- Long Marine Laboratory, UC Santa Cruz:
This lab started around 1980 and in the early 90's our field engineer was hired to solve a major problem with their Seawater Delivery System
(1,000 Gallons/Minute, 8" dia. pipe) which is the lifeblood of every aspect of operations at the laboratory.
They had hired a caretaker to watch over the pump system and do whatever was needed to keep the pumps working and it was a constant battle for him to keep it working properly.
The Seawater Delivery System consists of two seawater intakes below the surf at the bottom of the cliff and out into the water about 20ft or so, that enter a caisson drilled 13 Ft always from the surf from the top of the cliff to below sealevel where two 3-Phase, 480VAC, 40Amp centrifugal pumps sit and pump seawater up from the bottom of the caisson to the top of the cliff where the contactors and power are housed in a "birdhouse" (control cabinet with a roof) to switch the high power needed to run the pumps, and across the land inland at the top of the giant silo looking tank that holds the fresh seawater reserve or ballast, and the second tank is the filtered seawater for the labs.
The controls used two PSIG tank level sensors that expand a coiled tube that uncoils as the water pressure increases in the tank do to gravity and has a mercury switch attached to the coil so when the pressure reaches the setpoint it tips the mercury switch on, and one was set for the high limit and the other for the low limit, a couple of relays, an alarm, and a switch for each pump, that's it.
The problems were this:
The tank liquid level limit switches would oscillate back and forth as the seawater entered the tank at the top and agitated the water which fluctuated the pressure, which causes the switch to turn on and off randomly as it neared it's setpoint. The bad news is that would cause the huge pumps to rapidly turn on and off and burnout the contacts in the power switching contactor
(a high power-switching relay), and waste all that electrical power because the inrush current to start the motors is huge compared to just running them.
The intakes under the surf would sometimes get clogged and needed to be cleared periodically so it was a dangerous task for the divers, and also as the surf would pound up against the cliff it would cause bubbles to form in the seawater at the intake and those bubbles would get trapped in the centrifugal pumps and create an air pocket in the pump chamber and cause the pump to cavitate and stop pumping seawater and free-wheel, so the caretaker had to stop the pump, wait for the air to vent out so the pump could create a vacuum again and start pumping seawater, at all hours of the night mind you.
All of the electrical was done in metal conduit and metal control boxes, which was a very bad idea because the salt in the air being at the shoreline was very corrosive, plus they would wash things down with the saltwater since there is an abundance, and everything was rusted out.
First we tore out the old metal conduit and replaced it with PVC plastic conduit which will never corrode, pulled in new wires with extras for signaling, and the 2' x 2' metal control box was replaced with a NEMA4X box made of fiberglass that was 3 ft. wide and 4ft. tall so we could make the control panel face a GUI (Graphic User Interface), long before being an app on your smartphone, back before cellphones and the Internet even, but graphics of the tanks, filter system, pumps, with the switches and indicator lights in the physical location of the device depicted on the panel using chart tape and them covering the whole panel with a clear plastic coating to repel the saltwater and air.
We tried to get them to put in a PLC (programmable logic controller) but they were still too new on the seen to be adopted by the engineers, so we did all the controls using a standard type, easy to find relays, to do all the logic functions, such as when the seawater level in the tank approaches it's setpoint, the first tip of the mercury switch on the preassure gauge locks in a relay that keeps the signal of the sensor from oscillating, a stable state, until it is unlatched by the other sensor through the control logic, and never chattering the contactor to destruction again.
Unfortunately, the estimate was over budget so the push was on to wrap it up and we had to scale back one of our schemes to monitor the intakes for clogging, but we were able to completely stop the cavitation problem of the pumps and alleviating the need for a caretaker at all by using a current-relay with setpoint, two time-delay relays, and a N.C. 1/4" solenoid valve mounted on the top of the pump chamber. When the pump gets the signal to run, it also starts a time-delay relay for about 3 seconds, just long enough for the pump to come up to speed and level-off the current used for pumping seawater, then the relay connects the current-relay to watch for the current to fall below a setpoint that happens when the pump cavitates and draws less current for not pumping anymore. When the setpoint is reached that current-relay interrupts the signal to the starter contactor of the pumps, and at the same time switches the solenoid valve open to vent the chamber, and at the same time starts the 2nd time-delay relay and waits a pre-determined time to completely vent the chamber which is a constant time, and when the 2nd timer has timed out the pump chamber is filled with seawater again and the circuit resets closing the valve and reconnecting the pump signal again. When this was implemented the pumps ran unattended for years and years without a hitch again...
