Search

EP-4557187-B1 - A METHOD FOR PREDICTING ELECTRIC ENERGY CONSUMPTION IN AN ELECTRIC GRID

EP4557187B1EP 4557187 B1EP4557187 B1EP 4557187B1EP-4557187-B1

Inventors

  • LAURICELLA, MARCO
  • Mandelli, Giulia
  • Fenili, Lorenzo

Dates

Publication Date
20260506
Application Date
20231115

Claims (17)

  1. Method (100) for predicting electric energy consumption in an electric grid (1), said method being characterised by the following steps: - acquiring (101) first detection data ( D S1 ) including detection values related to an actual electric energy consumption in said electric grid; - acquiring (102, 103) additional detection data ( D S2 , D S3 ) including detection values related to the energy consumption in said electric grid during at least a time window ( TW 1 , TW 2 ) preceding a reference instant ( t R ); - acquiring (104) calendar data ( D C ) including chronological information associated to the operation of said electric grid; - calculating (105) training data ( D T ) based on the acquired detection data and calendar data; - based on said training data ( D T ), setting (106) a linear auto-regressive mathematical model ( M R ) describing the trend of the electric energy consumption in said electric grid, said linear auto-regressive mathematical model ( M R ) being configured to process at least a set of exogenous input values (U(k), U l (k)) indicative of at least a periodic function (U(t), U l (t) ) approximating the profile of the electric energy consumption in said electric grid over said at least a time window ( TW 1 , TW 2 ) preceding said reference instant ( t R ); - based on said linear auto-regressive model ( M R ), calculating (107) prediction data ( D P ) including prediction values related to the electric energy consumption in said electric grid during a time window ( TW 3 ) following said reference instant ( t R ).
  2. Method, according to claim 1, characterised in that it comprises the step of acquiring (102) second detection data ( D S2 ) including detection values related to the energy consumption in said electric grid during a first time window ( TW 1 ) preceding said reference instant ( t R ), wherein said linear auto-regressive mathematical model ( M R ) is configured to process first exogenous input values ( U(k) ) indicative of a first periodic function ( U(t) ) approximating the profile of the electric energy consumption in said electric grid over said first time window ( TW 1 ) .
  3. Method, according to one of the previous claims, characterised in that it comprises the step of acquiring (103) third detection data ( D S3 ) including detection values related to the energy consumption in said electric grid during a second time window ( TW 2 ) preceding said reference instant ( t R ), wherein said linear auto-regressive mathematical model ( M R ) is configured to process second exogenous input values ( U l (k) ) indicative of a second periodic function ( U l (t) ) approximating the profile of the electric energy consumption in said electric grid over said second time window ( TW 2 ) .
  4. Method, according to one of the previous claims, characterised in that the step (105) of calculating said training data ( D T ) includes processing the acquired first detection data ( D S1 ) to check the correctness of said data.
  5. Method, according to claim 2, characterised in that the step (105) of calculating said training data ( D T ) includes processing the acquired second detection data ( D S2 ) to identify the trend of the electric energy consumption in the electric grid during the first time window ( TW 1 ) .
  6. Method, according to claim 3, characterised in that the step (105) of calculating said training data ( D T ) includes processing the acquired third detection data ( D S3 ) to identify the trend of the electric energy consumption in the electric grid during the second time window ( TW 2 ).
  7. Method, according to one of the previous claims, characterised in that said prediction data ( D P ) are cyclically calculated with a predefined time granularity ( T P ) and with a predefined time horizon ( T H ).
  8. Method, according to one of the previous claims, characterised in that said linear auto-regressive mathematical model ( M R ) is a linear ARX mathematical model with one or more exogeneous inputs.
  9. Method, according to one of the previous claims, characterised in that setting said linear auto-regressive mathematical model ( M R ) includes: - setting a regression order ( m ) and a maximum number ( T max ) of training steps for said linear auto-regressive mathematical model ( M R ); - iteratively calculating one or more parameters ( θ ) of said linear auto-regressive mathematical model ( M R ) based on said training data ( D T ) during said training steps by solving an unconstrained linear problem established basing on the set regression order ( m ) and maximum number ( T max ) of training steps.
  10. Method, according to one of the previous claims, characterised in that setting said linear auto-regressive mathematical model ( M R ) includes tuning one or more parameters ( θ Ul ) of said linear auto-regressive model ( M R ) based on corresponding parameters ( θ Ul ) calculated during said training steps and one or more parameters ( θ " Ul ) calculated for previously set linear auto-regressive mathematical models ( M R ).
  11. Method, according to one of the previous claims, characterised in that it comprises the step (108) of carrying out a first check procedure to check the computational performances of said mathematical model.
  12. Method, according to claim 11, characterised in that said first check procedure (108) comprises: - comparing (108a) the first detection data ( D S1 ) acquired during a predefined checking period and the prediction data ( D P ) calculated during said checking interval; - calculating (108b) an error function (E) indicative of differences between the detection values included in said first detection data ( D S1 ) and the prediction values included in said prediction data ( D P ); - updating (108c) said auto-regressive mathematical model ( M R ), if said error function ( E ) takes values exceeding a threshold error value ( E TH ).
  13. Method, according to one of the previous claims, characterised in that it comprises the step (109) of carrying out a second check procedure to check the electric energy consumption predicted by said mathematical model.
  14. Method, according to claim 13, characterised in that said second check procedure (109) comprises: - processing (109a) the calculated prediction data ( D P ) to calculate a prediction function ( P ) indicative of a predicted trend of the electric energy consumption in said electric grid; - generating (109b) an alert signal, if said prediction function ( P ) takes values higher a maximum confidence value ( P max ) or lower than a minimum confidence value ( P min ) .
  15. A computer program, which is stored or storable in a storage medium, characterised in that it comprises software instructions to implement a method (100), according to one or more of the previous claims.
  16. A computerized device characterised in that it comprises data processing resources configured to execute software instructions to implement a method (100), according to one or more of the claims from 1 to 14.
  17. A computerised device, according to claim 16, characterised in that it is an intelligent electronic device (5) for an electric power distribution grid (1).

