc_model Use the TMC-based model instead of the basic model
-c, --client Write the data to a separate server as a Python client
-e ENDPOINT Endpoint to host / Endpoint to target as client
-n NAME Name to use for Python server (if hosting)
-f FEED_FILE Filename of CSV with simulation data
-s TIMESTAMP Skip to given time in feed file before writing anything
-b TIMESTAMP Begin live playback by fast-forwarding to given time
-x SPEED Playback speed multiplier
Input consists of a list of factory events produced from a simulation of an electric motor assembly line with 10 stations, some working in parallel, and one working on batches of parts:
The average process times and changeover times depend on the station and the part type, and the amount of each part type depends on a weekly production schedule:
The data is formatted as a CSV included in /data with 1 year of events.
Timestamps are represented as minutes relative to the beginning of the simulation.
Additional properties of the work item involved in the event are included as well.
Data sample:
| Object | Timestamp | Event_Name | Event_Value | Serial_Num | Type_ID | OvenBatch | Reworked | ShiftID |
|---|---|---|---|---|---|---|---|---|
| ... | ||||||||
| Cleaning | 380.91313 | Work_In | 1 | 3 | 1 | 1 | ||
| Cleaning | 380.91313 | State | 1 | 3 | 1 | 1 | ||
| Insulating | 386.39065 | State | 2 | 1 | 1 | 1 | ||
| Insulating | 386.39065 | Work_Out | 1 | 1 | 1 | 1 |
Description of event names and values:
| Event_Name | Event_Value description |
|---|---|
| Work_In | Number of parts contained in the factory station |
| Work_Out | Number of parts remaining in the factory station |
| State_Change | New state of event |
State values (later converted to PackML enum in OPC UA):
| Event_Value | Decription |
|---|---|
| 0 | Starved |
| 1 | Working |
| 2 | Blocked downstream |
| 3 | Changeover between part types |
| 5 | Off shift |
| 6 | Lack of staff |
| 7 | Scheduled maintenance |
The script will provide the data through OPC UA representations of each station in the assembly line:
Can be added/edited in presets.cfg, determines the base settings
for how to write the data and where to write it to.
[basic]
server_name = Python Sim
endpoint = opc.tcp://0.0.0.0:4840/freeopcua/server/
[basic_client]
write_as_client = True
endpoint = opc.tcp://localhost:53530/OPCUA/SimulationServer
[tmc_client]
use_tmc = True
write_as_client = True
endpoint = opc.tcp://localhost:53530/OPCUA/SimulationServer
The simulation can write to one of two information models:
- One leveraging the TMC companion spec (https://github.com/OPCFoundation/UA-Nodeset/tree/latest/TMC)
- A more basic nodeset mirroring the TMC structure using base OPC data types
If using a separate OPC UA server, import the instance nodeset for the desired
information model, which will add a folder of AssemblyLine stations to
Objects corresponding to the data feed:
- Basic model types + instances -
nodesets/SimBasic.xml - TMC instances -
nodesets/SimInstancesTMC.xml
If using the TMC-based model, the required companion nodesets are also included:
- DI 1.03.0 -
nodesets/DI.xml - PackML 1.01 -
nodesets/PackML.xml - TMC v2 2.00.1 -
nodesets/TMC.xml