Modeling power usage on a boat

[evm1024]

I've been working on a model to calculate the power usage on boats.

Computer models in general take a bunch of inputs and then step through time. At each time-step the model return the state and that state is used for the input to the next time-step.

The fun thing about models is that they do not have to be accurate to be useful. Just changing initial parameters and running the model again allows you to get some insight as to what changes in the real word might be like.

of course a good modeler will take a model run and compare it to the real-world and then make changes to the model to try to get closer to reality.

IN my last job we were modeling the interaction of the Columbia River and the near Pacific Ocean. (mostly I just made the computer cluster run fast, not so much modeling)

The real fun was that in order to test the model we needed data from the ocean. And to that end we had research ships that we could go out on and take measurements from. Nothing like a cruise on Wecoma, Pt Sur, New Horizons and other ships.

Anyway, this model is written in perl and is not quite ready for prime time. That is a nice way to say that I just threw the code together and any structure in more or less by chance. Not my best example of coding.

It is currently set up to model FLA and LiFePO4 battery types. It has solar and an alternator for charge sources and just a general table for ships loads.

I'll be adding in other parameters as time goes on.

At some point I'll be asking for parameters from some cruisers so that I can do a model run and see if the results are anywhere near what you see on your boat.

Initial runs are showing how much of the time that FLA spends in PSOC....

OF course when the code is not too much of an embarrassment to me I'll publish the code.

What factors would you like to see in the model?

[isuee94]

A few things come to mind:
- Battery bank type / capacity
- Charging inputs (i.e. solar / wind / tow generator)
- List of loads, consumptions, typical usage per day

[S/V Illusion]

Academic exercises can sometimes be fun but given the ubiquitous nature of power monitoring devices on boats these days which can tell everything critical about power management, why do you think this is any utility?

[evm1024]

Originally Posted by S/V Illusion

Academic exercises can sometimes be fun but given the ubiquitous nature of power monitoring devices on boats these days which can tell everything critical about power management, why do you think this is any utility?

Yes, they can be fun. Which may be all the reason needed.

And the other answer is: What if?

What if I add in this load? Will I have to run the generator?

What if I add in 100 watts more solar. Will my batteries last longer?

What if I lost half my solar on passage? How much would I need to run the engine or honda?

[Singularity]

For data acquisition, consider looking into DIY home automation systems. Cheap, options are limited to your imagination (e.g. monitor your old diesel engine, leak sensors, cameras, sun levels, etc).

Otherwise when it comes to power on a boat I would not fall prey to false scarcity thinking/engineering and/or false economy.

Is saving weight they main goal? Saving money? Increasing safety? Pure research/curiosity? What utility will this endeavor provide that people need and can use? Some examples of how the finished product changes human behavior?

[Paul L]

Multiple modes of operation, e.g. at anchor, sailing with AP and instruments, day/night

[Mirage35]

Originally Posted by S/V Illusion

Academic exercises can sometimes be fun but given the ubiquitous nature of power monitoring devices on boats these days which can tell everything critical about power management, why do you think this is any utility?

Maybe I don’t have the latest technology, but the power monitoring on my boat tells me what’s happening, but not why it’s happening. So I have started paying a lot more attention to which systems take how much power on what cycle in what circumstances. Basically building a mental model to help me manage and understand.

[evm1024]

Originally Posted by AnglaisInHull

Maybe I don’t have the latest technology, but the power monitoring on my boat tells me what’s happening, but not why it’s happening. So I have started paying a lot more attention to which systems take how much power on what cycle in what circumstances. Basically building a mental model to help me manage and understand.

That's the ticket. A tool for fun and to help predict what could be going on.

[Singularity]

Originally Posted by evm1024

That's the ticket. A tool for fun and to help predict what could be going on.

On a hopefully related tangent, a smart system that I would like to have would be one that could provide feedback data for performance under sail. Perhaps such systems exist for the racers, but one that might measure accelerations, heel, etc, allow data entry for sail position, load distribution. Basically a system to provide more granular feedback of the consequence of user inputs (to rudder, trims) with the goal of really finding the sweet spots for the given design, sails, and vessel loading, sea conditions, etc.

If this makes sense, something akin to the brains of a fly-by-wire system used in an aircraft, only in the boat the providing data to the helmsman to help supplement what his senses and experience is otherwise telling him.

