Method and device for managing electric energy consumption and/or production

Abstract

A method of managing electric energy consumption and/or production dynamics, includes the steps of: sampling an electric energy flow, adapted to be measured by an electricity metering device (1) within a network (15) of an electrical company; calculating a variation ΔE.sub.i=(E.sub.i−E.sub.i−1)/Δt at regular intervals, where E.sub.i and E.sub.i−1 are two integral sum values of electric energy amounts consolidated over a given number of cycles and Δt represents a time interval between the respective times at which the two values are obtained; adding the variation (ΔE.sub.i) to analogous variations calculated at previous times, to obtain a cumulative sum of such variations; determining whether the variation and/or the cumulative sum exceed a predetermined threshold value (δ.sub.DE); transmitting a message over the network (15) from the electricity metering device (1) if the variation and/or cumulative sum exceed the predetermined threshold value (δ.sub.DE).

Claims

1. A method of managing electric energy consumption and/or production dynamics in an electricity metering device, comprising the steps of: sampling an electric energy flow through a given node of a network of an electric company by an electricity metering device belonging to said network; calculating at said node by a digital metering circuit of said electricity metering device a time-divided energy variation ΔE.sub.i=(E.sub.i−E.sub.i−1)/Δt at regular intervals, where E.sub.i is an integral sum value of electric energy consolidated over a first number of cycles and E.sub.i−1 is an integral sum value of electric energy consolidated over a second number of cycles that occur after the first number of cycles and Δt represents a time interval between the respective times of the first number of cycles and the second number of cycles at which said two integral sum values are obtained; adding said variation (ΔE.sub.i) to analogous variations calculated by said electricity metering device at the same node at previous time instants, to obtain a cumulative sum of such variations; determining by said electricity metering device whether said variation and/or said cumulative sum detected at said node exceed a predetermined threshold value (δ.sub.DE), wherein said metering device is configured to meter the electric energy consumed and/or the electric energy produced by a user of the electric company; and transmitting by a communication unit of said electricity metering device a message over said network if said variation and/or cumulative sum detected at said node exceed said predetermined threshold value (δ.sub.DE).

2. The method as claimed in claim 1, wherein said predetermined threshold value (δ.sub.DE) is a constant value.

3. The method as claimed in claim 1, wherein said predetermined threshold value (δ.sub.DE) is a percentage value of a measured amount of electric energy.

4. The method as claimed in claim 1, wherein said predetermined threshold value (δ.sub.DE) is obtained empirically, through the observation of a collection of previously captured load shapes.

5. The method as claimed in claim 1, wherein said given number of cycles is a value ranging from 5 to 10.

6. The method as claimed in claim 1, wherein said message comprises: an attribute indicating the identity of said electricity metering device; information about the electric energy values that triggered the transmission of said message; a timestamp.

7. The method as claimed in claim 6, wherein said message further comprises one or more message recipients and/or the type of variation that triggered the transmission of said message.

8. The method as claimed in claim 1, wherein said electricity metering device is associated with an electricity user and, if the variations and/or said cumulative sum exceed a maximum electric energy value for a given time, then such metering device cuts off electricity.

9. The method as claimed in claim 8, wherein, prior to said cut-off, said electricity metering device emits a notification, particularly an acoustic or optical notification.

10. The method as claimed in claim 9, wherein said notification is transmitted to a network manager, to prevent said cut-off, by increasing said maximum electric energy value for a period of time.

11. The method as claimed in claim 1, wherein said network comprises a network computer which is adapted to manage said consumption and/or production dynamics, said network computer being able to process the data that come from a network segment or a group of network nodes for balancing purposes.

