Method for controlling power ramps with prediction in intermittent power generation plants

Abstract

The present invention relates to a method for controlling power ramps with prediction in intermittent power generation plants, such as, for example, a photovoltaic solar plant, which minimizes the storage capacity required for compliance with the maximum ramp requirements for power fluctuation as well as the cycling of said storage systems, thus extending its lifespan and also reducing associated energy losses, thus reducing investment costs in the plant, such that, in order to achieve the same maximum fluctuation ramp, a minor use is made of the energy storage system.

Claims

1. A method for controlling power ramps with prediction in intermittent power generation plants, which minimizes energy storage requirements in intermittent power generation plants guaranteeing at all times compliance with a maximum admissible ramp value for the power being fed into a grid, P.sub.G(t), given by a grid code regulation, or other instrument, wherein the method comprises: a stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), in a prediction time horizon, the maximum plant power being, P.sub.Max(t), and the minimum plant power being, P.sub.Min(t), the maximum and minimum power, respectively, being those that can be produced in the intermittent power generation time during that prediction time horizon, wherein in each t moment the energy required to achieve the maximum plant power, P.sub.Max(t), and the minimum power of the plant, P.sub.Min(t), is also calculated in said prediction time horizon; a stage for determining the maximum possible fluctuation that can occur in the instantaneous power generated by the intermittent power generation plant, P(t), wherein the maximum possible fluctuation is one of the two following fluctuations: a positive fluctuation, calculated between the instantaneous power, P(t), and the maximum plant power, P.sub.Max(t), and a negative fluctuation calculated between the instantaneous power, P(t), and the minimum plant power, P.sub.Min(t); a stage for dynamically calculating the state of charge of an energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t); and a stage for determining a stabilized state of charge, SOC.sub.sta(t), of the storage system, wherein the stabilized state of charge, SOC.sub.sta(t), is calculated as the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t), plus a term that takes into account the energy required by the power being fed into the grid, P.sub.G(t), to reach the instantaneous power, P(t), generated by the intermittent power generation plant.

2. The method of claim 1, further comprising a control stage wherein the state of charge, SOC(t), of the energy storage system associated with the difference between the power being fed into the grid, P.sub.G(t), and the instantaneous power generated by the intermittent power generation plant, P(t), is modified according to the stabilized state of charge, SOC.sub.sta(t), if the previous value of the stabilized state of charge SOC.sub.sta(t−1), is not able to support the maximum possible fluctuation, or is not modified if the previous value of the stabilized state of charge, SOC.sub.sta(t−1), is capable of supporting the maximum possible fluctuation.

3. The method of claim 2, further comprising a stage for adjusting a dynamic component of the ramp defined as the rate at which the power in the plant must be modified in the control stage, according to the state of charge SOC(t) of the energy storage system.

4. The method of claim 2 further comprising an additional stage between the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during a prediction time horizon, and the stage for determining the maximum possible fluctuation that can occur in the instantaneous power generated by the intermittent power generation plant, P(t), wherein said additional stage is a stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon.

5. The method of claim 4 wherein the maximum plant power, P.sub.Max(t) and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of irradiance, G(t), and measured temperatures of the cell (T.sub.c), or optionally, with a non-parametric method, in which the intermittent power generation plant is a photovoltaic plant.

6. The method of claim 4 wherein the maximum plant power, P.sub.Max(t), and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of the following meteorological variables: wind speed, {right arrow over (ν)}w(t), and temperature, T(t), or optionally with a non-parametric method, in which the intermittent power generation plant is a wind turbine or a wind farm.

7. The method of claim 2 wherein the maximum plant power, P.sub.Max(t) and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of irradiance, G(t), and measured temperatures of the cell (T.sub.c), or optionally, with a non-parametric method, in which the intermittent power generation plant is a photovoltaic plant.

8. The method of claim 2 wherein the maximum plant power, P.sub.Max(t), and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of the following meteorological variables: wind speed, {right arrow over (ν)}w(t), and temperature, T(t), or optionally with a non-parametric method, in which the intermittent power generation plant is a wind turbine or a wind farm.

9. The method of claim 2, wherein the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), is performed for each of a group of plants in order to calculate the maximum power, P.sub.Max(t), and the minimum power, P.sub.Min(t), of the group of plants.

