Unlock instant, AI-driven research and patent intelligence for your innovation.
Energy storage management method and system based on random batch gradient descent algorithm
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A technology of gradient descent algorithm and management method, applied in the field of energy storage management method and system based on stochastic batch gradient descent algorithm, can solve problems such as inability to perform real-time data collection, periodic offline training and model iterative update, high power consumption, etc. , to achieve the effect of optimizing electricity cost and electricity cost
Pending Publication Date: 2022-03-08
SHANDONG INSPUR SCI RES INST CO LTD
View PDF0 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0003] Therefore, edge-end devices should collect data locally and perform incremental training. However, in the process of using machine learning algorithms for local data collection and incremental training, edge-end devices consume a lot of power and cannot perform real-time data collection and periodic offline Training and iterative model updates
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0050] combined with figure 1 , this embodiment proposes an energy storage management method based on the stochastic batch gradient descent algorithm, and its implementation includes:
[0051] (1) Collect historical power consumption data,
[0052](1.1) Take out the data of four consecutive weeks and decompose the time series to obtain the daily fluctuation cycle of electricity consumption, the trend of electricity consumption change, and the residual of random fluctuation of electricity consumption, and further obtain the continuous time series model through these three sets of data;
[0053] (1.2) Take out the power consumption data of the past ten weeks, divide them into seven groups according to seven days a week, and perform time series decomposition to obtain the power consumption fluctuation cycle, power consumption change trend, and power consumption random fluctuation residual of the same day, And through these three sets of data, a discrete time series model is furt...
Embodiment 2
[0067] combined with figure 2 , this embodiment proposes an energy storage management system based on a stochastic batch gradient descent algorithm, which includes:
[0068] Data collection module 1, used to collect historical power consumption data;
[0069] Data processing module 12 is used to take out the data of four consecutive weeks for time series decomposition to obtain the daily fluctuation cycle of electricity consumption, the trend of electricity consumption change, and the residual of random fluctuation of electricity consumption, and further obtain continuous Timing model;
[0070] Data processing module 2 3 is used to take out the electricity consumption data of the past ten weeks, divide them into seven groups according to the seven days of the week, and perform time series decomposition to obtain the electricity consumption fluctuation cycle, electricity consumption change trend, and electricity consumption of the same day The random fluctuation residual, an...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
PUM
Login to View More
Abstract
The invention discloses an energy storage management method and system based on a random batch gradient descent algorithm, and relates to the technical field of energy storage management, and the method comprises the steps: collecting historical electricity consumption data, taking out continuous surrounding data, and carrying out the time series decomposition, the method comprises the following steps: acquiring a daily electricity consumption fluctuation period, an electricity consumption change trend and an electricity consumption random fluctuation residual error, extracting electricity consumption data of past ten weeks, dividing the data into seven groups according to seven days in one week, and performing time sequence decomposition to obtain an electricity consumption fluctuation period, an electricity consumption change trend and an electricity consumption random fluctuation residual error in the same day; historical generating capacity data and weather data are collected, and linear regression modeling is carried out on generating capacity; and combining the obtained data and model and the residual electric quantity data of the electricity storage equipment, giving t, substituting into a cost function, and optimizing the cost function by using a random batch gradient descent algorithm so as to optimally select the power supply power according to the use time period. According to the invention, overall power supply and peak load shifting can be realized.
Description
technical field [0001] The invention relates to the technical field of energy storage management, in particular to an energy storage management method and system based on a stochastic batch gradient descent algorithm. Background technique [0002] In the real world, every new scene can generate data patterns that have never been seen before. When a machine learning model is deployed on an edge device, when the model encounters a new data pattern, the solidified model cannot be used for the new data pattern. The data schema responds correctly and therefore produces incorrect results. Furthermore, training a model on data from one context and deploying it to another context often does not yield the desired results. In fact, it is usually not feasible to train different models for different contexts, because each model needs to collect, label, process data and adjust the parameters of the model. [0003] Therefore, edge-end devices should collect data locally and perform incr...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More