12. An electricity metering device comprising: an analog module for detecting electric voltage and current at a given node of a network of an electric company; a communication unit for transmitting messages, and a digital metering circuit, which is configured to provide instantaneous electric voltage and current values at said node and to calculate a time-divided energy variation ΔE.sub.i=(E.sub.i−E.sub.i−1)/Δt occurring at said node at regular intervals, where E.sub.i is an integral sum value of electric energy consolidated over a first number of cycles and E.sub.i−1 is an integral sum value of electric energy consolidated over a second number of cycles that occur after the first number of cycles and Δt represents a time interval between the respective time instants of the first number of cycles and the second number of cycles at which said two integral sum values are obtained, and to add said variation (ΔE.sub.i) to analogous variations calculated at the same node at previous time instants, to obtain a cumulative sum of such variations, wherein said communication unit transmits a message over a network if said variation and/or said cumulative sum detected at said node exceed a predetermined threshold value (δ.sub.DE), wherein said metering device is configured to meter electric energy consumed and/or the electric energy produced by a user of the electric company.

13. The device as claimed in claim 12, wherein said analog module comprises a Rogowski coil.

14. The device as claimed in claim 12, wherein said message is transmitted over said network using a power line protocol.

15. A method of managing electric energy consumption and/or production dynamics in an electric energy distribution network, comprising a plurality of metering devices which operate according to the method as claimed in claim 1, and at least one computer for receiving and processing messages transmitted by said plurality of devices, wherein said computer: receives said messages and processes attributes contained therein; counts the number of messages received from said plurality of metering devices within a period of time, as an indicator of event intensity; totals the electric energy variations that took place in said period of time over said plurality of metering devices; calculates the cumulative trend of electric energy variations over said period of time; and if the amount of said received messages exceeds a first threshold (C.sub.MIN) and/or said totaled variations exceed a second threshold (E.sub.MAX) and/or said cumulative trend exceeds said second threshold (E.sub.MAX), it generates a control action on said distribution network.

16. The method as claimed in claim 15, wherein said control action includes one or more of the following actions: network load control, feed-forward control, assessment of network state and/or stability.

17. The method as claimed in claim 15, wherein said control action comprises a predictive calculation to predict the amount of energy for the next period, based on said variations.

18. The method as claimed in claim 15, wherein said computer shares its estimates with the nodes of said network for the purpose of general optimization.

19. The method as claimed in claim 15, wherein said method may be carried out at any node of said network, particularly on a metering device and/or a computer, which detects a variation in a single section of said network at two different times.

20. A method of managing electric energy consumption and/or production dynamics in an electricity metering device, comprising the steps of: sampling an electric energy flow through a given node of a network of an electric company by an electricity metering device belonging to said network; calculating at said node by a digital metering circuit of said electricity metering device a time-divided energy variation ΔE.sub.i=(E.sub.i−E.sub.i−1)/Δt at regular intervals, where E.sub.i is an integral sum value of electric energy consolidated over a first number of cycles and E.sub.i−1 is an integral sum value of electric energy consolidated over a second number of cycles that occur after the first number of cycles and Δt represents a time interval between the respective times of the first number of cycles and the second number of cycles at which said two integral sum values are obtained; adding said variation (ΔE.sub.i) to analogous variations calculated by said electricity metering device at the same node at previous time instants, to obtain a cumulative sum of such variations; determining by said electricity metering device whether said variation and/or said cumulative sum detected at said node exceed a predetermined threshold value (δ.sub.DE), wherein said metering device is configured to meter the electric energy consumed and/or the electric energy produced by a user of the electric company; and transmitting by a communication unit of said electricity metering device a message over said network if said variation and/or cumulative sum detected at said node exceed said predetermined threshold value (δ.sub.DE), wherein said message includes information about the electric energy values of said variation and/or said cumulative sum that triggered the transmission of said message.

