Industrial AI, without the licensing barrier. TDengine is free forever, up to 5,000 tags. Data ingestion, real-time analytics, visualization, AI-assisted analysis, natural-language exploration, root cause insights. One platform, no stitching things together. Take it for a spin. No sales call required: tdengine.com/free-tier/?utm_… #Industry40 #TDengine #FreeForever

Industrial knowledge is often scattered across maintenance records, manuals, SOPs, and standards. TDengine Knowledge Graph helps extract the knowledge, connect the relationships, and make that experience reusable. Try it free: tdengine.com/downloads/?utm_… #TDengine #KnowledgeGraph

One maintenance record can solve more than one maintenance problem. TDengine Knowledge Graph connects what was learned before such as causes, procedures, limits, spare parts, and more, so that experience can be reused when similar problems appear again. #TDengine #KnowledgeGraph #IndustrialAI

That’s a wrap on #ICCUnleashed! Three days of great conversations with teams looking for a modern alternative to the traditional data historian, with analytics and Industrial AI built for what comes next. Thanks to everyone who stopped by, and to @InductiveAuto for bringing the community together! #Ignition #TDengine

A battery can keep cycling even while usable capacity is already falling. In this case, cell level warning signs appeared about 3 weeks before traditional capacity calibration exposed the loss. See the full analysis: tdengine.com/battery-energy-… #EnergyStorage #BESS #TDengine

Day 2 at #ICCUnleashed! Great event, great conversations, and plenty of real industrial use cases being discussed. Huge thanks to the @InductiveAuto team for bringing the community together. One more day to go. Come see TDengine at Booth 4️⃣8️⃣! #Ignition #TDengine #OperationalTechnology

ICC 2026, Day 2. Started the morning with a 10K run in Sacramento. A good run always helps me clear my mind, recharge, and get ready for the day ahead. Now fully charged and ready for another busy day at ICC 2026 — meeting more people, sharing what we’re building at TDengine, and learning from the industrial community. See you at the TDengine booth!

An alarm tells you something changed. The real work is connecting the evidence before deciding what to do next. Follow the investigation from pressure drop to leak evidence to the operator’s shutoff decision. tdengine.com/urban-gas-safet… #TDengine #GasSafety

Day One at ICC 2026 by Inductive Automation — what a day! My team and I barely had a moment to stop. We spent the entire day talking with visitors, introducing TDengine, and doing live demos. It was exciting to see the TDengine booth packed with people throughout the day. What impressed many visitors most was how AI is already becoming part of the industrial data workflow. Our AI-powered Root Cause Analysis, Panel Insights, and Chat BI generated a lot of excitement. Instead of just storing and visualizing industrial data, TDengine is helping users understand what is happening, why it is happening, and what they should pay attention to. A big thank you to Jeff Winter, Michael Finocchiaro, Rafey Shahid, Rudford Hamon, Craig Resnick, Mike Bowers and many others who stopped by our booth today. We had many great conversations and received a lot of valuable feedback and suggestions. These discussions with real industrial users are incredibly important to us as we continue building TDengine as an AI-native industrial data platform. Looking forward to another exciting day at ICC 2026. If you are here, stop by the TDengine booth and say hello! #iccunleashed

Day 1️⃣ at #ICCUnleashed is a wrap! A busy day at Booth 48 with plenty of demos, questions, and great conversations about industrial data and AI. Thanks to everyone who stopped by. We’ll be back tomorrow. Come say hi and discuss the latest developments in industrial data and AI! #TDengine #IndustrialAI #Ignition

Just arrived in Sacramento for ICC 2026 hosted by Inductive Automation! After the drive, our first mission was getting the TDengine booth ready. Everything is now set — demos are ready, the team is ready, and of course, we’ve prepared plenty of TDengine swag. 😄 If you’re attending ICC 2026, stop by the TDengine booth and say hello. I’d love to meet you in person, hear about the industrial data challenges you’re working on, and show you what we’ve been building at TDengine. See you at ICC!