10. The method of claim 1, further comprising an additional stage between the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during a prediction time horizon, and the stage for determining the maximum possible fluctuation that can occur in the instantaneous power generated by the intermittent power generation plant, P(t), wherein said additional stage is a stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon.

11. The method of claim 10 wherein the maximum plant power, P.sub.Max(t) and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of irradiance, G(t), and measured temperatures of the cell (T.sub.c), or optionally, with a non-parametric method, in which the intermittent power generation plant is a photovoltaic plant.

12. The method of claim 11, wherein the stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, corrects the minimum plant power, P.sub.Min(t), at a moment of time around sunrise and corrects the maximum plant power, P.sub.Max(t), at a moment of time around sunset.

13. The method of claim 10 wherein the maximum plant power, P.sub.Max(t), and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of the following meteorological variables: wind speed, {right arrow over (ν)}w(t), and temperature, T(t), or optionally with a non-parametric method, in which the intermittent power generation plant is a wind turbine or a wind farm.

14. The method of claim 10, wherein the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), is performed for each of a group of plants in order to calculate the maximum power, P.sub.Max(t), and the minimum power, P.sub.Min(t), of the group of plants.

15. The method of claim 1, wherein the maximum plant power, P.sub.Max(t) and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of irradiance, G(t), and measured temperatures of the cell (T.sub.c), or optionally, with a non-parametric method, in which the intermittent power generation plant is a photovoltaic plant.

16. The method of claim 1, wherein the maximum plant power, P.sub.Max(t), and/or the minimum plant power, P.sub.Min(t), are calculated through specific values of the following meteorological variables: wind speed, {right arrow over (ν)}w(t), and temperature, T(t), or optionally with a non-parametric method, in which the intermittent power generation plant is a wind turbine or a wind farm.

17. The method of claim 1, wherein the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), is performed for each of a group of plants in order to calculate the maximum power, P.sub.Max(t), and the minimum power, P.sub.Min(t), of the group of plants.

18. The method of claim 1, further comprising an additional stage for calculating an error committed in the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, with respect to the instantaneous power values actually generated by the intermittent power generation plant, P(t).

19. The method of claim 18, wherein: if the calculated error exceeds a predetermined value, the method comprises as stage for controlling the state of charge, SOC(t), wherein the state of charge of the energy storage system required to support the maximum possible fluctuation, SOC.sub.mpf(t), is a target state of charge, SOC.sub.tar(t), and the method further comprises a control stage, wherein the state of charge, SOC(t), of the energy storage system associated with the difference between the power being fed into the grid, P.sub.G(t), and the instantaneous power generated by the intermittent power generation plant, P(t), is modified according to the target state of charge, SOC.sub.tar(t), if the previous value of the target state of charge, SOC.sub.tar(t−1), is not capable of supporting the maximum possible fluctuation, or is not modified if the previous target state of charge, SOC.sub.tar(t−1), is capable of supporting the maximum possible fluctuation; while if the calculated error in the stage for calculating error does not exceed the predetermined value, the method comprises a control stage wherein the state of charge, SOC(t), of the energy storage system associated with the difference between the power being fed into the grid, P.sub.G(t), and the instantaneous power generated by the intermittent power generation plant, P(t), is modified according to the stabilized state of charge, SOC.sub.sta(t), if the previous value of the stabilized state of charge SOC.sub.sta(t−1), is not able to support the maximum possible fluctuation, or is not modified if the previous value of the stabilized state of charge, SOC.sub.sta(t−1), is capable of supporting the maximum possible fluctuation.

20. The method of claim 19, wherein the predetermined value is in the range of 60%.

21. The method of claim 1, wherein in the stage for determining a stabilized state of charge, SOC.sub.sta(t), of the storage system, the calculation of the stabilized state of charge, SOC.sub.sta(t) as the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t), plus a term that takes into account the energy required by the power being fed into the grid, P.sub.G(t), is calculated when a fluctuation occurs.

22. The method of claim 1, wherein in the stage for determining a stabilized state of charge, SOC.sub.sta(t), of the storage system, the calculation of the stabilized state of charge, SOC.sub.sta(t) as the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t), plus a term that takes into account the energy required by the power being fed into the grid, P.sub.G(t), is calculated while the stabilized state of charge, SOC.sub.sta(t), is always maintained above 0%.

