Power Router Queuing for Coincident Peak Demand
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The electric utility industry faces challenges in managing coincident peak demand, where multiple appliances draw power simultaneously, leading to increased costs for utility providers and consumers due to the need for additional generating capacity, despite this event occurring less than 5% of the time.
Innovation Solution
A method involving a power router that uses a queuing system to manage access to a shared power supply among independent electrical access points, employing a transformed Erlang C probability distribution to prioritize access and minimize coincident peak demand by controlling the turn-on times of appliances based on their power draw and operational states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If electric power supply capacity is increased to meet coincident peak demand, then reliability of power supply is improved, but cost of power infrastructure increases
Solution Approach 1:
The system performs preliminary actions by predicting future power demand using historical data and machine learning algorithms. It proactively schedules power distribution before coincident peak demand occurs, pre-positioning power allocation to prevent simultaneous high-demand events rather than reacting to them
Solution Approach 2:
The system implements periodic action through scheduled power distribution cycles and time-based pricing mechanisms. It creates periodic patterns in power consumption by encouraging off-peak usage through reduced pricing during low-demand periods, thereby smoothing demand fluctuations over time
2Quantity of substance
If power distribution is optimized to reduce coincident peak demand, then cost of power infrastructure decreases, but complexity of power management increases
Solution Approach 1:
The system introduces an intermediary layer of intelligent software agents and machine learning models that mediate between power supply and demand. These intermediaries automatically analyze patterns, make scheduling decisions, and coordinate power distribution without requiring direct complex control of individual appliances, thereby managing complexity centrally rather than distributedly
Solution Approach 2:
The system changes key parameters such as pricing structures, scheduling windows, and demand thresholds to optimize power distribution. By adjusting these parameters dynamically based on grid conditions and learned patterns, the system achieves efficient load management without increasing physical infrastructure complexity
3Ease of operation
If appliances are allowed unrestricted access to power supply, then ease of operation is improved, but coincident peak demand increases
Solution Approach 1:
The system implements dynamic power access control where appliance permissions and scheduling constraints are adjusted in real-time based on current grid conditions, predicted demand patterns, and user preferences. This dynamic approach maintains ease of operation during low-demand periods while automatically imposing scheduling restrictions during high-demand periods to prevent coincident peaks
Data Source
AI summary
A method of queuing access to a power supply shared by a set of electrical access points. The access points turn on independently from one another and thus have independent power draws. Each access point has a specific power draw when on. The on state and associated power draw of each of access point is identified, and a load duration curve for each access point is normalized (i.e., combined with load duration curve(s)) from the other access points) into a probability distribution function. The probability distribution function is a normalized load duration curve that thus accounts for a varying set of “operating states” that may occur with respect to the set of access points (when viewed collectively). Each operating state has an associated probability of occurrence. As the operating state of the set (of access points) changes, access to the power supply is selectively queued, or de-queued (if previously queued).