Cross-Coupled Active Disturbance Rejection Synchronization Control of Load-Side Heterogeneous Dual Hydraulic Cylinders for a Temporary Support Device
This paper proposes an IPSO-tuned cross-coupled Active Disturbance Rejection Control (LADRC + CCC) strategy to achieve high-precision synchronization of load-side heterogeneous dual hydraulic cylinders in temporary support devices by effectively compensating for load path differences, flexibility, and contact state variations.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are trying to push a giant, heavy shopping cart with two friends, one on the left handle and one on the right. If you both push with the exact same strength and the wheels on both sides roll perfectly smoothly, the cart moves straight. But what if the left wheel hits a patch of sticky mud while the right wheel rolls on smooth pavement? Or what if the left handle is slightly bent, making it harder to grip? Suddenly, your friend on the left has to work much harder, or maybe they get stuck, while your friend on the right zooms ahead. The cart starts to twist, wobble, and might even tip over. This is the daily struggle of heavy machinery in the real world: things are rarely perfectly symmetrical.
In the world of engineering, specifically for machines that dig tunnels and support the ground above them, this "twisting" problem is a big deal. These machines use pairs of hydraulic cylinders (think of them as super-strong, oil-powered pistons) to push the whole structure forward. If one piston moves faster than the other, the massive metal frame can get jammed, twist dangerously, or even break. Engineers have long tried to fix this with control systems—basically, the "brain" that tells the pistons how hard to push. Traditional brains use simple rules, like "if you're behind, push harder." But when the ground is uneven and the machine parts are slightly different from each other, these simple brains often get confused. They need a smarter way to handle the messiness of the real world, where friction, bending metal, and uneven loads are the norm, not the exception.
This is where a team of researchers from Shanxi Datong University and Inner Mongolia University of Science and Technology stepped in with a new idea for a "temporary support device" used in coal mine roadways. They tackled a specific headache: what happens when the two hydraulic cylinders are identical on the inside, but the world they are pushing against is totally different on the left and right sides? They built a digital twin of this system—a computer simulation where they could make the left side "sticky" and the right side "slippery" to see how different control strategies would handle the chaos.
The researchers discovered that the old way of tuning these controllers (basically guessing the right settings and hoping for the best) wasn't good enough. So, they invented a new "brain" for the machine. They combined three powerful tools: a system that can guess and cancel out disturbances (like a noise-canceling headphone for mechanical vibrations), a "cross-coupling" link that lets the two cylinders talk to each other (so if the left one gets stuck, the right one knows to slow down), and a super-smart optimization algorithm called "Improved Particle Swarm Optimization" (IPSO). You can think of IPSO as a swarm of digital birds searching for the best nest; they fly around the solution space, sharing information to find the perfect settings for the controller much faster and better than a human ever could.
In their simulations, this new "IPSO-LADRC+CCC" brain showed off some impressive moves. When the machine faced a sudden bump, a sensor glitch, or a sudden change in how hard it had to push, the new system kept the two cylinders moving in perfect lockstep. Under normal conditions, the maximum difference in how far the two cylinders moved was just 0.006503 meters (about 6.5 millimeters). To put that in perspective, the total distance they needed to travel was 0.8 meters. This means the error was only about 0.813% of the total trip. In contrast, older methods allowed errors of over 2%, which is like the shopping cart twisting enough to scrape the wall.
The paper suggests that this method is particularly good at handling the "messy" reality of mining, where one side might hit a rock (a sudden load change) or the electrical signal to the valve might get a bit weak (input gain attenuation). The simulation showed that even when these things happened, the new system recovered quickly and kept the error low. However, the authors are careful to note that this is all based on computer simulations. They haven't built a physical robot to test this in a real mine yet. They believe their math proves the system is stable and safe, but the real-world test is the next step.
The main takeaway is that by acknowledging that the left and right sides of a machine will never be perfectly identical in the real world, and by using a smart, self-tuning brain to manage those differences, engineers can keep massive support structures moving straight and safe. The study rules out the idea that simple, fixed settings are enough for these complex, uneven environments. Instead, it suggests that a dynamic, self-correcting approach is the key to preventing those dangerous twists and jams that could stop a mine's progress or, worse, cause an accident. While the numbers look promising in the simulation, the true test will come when this "smart brain" is put to work in the dusty, unpredictable tunnels of a real excavation site.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.