Guest Blog by Brennan Harris, Systems Engineer at Sentient Science
In 1641, the French philosopher René Descartes wrote a treatise called Meditations on First Philosophy, in which he posted the idea that his entire experience of reality could be false; that the world around him and even his own internal sensations could be manipulated by some external force beyond his control. He called this force the “evil demon.”
The notion that “reality” might not be real at all has recently resurfaced in the form of simulation theory. A lot of smart people have taken the mic to explain their position on the hypothesis, including Tesla/SpaceX CEO Elon Musk. Nick Bostrom is the most well-known and most quoted expert in the field.
The simulation hypothesis, as framed by Bostrom, basically says, “If we assume that any advanced civilization ever has achieved technology that is capable of simulating ‘a universe’ like the one we live in, we are almost certainly living in a simulated reality.”
One extreme implication of the hypothesis is that every imaginable phenomenon, particle, law, or experience that is observable in the universe is simulated; from galaxies to planets to people to quarks.
The hardware needed to perform such a simulation would be much, much more advanced than the largest supercomputers available today. The idea of processing such an amount of data is physically possible, though.
One essential aspect of a universe simulation is that macro-level entities would inherit their characteristics from micro-level ones. The interactions of the smallest imaginable particles would lead to the human-level and even galaxy-level realities. Designing such a simulation would be the most ambitious undertaking in physics.
Sentient Science is the pioneer on the simulation frontier.
Although Sentient doesn’t boast to integrate quantum theory with relativity or simulate human behavior or scale the characteristics of fundamental particles, it does have perhaps the world’s most exciting integrated approach to simulation. At Sentient, advanced simulation technology is used to predict the life of mechanical systems like drivetrains, helicopter rotors, spacecraft, and other advanced rotating systems.
It starts with a system-level simulation that uses validated technology to model the working environment of the mechanical system. We then simulate the propagation of loads through the mechanical system, extracting the resulting loads on each component.
A new kind of simulation is then performed, using the loads from the system model along with microscopic material characteristics of each component. With these inputs, we simulate the microscopic processes of crack propagation and micro-pitting: early damage events that are not detected by system performance changes or sensors.
These micro-damage models allow owners and operators of mechanical systems to make decisions about derating equipment and replacing components that can save their mechanical assets from catastrophic failure. The cost of replacing a $3,000 bearing in its early stages of damage (when the rest of the system is not affected) is in contrast to the loss of an entire gearbox (which costs about $450,000 in a 1.5 MW wind turbine).
Sentient Science is not at the door of proving the simulation hypothesis, but they are pushing in that direction. Right now, Sentient is implementing “bridges of simulation” between macroscopic and microscopic interactions. Hopefully, one day, these bridges will give way to a unified simulation that simulates human-level and atomic-level realities in one fell swoop.
Sentient will host a webinar on Dec. 7 at 1 p.m. on choosing the right components for Repowered Machines.