Description

The present invention relates to the field of electric power distribution grids. More particularly, the present invention relates to a method for predicting electric energy consumption in an electric grid. As is known, the management of electric grids generally requires an accurate prediction activity of the electric energy consumption to allow system operators to properly plan the use of electric energy over time, thereby preventing or limiting demand peaks and making more favourable purchase plans of electric energy. Most common forecast methods are based on machine learning (ML) techniques and require that relevant amounts of data are processed to provide accurate predictions. Additionally, these methods typically provide for carrying out a computationally intensive training phase of an artificial intelligence unit (e.g., a neural network) executing the ML algorithms. ML-based prediction methods can thus be hardly implemented by computing systems typically managing the operation of field devices and switchboards in electric grids, which often have relatively limited storage and computational resources unsuitable to process huge amounts of data. These computing systems are, in fact, commonly based on Edge computing architectures and are basically aimed at bringing computation and data storage closer to the sources of data to improve response times and save bandwidth rather than processing large datasets. In the state of the art, there have been developed prediction methods (e.g., based on linear regression analysis techniques), which normally require lighter computational and data storage resources compared to ML-based forecast methods and which would therefore be adapted for being implemented computing systems commonly used to manage electric grids. An example of these prediction methods is described in US10515308B2. However, available prediction methods of this type often provide relatively poor performances in terms of reliability and prediction accuracy compared to ML-based prediction techniques. The main task of the present invention is to provide to a method for predicting electric energy consumption in an electric grid, which can overcome the limitations of the prior art highlighted above. Within this aim, another purpose of the present invention is to provide a prediction method, which can ensure high level performances in terms of reliability and prediction accuracy. A further aim of the invention is to provide a prediction method, which can be easily implemented even when limited computational and data storage resources are available and which is therefore suitable for being implemented in computing systems commonly used for managing the operation of electric grids, for example in computing systems based on Edge computing architectures. This task and these aims, as well as other aims that will appear evident from the subsequent description and from the attached drawings, are achieved, according to the invention, by a prediction method, according to claim 1 and to the related dependent claims proposed below. In a general definition, the method, according to the invention, comprises the following steps: acquiring first detection data including detection values related to an actual electric energy consumption in said electric grid;acquiring additional detection data including detection values related to the energy consumption in said electric grid during at least a time window preceding a given reference instant;acquiring calendar data including chronological information associated to the operation of said electric grid;calculating training data based on the acquired detection data and calendar data;based on said training data, setting a linear auto-regressive mathematical model describing the trend of the electric energy consumption in said electric grid. Such a linear auto-regressive mathematical model is configured to process at least a set of exogenous input values indicative of at least a periodic function approximating the profile of the electric energy consumption in said electric grid over said at least a time window preceding said reference instant;based on said linear auto-regressive model, calculating prediction data including prediction values related to the electric energy consumption in said electric grid during a time window following said reference instant. Preferably, the method according to the invention comprises the step of acquiring second detection data including detection values related to the energy consumption in said electric grid during a first time window preceding said reference instant. In this case, the linear auto-regressive mathematical model is configured to process first exogenous input values indicative of a first periodic function approximating the profile of the electric energy consumption in said electric grid over said first time window. Preferably, the method according to the invention comprises also the step of acquiring third detection data including detec