Otherwise for the power app:
I can't imagine how to model, for example, the myriad possible solar factors and outcomes, from panel size and efficiency, orientation, orientation with respect to hemisphere, all of this vs infinite weather calculations, etc. The closest I ever came to your (OP) line of work was seeing some human factors research at NASA. They had multiple teams devising models, mad Monte Carlo runs for problems very simple in comparison to what you are proposing.

The problem with boats is that a backup power system costs the average boat owner less than it costs to buy a program to see when you will need a backup power system.

[Paul Elliott]

I did something (I suppose) similar with a fairly simple spreadsheet. I had my solar charging sources, my diesel burn-rate while charging and the amps delivered during alternator charging, Fuel tank capacity, all my loads and anticipated usage for each. The result was a daily Ah usage, and by varying the solar output (due to cloudy days, shading, etc) I would get the number of hours per day or week I would need to run my engine for charging. I had at-anchor and at-sea power drain configurations.

The spreadsheet did a pretty good job of modeling the actual real-world results I got. Will your program do something similar? I suppose a program might make it easier to play "what if" games than my spreadsheet...

[Singularity]

Originally Posted by Paul Elliott

I did something (I suppose) similar with a fairly simple spreadsheet. I had my solar charging sources, my diesel burn-rate while charging and the amps delivered during alternator charging, Fuel tank capacity, all my loads and anticipated usage for each. The result was a daily Ah usage, and by varying the solar output (due to cloudy days, shading, etc) I would get the number of hours per day or week I would need to run my engine for charging. I had at-anchor and at-sea power drain configurations.

The spreadsheet did a pretty good job of modeling the actual real-world results I got. Will your program do something similar? I suppose a program might make it easier to play "what if" games than my spreadsheet...

Consider checking out the Home Assistant page. DIY home automation freeware app. You can let your smart device continually monitor and log in real time (as often as you'd like) all your fluid and electron flows, capacities, production, etc. Run reports. Run protocols "turn on and record bilge light and camera whenever bilge pump is triggered" etc. Up to you and your imagination. Costs next to nothing.

https://www.home-assistant.io/cookbook/automation_for_rainy_days/

[Paul Elliott]

I find that as I get more familiar with my usage patterns, I'm less and less interested in fancy instrumentation. I've got some good boat-geek friends who have their boats totally wired for this stuff, and while I originally was leaning in that direction (I'm an engineer and like the geeky stuff), at this point I'm contemplating ripping out some of the monitoring gear I already have. It's fun to look at, but in practice I just don't need it.

I do enjoy seeing what other people are doing in this area though.

[RaymondR]

I can see some value in what you are doing and would like to suggest that if you are going to go to the trouble of writing code you do it in a language which would allow the modeling to be incorporated in a boat power monitoring and control system at a later date.

The boat I have now came with a battery monitoring system for two banks, house and cranking, a good battery charger, an MPPT solar controller and a smart regulator on the engines 12V DC charging system.

The battery monitoring system appears to work ok but does not log any data.

The battery charger is installed in the lazarette where I can't get at it and may or may not have a data logging function. Since it's hard to access I have not bothered to research it until I relocate it.

The smart regulator has no data logging function.

The MPPT solar regulator has no data logging function.

I could implement data logging for the various functions and use the data in a spread sheet but it's just too much mucking about to be used on a regular basis.

Modelling may have some value however if the modelling could be tied into a system which both modeled the DC system and logged data on the performance of the individual components of the DC generation, storage and load performances which it could then compare the modeled performance, and perhaps even manage warnings and load shedding, it would have real value to us cruisers.

[akprb]

I must be getting old.

First, I admire the shear brain power exhibited on this thread and thank every one of you who make stuff better and more efficient.

For me it’s like finances. Use less than I make. I probably should have modeled from time to time in the past but just not that bright.

Was more fun to be a electricity tyrant and now even those days are gone thanks to LEDs.

What I really liked about the original post was evm1024’s technique for getting out of the lab and onto some boats! Brilliant!

[Singularity]

Originally Posted by RaymondR

Modelling may have some value however if the modelling could be tied into a system which both modeled the DC system and logged data on the performance of the individual components of the DC generation, storage and load performances which it could then compare the modeled performance, and perhaps even manage warnings and load shedding, it would have real value to us cruisers.

If you get away from boat electronics and look at household system development you'l find more than you ever imagined. I'm not saying that anyone should want any of the data, rather that this technology is available for next to free (free software coupled with generic sensors (temp, pressure, volumes, flows, quantities, accelerations, direction, fire systems, cameras, tracking cameras)).

The boating community is only a fraction of a percent of the global community interested in these topics (modelling, efficiency, etc).