21. The method as claimed in claim 20, wherein said message comprises: an attribute indicating the identity of said electricity metering device; and a timestamp.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The above objects will appear more clearly upon reading the detailed description of the method of managing electric energy consumption and/or production and the device therefor, particularly referring to the accompanying Figures, in which:

(2) FIG. 1 shows a block diagram of an electricity metering device according to the invention;

(3) FIG. 2 shows a flow diagram of a method of managing electric energy consumption and/or production dynamics according to the invention;

(4) FIG. 3 shows an exemplary meta-code implementing the method as shown in the flow diagram of FIG. 2;

(5) FIGS. 4a and 4b show two charts, which highlight advantageous aspects of the method of the present invention as compared with the prior art;

(6) FIG. 5 shows a system comprising a plurality of metering devices according to the invention and at least one network computer;

(7) FIG. 6 shows an exemplary meta-code to be used by the network computer of FIG. 5.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(8) Referring to FIG. 1, an electricity metering device 1 is shown, which comprises: a source 3 of an electric energy signal to be measured; a digital measuring circuit 5 comprising an integrated microchip with an onboard microprocessor; a communication unit 7 for transmitting messages to a network 15, preferably operating with a power line technology, as the protocol in use should ensure timely delivery of messages to the recipient; voltage and current sensing means 9; memory means 11; power supply means 19 for the device 1; and a user interface (not shown), such as a LCD screen.

(9) The voltage and current sensing means 9 include an analog sensor or module, which preferably consists of a Rogowski coil, a current transformer, a resistive shunt, a resistive voltage divider or another suitable sensor.

(10) The electricity metering device 1 may further include indicator means 17, namely an acoustic or optical alarm or a display (e.g. of LED, LCD, or other type).

(11) Referring to FIG. 2, a method 100 will be now described for managing electric energy consumption and/or production dynamics according to the invention.

(12) In step 102, parameters are initialized for determining the following quantities, whose meaning will be better explained below: “E.sub.prev”: amount of previously metered energy; “E.sub.i”: amount of currently metered energy; “ΔE.sub.prev”: previously metered variation; “ΔE.sub.i”: currently metered variation; “sum”: sum of minor variations to current time.

(13) Also, in step 102, a constant is selected for a threshold value δ.sub.DE, which sets the amplitude of an admissible variation range, and a minor variation filtering threshold is also selected, for such variations to be deemed as individually negligible.

(14) Various options are allowed for the threshold value δ.sub.DE, which may consist of a fixed constant or a predetermined percentage of the amount of metered energy.

(15) In residential applications, threshold values δ.sub.DE may be 50, 100, and 200 Wh (watthour).

(16) For any particular use of the inventive method, the threshold value δ.sub.DE may be selected either empirically, through the observation of a collection of previously captured load shapes, or through the use of advanced optimization techniques, such as “soft computing”.

(17) The purpose of the threshold value δ.sub.DE is to set a compromise between the sensitivity of the instrument and the ability of compressing the data volume to be transmitted.

(18) A proper threshold value δ.sub.DE will be selected by maintaining any nomothetic (i.e. repetitive) cycle, and by providing improved filtering of stochastic “noise” events, such as turning on of low-power-consumption lamps or the like, which events are negligible for the change of the energy system state.

(19) Assuming for example a house having lamps (absorbing 7, 11, 15 and 18 Wh each), TV sets and other similar small appliances, and assuming the predominance of energy-consuming appliances, such as an air conditioning system (absorbing 2 kWh), an iron (absorbing 2 kWh), a washing machine (absorbing 2 kWh for heating and 300 Wh for its motors), a dishwasher (absorbing 2 kWh for heating and 120 Wh for its motors and pumps), a refrigerator (absorbing 120 Wh) and similar heating, ventilation, and air conditioning devices, a proper threshold δ.sub.DE would be higher than the consumption of each lamp and lower than the power absorbed by the motors of the washing machine, dishwasher and refrigerator, and hence, for instance 99 Wh.

(20) This threshold value δ.sub.DE may be also dynamically set using intelligent self-learning techniques, as described in the art.

(21) In step 104, the value of the meter i is increased by 1.

(22) In step 106, the electricity metering device 1 measures and totals “k” instantaneous electric energy flow values, and particularly instantaneous voltage values V.sub.i and instantaneous current values I.sub.i. In the same step, instantaneous electricity values E.sub.i=V.sub.i.Math.I.sub.i.Math.Δt, are also calculated, where Δt=t.sub.i−t.sub.i−1, i.e. a time interval between two successive times.