Description

DESCRIPTION OF THE DRAWINGS

(1) To implement the present description and for the purpose of providing a better understanding of the characteristics of the invention, according to a preferred embodiment thereof, a set of drawings is attached as an integral part of said description, which by way of illustration and not limitation represent the following:

(2) FIG. 1 show a graph of the stage in which the maximum P.sub.Max(t) and the minimum P.sub.Min(t) production limits are instantaneously calculated, which are set to occur during the prediction time horizon (H).

(3) FIG. 2 shows a graph depicting the stage for dynamically calculating the state of charge of the energy storage system required for supporting the maximum possible fluctuation SOC.sub.mpf(t) according to the method presented in patent application EP3026774A1 (without prediction). It shows how the SOC becomes negative, which implies that the BESS is completely discharged, the ramp limitation would be lost and therefore the grid code would not be complied with.

(4) FIG. 3 shows a graph depicting the stage for dynamically calculating the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t) according to the method presented in International application WO02016/055658A1, performing short-term prediction with a prediction time horizon (H) of 45 minutes. It shows how the SOC becomes negative, which implies that the BESS is completely discharged, the ability to limit ramps would be lost and therefore the grid code would not be complied with.

(5) FIG. 4 shows a graph depicting the stage for determining the stabilized state of charge, SOC.sub.sta(t), according to the present invention, of the storage system performing short-term prediction, with a prediction time horizon (H) of 45 minutes. The values shown in FIG. 3 are represented by dashed lines. It shows how the BESS does not discharge now and therefore the grid code is complied with.

(6) FIG. 5 shows a graph depicting the cycling degradation of the methods found in Patent application EP3026774A1 (without prediction, H equals zero) and those of the International application WO2016/055658A1 (without prediction, H equals zero, and between 15 and 180 min). In all the situations, the total discharge of the battery takes place in the same way as shown in FIG. 3 corresponding to a prediction time horizon (H) of 45 minutes and for prediction time horizons (H) of between 15 and 180 minutes, applying said horizon to the method found in International application WO2016/055658A1.

(7) FIG. 6 shows a graph depicting cycling degradation for the graph depicted in FIG. 4 corresponding to a prediction time horizon (H) of 45 minutes and for prediction time horizons (H) of between 15 and 180 minutes, by applying the method of the present invention.

(8) FIG. 7 shows a graph depicting certain inconsistencies presented by the method of the present invention when the day is clear, wherein peeks are observed in the calculation of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, and which cause excessive cycling.

(9) FIG. 8 shows a graph after applying the stage for correcting the maximum plant power, P.sub.Max(t), y and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, in the method of the present invention.

(10) FIG. 9 shows a graph depicting the cycling degradation shown in FIG. 6 after applying the additional stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon of the present invention. It can therefore be observed that in no case is the BESS completely discharged and the cycling degradation is inferior to the methods of the state of the art.

PREFERRED EMBODIMENT OF THE INVENTION

(11) A detailed explanation of a preferred embodiment of the invention is described hereinafter according to FIGS. 1 to 9 mentioned above.

(12) The method for controlling power ramps with prediction in intermittent power generation plants that minimizes energy storage requirements in intermittent power generation plants, guaranteeing at all times compliance with a maximum admissible ramp value for the power being fed into the grid, PG(t), given by a GRID CODE regulation, or another instrument, which also improves the lifespan and losses of the storage system, which can be applied to, for example, a photovoltaic solar plant, and wherein the energy storage system is, for example, a battery, comprises: a stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), in a prediction time horizon, the maximum plant power being, P.sub.Max(t), and the minimum plant power being, P.sub.Min(t), the maximum and minimum power, respectively, being those that can be produced in the intermittent power generation time during that prediction time horizon; a stage for determining the maximum possible fluctuation that can occur in the instantaneous power generated by the intermittent power generation plant P(t), wherein the maximum possible fluctuation is one of the two following fluctuations: a positive fluctuation which is calculated between the instantaneous power P(t) and the maximum plant power, P.sub.Max(t), and a negative fluctuation which is calculated between the instantaneous power, P(t), and the minimum plant power, P.sub.Min(t); a stage for dynamically calculating the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t); and a stage for determining a stabilized state of charge, SOC.sub.sta(t), of the storage system, wherein the stabilized state of charge, SOC.sub.sta(t), is calculated as the state of charge of the energy storage system required to support the maximum possible fluctuation, SOC.sub.mpf(t), plus a term that takes into account the energy required by the power being fed into the grid, P.sub.G(t), in order to achieve the instantaneous power, P(t), generated by the intermittent power generation plant;

