Method for improving automated scheduling, scheduling server and system for automated scheduling
A method, scheduling server and system for improving automated scheduling are provided. The automated scheduling is at least partially based on user settings input by users includes: obtaining user settings for the users, the user settings defining user preferences and personal constraints of each user, obtaining system boundary conditions from a schedule goal and technical conditions to achieve a desired goal and calculating a schedule. The acceptance probabilities for alternative user settings are predicted. At least one alternative user setting is selected and an alternative schedule for each selected alternative setting is calculated. The alternative schedules are compared to the initially calculated schedule with respect to quality measure. After reading in the user's response to the suggested alternative setting another alternative setting is selected in case the user declined, or the systems proceeds with accepting the alternative setting as obtained user specific setting.
1 . A method for improving automated scheduling of charging sessions executed in a system including a charging system for electric vehicles, wherein the automated scheduling is at least partially based on user settings input by a plurality of users, comprising the following steps:
obtaining, via a user interface, user settings for the plurality of users, the user settings defining user preferences and personal constraints of each user;
obtaining, via a system interface, system boundary conditions derived from a schedule goal and technical conditions of a technical equipment of the charging system involved to achieve a desired goal that shall be achieved by applying a schedule to be determined;
calculating, by at least one processor, the schedule by solving a scheduling problem specified from a combination of user specific settings and the system boundary conditions;
predicting, by the at least one processor, acceptance probabilities for alternative user settings, which are settings differing from the obtained user settings;
selecting, by the at least one processor, at least one alternative user setting and calculating an alternative schedule for each selected alternative user setting;
evaluating, by the at least one processor, the calculated alternative schedules and an initially calculated schedule with respect to a quality measure in order to determine an improvement or deterioration for each alternative schedule;
determining, by the at least one processor, based on a result of an evaluation and the predicted acceptance probabilities, an alternative setting to be suggested to a user whose user setting is concerned by the alternative setting;
reading in, by the at least one processor, a user's response to the suggested alternative setting via the user interface and selecting another alternative setting in case the user declined, or proceeding with accepting the alternative setting as obtained user specific setting and repeating calculation of a schedule, selection of alternative settings, calculation of alternative schedules, evaluation and selecting and suggesting an alternative user setting and reading in of the user's response, until a stop criterion is met;
outputting, via an output interface, a last calculated schedule to the plurality of users and to the charging system for controlling charging parameters of the charging system for charging the vehicle vehicles, wherein the last calculated schedule adapts charging parameters of the charging system; and
controlling, by the charging system, the technical equipment using the output last calculated schedule for directly controlling charging sessions of a plurality of charging stations based on the adapted charging parameters regarding technical limitations including at least one of peak electricity loads and peak energy costs.
2 . The method according to claim 1 ,
wherein prediction of the acceptance probabilities is based on at least one of a user's behavior history, a previous acceptance rate, user input information, and general domain knowledge.
3 . The method according to claim 2 ,
wherein the method determines from the alternative user settings having an acceptance probability above a set threshold the one with a greatest quality measure improvement.
4 . The method according to claim 1 ,
wherein the method determines from the alternative user settings having an acceptance probability above a set threshold the one with a greatest quality measure improvement.
5 . The method according to claim 4 ,
wherein the method estimates an inconvenience measure for the users and takes the inconvenience measure for determination of the alternative user setting to be suggested into account.
6 . The method according to claim 4 ,
wherein the selection prioritizes for suggesting an alternative user setting of a user having a lower inconvenience measure.
7 . The method according to claim 6 , wherein the inconvenience measure is accumulated over time.
8 . The method according to claim 1 ,
wherein the method estimates an inconvenience measure for the users and takes the inconvenience measure for determination of the alternative user setting to be suggested into account.
9 . The method according to claim 8 , wherein the inconvenience measure is accumulated over time.
10 . A system including a scheduling server for improving automated scheduling of charging sessions in a charging system for electric vehicles, and the charging system, wherein the automated scheduling is at least partially based on user settings input by a plurality of users, wherein the scheduling server comprises a user interface, a system interface, at least one processor, and an output interface,
wherein the user interface is configured to obtain user settings for the plurality of users, the user settings defining user preferences and personal constraints of each user,
wherein the system interface is configured to obtain system boundary conditions derived from a schedule goal and technical conditions of a technical equipment of the charging system involved to achieve a desired goal that shall be achieved by applying a schedule to be determined,
wherein the at least one processor is configured to:
calculate the schedule by solving a scheduling problem specified from a combination of user specific settings and the system boundary conditions;
predict acceptance probabilities for alternative user settings, which are settings differing from the obtained user settings;
select at least one alternative user setting and calculating an alternative schedule for each selected alternative user setting;
evaluate the calculated alternative schedules and an initially calculated schedule with respect to quality measure in order to determine an improvement or deterioration for each alternative schedule;
determine, based on a result of an evaluation and the predicted acceptance probabilities, an alternative setting to be suggested to a user whose user setting is concerned by the alternative setting; and
read in a user's response to the suggested alternative setting and select another alternative setting in case the user declined, or proceed with accepting the alternative setting as obtained user specific setting and repeating calculation of a schedule, selection of alternative settings, calculation of alternative schedules, evaluation and selecting and suggesting an alternative user setting and read in of the user's response, until a stop criterion is met,
wherein the output interface is configured to output a last calculated schedule to the plurality of users and to the charging system for controlling charging parameters of the charging system for charging the electric vehicles, wherein the last calculated schedule adapts charging parameters of the charging system, and
wherein the charging system is configured to control the technical equipment using the output last calculated schedule for directly controlling charging sessions of a plurality of charging stations based on the adapted charging parameters regarding technical limitations including at least one of peak electricity loads and peak energy costs.
11 . A system for improving automated scheduling comprising the scheduling server according to claim 10 and a user communication interface for outputting a request generated by the scheduling server and concerning an alternative user setting to a user.