CN-122000939-A - Peak-to-valley self-adaptive scheduling method and system for household power load

CN122000939ACN 122000939 ACN122000939 ACN 122000939ACN-122000939-A

Abstract

The invention discloses a peak-valley self-adaptive scheduling method and system for household electrical loads, and relates to the technical field of load optimization. The method comprises the steps of collecting historical electricity utilization data of all electric equipment in a household, analyzing electricity utilization characteristics to carry out marking of rigid load, transferable load and interruptible load to generate electricity utilization characteristic marks, establishing an electricity utilization load prediction model of a household in a future preset period, and solving an optimal scheduling strategy of all the electric equipment in the future preset period by combining the electricity utilization characteristic marks based on output of the electricity utilization load prediction model and peak-valley electricity price period information and with the aim of minimizing electricity utilization cost, minimizing user comfort level deviation and minimizing peak-valley difference of a load curve. The technical problem that in the prior art, household electricity load scheduling is difficult to combine peak-valley electricity price and user preference to perform dynamic optimization adjustment, so that the electricity cost is high is solved, and the technical effect of reducing the household electricity cost on the premise of meeting the electricity demand of a user is achieved.

Inventors

CAI RUINING
LIU CHANG
WANG HAIJUN

Assignees

南京信息工程大学

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (10)

1. A peak-to-valley adaptive scheduling method for household electrical loads, the method comprising: Collecting historical electricity utilization data of each electric equipment in a household, analyzing the electricity utilization characteristics to carry out marking of rigid load, transferable load and interruptible load, and generating electricity utilization characteristic marks; Collecting historical load data and user setting preference data of each electric equipment in a household, and peak-valley electricity price time period information issued by a power grid in real time, and establishing an electricity load prediction model of a preset time period in the future of the household based on the historical load data and the user setting preference data; and based on the output of the electricity load prediction model and the peak-valley electricity price time period information, combining the electricity characteristic marks, aiming at minimizing electricity cost, minimizing user comfort level deviation and minimizing peak-valley difference of a load curve, and solving an optimal scheduling strategy of each electric equipment in a future preset time period.
2. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 1, wherein, based on the output of the electrical load prediction model and the peak-to-valley electricity price period information, in combination with the electrical characteristic label, solving an optimal scheduling policy of each electrical device in a future preset period with the objective of minimizing electrical cost, minimizing user comfort deviation and minimizing peak-to-valley difference of load curve, comprises: acquiring a predicted electric equipment scheduling time sequence based on the output of the electric load prediction model; Extracting candidate scheduling equipment with marks of transferable loads and interruptible loads based on the electrical characteristic marks, determining the schedulable priority to be arranged in a descending order, and generating a scheduling equipment sequence; and establishing an objective function with minimized electricity cost, minimized user comfort deviation and minimized peak-valley difference of a load curve based on the peak-valley electricity price time period information, and performing equipment electricity utilization time adjustment according to the scheduling equipment sequence based on a preset scheduling mechanism to generate the optimal scheduling strategy.
3. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 2, wherein the preset scheduling mechanism comprises: The transferable load translates to a low-cost period of time under the constraint of meeting a time window, and the interruptible load adjusts the operating power or starts and stops within a range meeting a preset comfort level.
4. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 2, wherein before performing the device electrical time adjustment by the electrical characteristic marking preset scheduling mechanism, further comprising: determining a current electricity consumption peak period and an electricity consumption valley period based on the peak-valley electricity price period information; Analyzing load balance indexes of the electricity consumption peak period and the electricity consumption valley period according to the predicted dispatching time sequence of the electric equipment; And if the load balance index meets the preset balance index, not performing equipment power utilization time adjustment, and if the load balance index does not meet the preset balance index, performing equipment power utilization time adjustment through a preset scheduling mechanism of the power utilization characteristic mark.
5. The peak-to-valley adaptive scheduling method for household electrical loads according to claim 2, further comprising, after generating said optimal scheduling policy: receiving response feedback information of the family user to the optimal scheduling strategy; according to the response feedback information, counting manual change records of the user on the optimal scheduling strategy; and adjusting the weight of each evaluation item in the objective function based on the manual change record, and synchronously carrying out the schedulable priority adjustment of the equipment.
6. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 1, wherein the schedulable priority of the candidate scheduling device is set based on the power consumption and time resilience of the corresponding device.
7. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 1, wherein collecting historical electrical data of each electrical consumer in a household, analyzing electrical characteristics for marking rigid load, transferable load and interruptible load, generating electrical characteristic marks, comprises: Based on the historical electricity utilization data, extracting a power change mode, a start-stop rule and user operation time distribution of each electric equipment as key characteristic parameters; And matching the key characteristic parameters with a preset classification rule base, and distributing load type marks for each electric equipment according to the matching result to generate the electric characteristic marks.
8. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 7, wherein the transferable load and the interruptible load are simultaneously marked on a same electrical device when the load type marks are assigned to each electrical device.
9. The peak-to-valley adaptive scheduling method of household electrical loads according to claim 1, wherein establishing the electrical load prediction model for the future preset period of the household based on the historical load data and the user setting preference data comprises: Constructing a hybrid neural network prediction model, wherein the hybrid neural network prediction model comprises a long-term and short-term memory network layer for capturing a load time sequence change rule and a full-connection layer for fusing user preferences and environmental characteristics; And performing time sequence processing on the historical load data, combining corresponding user setting preference data as a training sample, performing supervision training on the hybrid neural network prediction model, and establishing the electricity load prediction model.
10. A peak-to-valley adaptive scheduling system for a household electrical load, for implementing a peak-to-valley adaptive scheduling method for a household electrical load according to any one of claims 1-9, said system comprising: The data analysis module is used for collecting historical electricity utilization data of each electric equipment in the household, analyzing the electricity utilization characteristics to carry out marking of rigid load, transferable load and interruptible load, and generating electricity utilization characteristic marks; The model building module is used for collecting historical load data of each electric equipment in the household, user setting preference data and peak-valley electricity price time period information issued by the power grid in real time, and building an electricity load prediction model of a preset time period in the future of the household based on the historical load data and the user setting preference data; and the solving module is used for solving the optimal scheduling strategy of each electric equipment in a future preset period by combining the electricity utilization characteristic mark based on the output of the electricity utilization load prediction model and the peak-valley electricity price period information and taking the minimization of electricity utilization cost, the minimization of user comfort level deviation and the minimization of peak-valley difference of a load curve as targets.