(23) The value Ei can be calculated with a discrete numerical method, i.e. with

(24) $E_i=\underset{i=0}{\overset{k}{.Math.}}\left[V\left(t_i\right).Math.I\left(t_i\right).Math.\Delta t\right],$
where the values E.sub.i are consolidated every k cycles (50 Hz alternate current in Europe or 60 Hz in certain other countries), and in which the k value preferably ranges from 5 to 10.

(25) Unlike PMUs, whose operation is based on phasors, the device of the present invention operates with scalar values, and is required to total sums by integral calculus.

(26) Since integral sums are periodic functions with period ω, they shall be mutually comparable. For example, this may occur by monitoring zero-crossing times, or by extending the duration of multiple integration of the period ω.

(27) Nevertheless, the k value shall maintain compatibility with the system control function, allowing the user to respond before experiencing an adverse effect, such as under-frequency.

(28) The less computationally expensive option is selected, i.e. the integral sum in k cycles, where k preferably ranges from 5 to 10.

(29) The integration interval is long enough to allow comparison of the sums and allows timely response by a control system.

(30) A second consideration to be made about the selection of the k parameter is associated with exchanged data volumes and controllability. If a k value=10 is selected, the maximum frequency of DE messages in Europe will be 0.2 seconds (a k value=50 would increase the maximum frequency to 1 second and so on).

(31) The “event-driven” method of the present invention is thus expected to provide exchanged data volumes similar to state-of-the-art “time-driven” methods, with longer time intervals (e.g. 15-minute intervals in current metering).

(32) At step 108 a variation ΔE.sub.i is calculated, with ΔE.sub.i=(E.sub.i−E.sub.prev)/Δt, where E.sub.prev is instantaneous energy value calculated at time t−1, or at the previous cycle.

(33) In step 110 the parameter “sum” is increased by the value ΔE.sub.i as determined in step 108, to obtain a cumulative sum of values.

(34) In step 112 a check is made to see whether the absolute value of an individual variation ΔE.sub.i is higher than the predetermined threshold δ.sub.DE (fast trend). If it is, the flow diagram moves to step 116, if it is not, the flow diagram continues with step 114.

(35) In step 114 a check is made to see whether the absolute value of the parameter “sum”, i.e. the sum of minor variations, is higher than the predetermined threshold δ.sub.DE (slow trend). If it is not, the flow diagram moves to step 104. If it is, the process moves to step 116.

(36) In step 116, a message is sent over the network 15 by the electricity metering device 1. If the variation as described in steps 112 and 114 is higher than the threshold δ.sub.DE, this will be indicative of dynamics that might affect the steady state of the local power system comprising the device 1.

(37) The format of the electronic message may change according to actual implementations. Nevertheless, the message shall contain at least: an attribute indicating the identity of a remote node of the network 15, e.g. the ID of the device 1; a timestamp for the measuring time; information about the electric energy values that triggered message transmission, such as one of the pairs of attributes: (E.sub.prev, E.sub.i); (E.sub.prev, ΔE.sub.i); (ΔE.sub.i, E.sub.i).

(38) The “push” mode, i.e. the transmission of data from the device 1 to the network 15 advantageously allows control of the system comprising the device 1. Due to message timestamps, messages may be sorted by time and be processed by a provider's server in “stateless” mode. For this purpose, two instantaneous electric energy values are provided in the message, i.e. previous and current values, which are related by the formula E.sub.i=E.sub.prev+ΔE.sub.i, and allow determination of the third value of the formula.

(39) The message may further contain: an attribute indicating the message recipient (one recipient for unicast, a group for groupcast, or all recipients for broadcast); a variation attribute indicating the variation type that triggered the message; if the message was triggered by the test of step 112, then the variation attribute is “fast”; if the message was triggered by the test of step 114, then the variation attribute is “slow”.

(40) The information in the message is appropriately examined by the network manager 15, or by other nodes of the network operated by the network manager 15, as being useful for control purposes (SCADA) or else.