(13) Preferably, the method further comprises a control stage wherein the state of charge, SOC(t), of the energy storage system associated with the difference between the power being fed into the grid, P.sub.G(t), and the instantaneous power generated by the intermittent power generation plant, P(t), is modified according to the stabilized state of charge, SOC.sub.sta(t), if the previous value of the stabilized state of charge, SOC.sub.sta(t−1), is not able to support the maximum possible fluctuation, or is not modified if the previous value of the stabilized state of charge, SOC.sub.sta(t−1), is capable of supporting the maximum possible fluctuation.

(14) The method for controlling power ramps of the present invention, in addition to working with the minimum necessary storage, can reduce cycling by more than 50% with respect to the state of the art and thus double the lifespan of the storage system. The method is based on a control of the state of charge SOC(t) of the battery based on preferably short-term production prediction data, and more preferably on data of between 15 and 180 minutes. Particularly, in the first stage of the method, the maximum production limits P.sub.Max(t) and the minimum production limits P.sub.Min(t) that are going to take place within the prediction time horizon, are calculated instantaneously.

(15) FIG. 1 shows a representation associated to said stage wherein the maximum production limits P.sub.Max(t) and the minimum production limits P.sub.Min(t) that are going to take place within the prediction time horizon (H), which in said FIG. 1 is 30 minutes, are calculated instantaneously. This prediction time horizon (H) shifts with time such that for each moment of time t, maximum production limits P.sub.Max(t) and minimum production limits P.sub.Min(t) are obtained.

(16) The method then determines the maximum possible fluctuation that can occur in the instantaneous power generated by the intermittent power generation plant, P(t), in said prediction time horizon, wherein the maximum possible fluctuation is one of the two following fluctuations: a positive fluctuation, calculated between the instantaneous power P(t) and the maximum plant power, P.sub.Max(t), and a negative fluctuation, calculated between the instantaneous power P(t) and the minimum plant power, P.sub.Min(t).

(17) Next, the method dynamically calculates the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t).

(18) FIG. 2 shows a graph depicting the stage for dynamically calculating the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t). Said FIG. 2 represents the value of the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t). Up to this stage, the method of the present invention would be that of the method found in patent application EP3026774A1.

(19) FIG. 3 shows a graph depicting the stage for dynamically calculating the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t). Said FIG. 3 represents the value of the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t), by carrying out a short-term prediction with a prediction time horizon (H) of 45 minutes. Up to this stage, the method of the present invention would be the result of applying short-term prediction with a prediction time horizon to the method found in International application WO2016/055658A1.

(20) FIGS. 2 and 3 show how the battery discharges completely, since there is an area wherein the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t) is below 0%. This implies that the capacity for controlling ramps is lost and therefore the grid code would not be complied with.

(21) The method of the present invention solves the problem associated with the discharge of the battery by means of a stage for determining a stabilized state of charge, SOC.sub.sta(t) of the storage system, wherein the stabilized state of charge, SOC.sub.sta(t), is calculated as the state of charge of the energy storage system required to support the maximum possible fluctuation SOC.sub.mpf(t), plus a term that takes into account the energy required by the power being fed into the grid, P.sub.G(t), in order to achieve the instantaneous power, P(t), generated by the intermittent power generation plant. Similarly, the battery is never fully charged in order to avoid loss of energy that is not fed into the grid.

(22) FIG. 4 shows a graph depicting the stage for determining the stabilized state of charge, SOC.sub.sta(t), of the storage system. Said FIG. 4 represents the value of the stabilized state of charge, SOC.sub.sta(t), of the storage system by carrying out a short-term prediction with a prediction time horizon (H) of 45 minutes. It shows that the battery never fully discharges, since there is no area wherein the stabilized state of charge, SOC.sub.sta(t), is below 0%. Therefore ramp control is complied with and thereby so is the grid code.

(23) FIG. 5 shows a graph depicting cycling degradation in the methods found in patent application EP3026774A1 (without prediction, H equals zero) and International application WO02016/055658A1 (without prediction, H equals zero, and between 15 and 180 min). Said FIG. 5 shows how the battery discharges completely for all the prediction time horizons. This implies that ramp control is lost and therefore the grid code would not be complied with.

