Automated method of generalized building automation from atomic physical models and control loops thereof
Disclosed is a system and method for classifying units of control and concomitant model units from a larger physical system model. The system enables model-based simulation of decoupled system units for improved simulation efficiency. Further, the system allows for automated reasoning about the classification of said units, producing both machine-interpretable and human-readable, natural language descriptions of the decisions and reasons for said decisions made during the classification process, for use in heuristics and user feedback.
1. A method comprising:
classifying individual atomic units of a physical model proxying a physical system under control, wherein each atomic unit comprises a source, a sink, a transport, a sensor, a threshold, and a control loop;
inferring atomic units of control from the physical model;
consuming, by an automated reasoning system, the classified atomic units as reasoning entities in a knowledge graph to provide a reasoned description of and motivation for the classifying of the individual atomic units;
using the knowledge graph to make actionable decisions using the atomic units of the physical model to actuate the inferred atomic units of control; and
whereby:
the source comprises a physical model of at least one source;
the sink comprises a physical model of at least one sink;
the transport comprises a physical model of at least one actuated means of transport; and
wherein:
the transport is interposed between the source and the sink, so that it forms a controlled system;
the sensor comprises at least one sensor providing a means of feedback for the controlled system;
the atomic unit simulates a physical behavior of the individual atomic units of the physical model;
the physical model infers the control loop;
the sensor forms a feedback data source to simulate the physical model;
the threshold is a function of a state of the physical model; and
the transport is actuated by the threshold such that the control loop is formed.
2. The method of claim 1 , wherein the source comprises a producer, originator, or input of a measurable resource.
3. The method of claim 1 , wherein the source is comprised of at least one thermal, energy, air, or water source.
4. The method of claim 1 , wherein the source is comprised of utility generated electricity, site generated electricity, boiler, steam generator, gas turbine, gas heater, chiller, heat pump, adsorption heat pump, ground source heat pump, furnace, air conditioner, evaporative cooler, photovoltaics, solar hot water collector, wind turbine, hydro turbine, liquid or solid thermal storage tanks, mass thermal storage well, thermo-electric generators including Peltier junctions, Carnot cycle engines, Stirling engines, or water sources of irrigation.
5. The method of claim 1 , wherein the sink comprises a consumer, terminator, or output to a measurable resource.
6. The method of claim 1 , wherein the sink is comprised of at least one thermal, energy, air, or water sink.
7. The method of claim 1 , wherein the sink is comprised of buildings, building zones, building surfaces, building surface interlayers, electric batteries, electric loads, outdoor surfaces including snow melt surfaces, irrigation consuming masses, HVAC system equipment, functional control equipment, lights, motors, liquid or solid thermal storage tanks, mass thermal storage, or phase change materials.
8. The method of claim 1 , wherein the automated reasoning system may be used algorithmically to select only the relevant atomic units of control to simulate system behavior.
9. The method of claim 1 , wherein the reasoned description may be machine-interpretable and usable as a heuristic.
10. The method of claim 1 , wherein the reasoned description provides a human-readable or natural language description that may be used to inform users, either preemptively or retrospectively.
11. The method of claim 2 , wherein the resource has a value applied to it.
12. The method of claim 11 , wherein the value is monetary, economic, comfort, equipment longevity, or resource utilization.
13. The method of claim 11 , wherein the value has a discount and/or compound rate applied to it.
14. The method of claim 11 , wherein the threshold is computed from the value.
15. The method of claim 1 , wherein the knowledge graph comprises: heuristic classification methods, probabilistic models, graphical models, Bayesian networks, neural networks, knowledge base, BDI models, factor graphs, Markov random fields, Markov chains, Markov decision processes, partially observable Markov decision processes, hidden Markov models, or self-organizing maps.
16. The method of claim 1 , wherein the knowledge graph includes heuristic classification methods comprised of decomposition methods, inductive methods, reduction methods, constructive methods, or local search methods.