(41) Once the message has been transmitted, in step 118, the parameters “sum” and the meter “i” are set to zero, whereas the parameter “ΔE.sub.prev” is set to “ΔE.sub.i” and the parameter “E.sub.prev” is set to the last value “E.sub.i”.

(42) One exemplary meta-code implementing the method 100 is shown in FIG. 3.

(43) Certain contracts restrict maximum energy consumption by a user, e.g. to 3 kWh. Some tolerance is typically allowed as this limit is exceeded, which tolerance is also expressly set forth in contract documents. For example, the maximum energy limit may be allowed to be exceeded by not more than 30% for 5 minutes, or by more than 30% for not more than 10 successive times. In these cases, the user should be acoustically warned about the occurrence of these conditions, to avoid cut-off from the network.

(44) If the electricity metering device 1 is associated with a user subjected to contract limitation of maximum consumed power (e.g. 3 kWh), and if the variations ΔE.sub.i or the parameter “sum” are equal to or higher than the maximum allowed values for a time e.g. equal to half the time after which the electric company actually cuts off electricity (here 2.5 minutes), the electricity metering device 1 acoustically warns the user about the forthcoming risk of electricity cut-off, thereby leaving enough time (here 2.5 minutes) to reduce consumption.

(45) Upon occurrence of such condition, the electricity metering device 1 may emit an acoustic and/or optical alarm by the alarm means 17, to notify the user of a possible forthcoming electricity cut-off.

(46) Such notification may be transmitted to the network manager, to possibly prevent such cut-off, e.g. by increasing the maximum electric energy value allowed to the user for a period of time (dynamic threshold).

(47) Referring to FIGS. 4a and 4b, two comparative charts are shown. The top chart of FIG. 4a describes the behavior of a prior art electricity metering device, in which the metering device transmits one sample of measured values per second (3600 samples per hour) over the network, because such samples are measured with a “time-driven” logic, i.e. at predetermined time intervals; the bottom chart of FIG. 4b describes an electricity metering device 1 implementing the method of the present invention, in which the samples transmitted in the same period of time are dramatically decreased to 77 (in 4 hours and a half), because the method of the present invention uses an “event-driven” logic, which means that the values are only transmitted upon occurrence of given events, i.e. when the measured values fall out of the admitted range δ.sub.DE.

(48) A network computer, or server, receives and processes dynamics from the networked devices 1. The server can process the data that come from a network segment or a group of network nodes for balancing or other business purposes.

(49) Referring to FIG. 5a, the server 20 in the network continuously listens to the messages from the metering devices 1 which calculate variations at the remote nodes, as the latter use the push mode. Unlike PDC servers and other servers that operate with instantaneous datasets, the server 20 accomplishes various artificial intelligence tasks, and processes historical data series, past and present dynamics. As it continuously listens to infinity, it divides time into time frames, i.e. blocks of a given duration (above the k value of the devices 1), e.g. from 100 to 200 ms. At the start of the time frame, its counters are set to zero. The server counts the messages it receives, n=Count(DE.sub.i(t), and analyzes them.

(50) Each message it receives increases the total number of variations: by definition, this amount of dynamics indicates the intensity of the variation flow. The intensity of the variation flow Count(DE.sub.i(t)) is considered, as it is deemed to be significant for network stability. At the end of the time-frame the overall energy unbalance of the network topology is assessed, considering that certain positive variations may be nullified by negative variations from different nodes.

(51) In addition to the counter for time frame variations, a total counter is used, which stores cumulative variations, as well as the trend. Cumulative values and variation intensities are useful for the purpose of network control, as a great number of events and/or their relevance may affect stability and/or quality of energy.

(52) A fuzzy logic predicate (FIG. 6, right column), which determines whether there are too many events (IF C>=C.sub.min) or whether there are excessively abrupt variations, or whether there is an accumulating trend (IF Ep>=Emax) triggers anomalies, thereby providing a control function.

(53) Any intelligent control action based on these counters and/or any interaction with SCADA control systems are as provided in prior art load management algorithms, or in feed-forward control, where possible.

