Computerized system and method for distributed energy resource scheduling
The present disclosure provides for distributed resource scheduling performed by an advanced resource scheduling (ARS) module implemented on a distributed grid management (DGM) server in a power system. The ARS module is configured to automatically generate a resource schedule for controllable distributed energy resources (e.g., resources that are remotely controllable by DGM server and ARS module) in a distribution network of the power system, such as power generation resources and energy storage resources, to provide power in a cost-effective (e.g., optimal) manner. The ARS module is configured to take into account the operating limits of the distributed energy resources (DERs), the cost curves of the DERs, the system load demand, and other operating constraints to determine the most economical operating plan for the DERs, using an optimization technique such as the particle swarm optimization (PSO) algorithm.
1. A method performed by a computer system for generating a resource schedule, the method comprising:
automatically generating a resource schedule for a set of resources in a distribution network, wherein the automatically generating comprises
determining a solution using a particle swarm optimization technique,
the resource schedule is automatically generated using the solution,
the solution indicates calculated power output levels for the set of resources in the distribution network, and
the calculated power output levels of the set of resources, in aggregate, satisfy a power demand placed on the distribution network; and
deriving operating statuses of the set of resources from the calculated power output levels of the set of resources, wherein
if a calculated power output level falls within threshold power output levels, a corresponding operating status is derived to be online, and
if the calculated power output level falls outside the threshold power output levels, the corresponding operating status is derived to be offline.
2. The method of claim 1 , wherein the determining the solution further comprises:
determining an operating status of each of the set of resources;
initializing a particle swarm to initial conditions, wherein
the initial conditions are based on the operating status of each of the set of resources; and
solving an objective function for the solution, using the particle swarm, wherein
the objective function represents a total production cost incurred by the set of resources.
3. The method of claim 2 , further comprising:
determining whether the solution is feasible; and
in response to a determination that the solution is not feasible,
determining modified initial conditions of the particle swarm,
re-initializing the particle swarm to the modified initial conditions, and
re-solving the objective function for a second solution, using the particle swarm.
4. The method of claim 3 , wherein the determining modified initial conditions comprises:
changing an initial operating status of a first resource of the set of resources to a modified operating status.
5. The method of claim 4 , wherein
the initial operating status indicates the first resource is offline, and
the modified operating status indicates the first resource is online.
6. The method of claim 1 , further comprising:
building an objective function, wherein
the set of resources comprises a plurality of power generation devices and a plurality of energy storage devices,
the objective function represents a total production cost incurred by the plurality of power generation devices and by the plurality of energy storage devices, and
the particle swarm optimization technique minimizes the total production cost over a dispatch horizon.
7. The method of claim 6 , wherein
the objective function is subject to operating constraints of the plurality of power generation devices and the plurality of energy storage devices.
8. The method of claim 1 , further comprising:
determining a load profile of the distribution network, wherein the load profile indicates the power demand placed on the distribution network.
9. A non-transitory computer-readable storage medium configured to store program instructions that, when executed on a processor, are configured to cause the processor to at least perform:
automatically generating a resource schedule for a set of resources in a distribution network, wherein the automatically generating comprises
determining a solution using a particle swarm optimization technique,
the resource schedule is automatically generated using the solution, the solution indicates calculated power output levels for the set of resources in the distribution network, and
the calculated power output levels of the set of resources, in aggregate, satisfy a power demand placed on the distribution network; and
deriving operating statuses of the set of resources from the calculated power output levels of the set of resources, wherein
if a calculated power output level falls within threshold power output levels, a corresponding operating status is derived to be online, and
if the calculated power output level falls outside the threshold power output levels, the corresponding operating status is derived to be offline.
10. The non-transitory computer-readable storage medium of claim 9 , wherein the determining the solution further comprises:
determining an operating status of each of the set of resources;
initializing a particle swarm to initial conditions, wherein
the initial conditions are based on the operating status of each of the set of resources; and
solving an objective function for the solution, using the particle swarm, wherein
the objective function represents a total production cost incurred by the set of resources.
11. The non-transitory computer-readable storage medium of claim 10 , wherein the method further comprises:
determining whether the solution is feasible; and
in response to a determination that the solution is not feasible,
determining modified initial conditions of the particle swarm,
re-initializing the particle swarm to the modified initial conditions, and
re-solving the objective function for a second solution, using the particle swarm.
12. The non-transitory computer-readable storage medium of claim 9 , wherein the method further comprises:
building an objective function, wherein
the set of resources comprises a plurality of power generation devices and a plurality of energy storage devices,
the objective function represents a total production cost incurred by the plurality of power generation devices and by the plurality of energy storage devices, and
the particle swarm optimization technique minimizes the total production cost over a dispatch horizon.
13. An apparatus comprising:
a processor; and
a memory coupled to the processor and configured with stored instructions executable by the processor, the instructions when executed by the processor cause the processor to:
automatically generate a resource schedule for a set of resources in a distribution network, wherein
the instructions are further configured to
determine a solution using a particle swarm optimization technique,
the resource schedule is automatically generated using the solution,
the solution indicates calculated power output levels for the set of resources in the distribution network, and
the calculated power output levels of the set of resources, in aggregate, satisfy a power demand placed on the distribution network; and
derive operating statuses of the set of resources from the calculated power output levels of the set of resources, wherein
if a calculated power output level falls within threshold power output levels, a corresponding operating status is derived to be online, and
if the calculated power output level falls outside the threshold power output levels, the corresponding operating status is derived to be offline.
14. The apparatus of claim 13 , wherein the instructions further include instructions that when executed by the processor cause the processor to:
determine an operating status of each of the set of resources;
initialize a particle swarm to initial conditions, wherein
the initial conditions are based on the operating status of each of the set of resources; and
solve an objective function for the solution, with use of the particle swarm, wherein
the objective function represents a total production cost incurred by the set of resources.
15. The apparatus of claim 14 , wherein the instructions further include instructions that when executed by the processor cause the processor to:
determine whether the solution is feasible; and
in response to a determination that the solution is not feasible,
determine modified initial conditions of the particle swarm,
re-initialize the particle swarm to the modified initial conditions, and
re-solve the objective function for a second solution, with use of the particle swarm.
16. The apparatus of claim 13 , wherein the instructions further include instructions that when executed by the processor cause the processor to:
build an objective function, wherein
the set of resources comprises a plurality of power generation devices and a plurality of energy storage devices,
the objective function represents a total production cost incurred by the plurality of power generation devices and by the plurality of energy storage devices, and
the particle swarm optimization technique minimizes the total production cost over a dispatch horizon.