(24) FIG. 6 shows a graph depicting cycling degradation for the graph in FIG. 4 for prediction time horizons (H) of between 15 and 180 minutes, by applying the method of the present invention. Said FIG. 6 shows how the battery is not completely discharged for any of the prediction time horizons. Therefore, ramp control and the associated grid code would be complied with.

(25) Said FIG. 6 shows that by applying the present invention with a battery, said battery presents high cycling degradation, which increases as the prediction time horizon increases (H). This is because the method of the present invention has certain inconsistencies when it is a clear day, as shown in FIG. 7, where peaks are observed in the calculation of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, specifically, a peak at dawn, shown on the left side of the graph, and a peak at dusk, shown on the right side of the graph. In order to solve this high cycling degradation, which does not affect the state of charge of the battery, since at no moment is said battery completely discharged, the method of the present invention comprises an additional stage between the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant P(t), during a prediction time horizon, and the stage for determining the maximum possible fluctuation that can occur in the instantaneous power generated by the intermittent power generation plant, P(t), wherein said additional stage is a stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant P(t), during the prediction time horizon.

(26) Preferably, the stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant P(t), during the prediction time horizon, corrects the minimum plant power P.sub.Min(t) at a moment of time around sunrise and corrects the maximum plant power, P.sub.Max(t) at a moment of time around sunset.

(27) In the case of the correction of the minimum plant power, P.sub.Min(t), at the moment of time around sunrise, said correction must be carried out to avoid excessive cycling degradation of the battery, since it makes no sense that at any moment of the day, and at that particular moment during sunrise, the minimum plant power, P.sub.Min(t), exceeds the estimated power for a completely clear and cloudless day.

(28) In the case of the correction of the maximum plant power, P.sub.Max(t), at the moment of time around sunset, said correction must be carried out to avoid excessive cycling degradation of the battery, since it makes no sense that at any moment of the day, and at that particular moment during sunset, the difference between the value of the estimated power for a completely clear and cloudless day and the value of the maximum plant power, P.sub.Max(t), exceeds the difference between the actual instantaneous power and the estimated power for a cloudy day.

(29) By implementing said stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, into the method of the present invention, both the peak at sunrise and the peak at sunset are corrected, of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, as shown in FIG. 8.

(30) Thus, the method of the present invention with the additional stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant P(t), during the prediction time horizon, reduces cycling degradation of the battery, as shown in FIG. 9.

(31) The following table shows the cycling degradation of the battery between the method of the present invention, by applying the additional stage for correcting the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, for different prediction time horizons, and the methods of European patent EP3026774A1 and International application WO02016/055658A1, wherein it shows that cycling degradation is improved for any prediction time horizon used in the present invention with respect to the methods used in European patent EP3026774A1 and International patent WO2016/055658A1.

(32) TABLE-US-00002 WO2016/055658A1 Invention H = H = H = H = MERIT 15 180 15 180 INDICES EP3026774A1 min min min min Capacity 0.32 0.32 . . . 0.32 0.32 . . . 0.32 C.sub.BAT (h) Cycling 2.56 1.1  . . . 5.26 1.08 . . . 1.73 degrada- tion (annual %)

(33) In other words, at best, the battery would last approximately twice as long in terms of cycling when using the method of the present invention with respect to the methods found in European patent EP3026774A1 and International application WO2016/055658A1.

(34) Optionally, the method further comprises an additional stage for calculating the error committed in the stage for calculating the maximum plant power, P.sub.Max(t), and the minimum plant power, P.sub.Min(t), of the instantaneous power generated by the intermittent power generation plant, P(t), during the prediction time horizon, with respect to the instantaneous power values actually generated by the intermittent power generation plant, P(t).

(35) The method of the present invention can be combined with the method found in European patent EP3026774A1, included herein as reference, depending on the error calculated in the stage for calculating the error of the method of the present invention, such that if the calculated error exceeds a predetermined value, preferably in the range of 60%, more preferably it is 60%, the stage for controlling the state of charge, SOC(t), is the stage for controlling the target state of charge SOC.sub.tar(t) of European patent EP3026774A1, and vice versa, as long as the calculated error does not exceed the predetermined value, preferably in the range of 60%, more preferably it is 60%, the stage for controlling the state of charge, SOC(t), is the stage for controlling the stabilized state of charge, SOC.sub.sta(t), of the present invention.