(54) The control action may also include one or more of the following actions: network load limitation, assessment of states and/or state transitions and/or network stability, prediction of energy amount for the next period based on variations, addition of generators.

(55) Advantageously, the control action may be carried out at any node of the network, particularly on a metering device (1) and/or a computer (20), which detects a variation in a single network section at two different times.

(56) The computer 20 may share its estimates with the nodes of the distribution network for general optimization of the nodes.

(57) An example of the natural phenomenon of variable cloudiness, which affects photovoltaic energy generation is addressed in M. Simonov, F. Grimaccia, M. Mussetta, R. Zich, Modello di gestione dell'energia solare in tempo reale, in Journal AEIT, 1 (2011), pp. 24-29, Milano, Italy.

(58) The flow of events, which have been only used heretofore as independent events, is used to calculate a synergy between nodes, deriving from mutual knowledge of the times at which production losses occur. The observation of propagation features, as well as the phenomenon-describing rules, provide control criteria: for example, if a loss occurs in node A at time t.sub.1, and a loss occurs in node B at time t.sub.2, then a loss will also occur in node C at time t.sub.3, and hence the control action may consist in turning on a generator D before time t.sub.3.

(59) The invention provides a flow of events from one or more devices 1, that normally occur during actual electricity use. The flow of actual use events DE introduces new semantic knowledge into the digital world, allowing for novel decision-making processes in the so-called smart grids.

(60) Referring now to FIG. 6, an exemplary meta-code implemented in the server 20 of FIG. 5 is shown.

(61) The above disclosure clearly shows the characteristics of the present invention, as well as its advantages.

(62) A first advantage of the present invention is that data exchange between an electricity metering device and an electricity manager that manages such device is dramatically reduced, because data are only transmitted upon occurrence of an event, i.e. when a typically predetermined threshold (e.g. selected by the network manager, the electrical company or the user of the device) is exceeded.

(63) While a prior art meter provides a time-driven load shape of n objects {(E.sub.1,t.sub.1), . . . , (E.sub.i, t.sub.i), . . . , (E.sub.n, t.sub.n)}, in which |t.sub.i−t.sub.i−1|=constant, the device according to the invention provides a series with a smaller number of terms {(ΔE.sub.1,t.sub.1, . . . ), . . . , (ΔE.sub.i,t.sub.i, . . . ), . . . }, because if no variation occur during certain periods [t.sub.k;t.sub.k+Δt.sub.k], [t.sub.l;t.sub.l+Δt.sub.l], [t.sub.m;t.sub.m+Δt.sub.m] and so on, then their values, and all the individual components in the formula, remain unchanged. Therefore, less data are provided as compared with a “time-driven” series.

(64) A combined observation method, comprising a prior art time-driven metering component {. . . , (E.sub.i, t.sub.i), . . . } at less frequent intervals |t.sub.i−t.sub.i−1|=10-20 minutes, and possibly with variable sampling times, i.e. not with regular and constant time intervals, and a novel event-driven real-time continuous observation component, which interposes additional elements (ΔE.sub.k,t.sub.k,[E.sub.prev], . . . ) is highly advantageous, as it involves a minor increase of data traffic volumes, as compared with the prior art.

(65) Experimental comparison of time-driven methods at 15 minutes and 0.1 seconds, i.e. with 96 and 864,000 daily data elements respectively, with the above mentioned combined method, shows that the latter adds a few hundreds of significant power system dynamics, corresponding to as many actual events, thereby avoiding the prior art large data volumes.

(66) A second advantage of the present invention is that the device according to the present invention affords management of both “consumer” power system dynamics, and “producer” dynamics, both measures being taken at the same time (“prosumer”). Any flow reversal will only cause a change of sign of the measured scalar value. Therefore, the system can stay close to a balance condition defined by the range −δ.sub.DE<ΔE.sub.i<+δ.sub.DE, and conveniently manage a combined user type (“prosumer”).