What makes AI actually useful in industrial operations? It needs context to understand your plant, experience to know how it works, and the capabilities to act. Industrial Ontology → Knowledge Management → Agent Runtime Context → Experience → Action #TDengine #IndustrialAI

Sand content rises in a pump. 14 hours later, winding temperature crosses the warning line and the unit shuts down. See how this use case traces the fault through the gathering system. tdengine.com/oilfield-produc… #TDengineUseCases #OilfieldOperations

Flow falls at the metering station. Which well is responsible? A 60% drop in polished-rod load may be sensor drift or a real break. See how this oilfield use case uses AI to trace the anomaly. tdengine.com/oilfield-produc… #TDengine #OilAndGas

Power-grid alarms sit on separate screens. See how engineers connect equipment data and trace downstream effects: tdengine.com/urban-distribut… #TDengineUseCases #UtilityManagement

Transformer oil temperature rises to 88°C as imbalance climbs from 5% to 21%. Summer heat or fault? See how engineers connect the signals: tdengine.com/urban-distribut… #TDengine #GridOperations

I recently had a fascinating conversation about AI with science writer Wan Weigang. We both studied physics, which led us to an interesting question. AI has been astonishing in areas like programming and mathematics. But when we look at physics — or industrial data, the field I work in — the progress feels much less dramatic. Why? One explanation we discussed is the ability to close the loop. In programming, AI can write code, execute it, see the result, identify errors, and improve. In mathematics, it can generate a solution and increasingly verify whether the reasoning or answer is correct. The feedback loop can largely happen within the digital world. Physics is different. Industrial operations are even more so. To know whether an answer is right, AI often needs observations, experiments, domain knowledge, or actions in the physical world. Humans — and the physical environment — are still part of the loop. This difference may turn out to be very important as we think about where AI will advance fastest, and what is still missing for AI to make a real impact in industry. We also discussed many other interesting topics around AI, knowledge, writing, and how humans work with machines. A very enjoyable conversation between two former physics students trying to understand where AI is taking us.

When water data sits in different systems, faults are harder to trace. See how this use case connects flow, pressure, and water-quality data on one timeline to locate leaks. tdengine.com/water-utilities… #TDengineUseCases #WaterOperations

Flow up 10%, pressure down 6%: leak or morning demand? A water-utilities use case on connecting flow, pressure, and chlorine data to investigate what happens next. tdengine.com/water-utilities… #TDengine #WaterUtilities

I’m a C/C++ programmer on Linux. I’ve never programmed in Go, and I’ve never touched Windows PowerShell. But today, with the help of AI, I can review and fix bugs in the TDengine installer — even though the codebase includes Go, Linux shell scripts, Windows PowerShell, and YAML. As the screenshot shows, AI can even split the work across multiple agents, inspect different parts of the codebase in parallel, and bring the findings back together. I think the same thing is going to happen with industrial operations data. Today, extracting insights from industrial data often requires specialized knowledge. You need to understand root cause analysis, process behavior, batches, SPC, time-series analytics, and sometimes the details of a particular historian or analytics tool. AI is going to change this. Imagine simply asking: “Why did this batch take 20% longer than normal?” “What caused the quality deviation?” “Why has the efficiency of this compressor been declining?” “What is unusual in my plant today?” You don’t necessarily need to know which statistical method to use, how to perform root cause analysis, or how SPC works. AI can choose the appropriate analysis, examine the relevant signals and events, compare historical operating conditions, and explain what is happening. But there is one critical prerequisite: AI must understand what your industrial data means. Collecting time-series data is not enough. We need an industrial ontology that gives the data context and meaning — what assets exist, how they are related, what each signal represents, its unit, operating state, events, batches, processes, and relationships with other data. The industrial ontology becomes a semantic layer between raw operational data and AI. Once the operational data is collected and organized through this ontology, AI can reason over the plant much like it reasons over a software codebase today. You don’t need to master every analytical technique. You need to know what you want to understand. That, to me, is one of the biggest changes industrial AI will bring.

TDengine is heading to ICC2026 in Sacramento, September 22–24. Come see how TDengine provides a modern alternative to traditional data historians for Ignition architectures. Find us at Booth 4️⃣8️⃣ @InductiveAuto #ICCUnleashed #Ignition