Description

Peak-to-valley self-adaptive scheduling method and system for household power load Technical Field The invention relates to the technical field of load optimization, in particular to a peak-valley self-adaptive scheduling method and system for household electrical loads. Background With the improvement of living standard of residents, the types and the number of household electric equipment are continuously increased, and household electric loads have the characteristics of diversified electric behaviors, centralized time period, obvious load fluctuation and the like. Especially in the period of higher peak-time electricity price, a plurality of high-power electric equipment operates simultaneously, so that the household electricity cost is easy to rise, and a larger pressure is brought to the operation of the power distribution network. The existing household power load scheduling scheme is controlled based on fixed time interval rules or simple electricity price thresholds, and usually only starts and stops or delays running of partial electric equipment according to preset peak-valley time intervals, so that effective distinction of load characteristic differences of different electric equipment in a household is lacking. Meanwhile, the existing scheme often fails to fully combine the personalized electricity consumption preference of the user and the prediction information of the household electricity consumption load in the scheduling process, so that the scheduling strategy is insufficient in adaptability, and dynamic adjustment is difficult to carry out according to electricity consumption behavior change and electricity price fluctuation. In addition, in the household electric field scene of parallel operation of multiple devices, the prior art is mainly based on a single optimization target, focuses on reducing electricity consumption cost, ignores factors such as comfort level of users and smooth load curve, and is easy to cause the reduction of electricity consumption experience or the excessive concentration of load in low price period, so that the adjusting effect of peak-to-valley electricity price mechanism is weakened. Disclosure of Invention The application provides a peak-valley self-adaptive scheduling method and system for household electricity loads, which solve the technical problem that in the prior art, household electricity load scheduling is difficult to dynamically optimize and adjust in combination with peak-valley electricity price and user preference, so that electricity cost is high. In a first aspect of the present application, there is provided a peak-to-valley adaptive scheduling method for household electrical loads, the method comprising: The method comprises the steps of collecting historical electricity utilization data of all electric equipment in a household, analyzing the electricity utilization characteristics to carry out marking of rigid load, transferable load and interruptible load to generate electricity utilization characteristic marks, collecting historical load data of all electric equipment in the household, user setting preference data and peak-valley electricity price time period information issued by a power grid in real time, establishing an electricity utilization load prediction model of a household in a future preset time period based on the historical load data and the user setting preference data, and solving an optimal scheduling strategy of all electric equipment in the future preset time period by combining the electricity utilization characteristic marks based on output of the electricity utilization load prediction model and the peak-valley electricity price time period information and taking minimization of electricity utilization cost, user comfort deviation minimization and load curve peak-valley difference minimization as targets. In a second aspect of the present application, there is provided a peak-to-valley adaptive scheduling system for household electrical loads, the system comprising: The system comprises a data analysis module, a model establishment module, a solving module and a power utilization characteristic marking module, wherein the data analysis module is used for collecting historical power utilization data of all electric equipment in a household, analyzing power utilization characteristics to carry out marking of rigid load, transferable load and interruptible load and generating power utilization characteristic marking, the model establishment module is used for collecting historical load data of all electric equipment in the household, user setting preference data and peak-valley power price time period information issued by a power grid in real time, establishing a power utilization load prediction model of a household in a future preset time period based on the historical load data and the user setting preference data, and the solving module is used for solving an optimal scheduling strategy of all electric equipment