(67) A third advantage of the present invention is anticipation of the expected value. Adjacent energy flow variations caused by an external agent are always distant in time, and hence a variation ΔE.sub.i qualifies a past period, and also predicts the next time frame E.sub.i+1=E.sub.i in the event that no further variations exist.

(68) A further advantage of the present invention is that the volume of significant data transmitted over the network is dramatically reduced. Due to the maintenance of the previous values when no variation occurs, there is no need to repeat the same values, like in all data compression algorithms. Any redundancy would be reflected in the additional attribute E.sub.next, as well as the variation ΔE.sub.i itself, in the data package.

(69) The method and device of the present invention may be used by: a manager of a network for distribution of any level of electric energy (low voltage, medium voltage or high voltage); any local sub-topology; owners of renewable energy sites; various participants in the electricity value chain.

(70) A use of the method of the present invention involves the employment of a plurality of networked devices, interoperable by DE message broadcasting, to provide real-time proactive load balancing.

(71) Instead of measuring energy quality based on the occurrence of under-frequency, i.e. a frequency of less than 50 or 60 Hz/rated values, which is a consequence of excessive electricity consumption and not a cause, the invention uses a central computer (the equivalent of the PDC of prior art phasor measurement by PMUs) to measure the individual dynamics of the electricity metering devices and their superpositions (sums) at any level.

(72) Since the method of the present invention can capture events susceptible of causing network imbalance (i.e. their causes), such capture occurs before their consequences (i.e. under-frequency).

(73) Typically, the power flow is directed from high and medium voltage segments to low voltage segments in the electric network. In case of distributed electricity generation, flows might be reversed. This might be detrimental for users.

(74) Calculation of power system dynamics with the method of the present invention provides computationally easier acknowledgement of any unbalance between directional flows anytime. The system remains in its steady state until dynamics-altering events occur. Dynamics are detected by the method of the present invention. Therefore, the system has real-time knowledge of such dynamics and can calculate flow reversals and take the required actions.

(75) The method of the present invention is particularly useful for energy brokers operating in the free market, as it allows real-time observation of electric energy consumption/production dynamics, and can capture the causes of such dynamics, instead of analyzing their consequences. Forward energy demand calculations and optimization of purchases over the liberalized market can advantageously provide considerable savings.

(76) Also, electric network segments may be dynamically partitioned to obtain a desired energy behavior, e.g. by avoiding electricity injection into the backbones (total local consumption).

(77) The method of the present invention may be further used by a panel of renewable-source energy producers (producing energy from solar or other similar sources), operating in autonomous, independent manners. The method of the present invention advantageously allows detection of events such as production losses (e.g. due to a weather factor, such as a cloud in case of photovoltaic energy generation, or any other external agent).

(78) Instead of calculating a separate probability forecast for each network node, the method of the present invention allows utilization of the retrieved knowledge for global optimization. Therefore, the first node affected by a detrimental phenomenon may transmit a message over the network to warn the neighboring nodes. This event is received by other network partners and may be used to anticipate the decisions and/or control required to address the forthcoming loss also by said partners, due to the displacement of the phenomenon.

(79) In the prior art, load balance is based on probability forecasts. The method of the present invention increases prediction accuracy, due to non-probability components. Improved load prediction can reduce imbalance.

(80) For example, electricity production from wind and solar sources in Italy, in 2009, was 66 TWh, amounting to more than 660 million euros (10 eurocents/kWh). The estimated 2% imbalance may be further reduced. The method of the present invention, assuming a 50% contribution (as random components are replaced by certain terms), will provide savings for 0.13 TWh/year, i.e. about 6.6 million euros/year.

(81) Many variants may be envisaged for the method of managing electric energy consumption and/or production dynamics and the device therefor, as disclosed herein as an example, without departure from the novel concepts of the invention, and the disclosed details may be implemented in different manners, or replaced by technically equivalent parts.

(82) Therefore, it will be appreciated that the present invention is not limited to a method of managing electric energy consumption and/or production dynamics and a device therefor, but is susceptible to various modifications, improvements, replacements of equivalent elements and parts, without departure from the scope of the invention, as better defined in the following claims.