Today, we can put the entire control system on a wireless module the size of a quarter, and we can do so much more using the built-in communications allowing supervisory control from AI
(Artificial Intelligence) in the cloud if that's what is needed at the time to automatically asses and even effect modifications...
Now, in example 1. we did power monitoring with special relays, the best approach at the time based on technology in the early 90's, and AI is the way it's done today in 2017.
Here is what we can do today and in the future:
Our AI system transforms large facilities into intelligent environments:
Advanced embedded systems boasting technologies like Artificial Intelligence (AI), artificial neural networks (ANNs) and deep neural networks (DNNs), machine learning, and cognitive (thinking, reasoning) seem to be popping up all over the place in 2017 -- often targeting application areas that many people would never have considered.
Our platform empowers buildings to monitor what is happening inside themselves and to easily communicate with the humans who run them, thereby reducing energy costs and power consumption, and predicting and solving problems before they happen.
Our solution is far cleverer than you might initially think. Our intelligent sensors learn the individual electrical 'fingerprints' of every single electric device within a building’s walls, down to the smallest iPhone charger, and the knee-jerk reaction is to assume that vast numbers of sensors have to be deployed, with individual units associated with every single power socket in a building. This would be incredibly expensive, time-consuming, and complex to implement, so it's fortunate we don’t do it this way.
In fact, the sensors are only associated with the building's main power supply panels. We start with the non-invasive Verdigris smart sensor, which is based on the Hall effect, and which simply clips over one of the feeds coming out of the building's main circuit breaker power panel.
Individual smart sensors are installed non-intrusively on every feed coming out of the electrical power panel, with all the sensors being daisy-chained together.
The smart sensors employ 8KHz analog-to-digital converters (ADCs) to sample the current profile with a high-degree of accuracy and resolution. The last sensor in the chain is connected to our wireless data transmitter (Our LoRa Radio Gateway-to-Internet device, as far as 1.25 miles away in optimum conditions), which communicates with the cloud. The sensors can send hundreds of thousands of data points per second to the cloud-based AI at our facility, which continuously analyzes and interprets the data, and then communicates its findings to your mobile app, where you can also effect controls with authorization.
Using machine learning techniques based on an artificial neural network, the our system is trained to recognize the power profiles associated with of a wide variety of devices, from the large motors used in elevators and HVAC systems to tiny smartphone chargers.
When the system is deployed into a real-world facility, all the data from the sensors is constantly uploaded to the our AI through the cloud. If an unknown power profile is detected, the building's engineers are prompted to identify and label the equipment associated with this profile, and this intelligence is added to the AI's knowledge database for subsequent use in existing and future deployments.
The end result is that, even when a building is fully occupied and its power systems are highly loaded, it's still possible for our system to tease out the fact that someone on the third floor just plugged in an electric shaver, for example.
The tracker is a mobile web app that provides real-time event tracking and notifications to the facility's engineering team. If the AI detects any abnormalities or building drift, it will use the tracker to inform the building's engineers.
If monitoring power consumption and identifying which devices are active at any particular time were all our system could do, it would still be a valuable tool, but it goes much farther than this. Using a physics-based modelling approach, the AI is also trained to recognize the power profiles associated with various failure modes associated with different pieces of equipment. This allows it to detect and identify units that are causing spikes in energy usage, that are potentially going to fail, or that have already failed, and to immediately alert the associated maintenance team.
Automated fault detection with notifications that can prevent unexpected equipment breakdown. Our system can save millions of dollars in reduced energy costs and equipment breakdowns for customers which is researching ways to apply our technology to predict failure on potential future operations.
Our AI becomes increasingly smart and more connected as existing customers use the system and as more facilities come online. The end result is to provide the AI with a massive repository of incredibly-specific, granular information about buildings, their energy use, the way in which their internal systems and appliances work, and the signals these systems send when they start to degrade and fail.
At this time, we are targeting our platform for use in large buildings and facilities, but don't be at all surprised to see our residential equivalents being deployed in the not-so-distant future -- in home retro-fitting and especially in new home construction -- as part of the migration to smarter, more efficient abodes.
What do you think about all of this?
We can apply our technology to literally anything...
We are Custom Data Solution and we would love to help you save time, money, and frustration, and increase your safety and security, in your daily operations.
Give us a call at Toll-Free: 855-415-7273 to discuss what we can do for you.