Smart House Load Management Scheme Using Scoring Rule Based Optimal Time Dependent Pricing

This paper presents an optimized load management scheme for smart house facilitating a scoring rule operated reward based time dependent pricing mechanism. Smart houses are capable establishing two-way communications between consumers and energy providers and thus enabling a feedback-control loop using a time dependent pricing (TDP). The energy provider (EP) persuades the consumers to manage their loads by charging them more in peak and less in off-peak hours. EP monitors the network load to estimate consumers' willingness to shift their demand and utilizes it to announce the price of the next slot of periods. The response of consumers to these prices are then used to refine EP's estimation. Such response is comprised by their estimated device shifting periods associated with corresponding confidence for the coming periods. Additionally, the actual margin of shifting for previous periods are also recorded as a parameter of response. The EP incentivizes the consumers by officering rewards (or discount) over the price depending on the margin of accuracy of shifting-period forecast and closeness of the device schedule towards lower demand period. Such reward function is formed based on a Continuous Ranked Probability Score (CRPS). CRPS has the ability to be a strictly proper scoring rule as well as to assess the closeness of estimation. These properties are essential to design the reward mechanism of EP. EP optimizes the TDP by minimizing the total cost of offering rewards and satisfying consumers' demand. The simulation results will show the effectiveness of the proposed method. The method is tested on real data provided by Ontario power company. As of preliminary simulation, it will be shown that the method is capable of reducing the hourly demand by properly incentivizing the consumers on their ability to shift the household devices.